© 2003 American Thoracic Society DOI: 10.1165/rcmb.4840 Recruitment of Antigen-Specific Th1-Like Responses to the Human Lung following Bronchoscopic Segmental Challenge with Purified Protein Derivative of Mycobacterium tuberculosisDivision of Pulmonary and Critical Care Medicine, Case Western Reserve University School of Medicine, and University Hospitals of Cleveland, Cleveland, Ohio Address correspondence to: Richard F. Silver, M.D., Division of Pulmonary and Critical Care Medicine, Biomedical Research Building, Room 1030, Case Western Reserve University School of Medicine, Cleveland, OH 44106-4984. E-mail: rfs4{at}po.cwru.edu
Although the mechanisms of specific immunity to Mycobacterium tuberculosis in humans are poorly understood, responses of Th1-like CD4+ T cells appear to be essential for protection. We hypothesized that healthy individuals displaying positive skin-test responses to purified protein derivative of M. tuberculosis (PPD) would have the capacity to mobilize M. tuberculosis–specific Th1 cells to the lung in response to bronchoscopic challenge with PPD. Local instillation of 0.5 tuberculin units of PPD was followed 48 h subsequently by bronchoalveolar lavage (BAL) of PPD-challenged and control segments. In PPD-positive subjects, PPD challenge resulted in a 2.7-fold increase in total BAL cells and in an increase in the percentage of lymphocytes in BAL from 10 to 19%. The BAL lymphocytosis observed in PPD-challenged segments was characterized by an increased percentage of CD4+ T cells and by increased numbers of cells capable of antigen-specific interferon-  production. In contrast, PPD-negative subjects did not develop local inflammation following PPD challenge. These findings indicate that bronchoscopic challenge with PPD results in recruitment of antigen-specific recall responses to the lung. This novel approach may be useful in clarifying the basis of local immunity against M. tuberculosis, and could serve more generally as a model of the development of Th1-like responses in the human lung.
Abbreviations: alveolar macrophages, AM • bronchoalveolar lavage, BAL • bacillus of Calmette and Guerin, BCG • interferon-
Tuberculosis remains a worldwide threat to public health, as it is estimated that one third of the world's population is currently infected with the intracellular pathogen Mycobacterium tuberculosis (1). In the great majority of infected individuals, aerosol infection with M. tuberculosis is followed by the development of antigen-specific cell-mediated immunity that serves to contain the organism. Although the mechanisms that underlie cell-mediated immunity to M. tuberculosis in humans are poorly understood, clinical observations suggest that the responses of Th-1 like CD4+ T cells capable of producing interferon- (IFN-) are critical for protection against active disease. Cell-mediated immunity to M. tuberculosis is manifested clinically by the development of a positive skin-test response to purified protein derivative of M. tuberculosis (PPD), and is associated with relative protection from new infection with the organism (2). Because bronchoalveolar lavage (BAL) cells of healthy PPD-positive subjects do not display increased numbers of lymphocytes (3), one aspect of protective immunity to M. tuberculosis would presumably be the ability to mobilize antigen-specific Th1-like CD4+ T cells to the lung in response to inhalation of the organism. We therefore hypothesized that pulmonary challenge of healthy M. tuberculosis–infected individuals with appropriate mycobacterial antigens could induce the development of a localized Th1 response in the human lung. The development of Th2-like immune responses in the human lung has been studied extensively using the procedure of bronchoscopic segmental antigen challenge (4, 5). This procedure involves local instillation of allergens via a fiberoptic bronchoscope into subsegmental bronchi of subjects with atopic asthma to induce responses of CD4+ T cells that produce interleukin (IL)-4 and IL-5 and stimulate local IgE- and eosinophil-mediated effects. In these procedures, segmental allergen challenge induces well-localized immune responses that are more intense than those induced by whole lung antigen challenge, but do not result in greater physiologic disturbance than generalized bronchoprovocation (6). In the current study, we report the adaptation of the technique of segmental antigen challenge to the evaluation of local immunity to M. tuberculosis. Our studies provide the first demonstration that bronchoscopic instillation of PPD is a well-tolerated and effective means of inducing local M. tuberculosis–specific Th1-like responses in the lungs of healthy PPD-positive individuals. The specificity of this finding was confirmed by the observation that bronchoscopic challenge did not induce local inflammatory responses in PPD-negative subjects. Segmental challenge with PPD therefore provides a means to assess the role of local recall responses in protective immunity against M. tuberculosis. In addition, this method may serve more generally as a model for the development and regulation of Th1 responses in the human lung.
Human Subjects Skin-testing subjects were healthy volunteers aged 23–55 yr who were self-reported as having a previous positive PPD skin test and whose history was suggestive of previous infection with M. tuberculosis. None of the subjects had a history of active tuberculosis. Subjects for bronchoscopy with segmental antigen challenge were healthy non-smoking volunteers aged 26–50. Of the PPD-positive subjects, none had a history of active tuberculosis or of any symptoms suggestive of active tuberculosis, such as cough, night sweats or weight loss. One of the subjects received bacillus of Clamette and Guerin (BCG) vaccination in childhood, but subsequently was repeatedly found to be PPD-negative until re-converting during the course of his medical training. None of the other subjects had received BCG. For both PPD-positive and PPD-negative subjects, exclusion criteria included history of asthma or other chronic lung disease, history of cardiac disease, and history of adverse reactions to topical anesthetic agents. All protocols involving human subjects were approved by the Institutional Board of Review of Case Western Reserve University and University Hospitals of Cleveland.
PPD PPD for in vitro use was obtained from the Staten Serum Institut (Copenhagen, Denmark).
PPD Skin Testing For subjects who reported themselves as having no history of skin-test reactivity to PPD, confirmatory skin-testing was performed at least 1 mo before bronchoscopic challenge using the standard dose of 5 TU only. At examination 48 h subsequently, none of the subjects exhibited any induration in response to PPD.
Bronchoscopy Protocol Before each procedure, subjects received aerosolized lidocaine and gargled with 2% lidocaine solution for 60 s. Topical anesthesia of the nasal passage was performed using viscous lidocaine. Further anesthesia was provided by topical application of 2% lidocaine to the airways via the bronchoscope. In the initial bronchoscopy procedure, BAL of a right middle lobe (RML) subsegment was performed with up to four 60-cc aliquots of pre-warmed normal saline. A control instillation of 10 cc of normal saline (NS) was placed into a different subsegment of the RML. The challenge dose of PPD was then instilled, in a volume of 10 cc normal saline, into a subsegment of the lingula. All subjects were observed in the GCRC for at least 30 min following the procedure. Subjects were provided the hospital pager number of one of the investigators, and were advised to call if any symptoms of concern arose. Repeat bronchoscopy was performed 48 h after the challenge procedure. Before the procedure, subjects were questioned regarding interval symptoms of cough, sputum production dyspnea, or chest pain, and examination of the heart and lungs was repeated. BAL of the saline control subsegment of the RML was performed first using four aliquots of 60 cc NS. BAL of the PPD-challenged subsegment of the lingula was then performed, also using four 60-cc aliquots of NS. Subjects were again observed for 30 min and instructed to contact the investigators if any symptoms developed.
Processing of BAL Fluid
Cytospin preparations were made using
Assessment of Lymphocyte Subjects
Determination of Antigen Specificity using ELISPOT
Determination of Minimally Reactive Dose of PPD To minimize risks to subjects, we first sought to determine the lowest dose of PPD that could elicit any detectable amount of skin-test reactivity. Skin testing was performed on 20 PPD-positive subjects with a history of aerosol exposure to M. tuberculosis using the standard 5 TU dose, as well as with 10-fold serial dilutions of PPD. The results are represented as overlapping histograms in Figure 1. As illustrated, 40% of the subjects displayed reactions of 10–14 mm induration in response to the standard dose, whereas the remainder had responses of 15 mm or greater. Ninety percent of subjects were found to display some degree of skin-test response to 0.5 TU. More than 50% of the subjects displayed minimal responses of 1–4 mm induration in response to the 0.05 TU dose. The 0.005 TU dose was administered only to the first 10 subjects tested, none of whom displayed any induration in response to this dose.
Local Inflammation in Response to Bronchoscopic Challenge with PPD in Skin-Test–Positive Subjects All PPD-positive bronchoscopy subjects were chosen from the initial group of 20 skin-testing subjects. For initial dose escalation studies, we specifically enrolled subjects with large skin test responses (> 15 mm of induration in response to 5 TU) to avoid underestimating the inflammatory response to bronchoscopic challenge with PPD. Based on our skin test findings, we performed initial PPD challenge procedures using a dose of 0.01 TU, or 1/500th of the standard skin-test dose of PPD. Subsequent dose escalation did not result in development of alveolar inflammation until a challenge dose of 0.5 TU (1/10th of the standard skin-test dose) was reached. Immediately following challenge procedures, lung examination of all subjects displayed mild scattered rhonchi without wheezing. In each subject, rhonchi resolved within 30 min, by which time the subjects reported feeling at baseline. At the time of presentation for the second procedure 48 h later, subjects were questioned regarding symptoms including sputum production, chest pain, fevers, chills, and myalgias. No subjects reported any symptom other than mild nonproductive cough that in each case had resolved within several hours after the initial procedure. All subjects had normal lung examinations before the second procedure. One subject developed productive cough shortly after the second bronchoscopy. The subject, who was a physician, took a single dose of antibiotics before sputum could be collected for laboratory examination and subsequently had no further symptoms. Because it was not clear if the inflammatory response observed in this subject was due to challenge with PPD or to a procedure-related infection, BAL findings of this subject were excluded from the analysis. Figure 2 shows the total number and composition of BAL cells obtained from control and challenged segments of the five other PPD-positive subjects 48 h after instillation of 0.5 TU of PPD. This subject group consisted of four men and one woman. Three of the subjects were white, one was Indian, and another was of Middle Eastern origin. Mean BAL return volume was 167.8 ± 30.8 ml (70%) from control segments and 160.9 ± 39.3 (67%) from PPD-challenged segments (P = 0.614). As shown, instillation of PPD resulted in a statistically significant 2.70-fold increase in the total number of cells present in BAL compared with BAL obtained from control segments during the same procedure (P = 0.016 by paired t test). Because lymphocytes accounted for 19 ± 7% of BAL cells obtained from challenged segments as opposed to 10 ± 4% of cells from control segments (P = 0.011), PPD challenge resulted in an overall 5.24-fold increase in total BAL lymphocytes (P = 0.016). A corresponding decrease was observed in the percentage of alveolar macrophages (AM) in BAL from 88 ± 2% in control segments to 77 ± 7% in PPD-challenged segments (P = 0.012). Due to the overall increase in BAL cellularity, however, the total number of AM increased 2.35-fold in response to challenge (P = 0.018). PPD challenge did not result in changes in the proportions of BAL neutrophils or eosinophils compared with those observed in control segments.
The lack of generalized pulmonary inflammation following challenge with PPD was confirmed by comparison of these results with pre-challenge BAL findings, as illustrated in Figure 3. Because a lower volume of saline was used in pre-challenge BAL than in post-challenge procedures in some of the subjects, these results are expressed as cells per cc of recovered BAL fluid. As indicated, total BAL cells, AM, and lymphocytes of control segments at 48 h showed no significant differences from those of pre-challenge BAL (P = 0.267, 0.278, and 0.938, respectively). Responses to PPD instillation therefore were effectively localized to the challenged segments.
Evaluation of Lymphocyte Subsets Recruited to the Lung in Response to PPD Challenge The make-up of various lymphocyte populations found in BAL of control and PPD-challenged segments is illustrated in Figure 4. A statistically significant increase in the percentage of CD4+ T cells was observed in the PPD-challenged segments as compared with control segments (69.2 ± 9.7% versus 62.8 ± 12.5%, P = 0.043). Although the percentage of CD8+ T cells observed PPD-challenged segments was slightly lower than that of control segments (18.5 ± 3.9% versus 21.8 ± 7.5%), this finding was not statistically significant for this number of subjects (P = 0.062). No differences were observed in the percentages of  T cells or NK cells in BAL lymphocytes obtained from challenged and control segments.
Because PPD-challenged segments yielded both a greater number of total cells and a higher percentage of lymphocytes than did control segments, the total numbers of CD4, CD8, , and NK cells all were increased in PPD-challenged segments as compared with control segments, as illustrated. This increase was most marked with regard to the total number of CD4+ T cells in BAL, which increased 6-fold from 7.05 x 105 (± 3.61 x 105) in control segments to 4.23 x 106 (± 2.43 x 106) in PPD-challenged segments (P = 0.021), as shown.
IFN-
PPD Challenge of Skin Test–Negative Subjects Bronchoscopic challenge with 0.5 TU of PPD was performed in four PPD-negative subjects. This group consisted of one man and three women. Two were white and the other two were of Asian descent. Following the initial bronchoscopy procedure, the subjects all displayed mild rhonchi that rapidly resolved. Again, subjects all reported feeling at baseline 30 min after the procedure. Other than transient and nonproductive post-procedure cough, no symptoms were reported in the subsequent 48 h. The total number of BAL cells and cell differential observed in these subjects 48 h after PPD challenge is displayed in Figure 6. The volume of BAL fluid recovered from RML (control) segments was 192.5 ± 21.6 ml, and from lingular (PPD-challenged) segments, 198.5 ± 16.4 ml (P = 0.382 by paired t test). As illustrated, in contrast to the findings of PPD-positive subjects, there was no significant difference in the total number of BAL cells recovered from control and challenged segments (P = 0.359). Likewise, no differences were observed in the percentage of BAL lymphocytes or AM obtained from the two segments (P = 0.443 and P = 0.446, respectively).
In this study, we report the first use of PPD in bronchoscopic segmental antigen challenge procedures in human subjects. Using PPD challenge, we demonstrated that subjects with a history of natural infection with M. tuberculosis have the capacity to develop localized lymphocytic inflammation in the lung following exposure to protein antigens of the organism. PPD challenge resulted in the accumulation of cells capable of producing IFN- in response to in vitro restimulation with PPD. As the lymphocyte population that accumulated in response to PPD was composed predominantly of CD4+ T cells, our findings most likely indicate that segmental antigen challenge with PPD resulted in the recruitment of M. tuberculosis–specific Th1 cells to the lung.
Despite strong clinical evidence for the development of specific acquired immunity to M. tuberculosis in humans, the mechanisms that underlie protective immunity remain unclear. The importance of Th1 responses in defense against M. tuberculosis is emphasized by the remarkable susceptibility of HIV-infected individuals to the development of active tuberculosis (8), however. The susceptibility of individuals with deficiencies in functional receptors for IL-12 and IFN- M. tuberculosis infection is acquired via the inhalation of infectious droplet particles, and pulmonary disease is the most common manifestation of active tuberculosis. Assessment of local immunity within the lung is therefore an important component of understanding protective responses to infection with M. tuberculosis. Previous studies have examined BAL findings in active tuberculosis (3, 11, 12). It is not clear that this work can shed light on the nature of protective immunity as it exists in infected individuals in whom the containment of the organism has occurred in association with the development of specific cell–mediated immunity, however. BAL cells obtained from healthy PPD-positive individuals do not contain increased numbers of lymphocytes at baseline (3), suggesting that the capacity to mobilize protective Th1 responses to the lung in response to inhalation of M. tuberculosis may be a critical component in the development of immune responses that protect against new infection with the organism. Rapid recruitment of Th1 responses to the lung has been shown to correlate with successful containment of aerosol challenge with the M. tuberculosis in mice (13). As this method is not amenable to human studies, we instead chose to use antigenic proteins of M. tuberculosis to assess the ability of naturally infected PPD-positive subjects to recruit antigen-specific Th1 responses to the lung. PPD was clearly the mycobacterial antigen of choice for use in challenge studies in that it is the only M. tuberculosis antigen preparation that is commercially available in sterile and well-standardized form. Use of PPD is relevant in assessment of protective immunity to M. tuberculosis in that this preparation is largely composed of secreted protein components of the organism (14). PPD is thus similar in composition to culture filtrates of M. tuberculosis, which have been used successfully as tuberculosis vaccines in animals (15, 16). A study of aerosol delivery of PPD to sensitized guinea pigs suggested that the kinetics of PPD-induced pulmonary inflammation were similar to those of skin test responses, in that the intensity of the observed lymphocytic pneumonitis peaked 48–72 h following challenge and was followed by resolution over the subsequent 2–3 wk (17). Previously reported studies of local instillation of PPD into the lungs of humans involved administration of PPD as an aerosol to skin test–positive patients with active pulmonary tuberculosis and with the clinical diagnosis of chronic bronchitis (18, 19). Although these studies indicated that aerosolized PPD caused no impairment in pulmonary function of patients who had normal spirometry before challenge, they did not make any assessment of local inflammatory responses. Our study is thus the first to use PPD in segmental challenge studies and the first to demonstrate that local challenge with PPD results in the recruitment of antigen-specific recall responses to the lungs of healthy M. tuberculosis–infected humans. Additional studies will be required to determine whether the inflammation we observed at 48 h represents the peak of local responses to PPD within the lung as it does in PPD skin testing. The limitations of current vaccination against tuberculosis with BCG emphasize the potential importance of local immunity to M. tuberculosis in vaccine efficacy. BCG vaccination provides protection against disseminated forms of tuberculosis, but does not clearly serve to prevent pulmonary tuberculosis (20). As lymphocytes demonstrate preferential homing back to the site of their initial activation (21), the lack of effectiveness of BCG in preventing pulmonary tuberculosis may reflect the inability of standard intradermal vaccination to stimulate an immune response that can be optimally localized to the lung in response to inhalation of M. tuberculosis. This possibility is supported by the observation that the route of administration of BCG to humans can affect the expression of organ-specific adhesion molecules on M. tuberculosis–reactive lymphocytes (22). Limited studies in both monkeys (23) and guinea pigs (24) suggest that aerosolized delivery of BCG may be more effective than intradermal vaccination at inhibiting the establishment of pulmonary infection with M. tuberculosis. Segmental antigen challenge with PPD is likely to be useful in clarifying the importance of suboptimal lymphocyte recruitment to the lung as opposed to other factors in the limitations of current vaccination with BCG. This procedure may also serve as a useful assay for the initial evaluation of the efficacy of new candidate vaccines against tuberculosis. Studies involving segmental antigen challenge of subjects with atopic asthma have provided extensive information about the development of Th2-like responses in the lung, preliminary evaluation of the efficacy of pharmacologic interventions in modulating these responses, and initial evidence for genetic differences in regulation of Th2 responses in the lung (25–29). The development and regulation of Th1-like responses in the human lung have not previously been amenable to this type of study. Nevertheless, the mechanisms that underlie Th1 responses in the lung are likely to contribute to pulmonary defenses against a wide range of infectious pathogens (30–32), as well as to pathogenesis of the interstitial lung disease observed in the systemic granulomatous disorder sarcoidosis (33) and the development of various forms of acute lung injury (34, 35). Recent studies suggest also that the progressive respiratory impairment typical of idiopathic pulmonary fibrosis may result from an imbalance between Th1-like inflammatory stimuli and Th2-mediated mechanisms of lung repair (36). Thus, in addition to its applicability to the study of protective immunity to M. tuberculosis, PPD challenge may be more generally relevant to the understanding of the many Th1-dependent processes that affect the human lung.
The authors thank Dr. Mark Liu of Johns Hopkins University for many helpful discussions regarding establishment of the antigen-challenge protocol. They also greatly appreciated the advice and encouragement of Dr. David Jacoby of Johns Hopkins University and Drs. W. Henry Boom, Jeffrey Kern, and E. Regis McFadden, Jr. of Case Western Reserve University. In addition, they thank the staff of the Case Western Reserve University General Clinical Research Center for their assistance. They are especially grateful to the research volunteers who participated in this study. This work was supported by NIH R01 HL59858. Additional funding was provided by a Developmental Award from the NIH-funded Case Western Reserve University Center for AIDS Research (CFAR), NIH AI-36219. R.F.S. was also supported by an American Lung Association Career Investigator Award (CI-024-N). All bronchoscopy procedures were performed in the NIH-sponsored General Clinical Research Center of Case Western Reserve University (NIH M01 RR00080). Received in original form February 13, 2002 Received in final form November 4, 2002
This article has been cited by other articles:
|
|||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
Top
Abstract
Introduction
25–50,000 cells from each BAL sample. Resuspended BAL cells were placed in a slide centrifugation apparatus (Cytofunnel, #5991040; Shandon, Pittsburgh, PA) and centrifuged at 800 x g for 8 min in a Shandon Cytospin 3 centrifuge. Slides were then stained with a rapid Wright-Giemsa stain method (LeukoStat, #C430; Fisher Diagnostics, Pittsburgh, PA). Cell differentials were determined by counting 300 cells from each sample under light microscopy. 
