Malondialdehyde and superoxide dismutase correlate with FEV1in patients with COPD associated with wood smoke exposure and tobacco smoking

Tobacco smoking is the primary risk factor for chronic obstructive pulmonary disease (COPD). However, recent epidemiological studies have established domestic exposure to wood smoke and other biomass fuels as additional important risk factors, characteristic in developing countries. Oxidative stress is one of the mechanisms concerned with pathogenesis of COPD. However, the molecular mechanisms involved in the onset and progress of COPD associated with biomass and specifically that derived from wood smoke exposure remain unknown. We analyzed the relationship between forced expiratory volume in first second (FEV₁) with plasma malondialdehyde (MDA) concentration and activities of superoxide dismutase (SOD), glutathione peroxidase (GPx), glutathione reductase (GR), and glutathione-S-transferase (GST) in COPD patients associated with wood smoke (WSG; n = 30), tobacco smoking (TSG; n = 30), and healthy control subjects (HCG; n = 30). Differences between FEV₁ from WSG and TSG (58 ± 22% and 51 ± 24%, respectively) with HCG (100 ± 6%) were observed (P < 0.01). Plasma MDA concentration was higher in both WSG and TSG (1.87 ± 0.81 and 1.68 ± 0.82 nmol/mL, respectively) compared with HCG (0.42 ± 0.17 nmol/mL; P < 0.01). SOD activity showed a significant increase in both WSG and TSG (0.36 ± 0.12 and 0.37 ± 0.13 U/mL) compared with HCG (0.19 ± 0.04 U/mL; P < 0.01). No differences were shown regarding GPx, GR, and GST activities between COPD and control groups. Inverse correlations were founded between MDA and SOD with FEV₁ in both COPD patients and control subjects (P < 0.001). These results indicate a role for oxidative stress in COPD associated with wood smoke similar to that observed with tobacco smoking in subjects who ceased at least 10 years previous to this study.

Keywords: Biomass; COPD; malondialdehyde; oxidative stress; superoxide dismutase; tobacco smoking

Chronic obstructive pulmonary disease (COPD) is a leading cause of mortality and morbidity worldwide (Lopez et al., [20]). Although tobacco smoking is the highest known risk factor for the disease, exposure to wood smoke and other biomass fuel products have also been described as additional risk factors (Rabe et al., [28]). Biomass is any material derived from living or recently living organism, including animal dung, twigs, grass, crop wastes, wood, and charcoal (Bruce et al., [5]; Diaz et al., [11]), from these biomass forms wood is the main material used by more than a half of people around the world as energetic material, mainly in developing countries (Pérez-Padilla et al., [26]; Ramírez-Venegas et al., [31]). It is important to differentiate biomass from fossil fuels, which are materials derived of biomass from organisms that lived up to 300 million years ago. The major forms of fossil fuels are petroleum, natural gas, and coal. Biomass exposure refers to human inhalation of any gaseous or particulate emissions from biomass burning (Ramlogan, [32]; Morandi et al., [23]), consequently wood smoke exposure consist in the human inhalation of gaseous or particulate emissions from wood burning (Ramlogan, [32]; Morandi et al., [23]).

COPD associated with both wood smoke exposure and tobacco smoking occurs even many years after exposure has disappeared or subjects stopped smoking. Mechanisms for this phenomenon are not known.

The clinical profile of COPD associated with wood smoke exposure and its prognostic factors have recently been described (Ramírez-Venegas et al., [31]). These patients were predominantly female and had less airflow obstruction than tobacco smokers. However, both groups had similar bronchial symptoms, exercise capacity, quality of life, and use of healthcare services and supplemental oxygen. Nevertheless, many questions remain about the pathogenesis and its mechanisms for the development of COPD in comparison with that due to tobacco smoking (Ramírez-Venegas et al., [31]). For instance, there is increasing evidence that oxidative stress plays an important role in the pathogenesis of COPD (Rahman, [29]). In this regard, the presence of both oxidants and antioxidants are of particular value to demonstrate the participation of the oxidative process. Polyunsaturated fatty acids that may undergo lipid peroxidation are found in membranes of cells participating in inflammation; one of the end products of this peroxidation is malondialdehyde (MDA), which is used as biomarker of oxidative stress (Rahman, [29]; Chung and Adcock, [9]). In response to this attack, several enzymatic and nonenzymatic antioxidant defense mechanisms exist in cells; enzymes include superoxide dismutase (SOD), glutathione peroxidase (GPx), glutathione reductase (GR), and glutathione-S-transferase (GST).

The MDA and several antioxidant enzymes have been described as biomarkers of inflammatory response in COPD associated with tobacco smoking (Chung and Adcock, [9]). However, despite observed similarities with the disease associated with wood smoke exposure, there are few studies describing the oxidative and anti-oxidative events in COPD associated with wood smoke exposure. Therefore, in this study we first determined the presence of MDA, SOD, GPx, GR, and GST in females with COPD associated with wood smoke exposure or tobacco smoking and compared them with a group of healthy, female, nonsmoking controls. Results of associations with forced expiratory volume in first second (FEV1) of these subjects were subsequently pursued.

Sixty consecutive women with a clinical and functional diagnosis of COPD associated with wood smoke exposure or tobacco smoke were recruited from the COPD Clinic of the National Institute of Respiratory Diseases in Mexico City from May 2008 to May 2009, according to their willing to participate in this study. All women included in this report had as source of biomass exposure of burning of wood while cooking or heating their homes. The level of exposure to wood smoke or tobacco smoke was determined by a clinical interview using a standardized Spanish version of the American Thoracic Society questionnaire (Menezes et al., [22]) supplemented by additional questions directly related to cooking and warming fuels The main inclusion criterion was a history of daily wood smoke exposure for at least 200 h/years (Pérez-Padilla et al., [26]) or a history of tobacco smoking >10 pack/years. Cumulative exposure to wood smoke was expressed as hour/years, which was derived by multiplying the number of years of cooking with wood stoves by the average daily hours spent in the kitchen. None of the COPD patients was exposed to either wood smoke or tobacco smoke, respectively, during the previous 10 years. Subjects with COPD with a history of wood smoke exposure were matched one-on-one with those subjects with a history of tobacco smoking in regard to % predicted forced expiratory volume in first second (FEV1 % predicted). We excluded those subjects who had both wood and tobacco smoke exposure or had a history of other chronic pulmonary conditions such as asthma, tuberculosis or bronchiectasis. Patients with COPD were clinically stable and without exacerbation for at least 6 weeks prior to the study. A group of 30 healthy females without history of tobacco smoking or wood smoke exposure were recruited as the healthy control group. A brief health questionnaire including questions regarding chronic diseases was administered. The study was approved by the Research Ethics Committee of the INER.

Subjects were evaluated by spirometry both pre- and post-bronchodilation following the procedures recommended by the American Thoracic Society/European Respiratory Society (Celli and MacNee, [7]) with a dry rolling-seal volume spirometer (Sensormedics, Yorbalinda, CA) and using Mexican standard reference equations (Pérez-Padilla et al., [27]). These reference equations are similar to the National Health and Nutrition Examination Survey III values for Mexican-Americans (Hankinson et al., [15]).

Diagnosis of COPD was established according to the history of tobacco smoking or wood smoke exposure and pulmonary function tests with FEV1/forced vital capacity (FVC) <70% (Rabe et al., [28]).

Plasma was obtained from whole blood in subjects for whom HIV status was tested according to subjects' previous consent. Samples were centrifuged at 5000g at 4°C in tubes with EDTA. Plasma was kept at 20°C until analysis.

MDA, as an end product of lipid peroxidation, was measured in plasma using the reactions of 1-methyl-2-phenylindole (1M2P) as previously described (Gérard-Monnier et al., [13]). Briefly, 10-µL aliquots of plasma were mixed in 0.65 mL of a solution of 1M2P (15.4 mM) dissolved in acetonitrile/methanol (3:1). These samples were then mixed with 0.35 mL of 50 mM potassium phosphate buffer, pH 7.4, containing 18 mM HCl. Samples were incubated in a water bath at 45°C for 40 min and centrifuged at 3000g for 5 min. Absorbance of the developed color was measured at 586 nm using a Beckman DU 640 spectrophotometer (Beckman Coulter, Fullerton, CA). MDA concentration was expressed as nmol/mL plasma after converting the average absorbance of the samples using a standard curve of MDA. Association between plasma MDA levels and FEV1 within-subjects from COPD and control groups was performed with Pearson correlation coefficients (r).

Activity of extracellular SOD in plasma samples was measured with a SOD assay kit (#706002; Cayman Chemicals, Ann Arbor, MI) according to the manufacturer's protocol. For analysis, 10-µL serum samples were treated with 190 µL of tetrazolium. The reaction was initiated by adding 20 μL of xanthine oxidase followed by incubation for 20 min at room temperature. Absorbance change was read at 450 nm using a microplate reader. SOD activity was expressed as U/mL, calculated from a standard curve constructed with known amounts of standards processed with samples. One unit of SOD activity was defined as the amount of enzyme needed to exhibit 50% dismutation of the superoxide radical. Association between plasma SOD activity and FEV1 among subjects from COPD and control groups was performed with Pearson correlation coefficients (r).

Plasma GPx activity was assayed by a previously described method (Pedraza-Chaverrí et al., [25]). Reaction mixture consisted of 50 mM potassium phosphate (pH 7.0), 1 mM EDTA, 1 mM sodium azide, 0.2 mM NADPH, 1 U/mL of GR, and 1 mM GSH. One hundred microlitre of the appropriate dilution of plasma was added to 0.8 mL of mixture and incubated for 5 min at room temperature before initiation of the reaction by the addition of 0.1 mL 2.5 mM H2O2 solution. Absorbance at 340 nm was recorded for 3 min and the activity was calculated from the slope of these lines as μmol of NADPH oxidized/min, taking into account that the millimolar absorption coefficient for NADPH is 6.22 mM/cm. Blank reactions with homogenates replaced by distilled water were subtracted from each assay. Results were expressed as units/mL of plasma.

Plasma GST activity was assayed according to the method of Habig et al. ([14]). The reaction mixture consisted of 1.475 mL of 0.05 M phosphate buffer (pH 6.5), 0.2 mL of 1 mM GSH, 0.025 mL of 1 mM 1-chloro-2,4,dinitrobenzene (CDNB) dissolved in dimethylsulphoxide, and 0.3 mL post-mitochondrial supernatant (10% w/v) in a total volume of 2.0 mL. Changes in absorbance were recorded at 340 nm, and enzyme activity was calculated as nmoL of GSH CDNB conjugate formed/min/mL of plasma using a molar extinction coefficient of 9.6 mM/cm. Results were expressed as units of GST/mL of plasma.

Plasma GR activity was measured according to the technique of Carlberg and Mannervik ([6]). A 50-μL aliquot of plasma was added to a cuvette containing KH2PO4 buffer (0.2 M, pH 7.6) plus 0.5 mM EDTA-Na2, 1.25 mM oxidized glutathione and 0.1 mM NADPH. Change in absorbance at 340 nm was monitored for 3 min. Oxidation of 1 pmol of NADPH/min under these conditions is used as a unit of GR activity. GR activity was expressed as units/mL of plasma. GR units are defined as the amount of enzyme catalyzing the reduction of 1 nmol NADPH/min.

Data were expressed as mean ± SD of at least three independent experiments. One-way analysis of variance followed by Dunnett's test that was used to adjust for multiple comparisons between groups. Associations between variables were performed using Pearson correlation coefficients (r). Statistical analyses were performed using SPSS for Windows (Chicago, IL); P < 0.05 was considered significant.

General characteristics of patients formerly exposed to wood smoke and former smokers with COPD and controls are shown in Table 1. Patients exposed to wood smoke were older and shorter than smokers with COPD and healthy controls. Mean exposure to wood smoke was 361 ± 177 h/years, whereas smokers had a mean cumulative consumption of tobacco smoking of 36 ± 23 pack/years. Both pulmonary function tests, FVC (% predicted) and FEV1 (% predicted), showed significant differences between both groups of patients with COPD vs. control group; P < 0.01 (Table 1). FEV1/FVC ratio also showed significant difference with control group (P < 0.05). No significant difference was observed when FEV1 and FVC were compared in patients with COPD secondary to wood smoke in relation to patients with COPD due to tobacco smoke (Table 1). Patients with COPD secondary to wood smoke were older and showed differences with both tobacco smoking patients and control subjects (Table 1, P < 0.01). Height also was statistically different in patients with COPD due to wood smoke exposure. Patients with COPD and exposure to wood smoke are currently nonsmokers.

Table 1.  General and functional characteristics of study subjects.

3 Comparison of age, height, weight, BMI (body mass index), wood smoke, and tobacco smoke exposure, FEV1 (forced expiratory volume in the 1st second), FVC (forced vital capacity), and FEV1/FVC. Date are expressed as means ± SD. One-way ANOVA test followed by Dunnet post hoc test. *P < 0.01 and +P < 0.05 compared with control group. NS, not significant.

There was a significant increase in MDA concentration both in patients with COPD associated with wood smoke exposure and those with a history of tobacco smoking in comparison to control subjects (1.87 ± 0.81, 1.68 ± 0.82, and 0.42 ± 0.17 nmol/mL of plasma, respectively; P < 0.01). However, no significant differences were found among groups with COPD (Table 2 and Figure 1). Furthermore, an inverse correlation (r = 0.47 and P < 0.001) was found between plasma MDA concentration and FEV1 when all subjects were compared (Figure 2).

Table 2.  Parameters of oxidative stress in plasma of patients with COPD and control subjects.

4 Malondialdehyde (MDA) concentration and activities of superoxide dismutase (SOD), glutathione peroxidase (GPx), glutathione-S-transferase (GST), and glutathione reductase (GR) in plasma of COPD and control subjects. MDA concentration is expressed in nmol/mL, whereas enzymatic activities are expressed as U/mL. Date are expressed as means ± SD from three different experiments. One-way ANOVA test followed by Dunnet post hoc test. *P < 0.01 compared with control group. NS, not significant.

Graph: Figure 1. Plasma concentration of malondialdehyde (nmol/mL) in COPD and control groups. One-way ANOVA test followed by Dunnet post hoc test showed significant difference when both WSG and TSG were compared against HCG, *P < 0.01. However, no differences were found between WSG and TSG. HCG, Healthy control group(filled squares); TSG, tobacco smoke group (filled triangles); WSG, wood smoke group (filled circles).

Graph: Figure 2. Relationship between plasma malondialdehyde (nmol/mL) and forced expiratory volume in the first second (FEV1 % predicted) in chronic obstructive pulmonary disease and control subjects. Associations among variables were performed using the Pearson correlation coefficient (r); r = 0.47 and P < 0.001. Figure indicates that control subjects (filled squares) with the higher FEV1 are grouped to the right, whereas subjects with COPD associated with wood smoke (filled circles) and to tobacco smoking (filled triangles) are dispersed to the left, showing that subjects with the lower FEV1 have the higher MDA.

A significant increase was found in SOD activity both in the wood smoke group and the tobacco smoking group in comparison to normal subjects (0.36 ± 0.12, 0.37 ± 0.13, and 0.19 ± 0.04 U/mL of plasma, respectively; P < 0.01). However, no significant difference was found among groups with COPD (Table 2 and Figure 3). Additionally, an inverse correlation (r = 0.54 and P < 0.001) was found among plasma SOD activity and FEV1 when all subjects were compared (Figure 4).

Graph: Figure 3. Plasma superoxide dismutase (SOD) activity (U/mL) in patients with COPD and control groups. One-way ANOVA test followed by Dunnet post hoc test showed significant difference when both subjects in the WSG and TSG were compared against HCG, *P < 0.01. However, no differences were found between WSG and TSG. HCG, Healthy control group(filled squares); TSG, tobacco smoke group (filled triangles); WSG, wood smoke group (filled circles).

Graph: Figure 4. Relationship between plasma SOD activity (U/mL) and forced expiratory volume in the 1st second (FEV1 % predicted) in chronic obstructive pulmonary disease and control subjects. Associations between variables were performed using the Pearson correlation coefficients (r); r = 0.54 and P < 0.001. Figure indicates that control subjects (filled squares) with the higher FEV1 are grouped to the right, whereas subjects with COPD associated with wood smoke (filled circles) and to tobacco smoking (filled triangles) are dispersed to the left, showing that subjects with the lower FEV1 have the higher SOD.

No significant differences were found with any of the GSH-dependent enzymes among groups (Table 2).

The main finding of this study is that levels of MDA and SOD are significantly increased in subjects with COPD associated with wood smoke in comparison to healthy controls and that these markers showed an inverse correlation with FEV1 (Figures 2 and 4). Likewise, these products are similarly elevated in COPD associated with tobacco smoking (Boots et al., [3]).

No differences were observed in FEV1 and FVC when groups of patients with COPD secondary to wood smoke and tobacco smoke were compared (Table 1). An important fact is that patients with COPD secondary to wood smoke showed a significant difference with respect to age compared with tobacco smoking patients with COPD and control subjects (Table 1, P < 0.01). This feature is significant because it is related to the natural history of COPD development with wood smoke. It has been shown that wood smoke characteristically needs a longer time period to develop COPD in comparison to tobacco smokers (Brauer et al., [4]; Naeher et al., [24]). Female subjects with COPD secondary to wood smoke included in this study had the typical clinical and functional characteristics of COPD without a history of tobacco smoking and are a representative sample of the patient population with COPD treated at our clinic. The socio-economical condition of these subjects who were exposed to wood smoke seems to be another factor favouring the COPD development due to the fact that these subjects do not use gas or electric stoves because of poverty and most live in rural areas (Pérez-Padilla et al., [26]; Ramírez-Venegas et al., [31]).

The activity of the GSH-dependent enzymes measured did not show significant differences among the three groups despite the marked tendency to be higher in COPD patients than in controls (Drost et al., [12]; MacNee [21]). Tobacco smoking is the primary risk factor for COPD, and oxidative stress is one of the mechanisms involved in its pathogenesis (Rahman, [29]; Chung and Adcock, [9]). Recent epidemiological studies have established domestic exposure to wood smoke and other biomass solid fuels as additional important risk factors, characteristic in developing countries (Pérez-Padilla et al., [26]). However, cellular and molecular mechanisms involved in the onset and progression of COPD associated with wood smoke remain poorly investigated. Consequently, in this study we analyzed MDA, a marker of oxidative stress, and the activity of the antioxidant enzymes SOD, GPx, GR, and GST in plasma of patients with COPD associated with wood smoke and tobacco smoking.

The potential impact of oxidative stress induced by wood smoke has been shown by using MDA levels, protein carbonyl content and DNA damage in women (Isik et al., [17]; Macnee, [21]; Ceylan et al., [8]; Danielsen, et al., [10]). Although SOD and GSH-dependent enzymes are increased in plasma and sputum of patients with COPD associated with tobacco smoking (MacNee, [21]), neither extracellular SOD activity nor the activity of GSH-dependent enzymes has been previously evaluated in COPD due to wood smoke exposure.

SOD and GSH are potent antioxidants that play an important role in controlling oxidative-mediated stress and inflammation. In this study SOD activity was elevated, and GPx, GR, and GST activities were within normal ranges. In a study by Agacdiken et al. ([1]), GSH levels were also found within the same range observed during exacerbations, suggesting according to these results that oxidant/antioxidant activity is on, not only during exacerbations but also throughout the stable phase of the disease (Drost et al., [12]; Rahman, [30]; Kluchová et al., [19]). Barregard et al. ([2]) reported that healthy humans exposed to wood smoke showed signs of oxidative stress. An interesting finding of this study is that limited exposure to wood smoke (4 h) increases MDA levels immediately and also 20 h after exposure. Other study showed that after exposure to wood smoke, DNA strand breaks significantly decreased and the mRNA levels of oxoguanine glycosylase-1 were significantly increased in peripheral blood mononuclear cells, suggesting that wood smoke causes short-term systemic effects (Danielsen et al., [10]). These data support the possibility that oxidative insult secondary to wood smoke occurs when women are domestically exposed. Therefore, if this attack persists, which occurs during cooking, the inflammatory response may perpetuate. In the long term, this explains the emergence of the disease. It is worth noting that MDA values in our patients were much lower in relation to other studies (Ceylan et al., [8]; Hanta et al., [16]; Joppa et al., [18]); however, several reports also suggest that MDA concentrations may be related to the severity of obstructive impairment in COPD (Joppa et al., [18]). In this sense, our COPD patients had better lung function (mean FEV1 of 58 ± 22%) than that reported by Ceylan et al. (FEV1 35%), and MDA value ranged from 4.78 to 4.34 mol/L, respectively. Nevertheless, despite those differences, our results showed that the higher the concentration of MDA or SOD the lower the FEV1, suggesting a real relationship between oxidative attack and damage to lung function both in subjects exposed to wood and tobacco smoke. The most important contribution of this study is in relation to the implications for pathogenesis of COPD, in particular with oxidative stress when the last exposure occurred at least 10 years prior to this study both for wood smoke or tobacco smoking. This means that the disease progressed and oxidative stress continued even after cessation. This phenomenon is well known in COPD observed in smokers who, despite having stopped smoking, continue with a decline of FEV1. This can be appreciated in one of the large clinical trials (Tashkin et al., [33]) during follow-up, in which more of 70% of smokers had stopped smoking but disease progression persisted. In this sense, in our study the oxidative stress in both group of risk is increased, despite that these women stopped the exposure to wood smoke at least 10 years previously to come at the Institute.

It is not surprising that both wood smoke and tobacco smoke have similar results, in terms of oxidative, stress among them. For instance, in terms of particulate and gases emissions, the combustion of wood and other biomass is qualitatively similar to the burning of tobacco (Bruce et al., [5]), except by the lack of nicotine in the biomass fuels.

In summary, in this study we demonstrated an oxidative phenomenon associated with a significant reduction in FEV1 in subjects who were exposed to biomass, specifically wood smoke, over a span of many years. This is similar to that observed in COPD of former smokers. Female subjects included in this study had a typical clinical and functional picture of COPD without history of tobacco smoking and are representative of the patient population with COPD treated at our clinic.

This work was supported and financed by a grant from the Consejo Nacional de Ciencia y Tecnología (CONACYT) México, project number III-90942. Institution where the work was performed: Instituto Nacional de Enfermedades Respiratorias "Ismael Cosío Villegas" (INER). The authors declare that there are no conflicts of interest related to the article or the research described.