Dysfunction of Nonadrenergic Noncholinergic Inhibitory System after Antigen Inhalation in Actively Sensitized Cat Airways1 , 2
M. MIURA, M. ICHINOSE, K. KIMURA, U. KATSUMATA, T. TAKAHASHI, H. INOUE, and T. TAKISHIMA
Introduction
The nonadrenergic noncholinergic inhibitory nervous system (NANCIS) is the only inhibitory nervous system distributed in human airways (1), and the functional existence of NANCIS in normal humans has been demonstrated (2, 3). Therefore, dysfunction of this nervous system may possibly cause airway hyperreactivity, which is the most common feature of bronchial asthma. Actually, vasoactive intestinal peptide (VIP), a putative neurotransmitter of NANCIS (4,5), has recently been reported to be absent in asthmatic patients (6). On the other hand, it is also speculated that NANC inhibitory nerve activity can be impaired during the inflammatory process because possible peptide neurotransmitters of NANCIS such as VIP may be degenerated by proteases released from inflammatory cells (7). It has been reported that a mast-cell-derivedprotease, tryptase, can degenerate VIP (8) and reverse its dilatatory effects (9). In in vitro studies, there are some reports describing the effects ofpeptidases on NANCIS (10,11). However, it has never been determined whether the dilatatory effects of NANCIS can actually be affected by endogenous proteases released during airway inflammation in vivo. Thus, in the present study, we examined this hypothesis by using immediate allergic reaction (IAR) as a tool to induce airway inflammation on actively sensitized cat airways in vivo. Wechose IAR as an inducer of airway inflammation because proteases released from mast cells by IgE-mediated mechanism may possibly cause dysfunction of NANCIS (12). Moreover, we compared the effects of IAR on inhibitory activity of NANCIS with that on adrenergic inhibitory activities, which has been reported to have some parallels in their effective sites and potencies but have different types of neurotransmitters in cat airways (13). Next, 70
SUMMARY Wehave investigated whether proteases released during antigen inhalation cause dysfunction of the nonadrenergic noncholinerglc inhibitory nervous system (NANCIS). Frequencyresponse (F-R)studies of NANCIS were performed before and afterAscarls antigen (ASC) Inhalation using actively sensitized cats. NANC dllatatory effecta were obtained by stimulating bilateral cervical vegl under cholinergic and p-adrenerglc blockade and serotonln·lnduced bronchoconstrlctlon, and assessed by maximal percent relaxation (rmex) and the frequency causing 50% of maximal relaxation (EF••). ASC Inhalation caused a transient Increase In pUlmonary resistance In all animals. One hour after ASC Inhalation, pulmonary resistance returned to the baseline velue, but ASCInhalation significantly attenuated NANC Inhibitory activities: rmexdecreased from 82.2 ± 4.7 (mean ± SE) to 64.3 ± 11.2% (p < 0.05), and the geometric mean of EF•• increased from 1.7 to 4.3 Hz (p < 0.05). Dllatatory effects of Infused VIP,a possible neurotransmitter of NANCIS, wes also attenuated after ASC Inhalation. Pretreatment with leupeptin (3 mg/kg) abolished ASC-Induced Impairment of NANC Inhibitory activities. By contrast, dllatatory effects of adrenergic nerve stimulation were not affected by ASC Inhalation. These results suggest that NANC Inhibitory activities can be Impaired after ASC Inhalation, and that this Impairment of NANCIS may be due to effects of proteases released during allergic reaction. AM REV RESPIR DIS 1992: 145:70-74
we examined whether the dilatatory effects of a possible neurotransmitter, VIP, are also impaired after IAR or not. Finally, we examined whether a protease inhibitor can suppress the dysfunction of NANCIS induced by IAR in order to elucidate the role of proteases on impairment of NANCIS after antigen inhalation. Methods Experimental Animals We used 20 adult cats weighing 2.0 to 3.0 kg; they were separated into four groups for different study protocols. All animals were activelysensitized by injectingintraperitoneally a mixture of 1 ml of Ascaris suum antigen (1:100 dilution; Greer Laboratories, Lenoir, NC) and 150 mg of aluminum hydrochloride dissolved in 3 ml of physiologic saline 3 wk and 1 wk before antigen inhalation. Skin tests under anesthesia with ketamine (25 mg/kg given intramuscularly) were performed in all animals on the experimental day to determine whether sensitization had been effective. The procedure was described in detail in our previous report (14)in which wedemonstrated that there is a significant correlation between immediate intradermal reaction and pulmonary responses induced by the antigen. Therefore, we confirmed that all experimental animals
were expected to show IAR to Ascaris antigen by the skin test preceding the antigen inhalation. Briefly, 0.5070 Evans-blue was intravenously injected into each animal, and 0.02 ml of Ascaris antigen (1:100 dilution) was administered intradermally 15min later. Evaluation of the test was performed 30 min after antigen injection.
Physiologic Measurements After additional anesthesia with ketamine (25 mg/kg given intramuscularly), a femoral artery and bilateral femoral veins were cannulated with polyethylene tubes. Venous catheters were used to inject drugs, and an arterial catheter was used to monitor systemic blood pressure and to collect blood samples for gas analysis. An endotracheal tube (outer diameter, 6 mm; length,S em) was then inserted via the tracheostoma, and the chest (Receivedin originalform November 26, 1990and in revised form August 7, 1991) I From the First Department of Internal Medicine, Tohoku University School of Medicine, Sendai, Japan. 2 Correspondence and requests for reprints should be addressed to Tarnotsu Takishima, M.D., Professor and Chairman, First Department of Internal Medicine, 'Iohoku UniversitySchool of Medicine, 1-1 Seiryo-machi, Aoba-ku, Sendai 980, Japan.
71
ANTIGEN·INDUCED ATTENUATION OF NONADRENERGIC BRONCHODILATION
was opened. For further anesthesia and muscle paralysis, pentobarbital (20 mg/kg) and succinylcholine (25mg/kg) were intravenously injected. The lungs were then artificially ventilated with a constant-volume respirator (Model 661; Harvard Apparatus Co., South Natick, MA) at a tidal volume of 15 to 20 ml/kg and a frequency of 15breaths/min. Another catheter (inner diameter, 1 mm; length, 7 ern) was inserted into the distal end of the tracheal tube to obtain airway pressure directly and connected to a differential pressure transducer (MP 45, ± 50 em H 2 0 ; Validyne Co., Northridge, CA) to monitor transpulmonary pressure. Airflow at the opening of the endotracheal tube was measured with a pneumotachograph (FleischNo. (0) and a differential pressure transducer (MP 45, ± 5 ern H 2 0 ; Validyne). Pulmonary resistance (RL)was obtained by the method of Mead and Whittenberger (15). Arterial P0 2 , Pco., and pH were determined intermittently with a pH/blood gas analyzer (Model 1302; Instrumentation Laboratories, Lexington, MA). The animals' pH and blood gas tension were maintained within their normal range by intermittent administration of sodium bicarbonate and adjustment of the tidal volume of the respirator.
Animal Preparation for Nerve Stimulation For nerve stimulation, the bilateral vagosympathetic trunk was exposed and separated into two component bundles. The identities of these two components were then verified by testing for a vagal effect on heart rate and for a sympathetic effect on pupil size and systemic blood pressure (11). Liquid paraffin was applied to keep the nerves from drying. Separated bilateral cervical vagi or sympathetic nerves wereplaced on two sets of bipolar silver electrodes only during stimulation. An isolated stimulator (SEN 3201;Nihon-Koden, Tokyo, Japan) was employed for electrical stimulation with different stimulating conditions.
Effect of IAR on NANCIS In cats of Group I (n = 5), NANC inhibitory activity was examined before and after Ascaris antigen inhalation. In order to evaluate and compare NANC inhibitory activity under different conditions in the same animal, the frequency-response study was repeated. First, atropine (3 mg/kg) and propranolol (2 mg/ kg) were intravenously administered to animals to avoid cholinergic and p-adrenergic influences. After measuring baseling RL, serotonin (25 to 100ug/kg/min) was continuously infused to achieve steady-state bronchoconstriction. Bilateral cervical vagi were then electrically stimulated for 20 s with pulses of 12 V and I ms duration. After obtaining the peak dilatatory response, serotonin infusion was stopped. This procedure was repeated every 15min with stimulating frequency varied from 1,3, and 10 to 30 Hz to construct frequencyresponse curves. Thirty minutes after constructing the first frequency-response curve, Ascaris antigen (1:100 dilution) was delivered
to animals for 3 min with an ultrasonic nebulizer (Omron NE-UllB; Tateishi Co., Tokyo, Japan) (output, 0.6 ml/min) interposed between the inspiratory port of the respirator and the cannula to the lung. A frequencyresponse study after antigen inhalation was performed immediately after increased RL by antigen inhalation had returned to its baseline value (approximately 1 h after antigen inhalation). To evaluate NANC inhibitory effects quantitatively, maximal relaxation (r max) and EF 50 were obtained from each frequency-response curve; r max was defined as percent recovery of RL at 30 Hz stimulation, and EF 5 0 was defined as the frequency causing 50070 relaxation of r max • We preliminarily confirmed reproducibility of dilatatory effects elicited by electrical NANCIS stimulation by using cats not showing intradermal hyperreactivity to Ascaris antigen (n = 3). Antigen inhalation did not cause an increase in RL, and frequency-response curves showed good reproducibility in all of these animals.
Effects of IAR on VIP-induced Bronchodilation In cats of Group 2 (n = 5), dilatatoryeffects of infused VIP were examined before and after antigen inhalation. Dilatatory effects of VIP were assessed by constructing doseresponse curves. Incremental doses of VIP from 0.1,0.3, I, and 3 to 10 ug/kg were intravenously administered under cholinergic and p-adrenergic blockade and steady-state bronchoconstriction at 15-min intervals. Ascaris antigen inhalation was performed in the same way as with the other groups 30 min after the first dose-response study. A second doseresponse curve was obtained after RL, increased by antigen inhalation, had returned to its baseline value.
Effect of Protease Inhibitor on IAR-induced Dysfunction of NANCIS In cats of Group 3 (n = 5), NANC inhibitory activity was examined before and after Ascaris antigen inhalation in the same way as in those of Group 1. In this group, however, a serine protease inhibitor, leupeptin (3 mg/ kg) was intravenously administered 15 min after the antigen inhalation. Infusion of leupeptin did not influence systemic blood pressure or RL in any animal studied.
Effect of IAR on Adrenergic Nerves In cats of Group 4 (n = 5), adrenergic inhibitory activity was examined before and after the antigen inhalation by constructing frequency-response curves.The study protocol for this group was almost the same as that for Group I except for a pretreatment with propranolol. In this group, bilateral cervical sympathetic nerves instead of vagi were stimulated with the same stimulating conditions and intervals as in NAN CIS stimulation. Quantitative assessment of adrenergic activ-
ity was also made with Jmax and EF 5 0 obtained from each frequency-response curve.
Assessment of Dilatatory Responses We assessed the dilatatory effects of nerve stimulation and VIP infusion as follows. A raw constrictive response was initially obtained as the difference between RL during steady-state bronchoconstriction and baseline RL. An inhibitory response was then obtained as the difference between RL during steadystate bronchoconstriction and recovered RL caused by nerve stimulation or VIP infusion. We thus defined percent relaxation as the ratio of these differences, which indicates the degree of inhibition normalized by the raw response.
Statistical Analysis The data were expressed as means ± SE and analyzed using Student's paired t test for parametric data and Wilcoxon's signed-ranks test for nonparametric data (16). The level of statistical significance for mean values was accepted at p < 0.05.
Results
Antigen inhalation caused an increase in RLand a decrease in systemic blood pressure in all animals in Groups 1to 4. Mean maximal increases in RLafter antigen inhalation in Groups 1 to 4 were 98.9 ± 52.5, 147 ± 61.6, 85.6 ± 25.4, and 125 ± 31.5070, respectively, and there was no significant difference between these values. This immediate airway response induced by antigen inhalation depressed NANC inhibitory activity (Group 1). Individual frequency-response curves before and after antigen inhalation obtained from animals of Group 1 are shown in figure 1. In all animals of this group, frequency-response curves after antigen inhalation shifted rightward, and rmax was decreased compared with that before antigen inhalation. Mean values of r max and EF5 0 were significantly changed from 82.2 ± 4.7% and 1.7 Hz (geometric mean for EF5 0 ) to 64.3 ± 11.2% and 4.3 Hz (p < 0.05), respectively, with antigen inhalation (figure 2a). Dilatatory effects of infused VIP were weakened after IAR, as demonstrated in animals of Group 2. Mean dose-response curves before and after antigen inhalation in this group are shown in figure 3. Percent relaxation at 1, 3, and 10 ug/kg weresignificantly decreased from 40.2 ± 13.9,76.6 ± 6.2, and 87.5 ± 5.2% to 13.4 ± 11.0(p < 0.01), 49.9 ± 14.5(p < 0.05), and 63.6 ± 13.3% (p < 0.05), respectively. Pretreatment with the protease inhibitor leupeptin significantly depressed IAR-induced impairment of NANC dilatatory activity. Individual frequency-
72
MIURA, ICHINOSE, KIMURA, KATSUMATA, TAKAHASHI, INOUE, AND TAKISHIMA
100
50
,... #. c
0
.Q
a; )(
/.-. 0/°/
1-1
0-0
100
1-3
~.-.
;//' 1/ •
• ./
3 100
III
1-4
10
0
..'" "i II:
30
50
3
10
30
3
10
30
Frequency (Hz)
response curves for NANCIS before and after antigen inhalation in leupeptintreated animals are shown in figure 4. Mean values of r max and EFso were 85.2 ± 5.10,70 and 1.7Hz before antigen inhalation and 85.7 ± 2.7% and 1.8Hz after antigen inhalation, respectively. There was no significant difference between these values (figure 2b).
100
P
M. MIURA, M. ICHINOSE, K. KIMURA, U. KATSUMATA, T. TAKAHASHI, H. INOUE, and T. TAKISHIMA
Introduction
The nonadrenergic noncholinergic inhibitory nervous system (NANCIS) is the only inhibitory nervous system distributed in human airways (1), and the functional existence of NANCIS in normal humans has been demonstrated (2, 3). Therefore, dysfunction of this nervous system may possibly cause airway hyperreactivity, which is the most common feature of bronchial asthma. Actually, vasoactive intestinal peptide (VIP), a putative neurotransmitter of NANCIS (4,5), has recently been reported to be absent in asthmatic patients (6). On the other hand, it is also speculated that NANC inhibitory nerve activity can be impaired during the inflammatory process because possible peptide neurotransmitters of NANCIS such as VIP may be degenerated by proteases released from inflammatory cells (7). It has been reported that a mast-cell-derivedprotease, tryptase, can degenerate VIP (8) and reverse its dilatatory effects (9). In in vitro studies, there are some reports describing the effects ofpeptidases on NANCIS (10,11). However, it has never been determined whether the dilatatory effects of NANCIS can actually be affected by endogenous proteases released during airway inflammation in vivo. Thus, in the present study, we examined this hypothesis by using immediate allergic reaction (IAR) as a tool to induce airway inflammation on actively sensitized cat airways in vivo. Wechose IAR as an inducer of airway inflammation because proteases released from mast cells by IgE-mediated mechanism may possibly cause dysfunction of NANCIS (12). Moreover, we compared the effects of IAR on inhibitory activity of NANCIS with that on adrenergic inhibitory activities, which has been reported to have some parallels in their effective sites and potencies but have different types of neurotransmitters in cat airways (13). Next, 70
SUMMARY Wehave investigated whether proteases released during antigen inhalation cause dysfunction of the nonadrenergic noncholinerglc inhibitory nervous system (NANCIS). Frequencyresponse (F-R)studies of NANCIS were performed before and afterAscarls antigen (ASC) Inhalation using actively sensitized cats. NANC dllatatory effecta were obtained by stimulating bilateral cervical vegl under cholinergic and p-adrenerglc blockade and serotonln·lnduced bronchoconstrlctlon, and assessed by maximal percent relaxation (rmex) and the frequency causing 50% of maximal relaxation (EF••). ASC Inhalation caused a transient Increase In pUlmonary resistance In all animals. One hour after ASC Inhalation, pulmonary resistance returned to the baseline velue, but ASCInhalation significantly attenuated NANC Inhibitory activities: rmexdecreased from 82.2 ± 4.7 (mean ± SE) to 64.3 ± 11.2% (p < 0.05), and the geometric mean of EF•• increased from 1.7 to 4.3 Hz (p < 0.05). Dllatatory effects of Infused VIP,a possible neurotransmitter of NANCIS, wes also attenuated after ASC Inhalation. Pretreatment with leupeptin (3 mg/kg) abolished ASC-Induced Impairment of NANC Inhibitory activities. By contrast, dllatatory effects of adrenergic nerve stimulation were not affected by ASC Inhalation. These results suggest that NANC Inhibitory activities can be Impaired after ASC Inhalation, and that this Impairment of NANCIS may be due to effects of proteases released during allergic reaction. AM REV RESPIR DIS 1992: 145:70-74
we examined whether the dilatatory effects of a possible neurotransmitter, VIP, are also impaired after IAR or not. Finally, we examined whether a protease inhibitor can suppress the dysfunction of NANCIS induced by IAR in order to elucidate the role of proteases on impairment of NANCIS after antigen inhalation. Methods Experimental Animals We used 20 adult cats weighing 2.0 to 3.0 kg; they were separated into four groups for different study protocols. All animals were activelysensitized by injectingintraperitoneally a mixture of 1 ml of Ascaris suum antigen (1:100 dilution; Greer Laboratories, Lenoir, NC) and 150 mg of aluminum hydrochloride dissolved in 3 ml of physiologic saline 3 wk and 1 wk before antigen inhalation. Skin tests under anesthesia with ketamine (25 mg/kg given intramuscularly) were performed in all animals on the experimental day to determine whether sensitization had been effective. The procedure was described in detail in our previous report (14)in which wedemonstrated that there is a significant correlation between immediate intradermal reaction and pulmonary responses induced by the antigen. Therefore, we confirmed that all experimental animals
were expected to show IAR to Ascaris antigen by the skin test preceding the antigen inhalation. Briefly, 0.5070 Evans-blue was intravenously injected into each animal, and 0.02 ml of Ascaris antigen (1:100 dilution) was administered intradermally 15min later. Evaluation of the test was performed 30 min after antigen injection.
Physiologic Measurements After additional anesthesia with ketamine (25 mg/kg given intramuscularly), a femoral artery and bilateral femoral veins were cannulated with polyethylene tubes. Venous catheters were used to inject drugs, and an arterial catheter was used to monitor systemic blood pressure and to collect blood samples for gas analysis. An endotracheal tube (outer diameter, 6 mm; length,S em) was then inserted via the tracheostoma, and the chest (Receivedin originalform November 26, 1990and in revised form August 7, 1991) I From the First Department of Internal Medicine, Tohoku University School of Medicine, Sendai, Japan. 2 Correspondence and requests for reprints should be addressed to Tarnotsu Takishima, M.D., Professor and Chairman, First Department of Internal Medicine, 'Iohoku UniversitySchool of Medicine, 1-1 Seiryo-machi, Aoba-ku, Sendai 980, Japan.
71
ANTIGEN·INDUCED ATTENUATION OF NONADRENERGIC BRONCHODILATION
was opened. For further anesthesia and muscle paralysis, pentobarbital (20 mg/kg) and succinylcholine (25mg/kg) were intravenously injected. The lungs were then artificially ventilated with a constant-volume respirator (Model 661; Harvard Apparatus Co., South Natick, MA) at a tidal volume of 15 to 20 ml/kg and a frequency of 15breaths/min. Another catheter (inner diameter, 1 mm; length, 7 ern) was inserted into the distal end of the tracheal tube to obtain airway pressure directly and connected to a differential pressure transducer (MP 45, ± 50 em H 2 0 ; Validyne Co., Northridge, CA) to monitor transpulmonary pressure. Airflow at the opening of the endotracheal tube was measured with a pneumotachograph (FleischNo. (0) and a differential pressure transducer (MP 45, ± 5 ern H 2 0 ; Validyne). Pulmonary resistance (RL)was obtained by the method of Mead and Whittenberger (15). Arterial P0 2 , Pco., and pH were determined intermittently with a pH/blood gas analyzer (Model 1302; Instrumentation Laboratories, Lexington, MA). The animals' pH and blood gas tension were maintained within their normal range by intermittent administration of sodium bicarbonate and adjustment of the tidal volume of the respirator.
Animal Preparation for Nerve Stimulation For nerve stimulation, the bilateral vagosympathetic trunk was exposed and separated into two component bundles. The identities of these two components were then verified by testing for a vagal effect on heart rate and for a sympathetic effect on pupil size and systemic blood pressure (11). Liquid paraffin was applied to keep the nerves from drying. Separated bilateral cervical vagi or sympathetic nerves wereplaced on two sets of bipolar silver electrodes only during stimulation. An isolated stimulator (SEN 3201;Nihon-Koden, Tokyo, Japan) was employed for electrical stimulation with different stimulating conditions.
Effect of IAR on NANCIS In cats of Group I (n = 5), NANC inhibitory activity was examined before and after Ascaris antigen inhalation. In order to evaluate and compare NANC inhibitory activity under different conditions in the same animal, the frequency-response study was repeated. First, atropine (3 mg/kg) and propranolol (2 mg/ kg) were intravenously administered to animals to avoid cholinergic and p-adrenergic influences. After measuring baseling RL, serotonin (25 to 100ug/kg/min) was continuously infused to achieve steady-state bronchoconstriction. Bilateral cervical vagi were then electrically stimulated for 20 s with pulses of 12 V and I ms duration. After obtaining the peak dilatatory response, serotonin infusion was stopped. This procedure was repeated every 15min with stimulating frequency varied from 1,3, and 10 to 30 Hz to construct frequencyresponse curves. Thirty minutes after constructing the first frequency-response curve, Ascaris antigen (1:100 dilution) was delivered
to animals for 3 min with an ultrasonic nebulizer (Omron NE-UllB; Tateishi Co., Tokyo, Japan) (output, 0.6 ml/min) interposed between the inspiratory port of the respirator and the cannula to the lung. A frequencyresponse study after antigen inhalation was performed immediately after increased RL by antigen inhalation had returned to its baseline value (approximately 1 h after antigen inhalation). To evaluate NANC inhibitory effects quantitatively, maximal relaxation (r max) and EF 50 were obtained from each frequency-response curve; r max was defined as percent recovery of RL at 30 Hz stimulation, and EF 5 0 was defined as the frequency causing 50070 relaxation of r max • We preliminarily confirmed reproducibility of dilatatory effects elicited by electrical NANCIS stimulation by using cats not showing intradermal hyperreactivity to Ascaris antigen (n = 3). Antigen inhalation did not cause an increase in RL, and frequency-response curves showed good reproducibility in all of these animals.
Effects of IAR on VIP-induced Bronchodilation In cats of Group 2 (n = 5), dilatatoryeffects of infused VIP were examined before and after antigen inhalation. Dilatatory effects of VIP were assessed by constructing doseresponse curves. Incremental doses of VIP from 0.1,0.3, I, and 3 to 10 ug/kg were intravenously administered under cholinergic and p-adrenergic blockade and steady-state bronchoconstriction at 15-min intervals. Ascaris antigen inhalation was performed in the same way as with the other groups 30 min after the first dose-response study. A second doseresponse curve was obtained after RL, increased by antigen inhalation, had returned to its baseline value.
Effect of Protease Inhibitor on IAR-induced Dysfunction of NANCIS In cats of Group 3 (n = 5), NANC inhibitory activity was examined before and after Ascaris antigen inhalation in the same way as in those of Group 1. In this group, however, a serine protease inhibitor, leupeptin (3 mg/ kg) was intravenously administered 15 min after the antigen inhalation. Infusion of leupeptin did not influence systemic blood pressure or RL in any animal studied.
Effect of IAR on Adrenergic Nerves In cats of Group 4 (n = 5), adrenergic inhibitory activity was examined before and after the antigen inhalation by constructing frequency-response curves.The study protocol for this group was almost the same as that for Group I except for a pretreatment with propranolol. In this group, bilateral cervical sympathetic nerves instead of vagi were stimulated with the same stimulating conditions and intervals as in NAN CIS stimulation. Quantitative assessment of adrenergic activ-
ity was also made with Jmax and EF 5 0 obtained from each frequency-response curve.
Assessment of Dilatatory Responses We assessed the dilatatory effects of nerve stimulation and VIP infusion as follows. A raw constrictive response was initially obtained as the difference between RL during steady-state bronchoconstriction and baseline RL. An inhibitory response was then obtained as the difference between RL during steadystate bronchoconstriction and recovered RL caused by nerve stimulation or VIP infusion. We thus defined percent relaxation as the ratio of these differences, which indicates the degree of inhibition normalized by the raw response.
Statistical Analysis The data were expressed as means ± SE and analyzed using Student's paired t test for parametric data and Wilcoxon's signed-ranks test for nonparametric data (16). The level of statistical significance for mean values was accepted at p < 0.05.
Results
Antigen inhalation caused an increase in RLand a decrease in systemic blood pressure in all animals in Groups 1to 4. Mean maximal increases in RLafter antigen inhalation in Groups 1 to 4 were 98.9 ± 52.5, 147 ± 61.6, 85.6 ± 25.4, and 125 ± 31.5070, respectively, and there was no significant difference between these values. This immediate airway response induced by antigen inhalation depressed NANC inhibitory activity (Group 1). Individual frequency-response curves before and after antigen inhalation obtained from animals of Group 1 are shown in figure 1. In all animals of this group, frequency-response curves after antigen inhalation shifted rightward, and rmax was decreased compared with that before antigen inhalation. Mean values of r max and EF5 0 were significantly changed from 82.2 ± 4.7% and 1.7 Hz (geometric mean for EF5 0 ) to 64.3 ± 11.2% and 4.3 Hz (p < 0.05), respectively, with antigen inhalation (figure 2a). Dilatatory effects of infused VIP were weakened after IAR, as demonstrated in animals of Group 2. Mean dose-response curves before and after antigen inhalation in this group are shown in figure 3. Percent relaxation at 1, 3, and 10 ug/kg weresignificantly decreased from 40.2 ± 13.9,76.6 ± 6.2, and 87.5 ± 5.2% to 13.4 ± 11.0(p < 0.01), 49.9 ± 14.5(p < 0.05), and 63.6 ± 13.3% (p < 0.05), respectively. Pretreatment with the protease inhibitor leupeptin significantly depressed IAR-induced impairment of NANC dilatatory activity. Individual frequency-
72
MIURA, ICHINOSE, KIMURA, KATSUMATA, TAKAHASHI, INOUE, AND TAKISHIMA
100
50
,... #. c
0
.Q
a; )(
/.-. 0/°/
1-1
0-0
100
1-3
~.-.
;//' 1/ •
• ./
3 100
III
1-4
10
0
..'" "i II:
30
50
3
10
30
3
10
30
Frequency (Hz)
response curves for NANCIS before and after antigen inhalation in leupeptintreated animals are shown in figure 4. Mean values of r max and EFso were 85.2 ± 5.10,70 and 1.7Hz before antigen inhalation and 85.7 ± 2.7% and 1.8Hz after antigen inhalation, respectively. There was no significant difference between these values (figure 2b).
100
P
