Adaptive immune processes have been implicated in the pathogenesis of chronic obstructive pulmonary disease (COPD). We hypothesized that peripheral T cell abnormalities may be present in afflicted patients. We tested this hypothesis by characterizing circulating T cells in COPD patients and correlated these findings with disease severity, smoking status, and use of inhaled glucocorticosteroids (ICS). Compared with normal controls, a lesser proportion of peripheral CD4 T cells from COPD subjects produced IL-10, whereas the CD8 T cells from these patients were more often activated and more frequently produced both IFN-γ and IL-4. COPD severity was significantly and inversely associated with the proportion of circulating CD4 T cells and directly correlated with CD4 production of IL-2, as well as frequency of CD8 T cell activation and CD8 IFN-γ production. Adjustments for current smoking status and ICS use by linear regression showed independent, and generally inhibitory, effects of these clinical variables on the abnormal T cell functions of these patients. We conclude that circulating T cells from COPD patients are abnormally activated and elaborate proinflammatory mediators with admixed features of Th1 and Th2 responses. Furthermore, many of these effector processes are significantly correlated with disease severity. These findings further implicate adaptive immune processes in COPD progression and indicate that facile assays of peripheral lymphocytes may provide useful insights into disease mechanisms. Current smoking and ICS use had independent effects on T cell functions among the COPD subjects, illustrating the importance of controlling for clinical parameters as covariates in immunological studies of patients afflicted with this disease.

Chronic obstructive pulmonary disease (COPD)4 is a leading cause of death in the United States, and one of the few major diseases in which mortality continues to increase worldwide (1). A growing body of evidence suggests that adaptive immune responses play an important role in the pathogenesis and progression of COPD, as evidenced by correlations of lung histological and morphometric measurements with the characteristics of intrapulmonary lymphocyte infiltrations (2, 3, 4, 5, 6, 7, 8, 9, 10). Moreover, activated T cells can cause a variety of the tissue injuries that typify COPD by direct cytopathic effects, elaboration of diverse proinflammatory and deleterious mediators, and/or recruitment and activation of other immune and parenchymal effector cells (11, 12).

Most previous studies have examined functions of intrapulmonary T cell populations in situ or after isolation of these cells from the lungs of patients with COPD (2, 3, 4, 5, 6, 7, 8, 9, 10). However, it is also widely appreciated that lymphocytes readily traffic between inflammatory sites (including lungs), regional lymph nodes, and the systemic circulation, where they can be easily sampled (13). The ability to study disease phenomena using specimens procured by minimally invasive means could be a boon for subsequent biological and clinical investigations.

The characteristics of circulating lymphocytes in COPD patients have not yet been comprehensively defined, and the limited numbers of these previous studies are notable for often discrepant results (8, 9, 10, 14, 15, 16). We undertook an evaluation of peripheral blood lymphocytes from COPD patients for the purpose of determining whether characteristics of these cells could be reflective of the pulmonary physiological abnormality (expiratory airflow obstruction) that defines this disease (1). We found that circulating T cells of COPD patients are abnormally activated and elaborate proinflammatory cytokines that likely have relevance in the pathogenesis of this disease. Moreover, some measurements of T cell effector functions are significantly correlated with disease severity and, at least in some cases, appear to be independently affected by smoking status and inhaled glucocorticosteroid (ICS) use.

These data provide further evidence for a role of adaptive immunity in the progression of COPD and raise possibilities that focused immunomodulation could eventually be used as a specific treatment for this disease. In addition, these findings imply that easily procured circulating lymphocytes could be useful for study of disease pathogenesis and/or possibly utilized as surrogate biomarkers in early phase trials of novel therapeutics for COPD. The identification here of multiple significant clinical covariates also illustrates the importance of controlling for disease severity and other relevant parameters in immunological studies of afflicted patients.

Current and former smokers (n = 81) with no history of asthma and at least a 10-pack-year tobacco smoking history were recruited through the Emphysema-COPD Research Registry at the University of Pittsburgh (Pittsburgh, PA). All subjects included in these analyses were stable at the time of the examination and had no current clinical evidence of active infections or current or previous diagnoses of rheumatological, malignant, or other systemic inflammatory disease. COPD subjects with a dominant restrictive impairment by spirometry, a significant allergic history, completely reversible airflow obstruction, a history of clinical asthma, or current use of oral steroids were excluded. Exsmokers herein denotes COPD subjects who quit smoking at least 6 mo previously. Not using ICS denotes COPD subjects who were either never prescribed these agents or had discontinued their use at least 3 mo previously. Spirometry was performed using standard methodology and reference equations (17). Global Initiative for Obstructive Lung Disease (GOLD) scores were assigned on the basis of current pulmonary function test criteria (18). Age-matched, healthy, lifelong nonsmoking, normal subjects (n = 22) were recruited by solicitation from among hospital personnel.

Written informed consent was given by all subjects, in accordance with the University of Pittsburgh Institutional Review Board.

Blood (5–10 ml) was obtained by venipuncture, and PBMCs were isolated by density gradient centrifugation. Cell surface molecule expressions were ascertained using fluorochrome-conjugated mAbs. Unless otherwise stated, mAb and other reagents for flow cytometry, including negative isotype controls, were purchased from BD Pharmingen. Intracellular cytokine elaborations were determined in PBMC stained with anti-CD4 and anti-CD8 mAb and stimulated for 5 h with PMA (5 ng/ml) and calcium ionomycin (500 ng/ml) in the presence of GolgiStop. Cells were fixed, permeabilized, and stained with fluorochrome-conjugated mAbs against selected cytokines.

Flow cytometry was performed on ≥10,000 live cells after gating on T cell subpopulations of interest (CD4+ and CD8+) and analyzed using a BD FACSAria (BD Biosciences). In particular, phenotypic and functional characteristics were determined specifically among the CD4bright and CD8bright cell populations, and their T cell identities were confirmed by anti-CD3 staining (Fig. 1).

FIGURE 1.

Flow cytometry methodology. A, Characterizations of circulating CD4 T cells (in this example, the proportion of these cells that coexpress MHC class II) were performed by gating on CD4bright cells (upper and lower right quadrants). B, Further analysis of these CD4bright cells (———) showed that they also ubiquitously expressed CD3 (confirming their identity as CD4 T cells). The CD4dim and CD4 subpopulations here (– – – –), gated from A (upper and lower left quadrants), also included a comparatively small proportion of CD3+ cells (i.e., the CD8 T cells). C, Characterizations of circulating CD8 T lymphocytes (in this example, the proportion of these cells that coexpress MHC class II) were performed by gating on CD8bright cells (upper and lower right quadrants). D, Further analysis of these CD8bright cells (———) showed that they ubiquitously express CD3, confirming their identity as CD8 T cells. The CD8dim and CD8 (– – – –), gated from C (upper and lower left quadrants) also included another CD3+ population (i.e., the CD4 T cells). APC, Allophycocyanin.

FIGURE 1.

Flow cytometry methodology. A, Characterizations of circulating CD4 T cells (in this example, the proportion of these cells that coexpress MHC class II) were performed by gating on CD4bright cells (upper and lower right quadrants). B, Further analysis of these CD4bright cells (———) showed that they also ubiquitously expressed CD3 (confirming their identity as CD4 T cells). The CD4dim and CD4 subpopulations here (– – – –), gated from A (upper and lower left quadrants), also included a comparatively small proportion of CD3+ cells (i.e., the CD8 T cells). C, Characterizations of circulating CD8 T lymphocytes (in this example, the proportion of these cells that coexpress MHC class II) were performed by gating on CD8bright cells (upper and lower right quadrants). D, Further analysis of these CD8bright cells (———) showed that they ubiquitously express CD3, confirming their identity as CD8 T cells. The CD8dim and CD8 (– – – –), gated from C (upper and lower left quadrants) also included another CD3+ population (i.e., the CD4 T cells). APC, Allophycocyanin.

Close modal

Although the numbers of available cells, or reagents in some cases, did not allow all assays to be performed in all specimens, these measures were made concurrently in consecutive specimens from both study groups, while laboratory investigators were blinded to the subject’s identities, clinical presentations (including disease severities), and demographic characteristics.

T cell activations and mediator productions were compared between normal controls and COPD subjects by the Mann-Whitney U test. Comparisons of parametrically distributed continuous data among three or more groups (e.g., ages among GOLD stages) were analyzed by ANOVA, with post hoc comparisons by Bonferroni-Dunn. Dichotomous variables (e.g., proportions of smoking patients) were analyzed by χ2. Associations between continuous values (e.g., forced expiratory volume in 1 s, as a percent of predicted values; FEV1%p) and T cell expressions or cytokine productions were first established by linear regression, with adjustment for current smoking and ICS use by multivariate linear regression. Data were analyzed using StatView (SAS) or Stata Statistical Software: Release 9 (StataCorp). Data are depicted as mean ± SEM. Significance was defined as p < 0.05.

COPD patient characteristics are shown in Table I. There were no systematic associations between either age or smoking duration and the magnitude of physiological impairments. Subjects with the most severe physiological impairments, however, were more likely to have quit smoking and to use ICS. The ages of normal controls (56 ± 1 years old) were similar to those of COPD subjects. All of the subjects studied here were Caucasian except one African American with COPD (GOLD 2).

Table I.

Demographics of the COPD populationa

GOLD Disease Severity ScoresnAge (yr)bGender (% male)FEV1%pcFEV1/FVCcSmoking History (pack-yr)% Currently Smokingd% ICS Used
12 59 ± 2 58 92 ± 2 0.63 ± 0.02 53 ± 8 58 17 
26 65 ± 1 62 65 ± 2 0.56 ± 0.02 61 ± 7 35 31 
26 67 ± 1 62 39 ± 1 0.39 ± 0.02 54 ± 4 31 58 
17 61 ± 2 53 21 ± 1 0.29 ± 0.01 52 ± 4 65 
Total (aggregate population) 81 64 ± 1 59 51 ± 3 0.46 ± 0.02 56 ± 3 31 44 
GOLD Disease Severity ScoresnAge (yr)bGender (% male)FEV1%pcFEV1/FVCcSmoking History (pack-yr)% Currently Smokingd% ICS Used
12 59 ± 2 58 92 ± 2 0.63 ± 0.02 53 ± 8 58 17 
26 65 ± 1 62 65 ± 2 0.56 ± 0.02 61 ± 7 35 31 
26 67 ± 1 62 39 ± 1 0.39 ± 0.02 54 ± 4 31 58 
17 61 ± 2 53 21 ± 1 0.29 ± 0.01 52 ± 4 65 
Total (aggregate population) 81 64 ± 1 59 51 ± 3 0.46 ± 0.02 56 ± 3 31 44 
a

GOLD disease severity scores were assigned on the basis of current pulmonary function test criteria (17, 18). FEV1/FVC, ratio of FEV1 and forced expiratory volume.

b

p = 0.004, GOLD 3 vs GOLD 1 (ANOVA);

c

p < 0.001 (GOLD 4 vs all others);

d

p < 0.015 in comparisons among GOLD stages (χ2).

Quantitative assays of peripheral T cells were initially performed to test for global differences between COPD subjects and healthy volunteers.

The proportion of CD4 T cells among PBMCs was slightly, albeit significantly, decreased in the COPD subjects compared with normal subjects, whereas proportions of CD8 T cells were equivalent in both groups (Fig. 2 A). These decrements in CD4 T cells were reflected by a possible (albeit insignificant) trend for decreased CD4:CD8 ratios in the COPD subjects (3.3 ± 0.3 vs 4.3 ± 1.0, for COPD and normal subjects, respectively).

FIGURE 2.

Characteristics of circulating T cells in COPD patients. Comparisons between the aggregate COPD population (composed of subjects with varying disease severity; see Table I) vs healthy normal subjects. A, Proportions of CD4 among PBMCs (PBMNC) were decreased in COPD subjects (p = 0.009), whereas there were no significant differences in CD8 percentages. B, Activation (per MHC class II coexpression) of CD8 T cells were greater in COPD patients (n = 49) than in controls (p = 0.0002). There may also be a trend for increased activation of CD4 T cells among COPD subjects, but this comparison with controls did not reach statistical significance (p = 0.07). C, Circulating CD4 T cells in COPD patients less frequently produced IL-10 (n = 29) than those of normal subjects (p = 0.009). D, IFN-γ and IL-4 production was more frequent in COPD CD8 T cells (n = 49) than in normal subjects (p = 0.0006 and p = 0.001, respectively). All intergroup comparisons were made using Mann-Whitney rank-sum tests.

FIGURE 2.

Characteristics of circulating T cells in COPD patients. Comparisons between the aggregate COPD population (composed of subjects with varying disease severity; see Table I) vs healthy normal subjects. A, Proportions of CD4 among PBMCs (PBMNC) were decreased in COPD subjects (p = 0.009), whereas there were no significant differences in CD8 percentages. B, Activation (per MHC class II coexpression) of CD8 T cells were greater in COPD patients (n = 49) than in controls (p = 0.0002). There may also be a trend for increased activation of CD4 T cells among COPD subjects, but this comparison with controls did not reach statistical significance (p = 0.07). C, Circulating CD4 T cells in COPD patients less frequently produced IL-10 (n = 29) than those of normal subjects (p = 0.009). D, IFN-γ and IL-4 production was more frequent in COPD CD8 T cells (n = 49) than in normal subjects (p = 0.0006 and p = 0.001, respectively). All intergroup comparisons were made using Mann-Whitney rank-sum tests.

Close modal

We further hypothesized that if T cells were involved in ongoing inflammatory responses in the COPD patients, there might be evidence of activation among the circulating lymphocytes. Peripheral CD8 T cells of the COPD subjects were more frequently activated than those among normal subjects, as ascertained by MHC class II expression (HLA-DR; Fig. 2 B). There was a trend too for greater relative proportions of activated CD4 in the COPD patients, although this difference did not reach statistical significance.

Intracellular cytokine determinations also often showed intergroup differences of mediator elaborations in COPD peripheral lymphocytes. CD4 T cells of COPD subjects less frequently produced IL-10 than corresponding cells of healthy controls (Fig. 2 C). The proportion of CD4 T cells that produce IL-2 (31.7 ± 2.1%), IFN-γ (8 ± 2%), and IL-4 (<1.5%) were near identical in both subject populations.

The frequency of IL-2 productions by CD8 T cells (16.0 ± 1.3%) tended to be greater among COPD patients than those from normal subjects (11.5 ± 1.8%), and this difference approached but did not reach significance (p = 0.06). Both IFN-γ and IL-4 were more frequently produced by the CD8 T cells of COPD patients than the corresponding cells from normal subjects (Fig. 2 D).

We hypothesized that if adaptive immune mechanisms are important in COPD pathogenesis, associations of immunological assays with disease severity may be evident. Moreover, the presence of correlations between T cell functions would not only implicate these immune responses in disease progression but also show that immunological assays in heterogeneous COPD populations could be confounded by the particular case-mix of disease severities.

Our analyses showed that significant correlations between several T cell parameters and disease extent were present in the COPD patients. The percentages of CD4 T cells among PBMCs decreased, and the frequency of their IL-2 production increased, in proportion to the extent of pulmonary dysfunction (Fig. 3, A and B, respectively). Both HLA-DR expression and IFN-γ production of circulating COPD CD8 T cells were also directly associated with disease severity (Fig. 3, C and D). There were no significant correlations between FEV1%p and CD8 proportions among PBMCs, CD4 MHC class II or IFN-γ frequencies, CD8 IL-2 production, or IL-4 and IL-10 elaborated by either T cell subpopulation.

FIGURE 3.

Immune functions of COPD peripheral T cells correlate with disease severity. Significant correlations (by linear regression analyses) were evident between pulmonary function (FEV1) of COPD patients and characteristics of circulating T cells in these subjects. The FEV1 values of individual COPD patients were significantly correlated with the proportions of CD4 T cells among their circulating PBMCs (PBMNC; A), the frequency of IL-2 production by their CD4 T cells (B), and the frequency with which their CD8 T cells expressed MHC class II (C) and produced IFN-γ (D).

FIGURE 3.

Immune functions of COPD peripheral T cells correlate with disease severity. Significant correlations (by linear regression analyses) were evident between pulmonary function (FEV1) of COPD patients and characteristics of circulating T cells in these subjects. The FEV1 values of individual COPD patients were significantly correlated with the proportions of CD4 T cells among their circulating PBMCs (PBMNC; A), the frequency of IL-2 production by their CD4 T cells (B), and the frequency with which their CD8 T cells expressed MHC class II (C) and produced IFN-γ (D).

Close modal

Although not extensively studied within COPD populations per se, cigarette smoking has been shown elsewhere to exert multiple effects on many adaptive immune functions (19). Moreover, in our population of COPD patients, those subjects with the most severe impairment in lung function were also less likely to have continued smoking. Therefore, to examine the potential that continued smoking may have confounded the relation between pulmonary impairment and T cell functions among COPD patients, we performed multivariate analyses, using FEV1%p and smoking status (current vs former) as independent covariates, in evaluations of those immune parameters that were associated with disease severity.

Direct comparisons of still smoking and ex-smoking cohorts stratified by disease severity suggested that current, continued smoking was associated with tendencies to blunt cytokine elaborations and CD8 T cell activation, particularly among those subjects with the most advanced disease (Fig. 4,A). The addition of current smoking status as a second independent variable strengthened the associations between FEV1%p with CD4 proportions among PBMCs (r = 0.35, p = 0.007) and CD8 activation (r = 0.42, p = 0.012; compare with Fig. 3, A and C, respectively). Post hoc regression analyses limited solely to exsmokers also showed that the correlations between productions of IL-2 (among CD4 T cells) and IFN-γ (by CD8 T cells) with disease severity tended to be stronger in this subpopulation (Fig. 4, B and C) than in the aggregate COPD cohort (compare with Fig. 3, B and D, respectively).

FIGURE 4.

Smoking status affects immune functions of COPD patients with severe disease. A, In addition to multivariate analyses (see text), evaluations of immunological functions among cohorts of COPD patients within a given disease severity category indicated that continued smoking tended to affect assays of T cell functions. These comparisons necessarily used subjects with similar disease severities, because pulmonary function abnormalities were, of themselves, independent correlates of many T cell functions (see Fig. 3). In general, subjects who were currently smoking had less T cell activation and cytokine production than patients with similar disease severity who had quit smoking. Given the comparatively small numbers of subjects within any particular disease severity category (in this case those subjects with GOLD 3 disease), however, only one of the comparisons here (the frequency of MHC class II expression by CD8 T cells) approached statistical significance (p = 0.054) using nonparametric comparisons (Mann-Whitney rank-sum test). B, Post hoc linear regression analyses limited to nonsmoking COPD subjects often showed stronger immunological-clinical correlations than the analogous evaluations of the aggregate COPD population (that included both smoking and nonsmoking subjects), further indicating that current smoking alters and confounds these assays. As an example, the correlation between FEV1 (as percent predicted values) of individual, nonsmoking COPD patients and the proportions of CD4 T cells among their PBMCs (PBMNC) was greater than the corresponding correlation of the total, aggregate COPD patient group (both smokers and nonsmokers) depicted in Fig. 3,B. C, Similarly, the association between FEV1%p and the frequency of IFN-γ (IFN-g) production by their CD8 T cells was greater for nonsmoking COPD subjects than corresponding analyses of these parameters among the aggregate COPD population (compare with Fig. 3 D).

FIGURE 4.

Smoking status affects immune functions of COPD patients with severe disease. A, In addition to multivariate analyses (see text), evaluations of immunological functions among cohorts of COPD patients within a given disease severity category indicated that continued smoking tended to affect assays of T cell functions. These comparisons necessarily used subjects with similar disease severities, because pulmonary function abnormalities were, of themselves, independent correlates of many T cell functions (see Fig. 3). In general, subjects who were currently smoking had less T cell activation and cytokine production than patients with similar disease severity who had quit smoking. Given the comparatively small numbers of subjects within any particular disease severity category (in this case those subjects with GOLD 3 disease), however, only one of the comparisons here (the frequency of MHC class II expression by CD8 T cells) approached statistical significance (p = 0.054) using nonparametric comparisons (Mann-Whitney rank-sum test). B, Post hoc linear regression analyses limited to nonsmoking COPD subjects often showed stronger immunological-clinical correlations than the analogous evaluations of the aggregate COPD population (that included both smoking and nonsmoking subjects), further indicating that current smoking alters and confounds these assays. As an example, the correlation between FEV1 (as percent predicted values) of individual, nonsmoking COPD patients and the proportions of CD4 T cells among their PBMCs (PBMNC) was greater than the corresponding correlation of the total, aggregate COPD patient group (both smokers and nonsmokers) depicted in Fig. 3,B. C, Similarly, the association between FEV1%p and the frequency of IFN-γ (IFN-g) production by their CD8 T cells was greater for nonsmoking COPD subjects than corresponding analyses of these parameters among the aggregate COPD population (compare with Fig. 3 D).

Close modal

Comparisons of cohorts stratified by disease severity indicated that current ICS use tended to diminish cytokine production and CD8 T cell activation among severely afflicted subjects (Fig. 5,A). Multivariate analyses, with current ICS usage as an independent variable, strengthened the association between FEV1%p and IL-2 production by CD4 T cells of COPD patients (r = 0.39, p = 0.017; compare with Fig. 3,B). Additional post hoc analyses, limited solely to the COPD subjects not currently taking ICS agents, again showed that the correlation between CD8 IFN-γ production and FEV1%p was strengthened by exclusion of ICS-treated subjects (Fig. 5,B). Although there were too few subjects to rigorously analyze and adjust for combinations of independent variables, the potential for complex, simultaneously confounding effects of both smoking status and ICS use could be indicated by the stronger correlations between immune functions and disease severity in analyses limited to exsmoking COPD subjects who also were not using ICS (Fig. 5 C).

FIGURE 5.

ICS use affects immune functions of COPD patients with severe disease. A, In addition to multivariate analyses (see text), evaluations of immunological functions among COPD patients with similar magnitudes of disease severities indicated that ICS use tended to have systemic effects on T cell functions. As an example: an analysis of the COPD cohort with the most severe disease (GOLD 4) suggested that T cells among the subjects within this population who were taking ICS (ICS Use) had decreased T cell activation and cytokine productions, compared with similarly affected patients who were not taking these agents (No ICS). Again, however, given the comparatively small numbers of subjects within any particular disease severity category, only one of these comparisons (the frequency of IFN-γ (IFN-g) production by circulating CD8 T cells) approached statistical significance (p = 0.08) using nonparametric comparisons (Mann-Whitney rank-sum test). B, Linear regression analyses limited to COPD subjects not taking ICS often showed strengthened immunological-clinical correlations, compared with evaluations of the aggregate COPD population (further indication of potential effects and confounding by ICS). As an example, the correlation of FEV1 and CD8 T cell expression of MHC class II was stronger and more significant among the COPD subjects not taking ICS depicted here, compared with the aggregate, total COPD population that included both medicated and nonmedicated subjects (compare with Fig. 3,D). C, Post hoc analyses limited to COPD subjects who were neither smoking nor taking ICS often resulted in the strongest and most significant correlations of T cell measures with pulmonary function, further illustrating the complex effects and potential confounding of both smoking and drug use on measures of T cell functions. As an example, the association between FEV1%p and CD8 T cell MHC class II expression of the nonsmoking COPD patients who were also not taking ICS was stronger than analogous analyses using the aggregate COPD population (see Fig. 3 C) or those limited to the nonsmoking patients (with variable ICS use), as depicted in B. PBMNC is PBMC.

FIGURE 5.

ICS use affects immune functions of COPD patients with severe disease. A, In addition to multivariate analyses (see text), evaluations of immunological functions among COPD patients with similar magnitudes of disease severities indicated that ICS use tended to have systemic effects on T cell functions. As an example: an analysis of the COPD cohort with the most severe disease (GOLD 4) suggested that T cells among the subjects within this population who were taking ICS (ICS Use) had decreased T cell activation and cytokine productions, compared with similarly affected patients who were not taking these agents (No ICS). Again, however, given the comparatively small numbers of subjects within any particular disease severity category, only one of these comparisons (the frequency of IFN-γ (IFN-g) production by circulating CD8 T cells) approached statistical significance (p = 0.08) using nonparametric comparisons (Mann-Whitney rank-sum test). B, Linear regression analyses limited to COPD subjects not taking ICS often showed strengthened immunological-clinical correlations, compared with evaluations of the aggregate COPD population (further indication of potential effects and confounding by ICS). As an example, the correlation of FEV1 and CD8 T cell expression of MHC class II was stronger and more significant among the COPD subjects not taking ICS depicted here, compared with the aggregate, total COPD population that included both medicated and nonmedicated subjects (compare with Fig. 3,D). C, Post hoc analyses limited to COPD subjects who were neither smoking nor taking ICS often resulted in the strongest and most significant correlations of T cell measures with pulmonary function, further illustrating the complex effects and potential confounding of both smoking and drug use on measures of T cell functions. As an example, the association between FEV1%p and CD8 T cell MHC class II expression of the nonsmoking COPD patients who were also not taking ICS was stronger than analogous analyses using the aggregate COPD population (see Fig. 3 C) or those limited to the nonsmoking patients (with variable ICS use), as depicted in B. PBMNC is PBMC.

Close modal

The primary focus of these investigations was to determine whether peripheral adaptive immune functions in COPD patients correlated with the severity of airflow obstruction. Indeed, we observed several associations between circulating T cell characteristics and pulmonary physiology in these patients (Fig. 3). These correlations are consistent with, and further implicate, a role of adaptive immune processes in COPD progression.

COPD is a complex heterogeneous syndrome with an incompletely understood pathogenesis (2, 3, 4, 5, 6, 7, 8, 9, 10, 13, 14, 15, 16, 20, 21, 22, 23, 24, 25, 26, 27). Long-term exposure to tobacco smoke is unquestionably the primary risk factor for development of the abnormal expiratory airflow obstruction that characterizes this disease (1). Nonetheless, severe COPD occurs in only a minority of long-term cigarette smokers, and illness susceptibility appears to be highly variable, even among subjects with similar ages and smoking histories (Table I), indicating that superimposed processes are the final determinants of disease development and/or progression (1, 20). Furthermore, abnormal intrapulmonary inflammation persists, and COPD typically progresses, long after smoking cessation (21). In addition, numerous systemic disease manifestations with a potential inflammatory basis have been associated with COPD, including skeletal muscle wasting and osteoporosis, as well as increased risks of cardiovascular disease that cannot be fully accounted for by cigarette smoking per se (22).

The potential importance of adaptive immunological processes in COPD has lately become a focus of considerable attention (2, 3, 4, 5, 6, 7, 8, 9, 10, 13, 14, 15, 16, 23, 24, 25, 26, 27, 28, 29). Among other observations, T lymphocytes are the most numerous inflammatory cell within alveolar walls of COPD patients, and the extent of intrapulmonary lymphocyte infiltrations in situ is closely correlated with various morphometric and physiological measures of disease severity (2, 3, 4, 5, 6). In general, CD8 T cells predominate in the diffuse parenchymal infiltrates within COPD lungs, although large numbers of CD4 lymphocytes are present proximate to focal lymphoid aggregates (6, 23). Findings of oligoclonal expansions among lymphocytes from COPD patients show these cells are activated and undergo proliferations in response to conventional peptide Ags (23, 24, 25). The identity(ies) of the Ag(s) that trigger and fuel these immune processes has not yet been established, although microbes that frequently (and/or chronically) colonize or infect COPD lungs have been indirectly implicated as potential initiators of immune responses (21, 28, 29). In turn, these antimicrobial responses could possibly generalize via processes of mimicry and/or epitope spreading (30). Autoreactive cellular and humoral responses directed against lung endothelium, epithelium, and connective tissue elements appear to be frequent in COPD patients (26, 27), and these processes could also conceivably contribute to or cause systemic manifestations of the disease (22).

Functional characterizations of intrapulmonary lymphocytes in situ, or among cells directly isolated from pulmonary specimens of diseased patients, include observations of activation and specific, and usually Th1 predominant, mediator elaborations, most consistently and notably including production of IFN-γ (2, 3, 4, 5, 6, 7). However, contrary findings have also been reported (8, 10).

Only a limited number of previous investigations have specifically examined the characteristics and functions of circulating T cells among COPD patients, and findings of these studies are often considerably disparate (8, 9, 10, 13, 14, 15, 16). However, direct comparisons of these reports may be complicated by their considerable differences of experimental methodologies, frequent use of small COPD subject cohorts, and variable extents of controlling for clinical parameters, including disease severities.

Circulating CD4 T cells from 20 COPD patients (all were exsmokers) with highly varying disease severities (FEV1%p range, 19–80), were reported in one study to more frequently elaborate IFN-γ, but less frequently produce IL-4, compared with normal preparations, whereas there were no apparent abnormalities of CD8 T cell cytokine productions among the former (14). In the largest previous study to date (n = 70), but limited to exsmoking subjects with GOLD stage II (FEV1%p, 54 ± 8), the frequency of T cell IL-2 and IL-4 productions among the COPD patients were equivalent to those of normal subjects (16). These investigators also reported the seemingly paradoxical finding that CD8 T cells from their diseased subjects less frequently produced IFN-γ. More recently, analyses of circulating T cells from four exsmoking patients with predominantly severe COPD (median FEV1%p, 36) did not find significant differences of either IFN-γ or IL-4 production, in comparison with four subjects who had normal pulmonary function (two current smokers and two exsmokers; Ref. 10). Another study of 14 patients with moderate lung disease (FEV1%p, 57 ± 2), none of whom had smoked within 12 h of testing (6 subjects were taking ICS), similarly did not find significant intergroup differences in elaborations of IL-2, IFN-γ, IL-4, or IL-10 by circulating COPD T cells, in comparison with specimens from healthy normal subjects (8). A characterization of peripheral T cell subsets among 20 currently smoking COPD subjects (FEV1%p, 45 ± 3) found no overall changes in T cell expression of HLA-DR, proportions of CD4+ or CD8+ T cells, or the CD4:CD8 ratio (15).

We believe our study of circulating T cells in COPD subjects has several unique features. The use of a large number of COPD subjects increases the power of the present analyses to detect intergroup differences and minimize type II (β) errors. Inclusion here of subjects with a broad (and delineated) range of disease severities (Table I) also facilitates analyses of clinical-immunological correlations (Figs. 3; 4, B and C; and 5, B and C).

The findings here, showing that the magnitude of physiological abnormalities are associated with T cell functions, notably demonstrate that global analyses of adaptive immune parameters in COPD-affected patients can be heavily skewed by the case-mix distributions of disease severities in these study populations. As an example using the current data, the frequency of IL-2 production by CD4 T cells among the aggregate COPD population here with diverse disease severities was not significantly greater than that of the normal controls (28.3 ± 2.7%). However, CD4 IL-2 production was directly correlated with COPD severity (Figs. 3,A and 4 B). Hence, study here of a patient cohort composed of only those with severe disease (i.e., GOLD 3 and 4) would have resulted in assertions that 37.8 ± 3.3% of circulating CD4 T cells among COPD subjects produce IL-2, and this value is significantly and abnormally greater than that of healthy controls (p < 0.05). Conversely, however, study of a COPD subpopulation instead composed solely of those with mild-to-moderate disease severity (i.e., GOLD 1–2) would have resulted in finding the frequency of IL-2 production among their circulating CD4 T cells was 27.3 ± 2.5% and led us to an equally (albeit opposite) erroneous conclusion that the value for COPD patients is nearly identical with that of normal individuals (16). Thus, attempts to define particular immunological measures as a constant or meaningful function for all COPD patients seem problematic and possibly misleading. The potential ascertainment biases due to differences of disease severities among their experimental populations may explain at least some of the discrepant results of earlier COPD studies (8, 9, 10, 13, 14, 15, 16).

The present data showing generally increased activation among circulating T cells in COPD subjects (Fig. 2,B) also appear to be an unprecedented finding and a possible result of the power afforded by study of comparatively greater subject numbers. Moreover, this particular observation is also consistent with previous findings of increased activations of in situ T cells within COPD lungs (2, 3, 7, 9). Proportions of circulating CD4 T cells were also seen here to decrease among those more severe COPD (Fig. 3 A), a possible result of these cells being sequestered within the bronchus-associated lymphoid aggregates most often found in patients with advanced disease (6). The present findings of increased IFN-γ production by circulating COPD CD8 T cells, although distinctly contrary to results of an analysis limited to GOLD 2 subjects (16), is consistent with most in situ observations (2, 5, 7). The likely singular importance of IFN-γ in COPD pathogenesis is underscored by the many demonstrable proinflammatory effects of this mediator, including activation and recruitment of epithelial and inflammatory cells, up-regulation of matrix metalloproteases and other injurious mediators (2, 12), as well as inferences derived from animal models (31).

Overall, the current data suggest that a complex cytokine network process is mediated by the T cells of COPD patients. A Th1 bias is indicated here by findings of increased IFN-γ- and IL-2-producing lymphocytes, along with concurrent decreases of IL-10-producing cells in the COPD subjects, and these observations are consistent with previous observations of a Th1 predominance in situ (2, 5, 7). However, the increased frequency of IL-4-producing CD8 T cells in the COPD patients also reflect a Th2-associated process and corroborates a trend reported previously (10). This particular observation is also concordant with earlier reports of heightened productions of IL-4 and/or other Th2 cytokines among COPD bronchoalveolar lavageT cells (8, 10) and could have important pathogenic implications, given the diverse, but generally profibrotic, and other injurious effects of this cytokine (32). Present data showing that IL-10 is less frequently elaborated by circulating COPD CD4 T cells could also have implications relevant to modulation (e.g., down-regulation) of adaptive immune responses among the diseased subjects (33). Corroboration and further study of potential immunoregulatory mechanisms in COPD seem indicated, including more definitive numerical and functional ascertainments of regulatory T cell subpopulations (26, 34).

Our analysis of smoking effects on T cell functions among a population of afflicted COPD patients also appears to be a novel approach. Tobacco smoking has been previously shown, by in vitro investigations and comparative studies of other subject populations (e.g., individuals with normal pulmonary function), to depress a variety of T cell functions, putatively due to toxicities of acrolein and/or other constituents of complex smoke aerosols (19, 35). Associations between disease severity and T cell parameters here were generally more rigorous in post hoc analyses limited to the nonsmoking COPD patients (Figs. 4, B and C; and 5 C), suggesting that current smoking has substantial confounding effects on some measurements of immune parameters. Results of the multivariate linear regression models here also confirm an independent influence of smoking status on immunological processes in COPD patients, even after adjusting for disease severity, including decrements in the proportion of circulating CD4 T cells and frequency of CD8 T cell activation.

The present study is also relatively unique in evaluations of inhaled corticosteroid effects by use of comparatively large numbers of afflicted COPD patients, with adjustments for disease severity. Previous investigations of ICS have yielded conflicting results regarding systemic immunological effects of these drugs (36, 37, 38, 39). We found statistically significant effects of ICS use on the frequency of IL-2 production among circulating CD4 T cells of COPD patients. In addition, possible trends for other systemic immunological effects of these agents were also indicated (Fig. 5), suggesting a possible focus for future and more powerful studies with even larger numbers of subjects that also need to incorporate considerations for confounding by disease severity and smoking status.

Assays using relevant immune effector cells from blood may ultimately prove to be a facile methodology for study of COPD pathogenesis, particularly in exploratory investigations of disease mechanisms, or in patients from whom more invasive specimen procurements are not feasible (e.g., those with severe and/or unstable pulmonary dysfunction). These particular measures may be especially useful for replicate, serial assays among individual subjects in longitudinal cohort studies. The correlations of immune functions with disease severity imply that certain of these assays could possibly be useful as clinically relevant biomarkers. As an example, the central importance of IFN-γ in modulating key pathophysiological processes in COPD patients seems increasingly likely (2, 5, 7, 12, 31), and longitudinal determinations of this cytokine production by peripheral T cells might eventually be proven a simple and useful correlate of disease progression.

Furthermore, detection of clinical effects in COPD treatment trials is inherently difficult and expensive, requiring large numbers of subjects and protracted observations, because most disease manifestations, including decrements of expiratory airflow, occur relatively slowly, and often have considerable interindividual variability (1, 40). If further validated by longitudinal study, it is conceivable that assays of peripheral immune functions (e.g., IFN-γ or other mediators) could eventually be used as indications of treatment efficacy in early-phase therapeutic trials (e.g., changes of IFN-γ production with therapy) and hence be a valuable adjunct to other clinical and physiological end-point determinations.

In summary, these data show that numerous, pathologically relevant immunological functions of circulating T cells are abnormal and correlated with COPD severity and thus further support paradigms that posit that adaptive immune processes contribute to progression of this disease. These findings also show that immunological assays using COPD patient specimens can be heavily influenced by the distribution of disease severities, as well as smoking status and ICS use. Hence, studies to define immunological responses among patients with this complex disease need to incorporate appropriate considerations for these confounding clinical factors, and/or inclusion of much larger subject numbers than has been typical in most previous investigations. Ultimately, better understanding of the role(s) that adaptive immune mechanisms play in COPD progression will likely lead to development of novel treatments that could have greater therapeutic efficacy for patients with this otherwise inexorable disease.

We appreciate the technical assistance of Dr. Yifang Song, Emily Lindsay, and Alison Logar, and the personnel of the Emphysema-COPD Research Center at the University of Pittsburgh who made these studies possible.

Dr. Sciurba receives less than $10,000 annually in consultancies from AstraZeneca, GlaxoSmithKline, and Novartis. None of the other coauthors have potential COI.

The costs of publication of this article were defrayed in part by the payment of page charges. This article must therefore be hereby marked advertisement in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.

1

This research was supported in part by National Institutes of Health Grants 1R01HL073241 and 1 P50 HL084948.

4

Abbreviations used in this paper: COPD, chronic obstructive pulmonary disease; ICS, inhaled glucocorticosteroid; GOLD, Global Initiative for Obstructive Lung Disease; FEV1%p, forced expiratory volume in 1 s, as a percent of predicted values.

1
Mannino, D. M..
2002
. COPD: epidemiology, prevalence, morbidity and mortality, and disease heterogeneity.
Chest
121
:
121S
-126S.
2
Grumelli, S., D. B. Corry, L. Z. Song, L. Song, L. Green, J. Huh, J. Hacken, R. Espada, R. Bag, D. E. Lewis, F. Kheradmand.
2004
. An immune basis for lung parenchymal destruction in chronic obstructive pulmonary disease and emphysema.
PLoS Med.
1
:
75
-83.
3
O'Shaughnessy, T. C., T. W. Ansari, N. C. Barnes, P. K. Jefferey.
1997
. Inflammation in bronchial biopsies of subjects with chronic bronchitis: inverse relationship of CD8+ T-lymphocytes with FEV1.
Am. J. Respir. Crit. Care Med.
155
:
852
-857.
4
Di Stefano, A., A. Capelli, M. Lusuardi, P. Balbo, C. Vecchio, P. Maestrelli, C. E. Mapp, L. M. Fabbri, C. F. Donner, M. Saetta.
1998
. Severity of airflow limitation is associated with severity of airway inflammation in smokers.
Am. J. Respir. Crit. Care Med.
158
:
1277
-1285.
5
Saetta, M., M. Mariani, P. Panina-Bordignon, G. Turato, C. Buonsanti, S. Baraldo, C. M. Bellattato, A. Papi, L. Corbetta, R. Zuin, F. Sinigaglia.
2002
. Increased expression of the chemokine receptor CXCR3 and its ligand CXCL10 in peripheral airways of smokers with chronic obstructive pulmonary disease.
Am. J. Respir. Crit. Care Med.
165
:
1404
-1409.
6
Hogg, J. C., F. Chu, S. Utokaparch, R. Woods, W. M. Elliott, L. Buzatu, R. M. Cherniack, R. M. Rogers, F. C. Sciurba, H. O. Coxson, P. D. Pare.
2004
. The nature of small-airway obstruction in chronic obstructive pulmonary disease.
N. Engl. J. Med.
350
:
2645
-2653.
7
Di Stefano, A., G. Caromori, A. Capelli, I. Gnemmi, F. Ricciardolo, T. Oates, C. F. Donner, K. F. Chung, P. J. Barnes, I. M. Adcock.
2004
. STAT4 activation in smokers and patients with chronic obstructive pulmonary disease.
Eur. Respir. J.
24
:
78
-85.
8
Barcelo, B., J. Pons, A. Fuster, J. Sauleda, A. Noguera, J. M. Ferrer, A. G. Augusti.
2006
. Intracellular cytokine profile of T lymphocytes in patients with chronic obstructive pulmonary disease.
Clin. Exp. Immunol.
145
:
474
-479.
9
Leckie, M. J., G. R. Jenkins, J. Khan, S. J. Smith, C. Walker, P. J. Barnes, T. T. Hansel.
2006
. Sputum lymphocytes in asthma, COPD, and healthy subjects have the phenotype of activated intraepithelial T cells (CD69+CD103+).
Thorax
58
:
23
-29.
10
Barczyk, A., W. Pierzchaia, O. M. Kon, B. Cosio, I. M. Adcock, P. J. Barnes.
2006
. Cytokine production by bronchoalveolar lavage T lymphocytes in chronic obstructive pulmonary disease.
J. Allergy Clin. Immunol.
117
:
1484
-1492.
11
Monaco, C., E. Andreakos, S. Kiriakidis, M. Feldman, C. Paleolog.
2004
. T-cell mediated signaling in immune, inflammatory and angiogenic processes: the cascade of events leading to inflammatory diseases.
Curr. Drug Targets Inflamm. Allergy
3
:
35
-42.
12
Curtis, J. L..
2005
. Cell-mediated adaptive immune defense of the lungs.
Proc. Am. Thorac. Soc.
2
:
412
-416.
13
Lehmann, C., A. Wilkening, D. Leiber, A. Markus, N. Krug, R. Pabst, T. Tschernig.
2001
. Lymphocytes in the bronchoalveolar space reenter the lung tissue by means of the alveolar epithelium, migrate to regional lymph nodes, and subsequently rejoin the systemic immune system.
Anat. Rec.
264
:
229
-236.
14
Majori, M., M. Corradi, A. Caminati, G. Cacciani, S. Bertacco, A. Pesci.
1999
. Predominant TH1 cytokine pattern in peripheral blood from subjects with chronic obstructive pulmonary disease.
J. Allergy Clin. Immunol.
103
:
458
-462.
15
Kim, W. D., W. S. Kim, Y. Koh, S. D. Lee, C. M. Lim, D. S. Kim, Y. J. Cho.
2002
. Abnormal peripheral blood T-lymphocyte subsets in a subgroup of patients with COPD.
Chest
122
:
437
-444.
16
Reyes, E., A. Prieto, A. de la Hera, P. de Lucas, R. Alvarez-Sala, J. L. Alvarez-Sala, M. Alvarez-Mon.
2006
. Treatment with AM3 restores defective T-cell function in COPD patients.
Chest
129
:
527
-535.
17
Hankinson, J. L., J. R. Odencrantz, K. B. Fedan.
1999
. Spirometric reference values from a sample of the general U.S. population.
Am. J. Respir. Crit. Care Med.
159
:
179
-187.
18
Celli, B. R., W. MacNee.
2004
. Standards for the diagnosis and treatment of patients with COPD: a summary of the ATS/ERS position paper.
Eur. Respir. J.
23
:
932
-946.
19
Sopori, M..
2002
. Effects of cigarette smoke on the immune system.
Nat. Rev. Immunol.
2
:
372
-376.
20
Lange, P., J. Parner, J. Vestbo, P. Schnor, G. Jensen.
1998
. A 15-year follow-up study of ventilatory function in adults with asthma.
N. Engl. J. Med.
339
:
1194
-1200.
21
Retamales, I., W. M. Elliott, B. Meshi, H. O. Coxson, P. D. Pare, F. C. Sciurba, R. M. Rogers, S. Hayashi, J. C. Hogg.
2001
. Amplification of inflammation in emphysema and its association with latent adenoviral infection.
Am. J. Respir. Crit. Care Med.
164
:
469
-473.
22
Gan, W. Q., S. F. Man, A. Senthilselvan, D. D. Sin.
2004
. Association between chronic obstructive pulmonary disease and systemic inflammation: a systematic review and a meta-analysis.
Thorax
59
:
574
-580.
23
Van der Strate, B. W. A., D. S. Postma, C.-A. Brandsma, B. N. Melgert, M. A. Lulnge, M. Geerlings, M. N. Hylkema, A. van den Berg, W. Timens, H. A. Kerstjens.
2006
. Cigarette smoke-induced emphysema: a role for the B cell?.
Am. J. Respir. Crit. Care Med.
173
:
751
-758.
24
Sullivan, A. K., P. L. Simonian, M. T. Falta, J. D. Mitchell, G. P. Cosgrove, K. K. Brown, B. L. Kotzin, N. F. Voelkel, A. P. Fontenot.
2005
. Activated oligoclonal CD4+ T cells in the lungs of patients with severe emphysema.
Am. J. Respir. Crit. Care Med.
172
:
590
-596.
25
Korn, S., R. Wiewrodt, Y. C. Walz, K. Becker, E. Mayer, F. Krummenauer, R. Buhl.
2005
. Characterization of the interstitial lung and peripheral blood T cell receptor repertoire in cigarette smokers.
Am. J. Respir. Cell Mol. Biol.
32
:
142
-148.
26
Lee, S.-H., S. Goswami, A. Grudo, L.-Z. Song, V. Bandi, S. Goodnight-White, L. Green, J. Hacken-Bitar, J. Huh, F. Bakaeen, et al
2007
. Antielastin autoimmunity in tobacco smoking-induced emphysema.
Nat. Med.
13
:
567
-569.
27
Feghali-Bostwick, C. A., A. S. Gadgil, L. E. Otterbein, J. M. Pilewski, M. W. Stoner, E. Csizmadia, Y. Zhang, F. C. Sciurba, S. R. Duncan.
2008
. Autoantibodies in patients with chronic obstructive pulmonary disease.
Am. J. Respir. Crit. Care Med.
177
:
156
-163.
28
Adlowitz, D. G., C. Kirkham, S. Sethi, T. F. Murphy.
2006
. Human serum and mucosal antibody responses to outer membrane protein G1b of Moraxella catarrhalis in chronic obstructive pulmonary disease.
FEMS Immunol. Med. Microbiol.
46
:
139
-146.
29
King, P. T., P. E. Hutchinson, P. D. Johnson, P. W. Homes, N. J. Freezer, S. R. Holdsworth.
2003
. Adaptive immunity to nontypeable Haemophilus influenzae.
Am. J. Respir. Crit. Care Med.
167
:
587
-592.
30
Oldstone, M. B..
2005
. Molecular mimicry, microbial infection and autoimmune disease: evolution of the concept.
Curr. Top. Microbiol. Immunol.
296
:
1
-17.
31
Ma, B., M.-J. Kang, C. G. Lee, S. Chapoval, W. Liu, Q. Chen, A. J. Coyle, J. M. Lora, D. Picarella, R. J. Homer, J. A. Elias.
2005
. Role of CCR5 in IFN-γ-induced and cigarette smoke-induced emphysema.
J. Clin. Invest.
115
:
3460
-3472.
32
Jakubzick, C., E. S. Choi, J. G. Carpenter, S. L. Kunkel, H. Evanoff, F. J. Martinez, K. R. Flaherty, G. B. Toews, T. V. Colby, W. D. Travis, et al
2004
. Human pulmonary fibroblasts exhibit altered interleukin-4 and interleukin-13 receptor subunit expression in idiopathic pulmonary fibrosis.
Am. J. Pathol.
164
:
1989
-2001.
33
Moore, K. W., R. de Waal Malefyt, R. L. Coffmen, A. O'Garra.
2001
. Interleukin-10 and the interleukin-10 receptor.
Annu. Rev. Immunol.
19
:
683
-765.
34
Smyth, L. J. C., C. Starkey, J. Vestbo, D. Singh.
2007
. CD4-regulatory cells in COPD patients.
Chest
132
:
156
-163.
35
Lambert, C., J. McCue, M. Portas, Y. Ouyang, J. Li, T. G. Rosano, A. Lazis, B. M. Freed.
2005
. Acrolein in cigarette smoke inhibits T-cell responses.
J. Allergy Clin. Immunol.
116
:
916
-922.
36
Sin, D. D., P. Lacy, E. York, S. F. Man.
2004
. Effects of fluticasone on systemic markers of inflammation in chronic obstructive pulmonary disease.
Am. J. Respir. Crit. Care Med.
170
:
760
-765.
37
Sin, D. D., S. F. Man, D. D. Marciniuk, G. Ford, M. FitzGerald, E. Wong, E. York, R. R. Mainra, W. Ramesh, L. S. Melenka, et al
2008
. The effects of fluticasone with or without salmeterol on systemic biomarkers of inflammation in chronic obstructive pulmonary disease.
Am. J. Respir. Crit. Care Med.
177
:
1207
-1214.
38
Keatings, V. M., A. Jatakanon, Y. M. Worsdell, P. J. Barnes.
1997
. Effects of inhaled and oral glucocortiocosteroids on inflammatory indices in asthma and COPD.
Am. J. Respir. Crit. Care Med.
155
:
542
-548.
39
Stelmach, I., J. Jerzynska, P. Kuna.
2001
. Markers of inflammation in peripheral blood of children with asthma after treatment with inhaled triamcinolone acetonide.
Ann. Allergy Asthma Immunol.
87
:
319
-326.
40
Sciurba, F. C..
2004
. Physiologic similarities and differences between COPD and asthma.
Chest
126
:
117S
-124S.