Pulmonary Pharmacology (1990) 3 137-144 ©1990 Longman Group UK Ltd
0952-0600/90/0003-0137/$10 .00
PULMONARY PHARMACOLOGY
Peripheral Type Benzodiazepine Receptors and Response to Adenosine on the Guinea-pig Isolated Trachea C . Advenier*, P . Devillier*t, M . Blanc*, J . P . Gnassounou* *Department of Pharmacology, Faculty of Medicine, 15, rue de l'Ecole de Medecine, F-75270 Paris Cedex 06 tPasteur Vaccins, 3 avenue Pasteur, 92430 Marnes la coquette, France SUMMARY . It has been reported that dipyridamole, an adenosine uptake inhibitor, and some benzodiazepines potentiate the responses to adenosine in peripheral organs and in particular, on guinea-pig isolated atria . In this paper, we have examined the potentiation of responses to adenosine produced by dipyridamole, diazepam and four compounds with selective agonistic activity towards the central (clonazepam) or peripheral (Ro5-4864) type benzodiazepine receptors or antagonistic activity towards the central (flumazenil) or peripheral (PK 11195) benzodiazepine receptors in guinea-pig trachea in vitro. In preparations under basal tone and in the absence of adenosine, dipyridamole (10 - ' M) and benzodiazepines (10 -4 M) with the exception of flumazenil induced a relaxation of the airway smooth muscle . In addition, diazepam (10 -4 M) attenuated the phasic response to histamine (10 - ' M) . Dipyridamole, and the benzodiazepine agonists diazepam, Ro5-4564 and clonazepam (10 - ' to 10 -4 M) produced potentiation of the tracheal response to adenosine, the rank order of potency being dipyridamole (pK1 = 7 .77 ± 0.12, n = 8) > Ro5-4864 (pK 1 = 5 .43 ± 0 .18, n = 6) >, diazepam > clonazepam (pK 1 = 4.84 ± 0 .11, n = 6). The two benzodiazepine receptor antagonists, flumazenil and PK 11195, gave a significant but small potentiation to adenosine only at 10 -4 M . In the presence of dipyridamole (10 - ' M), diazepam (10 -4 M) did not cause any further potentiation to adenosine . Additionally, the potentiation produced by diazepam was not antagonised by flumazenil, whereas it was potently antagonised by PK 11195 . Similarly, PK 11195 potently inhibited the adenosine potentiation produced by Ro5-4864 . In the presence of dipyridamole, the antagonism between adenosine and theophylline appeared to be of the competitive type whereas it did not in the presence of diazepam or in the absence of dipyridamole . Mechanical removal of the tracheal epithelium significantly potentiated the relaxation of guineapig trachea to adenosine but did not alter the effect of dipyridamole or diazepam . Benzodiazepines therefore appear to potentiate the response of guinea-pig trachea to adenosine by inhibition of adenosine uptake, and this might occur through activation of a peripheral type benzodiazepine receptor .
related to the activation of a peripheral type benzodiazepine receptor .' Numerous recent reports have focused on the interaction of benzodiazepine with adenosine at central and peripheral levels 9 ' 10 and it has been suggested that potentiation of adenosine effects is one mode of action of benzodiazepines . This may occur through inhibition of adenosine uptake and/or enhancement of adenosine release . 10 At the peripheral level, benzodiazepines have been shown to potentiate the inhibitory response to adenosine on guinea-pig isolated left atria and rat vas deferens by inhibition of adenosine uptake .' 1 .12 In airway tissue, adenosine plays a significant role in the regulation of bronchomotor tone 13-16 and it has been reported that inhibition of adenosine uptake potentiates the relaxant response to adenosine in guinea-pig trachea . 13'1' The first objective of our study therefore was to investigate the adenosine potentiating effect of diazepam on isolated guinea-pig trachea . Diazepam binds to both peripheral and central type benzodiazepine receptors .
INTRODUCTION Non-neural binding sites for benzodiazepines have been demonstrated in many peripheral tissues and have been designated peripheral type benzodiazepine receptors 1-4 or S23 receptors .' To date, no function has been attributed to these peripheral binding sites .' The existence of the peripheral type receptor has been demonstrated by radioligand binding studies both in heart' and trachea .' The peripheral effects of benzodiazepines include negative inotropism and vasodilatation in cardiac and vascular smooth muscles'" and relaxation of airway smooth muscles,' although these effects have been ascribed at least in part to interference with Ca ++ movements and did not appear to be
Address correspondence to : C . Advenier, Dept of Pharmacology, Faculty of Medicine, 15, rue de l'Ecole de Medecine, F-75270 Paris Cedex 06 . 137
138
Pulmonary Pharmacology
Since there is now convincing evidence that peripheral type benzodiazepine receptors are localized in central structures and, conversely, central type receptors in peripheral structures,', 111-20 we have also studied the adenosine potentiating effects of benzodiazepine agonists and antagonists displaying high affinity and selectivity for the central or peripheral receptor type . Accordingly, in addition to diazepam we have used four compounds, clonazepam and the benzodiazepine antagonists flumazenil (Ro 15-1788) 21 which bind selectively to the central type receptor, and Ro5486422 and PK 11195, isoquinoline-carboxamide compounds, 1,8,23 which bind selectively to the peripheral type receptor . In addition, since we have previously demonstrated that adenosine uptake interferes with the pharmacological evaluation of the adenosine-theophylline antagonism, 24 our second objective was to examine the effect of diazepam on this antagonism . Finally, since epithelium removal potentiates the relaxant effect of adenosine, 24 we tried to determine whether the epithelium modulates the benzodiazepine interaction with adenosine .
Protocols
METHODS
In all experiments, tracheal strips were first contracted to maximal tension with carbachol 10 -4 M and then relaxed with theophylline (3 x 10 - ' M). After 1 h rest, with washing every 15 min, 10 min pretreatments with saline, dipyridamole or benzodiazepine ligands were made . The strips were contracted with histamine (10 -5 M) and after equilibration, cumulative concentration-response curves to adenosine were obtained by increasing the concentration of adenosine at 10- 15 min intervals in logarithmic increments . When the concentration-response curves to adenosine were completed, theophylline 3 x 10 - ' M was added to the bath to determine maximal relaxation . Owing to the development of tachyphylaxis, only one series of adenosine concentrations was used for each tracheal preparation . The adenosine-induced relaxation was expressed as a percentage of the maximal effect of theophylline 3 x 10 -3 M . The -log EC50 values (defined as the negative log of the drug concentration that caused 3 50% of maximal effect of theophylline 3 x 10_ M), were derived from the log concentration-effect curves . These values were evaluated graphically from each experiment. pA 2 values were determined according to Arunlakshana and Schild .29
Tissue preparation
Measurement of potentiation
Male guinea-pigs (250-350 g) were killed by a blow to the head and exsanguinated . The trachea was removed and placed in Krebs-Henseleit solution (composition mM : NaCl 114, KCI 4 .7, CaCl 2 2.5, KH2PO4 1 .2, MgSO 4 1 .4, NaHCO 3 25 .0, glucose 11 .7) . Following removal of adhering fat and connective tissue the trachea was slit open along its longitudinal axis, directly opposite the smooth muscle, and three strips were prepared according to the zig-zag method of Emmerson and Mackay . 25 The strips were placed in parallel in 25 ml organ baths containing Krebs-Henseleit solution at 37 ° C, gassed with 95%0 2 and 5%CO 2 and tied to transducers . The tissues were equilibrated under an initial tension of 1 .80 g and washed every 15 min before the start of the experiments . After equilibration for 1 .25 h, the resting tension was between 0 .8 and 1 .4 g . Under these conditions, responses to agonists were reproducible . Tensions were measured isometrically with Celaster strain gauges and were displayed on Linseis recorders . In some experiments the epithelium was removed by gently rubbing the luminal surface (over both the smooth muscle and cartilage areas) with a cottontipped applicator . 24 ' 26-28 Comparisons between epi thelium containing and epithelium-denuded preparations were made on adjacent strips from the same area of trachea .
A method previously described was used to quantitate potentiation of peripheral tissues to adenosine and study inhibition by diazepines of adenosine uptake . 12,30 Sensitivity of trachea to adenosine was quantitated as the molar concentration of agonist producing 50% of maximal response (EC 50) in the presence of various concentrations of benzodiazepines or dipyridamole . Potentiation to the agonist, denoted as x, was defined as the EC 50 in the absence of uptake inhibitor divided by the EC 50 in the presence of uptake inhibitor . Maximal potentiation, denoted y, was defined as the potentiation produced by 10 -5 M dipyridamole . It was not possible to determine the potentiating effect of higher concentration of dipyridamole (10 -4 M) because such concentration abolished the responses to histamine . Benzodiazepines were considered to be devoid of antagonistic effect vs adenosine as previously demonstrated by Kenakin 12 for diazepam and clonazepam . Experiments conducted with Ro5-4864, PK 11195 and flumazenil -5 4 showed that 10 M and 10" M concentrations of these drugs did not exert an antagonistic effect versus 2-chloroadenosine . The estimates of potentiation to adenosine were then utilized in the following equation" log Y(x-1) = log [I]-log K . (Y - x) where y is the maximal potentiation in the same tissue
Benzodiazepine Interaction with Adenosine
from which estimates of x were made, [I] is the molar concentration of uptake inhibitor and K ; theoretically is the equilibrium dissociation constant of the inhibitor for the site of uptake (assuming competitive antagonism of adenosine removal from the receptor compartment) . Regressions of log (y(x-1)/(y- x)) on log [I] yielded estimates of the pK ; (-log K ;) as the intercepts as well as the slopes of the regression lines and the pK ; .
1 39
Influence of benzodiazepines and dipyridamole on histamine induced contraction Phasic responses to histamine 10 -5 M were unaffected by pretreatment with dipyridamole or benzodiazepines except for diazepam which at a concentration of 10 -4 M caused a weak but significant reduction of the phasic contractile response to histamine 10 - 5 M (39 .8 ± 4 .1 % of ACh maximal effect versus control 61 .9 ± 6.6%, p < 0 .05) .
Statistical analysis of results Statistical analysis of the results was performed using analysis of variance and then Student's t-test for paired or unpaired data . All values in the text and table are expressed as mean±s .e .mean . Drugs The drugs used were : adenosine (Merck, D-Darmstadt), histamine (Sigma, St Louis, USA), dipyridamole (Boehringer-Ingelheim, F-Reims), theophylline sodium anisate (Bruneau, F-Paris), diazepam, Ro5-4864 (7-chlor-5-(4-chlorphenyl)-1,3-dihydro-l-methyl-2H1,4-benzodiazepin-2-on), flumazenil (ethyl 8-fluoro5,6-dihydro-5-methyl-6-oxo-4H-imidazo[1,5-a][1,4]benzodiazepine-3-carboxylate) (Roche, CH-Basle) and PK 11195 (N-methyl-N-(methyl-l-propyl)(chloro-2 phenyl)-1 isoquinoleine-3 carboxamide) (Rhone-Poulenc, F-Vitry sur Seine) . Theophylline was used as proprietary injectable solution (Theophylline Bruneau) ; dipyridamole was dissolved daily in ethanol and the solution was further diluted with Krebs solution. Diazepines were dissolved in sulfolane (tetramethylene sulfone, Aldrich Chemie, FRG-Steinheim) . The appropriate concentration of drugs was added to the organ baths so that at no time did the volume of sulfolane or ethanol exceed 0 .1 ml (0,4% by volume) . The solvents in such maximal concentrations produced no direct effect on basal tone or on responses to adenosine .
Influence of dipyridamole and benzodiazepines on adenosine concentrations-response curves In the presence of dipyridamole 10' M to 10 -5 M, concentration-effect curves to adenosine were shifted to the left, with a maximal effect at 10 -5 M corresponding to a shift of 1 .89 ± 0 .12 log unit, that is a 79.4 fold potentiation (Fig . 1) . The apparent pK ; of dipyridamole as inhibitor of adenosine uptake calculated according to Kenakini 2 was 7 .77±0 .12 (Fig . 2, Table 1 In the presence of benzodiazepines, concentrationresponse curves of adenosine were also shifted to the left (Fig . 1) with potentiation of 47 .8, 25 .1, 4 .46, 2 .51 and 2 .37 fold respectively for diazepam (10 -4 M), Ro5-4864 (10 -4 M), clonazepam (10 -4 M), flumazenil (10 -4 M) and PK 11195 (10 -4 M) . However, the shifts of the dose responses curves in the presence of
RESULTS Influence of benzodiazepines and dipyridamole on basal tone Dipyridamole, at concentration of 10 -5 M, caused a significant relaxation to 88+5 .5% (p
0952-0600/90/0003-0137/$10 .00
PULMONARY PHARMACOLOGY
Peripheral Type Benzodiazepine Receptors and Response to Adenosine on the Guinea-pig Isolated Trachea C . Advenier*, P . Devillier*t, M . Blanc*, J . P . Gnassounou* *Department of Pharmacology, Faculty of Medicine, 15, rue de l'Ecole de Medecine, F-75270 Paris Cedex 06 tPasteur Vaccins, 3 avenue Pasteur, 92430 Marnes la coquette, France SUMMARY . It has been reported that dipyridamole, an adenosine uptake inhibitor, and some benzodiazepines potentiate the responses to adenosine in peripheral organs and in particular, on guinea-pig isolated atria . In this paper, we have examined the potentiation of responses to adenosine produced by dipyridamole, diazepam and four compounds with selective agonistic activity towards the central (clonazepam) or peripheral (Ro5-4864) type benzodiazepine receptors or antagonistic activity towards the central (flumazenil) or peripheral (PK 11195) benzodiazepine receptors in guinea-pig trachea in vitro. In preparations under basal tone and in the absence of adenosine, dipyridamole (10 - ' M) and benzodiazepines (10 -4 M) with the exception of flumazenil induced a relaxation of the airway smooth muscle . In addition, diazepam (10 -4 M) attenuated the phasic response to histamine (10 - ' M) . Dipyridamole, and the benzodiazepine agonists diazepam, Ro5-4564 and clonazepam (10 - ' to 10 -4 M) produced potentiation of the tracheal response to adenosine, the rank order of potency being dipyridamole (pK1 = 7 .77 ± 0.12, n = 8) > Ro5-4864 (pK 1 = 5 .43 ± 0 .18, n = 6) >, diazepam > clonazepam (pK 1 = 4.84 ± 0 .11, n = 6). The two benzodiazepine receptor antagonists, flumazenil and PK 11195, gave a significant but small potentiation to adenosine only at 10 -4 M . In the presence of dipyridamole (10 - ' M), diazepam (10 -4 M) did not cause any further potentiation to adenosine . Additionally, the potentiation produced by diazepam was not antagonised by flumazenil, whereas it was potently antagonised by PK 11195 . Similarly, PK 11195 potently inhibited the adenosine potentiation produced by Ro5-4864 . In the presence of dipyridamole, the antagonism between adenosine and theophylline appeared to be of the competitive type whereas it did not in the presence of diazepam or in the absence of dipyridamole . Mechanical removal of the tracheal epithelium significantly potentiated the relaxation of guineapig trachea to adenosine but did not alter the effect of dipyridamole or diazepam . Benzodiazepines therefore appear to potentiate the response of guinea-pig trachea to adenosine by inhibition of adenosine uptake, and this might occur through activation of a peripheral type benzodiazepine receptor .
related to the activation of a peripheral type benzodiazepine receptor .' Numerous recent reports have focused on the interaction of benzodiazepine with adenosine at central and peripheral levels 9 ' 10 and it has been suggested that potentiation of adenosine effects is one mode of action of benzodiazepines . This may occur through inhibition of adenosine uptake and/or enhancement of adenosine release . 10 At the peripheral level, benzodiazepines have been shown to potentiate the inhibitory response to adenosine on guinea-pig isolated left atria and rat vas deferens by inhibition of adenosine uptake .' 1 .12 In airway tissue, adenosine plays a significant role in the regulation of bronchomotor tone 13-16 and it has been reported that inhibition of adenosine uptake potentiates the relaxant response to adenosine in guinea-pig trachea . 13'1' The first objective of our study therefore was to investigate the adenosine potentiating effect of diazepam on isolated guinea-pig trachea . Diazepam binds to both peripheral and central type benzodiazepine receptors .
INTRODUCTION Non-neural binding sites for benzodiazepines have been demonstrated in many peripheral tissues and have been designated peripheral type benzodiazepine receptors 1-4 or S23 receptors .' To date, no function has been attributed to these peripheral binding sites .' The existence of the peripheral type receptor has been demonstrated by radioligand binding studies both in heart' and trachea .' The peripheral effects of benzodiazepines include negative inotropism and vasodilatation in cardiac and vascular smooth muscles'" and relaxation of airway smooth muscles,' although these effects have been ascribed at least in part to interference with Ca ++ movements and did not appear to be
Address correspondence to : C . Advenier, Dept of Pharmacology, Faculty of Medicine, 15, rue de l'Ecole de Medecine, F-75270 Paris Cedex 06 . 137
138
Pulmonary Pharmacology
Since there is now convincing evidence that peripheral type benzodiazepine receptors are localized in central structures and, conversely, central type receptors in peripheral structures,', 111-20 we have also studied the adenosine potentiating effects of benzodiazepine agonists and antagonists displaying high affinity and selectivity for the central or peripheral receptor type . Accordingly, in addition to diazepam we have used four compounds, clonazepam and the benzodiazepine antagonists flumazenil (Ro 15-1788) 21 which bind selectively to the central type receptor, and Ro5486422 and PK 11195, isoquinoline-carboxamide compounds, 1,8,23 which bind selectively to the peripheral type receptor . In addition, since we have previously demonstrated that adenosine uptake interferes with the pharmacological evaluation of the adenosine-theophylline antagonism, 24 our second objective was to examine the effect of diazepam on this antagonism . Finally, since epithelium removal potentiates the relaxant effect of adenosine, 24 we tried to determine whether the epithelium modulates the benzodiazepine interaction with adenosine .
Protocols
METHODS
In all experiments, tracheal strips were first contracted to maximal tension with carbachol 10 -4 M and then relaxed with theophylline (3 x 10 - ' M). After 1 h rest, with washing every 15 min, 10 min pretreatments with saline, dipyridamole or benzodiazepine ligands were made . The strips were contracted with histamine (10 -5 M) and after equilibration, cumulative concentration-response curves to adenosine were obtained by increasing the concentration of adenosine at 10- 15 min intervals in logarithmic increments . When the concentration-response curves to adenosine were completed, theophylline 3 x 10 - ' M was added to the bath to determine maximal relaxation . Owing to the development of tachyphylaxis, only one series of adenosine concentrations was used for each tracheal preparation . The adenosine-induced relaxation was expressed as a percentage of the maximal effect of theophylline 3 x 10 -3 M . The -log EC50 values (defined as the negative log of the drug concentration that caused 3 50% of maximal effect of theophylline 3 x 10_ M), were derived from the log concentration-effect curves . These values were evaluated graphically from each experiment. pA 2 values were determined according to Arunlakshana and Schild .29
Tissue preparation
Measurement of potentiation
Male guinea-pigs (250-350 g) were killed by a blow to the head and exsanguinated . The trachea was removed and placed in Krebs-Henseleit solution (composition mM : NaCl 114, KCI 4 .7, CaCl 2 2.5, KH2PO4 1 .2, MgSO 4 1 .4, NaHCO 3 25 .0, glucose 11 .7) . Following removal of adhering fat and connective tissue the trachea was slit open along its longitudinal axis, directly opposite the smooth muscle, and three strips were prepared according to the zig-zag method of Emmerson and Mackay . 25 The strips were placed in parallel in 25 ml organ baths containing Krebs-Henseleit solution at 37 ° C, gassed with 95%0 2 and 5%CO 2 and tied to transducers . The tissues were equilibrated under an initial tension of 1 .80 g and washed every 15 min before the start of the experiments . After equilibration for 1 .25 h, the resting tension was between 0 .8 and 1 .4 g . Under these conditions, responses to agonists were reproducible . Tensions were measured isometrically with Celaster strain gauges and were displayed on Linseis recorders . In some experiments the epithelium was removed by gently rubbing the luminal surface (over both the smooth muscle and cartilage areas) with a cottontipped applicator . 24 ' 26-28 Comparisons between epi thelium containing and epithelium-denuded preparations were made on adjacent strips from the same area of trachea .
A method previously described was used to quantitate potentiation of peripheral tissues to adenosine and study inhibition by diazepines of adenosine uptake . 12,30 Sensitivity of trachea to adenosine was quantitated as the molar concentration of agonist producing 50% of maximal response (EC 50) in the presence of various concentrations of benzodiazepines or dipyridamole . Potentiation to the agonist, denoted as x, was defined as the EC 50 in the absence of uptake inhibitor divided by the EC 50 in the presence of uptake inhibitor . Maximal potentiation, denoted y, was defined as the potentiation produced by 10 -5 M dipyridamole . It was not possible to determine the potentiating effect of higher concentration of dipyridamole (10 -4 M) because such concentration abolished the responses to histamine . Benzodiazepines were considered to be devoid of antagonistic effect vs adenosine as previously demonstrated by Kenakin 12 for diazepam and clonazepam . Experiments conducted with Ro5-4864, PK 11195 and flumazenil -5 4 showed that 10 M and 10" M concentrations of these drugs did not exert an antagonistic effect versus 2-chloroadenosine . The estimates of potentiation to adenosine were then utilized in the following equation" log Y(x-1) = log [I]-log K . (Y - x) where y is the maximal potentiation in the same tissue
Benzodiazepine Interaction with Adenosine
from which estimates of x were made, [I] is the molar concentration of uptake inhibitor and K ; theoretically is the equilibrium dissociation constant of the inhibitor for the site of uptake (assuming competitive antagonism of adenosine removal from the receptor compartment) . Regressions of log (y(x-1)/(y- x)) on log [I] yielded estimates of the pK ; (-log K ;) as the intercepts as well as the slopes of the regression lines and the pK ; .
1 39
Influence of benzodiazepines and dipyridamole on histamine induced contraction Phasic responses to histamine 10 -5 M were unaffected by pretreatment with dipyridamole or benzodiazepines except for diazepam which at a concentration of 10 -4 M caused a weak but significant reduction of the phasic contractile response to histamine 10 - 5 M (39 .8 ± 4 .1 % of ACh maximal effect versus control 61 .9 ± 6.6%, p < 0 .05) .
Statistical analysis of results Statistical analysis of the results was performed using analysis of variance and then Student's t-test for paired or unpaired data . All values in the text and table are expressed as mean±s .e .mean . Drugs The drugs used were : adenosine (Merck, D-Darmstadt), histamine (Sigma, St Louis, USA), dipyridamole (Boehringer-Ingelheim, F-Reims), theophylline sodium anisate (Bruneau, F-Paris), diazepam, Ro5-4864 (7-chlor-5-(4-chlorphenyl)-1,3-dihydro-l-methyl-2H1,4-benzodiazepin-2-on), flumazenil (ethyl 8-fluoro5,6-dihydro-5-methyl-6-oxo-4H-imidazo[1,5-a][1,4]benzodiazepine-3-carboxylate) (Roche, CH-Basle) and PK 11195 (N-methyl-N-(methyl-l-propyl)(chloro-2 phenyl)-1 isoquinoleine-3 carboxamide) (Rhone-Poulenc, F-Vitry sur Seine) . Theophylline was used as proprietary injectable solution (Theophylline Bruneau) ; dipyridamole was dissolved daily in ethanol and the solution was further diluted with Krebs solution. Diazepines were dissolved in sulfolane (tetramethylene sulfone, Aldrich Chemie, FRG-Steinheim) . The appropriate concentration of drugs was added to the organ baths so that at no time did the volume of sulfolane or ethanol exceed 0 .1 ml (0,4% by volume) . The solvents in such maximal concentrations produced no direct effect on basal tone or on responses to adenosine .
Influence of dipyridamole and benzodiazepines on adenosine concentrations-response curves In the presence of dipyridamole 10' M to 10 -5 M, concentration-effect curves to adenosine were shifted to the left, with a maximal effect at 10 -5 M corresponding to a shift of 1 .89 ± 0 .12 log unit, that is a 79.4 fold potentiation (Fig . 1) . The apparent pK ; of dipyridamole as inhibitor of adenosine uptake calculated according to Kenakini 2 was 7 .77±0 .12 (Fig . 2, Table 1 In the presence of benzodiazepines, concentrationresponse curves of adenosine were also shifted to the left (Fig . 1) with potentiation of 47 .8, 25 .1, 4 .46, 2 .51 and 2 .37 fold respectively for diazepam (10 -4 M), Ro5-4864 (10 -4 M), clonazepam (10 -4 M), flumazenil (10 -4 M) and PK 11195 (10 -4 M) . However, the shifts of the dose responses curves in the presence of
RESULTS Influence of benzodiazepines and dipyridamole on basal tone Dipyridamole, at concentration of 10 -5 M, caused a significant relaxation to 88+5 .5% (p
