Fentanyl Attenuates α^1B-Adrenoceptor–Mediated Pulmonary Artery Contraction

All experimental procedures and protocols were approved by the Institutional Animal Care and Use Committee of The Cleveland Clinic Foundation (Cleveland, Ohio).

Healthy mongrel dogs weighing 20–30 kg were anesthetized with intravenous pentobarbital sodium (30 mg/kg) and fentanyl citrate (15 μg/kg). After tracheal intubation, the lungs were mechanically ventilated. A catheter was placed in the femoral artery, and the dogs were exsanguinated by controlled hemorrhage. A left lateral thoracotomy was performed through the fifth intercostal space, and the heart was arrested with induced ventricular fibrillation. The heart and lungs were removed from the thorax en bloc . Lower right and left intralobar pulmonary arteries were dissected free. Intralobar pulmonary arteries (2–4 mm ID) were carefully dissected and immersed in cold modified Krebs-Ringer's bicarbonate (KRB) solution composed of 118.3 mm NaCl, 4.7 mm KCl, 1.2 mm MgSO⁴, 1.2 mm KH²PO⁴, 2.5 mm CaCl², 25 mm NaHCO³, 0.016 mm Ca-EDTA, and 11.1 mm glucose. The arteries were cleaned of connective tissue and cut into ring segments 4–5 mm in length. In all rings, the endothelium was removed by gently rubbing the intimal surface with a cotton swab. Removal of the endothelium was later verified by the absence of a vasorelaxant response to acetylcholine (10⁻⁶m).

Pulmonary arterial rings were vertically mounted between stainless steel hooks in organ chambers filled with 25 ml KRB solution (37°C) gassed with 95% oxygen and 5% carbon dioxide. One of the hooks was anchored, and the other was connected to a strain gauge (model FT03 force displacement transducer; Grass Instruments, West Warwick, RI) to measure isometric tension. The rings were stretched at 10-min intervals in increments of 1 g to achieve optimal resting tension. Optimal resting tension was defined as the minimum amount of stretch required to achieve the largest contractile response to 60 mm KCl and was determined to be 5 g for the size of arteries used in these experiments. After the rings had been stretched to their optimal resting tension, the contractile response to 60 mm KCl was measured. All rings were pretreated with the β-adrenoceptor antagonist propranolol (5 × 10⁻⁶m, incubated for 30 min) before phenylephrine administration to avoid any β-agonist effect of phenylephrine.

The effects of fentanyl (0.297 × 10⁻⁶, 7.86 × 10⁻⁶, 1.57 × 10⁻⁵m) on the phenylephrine concentration–response relation were assessed in size and positioned-matched pulmonary arterial rings. Fentanyl was added to the organ chambers 30 min before phenylephrine administration. The contractile responses to phenylephrine in fentanyl-pretreated rings were compared with responses in matched untreated rings.

To determine whether α²-adrenoceptor activation is involved in phenylephrine-induced contraction, pulmonary arterial rings were pretreated for 30 min with the α²-adrenoceptor antagonist rauwolscine (3 × 10⁻⁷m). The contractile responses to phenylephrine in rauwolscine-pretreated rings were compared with responses in matched untreated rings.

To assess the role of α¹-adrenoceptor activation in phenylephrine contraction and to identify the α¹-adrenoceptor subtypes involved in the response, pulmonary arterial rings were pretreated for 30 min with the following α¹-adrenoceptor antagonists: the non–subtype-selective α¹-adrenoceptor antagonist prazosin (3 × 10⁻¹⁰to 3 × 10⁻⁹m), the α^1A-adrenoceptor–selective antagonist 5-methylurapidil (10⁻⁷, 3 × 10⁻⁷, 10⁻⁶m), the α^1B-adrenoceptor–selective alkylating agent chloroethylclonidine (10⁻⁶, 2 × 10⁻⁵, 10⁻⁴m), and the α^1D-adrenoceptor–selective antagonist BMY 7378 (10⁻⁷, 3 × 10⁻⁷, 10⁻⁶m). Contractile responses to phenylephrine in antagonist-pretreated rings were compared with responses in matched untreated rings. In experiments with the irreversible α^1B-adrenoceptor–selective alkylating agent chloroethylclonidine, the antagonist was incubated for 30 min at 37°C, followed by six successive washouts with fresh KRB to remove chloroethylclonidine hydrolysis products. The phenylephrine concentration–response relation was then measured.

To determine whether fentanyl-induced changes in phenylephrine contraction involve α^1B-adrenoceptors, the rings were pretreated with fentanyl (7.86 × 10⁻⁶m), followed by a 30-min treatment with chloroethylclonidine (2 × 10⁻⁵m). Non–fentanyl-treated rings were suspended for 30 min in KRB and then were treated for 30 min with chloroethylclonidine (2 × 10⁻⁵m). After 30 min of exposure to chloroethylclonidine, all rings were repeatedly washed with fresh KRB to remove chloroethylclonidine hydrolysis products as described above, and concentration–response curves for phenylephrine were then measured.

Rat-1 fibroblasts were stably transfected with human α¹-adrenoceptor cDNA corresponding to the α^1A-, α^1B-, or α^1D-adrenoceptor subtypes (gift from GlaxoSmithKline, Uxbridge, United Kingdom). These cells were derived from clonal isolates. Cells were propagated in 75-cm²flasks in a humidified atmosphere (37°C) in Dulbecco's Modified Eagle Medium containing 5% fetal bovine serum, 10 U/ml penicillin, 100 μg/ml streptomycin, and 500 μg/ml of the selection antibiotic G418. Culture plates were scraped, and the stably transfected rat-1 fibroblast cells were collected in a sterile conical tube containing Hank's balanced salt solution. The cells were pelleted and then resuspended in a 5-ml volume of 0.25 m sucrose. After further centrifugation, the cells were resuspended in 10 ml of water containing protease inhibitors (leupeptin, phenylmethylmethylsulfonylfluoride, bacitracin, and benzamidine) and frozen at −70°C for 30 min. Cell membranes, disrupted by freezing and the hypotonic action of water, were processed further by Dounce homogenization. Nuclear debris was removed by pelleting under low-speed centrifugation (30,000g  for 15 min). Membranes were washed further by repeating the above step. Membranes were finally resuspended in HEM buffer (20 mm HEPES, pH 7.4, 1.4 mm EGTA, 12.5 mm MgCl²) containing 10% glycerol, aliquoted and stored at −70°C until use. Competition binding studies were performed to determine the affinity with which fentanyl inhibits the binding of the nonselective α¹-adrenoceptor antagonist ¹²⁵I-HEAT to the three α¹-adrenoceptor subtypes. Nonspecific binding of ¹²⁵I-HEAT was accounted for by the addition of 100 μm phentolamine and subtracted from total binding to achieve specific binding. Competition binding isotherms were generated by incubating a Ki (a measure of affinity of a ligand for a receptor) concentration of ¹²⁵I-HEAT with a fixed amount of stably transfected rat-1 fibroblast cell membranes together with various concentrations of fentanyl. All incubations were performed at 22°C in HEM buffer, in a total volume of 0.5 ml. Upon reaching steady state (approximately 60 min), bound radioactivity was separated from unbound by vacuum filtration through Whatman GF/C filter paper (Whatman Inc., Florham Park, NJ) using a Brandel Cell Harvestor (Brandel, Gaithersburg, MD). Filters were quickly washed with 20 ml ice-cold HEM buffer to remove further unbound radioligand. Bound radioactivity was counted on a gamma counter operating at 79.8% efficiency.

The following drugs and chemicals were used: phenylephrine, acetylcholine, propranolol, prazosin, phentolamine (Sigma Chemical, St. Louis, MO); 5-methylurapidil, chloroethylclonidine, BMY 7378 (8-[2-[4-(2-methoxyphenyl)-1-piperazinyl]ethyl]-8-azaspiro[4.5]decane-7,9-dione dihydrochloride), rauwolscine (Research BioChemical Incorporated, Natick, MA); fentanyl citrate (Abbott Laboratories, Abbott Park, IL); and ¹²⁵I-HEAT (Amersham Bioscience, Chicago, IL). All concentrations are expressed as the final molar concentration in the organ chamber. All drugs were dissolved in distilled water.

Values are expressed as mean ± SEM. Contractile responses to phenylephrine were calculated as a percentage of the maximum contractile response to 60 mm KCl.

The effects of fentanyl and the α-adrenoceptor antagonists on the phenylephrine concentration–response relation were assessed by calculating the phenylephrine concentrations that produced 50% of the maximal contractile response to 60 mm KCl (phenylephrine ED⁵⁰) using Graph Pad Prism version 3.0 (Graph Pad Software Inc., San Diego, CA). The phenylephrine ED⁵⁰values in the presence or absence of the antagonists were used to calculate the concentration ratio (CR). The pA²values represent the concentration of the antagonists necessary to displace the concentration–response curve of the agonist by twofold. The pA²values (−log M) were calculated by linear regression using Graph Pad Prism and were obtained from the x-intercept of the plot of log (CR-1) against log molar antagonist concentration, where the slope was not different from unity.9 

Competition binding data were analyzed by nonlinear regression using a noniterative curve-fitting program (GraphPad Prism). The best fit of the data was determined by comparison of the least squares value resulting from one- and two-site competition models in the F test from three to four individual experiments. The data fit best to the one-site competition model using the algorithm y = Bottom + (Top − Bottom)/(1 + 10^∧(X − LogEC⁵⁰)).

Statistical analysis was performed using the Student t  test for paired comparisons. Two-way analysis of variance, followed by the Tukey post hoc  test, was used when more than two means were compared. A P  value less than 0.05 was considered a statistically significant change. N refers to the number of dogs from which pulmonary arterial rings were studied in the isometric tension experiments.

We tested the hypothesis that fentanyl attenuates phenylephrine-induced contraction. Fentanyl alone had no effect on resting tension. Compared with the control condition, fentanyl caused a dose-dependent rightward shift (P < 0.05) in the phenylephrine concentration–response relation (fig. 1), i.e. , fentanyl attenuated phenylephrine contraction in canine pulmonary artery.

The selective α²-adrenoceptor antagonist rauwolscine had no effect on resting tension. Pretreatment of rings with rauwolscine had no effect on phenylephrine-induced contraction (fig. 2), which indicates that α²-adrenoceptor activation is not involved in phenylephrine contraction in canine pulmonary artery.

The selective α¹-adrenoceptor antagonist prazosin had no effect on resting tension. However, prazosin caused a rightward shift (P < 0.05) in the phenylephrine concentration–response curve (fig. 3A). The slope (0.905 ± 0.181) of the Arunlakshana and Schild plot for prazosin (fig. 3B) did not differ significantly from unity, and the pA²value for prazosin was 9.286 ± 0.095. This pA²value confirms that phenylephrine-induced contraction is mediated by α¹-adrenoceptors in canine pulmonary artery.10 

We investigated the role of the α^1A-adrenoceptor subtype in phenylephrine-induced contraction. 5-Methylurapidil, which has a higher affinity for the α^1A-adrenoceptor subtype, had no effect on resting tension. 5-Methylurapidil caused a parallel rightward shift (P < 0.05) in the phenylephrine concentration–response curve (fig. 4A). The slope (0.972 ± 0.113) of the Arunlakshana and Schild plot for 5-methylurapidil (fig. 4B) did not differ significantly from unity. The pA²for 5-methylurapidil was 7.185 ± 0.099, approximately 100-fold lower than the reported negative logarithm of the inhibitory constant6at α^1Aadrenoceptors, which suggests that the α^1A-adrenoceptor subtype does not play a primary role in phenylephrine-induced contraction in canine pulmonary artery.

We investigated the role of the α^1D-adrenoceptor subtype in phenylephrine-induced contraction. BMY 74378, which has a higher affinity for the α^1D-adrenoceptor subtype, had no effect on resting tension. BMY 74378 caused a rightward shift (P < 0.05) in the phenylephrine concentration–response curve (fig. 5A). The slope (0.919 ± 0.162) of the Arunlakshana and Schild plot (fig. 5B) did not differ significantly from unity, and the pA²value for BMY 7378 was 6.879 ± 0.098, approximately 100-fold lower than the reported pKi6at α^1Dadrenoceptors, which suggests that the α^1D-adrenoceptor subtype does not play a primary role in phenylephrine-induced contraction in canine pulmonary artery.

We investigated the role of the α^1B-adrenoceptor subtype in phenylephrine-induced contraction. Chloroethylclonidine, which selectively inactivates the α^1B-adrenoceptor subtype, had no effect on resting tension. Chloroethylclonidine at 10⁻⁶m had no significant effect on the phenylephrine concentration–response curve. However, chloroethylclonidine at 2 × 10⁻⁵m significantly decreased (P < 0.05) the maximal contractile response to phenylephrine (fig. 6), and chloroethylclonidine at 10⁻⁴m abolished phenylephrine-induced contraction. Together with the other subtype-selective antagonists, these results suggest that the α^1B-adrenoceptor subtype plays a primary role in phenylephrine-induced contraction in canine pulmonary artery.

If fentanyl attenuates phenylephrine-induced contraction via  an effect on the α^1B-adrenoceptor subtype, pretreatment with fentanyl before the administration of chloroethylclonidine should “protect”α^1Badrenoceptors from chloroethylclonidine and thereby potentiate phenylephrine contraction. As summarized in fig. 7, phenylephrine-induced contraction was potentiated (P < 0.05) in fentanyl-pretreated rings exposed to 2 × 10⁻⁵m chloroethylclonidine compared with the response observed with chloroethylclonidine alone.

Because fentanyl can have effects at the α^1Badrenoceptor through indirect mechanisms or cross-talk, we explored the possibility that fentanyl directly binds and occupies the activational pocket of α¹-adrenoceptor subtypes. Structural analysis of fentanyl reveals that it has pharmacophores similar to both α¹-adrenoceptor agonists and antagonists (fig. 8). Binding studies on cell membranes containing transfected individual α¹-adrenoceptor subtypes would determine whether fentanyl was subtype selective without the confounding effects of multiple subtypes as found in native tissue. As summarized in figure 9, fentanyl can compete with ¹²⁵I-HEAT binding at a single site of high affinity at each of the α¹-adrenoceptor subtypes. pKi values were 5.92 ± 0.032, 5.28 ± 0.031, and 5.53 ± 0.036, respectively, for the α^1B-, α^1D-, and α^1A-adrenoceptor subtypes, similar in potency to α¹-adrenoceptor agonists. The Ki of fentanyl for the α^1Badrenoceptor was approximately fivefold higher than that of the α^1Dadrenoceptor (P < 0.05) but was not significantly different from that of the α^1Aadrenoceptor.

Our laboratory is systematically examining the effects of general anesthetics on vascular mechanisms that regulate the pulmonary circulation. If an anesthetic agent either directly alters the pulmonary vasculature or alters agonist-induced changes in pulmonary vasomotor tone, this could have either a beneficial or deleterious effect on right ventricular function. The major finding in the current study is that fentanyl attenuates the pulmonary vasoconstrictor response to α¹-adrenoceptor activation via  a direct effect on the α^1B-adrenoceptor subtype. All other things being equal, this effect would decrease right ventricular afterload in the setting of sympathetic nervous system activation, which could improve right ventricular function.

Molecular pharmacology studies in rats and humans have demonstrated the expression of three subtypes of α¹-adrenoceptors (α^1A, α^1B, and α^1D).5However, only one study has assessed the relative contributions of these α¹-adrenoceptor subtypes to mediate phenylephrine-induced contraction in canine pulmonary artery7and reported that only one subtype (likely to be α^1B) had a high affinity for prazosin and was activated by phenylephrine. In the current study, pretreatment with the α¹-adrenoceptor antagonist prazosin caused a rightward shift in the phenylephrine concentration–response curve, which confirms that phenylephrine-induced contraction in canine pulmonary artery is mediated by α¹-adrenoceptor activation. Then, we investigated the contribution of each α¹-adrenoceptor subtype to phenylephrine-induced contraction. 5-Methylurapidil is a selective α^1A-adrenoceptor antagonist. The estimated affinity of 5-methylurpidil for the α^1Aadrenoceptor is approximately 50 times greater than for the α^1Dadrenoceptor and 100 times greater than for the α^1Badrenoceptor.11,12In the current study, the estimated affinity of 5-methylurapidil (pA²= 7.185 ± 0.099) was close to the value expected for an interaction at the cloned human α^1Bor α^1Dadrenoceptors rather than α^1Aadrenoceptors.11,12This suggests that phenylephrine contraction is not mediated by α^1A-adrenoceptor activation in canine pulmonary artery.

To determine whether the phenylephrine-mediated contraction in canine pulmonary artery was the α^1Bor the α^1Dadrenoceptor, BMY 7378 was used. BMY 7378 is a selective α^1D-adrenoceptor antagonist whose selectivity for α^1Dadrenoceptors is 100-fold greater than α^1Aor α^1Badrenoceptors.13In the current study, the estimated affinity of BMY 7378 was 6.879 ± 0.098, which is lower than the affinity reported for human recombinant α^1Dadrenoceptors expressed in rat-1 fibroblast (9.39) or in rat aorta (8.88), a tissue that exhibits a contractile function mediated predominantly by the α^1D-adrenoceptor subtype.14The pKⁱvalue reported for BMY 7378 for cloned human α^1Badrenoceptors stably expressed in rat-1 fibroblast was 6.7 ± 0.11, which is close to the pA²value for BMY 7378 in the current study.15Therefore, the inappropriately low affinity for BMY 7378 indicates that phenylephrine-induced contraction in canine pulmonary artery is not mediated by α^1D-adrenoceptor activation.

Chloroethylclonidine is an irreversible antagonist that preferentially alkylates and completely inactivates α^1Badrenoceptors, but it can also inactivate α^1Aor α^1Dadrenoceptors, depending on the treatment time and concentration used.8,11,12Chloroethylclonidine can be used to confirm the functional presence of α^1Badrenoceptors when used in conjunction with other more selective antagonists. In the current study, chloroethylclonidine at 10⁻⁶m had no effect on the phenylephrine concentration–response curve. However, chloroethylclonidine at 2 × 10⁻⁵m caused significant inhibition of phenylephrine contraction, and chloroethylclonidine at 10⁻⁴m completely abolished phenylephrine contraction. Taken together with the low affinity of BMY 7378 and 5-methylurapidil for α^1Badrenoceptors noted above, our results indicate that phenylephrine-induced contraction is primarily mediated by α^1Badrenoceptors in canine pulmonary artery. Chloroethylclonidine has been reported to cause concentration-dependent contraction in dog saphenous vein by stimulating α²adrenoceptors.16However, the α²-adrenoceptor antagonist rauwolscine had no effect on phenylephrine contraction in the current study, and treatment with chloroethylclonidine alone did not cause contraction. These results confirm the absence of functional α²adrenoceptors in canine pulmonary artery.

Fentanyl has been reported to attenuate phenylephrine contraction in isolated rabbit and rat aorta via  antagonism of α¹adrenoceptors.3,4However, the α¹-adrenoceptor subtypes involved in the fentanyl-induced attenuation of phenylephrine contraction have not been identified. We also observed that fentanyl attenuated phenylephrine contraction. Moreover, preincubation with fentanyl before exposure to chloroethylclonidine produced greater maximal phenylephrine-induced contraction compared with the response observed in rings pretreated with chloroethylclonidine alone, i.e. , fentanyl effectively protected α^1Badrenoceptors from the effects of chloroethylclonidine. These results suggest that fentanyl inhibits α^1Badrenoceptors in canine pulmonary artery by directly competing with chloroethylclonidine in the α¹-adrenoceptor activation pocket.

Fentanyl is a potent analgesic of the 4-anilidopiperidine class of opioids. It is a cationic selective full agonist at μ-opioid receptors.17As illustrated in figure 8, fentanyl is composed of both an α¹-adrenoceptor agonist functional group (pharmacophore) and an antagonist pharmacophore, suggesting that fentanyl should also bind and block the α¹-adrenoceptor activation pocket. In fact, common functional pharmacophores do exist between α¹adrenoceptors and opioids, because metabolites of some α¹-adrenoceptor antagonists form agonists at μ-opioid receptors.18Many ligands of the rhodopsin class of biogenic amine receptor family (i.e. , adrenergic, dopamine, serotonin, histamine) have similar binding pockets and can cross-activate or antagonize. This binding pocket similarity is also true of the μ-opioid receptor and the rhodopsin class of biogenic amine receptors.19,20We found that fentanyl can bind and compete for the activational binding pocket for all three α¹-adrenoceptor subtypes (fig. 9). The pKi values are similar to those of other α¹-adrenoceptor ligands. There was significant fivefold selectivity of fentanyl for the α^1Badrenoceptor over the α^1Dadrenoceptor.

Despite widespread distribution of the three α¹-adrenoceptor subtypes in many tissues, the α^1Badrenoceptor has not been linked to the activation of contraction in any of the systemic arteries in which it is expressed.21This has led to a theory that an α¹adrenoceptor can be expressed but cannot participate in contractile regulation, but may regulate other important functions in arterial smooth muscle (i.e. , growth). Our study is the first report that the α^1Badrenoceptor can predominantly regulate the contraction of pulmonary arterial smooth muscle. In contrast, the α^1Badrenoceptor is known to regulate venule contraction,22perhaps because of the higher receptor reserve in veins than in arteries.23Two studies have indicated that both the α^1Aand α^1Badrenoceptors regulate the human saphenous and umbilical veins.24,25The pulmonary artery shares functional similarity with veins in that both are low-pressure systems with relatively low blood oxygenation, which may influence the regulation of genes in those vessels. In fact, hypoxia is known to increase the expression of the α^1Badrenoceptor, which contains a hypoxia regulatory element on the α^1B-adrenoceptor promoter.26 

Concentrations of fentanyl used in this study are higher than typically encountered in clinical settings. However, rapid redistribution of fentanyl (octanol:water partition coefficient = 813)27to lipid-rich tissue, which also contain lipophilic G protein–coupled receptors, may create a discrepancy between serum concentration and actual tissue concentration. Because cerebrospinal fluid contains very little protein compared with plasma,28the active fentanyl concentration in cerebrospinal fluid averages approximately 46% of total plasma fentanyl concentration, which is more than twice the free fraction of plasma fentanyl.29Moreover, small changes in the amount or binding capacity of proteins in certain pathologic conditions (e.g. , liver disease, hemodilution, hypoproteinemia) could result in an increase in the free fraction of fentanyl. Because fentanyl is highly lipophilic, we believe the fentanyl concentrations (7.86 × 10⁻⁶, 1.57 × 10⁻⁵m) used in this in vitro  experiment are justified by the fact that fentanyl (6.9 ng/ml) in human blood corresponds to the fentanyl concentration (2.96 × 10⁻³m) calculated in the brain lipid.30 

In summary, our results indicate that phenylephrine-induced contraction in canine pulmonary artery is mediated primarily by α^1Badrenoceptors. Fentanyl attenuates phenylephrine contraction by binding to and directly inhibiting α^1Badrenoceptors.