Increase in markers of airway inflammation after ozone exposure can be observed also in stable treated asthmatics with minimal functional response to ozone

Ozone is a potent oxidant known to induce a variety of respiratory effects, including cough, increased airway reactivity, decrease in lung function and neutrophilic airway inflammation [1]. Recent evidence supports a role for environmental chronic exposure to ozone in the development of asthma and in triggering asthma attacks [2, 3]. Ozone exposure imposes an oxidative burden on the lung both by directly oxidizing biomolecules, thereby generating reactive oxygen species (ROS), and by inducing inflammatory mediator production and release, with activation of inflammatory cells and further release of ROS [4]; this process causes acute and chronic airway damage, which results in bronchoconstriction and bronchial hyperresponsiveness [5].

Many studies have reported an increase in markers of neutrophilic activation in induced sputum or in bronchoalveolar lavage (BAL) after ozone exposure in healthy ad asthmatic subjects [6, 7], as well as an increase in markers of oxidative stress in lung tissue and BAL fluid of animals exposed to high ozone concentrations [8]. There is, however, considerable interindividual variability in the magnitude of pulmonary response, both in terms of functional and inflammatory reaction to ozone exposure, with a large proportion of subjects showing no significant change in airway calibre after controlled experimental ozone exposure in laboratory [9]. No specific determinants of poor functional response to ozone have been demonstrated, apart from age and levels of exposure [10, 11]. Most experimental studies in humans have reported no correlation between functional response and severity of inflammatory response measured in induced sputum or bronchoalveolar lavage fluid [12‐14]. In particular, very few data have been reported in asthmatic subjects.

In humans, increased levels of malondialdehyde and 8-isoprostane have been measured in breath condensate after O3 exposure [15, 16].

Polymorphism of two oxidative stress related enzymes, Glutathione-S-Transferase M1 (GSTM1) and NAD(P)H:quinone oxidoreductase (NQO1), has been associated with an increased susceptibility to ozone exposure injury. There are studies showing that the deleted form of Glutathione-S-Transferase M1 enzyme (GSTM1 null), resulting in a complete lack of its enzymatic function, associated with the pro187 form of NAD(P)H:quinone oxidoreductase (NQO1) enzyme, induces greater acute airway and inflammatory responses to ozone [17, 18].

The aim of our study was to evaluate the effects of ozone exposure on functional and inflammatory airway responses in 23 mild-to-moderate, stable asthmatic subjects, regularly treated with inhaled corticosteroids. We and other authors have demonstrated that inhaled or oral corticosteroids can blunt neutrophilic airway inflammatory response to ozone in asthmatics [19‐21], but have little effect on airway calibre. Therefore, we would like to verify whether, in asthmatic subjects regularly treated with inhaled corticosteroids, a dissociation between functional and inflammatory airway response to ozone can be observed.

Eight (34.7%) of the 23 subjects studied were defined as responders to ozone. No difference was found between responders and nonresponders for the main baseline characteristics (Table 1). Six subjects (26%) had the NQO1wt e GSTM1null genotypic combination.

The main result of this study was that some asthmatic subjects defined as "nonresponders" showed however an increase in airway markers of neutrophilic inflammation and oxidative stress, thus suggesting that they are nonetheless sensitive to the effect of ozone. In nonresponders, eosinophils (but not neutrophils) and IL-8 in sputum and MDA in exhaled breath condensate increased after ozone exposure, despite clinically nonsignificant changes in pulmonary function. The inverse relationship between sputum neutrophils and FEV1 values after exposure was significant only when responders and nonresponders were taken all together.

Only a small subgroup of the 23 asthmatic subjects studied had a functional airway response to ozone, which was associated with airway neutrophil recruitment. This fact may be due to regular pharmacologic treatment with inhaled corticosteroids, which may have blunted the severity of restrictive ventilatory impairment in most patients studied. The data reported in literature about functional and inflammatory airway response to ozone in asthmatic subjects are controversial: while some studies showed a greater inflammatory response in asthmatics compared to normal subjects [28, 29], these data have not been reproduced by other authors [30], although all the patients studied were free from treatment with inhaled corticosteroids. In previous studies, we demonstrated that asthmatic patients with persistent asthma, regularly treated with inhaled corticosteroids, had a lower functional response to ozone [6], and that short-period pre-treatment with inhaled or oral corticosteroids had some effect on the airway response to ozone [19, 20].

In subjects defined as "nonresponders" to ozone, there was however a mild decrease in FEV1, FVC and VC at different time points after ozone exposure and of no clinical relevance; sputum neutrophils, however, did not significantly increase. Despite the lack of a relevant functional and inflammatory response, ozone exposure still induced a significant immediate increase in MDA in exhaled breath condensate, and a later increase in neutrophilic soluble markers in induced sputum in these subjects. This delayed acute inflammatory response to ozone may predispose asthmatic subjects to greater sensitivity to other specific or non specific triggers of acute bronchoconstriction. Indeed, a relationship between peaks of environmental ozone concentrations and asthma symptoms or Peak Expiratory Flow decrease in asthmatics has been reported, with a time lag of 2-3 days [31, 32], even in asthmatic subjects on regular pharmacologic treatment [33]. These observations suggest that regularly treated, well-controlled asthmatic subjects are not completely protected from an increase in airway inflammation and oxidative stress, and this may predispose to subsequent asthma deterioration.

The increase in MDA was observed immediately after ozone exposure but not 6 hours later, when neutrophilic recruitment in the airways was observed. Although no significant correlation was found between the immediate increase in MDA and the late increase in sputum neutrophilc after ozone exposure, we may hypothesize that the oxidative stress induced by ozone in the airway epithelium may be one of several mechanisms of recruitment and activation of neutrophils in the airways.

When both responders and nonresponders were considered together, a significant correlation between changes in FEV1 and changes in sputum neutrophils was observed. This relationship has never been previously shown by other authors, and in general no correlations between functional and biochemical changes induced by ozone have been reported [12‐14]. However, all studies have been performed in healthy volunteers with high airway reactivity to ozone, whereas our study included asthmatic subjects with very mild functional response to ozone and a blunted ozone-induced airway inflammation, probably due to the long-term regular treatment with inhaled corticosteroids.

The increase in sputum eosinophils in nonresponders found in the present study is an original observation, and requires to be confirmed by further investigations. Other authors have demonstrated an increase in sputum or BAL eosinophils or ECP levels, or in urinary EPX in asthmatic or healthy subjects exposed to ozone, but the relationship with functional response was not reported [34, 35]. In a previous experience, we observed that when ozone exposure is performed 24 hours after an allergen challenge that has induced a late response, sputum eosinophil percentages are increased in comparison to the exposure to air, suggesting that ozone may amplify allergen-induced eosinophil airway inflammation [36].

We studied patients with well controlled asthma, after a short-term withdrawal (24 hours) of the regular treatment with ICS, associated with LABA in the majority of them. This short duration of the treatment withdrawal before each exposure allows to avoid a direct effect of ICS and mainly of LABA on the airway response, but it does not exclude a possible interference from the long-term ICS use. It is well known that the effect of ICS on airway cells may persist up to some weeks after the withdrawal. Because a long-term withdrawal of ICS was not possible in our patients, most of them affected by moderate asthma, we preferred to withdraw both ICS and LABA treatment 24 hours before each exposure, in order to minimize and standardize the possible interference of asthma treatment. This strategy has been used previously by other authors [29].

We found no difference between subjects bearing NQO1wt e GSTM1null genotypic combination and those with other genotypic combinations, as regards both functional and biological response to ozone exposure. The few works published on this topic, mainly evaluating ozone-induced functional changes, are controversial. Our results are in agreement with two studies performed either after acute or chronic exposure to ozone [15, 37], which found no difference in lung function between different genotypic combinations. The small number of subjects bearing the NQO1wt and GSTM1null genotypic combination (n = 6) and the different time points chosen in our study might explain the lack of concordance with other studies reporting an ozone-induced increase in oxidative stress markers in this group of subjects [15, 17].

In conclusion, the increase in some markers of airway inflammation and/or oxidative stress can be observed also in asthmatic subjects who do not functionally respond to ozone. The low rate of functional response can be due to the blunting effect of the previous regular pharmacological treatment in this category of well controlled asthmatics.