INTRODUCTION

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

Progressive lung disease accounts for most of the morbidity and mortality in cystic fibrosis (CF) (1). There is increasing evidence that a marked and sustained inflammatory response occurs in the airways of patients with CF, even when they are in stable clinical conditions (1, 2). The most characteristic feature of this inflammatory response is a persistent influx of polymorphonuclear neutrophils into the airways; these cells release a variety of oxidants and granule-associated enzymes, particularly neutrophil elastase, and may contribute to the development of lung injury and the maintenance of chronic pulmonary infection (3, 4).

Leukocyte-endothelial adhesion molecules play a crucial role in the recruitment and migration of inflammatory cells from vascular compartments into the airways, as well as in inflammatory cell activation (5). Among endothelial cell adhesion molecules, intercellular adhesion molecule-1 (ICAM-1), vascular cell adhesion molecule-1 (VCAM-1), and E-selectin have been implicated in the development of airway inflammation (6).

ICAM-1 is a glycoprotein of 80 to 115 kD belonging to the immunoglobulin supergene family, which is expressed on a variety of hemopoietic and nonhemopoietic cells (5, 8); it is the counterreceptor for the ₂ leukocyte integrins lymphocyte function associated antigen-1 (LFA-1) and Mac-1 (9, 10). ICAM-1 is constitutively expressed at low level on normal endothelium in vivo, and can be upregulated by several cytokines such as interferon- (IFN-), interleukin-1 (IL-1) and tumor necrosis factor- (TNF-) (8). VCAM-1, which is also a member of the immunoglobulin supergene family, is absent from unstimulated vascular endothelium, but it can be induced by a number of cytokines, including IL-1, TNF-, and interleukin-4 (IL-4) (11, 12). Endothelial VCAM-1 is involved in lymphocyte, monocyte, and eosinophil, but not neutrophil adhesion to activated endothelium, through an interaction with the ₁ integrin, very late activation antigen-4 (VLA-4) (12). E-selectin (previously known as endothelial leukocyte adhesion molecule-1 [ELAM-1]) is expressed exclusively on endothelial cells after stimulation by IL-1, TNF-, or bacterial lipopolysaccharide (LPS) (13). E-selectin interacts with a sialy-fucosyl pentasaccharide determinant, Sialy-Le^x, on leukocyte surface, and it is involved in an early step of leukocyte binding to endothelium (14).

Recently, soluble forms of these adhesion molecules have been described (15, 16) and shown to increase in the peripheral blood and other body fluids in several inflammatory diseases (17). For instance, we and others have shown that in atopic asthmatic patients soluble ICAM-1 (sICAM-1) significantly increased in serum (18, 19) and bronchoalveolar lavage fluid (19) at the time of the late-phase asthmatic response, a condition known to be associated with the recruitment and activation of inflammatory cells into the airways. Elevated serum levels of sICAM-1 have also been reported in idiopathic pulmonary fibrosis, miliary tuberculosis, sepsis, and acute asthma (20). Soluble E-selectin (sE-selectin) levels have been found markedly elevated in sepsis, and have been detected in bronchoalveolar lavage fluid of patients with interstitial lung diseases (17). Elevated serum levels of soluble VCAM-1 (sVCAM-1) have been reported in rheumatoid arthritis and systemic lupus erythematosus (24).

The aim of the present study was to determine whether serum levels of sICAM-1, sE-selectin, and sVCAM-1 were increased in patients with CF in stable clinical conditions and to examine whether the concentrations of these circulating adhesion molecules were related to disease severity in these patients. To this aim we determined serum levels of sICAM-1, sE-selectin, and sVCAM-1 in CF patients without signs and symptoms of acute exacerbation and compared them with those in healthy volunteers. We also examined the relationship between serum levels of these soluble adhesion molecules and clinical markers of disease severity.

	METHODS

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

Patients

Twenty-nine adult patients with CF (12 males, 17 females) 18 to 50 yr of age (mean ± SEM: 25.8 ± 1.5 yr) were studied. The diagnosis of cystic fibrosis was based on a typical clinical profile with a positive sweat test; sweat test was carried out by quantitative pilocarpine iontophoresis according to Gibson and Cooke (25), and repeatedly elevated sweat chloride levels (> 70 mEq/L) were considered positive. Six patients were homozygous for the F508 mutation: 21 were heterozygous for F508 with an associated other mutation, 18 of these 21 patients had known mutations; one patient was homozygous for the G542X mutation, and in one patient the genotype was N1303K/?. All patients studied had evidence of lung disease, based on typical clinical and radiographic findings. Microbiological analyses of sputum were routinely performed at each monthly control visit in the outpatient clinic of the adult section of the CF center, in Turin, Italy. Of the 29 patients, 19 had chronic infection of the airways with Pseudomonas aeruginosa, and eight with Staphylococcus aureus; chronic infection was defined as persistent growth of these bacteria for 6 mo or more, continuously. The remaining two patients had sterile sputum cultures. All patients were clinically stable at the time of the study, and with no evidence of infectious exacerbations during the previous 2 mo. None of them was taking antibiotics, nonsteroidal antiinflammatory drugs, nor systemic or topical corticosteroids for at least 2 mo before the study. Patients with CF-related liver disease, pancreatitis, diabetes mellitus, or allergic bronchopulmonary aspergillosis were excluded from the study.

Clinical, spirometric, microbiological, and hematological assessments were made in all subjects at the time of the study. The clinical scoring system of Schwachman and Kulczycki (26) was carried out on all patients. This system of clinical evaluation is based on the general activity of the patient, findings on physical examination, nutritional status, and chest roentgenogram findings. Each of these four items is given equal weight in a scoring system in which 100 represents a perfect score. All clinical assessments were performed by one of two clinicians who had no knowledge of the laboratory data. Pulmonary function tests for FEV₁ and FVC were performed using a computer-assisted spirometer (MasterLab; Jaeger, Wurzburg, Germany) and results were expressed as percentage of predicted values (27). Sputum samples were collected for microbiological analysis. Hematological assessments included an erythrocyte sedimentation rate, a total and differential blood white cell count, and levels of C-reactive protein.

Twelve healthy volunteers (5 males, 7 females) 24 to 40 yr of age (mean ± SEM: 27.6 ± 1.5) were evaluated as control subjects. None had a history of respiratory diseases, and all had normal physical examination, chest radiograph, lung function tests, and sterile sputum cultures. All subjects were nonsmokers. None had a history of infection or use of antibiotics or anti-inflammatory drugs within 2 mo before the study.

Ten-milliliter venous blood samples were obtained from each subject; samples were allowed to clot, and serum was obtained after centrifugation, and frozen at 80° C for later analysis of soluble adhesion molecules.

In 7 of 29 CF patients we were able to obtain serial serum samples and to evaluate soluble adhesion molecule levels not only at the time of stable clinical conditions, but also at the time of an acute pulmonary exacerbation, both before and after a 2-wk course of intravenous antibiotic therapy. A pulmonary exacerbation was defined as the presence of at least three of the following signs and symptoms: an increase in sputum purulence and volume, and increased respiratory rate or dyspnea, increased cough, weight loss and anorexia, temperature above 38° C on more than one occasion in the previous week, new findings on chest examination, new findings on chest radiograph, or a decrease in FEV₁ of 10% or more from a previously recorded value when the patient was clinically stable.

The study conformed to the Declaration of Helsinki, and informed consent was obtained from all subjects.

sICAM-1, sE-selectin, sVCAM-1 Assay

Serum levels of the soluble adhesion molecules were determined by commercially available ELISA (Bender MED-Systems, Vienna, Austria for sICAM-1 and sE-selectin; R&D Systems Europe Ltd, Abingdon, Oxon, U.K. for sVCAM-1), according to the procedure recommended by the manufacturer. All measurements were performed in duplicate. The reported sensitivity of the assay was 3.3 ng/ml for sICAM-1, < 1.6 ng/ml for sE-selectin, and < 2 ng/ml for sVCAM-1.

Statistical Analysis

Mann-Whitney U test for nonpaired data was used to compare soluble adhesion molecule levels in CF patients and control subjects, and Wilcoxon's matched pairs test to compare the levels of soluble adhesion molecules in the same CF patients at stable clinical conditions and at the time of pulmonary exacerbations, both before and after antibiotic treatment. Correlations were determined by the Spearman's rank correlation test. Statistical significance was assumed for p values lower than 0.05. Data are presented as mean ± SEM.

	RESULTS

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

Characteristics of CF Patients

The clinical characteristics of CF patients are shown in Table 1. The mean value of Schwachman score was 63.31 ± 4.01. FEV₁ values ranged from 19% to 93% of predicted, with a mean of 55.17 ± 4.08%. Nineteen patients had chronic infection of the airways with P. aeruginosa, and the mean duration of infection was 8 ± 1 yr (range: 2 to 19 yr); eight patients were chronically infected with S. aureus, with a mean duration of infection of 5.6 ± 1.9 yr (range: 1.5 to 15 yr). The remaining two patients had sterile sputum cultures. One of these patients, age 19 yr, underwent repeated (one per month) sputum cultures over the previous 2 yr which were consistently negative, and had never had in the past P. aeruginosa or S. aureus; the other patient, age 50 yr, was first seen at our CF Center at age 49 yr, when the diagnosis of CF was made; since then sputum cultures were performed at monthly intervals over 1 yr which were consistently negative.

Soluble Adhesion Molecules

Serum levels of sICAM-1 were significantly higher in CF patients in stable clinical conditions as compared with those in normal subjects (594.4 ± 20.4 ng/ml versus 322.08 ± 15.02 ng/ ml, respectively, p < 0.001) (Figure 1A). Similar results were obtained for sE-selectin; serum levels of this adhesion molecule were also significantly higher in CF patients than in control subjects (40.1 ± 2.7 ng/ml versus 24.3 ± 1.4 ng/ml, respectively, p < 0.001) (Figure 1B). In contrast, no significant differences were observed in serum levels of sVCAM-1 between the two groups (CF: 618.4 ± 42.9 ng/ml, control subjects: 533.2 ± 82.9 ng/ml) (Figure 1C). No significant differences in serum levels of sICAM-1, sE-selectin, and sVCAM-1 were observed either between CF patients with chronic infection of the airways with P. aeruginosa or S. aureus or between both these groups and patients without chronic infection (Table 2).

View larger version (11K): [in this window] [in a new window]   Figure 1.   Serum levels of sICAM-1 (A), sE-selectin (B), and sVCAM-1 (C ) in 29 patients with CF (open circles: patients with chronic infection of the airways; closed circles: patients without chronic infection) and 12 healthy control subjects. Bars represent mean values.

Interestingly, there was a negative, although weak, correlation between serum levels of sICAM-1 and both FEV₁ values (r = 0.519, p = 0.004), and Schwachman score (r = 0.405, p = 0.03) in CF patients (Figure 2). One patient showed a marked difference from the trend, having the lowest sICAM levels (410 ng/ml) and one of the lowest FEV₁ (25% of predicted). The clinical history of this patient showed that 5 yr prior to the study he had a bilateral chest trauma, with left hemothorax and pneumothorax which induced persistent impairment of pulmonary function (10% decrease in FVC, and 13% decrease in FEV₁ after the trauma). When this patient was excluded from the analysis, a stronger correlation with FEV₁ was observed (r = 0.668, p < 0.001). No correlations were observed between serum levels of sE-selectin or sVCAM-1 and markers of disease severity. No correlation could be found either between serum levels of sICAM-1, sE-selectin, and sVCAM-1, in CF patients.

View larger version (9K): [in this window] [in a new window]   Figure 2.   Correlation between serum levels of sICAM-1 and FEV1 values (A) and Schwachman score (B) in patients with CF (closed circles: patients with chronic infection of the airways; open circles: patients without chronic infection).

In 7 of 29 CF patients, serial measurements of sICAM-1, sE-selectin, and sVCAM-1 were performed at the time of stable clinical conditions, and before and after antibiotic treatment for a pulmonary exacerbation. As shown in Figure 3, serum levels of both sICAM-1 and sE-selectin significantly increased in all patients at the time of the acute exacerbation (mean ± SEM: 747.5 ± 51.3 ng/ml, 119.9 ± 20.9 ng/ml, respectively) compared with levels at the time of stable clinical conditions (591.1 ± 32.8 ng/ml, 42.5 ± 5.3 ng/ml, respectively, p < 0.02 for both comparisons); antibiotic therapy induced a significant decrease of serum levels of both adhesion molecules (sICAM-1: 608.8 ± 40.7 ng/ml, sE-selectin: 63.9 ± 11.7 ng/ml, p < 0.02 as compared with levels before therapy). In contrast, no differences were observed in serum levels of sVCAM-1 at the different times (data not shown).

View larger version (18K): [in this window] [in a new window]   Figure 3.   Serum levels of sICAM-1 (A) and sE-selectin (B) in patients with CF (n = 7), at the time of stable clinical conditions and before and after antibiotic treatment for a pulmonary exacerbation. Serum levels of both sICAM-1 and sE-selectin significantly increased at the time of the acute exacerbation, compared with concentrations at the time of stable clinical conditions (p < 0.02 for both comparisons); after antibiotic treatment, levels of both adhesion molecules significantly decreased, compared with levels before treatment (p < 0.02). Bars represent mean values.

	DISCUSSION

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

This study shows that (a) serum levels of sICAM-1 and sE- selectin, but not sVCAM-1 are significantly increased in CF patients in stable clinical conditions, as compared with healthy control subjects, (b) serum levels of sICAM-1 are inversely related to disease severity in CF patients.

ICAM-1 is a cell adhesion molecule expressed on cells of multiple lineages, and upregulated at sites of inflammation (5, 8). It is important in the adhesion and migration of inflammatory cells through endothelium and epithelium and in the development of specific immune responses (5, 28). Several lines of evidence suggest that ICAM-1 plays a critical role in the inflammatory response. ICAM-1 can be induced in vitro on multiple cell types by inflammatory mediators such as IL-1, TNF-, and IFN- (5, 8). Increased expression of this molecule, as well as increased levels of its soluble isoforms, have been reported in several inflammatory conditions such as bronchial asthma, rheumatoid arthritis, interstitial lung diseases, miliary tuberculosis, and sepsis (17). Antibodies to ICAM-1 have been shown to inhibit leukocyte adhesion to endothelial cells and granulocyte migration through endothelium in vitro (29); furthermore, in vivo, antibodies to ICAM-1 inhibited neutrophil trafficking into inflamed lungs in rabbits (30). E-selectin is a cell adhesion molecule which, unlike ICAM-1, is expressed only on the surface of activated endothelial cells; it is involved in the early stage of leukocyte adhesion to endothelium (13, 14). In several models of skin inflammation, E-selectin expression was found to correlate with neutrophil influx (7, 31). Furthermore, an antibody directed against E-selectin was found to block neutrophil extravasation and vascular leak in the lungs of rats, following IgG immune complex induced damage (32).

In vitro studies have shown that both ICAM-1 and E-selectin molecules are shed from endothelial cells, after their exposure to mediators such as TNF-, IL-1, and LPS (33).

In this context, the increased concentrations of circulating ICAM-1 and E-selectin that we observed in CF patients might reflect the upregulation of these adhesion molecules on vascular endothelial cells in lung tissue, as a consequence of the marked inflammatory response occurring at this level in cystic fibrosis. Increased concentrations of a number of cytokines have been reported in the airways and in the circulation of patients with CF (3, 4). In particular, increased levels of IL-1, TNF-, and IL-8 have been found in bronchoalveolar lavage fluid of CF patients, even with stable disease (34). Such cytokines might be responsible for local activation of endothelial cells, leading to increased expression and release into the circulation of ICAM-1 and E-selectin. However, as ICAM-1 is expressed on a variety of cell types in the airways and in the circulation, the source of the soluble ICAM-1 detected in serum of CF patients remains unclear and we cannot exclude that it might be derived from other cells in addition to or instead of activated endothelial cells. Consistent with this hypothesis, Rothlein and coworkers showed that peripheral blood mononuclear cells released detectable levels of sICAM-1 in culture (16). Furthermore, in vitro studies have shown that ICAM-1 expression on cultured airway epithelial cells is significantly enhanced by stimulation with cytokines such as IL-1, TNF-, and IFN-, although the release of the soluble form of this molecule was not evaluated in these studies (28, 35).

Because all CF patients studied were adult patients, most of them were chronically infected with P. aeruginosa and S. aureus; therefore, as an alternative possibility, the increase in levels of circulating ICAM-1 and E-selectin might be a consequence of the chronic infection of the airways with these microorganisms. This possibility, however, seems less likely because serum levels of sVCAM-1 were not elevated in CF patients, and because serum levels of sICAM-1 and sE-selectin were not significantly different between patients with or without chronic infection of the airways with P. aeruginosa and S. aureus. Although our patients were closely monitored through monthly microbiological analysis of sputum, we cannot exclude that if bronchoalveolar lavage (BAL) had been performed, S. aureus or P. aeruginosa might have been found also in the two patients with repeatedly negative sputum cultures. Konstan and coworkers, who performed BAL in CF patients with mild lung disease, recently showed that these patients have bacteria in their BAL fluid and a marked local inflammatory response, even when they are clinically stable (2).

In the present study we observed a differential rise of sICAM-1 and sE-selectin, but not sVCAM-1; this finding may be due to the prevalence in CF of mediators and cytokines that are more suited to expression of the former two adhesion molecules. More speculatively, this finding seems physiologically reasonable, as VCAM-1 is involved in the binding of monocytes, lymphocytes, eosinophils, but not neutrophils to activated endothelium.

Interestingly, serum levels of sICAM-1 were inversely related to FEV₁ values and Schwachman score in CF patients, suggesting that they might reflect the severity of the inflammatory process in the disease. In contrast, no relationship was observed between serum levels of sE-selectin and markers of disease severity; the significance of this finding is not immediately clear, but it seems to support the hypothesis that serum levels of sICAM-1 may derive not only from activated endothelial cells, but also from other cells in the airways and/or in the circulation, and, therefore, they may better reflect the overall inflammatory state in CF.

In 7 of 29 CF patients we were able to perform serial measurements of sICAM-1, sE-selectin, and sVCAM-1, at the time of stable clinical conditions, and before and after antibiotic treatment for a pulmonary exacerbation. Interestingly, serum levels of sICAM-1 and sE-selectin increased in all patients at the time of the exacerbation; antibiotic treatment induced a significant decrease of both circulating adhesion molecules. These results are in line with previous reports showing that circulating concentrations of other inflammatory markers (such as TNF-, C reactive protein, or neutrophil elastase-1 antiproteinase complex) increased in CF at the time of acute exacerbations, and were significantly reduced by antibiotic treatment (36). Furthermore, they suggest that the measurement of these soluble adhesion molecules might prove to be of clinical utility in the follow-up of CF patients and in monitoring antibiotic therapy. However, additional clinical studies are needed to confirm this potential clinical utility and to define the specificity of these markers. In addition to performing serial measurements of levels of circulating ICAM-1 and E-selectin in a larger number of patients, it would be of considerable interest to evaluate a population of younger patients, and possibly infants, and to determine the relationship between serum levels of sICAM-1 and sE-selectin and the degree of infection and inflammation, as assessed by BAL bacterial counts and neutrophil counts.

It is now widely recognized that airway inflammation plays a crucial role in the development of lung injury in CF. In this context, reliable, noninvasive markers of airway inflammation should be very useful in the follow-up of CF patients. A major problem in the management of these patients involves judging when to initiate and how to assess antimicrobial treatment. Although several inflammatory markers in serum and plasma have been studied in CF patients, none have been shown to be highly predictive of acute changes. To evaluate serum levels of adhesion molecules as a predictive marker of acute exacerbation was not the aim of the present study; regardless, our observations that serum levels of sICAM-1 and sE-selectin are elevated in CF patients in stable clinical conditions, and further increase during acute exacerbations suggest that these adhesion molecules may play an important role in CF lung inflammation, and deserve further investigation into the potential clinical utility of these markers.

In conclusion, this study shows that serum levels of sICAM-1 and sE-selectin are increased in CF patients, even when they are in stable clinical conditions, and that serum levels of sICAM-1 are inversely related to disease severity. The increase of these adhesion molecules might be influenced by their upregulation on vascular endothelial cells and/or other activated cells and may reflect the marked and persistent inflammatory process in the disease.