INTRODUCTION

Top Abstract Introduction Methods Results Discussion References

ACUTE AND PROLONGED EXPOSURE to moderately high ambient concentrations of ozone has multiple deleterious effects on the lungs, including decrements in pulmonary function, cough, chest tightness, airway inflammation, airway hyperreactivity, increases in bronchial epithelial permeability, impaired ciliary clearance, and exacerbations of asthmatic symptoms, especially in children. Specifically, exposure to ozone causes humans and animals to develop rapid shallow breathing, chest tightness, a decrease in forced vital capacity (FVC) and forced expiratory volume in 1 s (FEV₁), and an increase in airway responsiveness (19, 25). These perturbations in breathing pattern and lung function have been attributed, in part, to activation of the lung rapidly adapting receptors (RARs). This attribution is based on the findings that the RARs are stimulated by ozone (5, 11) and evoke some of the same reflex effects on lung function and breathing pattern as those effects evoked by ozone exposure (4). Furthermore, the ozone-induced changes in breathing pattern and lung function are reduced by interruption of nerve conduction, by either inhalation of lidocaine (12) or cooling the vagus nerve to 7°C (24).

Much less is known regarding the consequences of repeated daily exposure to ozone on RAR responsiveness and the associated reflex effects in lung function and breathing pattern. Of related interest, however, are the observations that, with repeated daily exposure to ozone, the changes in lung function and breathing pattern caused by the initial ozone exposure diminish (6, 9a, 19, 25). These findings raise the possibility that RAR responsiveness may also decay with repeated daily exposure to ozone. The consequences of a change in the generalized excitability of RARs and their responsiveness to multiple stimuli after repeated daily exposure to ozone might have far-ranging consequences with respect to airway responses to air pollutants, chemical mediators, and pulmonary edema. Accordingly, this study examined consequences of a 1-wk period of repeated daily exposure to ozone on the generalized responsiveness of RARs. To determine whether the change in RAR responsiveness was global or limited to a specific stimulus, we measured changes in RAR impulse activity in response to four stimuli: substance P, methacholine, hyperinflation, and removal of positive end-expiratory pressure (PEEP). We chose substance P because it is a potent stimulus for RARs, most likely by increasing microvascular leak (3). The mechanism by which methacholine stimulates RARs has been indirectly attributed to bronchoconstriction (4, 13) but may also include increased microvascular leak (18). We chose hyperinflation and removal of PEEP because they represent two disparate mechanical stimuli. Global changes in RAR responsiveness to all stimuli would suggest that the ozone exposure altered some property of the sensory ending or its surrounding milieu. On the other hand, changes in RAR responsiveness limited to a specific stimulus would suggest a specific interaction between the stimulus and the sensory ending.

We also examined the effects of week-long ozone exposure on baseline lung function and airway responsiveness to substance P and methacholine. Finally, to determine whether the ozone-induced changes in RAR activity were mediated by changes in lung stiffness (5), we evaluated the temporal relationship between changes in dynamic compliance (Cdyn) and RAR activity.

	METHODS

Top Abstract Introduction Methods Results Discussion References

Week-long exposure to ozone or filtered air. Male Hartley guinea pigs ~4 wk old (Charles River Laboratories, Raleigh, NC) were allowed to acclimate in chambers for 7-10 days before beginning exposures. Within the chambers, the animals were housed in open mesh, stainless steel cages. Ventilation with filtered air (FA) was 30 chamber-volume changes/h. The guinea pigs were then randomly assigned to be exposed either to 0.5 parts/million (ppm) ozone or to FA for 8 h/day from 10 AM to 6 PM on 7 consecutive days. Ozone was generated from medical-grade oxygen by using an electric discharge ozonizer (Erwin Sander Elektroapparatebau, Uetze-Eltze, Germany) and was mixed with FA. The ozone concentration was measured with an ultraviolet ozone monitor (Dasibi, Glendale, CA). Calibration of the monitor was performed according to the national reference method and was traceable to National Institute of Standards and Technology absolute ozone photometer serial number 4 located at the California Air Resources Board Quality Assurance Standards Laboratory in Sacramento, CA. The concentration of ozone in the exposure chamber was measured every 24 s, and a proportional feedback control system maintained the desired concentration. The concentration of ozone for all the exposures was 0.506 ± 0.006 (SD) ppm. The animals were studied ~16 h after the end of the last exposure. Care and housing of animals before, during, and after exposure complied with the provisions of the Institute of Laboratory Animal Resources and conformed to practices established by the American Association for Accreditation of Laboratory Animal Care.

Experimental procedures. Each guinea pig was anesthetized with an injection of urethan (1.7 g/kg ip). Adequacy of anesthesia was assessed by a paw-pinch test in which the hindlimb paw was pinched and the animal was monitored for a hindlimb flinch or withdrawal and/or a sudden fluctuation in arterial blood pressure or heart rate. During the experiment, the animals were monitored for spontaneous fluctuations in arterial blood pressure or heart rate, and the paw-pinch test was repeated at least every 30 min. Pentobarbital sodium (4 mg iv) was administered as needed. Catheters were introduced into the jugular vein to administer fluids and drugs and were introduced into the carotid artery to monitor arterial blood pressure and withdraw samples for arterial blood gases. The trachea was cannulated below the larynx, and a catheter was connected to a side port of the endotracheal tube to monitor intratracheal pressure. The guinea pigs were prepared with bilateral pneumothoraxes by incisions made in the chest wall, and the animals were mechanically ventilated (Baby Bird, Palm Springs, CA) at constant peak inspiratory pressure with oxygen-enriched humidified air. The ventilator PEEP was set at 2 cmH₂O, and the peak inspiratory pressure was adjusted to provide an initial tidal volume of 8 ml/kg. The rate was set initially at 35 breaths/min. Arterial blood gases and pH were maintained so that the pH was between 7.35 and 7.45 and the arterial PCO₂ (Pa_CO2) was between 35 and 45 Torr, by adjusting the ventilator rate and by infusing sodium bicarbonate. Body temperature was maintained at 37°C with a water blanket. After tracheal intubation was performed, the animals were paralyzed with gallamine (3 mg/kg every hour as needed).

Experimental protocol. The investigators were blinded as to the exposure treatment of the 30 animals studied. At the beginning of the protocol, the lung was hyperinflated for 3 s, and the RAR response to hyperinflation and the adaptation index were determined. At the end of the hyperinflation, data on RAR activity, Cdyn, RL, and arterial blood were then collected over a 2-min baseline period. Substance P (Research Biochemicals International, Natick, MA) was then administered at one-fourth-log intervals (0.89, 1.57, 2.78, and 4.95 nmol/kg iv) every 4 min. On the basis of our previous work, these doses were at the low end of the dose-response curve for stimulation of RARs (2). Low doses were administered to minimize systemic effects of substance P. In pilot studies, we determined that ventilation was adequate during the substance P dose-response curve. RAR activity was collected over a 2-min baseline period, then over a 30-s period immediately after each dose of substance P was administered, and the average values for the 30-s period were reported. A 30-s period was chosen because the RAR response was usually complete within that time frame (see Figs. 1, 2, and 5). Lung function was reported as the highest RL and lowest Cdyn that occurred during a 5-s bin in the 4 min after each dose of substance P was administered. The lowest mean arterial blood pressure was determined after each dose of substance P. The lungs were then hyperinflated to a pressure of 22 cmH₂O for 3 s to maintain a constant lung volume history and to reverse any atelectasis that may have occurred. The hyperinflation was repeated as needed until Cdyn returned to baseline. RAR activity was then recorded for a 2-min baseline period and a 9-s period after withdrawal of PEEP.

Statistical analysis. Body weights, blood gases, adaptation indexes of RARs, RAR response to hyperinflation, and initial RAR activity, RL, and Cdyn of the animals in the ozone- and the FA-exposed groups were compared with an unpaired Student's t-test. The substance P and methacholine dose-response curves for RAR activity, Cdyn, RL, and mean arterial blood pressure were analyzed with a repeated-measures analysis of variance, with exposure as a between-subject effect and dose level as a within-subject effect (SAS, SAS Institute, Cary, NC). A repeated-measures analysis of variance was also used to analyze the RAR response to withdrawal of PEEP. For the analysis of variance tests, data for RAR activity and RL were log transformed because variances differed by more than threefold. Statistical significance was claimed when the probability of a type I error was <0.05. All values are expressed as means ± SE.

	RESULTS

Top Abstract Introduction Methods Results Discussion References

RARs were recorded in 30 guinea pigs: 14 animals were exposed to ozone and 16 animals to FA. The weights of the FA-exposed group (417 ± 10 g) did not differ from weights in the ozone-exposed group (408 ± 6 g, P = 0.45). The pH in the FA group (7.39 ± 0.02) did not differ from that in the ozone-exposed group (7.38 ± 0.01, P = 0.67). The arterial PO₂ (Pa_O2) was slightly higher in the FA-exposed group (176 ± 3 Torr) than in the ozone-exposed group (161 ± 6 Torr, P = 0.04). The Pa_CO2 in the FA-exposed group (40.0 ± 2.1 Torr) was not different from that in the ozone-exposed group (41.7 ± 1.2 Torr, P = 0.50). All animals received injections of substance P. Four animals exposed to ozone and five animals exposed to FA also received injections of methacholine. The adaptation index (14) of the RARs also was not different between the groups, averaging 0.98 ± 0.01 in the FA-exposed group and 0.93 ± 0.03 in the ozone-exposed group (P = 0.13). All RARs were localized to the lung.

RAR responses. Substance P produced a dose-dependent increase in RAR activity that was greater in the ozone- than in the FA-exposed animals. Examples of the effects of the highest dose of substance P on RAR activity from guinea pigs exposed to either FA or to ozone are shown in Fig. 1. In Fig. 1A, from a FA-exposed animal, substance P activated the RAR and did not change blood pressure that had been lowered near maximally by the lowest dose of substance P. In Fig. 1B, from an ozone-exposed animal, the substance P produced a larger activation of the RAR and slightly lowered arterial blood pressure.

View larger version (34K): [in this window] [in a new window]   Fig. 1.   Examples of effects of hyperinflation and substance P on impulse activity of a rapidly adapting receptor (RAR) and on arterial blood pressure (ABP) from a filtered air (FA)-exposed guinea pig and from a guinea pig exposed to 0.5 parts/million (ppm) ozone for 8 h/day for 7 days. A: in FA-exposed guinea pig, RAR was activated by substance P (4.95 nmol/kg), and ABP that had been lowered by previous doses of substance P was not further affected. Inset: (2.5-s duration) 4 action potentials (AP) in response to maintained hyperinflation (22 cmH2O) with rapid adaption. B: in ozone-exposed guinea pig, same dose of substance P activated RAR to a greater extent, and ABP was slightly decreased. Inset: (2.5-s duration) 13 APs in response to maintained hyperinflation (22 cmH2O) with rapid adaption. , Injection of substance P.

View larger version (27K): [in this window] [in a new window]   Fig. 2.   Examples of effects of hyperinflation and methacholine on impulse activity of RAR and on ABP from FA-exposed guinea pig and from guinea pig exposed to 0.5 ppm ozone for 8 h/day for 7 days. A: in FA-exposed guinea pig, methacholine (42.2 nmol/kg) activated RARs and lowered ABP. Inset: (3.4-s duration) shows 20 APs in response to maintained hyperinflation (22 cmH2O), with rapid adaption. B: in ozone-exposed guinea pig, same dose of methacholine activated RAR to a greater extent and ABP was decreased. Inset: (3.4-s duration) shows 28 APs in response to maintained hyperinflation (22 cmH2O) with rapid adaption. , Injection of methacholine.

View larger version (13K): [in this window] [in a new window]   Fig. 3.   Summary of substance P and methacholine effects on RAR from guinea pigs exposed to FA or to ozone. Left: RARs from 14 ozone-exposed guinea pigs were more responsive to substance P than were those from 16 FA-exposed guinea pigs (*P = 0.002, exposure effect). Right: RARs from 4 ozone-exposed guinea pigs were also more responsive to methacholine than were those from 5 FA-exposed guinea pigs (*P = 0.03, exposure effect).

Cardiopulmonary mechanics responses. The grouped data for the airway responses to iv substance P and methacholine in the FA-exposed and the ozone-exposed guinea pigs are shown in Fig. 4. As shown in Fig. 4A, baseline Cdyn did not differ in the FA- and ozone-exposed groups. Cdyn was 0.479 ± 0.013 ml/cmH₂O in the FA-exposed group and 0.456 ± 0.009 ml/cmH₂O in the ozone-exposed group, P = 0.17. Substance P produced a dose-dependent decrease in Cdyn (P = 0.0001, dose effect). The substance P-induced decrease in Cdyn was slightly greater in the ozone-exposed animals than in the FA-exposed animals (P = 0.04, exposure effect).

View larger version (19K): [in this window] [in a new window]   Fig. 4.   Summary of substance P and methacholine effects on dynamic compliance (Cdyn) and on lung resistance (RL) from guinea pigs exposed to FA or to ozone. A, top: substance P decreased Cdyn slightly more in ozone- than in FA-exposed guinea pigs (*P = 0.04, exposure effect). A, bottom: substance P increased RL to same extent in ozone- and FA-exposed guinea pigs (P = 0.65, exposure effect). B, top: methacholine decreased Cdyn to same extent in ozone- and FA-exposed guinea pigs (P = 0.30, exposure effect). B, bottom: methacholine increased RL to same extent in ozone- and FA-exposed guinea pigs (P = 0.38, exposure effect).

1

1

1

1

Timing of the changes in RAR activity and Cdyn. Figure 5 shows the time course of the effects of substance P and methacholine on RAR activity superimposed on the associated Cdyn 15 s before and 30 s after injection. Cdyn was averaged over 5-s intervals, and RAR activity was averaged over 3-s intervals. In both FA- and ozone-exposed animals and for both substance P and methacholine, RAR activity began to increase at 6-9 s after injection, at a time when Cdyn was not changed. The peak change of RAR activity occurred at 9-15 s after injection, at a time when there was only a 3-4% decrease in Cdyn. The maximal decrease in Cdyn occurred at 20-25 s after injection, at a time when the RAR activity was almost back to baseline.

View larger version (42K): [in this window] [in a new window]   Fig. 5.   Time course for changes in RAR and Cdyn activity in 15 s before and first 30 s after injection of either substance P or methacholine. In all panels, values for Cdyn are superimposed on values for RAR activity. Time of injection is indicated as time 0. A, top: injection of substance P (2.78 nmol/kg) in ozone-exposed guinea pigs. A, bottom: injection of substance P (2.78 nmol/kg) in FA-exposed guinea pigs. B, top: injection of methacholine (42.2 nmol/kg) in ozone-exposed guinea pigs. B, bottom: injection of methacholine (42.2 nmol/kg) in FA-exposed guinea pigs. In all instances, change in RAR activity preceded change in Cdyn. RAR activity returned to baseline when change in Cdyn was maximal.

	DISCUSSION

Top Abstract Introduction Methods Results Discussion References

Although the consequences of exposure to ozone on RAR activity and lung function have been previously documented, until now there have been no data on the consequences of repeated daily exposure to ozone on baseline RAR activity or on the capacity of RARs to respond to other stimuli. In addition, the only studies investigating the consequences to lung function of repeated daily exposure to ozone have focused on characterizing the adaptive nature of those responses. The present study focused on the general responsiveness of RARs and lung function (Cdyn and RL) after repeated daily exposure to ozone by specifically examining RAR and lung function responses to stimuli other than ozone. The major finding was that a 7-day period of repeated daily exposure to ozone in guinea pigs exaggerated RAR responses to substance P, methacholine, and hyperinflation but had no important effect on lung function.

Exposure to ozone stimulates lung RARs and, moreover, evokes changes in breathing pattern and lung function that are the same as those evoked by stimulating the RARs. Those changes include cough, rapid breathing, and airway obstruction (6, 9a, 12, 16, 19, 20, 24, 25). Furthermore, the ozone-induced changes in pulmonary function in humans can be abolished by inhalation of lidocaine, which blocks nerve conduction (12). Finally, the ozone-induced rapid shallow breathing, increased RL, and decreased Cdyn can be partially blocked by cooling the vagus to 7°C to block the nerve conduction of RARs (24). These findings suggest that the ozone-induced changes in lung function may be caused, in part, by RAR stimulation.

Interestingly, evidence has been accumulated to suggest that the robust changes in lung function during the initial exposure to ozone may adapt or attenuate with repeated daily ozone exposure. Specifically, Hackney et al. (9) observed that persons exposed to ozone who lived in an area of low ambient-ozone exposure (Canada) developed more cough, substernal discomfort, and decreased FVC than did those who lived in an area of high ambient-ozone exposure (Los Angeles). The investigators also showed that ozone exposure decreased FVC and FEV₁ on the first 2 days of ozone exposure but failed to do so on subsequent repeated daily exposures (9a). This adaptation has been shown for a number of other physiological responses to ozone such as airway responsiveness (6, 20) and symptoms of cough and chest tightness (19). Thus the findings suggest that, although the initial exposure to ozone evokes changes in lung function, breathing pattern (including cough), and sensation, with continued repeated daily exposure, the system adapts and the ozone-evoked changes are less.

In summary, the findings that RAR stimulation appears to partially mediate the ozone-induced changes in pulmonary function and breathing pattern, coupled with the observations that the ozone-induced changes in pulmonary function and breathing pattern decline with repeated daily exposure to ozone, suggest that RAR responsiveness may decline. Thus we predicted that repeated daily ozone exposure for 7 days might decrease general RAR responsiveness to a variety of stimuli. Instead, repeated daily exposure to ozone enhanced RAR responsiveness. This was not a totally unexpected result, however, given that extended exposure to sidestream smoke, another air pollutant and stimulant of RARs, also increases RAR responses to substance P (2). When viewed collectively, the results suggest that, with repeated exposure to airway irritants, such as environmental tobacco smoke or ozone, the RARs may become hyperresponsive.

The most likely mechanisms underlying the enhanced RAR responsiveness include an increase in lung stiffness and an increase in airway microvascular leak. Other possible mechanisms include inflammation and inhibition of neutral endopeptidase, an enzyme that metabolizes substance P and is inhibited by ozone (21) and mainstream cigarette smoke (7). However inhibition of neutral endopeptidase would not explain the increase in RAR activity in response to methacholine or hyperinflation.

With regard to lung stiffness, RARs are stimulated by lung stiffness as measured by decreases in lung compliance. Jonzon et al. (13) showed that reducing lung compliance in steps, by briefly removing PEEP, increased the activity of RARs. In that regard, when RARs were acutely stimulated by ozone, the increased activity was abolished by hyperinflating the lungs to reverse an ozone-induced decrease in Cdyn (5). Thus it was possible that the ozone-induced changes in RAR activity might be caused by an increase in lung stiffness in ozone-exposed lungs. We do not believe that lung stiffness mediated the differences in stimulated RAR activity, because the temporal relationship (Fig. 5) of the changes in RAR activity and Cdyn after each dose of substance P and methacholine showed clearly that the increase in activity of the RARs preceded rather than followed the decrease in Cdyn.

With regard to microvascular permeability, RARs are stimulated by fluid flux produced by a variety of manipulations of Starling forces (23). Ozone exposure increases the concentrations of markers of vascular permeability found in bronchoalveolar lavage (16); this suggests that ozone exposure increases vascular permeability. We have shown that substance P stimulates RARs at least in part by enhancing microvascular leak (3). Methacholine may work by a similar mechanism, because it has been shown to be a vasodilator of the airway circulation and to increase the thickness of the mucosa (18). Thus it is possible that ozone exposure changes the permeability of the microvasculature, which, when stimulated with substance P or methacholine, leaks more fluid, thereby exaggerating the activation of RARs. The cellular mechanism by which microvascular leak stimulates the RARs is unknown, but the mechanism may include the mechanical distortion of the receptors caused by the fluid flux that opens stretch-activated ion channels in the sensory nerve endings, as has been suggested for the carotid baroreceptors (10). Perivascular edema probably does not change RAR activity by causing gross changes in lung function, because increasing left atrial pressure by 5 mmHg stimulates RARs by hydraulically induced perivascular fluid flux without changing peak tracheal pressure during volume ventilation (3). Therefore, if RARs are stimulated by stretch in the microenvironment, a combination of increased fluid (from ozone-induced permeability) and hyperinflation would functionally provide enhanced stretch and thereby increase RAR activity.

Finally, consideration should be given to inflammation as a possible cause of the enhanced RAR responsiveness. Tepper et al. (25) have shown repeated daily exposure to 0.5 ppm ozone initially caused rapid shallow breathing that disappeared by the fifth day of exposure. Interestingly, during the same time period, they saw a progressive pattern of epithelial damage and inflammation in the terminal bronchioles and an increase in protein in the bronchoalveolar lavage fluid. Ongoing inflammation could result in release of leukotriene C₄, prostaglandin F₂ , histamine, and/or bradykinin which have been shown to stimulate RARs (22). If these mediators were present near the RAR receptors, they may have enhanced their baseline activity as well as their increased responsiveness to stimuli. It should be noted that, although it was not statistically significant (P = 0.10), the baseline activity was twofold higher in the ozone-exposed group (0.24 ± 0.07 impulses/s) compared with the FA-exposed group (0.10 ± 0.04 impulses/s). In fact, before the methacholine dose-response curves, the difference in baseline activity between the exposure groups was statistically significant. This increase in baseline activity may have contributed to the general excitability and enhanced responsiveness of the RARs to substance P and methacholine.

With regard to lung function, week-long exposure to ozone did not cause a change in baseline function, nor did it cause an important change in lung responsiveness to substance P or methacholine. The enhancement in the substance P-induced effect on Cdyn was quite small and probably not physiologically important (Fig. 4). Our findings agree with studies in humans that show that 3-5 days of exposure to 0.4-0.5 ppm ozone (2-3 h/day with exercise) do not change baseline airway resistance or FEV₁ (6, 8, 19). In contrast, the only other study in guinea pigs (15) showed that exposure to 0.4 ppm ozone for 1 wk decreased quasistatic compliance of the respiratory system by ~25% (15). In that study, however, the guinea pigs were exposed to ozone continuously, an exposure pattern that does not mimic the human exposure pattern. Furthermore, lung function was measured immediately after the exposures and thus may have represented the acute effects of exposure to ozone. Regarding the effect of repeated daily ozone exposure on airway hyperresponsiveness, although acute exposure to ozone increases airway hyperresponsiveness (6, 20), in most studies, repeated daily ozone exposure does not (6, 17, 20). Our study similarly showed no effect of repeated daily ozone exposure on either substance P or on methacholine-induced changes in RL and only a small, most likely inconsequential effect on substance P-induced decreases in Cdyn.

In conclusion, repeated daily exposure of guinea pigs to ozone for 1 wk exaggerated the responsiveness of RARs to multiple stimuli, through mechanisms other than an increase in lung stiffness. The data raise the possibility that daily ozone exposure may increase the excitability of RARs to the extent that associated reflex changes in airway function may be enhanced and contribute to respiratory symptoms associated with ozone exposure.