The study included thirty-nine patients who underwent lung resection for non-obstructive peripheral lung tumours or were subjected to double lung transplantation for very severe COPD at Vall d’Hebron University Hospital (Barcelona, Spain). The Hospital's ethics committee approved the study and written informed consent was obtained from all subjects. Pulmonary function was assessed through spirometry according to ERS recommendations21 using standard equipment (Masterlab; Jaeger, Würzburg, Germany).

Subjects were divided into COPD (N=20) and non-COPD patients (N=19). All of them were free of acute lung infections in the last three weeks and none had received chemotherapy before surgery.

The excised lungs or lobes were inflated immediately after surgery at 30cm of water pressure for 1h with 4% (w/v) formaldehyde in 0.1M phosphate-buffered saline solution (pH 7.4). In patients with lung cancer, tissue samples were taken from the non-malignant lung tissue as demonstrated by pathologist examination.

Samples to be analyzed by light microscopy were then immersed in fixative for 24h. After that, 2cm×2cm×0.3cm randomly selected tissue blocks were excised, embedded in paraffin, cut into 4-μm sections and mounted on positive charged slides Starfrost Plus (Menzel-Gläser, Braunschweig, Germany). Four consecutive serial sections from a representative block from each patient were selected for histologic, histochemical and immunohistochemical analyses. Every first section was stained with haematoxylin and eosin (H&E) to reveal lung histology. The second and third sections were reserved for immunolocalization of EC-SOD and fibulin-5. The last section in each set was stained with Verhoeff's haematoxylin to detect pulmonary elastic fibres.22

On another front, after initial fixation, samples to be studied by transmission electron microscopy were cut into 1–2-mm thick pieces and fixed in 1% (w/v) paraformaldehyde, 0.1% (v/v) glutaraldehyde in 0.1M phosphate-buffered saline (PBS) solution for 2h at room temperature. After that, a post-fixation with 1% (w/v) osmium tetroxide in cacodylate buffer was performed for 1h. Embedding was performed in Lowycril K4M resin according to standard procedures.23

Samples to be analyzed by light microscopy were deparaffinised and processed at 121°C in a 2100 Antigen Retriever (PickCell Laboratories, Leiden, Netherland) in Tris–HCl 10mM-EDTA 0.5mM pH 9.0 buffer for EC-SOD slides and in Citrate 10mM pH 6.0 for fibulin-5 slides. After antigen unmasking, sections were washed three times in PBS and submerged in 3% hydrogen peroxide, 10% methanol in PBS for 15min. After washing three times with PBS, nonspecific binding was suppressed by incubation with blocking solution. Following that, samples were incubated at 4°C overnight with polyclonal anti-EC-SOD antibody (1:500, Abnova, Taipei, China) monoclonal anti-fibulin 5 antibody (1:750, Abcam, Cambridge, UK), rabbit polyclonal anti-4-HNE antibody (1:300; Calbiochem, San Diego, CA, USA), rabbit polyclonal anti-MDA antibody (1:400; Abcam Ltd., Cambridge, UK) or with rabbit polyclonal anti-NT antibody (1:90; Upstate Biotechnology, Inc., Lake Placid, NY, USA) in blocking solution. Negative controls were obtained by incubating a slide with rabbit IgG (1:500, ab37415, Abcam, Cambridge, UK) or mouse IgG (1:200, ab37355, Abcam). On the following day, sections were washed in PBS and incubated with biotinylated anti-Rabbit or ant-Mouse IgG antibody (Vector Laboratories; Burlingame, CA) for 1h at room temperature. Immunostaining was performed using the avidin–biotin complex (ABC) immunoperoxidase method (Vectastain Elite ABC kit; Vector Laboratories, Burlingame, CA, USA) with a 3,3′-diaminobenzidine tetrahydrochloride (DAB) reaction. Finally, the slides were counterstained in haematoxylin, dehydrated, and mounted. For analysis of immunostained lung tissue, two independent observers (JGV and JO) assessed the intensity and extent of immunoreactivity of EC-SOD in a double-blind fashion through a semi quantitative score (0=no staining; 1=weak/intermediate staining; 2=strong staining). At least twelve fields of each sample were randomly and systematically sampled to assess EC-SOD and fibulin-5 staining in alveolar interstitium.

Samples to be analyzed by transmission electron microscopy (TEM) were hydrated in a PBS 0.1M glycine 0.1M solution and non-specific binding was suppressed by incubation in 2% ovalbumin in PBS 0.1M. After that, samples were incubated with the primary antibody (anti-EC-SOD, 1:150, Abnova, Taipei, China) in blocking solution for2h. Then, samples were washed three times in PBS 0.1M glycine 0.1M solution and incubated with Goat Anti-Rabbit IgG with 10nm Gold preadsorbed (1:10, ab27234, Abcam) for1h. Finally, samples were washed in PBS 0.1M and jet-washed in 10ml of milli-Q water prior to counterstaining and mounting.

Isolation of total RNA from tissue samples was performed using the AllPrep DNA/RNA Mini Kit (QIAGEN, Hilden, Germany). After reverse transcription of mRNA into complementary DNA (cDNA), qRT-PCR was performed using primer and probe sets for ECSOD and fibulin-5 from TaqMan Gene Expression Assays purchased from Applied Biosystems (Foster City, CA, USA). Data from qRT-PCR reactions was analyzed by the comparative Ct (2-ΔΔCt) quantification method. Relative expression levels of ECSOD and fibulin-5 genes were determined using 18S mRNA as an endogenous control. Three independent experiments were performed and results were expressed as the normalized mean of the relative quantification (RQ) of the tested transcripts. Non-template controls exhibited no signal.

The descriptive statistical analysis included mean±SEM or SD for each parameter and patient group. Data were compared between the two groups using the Chi-square and Mann–Whitney U-tests. The significance level was set at p<0.05. All analyses were performed using Statgraphics Centurion XV (StatPoint, Inc., Warrenton, VA, USA).

The two clinical groups were similar regarding age and body-mass index, with a predominance of males and tobacco smoking in the COPD group (Table 1). As expected, the mMRC dyspnoea score, the number of previous exacerbations in the last year and the use of inhaled corticosteroids were higher in COPD patients. In this group, the microscopic (LM) and macroscopic emphysema (MED) degree was increased and the spirometry variables were reduced compared with non-COPD patients.

In order to determine whether the architecture of elastic fibres in alveolar walls was different among COPD patients and controls, we used Verhoeff's staining, a method that allows highlighting elastic fibres.22

As expected, sections from COPD patients exhibited an altered elastin network when compared to control subjects (Fig. 1). Thus, there were anomalous accumulations of elastic fibres in periemphysematous alveolar walls of COPD patients, suggesting a tissue collapse in periemphysematous areas (Fig. 1B, D and F). By contrast, control samples did not show these histological alterations (Fig. 1A, C and E).

Fig. 1.

Localization of elastic fibres in lung samples from COPD patients and control subjects. Verhoeff's staining of elastic fibres in alveolar ducts of a healthy control (A) and a COPD patient (B), with arrows indicating accumulation of elastic fibres in collapsing walls of emphysematous areas (bars=100μm). Detail of a tip of an alveolar wall of a control subject (C), showing the normal distribution of elastic fibres, while the alveolar walls of the emphysematous (D) areas appear collapsed in COPD patients (bars=25μm). Micrography of normal lung parenchyma of a control subject (E, bar=100μm) and emphysematous parenchyma of COPD patient (F, bar=50μm), showing prominent elastic accumulations. (G) Electron microscopy of a tip of a normal alveolar wall (bar=500nm). (H) Electron microscopy of a collapsing alveolar wall, showing vacuoles and a pierced pattern in the amorphous component (bar=500nm).

(0.35MB).

When the tips of alveolar walls were observed under TEM, COPD patients showed vacuolar changes in the amorphous component of elastic fibres, resulting in a pierced pattern (Fig. 1H). These alterations were absent in control samples (Fig. 1G).

Light microscopy study showed that EC-SOD was immunolocalised in airway and parenchymal epithelial cells (primarily in the apical area), as well as in alveolar macrophages and endothelial cells. EC-SOD was also present in the perimeter of elastic fibres of pulmonary vessels and alveolar walls, whereas collagen showed a weaker or no staining. Interestingly, EC-SOD immunostaining was higher in airway and vascular walls of control subjects (Fig. 2A and C) when compared to COPD patients (Fig. 2B and D). Regarding fibulin-5, it was also observed mainly in airway and vascular walls, as well as in alveolar interstitium of lung samples (Fig. S1).

Fig. 2.

Immunolocalization of EC-SOD in airway and vascular walls of COPD patients and control subjects. Presence of EC-SOD was markedly higher in airways (A, bar=25μm) and vascular walls (C, bar=10μm) of controls subjects when compared to those of COPD patients (B, bar=25μm, and D, bar=25μm, respectively).

(0.39MB).

Moreover, immunolocalization of EC-SOD and fibulin-5 allowed us to study the interstitial expression of these proteins. Alveolar, bronchiolar and arteriolar walls of patients with COPD showed a reduced expression of EC-SOD and fibulin-5 when compared to control subjects (Fig. 3 and Fig. S1). The additional immunohistochemical analysis by TEM confirmed the reduced expression of EC-SOD in interstitium of COPD patients, and revealed that EC-SOD was tightly associated with fibrillary components in the unaffected elastic fibres (Fig. 3 and Fig. S1). Moreover, in altered fibres, the decrease in EC-SOD labelling was associated with the loss of the fibrillary component and the partial degradation of elastin, which produced the pierced pattern (Fig. 3 and Fig. S1).

Fig. 3.

Immunolocalization of EC-SOD and fibulin-5 in alveolar walls of COPD patients and control subjects. Alveolar walls of control subjects showed increased presence of EC-SOD (A, bar=10μm) and fibulin-5 (E, bar=20μm) as compared to those of COPD patients (B, bar=10μm, and F, bar=20μm, respectively). (C) and (D) Electron microscopy of immunolocalised EC-SOD in elastic fibres from alveolar walls of a control subject (C, bar=200nm) and a COPD patient (D, bar=200nm). Colloidal gold labelling indicating EC-SOD immunolocalization was mostly absent in the fibrillary component alveolar elastic fibres of COPD samples. Dashed lines indicate loss of fibrillary component.

(0.25MB).

Furthermore, when the expression of EC-SOD and fibulin-5 in alveolar, bronchiolar and arteriolar walls of COPD patients and control samples was compared through a semi-quantitative score, we found that COPD patients have a significant reduced expression of both EC-SOD and fibulin-5 (0.59±0.64 and 0.62±0.61, respectively) compared to control subjects (1.39± 0.63 and 1.55±0.52, respectively, p<0.05, Fig. 4 and Fig. S1). Additionally, levels of EC-SOD and fibulin-5 were analyzed in patients without COPD by comparing smokers (n=10) and non-smokers (n=9) and males (n=11) versus females (n=8). Immunoscore values for EC-SOD in alveolar, arteriolar and bronchiolar wall were 1.44±0.23, 1.45±0.21 and 1.67±0.12 in non-smokers and 1.35±0.20, 1.8±0.13 and 1.65±0.11 in smokers without COPD (p: 0.63, 0.12 and 0.89 respectively). The immunoscore for fibulin-5 in these compartments were 1.56±0.18, 1.72±0.11 and 1.73±0.17 in non-smokers and 1.50±0.17, 1.60±0.18 and 1.66±0.13 in smokers without COPD (p: 0.96, 0.81 and 0.51 respectively). Thus, no significant differences were observed in the expression of the antioxidant markers when comparing smokers and non-smokers. Immunoscore values for EC-SOD in alveolar, arteriolar and bronchiolar wall were 1.50±0.20, 1.50±0.20 and 1.65±0.11 in non-COPD females and 1.28±0.20, 1.78±0.15 and 1.67±0.12 in non-COPD males (p=0.37, 0.23 and 0.89, respectively. Values for fibulin-5 in the same compartments were 1.65±0.13, 1.65±0.17 and 1.80±0.11 in non-COPD females and 1.44±0.21, 1.67±0.14 and 1.56±0.18 in non-COPD males (p=0.54, 0.92 and 0.25, respectively).

Fig. 4.

Semiquantitative score of the expression of EC-SOD and Fibulin-5 in the lung parenchyma. Semiquantitative score showed a significant reduction of EC-SOD and fibulin-5 in the connective tissues of the lung parenchyma from COPD patients when compared to control subjects (estimated mean±SEM).

(0.05MB).

Regarding oxidative stress markers, malondialdehyde was the most reliable to distinguish COPD patients. Indeed, the adducts of this lipid peroxidation product were mainly immunolocalised in the connectives tissues, such as adventitia layers, alveolar walls, arterioles and the parenchymal interstitium surrounding emphysematous areas of COPD patients, showing significant differences with non COPD samples (Fig. S2). Although 4-hydroxynonenal adducts were detected in all patients, they showed a significant increase in COPD groups (Fig. S3), while the 3-nitrotyrosine biomarker was primarily located in the tissues close to the respiratory tract and alveolar spaces of smokers (Figure S3).

Quantitative PCR analyses showed that both mRNA levels of EC-SOD and fibulin-5 were lower in COPD patients than in controls (Table 2), though these differences did not reach statistical significance (p>0.05, Table 2).