INTRODUCTION

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

Recruitment and activation of neutrophils is an important event in the pathogenesis of chronic obstructive pulmonary disease (COPD) (1, 2). The expression of several adhesion molecules, such as lymphocyte function-associated antigen-1 (LFA-1) (CD11a/CD18), Mac-1 (CD11b/CD18), and L-selectin on the neutrophil surface, and intercellular adhesion molecule-1 (ICAM-1) on the endothelium, is an important step in this process (3). Systemic oxidative stress occurs in peripheral blood of patients with COPD, particularly during exacerbations (4, 5). This can upregulate the expression of the above mentioned adhesion molecules (3), thus facilitating the recruitment of neutrophils into the lung parenchyma (1, 2). In this study we hypothesized that patients with COPD may show altered expression of these adhesion molecules, and that this may be modulated during exacerbations of the disease. Also, because guanine nucleotide binding proteins (G proteins) play a key role in the regulation of many cell signaling pathways (6, 7), including the activation (8) and adhesion of human neutrophils to tissues (9), we assessed the expression of two G protein subunits (Gs and Gi1/2) in the circulating neutrophils of these patients.

	METHODS

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

Population and Ethics

We studied 23 patients with clinically stable COPD and 18 patients with exacerbated COPD. All patients with COPD had significant (> 20 pack-years) smoking history and evidence of severe irreversible airflow obstruction (Table 1). Patients with stable COPD had not required medical attention and/or change in regular therapy over the previous 4 mo. Exacerbated patients were patients who required hospitalization because of acute on chronic respiratory failure (Table 1). Any COPD patient who suffered from another disease that could potentially interfere with their diagnosis and/or clinical course (such as pneumonia or lung cancer) was excluded, and none had received steroids before the study (inhaled or oral). Arterial blood gases (BG3; Instrumentation Laboratories, Izasa, Barcelona, Spain) and forced spirometry (GS, Warren E. Collins, Braintree, MA) were measured in all patients (Table 1). As a control group, we also studied 23 healthy male nonsmoking volunteers of similar age. All participants gave written consent having been fully informed of the nature, characteristics, risks, and potential benefits of the study. The study was approved by the research and ethical review board of the Hospital Universitari Son Dureta.

Isolation of Neutrophils

We obtained 30 ml of peripheral venous blood from each participant. In exacerbated patients, blood was obtained before intravenous treatment with steroids had been started. No subject participating in the study had smoked during the previous 2 h before blood sampling (10). The relatively small volume of blood obtained in each subject precluded the determination of every measurement in all participants. Leukocyte-rich plasma was obtained by mixing with an equal volume of endotoxin-free Hemoce reagent (Hoechst Iberica, Barcelona, Spain), followed by sedimentation during 1 h at 4° C. The resulting supernatant was kept at 80° C for determination of the soluble fraction of ICAM-1 (sICAM-1), as discussed subsequently. Neutrophils were separated from the leukocyte-rich plasma by centrifugation on 15-ml layer of endotoxin-free Ficoll-Paque Research Grade gradient (Pharmacia Biotech, Uppsala, Sweden) at 900 × g for 30 min at 22° C (11, 12). Residual erythrocytes were removed by mixing the neutrophil-rich pellet with 50 ml of ice-cold 0.15 M CINH₄ solution, which was gently rocked at 4° C for 10 min and then centrifuged at 750 × g for 10 min at 4° C. The neutrophil pellet was washed once with phosphate-buffered saline (PBS) and resuspended with 1 ml of PBS, counted by Sysmex K-4500 (Toa Medical Electronics Co. Ltd., Kobe, Japan) and adjusted to 4 × 10⁶ cells/ml with PBS. The neutrophil suspensions were routinely > 97% pure as assessed by Giemsa staining and 99% viable as assessed by trypan blue exclusion.

Immunofluorescence and Flow Cytometric Analysis

Neutrophil suspensions (100 µl) were mixed with 20 µl of fluorescein isothiocyanate (FITC)-labeled monoclonal antibody anti-LFA-1 (CD11a) (25.3.1 clone; Immunotech, Marseille, France), FITC-labeled anti-Mac-1 (CD11b) (BEAR 1 clone; Immunotech, Marseille, France), FITC-labeled anti-L-selectin (CD62L) (DREG56 clone; Immunotech, Marseille, France), and FITC-labeled IgG1 (2T8-2F5 MsIgG1; Coulter Immunology, Hialeah, FL) that acted as a nonspecific control antibody. This was incubated at 4° C for 30 min, washed twice in ice-cold PBS, resuspended in 1 ml of PBS and kept on ice until analyzed. Flow cytometry was performed on a Becton-Dickinson FACScan (Becton-Dickinson, Mountain View, CA). Data were collected for forward scatter, side scatter, and fluorescence on the FL1 channel (530 nm) with Becton-Dickinson LYSIS II software. Cells were analyzed with a gate setting for neutrophils on forward and side scatter diagrams. Ten thousand cell counts were accumulated for analysis. The nonspecific binding (IgG1) was measured at a mean fluorescence intensity (MFI) of 2.81 ± 0.17 units.

Determination of sICAM-1

The plasma levels of sICAM-1 were quantified by ELISA (Boehringer Mannheim, Mannheim, Germany). Standards of defined concentrations of recombinant sICAM-1 are run in each assay, allowing the construction of a calibration curve from which sICAM-1 values of plasma samples are calculated.

Quantitation of G Proteins

Neutrophils were prepared for immunodetection as previously described (13, 14). Briefly, pellets of neutrophils were homogenized in 1 ml of cold 40 mM TRIS buffer, pH 7.5 (1% Triton X-100, 1 mM ethylenediaminetetraacetic acid (EDTA), 1 mM MgCl₂, mM NaC1, 1 mM phenylmethylsulfonyl fluoride, and 40 µg/ml leupeptin). Samples were centrifuged at 12,000 × g for 10 min at 4° C. The resulting supernatant was mixed with an equal volume of electrophoresis loading buffer (62.5 mM TRIS, pH 6.8, 3% sodium dodecyl sulfate (SDS), 20% glycerol, 0.005% bromophenol blue), which was then boiled for 3 min. Proteins were determined by the method of Bradford (15). Five to 20 µl of the resulting suspension was loaded in a 10% polyacrylamide gel and submitted to electrophoresis. After transferring proteins to nitrocellulose membranes, these were incubated in PBS containing 4% nonfat dry milk and 0.1% Tween 20 (blocking solution), with the primary antibodies (anti-Gs [RM/1] and anti-Gi1/2 [AS/7]; Dupont/ New England Nuclear Corp., Hamburg, Germany) at 1:12,000 dilution, at room temperature for 2 h. The secondary antibody (horseradish peroxidase-linked sheep anti-rabbit IgG) was incubated at 1:5,000 dilution in blocking solution at room temperature for 2 h. Immunoreactivity was detected with a supersignal chemiluminescence detection system (Pierce, Rockford, IL) followed by exposure to film (Ambersham Hyperfilm ECL). Specific protein immunoreactivity was quantitated by scanning densitometry as described previously (16).

Statistical Analysis

Data are expressed as mean ± SEM. Because the data were normally distributed (Kolmogorov-Smirnov test), we used an unpaired t test (or analysis of variance [ANOVA] followed post-hoc by Scheffé test) to assess the statistical significance between values. A p value less than 0.05 was considered significant.

	RESULTS

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

Table 1 shows the main lung function data in the patients. The absolute value of circulating neutrophils was lower in control subjects (3.9 ± 0.5 × 10⁶/ml) than in patients with stable (7.4 ± 0.4 × 10⁶/ml, p < 0.001) or exacerbated COPD (12.0 ± 1.4 × 10⁶/ml, p < 0.001).

Figure 1 shows the individual and mean values of LFA-1, Mac-1, L-selectin, and sICAM-1 in the three groups of subjects studied. The most striking finding was that stable COPD patients showed higher expression of Mac-1 (p < 0.001) and lower levels of sICAM-1 (p = 0.002) than control subjects. In contrast, LFA-1 and L-selectin values were not significantly different in these two groups of subjects (Figure 1). In patients studied during exacerbations, these increased levels of Mac-1 expression were lower (p < 0.001) than in stable patients (Figure 1B). Similarly, during exacerbations LFA-1 values were lower than in control subjects (p < 0.05) and stable patients (p < 0.001) (Figure 1A). By contrast, neither L-selectin nor sICAM-1 were different during exacerbations as compared with stable COPD patients (Figures 1C and 1D).

View larger version (18K): [in this window] [in a new window]   Figure 1.   Scatter plots showing the surface expression on circulating neutrophils of the adhesion molecules LFA-1 (A), Mac-1 (B), L-selectin (C ), and the plasma levels of sICAM-1 (D). Each dot represents one single determination. Horizontal bars represent mean values. MFI = mean fluorescence intensity.

The expression of Gs was significantly reduced in the circulating neutrophils of patients with stable and exacerbated COPD (Figure 2B). This was not the case for the Gi1/2 subunit, whose immunoreactivity was not different from control subjects, irrespective of the clinical condition of the patient (Figure 2C).

View larger version (15K): [in this window] [in a new window]   Figure 2.   (A) Representative immunoblots of human neutrophils G protein subunits Gs (45 kD) and Gi1/2 (40 kD) signal intensity and protein content. The amount of total protein loaded per gel was 2 µg (lane 1), 4 µg (lane 2), 6 µg (lane 3), and 8 µg (lane 4). (B) Scatter plot showing the loss of expression of G protein Gs subunit in COPD patients compared with control subjects. Solid dots: stable COPD; open dots: exacerbated COPD. Horizontal bars represent mean values. (C ) Scatter plot showing the expression of G protein Gi1/2 subunit in COPD patients compared with control subjects. Solid dots: stable COPD; open dots: exacerbated COPD. Horizontal bars represent mean values.

	DISCUSSION

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

The main findings of our investigation were that: (1) compared with healthy subjects, patients with stable COPD showed abnormal expression of some neutrophil and endothelial adhesion molecules (Figure 1); (2) the pattern of expression of these adhesion molecules changes during exacerbations of the disease (Figure 1); and, (3) independently of their clinical state, patients with COPD show a consistent decrease in the expression of the Gs subunit (but not Gi1/2) on their circulating neutrophils (Figure 2).

Previous reports have shown increased expression of adhesion molecules in the bronchial mucosa of patients with COPD (17, 18). To our knowledge, no previous study has investigated their expression on circulating neutrophils in these patients. We observed that some (Mac-1), but not all (LFA-1 or L-selectin) of the neutrophil adhesion molecules investigated were significantly upregulated in patients with stable COPD (Figure 1). This indicates some degree of specificity of this response and, more importantly, that COPD is associated with systemic (i.e., extrapulmonary) effects, even when the condition is thought to be clinically stable. This interpretation would agree with recent studies showing evidence of systemic oxidative stress in peripheral blood in these patients (4, 5). Yet, because the latter was greatest during exacerbations of the disease (4, 5), the fact that the expression of the neutrophil adhesion molecules in our patients did not increase during exacerbations (Figure 1) requires some explanation. First, it has to be remembered that our study reports cross-sectional and not longitudinal data. Because two distinct groups of patients have been examined in stable conditions and in exacerbations, any interpretation of the dynamics of the inflammatory response is necessarily speculative. A potential downregulatory effect of steroids on adhesion molecules expression (3) can be ruled out because no patient received steroids before blood sampling. Alternatively, higher amounts of oxidative stress reaching the cells in transit in the microcirculation during exacerbations (4, 5) may lead to downregulation of adhesion molecules. However, this is unlikely because oxidative stress usually results in prolonged upregulation of adhesion molecules (19). A more likely explanation results from the increased sequestration of neutrophils in the pulmonary microcirculation during exacerbations of COPD (2). There is evidence from studies in experimental models of lung inflammation that the cells that are preferentially sequestered may be more activated than those that are circulating (20). This would leave a subpopulation of circulating cells which do not show such increased expression.

Firm adhesion of neutrophils to endothelial surfaces depends on the expression of the endothelial adhesion molecule ICAM-1 (3). The plasma level of sICAM-1 is considered a surrogate of its expression on the endothelium (21). Previous studies determining sICAM-1 in patients with chronic bronchitis and mild or no airflow obstruction have shown variable results (21, 22). No previous report has included patients with severe COPD, such as those herein studied, nor assessed patients during episodes of exacerbation. We observed reduced sICAM-1 values, both in patients who were clinically stable and during exacerbations (Figure 1). This is consistent with reduced expression of ICAM-1 on their endothelial surface and/or with a decrease in the total population of endothelial cells. There is evidence in the literature to support both possibilities. On the one hand, endothelial function appears to be abnormal in smokers (24, 25) and in patients with COPD (26, 27). On the other, the number of peripheral capillaries appears to be reduced in COPD (28). Although further studies will be required to dissect the relative importance of both hypotheses, our data do support the concept of some form of endothelial dysfunction (in the pulmonary and/or systemic circulation) in severe COPD.

G subunits (Gs and Gi1/2) are key proteins in the activation and adhesion of human neutrophils to tissues (29). No previous study has assessed their expression in COPD. The abnormal expression of some neutrophil adhesion molecules described previously prompted us to determine if the expression of these G subunits was also abnormal in patients with COPD. We found that, irrespective of the clinical condition of the patients with COPD, their circulating neutrophils showed significant loss of Gs (but not the Gi1/2) immunoreactivity (Figure 2). At present, the potential implications of this observation for the pathogenesis or treatment of COPD are unclear. Yet, the presence of these systemic abnormalities indicate that COPD can no longer be considered solely as a local lung disease. Rather, it supports previous indications (4, 5) of significant systemic abnormalities in these patients.