Pulmonary Pharmacology (1992) 5, 23-29 PULMONARY PHARMACOLOGY
Hypoxia Modulates Mediator Responses and Neurotransmission in Guinea-pig Trachea In Vitro H. Fujiwara*, N . Kurihara, K . Hirata, K. Ohta, S . Fujimoto, T. Takeda First Department of Internal Medicine, Osaka City University Medical School, 1-5-7 Asahi-machi, Abeno-ku, Osaka 545, Japan
SUMMARY : To discover whether hypoxia affects mediator responses and neurotransmission in tracheal smooth muscle, we studied in vitro tracheal segments from guinea-pigs under isometric conditions . Hypoxia itself did not alter the basal tone . The maximum response to acetylcholine and histamine under hypoxia was less than that under oxygenated conditions. The logarithm of 50% effective concentration (log EC,) of the response to acetylcholine under hypoxia was not altered, but the log EC so of the response to histamine decreased significantly . In contrast to the response to exogenous acetylcholine, the maximum contractile response to electrical field stimulation (EFS) under hypoxia was not different from that under oxygenated conditions, but the logarithm of 50% effective frequency of contractions caused by EFS under hypoxia decreased significantly . On the other hand, non-adrenergic-noncholinergic relaxation caused by EFS was unaffected by hypoxia . These observations suggest that hypoxia can modulate the responses of tracheal smooth muscle to acetylcholine, histamine and nerve stimulation .
INTRODUCTION
MATERIALS AND METHODS
Hypoxia increases airway resistance and decreases specific airway conductance in humans and dogs .' -' In addition, hypoxia enhances bronchial reactivity in sheep and dogs ." Breathing 30% oxygen decreases airway resistance and increases the maximum expiratory flow rate in patients with chronic obstructive lung disease and attenuates bronchial responsiveness to methacholine in asthmatics . 8-10 It is therefore possible that oxygen may play a role in regulation of bronchomotor tone . However, little is known about how oxygen tension affects the airway smooth muscle response in vitro ; conversely its effect on vascular smooth muscle has been much studied Thus, the primary purpose of this study was to investigate whether hypoxia changes the airway smooth muscle response to acetylcholine and histamine by evaluating isolated strips of guinea-pig trachea . To examine the changes in neurotransmission, we also investigated the effect of hypoxia on cholinergic contraction and non-adrenergic-non-cholinergic relaxation of guineapig trachea by using electrical field stimulation (EFS) .
Tissue preparation Guinea-pigs (250-500 g) were anaesthetized with pentobarbital sodium and the cervical portion of the trachea was rapidly removed and immersed in cold Krebs-Henseleit solution containing (in mm) : NaCl 118 ; KCl 4 .7 ; CaCl 2 1 .2 ; MgSO4 1 .2 ; KH 2 P04 1 .2 ; NaHCO 3 25 ; and glucose 11 .1 . Paired strips, each made up of three adjacent cartilage rings, were suspended vertically in l0-ml organ baths filled with Krebs-Henseleit solution at 37° C and bubbled with a mixture of 95% 02/5% CO 2 . All strips were initially equilibrated for at least 1 h while being washed every 15 min under a resting tension of 0 .5g ; this tension allowed the maximum response for tissues of this size (data not shown) . Throughout the experiments the cyclooxygenase inhibitor indomethacin (10 - 'm) was present in the bath to prevent the development of spontaneous resting tone .
Measurement of response and electrical field stimulation (EFS) The mechanical responses of the tissues were measured by isometric transducers (TB-612T ; Nihon Koden, Tokyo, Japan) and were recorded on a multichannel polygraph system (Nihon Koden) . Platinum electrodes were placed on both sides of the strips for
* For correspondence. 0952-0600/92/010023+07 $03 .00/0
23
J 1992 Academic Press Limited
24
H . Fujiwara et al
EFS produced by an electric stimulator (SEN 3301, Nihon Koden) . Biphasic square wave impulses of supramaximal voltage 50 V and a duration of 0 .3 ms were applied for 10 s every 3 min at frequencies of between 1 and 50 Hz . To assess the nature of the contractile and relaxant responses to EFS, separate preparations were incubated with tetrodotoxin (10`m), hexamethonium (10 -5 M) or atropine (10`m), and were electrically stimulated . Condition of the organ bath Hypoxia was produced by saturation of the buffer solution with a mixture of 7% 0 2 /5% CO2/88% N2 . Oxygenated (control) conditions were produced by saturation of the buffer with a mixture of 95% 02/ 5%CO 2 as described by Paterson et al) 2 The P0 2 , PCO2, and pH of the buffer solution in the baths were measured with a blood gas analyser (ABL-4, Radiometer, Copenhagen, Denmark) . With a 10-min exposure to the oxygenated gas mixture, the P0 2 was kept at > 55 kPa ; with a 10-min exposure to the hypoxic gas mixture, the P0 2 was kept at < 10 kPa . However, there were no differences in pH and PCO, in the two conditions (data not shown) . Changes with time in responses during hypoxia and reoxygenation To investigate the changes with time in the contractile responses during hypoxia for I h and the reoxygenation that followed, some strips were caused to contract by EFS (5 Hz) applied every 3 min, or the addition of acetylcholine (2 x 10 - 'm) or histamine (2 x 10 -6 M) every 10 min with washing to remove the agonist . New stable responses were reached after about 10 min of hypoxia, and no irreversible tissue damage occurred during I h of hypoxia . Therefore, in all the later experiments under hypoxic conditions observations were started after 10 min of hypoxia, and the total time of the hypoxia was less than 40 min . Concentration-response curves for acetylcholine and histamine One of the paired strips was tested under hypoxic conditions, and the other served as a control under oxygenated conditions . At the start of the experiment, the maximum contraction in response to each of the agonists (acetylcholine at 10 - 'm or histamine at 10 - 'm) was measured under initial oxygenated conditions for both strips to correct for the variation of the different tissues . After washing the strips several times, test strips were equilibrated with the hypoxic gas mixture for 10 min ; the control strips were equilibrated throughout the experiment with the oxygenated gas mixture . The cumulative concentration-
response curves for acetylcholine (10 - ' to approximately 10 M) or histamine (10 ' to approximately -4 10 M) were compared under these two conditions .
Frequency-response curves for EFS To investigate the effect of hypoxia on neurally mediated contractions, standard maximum contractions elicited by EFS at 50 Hz were obtained for both paired strips under the initial oxygenated conditions . Then the frequency-response curves to EFS (1-50 Hz) were obtained after 10 min of hypoxia for test strips, and compared with those of control strips under the oxygenated conditions . In some strips, we examined hypoxia-induced effects on the contractile response to EFS at 5 Hz in the presence of phentolamine 10 -6 M (a-receptor antagonist), dl-propranolol 10 - 'm (5receptor antagonist), pyrilamine 10 - ' m (H,-receptor antagonist), ketanserine 10 - ' m (5-HT,-receptor antagonist), theophylline 2 X 10 -5 M (adenosine receptor antagonist), and [D-Pro 4,D-Trp' •9]substance P(4-11) 10 -5 M (substance P receptor antagonist) ." To investigate the effects of hypoxia on non-adrenergic-non-cholinergic relaxations, the strips were caused to contract first with a concentration of histamine that caused about 50% of their maximum contraction under hypoxic or oxygenated conditions in the presence of dl-propranolol (10 - 'm) plus atropine (10 -6 M) . Then the frequency-response curves of nonadrenergic-non-cholinergic relaxation elicited by EFS (1-501-1z) were compared in the two different conditions .
Drugs The following agents were used : acetylcholine chloride, atropine sulphate, histamine hydrochloride, dlpropranolol, aminophylline (salt of theophylline), indomethacin, hexamethonium bromide (Sigma, St Louis, MO) ; phentolamine hydrochloride (CibaGeigy, Takarazuka, Japan) ; ketanserin (Janssen Pharmaceutica, Beerse, Belgium) ; and [D-Pro 4 ,DTrp'• 9 ]substance P(4-11) (Peptide Institute, Osaka, Japan) . The molar concentrations of drugs described in this paper refer to the bath concentrations .
Data analysis The contractile response to each agonist or EFS was expressed as a percentage of the standard response during initial oxygenated conditions . The relaxant response was expressed as a percentage of relaxation after contraction caused by histamine . Data are reported as means ± SE, and n indicates the number of animals in the different experiments . The logarithm of
Effect of Hypoxia on Guinea-pig Trachea 25
the concentration of agonists or the frequency of EFS that elicited 50% of the maximum response was designated as the log EC, or log EF SO , respectively . Statistical analysis was done by Student's paired and unpaired t-tests. A P value of less than 0 .05 was considered to be significant .
RESULTS Response to EFS EFS evoked a frequency-dependent, rapid contraction at resting tension . When the tone was increased with histamine in the presence of dl-propranolol (10 - 'm) plus atropine (10 -1 m), EFS evoked a frequency-dependent relaxation . Atropine or tetrodotoxin (10 - 'm) abolished contractile responses to EFS at 1-50 Hz, and the ganglionic blocker hexamethonium (10 - 'm) had no significant effects . Tetrodotoxin (10 -6 M) significantly reduced the relaxant response to EFS (data not shown) .
Effects of hypoxia and reoxygenation on basal tone Hypoxia and reoxygenation per se did not cause contraction or relaxation of the tracheal strips under resting conditions .
A
Changes with time in responses during hypoxia and reoxygenation As shown in Figures 1 and 2, after 10 min of hypoxia, contractile responses to acetylcholine (2 x 10 -6 M) were lower than in the oxygenated condition (78 .4 ± 3 .1 % of the response during initial oxygenation ; P
Hypoxia Modulates Mediator Responses and Neurotransmission in Guinea-pig Trachea In Vitro H. Fujiwara*, N . Kurihara, K . Hirata, K. Ohta, S . Fujimoto, T. Takeda First Department of Internal Medicine, Osaka City University Medical School, 1-5-7 Asahi-machi, Abeno-ku, Osaka 545, Japan
SUMMARY : To discover whether hypoxia affects mediator responses and neurotransmission in tracheal smooth muscle, we studied in vitro tracheal segments from guinea-pigs under isometric conditions . Hypoxia itself did not alter the basal tone . The maximum response to acetylcholine and histamine under hypoxia was less than that under oxygenated conditions. The logarithm of 50% effective concentration (log EC,) of the response to acetylcholine under hypoxia was not altered, but the log EC so of the response to histamine decreased significantly . In contrast to the response to exogenous acetylcholine, the maximum contractile response to electrical field stimulation (EFS) under hypoxia was not different from that under oxygenated conditions, but the logarithm of 50% effective frequency of contractions caused by EFS under hypoxia decreased significantly . On the other hand, non-adrenergic-noncholinergic relaxation caused by EFS was unaffected by hypoxia . These observations suggest that hypoxia can modulate the responses of tracheal smooth muscle to acetylcholine, histamine and nerve stimulation .
INTRODUCTION
MATERIALS AND METHODS
Hypoxia increases airway resistance and decreases specific airway conductance in humans and dogs .' -' In addition, hypoxia enhances bronchial reactivity in sheep and dogs ." Breathing 30% oxygen decreases airway resistance and increases the maximum expiratory flow rate in patients with chronic obstructive lung disease and attenuates bronchial responsiveness to methacholine in asthmatics . 8-10 It is therefore possible that oxygen may play a role in regulation of bronchomotor tone . However, little is known about how oxygen tension affects the airway smooth muscle response in vitro ; conversely its effect on vascular smooth muscle has been much studied Thus, the primary purpose of this study was to investigate whether hypoxia changes the airway smooth muscle response to acetylcholine and histamine by evaluating isolated strips of guinea-pig trachea . To examine the changes in neurotransmission, we also investigated the effect of hypoxia on cholinergic contraction and non-adrenergic-non-cholinergic relaxation of guineapig trachea by using electrical field stimulation (EFS) .
Tissue preparation Guinea-pigs (250-500 g) were anaesthetized with pentobarbital sodium and the cervical portion of the trachea was rapidly removed and immersed in cold Krebs-Henseleit solution containing (in mm) : NaCl 118 ; KCl 4 .7 ; CaCl 2 1 .2 ; MgSO4 1 .2 ; KH 2 P04 1 .2 ; NaHCO 3 25 ; and glucose 11 .1 . Paired strips, each made up of three adjacent cartilage rings, were suspended vertically in l0-ml organ baths filled with Krebs-Henseleit solution at 37° C and bubbled with a mixture of 95% 02/5% CO 2 . All strips were initially equilibrated for at least 1 h while being washed every 15 min under a resting tension of 0 .5g ; this tension allowed the maximum response for tissues of this size (data not shown) . Throughout the experiments the cyclooxygenase inhibitor indomethacin (10 - 'm) was present in the bath to prevent the development of spontaneous resting tone .
Measurement of response and electrical field stimulation (EFS) The mechanical responses of the tissues were measured by isometric transducers (TB-612T ; Nihon Koden, Tokyo, Japan) and were recorded on a multichannel polygraph system (Nihon Koden) . Platinum electrodes were placed on both sides of the strips for
* For correspondence. 0952-0600/92/010023+07 $03 .00/0
23
J 1992 Academic Press Limited
24
H . Fujiwara et al
EFS produced by an electric stimulator (SEN 3301, Nihon Koden) . Biphasic square wave impulses of supramaximal voltage 50 V and a duration of 0 .3 ms were applied for 10 s every 3 min at frequencies of between 1 and 50 Hz . To assess the nature of the contractile and relaxant responses to EFS, separate preparations were incubated with tetrodotoxin (10`m), hexamethonium (10 -5 M) or atropine (10`m), and were electrically stimulated . Condition of the organ bath Hypoxia was produced by saturation of the buffer solution with a mixture of 7% 0 2 /5% CO2/88% N2 . Oxygenated (control) conditions were produced by saturation of the buffer with a mixture of 95% 02/ 5%CO 2 as described by Paterson et al) 2 The P0 2 , PCO2, and pH of the buffer solution in the baths were measured with a blood gas analyser (ABL-4, Radiometer, Copenhagen, Denmark) . With a 10-min exposure to the oxygenated gas mixture, the P0 2 was kept at > 55 kPa ; with a 10-min exposure to the hypoxic gas mixture, the P0 2 was kept at < 10 kPa . However, there were no differences in pH and PCO, in the two conditions (data not shown) . Changes with time in responses during hypoxia and reoxygenation To investigate the changes with time in the contractile responses during hypoxia for I h and the reoxygenation that followed, some strips were caused to contract by EFS (5 Hz) applied every 3 min, or the addition of acetylcholine (2 x 10 - 'm) or histamine (2 x 10 -6 M) every 10 min with washing to remove the agonist . New stable responses were reached after about 10 min of hypoxia, and no irreversible tissue damage occurred during I h of hypoxia . Therefore, in all the later experiments under hypoxic conditions observations were started after 10 min of hypoxia, and the total time of the hypoxia was less than 40 min . Concentration-response curves for acetylcholine and histamine One of the paired strips was tested under hypoxic conditions, and the other served as a control under oxygenated conditions . At the start of the experiment, the maximum contraction in response to each of the agonists (acetylcholine at 10 - 'm or histamine at 10 - 'm) was measured under initial oxygenated conditions for both strips to correct for the variation of the different tissues . After washing the strips several times, test strips were equilibrated with the hypoxic gas mixture for 10 min ; the control strips were equilibrated throughout the experiment with the oxygenated gas mixture . The cumulative concentration-
response curves for acetylcholine (10 - ' to approximately 10 M) or histamine (10 ' to approximately -4 10 M) were compared under these two conditions .
Frequency-response curves for EFS To investigate the effect of hypoxia on neurally mediated contractions, standard maximum contractions elicited by EFS at 50 Hz were obtained for both paired strips under the initial oxygenated conditions . Then the frequency-response curves to EFS (1-50 Hz) were obtained after 10 min of hypoxia for test strips, and compared with those of control strips under the oxygenated conditions . In some strips, we examined hypoxia-induced effects on the contractile response to EFS at 5 Hz in the presence of phentolamine 10 -6 M (a-receptor antagonist), dl-propranolol 10 - 'm (5receptor antagonist), pyrilamine 10 - ' m (H,-receptor antagonist), ketanserine 10 - ' m (5-HT,-receptor antagonist), theophylline 2 X 10 -5 M (adenosine receptor antagonist), and [D-Pro 4,D-Trp' •9]substance P(4-11) 10 -5 M (substance P receptor antagonist) ." To investigate the effects of hypoxia on non-adrenergic-non-cholinergic relaxations, the strips were caused to contract first with a concentration of histamine that caused about 50% of their maximum contraction under hypoxic or oxygenated conditions in the presence of dl-propranolol (10 - 'm) plus atropine (10 -6 M) . Then the frequency-response curves of nonadrenergic-non-cholinergic relaxation elicited by EFS (1-501-1z) were compared in the two different conditions .
Drugs The following agents were used : acetylcholine chloride, atropine sulphate, histamine hydrochloride, dlpropranolol, aminophylline (salt of theophylline), indomethacin, hexamethonium bromide (Sigma, St Louis, MO) ; phentolamine hydrochloride (CibaGeigy, Takarazuka, Japan) ; ketanserin (Janssen Pharmaceutica, Beerse, Belgium) ; and [D-Pro 4 ,DTrp'• 9 ]substance P(4-11) (Peptide Institute, Osaka, Japan) . The molar concentrations of drugs described in this paper refer to the bath concentrations .
Data analysis The contractile response to each agonist or EFS was expressed as a percentage of the standard response during initial oxygenated conditions . The relaxant response was expressed as a percentage of relaxation after contraction caused by histamine . Data are reported as means ± SE, and n indicates the number of animals in the different experiments . The logarithm of
Effect of Hypoxia on Guinea-pig Trachea 25
the concentration of agonists or the frequency of EFS that elicited 50% of the maximum response was designated as the log EC, or log EF SO , respectively . Statistical analysis was done by Student's paired and unpaired t-tests. A P value of less than 0 .05 was considered to be significant .
RESULTS Response to EFS EFS evoked a frequency-dependent, rapid contraction at resting tension . When the tone was increased with histamine in the presence of dl-propranolol (10 - 'm) plus atropine (10 -1 m), EFS evoked a frequency-dependent relaxation . Atropine or tetrodotoxin (10 - 'm) abolished contractile responses to EFS at 1-50 Hz, and the ganglionic blocker hexamethonium (10 - 'm) had no significant effects . Tetrodotoxin (10 -6 M) significantly reduced the relaxant response to EFS (data not shown) .
Effects of hypoxia and reoxygenation on basal tone Hypoxia and reoxygenation per se did not cause contraction or relaxation of the tracheal strips under resting conditions .
A
Changes with time in responses during hypoxia and reoxygenation As shown in Figures 1 and 2, after 10 min of hypoxia, contractile responses to acetylcholine (2 x 10 -6 M) were lower than in the oxygenated condition (78 .4 ± 3 .1 % of the response during initial oxygenation ; P
