|
|
||||||||
Guest Access | Sign In via User Name/Password |
|||||||||
* From the Divisions of Clinical Immunology and Allergy (Drs. Wehlin and Lundahl) and Respiratory Medicine (Drs. Löfdahl and Sköld), Department of Medicine, Karolinska Institutet, Stockholm, Sweden.
Correspondence to: Lena Wehlin, PhD, Karolinska Institutet, Department of Medicine, Division of Clinical Immunology, Karolinska University Hospital, L2:04, S-171 76 Stockholm, Sweden; e-mail: Lena.Wehlin{at}medks.ki.se
| Abstract |
|---|
|
|
|---|
Methods: We enrolled 20 patients with stable, moderate COPD (FEV1, 33 to 69%). Ten asymptomatic smokers and 10 nonsmokers served as control groups. Flow cytometry and whole blood analysis were used to minimize unwanted ex vivo modulation. Oxidative burst and adhesion molecule mobilization were analyzed on freshly drawn cells and after in vitro activation.
Measurements and main results: We found reduced oxidative burst in neutrophils, monocytes, and eosinophils after in vitro stimulation with tumor necrosis factor (TNF) and the bacterial peptide N-formyl-methionyl-leucyl-phenylalanine (fMLP) in both COPD patients and asymptomatic smokers as compared to nonsmoking control subjects. Vascular involvement was determined as increased soluble intercellular adhesion molecule-1 (sICAM-1) in the COPD group. There were no differences in adhesion molecule expression among the three groups. However, in COPD patients who had smoked the same morning prior to blood sampling, we found a reduced ability to mobilize adhesion molecule CD11b after TNF plus fMLP activation in all investigated cell types. "Acute" smoking did not significantly alter respiratory burst measurements.
Conclusions: Both COPD patients and asymptomatic smokers have increased levels of sICAM-1 and a reduced intracellular oxidative burst in vitro, indicating a vascular endothelial activation and a possible state of refractoriness in circulating phagocytes in COPD. Although expression and mobilization of adhesion molecules were similar between groups, the acute smoke effect on CD11b points out the value of information on smoking behavior when analyzing function of peripheral inflammatory cells in a smoking population.
Key Words: adhesion molecules COPD eosinophil flow cytometry monocyte neutrophil oxidative burst
| Introduction |
|---|
|
|
|---|
. The increased oxidant load in the lungs due to an antioxidant/oxidant imbalance and reactive oxygen and nitrogen species in tobacco smoke is proposed to be a major cause for the systemic effects of COPD.4 Long-term cigarette smoking has several effects on immune cells. It causes neutrophilia with increased numbers of band cells indicating early bone marrow release.5 Neutrophils from smokers have an increased concentration and activity of myeloperoxidase,67 and they express more formyl peptide receptors8 than neutrophils from nonsmokers. Activation of peripheral monocytes has been reported in addition to increased concentrations of soluble intercellular adhesion molecule-1 (sICAM-1).9 In vitro cigarette smoke or nicotine exposure has inhibitory effects on neutrophil expression of adhesion molecules,1011 although in vivo no such clear relationship seems to exist in the literature. Regarded as a systemic disorder, one must consider the effect the disease may have on the function of the circulating inflammatory cells as well as adjacent inflammatory mediators.
Increased cellular infiltration into the lungs of COPD patients has been demonstrated for neutrophils, macrophages, dendritic cells, and T-lymphocytes,12 and during exacerbations also increased number of eosinophils.13 The inflammatory response includes multiple actions, of which the recruitment of leukocytes from the circulation constitutes a key event. Given the pivotal role for these cells at the inflammatory site, their activation state in the circulation is of vital importance; and studies141516 have demonstrated altered adhesion molecule expression, oxidative burst, and apoptosis on isolated peripheral neutrophils from COPD patients.
In view of these findings of activated peripheral neutrophils in COPD, we hypothesized that COPD may have an impact on a variety of circulating inflammatory cells claimed to play a pathophysiologic role in disease development. To test this hypothesis, we enrolled patients with stable COPD, and asymptomatic smokers and nonsmokers as control subjects. We investigated the expression of adhesion molecules and oxidative burst before and after in vitro stimulation in peripheral neutrophils, monocytes, and eosinophils.
Considering the potential influence of increased levels of circulating inflammatory mediators on peripheral cells, we used whole blood analysis and flow cytometry to minimize ex vivo modulation. Soluble inflammatory markers including C-reactive protein, sICAM-1, and TNF receptor II (TNF-RII) were run in parallel with cell analysis.
| Materials and Methods |
|---|
|
|
|---|
|
For in vitro stimulation, leukocytes were incubated for 30 min at + 37°C in TNF-
(final concentration, 10 ng/mL) [Genzyme Diagnostics; Cambridge, MA] or in buffer alone. The prestimulated cells were thereafter incubated for 15 min at + 37°C in N-formyl-methionyl-leucyl-phenylalanine (fMLP; final concentration, 106 mol/L) [Sigma; St. Louis, MO] or in buffer alone.
To assay the intracellular production of oxygen free radicals, leukocytes were incubated in 0.2 mL of 2',7'-dichlorofluorescein diacetate (final concentration, 5 µmol/L) for 15 min at + 37°C. Cells were then stimulated with fMLP, TNF-
, or buffer at times and temperatures as described above. Activation was terminated by adding 1 mL of ice-cold phosphate-buffered saline solution.
To evaluate the cell surface expression of CD11b and CD35, cells were incubated with monoclonal antibodies (100 µL; final concentration, 5 µg/mL) [CD11b-PE; DAKO A/S; Glostrup, Denmark; and CD35-FITC; Beckman Coulter; Fullerton, CA] and thereafter resuspended and kept on ice until analysis by flow cytometry. Isotype-matched control antibodies and fluorescein isothiocyanate- and phycoerythrin-conjugated IgG1 (DAKO A/S) were added in corresponding concentrations to determine the cutoff value for positive fluorescence.
Cell preparations were analyzed in a flow cytometer (EPICS XL; Beckman Coulter). Leukocyte subpopulations were distinguished by their characteristic light-scattering properties, and lymphocytes, monocytes, and granulocytes were detected as well-separated populations in a two-parameter histogram. To distinguish between neutrophils and eosinophils in the granulocyte population, measurements of depolarized side-scatter were made by inserting a dichroic plastic sheet in forward-scatter position. By subsequently shifting the following filters, analysis was made, as described previously.17
Serum Analysis
Enzyme-linked immunosorbent assay for sICAM-1 and TNF-RII (R&D Systems; Minneapolis, MN) and oxidized low-density lipoproteins (oxLDLs) [Mercodia AB; Uppsala, Sweden] were performed according to instructions of the manufacturer.
Statistics
Analyses of variance for repeated measures followed by the Tukey honest significant difference test were used unless otherwise stated. Statistical evaluation was made using statistical software (Statistica 6; StatSoft; Tulsa, OK). Differences were considered significant at p < 0.05.
| Results |
|---|
|
|
|---|
, the response to fMLP was further increased in nonsmokers. Stimulation with TNF-
alone did not affect production. There were no differences in oxidative burst in neutrophils between smokers and COPD patients (Fig 1, top left, A). The induced oxidative burst in monocytes was not as potent as in neutrophils (Fig 1, top right, B). The oxidative burst in eosinophils followed a similar pattern as neutrophils but at a lower level, and there was a reduced oxidative burst in smokers and COPD patients after fMLP stimulation, as compared to healthy control subjects (Fig 1, bottom, C). We found no correlations with measured lung function parameters (as determined by nonparametric Spearman rank-order correlation test).
|
Whole blood analysis of complement receptor CD11b did not reveal any differences among the three study groups on any of the cell populations. This was also true for cellular responsiveness, measured as mobilization of CD11b in response to in vitro stimulation with TNF-
and fMLP (Fig 2
, top left, A). All study groups mobilized the intracellular pool of CD35 equally in response to stimulation in all cell types with no synergistic effect of TNF-
and fMLP (data not shown).
|
and fMLP exposure. This was also observed in neutrophils exposed to TNF-
only (Fig 2, top right, B). There were no differences in oxidative burst in either cell population.
Serum Factors
sICAM-1 levels were significantly increased in serum from COPD patients and asymptomatic smokers compared to nonsmoking control subjects as determined by nonparametric Mann-Whitney U test (Fig 3
, left, A). Serum level of TNF-RII was similar in all study groups (Fig 3, right, B). oxLDL was measured as a marker for lipid peroxidation in plasma. We found no differences between the study groups in the total content of oxLDL or in the proportion of oxLDL in total low-density lipoprotein or cholesterol (data not shown). There were no correlations with measured lung function parameters (as determined by nonparametric Spearman rank-order correlation test).
|
| Discussion |
|---|
|
|
|---|
The airway inflammation in COPD is currently considered to generate a systemic inflammatory response. The bearing of this response on the function of peripheral leukocytes is an area poorly understood. A potential role for neutrophils in the pathophysiology of COPD is given by their substantial recruitment from circulation into the bronchial fluid, accompanied by elevated eosinophil numbers.18
An antioxidant/oxidant imbalance is proposed to be one of the major attributes of the systemic effects of COPD. This could be due to both increased production and release of oxygen free radicals from neutrophils and macrophages in the lungs, but also from oxidative particles in cigarette smoke per se. In the current study, we sought to investigate whether the imbalance in oxidative burst in COPD patients was present also in peripheral blood leukocytes.
We used TNF and fMLP as in vitro stimuli, separately and in combination. TNF is a potent proinflammatory cytokine, and fMLP is a chemo-attractant for neutrophils but can also activate eosinophils and monocytes. The rationale for using a combination of TNF and fMLP is given by the notion that TNF priming augments the fMLP-induced neutrophil response, with enhanced CD11b mobilization, chemotaxis, and cellular stiffness as a consequence.19 In line with this, neutrophils as well as eosinophils from the nonsmoking control group responded with increased oxidative burst after fMLP activation, a response that was further pronounced after the additional TNF priming. However, both COPD patients and smokers had a reduced response as compared to healthy control subjects. This was not restricted to neutrophils but was also present in monocytes and eosinophils. A reduced radical production is not in line with the general concept of increased oxidant load in COPD patients. Noguera et al15 also reported a different oxygen radical production in isolated neutrophils from COPD patients compared to control subjects. However, they reported an increased oxidative burst after stimulation with the receptor independent phorbol ester phorbol-myristate-acetate in COPD patients but not in smoking control subjects.15 The reason for this discrepancy is not fully understood but might be explained by the different methodologic approaches. Isolation of specific granulocyte population by gradient centrifugation will cause activation and degranulation,202122 and the use of different stimuli will, due to different intracellular signaling pathways, generate divergent responses.23 The choice of fluorescent probe also has impact on the result.24 In addition, 18 of 20 patients included in the present study were current smokers, while the 10 patients included in the study by Noguera et al15 were not. In this context, it is interesting to note that neutrophils exposed in vitro to extracts of cigarette smoke have been shown to decrease and delay the intracellular production and increase and accelerate the extracellular production of oxygen species in response to stimuli such as zymosan and phorbol-myristate-acetate.25 This indicates a time-dependent shift toward an extracellular production of oxygen radicals after tobacco smoke exposure. Also, alveolar macrophages from smokers have been shown to have a decreased intracellular radical production as compared to nonsmokers.26
Matheson et al8 found that smokers have increased numbers of fMLP receptors on isolated peripheral neutrophils. This might suggest an augmented responsiveness to formyl-peptides with subsequent release of oxidative products, but in the present study we did not gain data in support for this assumption. Also, the expression of complement receptors was not different between groups after fMLP activation. One might speculate that increased fMLP receptor expression on neutrophils from smokers is a consequence of decreased receptor turnover rate, which would affect the cellular response to fMLP exposure. fMLP receptor desensitization in neutrophils has been demonstrated previously,27 and we cannot rule out the possibility that an unknown mediator affects the responsiveness to fMLP in COPD patients and smokers. However, our data on the ex vivo effects of TNF priming followed by fMLP stimulation might suggest an existing refractory mechanism in leukocytes from smokers and COPD patients. The physiologic response to TNF is regulated by availability of its two receptors. The TNF receptor I (TNF-RI) is constitutively expressed on virtually all cells and is internalized on ligand binding, while TNF-RII is mostly expressed on inflammatory cells and is shed from activated cells. This given, it is possible that an ongoing inflammation would reduce the cellular TNF response. We did not find differences in serum TNF-RII levels between our study groups. Vernooy et al28 previously reported increased levels of TNF-RII and no differences in TNF-RI in clinically stable COPD patients as compared to control subjects. However, Dentener et al29 reported no differences in TNF-RII but increased levels of TNF-RI. Both these studies used ethylenediamine tetra-acetic acid plasma for analysis.
We found that both asymptomatic smokers and COPD patients have vascular involvement as measured by increased concentration of sICAM-1. This has been described earlier in serum and bronchial lavage fluid from COPD patients30 and in asymptomatic smokers.9 However, opposite results exist showing decreased sICAM-1 in both stable and exacerbated COPD.14
Neutrophil granulocytes encompass several distinct granules that are mobilized to the cell surface on appropriate stimulation. Also, monocytes and eosinophils have intracellular pools of CD11b and CD35 that can be mobilized on activation. In our study, we could not, in either cell population, find any dissimilarities among asymptomatic smokers, nonsmokers, or COPD patients in CD11b and CD35 expression at baseline or after in vitro activation. However, we found that TNF/fMLP-stimulated neutrophils, monocytes, and eosinophils from COPD patients who had smoked prior to blood sampling had a reduced CD11b mobilization capacity compared to patients who refrained from smoking. It has previously been shown that smoking generates an acute peripheral response irrespective of lung disease. For example, Selby et al10 showed that neutrophils exposed to cigarette smoke, in vivo and ex vivo, are less adherent than control cells without changes in adhesion molecule expression. The possibility that cigarette smoke affects the cytoskeleton of circulating leukocytes has also been proposed.3132 Speer et al11 demonstrated partially inhibited CD11b and totally inhibited CD11a expression in neutrophils after in vitro nicotine exposure. The few studies on acute effects of cigarette smoking in vivo on peripheral leukocyte adhesion molecule expression have shown unaltered expression of CD18 and unaltered or increased expression of L-selectin.3334 On isolated neutrophils, increased expression of CD11b has been reported previously in COPD patients both under basal and TNF stimulated conditions in smoking patients who did not smoke 8 h prior to blood sampling,14 or in patients who were nonsmokers.16
These discrepancies in cellular receptor expression and oxidative burst between different studies raise questions. First of all, current smoking habits of COPD patients and the timing of blood sampling in relation to smoking must evidently be taken into consideration when comparing studies. Secondly, when isolated by gradient centrifugation, neutrophils and monocytes degranulate with considerable up-regulation of several epitopes including CD11b on cell surfaces.202122 One ought therefore to consider that the possible subtle in vivo influence on peripheral cells in COPD patients will be enhanced or distorted after the isolation procedures. In this study, we used whole blood analysis by flow cytometry and separated different cell populations by their light-scattering properties only to minimize the impact of the cell-handling procedure on the cellular phenotype.
To summarize, COPD patients and smokers with normal lung function have a reduced intracellular oxidative burst. We found no differences in expression and mobilization of adhesion molecules, although a short-term effect of smoking was identified in cells from COPD patients after in vitro activation. Furthermore, COPD patients and smokers had increased sICAM-1. Taken together, this indicates a vascular endothelial involvement and a potential state of refractoriness in circulating phagocytes in COPD patients. And finally, these results suggest that smoking behavior must be taken into consideration while examining systemic effects of COPD.
| Acknowledgements |
|---|
| Footnotes |
|---|
This work was supported by grants from The Swedish Heart-Lung Foundation and Karolinska Institutet.
Received for publication January 31, 2005. Accepted for publication April 7, 2005.
| References |
|---|
|
|
|---|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |