Endogenous Sensory Neuropeptide Release Enhances Nonspecific Airway Responsiveness in Guinea Pigs1- 3
TZUEN.REN HSIUE,4 ALLAN GARLAND, DANIEL W. RAY, MARC B. HERSHENSON, ALAN R. LEFF, and JULIAN SOLWAY
Introduction
The pulmonary airways of rodents and of humans are innervated with sensory c-fibers (1-3), which provide an afferent limb for central neural reflex control of airway function. These neurons also synthesize and store neuropeptides in granules found within their sensitive terminal varicosities. A variety of chemical, physical, or electrical stimuli can cause sensory c-fibers to releasetheir neuropeptides locally into the innervated structures, where these substances often initiate important physiologic events. In the airways, c-fiber stimulation causes the release of two classes of neuropeptides: tachykinins, including substance P (SP) and neurokinin A (NKA), and calcitonin gene-related peptide. On the basis of studies of endogenously released or exogenously administered neuropeptides, it is established that these substances can cause bronchoconstriction, bronchial vasodilatation, bronchovascular hyperpermeability, mucus hypersecretion, and/or recruitment of inflammatory cells into the airway wall. The known actions of sensory neuropeptides in regulating airway function have been detailed in a recent review (1). One salient feature of asthma is the presence of nonspecific bronchial hyperresponsiveness. It has been proposed that cellular inflammation (4) and/or airway wall edema and thickening (5) may participate in the pathogenesis of airway hyperresponsiveness. Because sensory neuropeptides can provoke neutrophil migration into the airwaywall (6) and can provoke abnormal bronchovascular hyperpermeability (7) that might lead to airway wall edema and thickening, we hypothesized that endogenous sensory neuropeptide release might lead to airway hyperresponsiveness. Totest this hypothesis, we have administered capsaicin (a c-fiber stimulant) by aerosol to guinea pigs and assessed their airway bronchoconstrictor responsiveness the following day. Our results demonstrate 148
SUMMARY To test whether endogenous sensory neuropeptlde release results In airway hyper· responsiveness to exogenous bronchoconstrictor stimuli, male Camm·Hartley guinea pigs were exposed either to capsaicin aerosol for 10 min (CAP-AER) or to saline aerosol (SAL-AER)as a control condition. The following dey, animals were anesthetized, tracheostomlzed, and bets·adrenerglcally blocked with propranolol, and their bronchoconstrictor responses to Intrevenously administered acetylcholine (ACh), neurokinin A (NKA), or capselcln were measured. The bronchoconstrlctlon Induced by lsocapnlc dry gas hyperpnea also was assessed. Compared with the SAL-AERcontrol group, the CAp·AER·treated animals exhibited augmented bronchoconstrictor responses to ACh and NKA. In contrast, the SAL-AER and CAP-AER groups had equivalent bronchoconstrictor reo sponses to dry gas hyperpnea and to Intrevenously administered capsaicin. CAP-AERtreatment caused neutrophilic alrwey Inflammation, as reflected In Increased numbers of neutrophlls In bron· choalveolar levege fluid obtslned from CAP-AER·treated animals. Ablation of airway c·flber neuron function (by chronic pretreatment with capsaicin prior to capsaicin aerosol Inhalation) eliminated the ACh hyperresponslveness observed In the CAP-AER·treatedanimals, demonstrating that senso· ry nerve products play a key role In the development of this nonspecific hyperresponslveness. Our results demonstrate thst sensory nerve stimulation with capsaicin aerosolle8ds to nonspecific bron· choconstrlctor hyperresponslveness and cellular alrwey Inflammation, and thus disclose another potentially Important role of sensory nerves In regUlating alrwey function. Because CAP-AER·treated animals were hyperresponslve to exogenous NKA (a putative mediator of sensory·nerve-Induced bronchoconstrlctlon) but not to Intrevenously administered capsaicin or dry gas hyperpnea (each of which stimulate endogenous sensory nerve tachyklnln release), we also conclude that capsaicin aerosol exposure Impairs sensory nerve function through an uncertain mechanism. AM REV RESPIR DIS 1992; 148:141-153
that experimentally induced endogenous sensory neuropeptide release can elicit nonspecific airway hyperresponsiveness, and thus we identify another potential role of airway sensory nerves in controlling airway responses. Methods Study Design On Day 0 of each of the five Experiments described below,normal male Camm-Hartley guinea pigs underwent either inhalation of aerosolized capsaicin, to elicit endogenous sensory neuropeptide release within the airways(the CAP-AER group), or inhalation of saline aerosol (SAL-AER) as the control condition. 1\venty-four hours later (Day 1), the animals underwent airwayresponsivenesstesting or bronchoalveolar lavage, after which they werekilled by anesthetic overdose.In Experiment 2 only (see below) the CAP-AER or SAL-AER inhalation exposures were preceded in all animals by chronic capsaicin pretreatment, a regimen that eliminates neuropeptides from sensory c-fibers and thus prevents sensory neuropeptide release upon subsequent c-fiber stimulation.
Aerosol Inhalation (Day 0) Endotoxin-free capsaicin wasdissolvedin ethanol and diluted with normal saline to make a stock 10-2 M capsaicin solution containing 10070 ethanol; 10-3 and 10-4 M capsaicin solutions were made by diluting stock solution in normal saline. Animals were placed into a 26-L clear Plexiglas'" box through which nebulized (Pulmo-Sonic 25; DeVilbiss Co.,
(Received in original form April 1, 1991 and in revised form January 22, 1992) 1 From the Department of Medicine and Pediatrics, University of Chicago, Chicago, and the Department of Medicine,Evanston Hospital, Evanston, Illinois. 2 Supported by Grants HL-02205, HL-41009, HL-32495, HL-35718, HL-46368, and HL-02376 from the National Heart, Lung, and Blood Institute and by grants from the American Lung Association. 3 Correspondence and requests for reprints should be addressed to Julian Solway, M.D., Box MC6026, University of Chicago, 5841 S. Maryland Ave., Chicago, IL 60637. • Recipient of a grant from the National ChengKung University Hospital, 'Iainan, Taiwan.
149
SENSORY NEUROPEPTlDES AND AIRWAY REACTIVITY
Somerset, PA), capsaicinsolutions wereblown at 8 L/min. A mixing fan ensured uniform aerosol concentration within the exposure box. Awake animals spontaneously breathed capsaicin aerosols for 10 min (10-4 M for 5 min, followed by 10-3 M for 5 min), or until the development ofvisual evidence oflabored breathing, which was identified by the presence of repeated cough, deep chest wall indrawing, tachypnea, and/or cyanosis. Each animal was immediately returned to its cage, into which supplemental oxygen was blown for 5 min, by which time these signs always resolved. Every animal developed labored breathing within the lO-min exposure period; animals that received capsaicin aerosol on Day 0 are hereafter designated"CAP-AER" animals. Control animals were treated identically, except that for 10 min they breathed nebulized normal saline containing I % ethanol (i.e., capsaicin vehicle)instead of capsaicin solution. Animals that received saline aerosol on Day 0 are hereafter designated "SAL-AER" animals.
Animal Instrumentation and Preparation (Day 1) Each guinea pig was anesthetized intraperitoneally with sodium pentobarbital (75 rng/kg). A 13-gauge cannula was inserted through a high cervical tracheostomy, and the animal was mechanically ventilated at "quietbreathing" settings (6 ml/kg, 60 breaths/min) with humidified air at room temperature. As previously described (8-11), inspiratory and expiratory limbs of the ventilator circuit were attached to the tracheal cannula through a 2-cm common segment to minimize conditioning of inspired gas during subsequent dry gas hyperpnea challenges(seebelow). One carotid artery and one internal jugular vein werecannulated with PE-6O tubing. Each animal was then placed in an isothermal constant-mass whole-body plethysmography (1O.1-L displacement). Changes in lung volume were reflected as changes in plethysmographic pressure measured relative to a 20-L reference chamber with a differential pressure transducer (Honeywell MicroSwitch 163PCOlD36; Honeywell, Denver, CO). Airflow was obtained by electrical differentiation of the volume signal. Transrespiratory pressure (the pressure difference between tracheal pressure and plethysmograph pressure) was measured with a differential pressure transducer (Honeywell MicroSwitch 143PCOlD). Respiratory volume excursion, flow, transrespiratory pressure, and blood pressure were recorded on a chart recorder (Gould 2800S;Gould Instruments, Cleveland, OH). Respiratory system resistance (Rrs) was calculated from differences of flow and of transrespiratory pressure at isovolume points (8-11), an adaptation of the method of Goldman and coworkers (12). Each guinea pig received propranolol (1 mg/kg) intravenously to minimize potential changes in bronchoconstrictor response resulting from changes in levelsof circulating catecholamines (8-11); no change in Rrs resulted from this infusion. We
have previously shown that such beta-adrenergic blockade does not influence the magnitude of hyperpnea-induced bronchoconstriction in guinea pigs (10, 11).
Intravenously Administered Bronchoconstrictor Agonist Responsiveness Testing (Day 1) A single deep inflation of three times the "quiet-breathing" tidal volume was given to establish a fixed volume history, and baseline Rrs was obtained 1 min later. Then, intravenousbolus infusions (0.8 ml/kg) ofbronchoconstrictor agonist were sequentially administered in increasing half-log steps flushed by 0.2 ml normal saline, and Rrs measurements were obtained at the peak airway pressure after each such infusion. Each subsequent agonist dose was administered immediately after peak response from the previous dose was achieved; return to baseline Rrs was not allowed until all doses wereadministered. For intravenous administration, acetylcholine (ACh) and NKA weredissolved in normal saline; capsaicin was dissolved in normal saline containing < 1% ethanol.
Dry Gas Hyperpnea Stimulus-Response 'Testing (Day 1) Animals breathed fully humidified 50% oxygen in air at room temperature at the "quietbreathing" ventilator settings. A single deep inflation to three times tidal volume was given to establish a fixed volume history; 1 min later baseline Rrs was measured, and then isocapnic dry gas hyperpnea was mechanically imposed for 10min, using dry 95% 0,5% CO, at room temperature as inspired gas (8-11).Ventilator frequency was increased to 150 breaths/min, and tidal volume was adjusted to 3 ml. After 10 min of hyperpnea, the animal was returned to "quite-breathing" of humidified 50% oxygenin air. Rrs changes were monitored throughout hyperpnea and the posthyperpnea "recovery"period. The animal was allowed to recover to baseline Rrs, and the entire dry gas hyperpnea challenge/recovery sequence was repeated three more times, using 4, 5, and 6 ml tidal volumes during hyperpnea.
Bronchoalveolar Lavage (Day 1) After instrumentation and preparation as described above, animals were killed by exsanguination through the carotid artery; no physiologic studies were performed. The lungs, heart, and mediastinal structures were removed en bloc, and the esophagus was dissected away to reveal the main bronchi clearly.The trachea was sectioned above the main carina, and the right main bronchus was intubated with a polyethylene catheter that was tied securely in place. Normal saline (3 ml at 37° C) was instilled into the right lung, allowed to dwell 20 s, and then withdrawn. Lavage was repeated with another 3 ml, and the
returned fluid was combined with the first lavagate. Recovered bronchoalveolar lavage (BAL) fluid volume was measured, and the fluid was placed immediately on ice. Cells were sedimented by centrifugation at 200 x g for 10 min at 4° C. Supernatant was discarded, and cells were resuspended in 2 ml Hanks' balanced salt solution with 1% gelatin. Cell count was performed on a hemacytometer, and differential count of 300 cells was performed on cells transferred to a glass slide using a Cytospin centrifuge (Shandon, Pittsburgh, PA) and stained with eosin and methylene blue.
Chronic Capsaicin Pretreatment (Days -11 and - 4) Animals that were studied in Experiment 2 (see below) were depleted of sensory neuropeptides prior to aerosol inhalation at Day 0: Sensoryneuropeptide depletion was accomplished as previously detailed (10),bY"chronic" subcutaneous capsaicin pretreatment on two occasions; these pretreatments took place 11 and 4 days prior to aerosol inhalation (Days 11 and 4, respectively). Prior to capsaicin administration on these two days, guinea pigs wereanesthetized intramuscularly with ketamine (50 rng/kg) and xylazine (0.1 mg/kg) and treated intra peritoneally with aminophylline (10mg/kg) and subcutaneously terbutaline (0.1 mg/kg) to blunt severe bronchoconstriction caused by subcutaneously administered capsaicin. On Day 11, capsaicin (50 mg/kg) was administered subcutaneously, and a larger dose (100mg/kg) as given on Day 4. Oxygen was blown into the cage on each occasion until the animal recoveredfrom anesthesia.
Experiments Experiment 1. Intravenously administered ACh andhyperpnea responsiveness. In order to evaluate the effect of capsaicin-aerosolinduced endogenous neuropeptide releaseupon cholinergic airway responsiveness, CAPAER (n = 8, 533 to 620 g) and SAL-AER (n = 8; 550 to 610 g), animals were exposed to capsaicin aerosol or to saline aerosol (as described above) on Day O. On Day 1 each animal underwent intravenously administered ACh responsiveness testing (dose range, 10-to 10-7 mol/kg), followed 40 min later by dry gas hyperpnea stimulus-response testing. Experiment 2. Intravenously administered
ACh and hyperpnea responsiveness after chronic capsaicin pretreatment. As shown in RESULTS, the CAF-AERanimals in Experiment 1 exhibited exaggerated bronchoconstrictor responses to intravenously administered ACh compared with SAL·AER animals. To test whether the altered responsiveness induced by capsaicin aerosol inhalation depended upon intact sensory nerve function, Experiment 1 was repeated in additional guinea pigswhose airway sensorineural function was ablated by chronic capsaicin pretreatment (13). Thus, all animals studied in Experiment2 werechronically capsaicin-pretreated on Days
150 ::J (II
HSIUE, GARLAND, RAY, HERSHENSON, LEFF, AND SOLWAY
40
::J (II
TZUEN.REN HSIUE,4 ALLAN GARLAND, DANIEL W. RAY, MARC B. HERSHENSON, ALAN R. LEFF, and JULIAN SOLWAY
Introduction
The pulmonary airways of rodents and of humans are innervated with sensory c-fibers (1-3), which provide an afferent limb for central neural reflex control of airway function. These neurons also synthesize and store neuropeptides in granules found within their sensitive terminal varicosities. A variety of chemical, physical, or electrical stimuli can cause sensory c-fibers to releasetheir neuropeptides locally into the innervated structures, where these substances often initiate important physiologic events. In the airways, c-fiber stimulation causes the release of two classes of neuropeptides: tachykinins, including substance P (SP) and neurokinin A (NKA), and calcitonin gene-related peptide. On the basis of studies of endogenously released or exogenously administered neuropeptides, it is established that these substances can cause bronchoconstriction, bronchial vasodilatation, bronchovascular hyperpermeability, mucus hypersecretion, and/or recruitment of inflammatory cells into the airway wall. The known actions of sensory neuropeptides in regulating airway function have been detailed in a recent review (1). One salient feature of asthma is the presence of nonspecific bronchial hyperresponsiveness. It has been proposed that cellular inflammation (4) and/or airway wall edema and thickening (5) may participate in the pathogenesis of airway hyperresponsiveness. Because sensory neuropeptides can provoke neutrophil migration into the airwaywall (6) and can provoke abnormal bronchovascular hyperpermeability (7) that might lead to airway wall edema and thickening, we hypothesized that endogenous sensory neuropeptide release might lead to airway hyperresponsiveness. Totest this hypothesis, we have administered capsaicin (a c-fiber stimulant) by aerosol to guinea pigs and assessed their airway bronchoconstrictor responsiveness the following day. Our results demonstrate 148
SUMMARY To test whether endogenous sensory neuropeptlde release results In airway hyper· responsiveness to exogenous bronchoconstrictor stimuli, male Camm·Hartley guinea pigs were exposed either to capsaicin aerosol for 10 min (CAP-AER) or to saline aerosol (SAL-AER)as a control condition. The following dey, animals were anesthetized, tracheostomlzed, and bets·adrenerglcally blocked with propranolol, and their bronchoconstrictor responses to Intrevenously administered acetylcholine (ACh), neurokinin A (NKA), or capselcln were measured. The bronchoconstrlctlon Induced by lsocapnlc dry gas hyperpnea also was assessed. Compared with the SAL-AERcontrol group, the CAp·AER·treated animals exhibited augmented bronchoconstrictor responses to ACh and NKA. In contrast, the SAL-AER and CAP-AER groups had equivalent bronchoconstrictor reo sponses to dry gas hyperpnea and to Intrevenously administered capsaicin. CAP-AERtreatment caused neutrophilic alrwey Inflammation, as reflected In Increased numbers of neutrophlls In bron· choalveolar levege fluid obtslned from CAP-AER·treated animals. Ablation of airway c·flber neuron function (by chronic pretreatment with capsaicin prior to capsaicin aerosol Inhalation) eliminated the ACh hyperresponslveness observed In the CAP-AER·treatedanimals, demonstrating that senso· ry nerve products play a key role In the development of this nonspecific hyperresponslveness. Our results demonstrate thst sensory nerve stimulation with capsaicin aerosolle8ds to nonspecific bron· choconstrlctor hyperresponslveness and cellular alrwey Inflammation, and thus disclose another potentially Important role of sensory nerves In regUlating alrwey function. Because CAP-AER·treated animals were hyperresponslve to exogenous NKA (a putative mediator of sensory·nerve-Induced bronchoconstrlctlon) but not to Intrevenously administered capsaicin or dry gas hyperpnea (each of which stimulate endogenous sensory nerve tachyklnln release), we also conclude that capsaicin aerosol exposure Impairs sensory nerve function through an uncertain mechanism. AM REV RESPIR DIS 1992; 148:141-153
that experimentally induced endogenous sensory neuropeptide release can elicit nonspecific airway hyperresponsiveness, and thus we identify another potential role of airway sensory nerves in controlling airway responses. Methods Study Design On Day 0 of each of the five Experiments described below,normal male Camm-Hartley guinea pigs underwent either inhalation of aerosolized capsaicin, to elicit endogenous sensory neuropeptide release within the airways(the CAP-AER group), or inhalation of saline aerosol (SAL-AER) as the control condition. 1\venty-four hours later (Day 1), the animals underwent airwayresponsivenesstesting or bronchoalveolar lavage, after which they werekilled by anesthetic overdose.In Experiment 2 only (see below) the CAP-AER or SAL-AER inhalation exposures were preceded in all animals by chronic capsaicin pretreatment, a regimen that eliminates neuropeptides from sensory c-fibers and thus prevents sensory neuropeptide release upon subsequent c-fiber stimulation.
Aerosol Inhalation (Day 0) Endotoxin-free capsaicin wasdissolvedin ethanol and diluted with normal saline to make a stock 10-2 M capsaicin solution containing 10070 ethanol; 10-3 and 10-4 M capsaicin solutions were made by diluting stock solution in normal saline. Animals were placed into a 26-L clear Plexiglas'" box through which nebulized (Pulmo-Sonic 25; DeVilbiss Co.,
(Received in original form April 1, 1991 and in revised form January 22, 1992) 1 From the Department of Medicine and Pediatrics, University of Chicago, Chicago, and the Department of Medicine,Evanston Hospital, Evanston, Illinois. 2 Supported by Grants HL-02205, HL-41009, HL-32495, HL-35718, HL-46368, and HL-02376 from the National Heart, Lung, and Blood Institute and by grants from the American Lung Association. 3 Correspondence and requests for reprints should be addressed to Julian Solway, M.D., Box MC6026, University of Chicago, 5841 S. Maryland Ave., Chicago, IL 60637. • Recipient of a grant from the National ChengKung University Hospital, 'Iainan, Taiwan.
149
SENSORY NEUROPEPTlDES AND AIRWAY REACTIVITY
Somerset, PA), capsaicinsolutions wereblown at 8 L/min. A mixing fan ensured uniform aerosol concentration within the exposure box. Awake animals spontaneously breathed capsaicin aerosols for 10 min (10-4 M for 5 min, followed by 10-3 M for 5 min), or until the development ofvisual evidence oflabored breathing, which was identified by the presence of repeated cough, deep chest wall indrawing, tachypnea, and/or cyanosis. Each animal was immediately returned to its cage, into which supplemental oxygen was blown for 5 min, by which time these signs always resolved. Every animal developed labored breathing within the lO-min exposure period; animals that received capsaicin aerosol on Day 0 are hereafter designated"CAP-AER" animals. Control animals were treated identically, except that for 10 min they breathed nebulized normal saline containing I % ethanol (i.e., capsaicin vehicle)instead of capsaicin solution. Animals that received saline aerosol on Day 0 are hereafter designated "SAL-AER" animals.
Animal Instrumentation and Preparation (Day 1) Each guinea pig was anesthetized intraperitoneally with sodium pentobarbital (75 rng/kg). A 13-gauge cannula was inserted through a high cervical tracheostomy, and the animal was mechanically ventilated at "quietbreathing" settings (6 ml/kg, 60 breaths/min) with humidified air at room temperature. As previously described (8-11), inspiratory and expiratory limbs of the ventilator circuit were attached to the tracheal cannula through a 2-cm common segment to minimize conditioning of inspired gas during subsequent dry gas hyperpnea challenges(seebelow). One carotid artery and one internal jugular vein werecannulated with PE-6O tubing. Each animal was then placed in an isothermal constant-mass whole-body plethysmography (1O.1-L displacement). Changes in lung volume were reflected as changes in plethysmographic pressure measured relative to a 20-L reference chamber with a differential pressure transducer (Honeywell MicroSwitch 163PCOlD36; Honeywell, Denver, CO). Airflow was obtained by electrical differentiation of the volume signal. Transrespiratory pressure (the pressure difference between tracheal pressure and plethysmograph pressure) was measured with a differential pressure transducer (Honeywell MicroSwitch 143PCOlD). Respiratory volume excursion, flow, transrespiratory pressure, and blood pressure were recorded on a chart recorder (Gould 2800S;Gould Instruments, Cleveland, OH). Respiratory system resistance (Rrs) was calculated from differences of flow and of transrespiratory pressure at isovolume points (8-11), an adaptation of the method of Goldman and coworkers (12). Each guinea pig received propranolol (1 mg/kg) intravenously to minimize potential changes in bronchoconstrictor response resulting from changes in levelsof circulating catecholamines (8-11); no change in Rrs resulted from this infusion. We
have previously shown that such beta-adrenergic blockade does not influence the magnitude of hyperpnea-induced bronchoconstriction in guinea pigs (10, 11).
Intravenously Administered Bronchoconstrictor Agonist Responsiveness Testing (Day 1) A single deep inflation of three times the "quiet-breathing" tidal volume was given to establish a fixed volume history, and baseline Rrs was obtained 1 min later. Then, intravenousbolus infusions (0.8 ml/kg) ofbronchoconstrictor agonist were sequentially administered in increasing half-log steps flushed by 0.2 ml normal saline, and Rrs measurements were obtained at the peak airway pressure after each such infusion. Each subsequent agonist dose was administered immediately after peak response from the previous dose was achieved; return to baseline Rrs was not allowed until all doses wereadministered. For intravenous administration, acetylcholine (ACh) and NKA weredissolved in normal saline; capsaicin was dissolved in normal saline containing < 1% ethanol.
Dry Gas Hyperpnea Stimulus-Response 'Testing (Day 1) Animals breathed fully humidified 50% oxygen in air at room temperature at the "quietbreathing" ventilator settings. A single deep inflation to three times tidal volume was given to establish a fixed volume history; 1 min later baseline Rrs was measured, and then isocapnic dry gas hyperpnea was mechanically imposed for 10min, using dry 95% 0,5% CO, at room temperature as inspired gas (8-11).Ventilator frequency was increased to 150 breaths/min, and tidal volume was adjusted to 3 ml. After 10 min of hyperpnea, the animal was returned to "quite-breathing" of humidified 50% oxygenin air. Rrs changes were monitored throughout hyperpnea and the posthyperpnea "recovery"period. The animal was allowed to recover to baseline Rrs, and the entire dry gas hyperpnea challenge/recovery sequence was repeated three more times, using 4, 5, and 6 ml tidal volumes during hyperpnea.
Bronchoalveolar Lavage (Day 1) After instrumentation and preparation as described above, animals were killed by exsanguination through the carotid artery; no physiologic studies were performed. The lungs, heart, and mediastinal structures were removed en bloc, and the esophagus was dissected away to reveal the main bronchi clearly.The trachea was sectioned above the main carina, and the right main bronchus was intubated with a polyethylene catheter that was tied securely in place. Normal saline (3 ml at 37° C) was instilled into the right lung, allowed to dwell 20 s, and then withdrawn. Lavage was repeated with another 3 ml, and the
returned fluid was combined with the first lavagate. Recovered bronchoalveolar lavage (BAL) fluid volume was measured, and the fluid was placed immediately on ice. Cells were sedimented by centrifugation at 200 x g for 10 min at 4° C. Supernatant was discarded, and cells were resuspended in 2 ml Hanks' balanced salt solution with 1% gelatin. Cell count was performed on a hemacytometer, and differential count of 300 cells was performed on cells transferred to a glass slide using a Cytospin centrifuge (Shandon, Pittsburgh, PA) and stained with eosin and methylene blue.
Chronic Capsaicin Pretreatment (Days -11 and - 4) Animals that were studied in Experiment 2 (see below) were depleted of sensory neuropeptides prior to aerosol inhalation at Day 0: Sensoryneuropeptide depletion was accomplished as previously detailed (10),bY"chronic" subcutaneous capsaicin pretreatment on two occasions; these pretreatments took place 11 and 4 days prior to aerosol inhalation (Days 11 and 4, respectively). Prior to capsaicin administration on these two days, guinea pigs wereanesthetized intramuscularly with ketamine (50 rng/kg) and xylazine (0.1 mg/kg) and treated intra peritoneally with aminophylline (10mg/kg) and subcutaneously terbutaline (0.1 mg/kg) to blunt severe bronchoconstriction caused by subcutaneously administered capsaicin. On Day 11, capsaicin (50 mg/kg) was administered subcutaneously, and a larger dose (100mg/kg) as given on Day 4. Oxygen was blown into the cage on each occasion until the animal recoveredfrom anesthesia.
Experiments Experiment 1. Intravenously administered ACh andhyperpnea responsiveness. In order to evaluate the effect of capsaicin-aerosolinduced endogenous neuropeptide releaseupon cholinergic airway responsiveness, CAPAER (n = 8, 533 to 620 g) and SAL-AER (n = 8; 550 to 610 g), animals were exposed to capsaicin aerosol or to saline aerosol (as described above) on Day O. On Day 1 each animal underwent intravenously administered ACh responsiveness testing (dose range, 10-to 10-7 mol/kg), followed 40 min later by dry gas hyperpnea stimulus-response testing. Experiment 2. Intravenously administered
ACh and hyperpnea responsiveness after chronic capsaicin pretreatment. As shown in RESULTS, the CAF-AERanimals in Experiment 1 exhibited exaggerated bronchoconstrictor responses to intravenously administered ACh compared with SAL·AER animals. To test whether the altered responsiveness induced by capsaicin aerosol inhalation depended upon intact sensory nerve function, Experiment 1 was repeated in additional guinea pigswhose airway sensorineural function was ablated by chronic capsaicin pretreatment (13). Thus, all animals studied in Experiment2 werechronically capsaicin-pretreated on Days
150 ::J (II
HSIUE, GARLAND, RAY, HERSHENSON, LEFF, AND SOLWAY
40
::J (II
