Authors R. Dut and E. A. Dizdar have contributed equally to this manuscript.
Oxidative stress and its determinants in the airways of children with asthma
Article first published online: 12 NOV 2008
© 2008 The Authors. Journal compilation © 2008 Blackwell Munksgaard
Volume 63, Issue 12, pages 1605–1609, December 2008
How to Cite
Dut, R., Dizdar, E. A., Birben, E., Sackesen, C., Soyer, O. U., Besler, T. and Kalayci, O. (2008), Oxidative stress and its determinants in the airways of children with asthma. Allergy, 63: 1605–1609. doi: 10.1111/j.1398-9995.2008.01766.x
- Issue published online: 12 NOV 2008
- Article first published online: 12 NOV 2008
- Accepted for publication 22 March 2008
- exhaled breath condensate;
Background: There is ample evidence for the existence of a systemic oxidative stress in childhood asthma but relatively little information on the oxidant stress in the airways.
Objective: To determine the extent of oxidant/antioxidant imbalance and describe its determinants in the airways of asthmatic children including asthma severity and the genotype of the antioxidant enzymes.
Methods: One hundred and ten children with mild asthma, 30 children with moderate asthma and 191 healthy controls were included in the study. Exhaled breath condensate (EBC) was collected from all children with EcoScreen®. Levels of malondialdehyde were measured as the indicator of oxidative stress, and of reduced glutathione as the indicator of antioxidant defense. Children were genotyped for the presence of null variants of glutathione S transferase (GST) T1 and GSTM1, and ile105val variant of GSTP1. Risk factors were analyzed with multivariate logistic regression.
Results: EBC contained significantly higher levels of malondialdehyde and lower levels of reduced glutathione in asthmatic children compared with healthy controls (P < 0.001 for each), whereas there was no difference between mild and moderate asthmatics. Multivariate logistic regression identified asthma as the only independent factor contributing to oxidative stress. Genotypes of the antioxidant enzymes had no effect on the oxidative burden.
Conclusions: Asthma is associated with an extremely powerful oxidative stress not only in the systemic circulation but also in the airways.
glutathione S transferase
An imbalance between the oxidative forces and the antioxidant defense systems favoring an oxidative injury has been implicated in the pathogenesis of asthma (1, 2). Oxidative injury leads to increased lipid peroxidation, increased airway reactivity and secretions, production of chemoattractant molecules, and increased vascular permeability (3, 4), which collectively lead to an augmentation of the existing inflammation that is a hallmark of asthma.
There is ample evidence supporting the presence of a systemic oxidative injury in asthma. An increased production of reactive oxygen species was shown for eosinophils and macrophages obtained from the peripheral blood of patients with asthma (5–7). Very recently, in a large group of children, we were able to show that asthma is associated with a very strong systemic oxidative stress that increases in parallel with the severity of the disease (8). The increased oxidative burden was the result both of increased oxidative stress as evidenced by increased malondialdehyde; and of decreased antioxidant capacity as evidenced by the lowered reduced glutathione. From the glutathione S transferase (GST) supergene family (9), GSTP1 val/val genotype at GSTP1 Ile105Val locus was a significant determinant of the degree of oxidant injury in this population (8).
In addition to the systemic oxidative stress, some studies have also investigated the oxidative burden locally within the airways. In this context, inflammatory cells in the airways such as macrophages and eosinophils were shown to produce elevated amounts of reactive oxygen species (10, 11). Kelly et al. (12) in a study involving 20 asthmatics and 20 controls, have found that oxidized glutathione content in bronchoalveolar lavage was higher in asthma patients than in controls whereas reduced glutathione content was similar. In contrast, however, Dauletbaev et al. (13) found that sputum glutathione levels of stable asthmatic patients (n = 10) did not differ significantly from healthy controls (n = 10).
More recently, there has been an increasing interest in the use of exhaled breath condensate (EBC) as a noninvasive method to investigate lung diseases; and studies investigating the oxidant stress in the EBC of asthma patients have been published in the literature. In one of the initial studies, Antczak et al. (14) studied 10 healthy and 21 asthmatic subjects and found elevated levels of hydrogen peroxide and thiobarbituric acid-reactive products in the breath condensate of asthmatic patients. Similarly, Ganas et al. (15), in a study of 50 patients with stable asthma and 10 healthy controls, have shown that total NO2/NO3 levels in EBC are raised in patients with stable asthma and are significantly related to oxidative stress as assessed by hydrogen peroxide concentration. More recently, Shahid et al. (16) have found elevated levels of 8-isoprostane in 25 children with asthma compared with 13 healthy controls. Zanconato et al. (17) have also reported elevated 8-isoprostane levels in asthmatic children but failed to show any difference in stable vs unstable asthma (n = 9).
These studies provided extremely valuable insights for the oxidative insult observed in the asthmatic airways. However, they are uniformly performed in small groups and do not provide information on the determinants of the existing oxidative burden within the airways such as disease severity or the genotypes of the antioxidant enzymes. In addition, there is also very little evidence on the balance of oxidant and antioxidant forces within the airways.
Therefore, in order to determine the oxidant/antioxidant imbalance in a high number of individuals, and define the determinants of this imbalance, we measured a marker of oxidative stress, malondialdehyde, and a marker of antioxidant capacity, reduced glutathione, in the EBC obtained from a large cohort of children, involving those with mild and moderate asthma and healthy controls. In this cohort, we attempted to define the factors affecting the level of local oxidative stress including asthma severity and GST genotypes.
Patients with asthma
Children aged 6–18 years who were diagnosed with asthma at the Pediatric Allergy and Asthma Unit of Hacettepe University, School of Medicine, Ankara, Turkey between October 2005 and September 2006 have been prospectively included in the study. Asthma was diagnosed by a history of intermittent wheezing and the presence of reversible airway obstruction as defined by at least a 12% improvement in final expiratory volume in 1 s (FEV1) following bronchodilator administration, or an abnormal result in metacholine bronchoprovocation test (PC20<8mg/dl). Upon presentation, asthma diagnosis was confirmed and patients were graded for asthma severity by the examining pediatric allergist according to the Global Initiative for Asthma (GINA) guidelines (18). Children with mild asthma did not receive any controller medication. Children who had an asthma exacerbation or upper or lower airway infection within the last four weeks and those with a systemic disorder other than asthma were excluded from the study. Spirometric measurements, total immunoglobulin (Ig) E, and eosinophil counts were obtained, and skin testing was performed with a battery of 30 aeroallergens on the upper back of the children. Reactions with an induration >3 mm that of the negative control were considered positive, and children with at least one positive test result were considered atopic. DNA was extracted from whole blood by standard techniques.
The control group was composed of white Turkish school children who presented between October 2005 and September 2006 to the outpatient department of Hacettepe Children’s Hospital. They presented for reasons such as minor trauma or for their regular follow-up. They responded negatively to an established and validated asthma questionnaire (19), never had any diagnosis of asthma or allergic bronchitis by a physician, and never had any history of wheezing. They all had normal pulmonary function tests. All children underwent skin prick testing and had their total IgE measured in serum. All children in the asthma and control groups were genotyped for GSTM1 and GSTT1 (wild-type vs null) genotypes and for the presence of the single nucleotide polymorphism (SNP) in codon 105 (ile105val) of GSTP1.
All study procedures were performed in accordance with a protocol previously approved by the Ethics Committee of Hacettepe University. All parents provided written informed consent for the study procedures.
Study procedures and measurements
Collection of the EBC
A commercially available device, EcoScreen (Jaeger, Wurzburg, Germany), was used for collection of EBC as previously described (20). In this collection system, exhaled air enters and leaves the chamber through a two-way, nonrebreathing valve in which inspiratory and expiratory gas admixture is prevented, and saliva filtered. Expiratory air flows through a lamellar condenser that is surrounded by a cooling cuff. The interior temperature is −15°C or cooler. Immediately upon collection, the samples are frozen on dry ice and stored in −80°C until measurement. All measurements were performed within 6 months after collection.
EBC levels of reduced glutathione were measured by high-performance liquid chromatography (HPLC) with fluorescence detection using a recent method validated for the evaluation of glutathione in biologic samples at the femtomole level (21). All derivatized samples were filtered before HPLC. Samples (20 μl) were injected onto a reverse-phase HPLC column (C18-5 μm; 250 × 4 : 6mm; Phenomenex, CA, USA) kept at 30°C, with a mobile phase composed of 50 mM of sodium acetate buffer (pH 6.20)/acetonitrile (85/15, v/v). The detector was a spectra system FL3000 fluorescence detector (ThermoQuest, MT, USA). Fluorescence detector excitation and emission wavelengths were 340 and 420 nm, respectively. Data were collected and analyzed with a PC1000 software (ThermoQuest).
EBC malondialdehyde level was used as an indirect measure of oxidative stress using an HPLC-based method as previously described (22). The person performing the reduced glutathione and malondialdehyde assays was blind to the group to which the patient belonged. All samples were run in duplicate and triplicate.
Validation of reduced glutathione and malondialdehyde levels in the EBC
In order to validate the measurements in the exhaled breath, we have taken a few different approaches. First, we measured inter- and intra-assay coefficients of each measurement. Second, to assess the linearity of the assay, five samples were spiked with high concentrations of glutathione and malondialdehyde to produce samples with values within the dynamic range of the assay. Third, as the duration of storage may affect the redox state, we have measured the same mediators in the same samples of EBC almost 2 years after the initial analysis. After the first analysis, the same samples were again frozen at −80°C until they were thawed after 2 years for the second measurement. The measurements were done in 24 randomly selected samples. This analysis had 80% power to detect a 0.5 nmol/l of difference between the two measurements.
Spirometry (2130 Spirometer; Sensor Medics Co, Yorba Linda, CA, USA), IgE measurements, and eosinophil counts were done according to standard procedures, as previously described (8).
A multiplex polymerase chain reaction (PCR) method with the β-globin gene serving as the positive control was used to determine the wild or null genotype at GSTM1 and GSTT1 genes; and GSTP1 ile105val genotype polymorphism was determined by PCR-restriction fragment length polymorphism (RFLP) as previously described (8).
Statistical analyses were performed with SPSS 11.5 (spss, Chicago, IL, USA) for Windows. All data including age, eosinophil count, IgE levels, and FEV1 showed a non-normal distribution; therefore, data are given as medians and interquartile ranges, and all statistical comparisons were performed using nonparametric Mann–Whitney U-test or anova on ranks. For pair-wise comparisons of the non-normally distributed data, Mann–Whitney U-test with Bonferroni correction was used. All comparisons were adjusted for possible confounding factors, including age and sex and asthma severity where appropriate. A P-value <0.05 was considered significant.
Logistic regression was performed to establish the factors that were associated with oxidative stress and asthma severity. We examined the following variables: age, sex, age of onset, skin test positivity, IgE level, eosinophil counts, smoke exposure, pet ownership, family history of atopic diseases, asthma diagnosis, asthma severity, and polymorphisms at GSTM1, GSTT1, and GSTP1 genes. Factors that showed a significant association in the univariate analysis were included in the multivariate logistic regression to determine the variables showing an independent association. In this model, we treated the determinants of oxidative stress as dichotomous variables using a median split because various transformation techniques failed to normalize the malondialdehyde and reduced glutathione levels to allow the use of linear regression models. A two-sided P-value <0.05 was considered significant.
One hundred and ten children with mild asthma, 30 children with moderate asthma, and 191 healthy children were included in the study. Characteristics of the study population are summarized in Table 1. As expected, there were significant differences in IgE levels, skin test positivity, FEV1 values, and eosinophil counts among the three groups.
|Controls (n = 191)||Mild asthma (n = 110)||Moderate-severe asthma (n = 30)||P *|
|Female (%)||99 (52)||39 (36)||10 (33)||0.01†|
|Male (%)||92 (48)||71 (64)||20 (67)|
|Age‡||10.7 (8.1–13.1)||9.3 (8.1–12.4)||10.2 (8.7–11.7)||> 0.05§|
|Immunoglobulin E‡||35 (17–73)||87 (47–255)||147 (46–538)||0.001§|
|Positive (%)||31 (16)||54 (49)||10 (33)||< 0.001†|
|Negative (%)||160 (84)||56 (51)||20 (67)|
|Forced expiratory volume in 1 s (%)‡||101 (91–108)||95 (89–103)||78 (73–92)||0.001§|
|Eosinophils/mm3‡||146 (99–274)||219 (131–455)||284 (162–582)||0.001§|
Malondialdehyde and reduced glutathione levels in the EBC
There was a highly significant increase in the oxidant stress in the airways (increased malondialdehyde in the EBC) and a similar decrease in the antioxidant defense (decreased reduced glutathione) in patients with mild asthma compared with controls (Fig. 1). However, no significant difference was observed between patients with mild and moderate-severe asthma (Fig. 1).
In support of this finding, multivariate logistic regression analysis showed that both reduced glutathione [odds ratio (OR), 0.17; 95% confidence interval (CI), 0.09–0.33; P < 0.001] and malondialdehyde (OR, 22.0; 95% CI, 1.7–302; P = 0.018) are independently associated with asthma diagnosis but not with asthma severity.
Validation of reduced glutathione and malondialdehyde levels in the EBC
The inter- and intra-assay coefficients of variation for reduced glutathione measurements were 4.93% and 3.45%, and for malondialdehyde measurements 5.30% and 2.23%, respectively. Spiking five samples with high concentrations of glutathione and malondialdehyde showed a recovery rate of 98–103%.
The effect of glutathione transferase genotypes on malondialdehyde and reduced glutathione levels in the EBC
Logistic regression was performed to establish the factors that were associated with oxidative stress and asthma severity. We examined the following variables: age, sex, age of onset, skin test positivity, IgE level, eosinophil counts, smoke exposure, pet ownership, family history of atopic diseases, asthma diagnosis, asthma severity, and polymorphisms at GSTM1, GSTT1, and GSTP1 genes. The analyses were done in the asthmatic group and the control group separately. There was no effect of the genotype on malondialdehyde and reduced glutathione levels in either of the cohorts except a very small and marginally significant difference in the reduced glutathione levels between the null and the wild-type genotype of the GSTM1 within the asthmatic population [7.0 nmol/l (6.2–7.2) in the null genotype vs 6.8 (6.0–7.0) in the wild-type, P = 0.032].
Our study indicates that in addition to the systemic level, there is a very strong oxidative burden at the local level, i.e. in the airways of asthmatic children. The increased oxidative burden in the airways has two components: increased oxidative stress as evidenced by increased malondialdehyde and decreased antioxidant capacity as evidenced by the lowered glutathione.
Although there is ample evidence for the presence of a systemic oxidative burden (5–8), the studies that investigate the local oxidative burden are rare (10–13). Recently, there has been an increasing interest in the use of EBC as a noninvasive method to investigate the diseases of the airways (23, 24). Collection is a simple, safe, and noninvasive technique for assessment of the airways and requires minimal cooperation by the patient. Therefore, it is a very attractive medium for assessing airway especially in young children (25). Studies investigating the oxidative burden in the airways by assessing the mediators have been published. These studies provided initial evidence for the presence of increased markers of oxidant stress in the EBC (14–17). However, these studies have generally been performed in small groups and thus may not be able to detect the factors contributing to the existing oxidative burden. Our study confirms previous observations in a much bigger population of childhood and extends them by showing that, in addition to the ongoing oxidation, there is decreased antioxidant capacity in the airways.
Another novel aspect of our study was to search for the determinants of the oxidative burden within the airways. We examined the variables that had a potential to influence the oxidative stress such as age, sex, age of onset, skin test positivity, IgE levels, eosinophil counts, smoke exposure, pet ownership, family history of atopic diseases, asthma diagnosis, asthma severity, and polymorphisms at GSTM1, GSTT1, and GSTP1 genes. It turns out that the only factor that determines the oxidative stress is the presence of the airways disease, in this case, asthma. In this respect, there seems to be some difference between the systemic circulation and the airways. Although the val105val genotype of the GSTP1 enzyme was a significant determinant of the oxidative burden in the systemic circulation (8), it had no effect on the airways. Our study was not designed to provide any specific answer for this question and therefore the exact cause remains to be determined. However, one factor responsible for this may be the extreme dilution of the mediator concentration in the exhaled breath. EBC contains large amounts of pure water vapor. Therefore, changes in the mediator concentration may vary by a factor of 100 or more (26). In fact, the levels obtained for both malondialdehyde and reduced glutathione fell from the micromole range in the systemic circulation to the nanomole range in the EBC. Therefore, one can conclude that the levels of reduced glutathione and malondialdehyde are decreased by a factor of three log orders in the EBC. Because of this extreme dilution, subtle changes such as those caused by the different genotypes, may become very difficult to detect. The same explanation may also underlie the lack of a difference in the oxidative markers in the airways between mild asthma and moderate-severe asthma which was readily apparent in the systemic circulation. The severity-related difference in the oxidative stress was also much less apparent than the disease-related difference in the systemic circulation (8), and it is possible that this relatively smaller magnitude of difference did not translate into a difference in the diluted EBC. Alternatively, another underlying cause for this observation may be the relatively small number of children with moderate asthma in our study. Our study may simply be too underpowered to detect a severity-related difference. Combining all asthma patients into one group did not change the already strong differences and relationships (data not shown).
Higher levels of malondialdehyde, an end product of lipid peroxidation, show that there is a very strong oxidant attack in asthma. However, although there is a reduction of the antioxidant defense as evidenced by lowered glutathione, our study does not allow one to conclude whether this reduction is the cause of increased consumption of glutathione to counteract the oxidative stress, or whether this is a phenomenon with separate genetic or environmental causes that actually increases the susceptibility to asthma.
Taken together, our study shows that asthma is associated with an extremely powerful oxidative stress not only in the systemic circulation but also in the airways and that EBC is an appropriate medium to measure the extent of oxidative burden in asthmatic children. Any therapeutic approach targeting the oxidative burden in asthma should definitely take this into account because abolishing the oxidative insult in both the systemic and local compartments may be necessary to achieve a full therapeutic effect.
This study was supported by Hacettepe University grant #02 02 101 020.
- 18Global Initiative for Asthma. Global strategy for asthma management and prevention. National Institutes of Health. National Heart Lung and Blood Institute, Revised, 2002.