Bronchoconstriction Induced by Inhaled Sodium Metabisulfite in the Guinea Pig Effect of Capsaicin Pretreatment and of Neutral Endopeptidase Inhibition1-3
JAN O. LOTVALL, BENGT-ERIC SKOOGH, RICHARD J. LEMEN, WAYNE ELWOOD, PETER J. BARNES, and K. FAN CHUNG
Introduction
Sulfites and sulfur dioxide (S02) have been shown to induce bronchoconstriction in patients with asthma (1-14). Sodium metabisulfite (MBS), a sulfite frequently used as a preservative, can cause bronchoconstriction in a minority of asthmatics when ingested (3). On the other hand, inhalation of MBS aerosol or S02 gas induces bronchoconstriction in most asthmatics. It is likely that sulfites and S02 share a common mode of action in producing airway narrowing, both because of their close chemical relationship and because bronchoconstriction induced by either agent can be attenuated by inhaled sodium cromoglycate or anticholinergic agents (4-8, 12, 14, 15). The mechanisms by which sulfites and S02 induce bronchoconstriction are largely unknown. Anticholinergic agents inhibit S02-induced airway narrowing in normal subjects (4, 5, 14), implying involvement of parasympathetic pathways. However, anticholinergic agents have little inhibitory effect on the bronchoconstrictor response to S02 or MBS in most asthmatics (4, 5, 9), suggesting that other mechanisms may be contributing in this group. It is possible that airway sensory nerves may mediate the responses to both MBS and S02 for several reasons. Firstly, both S02 gas and MBS aerosol induce cough, suggesting an action on airway sensory nerves (5). Secondly, sodium cromoglycate, which inhibits sensory nerve discharge in the dog (16), attenuate both MBS- and S02-induced bronchoconstriction (4, 14). It is unlikely that MBS has a direct effect on airway smooth muscle because guinea pig tracheal strips do not exhibit a contractile response in vitro when exposed to MBS at concentrations as great as 1 mM (Ell1390
SUMMARY Sodium metabisulfite (MBS), a commonly used preservative, induces bronchoconstriction In asthmatics, probably through the release of sulfur dioxide (SO,). The mechanisms involved in MBS- and SO,-induced bronchoconstrlction are not yet certain. We aerosolized MBS or acid control solution (pH, 2.7) to anesthetized, tracheostomized guinea pigs pretreated intravenously with propranolol (1 mg/kg). MBS was given at Increasing doubling concentrations (0.01,0.02,0.04, and 0.08 M) every 5min. Steep concentration-response curves were observed, and most animals responded at 0.02 or 0.04 M. Tachyphylaxis was seen at high concentrations and during a subsequent MBS challenge 15 min later. For pharmacologic studies, we stopped the challenge when lung resistance (RL) had Increased by at least 350%; a second challenge was found to be reproducible. MBS response was measured as the concentration needed to increase RL by 350% (PC,s,). Atropine (1 mg/kg given intravenously) did not affect PC35Q or the peak RL response. Inhibition of neutral endopeptidase by inhaled phosphoramidon (7.5 nmol) administered before the repeated challenge did not alter PC35Q value to MBS or peak RL responses (phosphoramidon, 201 ± 49% of first peak; vehicle, 164 ± 35%). In addition, the Increase in RL was not prolonged in the phosphoramidontreated group. Animals treated subcutaneously with capsaicin (50 mg/kg) 1 wk before the experiment, so as to deplete neuropeptldes from airway sensory nerves, had PC35Q values similar to those of the control animals. Our data demonstrate that inhaled MBS causes bronchoconstriction in guinea pigs by mechanisms that are due neither to a cholinergic reflex nor to the release of tachykinins from airway sensory nerves. AM REV RESPIR DIS 1990; 142:1390-1395
wood Wand Chung KF, unpublished observation). This study was performed to investigate the involvement of cholinergic efferent nerves and of airway sensory nerves and tachykinins in MBS-induced bronchoconstriction in anesthetized guinea pigs. We did this by comparing the effects of aerosolized MBS in control animals with animals pretreated with capsaicin, which depletes peptides from sensory nerves (17). Wealso performed partial dose-response curves to MBS in the same animals before and after inhibition of neutral endopeptidase, which would enhance the response if tachykinins are involved (18, 19). We examined the involvement of cholinergic pathways in the response to MBS in this model by performing partial dose-response challenges before and after atropine. Because tachyphylaxis to repeated S02 and MBS challenges has been described (6,20), we
first studied the reproducibility of full and partial dose-response challenges to MBS repeated after a 15-min interval. Methods Preparation We studied 44 Dunkin-Hartley guinea pigs (Received in original form January 27, 1990 and in revised form June 8, 1990) 1 From the Department of Thoracic Medicine, National Heart and Lung Institute, Brompton Hospital, London, United Kingdom, and the Department of Pulmonary Medicine, Gothenburg University, Gothenburg, Sweden. 2 Supported by the Medical Research Council of Great Britain, the Clinical Research Committee of Brompton Hospital, AB Draco, a subsidiary of AB Astra, and the Swedish National Heart and Lung Foundation. 3 Correspondence and requests for reprints should be addressed to Dr. J. O. LiitvaII, Department of Thoracic Medicine, National Heart and Lung Institute, Dovehouse Street, London SW3 6LY, UK.
1391
AEROSOLIZED METABISULFITE IN GUINEA PIGS
weighing 350 to 580 g. After purchase the guinea pigs were fed guinea pig chow and tap water freely. On the day of study they were anesthetized with an initial dose of 6 to 8 mllkg of urethane (Sigma, Poole, Dorset, UK), diluted to 25% wt/vol in 0.9070 saline injected intraperitoneally. Additional urethane was given as required to maintain appropriate anesthesia. A tracheal cannula (10 mm long with a 2.7 mm inner diameter) was inserted into the upper part of the cervical trachea through a tracheostomy and tied in place with suture material. A catheter 130 mm long (inner diameter, 0.6 mm) was inserted into the left carotid artery to monitor blood pressure and heart rate with a transducer (PDCR 75 SIN 1527; Druck Ltd., UK). The right external jugular vein was cannulated for the administration of drugs or fluids. Guinea pigs were positioned supine with the intratracheal cannula connected to a constant-volume mechanical ventilator (Model 50-1718; Harvard Scientific, Edenbridge, Kent, UK), at a tidal volume of 12 milkg and a frequency of 60 breaths/min. Transpulmonary pressure was measured with a transducer (Model FCO 40; Furness Controls Ltd., Bexhill, Sussex, UK) with one side attached to a catheter inserted into the right pleural cavity and the other side attached to a catheter connected to a side port of the intratracheal cannula. The pleural cannula was carefully inserted during mechanical ventilation, thus avoiding the effects of spontaneous breathing. Then 0.5 ml of air was injected into the pleural cavity to acquire a pleural signal. Pleural pressure signals were checked repeatedly throughout the experiments. A ventilatory circuit with a total volume of 20 ml was used to measure lung resistance (RL). Airflow was measured with a pneumotachograph (Type FIL; Mercury, Glasgow, UK) connected to a transducer (Model FCO 40; Furness Controls Ltd.). Aerosols were generated from an ultrasonic nebulizer (Model 2512 PulmoSonic; DeVilbiss Co., Somerset, PA) at a mass median diameter of 3.8 Ilm (geometric standard deviation, 1.3). Aerosols were administered through a separate ventilatory system that bypassed the pulmonary function system. The volume of this circuit was 50 ml, and aerosol output was found to be 40 1l1/60 breaths. We used a six-channel recorder (Model MX 6/48; Devices Ltd., London, UK) for flow, transpulmonary pressure, blood pressure, and tidal volume measurements. RL was calculated using the method of von Neergaard and Wirz (21).
Protocols Four separate studies were performed. (1) We determined the short-term reproducibility of full and partial dose-response challenges to MBS; control experiments were performed with an acid solution at pH 2.7, corresponding to the pH of the MBS solutions (range, 2.7 to 3.1). (2) We examined the effect of intravenously administered atropine (1 mg/kg)
by performing two successive partial doseresponse challenges to MBS before and after cholinergic blockade. (3) We studied the effect of capsaicin pretreatment on MBSinduced bronchoconstriction. A separate group of guinea pigs was used as controls, and were injected with the vehicle for capsaicin (90070 saline, 5% Tween® 80,5% ethanol). (4) We performed partial dose-response challenges before and after inhibition of neutral endopeptidase with inhaled phosphoramidon (4 nmol = 0.1 mM for 60 breaths). Control animals inhaled saline alone. We also examined the effect of inhaled phosphoramidon or saline on a single inhaled dose of substance P.
MRS Challenge Animals were treated intravenously with propranolol (l mg/kg) in order to avoid the effects of catecholamines released during induction of anesthesia (22). Ten minutes later, baseline RL measurements were obtained. Dose-response challenges to MBS were performed by giving increasing concentrations of MBS from the nebulizer at 5-min intervals. Each MBS concentration was nebulized for 20 breaths. After each aerosol exposure we closed the nebulizer circuit, and the animal was reconnected to the pneumotachograph for RL measurements. To achieve a constant-volume history, hyperinflations with twice the tidal volume were performed twice between each dose by manually blocking the outflow of the ventilator. When the experiment included a second challenge, this was performed after a 15-min resting period. Previous to the second challenge, animals were hyperinflated on 10 separate occasions at I-min intervals. Full concentration-response challenges to MBS included nebulization of 0.01, 0.02, 0.04, and 0.08 M MBS. Partial concentrationresponse challenges were interrupted when the animals had responded with an increase of RL of at least 350%. The second challenge was then performed up to the same concentration. Effect of Atropine To investigate the importance of acetylcholine release in MBS-induced bronchoconstriction in guinea pigs, we injected atropine (l mg/kg) intravenously 10 min before the second dose-response challenge. Effect of Capsaicin Treatment Capsaicin or control pretreatments were given 1 wk before MBS challenges. We used the method previously described by Lundberg and coworkers (17). Briefly, the guinea pigs were given 0.1 mg/kg terbutaline subcutaneously and 25 mg/kg aminophylline intraperitoneally 20 and 5 min before a single dose of 50 mg/kg of capsaicin injected subcutaneously under anesthesia (ketamine, 50 mg/kg given intramuscularly, and xylazine, 0.1 mg/kg given intramuscularly). Control animals were injected with Tween 80 (Sigma), the diluent for capsaicin. A week later, MBS challenge was per-
formed, and 10 min after the end of challenge, a single dose of capsaicin, 1 mg/kg, was injected intravenously to determine whether tachykinins were depleted from airway sensory nerves.
Effect of Inhibition of Neutral Endopeptidase by Phosphoramidon For studies of the effects of inhibition of neutral endopeptidase, both the first and the second concentration-response challenges were started at the lower concentration of 2.5 mM of MBS so that any leftward shift of the concentration-response curves could be detected. Doubling increasing concentrations of MBS were administered to the animal at 5-min intervals until an increase in RL of at least 350% was seen (PC350). Phosphoramidon (0.1 mM, 60 breaths = 4 nmol) or vehicle (saline 0.9%, 60 breaths) was inhaled by aerosol 10 min after the first MBS challenge. The second challenge was then started 10 min later. Fifteen minutes after the second MBS challenge, substance P (0.3 mM) was given by inhalation for 20 breaths to investigate if the phosphoramidon-treated animals were hyperresponsive to this mediator. We chose a concentration of 0.3 mM because in preliminary experiments we found that this concentration of substance P increased RL approximately 200% above baseline. Drugs We used the following drugs: propranolol (Inderal ®; ICI Pharmaceuticals, Cheshire, UK); atropine (Phoenix Pharmaceuticals Ltd., Gloucester, UK); terbutaline (Bricanyl®; Astra Pharmaceuticals Ltd., Kings Langley, UK); aminophylline (Antigen Ltd., Roscrea, Ireland); ketamine (Ketalar®; Parke-Davis Research Laboratories, Eastleish, UK); xylazine (Rompun®; Bayer UK Ltd., Newbury, UK); urethane, diluted to 25% wt/vol in 0.9% saline; sodium metabisulfite (MBS); capsaicin; phosphoramidon; substance P; acetylcholine (Sigma Chemical). Capsaicin was dissolved in 5% ethanol, 5% Tween 80, and 90% normal salin~. As pH control solutions, we added HCIto normal saline to achieve a pH of 2.7. MBS was dissolved in sterile saline and diluted to a volume of 15 ml, either in the morning or on the day before the experiments, and kept in a closed, air-tight container with a total capacity of 25 ml. The container was shaken for 5 s before the MBS solution (pH, < 3.4) was placed in the nebulizer. When a second concentration-response challenge was included in the protocol, a second MBS solution with the same concentration taken from a separate container was used to avoid variability in results because of loss of S02 when the solutions are exposed to room air. This procedure was, in preliminary experiments, found to give most reproducible results. Statistical Analysis Student's t test and analysis of variance (ANOVA) were used to determine significant differences between groups. The level of sig-
1392
LCTVALL, SKOOGH, LEMEN, ELWOOD, BARNES, AND CHUNG
TABLE 1
three animals responded at the 0.02 M concentration of MBS with a steep concentration-response curve. There was a marked reduction in the response at the second challenge. On the other hand, when the first concentration-response challenge was stopped after an increase in RL of at least 350% was achieved, the peak RL and the PC 350 of the two successive challenges were not significantly different (figure 3A).
MEAN BASELINE LUNG RESISTANCE BEFORE THE FIRST AND SECOND SODIUM METABISULFITE CHALLENGES'
Groups
Baseline 2.
(two full challenges) (full first challenge)
5 3
0.28 ± 0.01 0.30 ± 0.02
0.27 ± 0.D1 0.40 ± 0.08
Control animals Atropine group
(partial first challenge) (partial first challenge)
8 5
0.29 ± 0.03 0.24 ± 0.01
0.32 ± 0.04 0.31 ± 0.02
Capsaicin-treated Tween-treated
(full first challenge) (full first challenge)
7
6
0.20 ± 0.01 0.30 ± 0.02
Phosphoramidon Vehicle
(partial first challenge) (partial first challenge)
5 5
0.23 ± 0.D1 0.25 ± 0.01
8- 1
0.41 ± 0.07 0.31 ± 0.01
Effect oj Atropine
± SEM.
nificant difference was set at the value of p < 0.05. Data were analyzed with a Macintosh Plus computer (Apple Computer Inc., Cupertino, CA) using standard statistical packages. Results
Baseline RL Baseline RL before both the first and the second dose-response challenges are shown in table 1. Mean baseline RL was found to be significantly higher before the second challenge (Baseline 2) in all groups, but analysis of variance (repeated measures) showed no significant differences among the groups. However, Baseline 1 in capsaicin-pretreated animals was significantly lower than Baseline 1 in animals treated with Tween 80 alone (p < 0.05). Full and Partial Dose Responses to MBS The mean increases in RL for a first con-
1A
centration-response challenge to MBS and pH-control solution are shown in figure 1. Saline with a pH of 2.7 increased RL to a maximum of 18 ± 4% (n = 5). On the other hand, MBS solutions increased RL in a concentration-dependent manner. RL started to increase during or immediately after nebulization of MBS. Peak RL was reached within 67 ± 5 s after the beginning of nebulization. The time course of MBS responses, expressed as percent of peak response for each individual animal, is shown in figure IE. The recovery was rapid, and 1 min after peak, the RL had returned to 53 ± 70/0 of the peak response. At 4 min, one hyperinflation with twice the tidal volume was performed, and 1 min later the animals had recovered to 22 ± 4% of the peak response. Concentration-response curves from three guinea pigs exposed to two successive full concentration-response challenges 15 min apart are shown in figure 2. All
100 ..J
Q)
The effects of placebo or atropine on the individual PC350 values are shown in figure 3A. There were no significant differences in PC350 values for the first and second concentration-response challenges in either control or atropine-treated animals. We also compared the peak responses in control and atropine-treated animals as a percentage of the first peak response seen at the same MBS concentration (figure 3B). Atropine (1 mg/kg) given intravenously had no significant effect on the peak RL responses at the second challenge (107 ± 17% of the first challenge).
Effect of Capsaicin Treatment Animals pretreated with a single subcutaneous injection of capsaicin (50 mg/kg) 1 wk before the experiment were as responsive to MBS when comparing PC 350 values with those of the control animals (figure 4). Intravenously administered capsaicin (1 mg/kg) produced a small increase in transpulmonary pressure in three capsaicinized animals (8 ± 3 cm H 2 0) compared with that in three control animals treated with Tween 80 alone (49 ± 8 cm H 2 0, p < 0.05), indicating that airway sensory nerves were depleted of bronchoconstrictor tachykinins.
II:
til
~ ~
~
...
Q)
Q)
Co
(,)
-= ;,!!
Baseline 1
pH Control animals Tachyphylaxis
• em H,O/ml x
2000
Animals (n)
1000
0 ;,!! 0
~
50
..J
II:
OJ
0
o
0.01
0.1
MBS [M]
/ 1
2 .
i
3
TIme (min)
I
4
"
r---'
Recovery (5 min)
Fig. 1. A. Increase in lung resistance (RL), as percent increase over baseline, induced by increasing concentrations of aerosolized MBS (closed circles) (n = 21) compared with pH control solution (open circles) (pH, 2.7; n = 5). For the MBS concentrations we have included data from total and partial challenges (0.01 M: n = 21; 0.02 M: n = 21; 0.04 M: n = 13; 0.08: n = 6). The pH control solution did not produce significant increases in RL. MBS, on the other hand, produced significant increases in RL at concentrations of 0.02 M and above. B. Time course of changes in RL induced by aerosolized MBS, expressed as percent of peak RL. One minute after peak, RL had returned to 53 ± 7% of peak RL. One minute after a single hyperinflation with twice the tidal volume, performed at 4 min, the animals had recovered to 22 ± 4% of peak responses.
Effect of Phosphoramidon Inhibition of neutral endopeptidase with phosphoramidon aerosol (0.1 mM; 60 breaths; 4 nmol) produced no significant shift in PC350 (figure SA), and did not change the peak RL when compared with that of the saline-treated animals (figure 5B). Furthermore, phosphoramidon aerosol did not prolong the duration of bronchoconstriction induced by MBS aerosol (figure sq. After the end of MBS challenges, substance P aerosol given for 20 breaths at a concentration of 0.3 mM produced an increase in RL of 269 ± 78% in three saline-treated animals, but in five phosphoramidon-treated animals it producedanincreaseinRLof3,190 ± 530% (p < 0.05). Thus, inhibition of neutral
1393
AEROSOLIZED METABISULFITE IN GUINEA PIGS
Fig. 2. Increase in RL, as percent increase over baseline, to two successive full concentration-response challenge to MBS separated by 15 min in three animals. During the first concentrationresponse challenge (closed circles), there was a smaller increase in RL at the higher doses in all three animals. When a second concentration-response challenge was performed (open circles), RL increased to a much smaller degree than after the first, showing that shortterm tachyphylaxis was present.
0.1
3000
~ VI
'f"!
." 5
C Ii!
2000
.-
~10 -"" 2
1000
l.O
U
c.
MBS[M]
Discussion In the present study, we have shown that inhaled MBS produces severe bronchoconstriction in the guinea pig. However, we found no support for the involvement of tachykinins and/or capsaicin-sensitive nerves in the airway narrowing induced by MBS in this species since neither capsaicin pretreatment nor inhibition of neutral endopeptidase altered the response. As in asthmatics (5,20), there was a pronounced short-term tachyphylaxis to MBS (figure 2), which interferes with duplicate measurements, although this
effect was dependent on the concentration of MBS inhaled. Thus, tachyphylaxis was minimized when repeated partial concentration-response curves to MBS were performed at lower concentrations and when we discontinued the challenge when RL had increased 3500/0 above baseline. We used this method to demonstrate that atropine or phosphoramidon did not affect the response to MBS, implying that cholinergic pathways and tachykinin release are not involved in the response to MBS in anesthetized guinea pigs. In addition, we compared the effects of MBS aerosol on animals pretreated with capsaicin with those treated with a control solution. This treatment has been shown
150
B
100 ....I
A
ex: ~
t':l
,
o In
-
0-
1Il
~
;;::
""""
M
",'
U
-
~
"
i" Q.
1Il
,,0
i.i:
~~
~
0
"
50
~ 0
10 ·2 10 ·2
~
o
~I-
endopeptidase by phosphoramidon was effective in our experiments.
10.1
~
Second PC350 (M)
10.1
o Controls
Atropine
Fig. 3. A. Individual PC 3•• obtained from two successive partial concentration-respons13 challenges to MBS (first and second challenges) in guinea pigs treated intravenously with placebo (open circles) or with atropine 1 mg 10 min before the second challenge (closed circles). The dashed line is the line of identity. The reproducibility of the PC". is shown to be good, and atropine had no effect on the responsiveness to MBS. B. Mean peak RL (± SEM) at the second MBS challenge, expressed as percent of the peak RL at the first challenge. This was calculated using the second peak RL response seen at the same MBS concentration as at the first challenge (control animals: six at 0.02 M and two at 0.04 M; atropine-treated animals: four at 0.02 M and one at 0.04 M). There was no significant change in peak RL after atropine.
0.01
~
~~ Controls
~~
¥ Capsaicin pretreated
Fig. 4. Comparison of individual PC 3•• to MBS concentration-response challenges in animals pretreated with capsaicin (closed triangles) (n = 7) or with control medium (open triangles) (n = 6) 1 wk earlier. There were no significant differences in mean PC 3•• between the two groups.
to block the noncholinergic component of bronchoconstrictor responses to electrical stimulation of the vagus sensory nerves (17,23,24). There was no significant difference in the PC 350 to MBS between the two groups, suggesting that activation of sensory nerves and tachykinin release is not involved in MBS-induced airway narrowing in guinea pigs. In vivo and in vitro studies in the guinea pig have shown that bronchoconstrictor responses elicited by electrical stimulation of the cervical vagus nerves and by capsaicin are mediated by both cholinergic and noncholinergic mechanisms (23, 24). The noncholinergic component of the responses is likely to be caused by neuropeptides released in the airways (for example, substance P and neurokinin A), perhaps through a local axon reflex involving airway sensory nerves (25). Thus, antagonists of substance P and pretreatment of guinea pigs with capsaicin both inhibit the noncholinergic bronchoconstrictor responses elicited by stimulation of the cervical vagus nerves (26). In addition, it has been· shown that the latter stimulus is enhanced by inhibition of neutral endopeptidase with inhaled phosphoramidon (19). In the present study, we observed little effect of capsaicin or phosphoramidon pretreatment on MBS-induced bronchoconstriction. It may be argued that a cholinergic component of MBS responses in the intact guinea pig cannot be completely excluded because anesthesia is known to
1394
laWAll, SKOOGH, lEMEN, ELWOOD, BARNES, AND CHUNG
l 10.
1
1A
II
B
I
100
I
I I
...•.
oon
U'"
~
//.~··O
Ui '-
u:
10
·2
Second PC350
.::- 100
1
a
I J
~ ~
..-6 10 .,
1 ~
I I
~
~/
C-
10 ·2
200
a: -'"
ro QJ
C-
50
'0
1
~
0
~
o~
Controls
Phosphoramidon pretreated
o
1
2
TIME (min)
T~
3"
/
~
-' 4 r - -Recover; IS min)
Fig. 5. A. Comparison of individual PC 3S• obtained from a first and a second concentration-response challenge to MBS in animals treated with phosphoramidon (4 nmol; inhaled), an inhibitor of neutral endopeptidase (closed circles), or saline (open circles), the vehicle for phosphoramidon (n = 5 for both groups). Phosphoramidon aerosol did not alter the PC 3S • to MBS compared with that in the control group. B. Comparison of peak Rl between a first and second MBS challenge as percent of the previous peak, using the response seen at the same MBS concentration as at the first challenge (control animals: one at 0.02 M and four at 0.04 M; phosphoramidon-treated animals: four at 0.04 M and one at 0.08 M). There was no significant increase in peak Rl at the second challenge after phosphoramidon. C. Comparison of the time course of MBS-induced bronchoconstriction after 4 nmol of inhaled phosphoramidon (closed circles), an inhibitor of neutral endopeptidase, or saline (open circles), both given by aerosol (60 breaths). Rl is expressed as percent of peak response. Animals treated with phosphoramidon (n = 5) did not show a prolonged increase in Rl when compared with those treated with saline (open circles) (n = 5), or to the time course observed before phosphoramidon treatment (open squares) (n = 10).
inhibit neural reflexes (27-29). Depth of anesthesia affects cortical pathways and ganglionic neurotransmission, but to a very small degree axonal conduction (27-29). Therefore, local axonal conduction such as the axon reflex would be affected to a negligible degree by the anesthesia in the present study (27). Although cholinergic reflexes could have been attenuated by general anesthesia (27-29), we do not believe they were abolished for several reasons. Firstly, Hey and coworkers (30) recently showed that anesthetized guinea pigs have an intact central nervous cholinergic innervation to the airways that can be stimulated by electrical stimulation of specific areas in the dorsal medulla. Secondly, bradykinin given to urethane-anesthetized guinea pigs causes bronchoconstriction that is partly inhibited by atropine (31), showing that cholinergic reflexes are functioning in animals given the anesthetic used in the present study. Despite this, care should be used when drawing conclusions from our data showing no effect of atropine onMBS-induced responses, although cholinergic reflexes as a major component are unlikely. The mechanism(s) behind the noncholinergic component of sulfite- and S02-induced airway narrowing in the guinea pig, and in humans, are still largely unclear. In both asthmatic subjects and in anesthetized guinea pigs, the cholinergic component of MBS-induced bronchoconstriction would appear to be minimal (4, 5, 7). In humans, the noncholinergic component of sulfite-induced
bronchoconstriction is largely inhibited by sodium cromoglycate (4,5, 7), a drug that inhibits both degranulation of mast cells and airway sensory nerve activation (16, 32). It has therefore been suggested that MBS in asthmatic subjects may cause bronchoconstriction through release of tachykinins from airway sensory nerves. However, in the present study, using the guinea pig, capsaicin pretreatment, which depleted sensory nerves, and phosphoramidon, which potentiated inhaled substance P, did not influence the responses to inhaled MBS. These data suggest strongly that release of tachykinin from airway sensory nerves is not an important component in MBS-induced bronchoconstriction in the guinea pig. Further studies are required to determine the cause of bronchoconstriction seen after inhaled MBS in both humans and guinea pigs. Acknowledgment The writers thank Dr. Geoff Nichol for helpful discussion during the planning of this study. References I. DeVries K, Giikemeijer JDM, Orie NGM, Peset
R, Sluiter HJ. Bronchial tree response of the allergic and nonallergic stimuli in patients with generalized obstructive lung disease. Bull Int Union Tuberc 1976; 51:617-9. 2. Sheppard D, Scott WW, Uehara CF, Nadel JA, Boushey HA. Lower threshold and greater bronchomotor responsiveness of asthmatic subjects to sulfur dioxide. Am Rev Respir Dis 1981; 122:873-8. 3. Stevenson DD, Simon RA. Sensitivity to ingested metabisulfites in asthmatic subjects. J Allergy
Clin Immunol 1981; 68:26-32. 4. Snashall PD, Baldwin C. Mechanisms of sulphur dioxide induced bronchoconstriction in normal and asthmatic man. Thorax 1982; 37:118-23. 5. Nichol GM, Nix A, Chung KF, Barnes PJ. Characterisation of bronchoconstrictor responses to sodium metabisulphite aerosol in atopic subjects with and without asthma. Thorax 1989; 44:1009-14. 6. Fine JM, Gordon T, Sheppard D. The roles of pH and ionic species in sulfur dioxide- and sulfiteinduced bronchoconstriction. Am Rev Respir Dis 1987; 136:1122-36. 7. Tan WC, Cripps E, Douglas N, Sudlow ME Protective effect of drugs on bronchoconstriction induced by sulphur dioxide. Thorax 1982; 37:671-6. 8. Delohery J, Simmul R, Castle WD, Allen DH. The relationship of inhaled sulfur dioxide reactivity to ingested metabisulfite sensitivity in patients with asthma. Am Rev Respir Dis 1984; 130:1027-32. 9. Jackson-Myers D, Bigby BG, Calvayrac P, Sheppard D, Boushey HA. Interaction of cromolyn and a muscarinic antagonist in inhibiting bronchial reactivity to sulfur dioxide and to eucapnic hyperpnea alone. Am Rev Respir Dis 1986; 133:1154-8. 10. Schwartz HJ, Chester EH. Bronchospastic responses to aerosolized metabisulfite in asthmatic subjects: potential mechanism and clinical implications. J Allergy Clin Immunol 1984; 74:511-3. 11. Bethel RA, Erie DJ, Epstein J, Sheppard D, Nadel JA, Boushey HA. Effect of exercise rate and route of inhalation on sulfur-dioxide-induced bronchoconstriction in asthmatic subjects. Am Rev Respir Dis 1983; 128:592-6. 12. Seale JP, Temple DM, Tennant CM. Bronchoconstriction by nebulized metabisulfite solutions (S02) and its modification by ipratropium bromide. Ann Allergy 1988; 61:209-13. 13. Petering DH, Shih NT. Biochemistry of bisulfite-sulfur dioxide. Environ Res 1975; 9:55-65. 14. Dixon CMS, Ind pw. Metabisulphite-induced bronchoconstriction: Mechanisms. Am Rev Respir Dis 1988; 137(Suppl:238). 15. Nadel JA, Salem H, Tamplin B, Tokiwa Y. Mechanism of bronchoconstriction during inhalation of sulfur dioxide. J Appl Physiol 1965; 20:164-7. 16. Dixon M, Jackson DM, Richards 1M. The action of sodium cromoglycate on "c" fibre endings
1395
AEROSOLIZED METABISULFITE IN GUINEA PIGS
in the dog lung. Br J Pharmacol 1980; 70:11-3. 17. Lundberg JM, Brodin E, Saria A. Effects and distribution of vagal capsaicin sensitive substance P neurons with special reference to the trachea and lungs. Acta Physiol Scand 1983; 19:243-52. 18. LotvallJO, Hui KP, Lemen RJ, Skoogh BE, Barnes PJ, Chung KF. Effects of aerosolized substance P in guinea pigs: contribution of airway microvascular leakage to bronchoconstriction. Eur Respir J 1989; 2:753s. 19. Dusser DJ, Umeno E, Graf PD, Djokic T, Borson DB, Nadel JA. Airway neutral endopeptidaselike enzyme modulates tachykinin-induced bronchoconstriction in vivo. J Appl Physiol 1989; 65:2585-91. 20. Sheppard D, Epstein J, Bethel RA, Nadel JA, Boushey HA. Tolerance to sulfur dioxide-induced bronchoconstriction in subjects with asthma. Environ Res 1983; 30:412-9. 21. von Neergaard K, Wirz K. Die messung den stromungswiderstiinde in den atemwegen des menschen; insbesondere bei asthma und emphysem. Z Klin Med 1927; 105:51-82.
22. Spriggs TLB. The effects of anaesthesia induced by urethane or phenobarbitone upon the distribution of peripheral catechol amines in the rat. Br J Pharmacol 1965; 24:752-8. 23. Martling CR, Saria A, Andersson P, Lundberg JM. Capsaicin pretreatment inhibits vagal cholinergic and non-cholinergic control of pulmonary mechanics in the guinea pig. Naunyn Schmiedebergs Arch Pharmacol1984; 325:343-8. 24. Lundberg JM, Saria A. Bronchial smooth muscle contraction induced by stimulation of capsaicinsensitive sensory neurons. Acta Physiol Scand 1982; 116:473-6. 25. Barnes PJ. Asthma as an axon reflex. Lancet 1986; 1:242-5. 26. Lundberg JM, Saria A, Brodin E, Rosell S, Folkers K. A substance P antagonist inhibits vagaily induced increase in vascular permeability and bronchial smooth muscle contraction in the guinea pig. Proc Nat! Acad Sci USA 1983; 80:1120-4. 27. Skoogh BE, Holtzman MJ, Sheller JR, Nadel JA. Barbiturates depress vagal motor pathway to ferret trachea at ganglia. J Appl Physiol 1982;
53:253-7. 28. Holtzman MJ, Hahn HL, Sasaki K, et at. Selective effect of general anesthetics on reflex bronchoconstrictor responses in dogs. J Appl Physiol 1982; 53:126-33. 29. Jackson DM, Richards 1M. The effects of pentobarbitone and chloralose on the vagal component of bronchoconstriction produced by histamine aerosol in the anesthetized dog. Br J Pharmacol 1977; 61:251-6. 30. Hey JA, del Prado M, Chapman RW. Activation of a novel medullary pathway elicits a vagal, cholinergic bronchoconstriction in guinea-pigs. Pulmon Pharmacol 1990; 3:53-4. 31. Ichinose M, Belvisi MG, Barnes PJ. Bradykinin-induced bronchoconstriction in guinea pig: role of neural mechanisms. J. Pharmacol Exp Ther 1990; 253:594-9. 32. Cox JSG, Altonunyan REC. Nature and mode of action of disodium cromoglycate (Lomudal). Respiration; 1979; 27 (Suppl:292-309).