Introduction

Top Abstract Introduction Materials and Methods Results Discussion References

Tachykinins such as Substance P (SP) and neurokinin A (NKA) can influence vascular tone, although the effect of these peptides may depend on the species and vessel studied. SP has been shown to cause relaxation in various human (Bodelsson and Stjernquist, 1992; Onoue et al., 1994; Wallerstedt and Bodelsson, 1997) and animal (D'Orléans-Juste et al., 1985; Bolton and Clapp, 1986; McCormack et al., 1989; Toda et al., 1991) vessels in vitro and to produce vasodilatation in human (McEwan et al., 1988; Strobel et al., 1996) and animal (Beattie et al., 1993; Hall and Brain, 1994) vascular beds in vivo. This relaxant action of SP is, however, not exclusive as there are also reports demonstrating SP to be capable of causing contraction of pulmonary arteries, including rabbit intralobar pulmonary arteries (McCormack et al., 1989; Shirahase et al., 1995).

Examination of the mechanisms underlying SP-induced vascular relaxation has revealed that the effect of SP is typically mediated through activation of tachykinin NK-1 receptors (Maggi et al., 1990; Constantine et al., 1991; Hall and Brain, 1994; Berthiaume et al., 1995; Corboz et al., 1998) and is dependent on the presence of an intact endothelium (D'Orléans-Juste et al., 1985; Bolton and Clapp, 1986; Bodelsson and Stjernquist, 1992; Onoue et al., 1994), although it has been suggested that in the dog endothelial-independent mechanisms may contribute as well (Enokibori et al., 1994). Several endothelial-dependent relaxant pathways have been implicated in mediating vascular relaxation produced by SP. These include the production of nitric oxide (Rosenblum et al., 1993), the production of relaxant prostanoids (Bodelsson and Stjernquist, 1994), and an endothelial-dependent hyperpolarization of the smooth muscle membranes (Petersson et al., 1995; Wallerstedt and Bodelsson, 1997). Again, involvement of specific relaxant mechanisms in SP-induced vasodilatation may depend on the particular species and vascular bed studied, so that different endothelium-dependent mechanisms may account for SP-induced relaxation in different vessels. Moreover, there is evidence that multiple relaxant pathways can be involved in the relaxation response to SP within a single vessel type (Enokibori et al., 1994; Petersson et al., 1995; Wallerstedt and Bodelsson, 1997).

SP has recently been reported to cause relaxation of human isolated pulmonary arteries via stimulation of NK-1 receptors (Corboz et al., 1998). The mechanism of this effect and pharmacological characteristics of the receptor, however, have not been thoroughly characterized. The present study was undertaken to investigate the mechanisms involved in tachykinin-mediated relaxation of human pulmonary arterial rings, and to provide additional pharmacological information on the characteristics of the receptor mediating the response.

	Materials and Methods

Top Abstract Introduction Materials and Methods Results Discussion References

Macroscopically normal human lung tissue was obtained from 42 organ donors (supplied by the International Institute for the Advancement of Medicine, Exton, PA or the Anatomical Gift Foundation, Woodbine, GA). Organ donor specimens were mainly obtained from victims of head trauma or cerebral vascular accidents. The donors were male (24) and female (18) with an average age of 33 ± 2 years (range 13 to 61 years). Lungs from donors with documented pulmonary pathology were not used in these studies. Organ donor specimens were placed in cooled (4°C) Eagle's minimum essential medium and transferred to the laboratory within 24 h. On reaching the laboratory, tissues were immediately placed in 4 liters of modified Krebs' bicarbonate solution of the following composition: NaCl, 118 mM; KCl, 5.4 mM; NaH₂PO₄, 1.0 mM; MgSO₄, 1.2 mM; CaCl₂, 1.9 mM; NaHCO₃, 25 mM; and D-glucose, 11.1 mM, and gassed with 95% oxygen and 5% carbon dioxide at 4°C.

Tissue Bath Studies. Intralobar pulmonary arteries (i.d. ~5 mm) were dissected free of surrounding parenchymal lung tissue and prepared as rings 8 to 10 mm in length. The rings were placed over stirrups made of tungsten wire and inserted over straight tungsten pins suspended in water-jacketed (37°C) tissue baths containing 10 ml of Krebs' solution. Preparations were gassed continuously with 95% oxygen and 5% carbon dioxide. Propranolol (10⁶ M) was added to the buffer at the beginning of experiments to block -adrenergic receptor activity. Adjacent human pulmonary artery ring segments were used as control or treated tissues. In each experiment at least two tissues served as control, one ring obtained from either side of the treated tissues.

Measurement of Mediator Release. Specimens of human intralobar arteries were dissected free from lung tissue, cleared of surrounding parenchyma, and prepared as rings 10 to 20 mm in length. These were pooled, divided into four samples (each containing approximately 100 mg of tissue), and placed into 2 ml of Krebs' bicarbonate buffer containing propranolol (1 µM). Tissues were incubated at 37°C in a water bath and were gassed continuously with 95% oxygen and 5% carbon dioxide. Arterial specimens underwent a 60-min equilibration period during which tissues were washed every 15 min with fresh buffer. At the end of the 60-min equilibration period, tissues were placed in 1 ml of buffer and all tissues received thiorphan (1 µM). After an additional 15 min, the buffer was removed and 1 ml of fresh buffer containing thiorphan (1 µM) was added to all tissues. Arterial specimens were then exposed to ASM-SP (0.1 µM) or vehicle control for 15 min at 37°C after which the supernatant was collected. Tissues were then gently blotted with a cotton swab to remove surface moisture, and weighed.

2

1

2

Analysis of Results. All numerical data are expressed as arithmetic mean ± S.E. The n values represent the number of separate experiments carried out with vascular tissue obtained from different donors. In studies of isolated human pulmonary artery log (M) EC₅₀ values were determined as the log M concentration of the agonist that produced 50% of the maximum relaxation in each concentration-effect curve. The apparent dissociation constant (K_b) was calculated for antagonists that caused a parallel rightward shift in the agonists concentration-response curves, using the standard equation, [antagonist]/(dose ratio  1), converted to the negative logarithm and expressed as log molar K_b (pK_b). Differences between means were determined using ANOVA or Student's t test; probability values (P) <.05 were considered significant.

Drugs and Solutions. SP, NKA, ASM-SP, [-Ala⁸]NKA(4-10), and senktide analog were obtained from Cambridge Biochemical (Wilmington, DE). Thiorphan, CP 99994, SR 140333, FK 888, RP67580, and L732138 were synthesized at Zeneca, Inc. (Wilmington, DE). SIN-1 was a gift from Casella AG, (Frankfurt, Germany). Phenylephrine, propranolol, indomethacin, L-NNA, papaverine, and barium chloride were obtained from Sigma Chemical Co. (St. Louis, MO). Indomethacin and thiorphan were dissolved in ethanol at a stock solution of 10 mM. Stock solutions (10 mM) of the NK-receptor antagonists and agonists were prepared in dimethyl sulfoxide and diluted to final concentration in Kreb's buffer solution. Papaverine, phenylephrine, and barium chloride were dissolved in distilled water.

	Results

Top Abstract Introduction Materials and Methods Results Discussion References

Neurokinin Receptor Agonists. Neither NKA nor SP, at concentrations up to 1 µM, contracted the human isolated arterial rings with or without endothelium (n = 3, data not shown). Both SP and NKA were potent and effective in relaxing arteries precontracted with phenylephrine (Fig. 1). There was no significant difference in the maximum response between the two agonists; however, SP was 5-fold more potent than NKA, with respective log EC₅₀ values of 9.0 ± 0.1 and 8.3 ± 0.3 (n = 42 and 8, Fig. 1).

View larger version (18K): [in this window] [in a new window]   Fig. 1.   Log concentration-response curves for SP (), NKA (), and ASM-SP () in causing relaxations of human intralobar pulmonary arteries precontracted with phenylephrine. The data are expressed as a percentage of maximum relaxation obtained with papaverine (1 mM) added at the end of the experiment. Vertical lines represent the S.E.M. of 42 (), 8 (), and 11 () experiments.

Neurokinin Receptor Antagonists. The selective NK-1 receptor antagonists CP 99994 and FK 888 shifted the ASM-SP concentration-response curves in a parallel rightward fashion (Fig. 2). The calculated pK_b values obtained with two different concentrations of CP99994 were the same, averaging about 9.0 (Table 1). The calculated pK_b values of FK 888 were also independent of antagonist concentration, averaging about 7.8 (Table 1). Two additional NK-1 selective receptor antagonists, RP 67580 and L732138, also caused rightward shifts in the ASM-SP concentration response curves, although they were considerably less potent than CP 99994. The calculated pK_b for L732138 was dependent on the antagonist concentration, suggesting effects of this compound in addition to simple competitive antagonism (Table 1). In three experiments, CP99994 also shifted the concentration-response curves for NKA in a parallel rightward fashion without affecting the maximum response (data not shown). In these experiments the estimated pK_b values were 9.5, 9.2, and 9.5, which is similar to that observed when ASM-SP was the agonist.

View larger version (25K): [in this window] [in a new window]   Fig. 2.   Log concentration-response curves for ASM-SP in causing relaxations of human intralobar pulmonary arteries, precontracted with phenylephrine, in the absence (closed symbols) or presence (open symbols) of the stated concentration of an NK-1 receptor antagonist. The antagonists were added 60 min before the addition of the agonists. D, concentrations of SR 140333 were 0.3 nM () 1 nM (), and 3 nM (). The data are expressed as a percentage of maximum relaxation obtained with papaverine (1 mM) added at the end of the experiment. The control and drug-treated tissues were adjacent segments of same artery each studied in a separate tissue bath. The vertical lines represents the S.E.M., n = 11 (A), 9 (B), 6 (C), and 6 (D).

Relaxation Mechanism. The relaxations to the ASM-SP were dependent on an intact endothelium. In six of seven experiments, there was no relaxant response to ASM-SP in endothelial denuded tissues, whereas the maximum response to ASM-SP averaged 47 ± 9% in the adjacent segment in which the endothelium was intact. In one of seven experiments ASM-SP caused a 34% relaxation in the tissue in which the endothelium was putatively removed.

View larger version (17K): [in this window] [in a new window]   Fig. 3.   Log concentration-response curves for ASM-SP in causing relaxation of human intralobar pulmonary arteries precontracted with phenylephrine. Sixty minutes before the addition of the first concentration of ASM-SP, the tissues were treated with vehicle (), indomethacin, 3 µM (), L-NNA, 100 µM (), or indomethacin and L-NNA (). The data are expressed as a percentage of maximum relaxation obtained with papaverine added at the end of the experiment. The vertical lines represent the S.E.M., n = 8 experiments. There was no significant effect (ANOVA) on the log M EC50 values between tissues treated with vehicle (8.9 ± 0.2), indomethacin (9.0 ± 0.2), and L-NNA (8.5 ± 0.4). The maximum response in the treated tissues was significantly (P < .05) less than that obtained in the vehicle-treated tissue. In five of eight experiments, the artery treated with the combination of indomethacin and L-NNA the maximum response to ASM-SP was less than 1%.

2

1

1

View larger version (18K): [in this window] [in a new window]   Fig. 4.   The release of prostanoids from human intralobar pulmonary arteries in the presence of vehicle (filled columns) and ASM-SP, 0.1 µM (open columns). The prostanoids were quantified using negative ion chemical ionization gas chromatography/mass spectrometry. Values are expressed as nanograms per gram wet weight of artery. Vertical lines represent the S.E.M., n = 6. *, significant difference in the amount of prostanoid released from the vehicle and ASM-SP-treated tissues. In some tissues the endothelium was removed (-endothelium) at the beginning of the experiment.

	Discussion

Top Abstract Introduction Materials and Methods Results Discussion References

This study demonstrates that the human isolated pulmonary artery is a useful functional assay for the study of human NK-1 receptor pharmacology. The NK-1 receptor-mediated relaxations are relatively consistent among tissues, and are not influenced by NK-2 or NK-3 receptor mediated responses. Moreover, the NK-1 dependent relaxations are not affected by direct contractile effects of neurokinins at the level of the smooth muscle.

The pharmacological characterization of human NK-1 receptors is based primarily on binding studies typically using cell lines expressing the human cloned receptor (Fong et al., 1992a,b; McLean et al.,1993; Cascieri et al., 1994; Walpole et al., 1998). Functional studies on human NK-1 receptors have been based mainly on rabbit and guinea pig isolated tissues. There are relatively few studies that have investigated the pharmacology of NK-1 receptors in human isolated tissues. The importance of developing a functional assay using isolated tissues for the human NK-1 receptors is underscored by the known species differences in the pharmacology of this receptor (Fong et al., 1992b; Aramori et al., 1994). The species differences in the receptor pharmacology can be noted with CP 99994. Supporting previous observations by Corboz et al. (1998), we found that CP 99994 antagonizes the NK-1 receptor-mediated relaxations with a dissociation constant in the nanomolar range. This reflects a potency that is approximately 1000 times greater than that noted in functional studies using the mouse stomach (Nsa Allogho et al., 1997), and about 5 to 10 times greater than its potency in functional studies using the guinea pig ileum (Hosoki et al., 1998). By contrast, FK888 was considerably less potent in inhibiting NK-1 responses in the human pulmonary artery (present study) than in the guinea pig ileum (Walpole et al., 1998). Likewise, RP 67580 was much less potent in inhibiting the NK-1-mediated responses in the human pulmonary artery (present study) than in the mouse stomach strip (Nsa Allogho et al., 1997). The potency of RP 67580 observed in the present study was similar, however, to that observed in the guinea pig ileum and rabbit iris preparations (Carruette et al., 1992; Hall et al., 1994).

Differences that arise between functional analysis and binding studies on human cloned receptors expressed in various cell systems also serve to point out the utility of including functional data in pharmacological characterizations of the human NK-1 receptor. All of the NK-1 receptor antagonists evaluated in the present study have estimated affinity constants or IC₅₀ values, obtained from receptor bindings studies, in the subnanomolar to nanomolar range (Fong et al., 1992a; McLean et al., 1993; Cascieri et al., 1994; Walpole et al., 1998). Thus the binding affinity categorically overestimated the functional potency of the compound on the human NK-1 receptors as evaluated using the human pulmonary artery. In some cases the difference was relatively minor (e.g., CP 99994 and SR 140333). In other cases, however, a major discrepancy was noted. For example, L 732,138 inhibits SP binding to human NK-1 receptors stably expressed in Chinese hamster ovary cells with an IC₅₀ of about 2 nM (Cascieri et al., 1994). This concentration is two to three orders of magnitude off its estimated dissociation constant in our functional studies with the human pulmonary artery.

Each of the NK-1 receptor antagonists studied in the present study are competitive antagonists in binding assays. Consistent with this, with the exception of SR 140333, the antagonism with these compounds could be surmounted by increasing agonist concentrations. The basis of the insurmountable antagonism afforded by SR 140333 is not clearly understood. It is unlikely to be explained by a nonselective effect of SR 140333 inasmuch as the compound had no effect on the relaxations caused by the nitric oxide donor molecule SIN-1. It is worth noting that even within a single species both surmountable and insurmountable antagonism of NK-1 receptor agonist-induced responses have been noted with SR 140333. In the guinea pig isolated trachea SR 140333 is an apparent competitive antagonist of ASM-SP-induced contractions causing a parallel rightward shift without affecting the maximum agonist response (Canning et al., 1998). In the guinea pig ileum, however, consistent with the observations presented here, SR 140333 inhibits NK-1 receptor mediated contractions in a insurmountable fashion (Croci et al., 1995).

NKA is often considered to be a selective NK-2 receptor agonist. In receptor binding studies, SP has an affinity that is 50 times greater than NKA for the wild-type NK-1 receptor (Fong et al., 1992a). NKA, however, was nearly as potent as SP in relaxing the human pulmonary artery. That NKA was acting on the NK-1 receptor to relax the artery is supported by the competitive antagonism observed with CP99994. There has been speculation that the NK-1 receptor may exist in two conformers. One conformer shows selectivity for SP whereas the other conformer can be stimulated by NKA and other septide-like agonists. The similarity in the potency of SP and NKA in relaxing the artery is consistent with the hypothesis that the NK-1 receptor involved in this response is the "general-tachykinin conformer" (Maggi and Schwartz, 1997). The average EC₅₀ value observed for NKA in the present study (8 nM) was virtually the same as the EC₅₀ value we observed for NKA in the NK-2-receptor-mediated contraction of human isolated bronchi (5 nM) (Ellis et al., 1997). Therefore, NKA is not an NK-2 selective agonist in human pulmonary tissues.

Our observation that in human pulmonary artery, ASM-SP-induced relaxation was endothelial dependent is in agreement with numerous prior studies in other types of human blood vessels (Onoue et al., 1994; Petersson et al., 1995; Wallerstedt and Bodelsson, 1997). The conclusion can thus be drawn from the data obtained with indomethacin and L-NNA (Fig. 3) that in human pulmonary artery it is a combination of the activities of both nitric oxide and cyclooxygenase that predominantly mediates the endothelium-dependent vascular relaxation evoked by NK-1 receptor stimulation.

The mass spectroscopy data presented in Fig. 4, considered with data from the tissue bath studies, leads to the conclusion that prostacyclin is most likely the cyclooxygenase metabolite, derived from endothelial cells, responsible for mediating the indomethacin-sensitive portion of the NK-1 receptor relaxation in human pulmonary artery. This is in contrast to the rabbit pulmonary artery where NK-1 receptor activation leads to endothelium-dependent thromboxane production and consequent contraction of the smooth muscle (Shirahase et al., 1995). It is possible that nitric oxide released from endothelial cells may itself be the stimulus for prostaglandin production. Several reports have shown that nitric oxide stimulates prostaglandin biosynthesis (Salvemini et al., 1993; Sautebin et al., 1995; Landino et al., 1996). It is thus possible that NO may contribute to the NK-1 receptor-mediated prostacyclin production in the pulmonary artery. It is unlikely, however, that NO is the major stimulus of prostacyclin in our studies, because the inhibitory effects of indomethacin and L-NNA were additive.