help button home button
AJRCCM
HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS
Published ahead of print on September 11, 2003, doi:10.1164/rccm.200306-837OC
This Article
Right arrow Abstract Freely available
Right arrow Full Text (PDF)
Right arrow Online Supplement
Right arrow All Versions of this Article:
200306-837OCv1
168/11/1346    most recent
Right arrow Alert me when this article is cited
Right arrow Alert me if a correction is posted
Services
Right arrow Similar articles in this journal
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Download to citation manager
Right arrow reprints & permissions
Citing Articles
Right arrow Citing Articles via HighWire
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by Lewinsohn, D. A.
Right arrow Articles by Lewinsohn, D. M.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Lewinsohn, D. A.
Right arrow Articles by Lewinsohn, D. M.
American Journal of Respiratory and Critical Care Medicine Vol 168. pp. 1346-1352, (2003)
© 2003 American Thoracic Society

Mycobacterium tuberculosis–specific CD8+ T Cells Preferentially Recognize Heavily Infected Cells

Deborah A. Lewinsohn, Amy S. Heinzel, James M. Gardner, Liqing Zhu, Mark R. Alderson and David M. Lewinsohn

Division of Infectious Diseases, Department of Pediatrics, Department of Molecular Microbiology and Immunology, Vaccine and Gene Therapy Center, and Division of Pulmonary and Critical Care Medicine, Department of Medicine, Oregon Health and Science University; and Portland Veterans Affairs Medical Center, Portland, Oregon; and Corixa Corporation, Seattle, Washington

Correspondence and requests for reprints should be addressed to Deborah Lewinsohn, M.D., Department of Pediatrics, Oregon Health and Science University, 707 S.W. Gaines Road, CDRCP, Portland, OR 97239. E-mail: lewinsde{at}OHSU.edu


   ABSTRACT
TOP
 ABSTRACT
METHODS
 RESULTS
 DISCUSSION
 REFERENCES
 
Both CD4+ and CD8+ T cells are important for successful immunity to tuberculosis and have redundant effector functions, such as cytolysis and release of potent antimycobacterial cytokines such as interferon-{gamma} and tumor necrosis factor-{alpha}. We hypothesized that CD8+ T cells play a unique role in host defense to Mycobacterium tuberculosis infection as well. Possibilities include preferential and/or enhanced release of granular constituents and/or preferential recognition of heavily infected cells. Utilizing human, Mycobacterium tuberculosis–specific, CD4+ and CD8+ T cell clones, we demonstrate that, after recognition of antigen-presenting cells displaying peptide antigen, CD4+ T cells preferentially release interferon-{gamma}, whereas CD8+ T cells preferentially lyse antigen-presenting cells. Furthermore, utilizing dendritic cells infected with Mycobacterium tuberculosis expressing green fluorescent protein, we show that CD8+ T cells preferentially recognize heavily infected cells that constitute the minority of infected cells. These data support the hypothesis that the central role of CD8+ T cells in the control of infection with Mycobacterium tuberculosis may be that of surveillance; in essence, recognition of cells in which the containment of Mycobacterium tuberculosis is no longer effective.

Key Words: antigen presentation • CD4-positive T lymphocytes • CD8-positive T lymphocytes • cytotoxic T lymphocytes

Mycobacterium tuberculosis (Mtb)-specific CD4+ and CD8+ T cells are critical for the effective control of Mtb infection. In the mouse model, passive transfer of CD4+ T cells to sublethally irradiated animals renders them less susceptible to Mtb infection (1). Mice in which the gene(s) for CD4 (CD4-/-) or for MHC Class II molecules are disrupted as well as wild-type mice depleted of CD4+ T cells demonstrate increased susceptibility to Mtb infection (2–5). In humans, human immunodeficiency virus–infected individuals are exquisitely susceptible to developing tuberculosis (TB) after exposure to Mtb, supporting an essential role for CD4+ T cells (6, 7). CD8+ T cells are also important for effective T cell immunity (reviewed in Lazarevic and Flynn [8]). Mice deficient in critical components of the MHC Class I processing and presentation pathway, including ß2-microglobulin (9), TAP1 (10, 11), CD8 (11), and Class Ia (Kb-/-/Db-/-) (12), are more susceptible to infection than wild-type animals. In humans, Mtb-specific CD8+ T cells have been identified in Mtb-infected individuals and include CD8+ T cells that are both classically HLA-Ia restricted (13–19) and nonclassically restricted by the HLA-Ib molecules HLA-E (20, 21) and Group 1 CD1 (22–24). Functionally, a role for human CD8+ T cells in the containment of Mtb infection has been suggested by Stenger and coworkers, who have demonstrated that CD1b-restricted CD8+ T cells are able to inhibit growth of Mtb in vitro (25).

CD4+ and CD8+ T cells share several effector functions that mediate antimycobacterial activities. First, on activation, both cell types secrete interferon-{gamma} (IFN-{gamma}), a potent cytokine that promotes immunologic control of Mtb infection. Mice deficient in IFN-{gamma} are highly susceptible to Mtb infection (26, 27). Similarly, individuals in whom the gene for the IFN-{gamma} receptor is mutated are prone to infection with atypical mycobacteria (28). IFN-{gamma} activates infected macrophages to kill intracellular bacteria and possesses additional immunomodulatory effects relevant to Mtb infection (29). Second, in humans, both CD4+ and CD8+ T cells release the contents of cytotoxic granules on activation (30–32). These granules contain perforin, granzyme B, and granulysin, all of which may play a role in the control of intracellular infection. Perforin establishes pores in the cytoplasmic membrane, allowing entry of granzyme B, which mediates apoptosis of infected cells. Although mice deficient in either perforin or granzyme B do not exhibit a dramatically increased susceptibility to Mtb infection (33, 34), it has been proposed that an apoptotic environment may be deleterious to mycobacteria (35–38). In addition, lysis of the cell membrane may release mycobacteria into the extracellular environment, which disfavors bacterial growth. Finally, granulysin is a bacteriostatic molecule that directly inhibits Mtb growth (39, 40).

We hypothesized that although redundancy of effector mechanisms exists between CD4+ and CD8+ T cells, these cell types have different thresholds for utilization of these pathways in the context of intracellular infection. In support of this hypothesis, we have noted that CD8+ T cells, as compared with CD4+ T cells, recognize a minority of Mtb-infected dendritic cells (DCs) under conditions in which nearly all the DCs had one or more intracellular bacteria. Thus, we further hypothesized that whereas CD4+ T cells recognize all infected cells equally well, CD8+ T cells may preferentially recognize only heavily infected cells.

Using Mtb-infected, peripheral blood-derived DCs, we have characterized both HLA-Ia (HLA-B14, HLA-B44)-restricted and HLA-Ib (HLA-E)-restricted Mtb-specific CD8+ T cells isolated from individuals with latent tuberculosis infection (13, 14). In this report, these T cell clones, as well as an Mtb-specific CD4+ T cell clone, are employed to elucidate the requirements for CD4+ and CD8+ T cell activation in response to antigen-presenting cells (APCs) expressing Mtb antigen, including Mtb-infected DCs. Some of the results of these studies have been previously reported in the form of abstracts (41, 42).


   METHODS
TOP
 ABSTRACT
 METHODS
RESULTS
 DISCUSSION
 REFERENCES
 
Human Subjects
Subjects were recruited from among employees at Harborview Medical Center (Seattle, WA), Corixa Corporation (Seattle, WA), and Oregon Health and Science University (Portland, OR) as previously described (13). The study was approved by institutional review boards at all sites.

Bacterial Strains
Mtb strain H37Rv was obtained from the American Type Culture Collection (ATCC, Manassas, VA) and prepared as described previously (13). H37Rv expressing enhanced green fluorescent protein (Mtb-EGFP) was kindly provided by J. Ernst (New York University, New York, NY).

Generation of Peripheral Blood DCs
Peripheral blood mononuclear cells were isolated from heparinized blood by centrifugation over Ficoll-Hypaque (Sigma, St. Louis, MO) or were obtained via leukapheresis and used to prepare monocyte-derived DCs according to the method of Romani and coworkers (43) and modified by Lewinsohn and coworkers (20).

Mtb Infection of Peripheral Blood-derived DCs
To generate Mtb-infected DCs, cells ( 1 x 106) were cultured in RPMI supplemented with granulocyte-macrophage colony-stimulating factor and interleukin-4 overnight in the presence of Mtb (multiplicity of infection, 50:1; or at various degrees of infection indicated) in low-adherence 16-mm wells (Costar 3473; Corning Life Sciences, Acton, MA). After 18 hours, the cells were harvested and resuspended in RPMI supplemented with 10% human serum.

Preparation of Peptide-loaded APCs
Epstein–Barr virus–transformed B lymphoblastoid cell lines (LCLs) were generated in our laboratory, using supernatants from the cell line 9B5-8 (ATCC). Synthetic peptides were kindly provided by Corixa (Seattle, WA). LCLs were cultured in RPMI supplemented with HEPES (25 mM), gentamicin (5 µg/ml), and 11% fetal calf serum. For preparation of APCs, LCLs or peripheral blood-derived DCs were incubated with peptide overnight at 37°C.

T Cell Clones
T cell Clones 23 and 29 are Class Ib, HLA-E-restricted CD8+ Mtb-specific T cell clones and have been described elsewhere (20, 21). Clone 1–1B is a Class Ia, HLA-B44-restricted CD8+ T cell clone specific for Mtb antigen CFP102–11, and recognizes the minimal epitope AEMKTDAATL (14). Clone 38.1-1 is a CD4+ T cell clone specific for CFP1073–87 and recognizes the epitope STNIRQAGVQYSRAD. To expand the CD8+ T cell clones, a rapid expansion protocol utilizing anti-CD3 monoclonal antibody stimulation was used (44).

IFN-{gamma} and Granzyme B Enzyme-linked Immunospot Assays
The IFN-{gamma} enzyme-linked immunospot (ELISpot) assay has been previously described (13, 45). The granzyme B ELISpot assay was performed with a BD ELISpot human granzyme B kit (BD Biosciences Pharmingen, San Diego, CA), according to the manufacturer's instructions with minor modifications (see the online supplement).

Cytotoxicity Assay
Target cell membrane damage was assessed in a standard 4-hour chromium release assay and percent specific lysis and lytic units were calculated according to standard methods (31).

Confocal Microscopy and Fluorescence-activated Cell Sorting
DCs infected with Mtb-EGFP were harvested and fixed with 0.5% paraformaldehyde. Confocal microscopy was performed with a Leica TCS confocal microscope (Leica Microsystems, Heidelberg, Germany). Flow cytometry was performed with a FACSCalibur (BD Biosciences Immunocytometry Systems, San Jose, CA). Data were collected from 104 viable cells. Fluorescence-activated cell sorting was performed with a FACSVantage (BD Biosciences Immunocytometry Systems) on viable cells as determined by forward and side scatter parameters; infected DCs were sorted on the basis of the intensity of green fluorescence.


   RESULTS
TOP
 ABSTRACT
 METHODS
 RESULTS
DISCUSSION
 REFERENCES
 
CD8+ T Cell Clones Preferentially Utilize the Granule Exocytosis Pathway
Classically restricted CFP10-specific CD4+ and CD8+ T cell clones (38.1-1 and 1–1B, respectively) were compared for their ability to lyse cognate peptide-pulsed target cells in a chromium release assay. Peptide was used at 10 µg/ml to ensure that antigen was not limiting. Peptide-pulsed target cells were efficiently lysed by both CD4+ and CD8+ T cell clones (Figure 1) . However, at each effector-to-target ratio, lysis was greater for the CD8+ T cell clone. In addition, we compared the ability of CFP10-specific CD4+ and CD8+ T cell clones (38.1-1 and 1–1B, respectively) to release either IFN-{gamma} or granzyme B (to directly evaluate the granule exocytosis pathway) in response to autologous LCLs incubated with peptide at various concentrations, using ELISpot assays (Figure 2) . The CD4+ T cell clone released IFN-{gamma} at lower peptide concentrations compared with granzyme B (Figure 2A), whereas the CD8+ T cell clone preferentially released granzyme B (Figure 2B).



View larger version (18K):
[in this window]
[in a new window]
  		Figure 1. Comparison of the granule exocytosis pathway for CFP10-specific CD4+ and CD8+ T cell clones. CFP10-specific CD4+ and CD8+ T cell clones were tested for lytic activity against autologous cognate peptide-pulsed LCL target cells in a chromium release assay. LCL targets were pulsed with excess peptide (10 µg/ml) so that antigen would not be limiting. Specific lysis is plotted against various effector-to-target (E:T) ratios. Each point represents the mean of triplicate determinations. Results are representative of two experiments. CD4 peptide, squares; CD8 peptide, circles; CD4 control, triangles; CD8 control, diamonds.

 


View larger version (23K):
[in this window]
[in a new window]
  		Figure 2. CFP10-specific CD8+ T cell clones preferentially utilize the granule exocytosis pathway. Using CFP10-specific CD4+ T cell clone 38.1-1 (A) or CFP-10-specific CD8+ T cell clone 1–1B (B) as a source of effector cells, antigen-specific IFN-{gamma} release, as determined by ELISpot, was compared with antigen-specific granzyme B release, as determined by ELISpot, over the range of peptide concentrations indicated. Data are expressed as a percentage of maximum number of spots demonstrated for highest peptide concentrations. Each data point reflects the mean and standard error of triplicate determinations. These data are representative of three independent experiments.

 
Mtb-specific CD4+ T Cells Preferentially Release IFN-{gamma}
We next compared T cell Clone 38.1-1 with Clone 1–1B in an IFN-{gamma} ELISpot assay under conditions in which neither antigen nor APCs were limiting (Figure 3) . T cell clones (500 cells) were cultured with autologous LCL (2 x 104 cells) previously incubated with cognate peptide (10 µg/ml). Nearly all the CD4+ T cells were activated, compared with approximately half the CD8+ T cells, and the amount of IFN-{gamma} released by CD8+ T cells, as reflected by spot area, was less than half that released by CD4+ T cells. Because the area per spot was twofold greater for the CD4+ T cell clone than for the CD8+ T cell clone, the decreased amount of IFN-{gamma} released reflected not only the activation of fewer CD8+ T cells, but also less IFN-{gamma} production on a per-cell basis.



View larger version (18K):
[in this window]
[in a new window]
  		Figure 3. CFP10-specific CD4+ T cell clones preferentially release IFN-{gamma}. CFP10-specific CD4+ and CD8+ T cell clones were tested for antigen-induced IFN-{gamma} release in an ELISpot assay. Effector T cells (500 cells per well) were cultured with autologous LCL target cells (20,000 cells per well) pulsed with cognate peptide (10 µg/ml) in an ELISpot assay. Data are displayed as number of spots per well, spot area per well, and area per spot. Each column reflects the mean of duplicate determinations. These data are representative of two independent experiments. CD4, black columns; CD8, gray columns.

 
To address the concern that these data reflect the properties of particular clones rather than the generalized phenotype of CD4+ and CD8+ T cells, we studied CFP10-specific T cells in peripheral blood. Peripheral blood mononuclear cells were obtained from adults with latent tuberculosis infection and CD4+ and CD8+ T cells were isolated by magnetic bead separation. Autologous peripheral blood-derived DCs were used as a source of APCs. Using the IFN-{gamma} ELISpot assay, T cells were assayed for their ability to recognize autologous DCs pulsed with pooled peptides (15-mers, overlapping by 11 amino acids, 10 µg/peptide per milliliter). For one individual, the effector cell frequency for both CD4+ and CD8+ effectors was comparable, 1:720 and 1:750, respectively, consistent with a primed T cell response (Figures 4A and 4C) . Despite the similarity in number of spots elicited, the area per spot was markedly reduced in CD8+ T cell wells (Figures 4B and 4C), suggesting that CD8+ T cells from peripheral blood release about 50% less IFN-{gamma} on a per-cell basis than do CD4+ T cells. Results were similar for two additional adults with latent tuberculosis infection (Figure 4D).



View larger version (27K):
[in this window]
[in a new window]
  		Figure 4. Comparison of IFN-{gamma} production from CFP10-specific CD4+ T cells and CD8+ T cells derived from peripheral blood. (A–C) CD4+ and CD8+ T cells derived from peripheral blood mononuclear cells were tested for antigen-induced IFN-{gamma} release in an ELISpot assay. (A) The number of spots was plotted versus the number of input cells and the slope of the line was determined by linear regression analysis. The effector cell frequency is represented by the slope of the line. Each data point represents the mean of duplicate determinations; CFP10, squares; control, triangle. (B) Data are expressed as the spot area per spot (µm2). (C) Photomicrograph of example duplicate ELISpot wells containing antigen and CD4+ and CD8+ effectors. (D) Results from two additional individuals are shown. Data are expressed as the spot area per spot (µm2). CD8, solid columns; CD4, open columns.

 
CD8+ Cells Preferentially Recognize Heavily Infected APCs
We next explored the hypothesis that CD8+ T cell effectors preferentially recognize cells heavily infected with Mtb. During the course of our original work identifying Mtb-specific CD8+ T cells in humans, we had noted that Mtb-infected DCs functioned as relatively inefficient APCs for CD8+ T cells, as compared with CD4+ T cells. This observation is demonstrated in the experiment shown in Figure 5 . Autologous DCs were infected with Mtb (H37Rv) at a multiplicity of infection of 10 for 18 hours and then tested for their ability to present antigen to Mtb-specific T cell clones, under conditions in which T cells were not limiting (1,000 cells per well). In this way, the ELISpot assay can be used to assess the functional number of APCs. CD4+ T cell Clone 38.1-1 recognized Mtb-infected DCs at all DC concentrations. Furthermore, recognition of Mtb-infected DCs by 38.1-1 was minimally enhanced by addition of cognate peptide. In contrast, recognition of Mtb-infected DCs by CD8+ T cell clones (HLA-E-restricted Clones 23 and 29, and HLA-B44-restricted Clone 1–1B) increased with increasing concentrations of DCs. In addition, the limitation in antigen presentation could be overcome by addition of cognate peptide for Clone 1–1B. These data suggested that only a minority of Mtb-infected DCs are capable of presenting antigen to CD8+ T cells. From these data, we hypothesized that the degree of recognition was related to the bacterial burden within the DCs. To directly test this hypothesis, we used DCs infected with Mtb-EGFP. Intracellular localization of the EGFP-Mtb was confirmed by confocal microscopy (Figure 6) . Mtb-infected DCs were sorted on the basis of the degree of green fluorescence, and each sorted DC population was assessed for the number of Mtb per cell, using fluorescence microscopy. As shown in Figure 6, DCs infected with EGFP-Mtb represented a heterogeneous population and could be sorted on the basis of the degree of intracellular infection. These sorted populations were then tested for their ability to present antigen to CD8+ Mtb-specific T cell clones (HLA-E-restricted Clones 23 and 29 and HLA-B44-restricted Clone 1–1B, 1,000 cells per well) in the IFN-{gamma} ELISpot assay (Figure 7) . For all T cell clones, CD8+ T cells recognized heavily infected DCs better than less infected DCs. By comparison, CD4+ T cells recognized all subsets of infected DCs better than did CD8+ T cells, and the greatest difference between CD4+ and CD8+ T cells was seen against the lightly infected DCs (DClow). Finally, more heavily infected DCs were less well recognized by CD4+ T cells than the lightly infected DCs, which may reflect decreased viability of heavily infected DCs.



View larger version (17K):
[in this window]
[in a new window]
  		Figure 5. Mtb-specific CD8+ T cells recognize a minority of Mtb-infected DCs. DCs were infected with Mtb at a multiplicity of infection of 10. CD4+ (38.1-1) and CD8+ (23, 29, and 1–1B) T cell clones (1,000 cells per well) were cultured with various numbers of Mtb-infected DCs in an IFN-{gamma} ELISpot assay. Where indicated, Mtb-infected DCs have been pulsed with cognate peptide (10 µg/ml). CD8 clone 23, solid squares; CD8 clone 29, solid circles; CD8 clone 1–1B peptide 1–15, solid triangles; CD8 clone 1–1B, solid diamonds; CD4 clone 38.1-1, open squares; CD4 clone 38.1-1 peptide 73–87, open circles.

 


View larger version (56K):
[in this window]
[in a new window]
  		Figure 6. DCs infected with EGFP-Mtb can be sorted on the basis of the degree of intracellular infection. Photomicrograph shows DCs infected with EGFP-Mtb as seen by fluorescence microscopy. By gating on viable cells, as determined by the forward and side scatter parameters shown, infected DCs were sorted by fluorescence-activated cell sorting on the basis of the intensity of green fluorescence. Two hundred cells from each population were then assessed for the number of DC-associated Mtb, using fluorescence microscopy. The mean number of Mtb per cell is presented.

 


View larger version (40K):
[in this window]
[in a new window]
  		Figure 7. Mtb-specific CD8+ T cells recognize heavily infected DCs. (A) EGFP-Mtb-infected DCs were sorted for green fluorescence and used as APCs (20,000 cells per well) for CD8+ T cell effectors (Clone 23, open columns; Clone 29, shaded columns; Clone 1–1B, solid columns; 1,000 cell/well) in the IFN-{gamma} ELISpot assay. Each column represents the mean of duplicate determinations. Error bars reflect the standard error. These data are representative of three independent experiments. (B) Photomicrograph shows individual ELISpot wells containing CD8+ T cell Clone 23 (1,000 cells) and either uninfected DCs or sorted EGFP-Mtb-infected DClow, DCmed, or DChi (1,000 cells per well). For comparison, ELISpot wells containing CD4+ T cell Clone 38.1-1 (1,000 cells per well) and either uninfected DCs or sorted EGFP-Mtb-infected DClow (2,500 cells per well), DCmed (2,500 cells per well), or DChi (1,800 cells per well) are shown.

 

   DISCUSSION
TOP
 ABSTRACT
 METHODS
 RESULTS
 DISCUSSION
REFERENCES
 
In this report, we have utilized Mtb-specific human CD4+ and CD8+ T cell clones to directly analyze the requirements for antigen-dependent T cell activation. We conclude that human CD4+ and CD8+ effectors share an ability to release IFN-{gamma} and to utilize the granule exocytosis pathway, consistent with previously published observations (31, 32). However, by carefully controlling the dose of antigen, we found that CD4+ T cell effectors preferentially secrete IFN-{gamma}, whereas CD8+ T cell effectors preferentially utilize the granule exocytosis pathway. Regarding these later results, Valitutti and coworkers (46) demonstrated that human influenza-specific CD8+ T cell clones are cytotoxic at peptide concentrations greater than 1,000-fold lower than those required for IFN-{gamma} production, indicating that our observations are not unique to Mtb-specific CD8+ T cells. Furthermore, Mtb-specific CD4+ T cells release an increased amount of cytokine on a per-cell basis compared with CD8+ T cells. These results differed from those of Shams and coworkers (47), who found, using ELISpot, that Mtb antigen-stimulated CD4+ and CD8+ T cells derived from the peripheral blood of individuals with latent tuberculosis infection produced similar amounts of IFN-{gamma} on a per-cell basis. Although the reason for these different results cannot be easily resolved, the decreased IFN-{gamma} production by CD8+ T cells seen in our study did not reflect a relatively poor APC because in our study we used a highly potent APC for CD8+ T cells, DCs infected with live Mtb, whereas Shams and coworkers used a less potent APC for CD8+ T cells, monocytes exposed to heat-killed Mtb. Thus, in summary, although CD4+ and CD8+ T cells share the effector functions relevant to immunologic control of Mtb infection, these cell types demonstrate preferential use of different effector functions as well.

Furthermore, although Mtb-specific CD4+ T cells recognize cells with a low, moderate, or high degree of infection equally well, both HLA-Ia– and HLA-E–restricted Mtb-specific CD8+ T cells preferentially recognize cells heavily infected with Mtb. Our observations suggest that Mtb antigens access the Class I processing pathway more efficiently in the context of a high degree of intracellular infection. These data are also consistent with the observation that in ß2-microglobulin–deficient, and hence CD8+ T cell–deficient, mice an increased number of heavily infected macrophages are observed relative to wild-type mice (9, 48). Normally, Mtb resides in an early endosomal compartment, an intracellular location considered discrete from the HLA-I cytosolic processing compartment. Although a noncytosolic, vesicular HLA-I processing pathway has been proposed for Mtb (49), both the HLA-E– and HLA-B44–restricted clones used in our study require proteosomal and hence cytosolic antigen processing (Lewinsohn and coworkers [20]; and D. M. Lewinsohn, unpublished data). In this regard, Teitelbaum and coworkers have suggested that Mtb infection induces increased permeability of the endosome, allowing "leakage" of Mtb antigen into the cytosol (50). Therefore a high multiplicity of infection may simply lead to increased entry of Mtb antigens into the cytosol. Alternatively, the Mtb-containing early endosome may undergo mechanical disruption in the face of high bacterial load, allowing unimpeded entry of Mtb antigens into the cytosol.

Taken together, our results suggest that CD4+ T cells and CD8+ T cells have distinctive roles in the control of Mtb infection. The macrophage serves both as the host cell for Mtb and, after activation, functions as the first line of defense, providing intracellular containment of infection. We postulate that so long as IFN-{gamma}– or tumor necrosis factor-{alpha}–driven macrophage activation is able to maintain a low degree of intracellular infection, CD4+ T cell–dependent cytokine release may be sufficient to contain the growth of Mtb. However, in some cases, we postulate that the activated macrophage is no longer able to contain this growth. In this case antigen could efficiently gain access to the cytosol, and then through enhanced HLA-I antigen processing these heavily infected macrophages could become increasingly visible targets for CD8+ CTL effectors. In this circumstance, utilization of the granule exocytosis pathway may become more important, both by creating an unfavorable apoptotic environment for mycobacteria (51) and by releasing the Mtb into an equally inhospitable extracellular environment. Finally, as there is both in vitro (52) and in vivo (53) evidence that extracellular Mtb can occasionally infect epithelial and/or endothelial cells, CD8+ T cells may contribute uniquely in their ability to recognize these Class I-expressing, but Class II-negative, cells. Hence, CD8+ T cells, both by preferential use of the granule exocytosis pathway and by preferential recognition of heavily infected cells, may control infection primarily through the recognition and eradication of heavily infected macrophages and secondarily through the recognition of Class II-negative cells.

How might CD8+ T cells play a role in latent TB infection, and hence in vaccine design? In contrast to latent viral infections in which viral protein, and hence antigen, may be unapparent for long periods of time, latent TB infection likely reflects chronic persistence of slowly replicating Mtb with persistent antigenic stimulation. As a result, latent TB infection is likely a reflection of active immune surveillance resulting in a state of immunologic equipoise. During the course of latent infection, we postulate that macrophages activated by potent CD4+ cell–derived helper T type 1 cytokines may occasionally be insufficient to contain the intracellular growth of Mtb. By recognizing these cells, the CD8+ T cell may limit the growth of Mtb such that active tuberculosis does not occur. With regard to vaccine design, CD8+ T cells may not be necessary in a vaccine that effectively prevents Mtb infection. However, at present, all current Mtb vaccines work not by the prevention of Mtb infection, but by augmenting protective cellular immunity. Hence, the immune surveillance provided by Mtb-specific CD8+ cells might be an essential component of any TB vaccine. In support of this model are data from Andersen and coworkers (54) and from Orme and coworkers (55), suggesting that CD8+ T cells may be important during the chronic rather than acute phase of Mtb infection.

Conversely, when immune surveillance fails, active tuberculosis ensues. In this regard, it would be anticipated that the number of heavily infected macrophages would increase. Driven by increased antigen load, we speculate that the frequency of Mtb-specific CD8+ T cells would increase proportionately. Furthermore, eradication of Mtb by effective treatment for latent TB infection or active disease may ultimately result in a decreased frequency of Mtb-specific CD8+ T cells.

In summary, our data suggest that CD8+ T cells, through preferential use of the granule exocytosis pathway, and through preferential recognition of heavily infected cells, may play an essential immune surveillance role in the human host response to infection with Mtb. As such, an effective vaccine for TB may require this critical immune surveillance capacity.


   Acknowledgments
 
The authors thank Joel Ernst for provision of EGFP-expressing Mtb and Sean Steen for tenacious peptide chemistry. The authors are indebted to Immunex for the provision of cytokine reagents.


   FOOTNOTES
 
Supported by NIH 1KO8AI01645 (D.A.L.), NIH 1K08AI01644 and R01AI48090 (D.M.L.), an American Lung Association Research Grant and Career Investigator Award (D.M.L.), and a Medical Research Foundation Research Grant (D.M.L.). Dr. D. A. Lewinsohn was supported in part as a Junior Investigator of the Oregon Child Health Research Center, NIH NICHHD HD33703. The Portland VA Medical Center has provided laboratory space and partial salary support (D.M.L.).

This article has an online supplement, which is accessible from this issue's table of contents online at www.atsjournals.org

Conflict of Interest Statement: D.A.L. has no declared conflict of interest; A.S.H. has no declared conflict of interest; J.M.G. has no declared conflict of interest; L.Z. has no declared conflict of interest; M.R.A. has no declared conflict of interest; D.M.L. has no declared conflict of interest.

Received in original form June 23, 2003; accepted in final form September 9, 2003


   REFERENCES
TOP
 ABSTRACT
 METHODS
 RESULTS
 DISCUSSION
 REFERENCES
 

  1. Orme IM. Characteristics and specificity of acquired immunologic memory to Mycobacterium tuberculosis infection. J Immunol 1988;140:3589–3593.[Abstract]
  2. Muller I, Cobbold SP, Waldmann H, Kaufmann SH. Impaired resistance to Mycobacterium tuberculosis infection after selective in vivo depletion of L3T4+ and Lyt-2+ T cells. Infect Immun 1987;55:2037–2041.[Abstract/Free Full Text]
  3. Flory CM, Hubbard RD, Collins FM. Effects of in vivo T lymphocyte subset depletion on mycobacterial infections in mice. J Leukoc Biol 1992;51:225–229.[Abstract]
  4. Leveton C, Barnass S, Champion B, Lucas S, De Souza B, Nicol M, Banerjee D, Rook G. T-cell-mediated protection of mice against virulent Mycobacterium tuberculosis. Infect Immun 1989;57:390–395.[Abstract/Free Full Text]
  5. Caruso AM, Serbina N, Klein E, Triebold K, Bloom BR, Flynn JL. Mice deficient in CD4 T cells have only transiently diminished levels of IFN-{gamma}, yet succumb to tuberculosis. J Immunol 1999;162:5407–5416.[Abstract/Free Full Text]
  6. Hirsch CS, Toossi Z, Othieno C, Johnson JL, Schwander SK, Robertson S, Wallis RS, Edmonds K, Okwera A, Mugerwa R, et al. Depressed T-cell interferon-{gamma} responses in pulmonary tuberculosis: analysis of underlying mechanisms and modulation with therapy. J Infect Dis 1999;180:2069–2073.[CrossRef][Medline]
  7. Eriki PP, Okwera A, Aisu T, Morrissey AB, Ellner JJ, Daniel TM. The influence of human immunodeficiency virus infection on tuberculosis in Kampala, Uganda. Am Rev Respir Dis 1991;143:185–187.[Medline]
  8. Lazarevic V, Flynn J. CD8+ T cells in tuberculosis. Am J Respir Crit Care Med 2002;166:1116–1121.[Free Full Text]
  9. Flynn JL, Goldstein MM, Triebold KJ, Koller B, Bloom BR. Major histocompatibility complex Class I-restricted T cells are required for resistance to Mycobacterium tuberculosis infection. Proc Natl Acad Sci USA 1992;89:12013–12017.[Abstract/Free Full Text]
  10. Behar SM, Dascher CC, Grusby MJ, Wang CR, Brenner MB. Susceptibility of mice deficient in CD1D or TAP1 to infection with Mycobacterium tuberculosis. J Exp Med 1999;189:1973–1980.[Abstract/Free Full Text]
  11. Sousa AO, Mazzaccaro RJ, Russell RG, Lee FK, Turner OC, Hong S, Van Kaer L, Bloom BR. Relative contributions of distinct MHC Class I-dependent cell populations in protection to tuberculosis infection in mice. Proc Natl Acad Sci USA 2000;97:4204–4208.[Abstract/Free Full Text]
  12. Rolph MS, Raupach B, Kobernick HH, Collins HL, Perarnau B, Lemonnier FA, Kaufmann SH. MHC Class Ia-restricted T cells partially account for ß2-microglobulin-dependent resistance to Mycobacterium tuberculosis. Eur J Immunol 2001;31:1944–1949.[CrossRef][Medline]
  13. Lewinsohn DM, Briden AL, Reed SG, Grabstein KH, Alderson MR. Mycobacterium tuberculosis-reactive CD8+ T lymphocytes: the relative contribution of classical versus nonclassical HLA restriction. J Immunol 2000;165:925–930.[Abstract/Free Full Text]
  14. Lewinsohn DM, Zhu L, Madison VJ, Dillon DC, Fling SP, Reed SG, Grabstein KH, Alderson MR. Classically restricted human CD8+ T lymphocytes derived from Mycobacterium tuberculosis-infected cells: definition of antigenic specificity. J Immunol 2001;166:439–446.[Abstract/Free Full Text]
  15. Lewinsohn DA, Lines RA, Lewinsohn DM. Human dendritic cells presenting adenovirally expressed antigen elicit Mycobacterium tuberculosis-specific CD8+ T cells. Am J Respir Crit Care Med 2002;166:843–848.[Abstract/Free Full Text]
  16. Lalvani A, Brookes R, Wilkinson RJ, Malin AS, Pathan AA, Andersen P, Dockrell H, Pasvol G, Hill AV. Human cytolytic and interferon {gamma}-secreting CD8+ T lymphocytes specific for Mycobacterium tuberculosis. Proc Natl Acad Sci USA 1989;5:270–275.
  17. Mohagheghpour N, Gammon D, Kawamura LM, van Vollenhoven A, Benike CJ, Engleman EG. CTL response to Mycobacterium tuberculosis: identification of an immunogenic epitope in the 19-kDa lipoprotein. J Immunol 1998;161:2400–2406.[Abstract/Free Full Text]
  18. Tan JS, Canaday DH, Boom WH, Balaji KN, Schwander SK, Rich EA. Human alveolar T lymphocyte responses to Mycobacterium tuberculosis antigens: role for CD4+ and CD8+ cytotoxic T cells and relative resistance of alveolar macrophages to lysis. J Immunol 1997;159:290–297.[Abstract]
  19. Turner J, Dockrell HM. Stimulation of human peripheral blood mononuclear cells with live Mycobacterium bovis BCG activates cytolytic CD8+ T cells in vitro. Immunology 1996;87:339–342.[CrossRef][Medline]
  20. Lewinsohn DM, Alderson MR, Briden AL, Riddell SR, Reed SG, Grabstein KH. Characterization of human CD8+ T cells reactive with Mycobacterium tuberculosis-infected antigen-presenting cells. J Exp Med 1998;187:1633–1640.[Abstract/Free Full Text]
  21. Heinzel AS, Grotzke JE, Lines RA, Lewinsohn DA, McNabb AL, Streblow DN, Braud VM, Grieser HJ, Belisle JT, Lewinsohn DM. HLA-E-dependent presentation of Mtb-derived antigen to human CD8+ T cells. J Exp Med 2002;196:1473–1481.[Abstract/Free Full Text]
  22. Beckman EM, Melian A, Behar SM, Sieling PA, Chatterjee D, Furlong ST, Matsumoto R, Rosat JP, Modlin RL, Porcelli SA. CD1c restricts responses of mycobacteria-specific T cells: evidence for antigen presentation by a second member of the human CD1 family. J Immunol 1996;157:2795–2803.[Abstract]
  23. Rosat JP, Grant EP, Beckman EM, Dascher CC, Sieling PA, Frederique D, Modlin RL, Porcelli SA, Furlong ST, Brenner MB. CD1-restricted microbial lipid antigen-specific recognition found in the CD8+ {alpha}ß T cell pool. J Immunol 1999;162:366–371.[Abstract/Free Full Text]
  24. Moody DB, Ulrichs T, Muhlecker W, Young DC, Gurcha SS, Grant E, Rosat JP, Brenner MB, Costello CE, Besra GS, et al. CD1c-mediated T-cell recognition of isoprenoid glycolipids in Mycobacterium tuberculosis infection. Nature 2000;404:884–888.[CrossRef][Medline]
  25. Stenger S, Mazzaccaro RJ, Uyemura K, Cho S, Barnes PF, Rosat JP, Sette A, Brenner MB, Porcelli SA, Bloom BR, et al. Differential effects of cytolytic T cell subsets on intracellular infection. Science 1997;276:1684–1687.[Abstract/Free Full Text]
  26. Flynn JL, Chan J, Triebold KJ, Dalton DK, Stewart TA, Bloom BR. An essential role for interferon {gamma} in resistance to Mycobacterium tuberculosis infection. J Exp Med 1993;178:2249–2254.[Abstract/Free Full Text]
  27. Cooper AM, Dalton DK, Stewart TA, Griffin JP, Russell DG, Orme IM. Disseminated tuberculosis in interferon {gamma} gene-disrupted mice. J Exp Med 1993;178:2243–2247.[Abstract/Free Full Text]
  28. Newport MJ, Huxley CM, Huston S, Hawrylowicz CM, Oostra BA, Williamson R, Levin M. A mutation in the interferon-{gamma}-receptor gene and susceptibility to mycobacterial infection. N Engl J Med 1996;335:1941–1949.[Abstract/Free Full Text]
  29. Pearl JE, Saunders B, Ehlers S, Orme IM, Cooper AM. Inflammation and lymphocyte activation during mycobacterial infection in the interferon-{gamma}-deficient mouse. Cell Immunol 2001;211:43–50.[CrossRef][Medline]
  30. Henkart PA, Williams MS, Zacharchuk CM, Sarin A. Do CTL kill target cells by inducing apoptosis? Semin Immunol 1997;9:135–144.[CrossRef][Medline]
  31. Lewinsohn DM, Bement TT, Xu J, Lynch DH, Grabstein KH, Reed SG, Alderson MR. Human purified protein derivative-specific CD4+ T cells use both CD95-dependent and CD95-independent cytolytic mechanisms. J Immunol 1998;160:2374–2379.[Abstract/Free Full Text]
  32. Canaday DH, Wilkinson RJ, Li Q, Harding CV, Silver RF, Boom WH. CD4+ and CD8+ T cells kill intracellular Mycobacterium tuberculosis by a perforin and Fas/Fas ligand-independent mechanism. J Immunol 2001;167:2734–2742.[Abstract/Free Full Text]
  33. Cooper AM, D'Souza C, Frank AA, Orme IM. The course of Mycobacterium tuberculosis infection in the lungs of mice lacking expression of either perforin- or granzyme-mediated cytolytic mechanisms. Infect Immun 1997;65:1317–1320.[Abstract]
  34. Laochumroonvorapong P, Wang J, Liu CC, Ye W, Moreira AL, Elkon KB, Freedman VH, Kaplan G. Perforin, a cytotoxic molecule which mediates cell necrosis, is not required for the early control of mycobacterial infection in mice. Infect Immun 1997;65:127–132.[Abstract]
  35. Laochumroonvorapong P, Paul S, Manca C, Freedman VH, Kaplan G. Mycobacterial growth and sensitivity to H2O2 killing in human monocytes in vitro. Infect Immun 1997;65:4850–4857.[Abstract]
  36. Fratazzi C, Arbeit RD, Carini C, Remold HG. Programmed cell death of Mycobacterium avium serovar 4-infected human macrophages prevents the mycobacteria from spreading and induces mycobacterial growth inhibition by freshly added, uninfected macrophages. J Immunol 1997;158:4320–4327.[Abstract]
  37. Keane J, Balcewicz-Sablinska MK, Remold HG, Chupp GL, Meek BB, Fenton MJ, Kornfeld H. Infection by Mycobacterium tuberculosis promotes human alveolar macrophage apoptosis. Infect Immun 1997;65:298–304.[Abstract]
  38. Balcewicz-Sablinska MK, Keane J, Kornfeld H, Remold HG. Pathogenic Mycobacterium tuberculosis evades apoptosis of host macrophages by release of TNF-R2, resulting in inactivation of TNF-{alpha}. J Immunol 1998;161:2636–2641.[Abstract/Free Full Text]
  39. Stenger S, Hanson DA, Teitelbaum R, Dewan P, Niazi KR, Froelich CJ, Ganz T, Thoma-Uszynski S, Melian A, Bogdan C, et al. An antimicrobial activity of cytolytic T cells mediated by granulysin. Science 1998;282:121–125.[Abstract/Free Full Text]
  40. Ernst WA, Thoma-Uszynski S, Teitelbaum R, Ko C, Hanson DA, Clayberger C, Krensky AM, Leippe M, Bloom BR, Ganz T, et al. Granulysin, a T cell product, kills bacteria by altering membrane permeability. J Immunol 2000;165:7102–7108.[Abstract/Free Full Text]
  41. Lewinsohn DM, Lines RA, Zhu L, Pefaur NB, Heinzel AS, Grabstein KH, Alderson MR. Mtb-specific CD8+ T lymphocytes preferentially recognize heavily infected cells. Abstract presented at the 2001 Meeting of the American Thoracic Society.
  42. Lewinsohn DA, Lines RA, Zhu L, Pefaur NB, Heinzel AS, Lewinsohn DA, Grabstein KH, Alderson MR. Mtb-specific CD8+ T lymphocytes preferentially recognize heavily infected cells. Abstract presented at the 36th Joint Tuberculosis/Leprosy Conference: United States–Japan Cooperative Medical Science Program, New Orleans, LA. 2001.
  43. Romani N, Gruner S, Brang D, Kampgen E, Lenz A, Trockenbacher B, Konwalinka G, Fritsch PO, Steinman RM, Schuler G. Proliferating dendritic cell progenitors in human blood. J Exp Med 1994;180:83–93.[Abstract/Free Full Text]
  44. Riddell SR, Watanabe KS, Goodrich JM, Li CR, Agha ME, Greenberg PD. Restoration of viral immunity in immunodeficient humans by the adoptive transfer of T cell clones. Science 1992;257:238–241.[Abstract/Free Full Text]
  45. Lalvani A, Brookes R, Hambleton S, Britton WJ, Hill AV, McMichael AJ. Rapid effector function in CD8+ memory T cells. J Exp Med 1997;186:859–865.[Abstract/Free Full Text]
  46. Valitutti S, Muller S, Dessing M, Lanzavecchia A. Different responses are elicited in cytotoxic T lymphocytes by different levels of T cell receptor occupancy. J Exp Med 1996;183:1917–1921.[Abstract/Free Full Text]
  47. Shams H, Wizel B, Weis SE, Samten B, Barnes PF. Contribution of CD8+ T cells to {gamma} interferon production in human tuberculosis. Infect Immun 2001;69:3497–3501.[Abstract/Free Full Text]
  48. Mogues T, Goodrich M, Ryan L, LaCourse R, North R. The relative importance of T cell subsets in immunity and immunopathology of airborne Mycobacterium tuberculosis infection in mice. J Exp Med 2001;193:271–280.[Abstract/Free Full Text]
  49. Canaday DH, Ziebold C, Noss EH, Chervenak KA, Harding CV, Boom WH. Activation of human CD8+ {alpha}ß TCR+ cells by Mycobacterium tuberculosis via an alternate Class I MHC antigen-processing pathway. J Immunol 1999;162:372–379.[Abstract/Free Full Text]
  50. Teitelbaum R, Cammer M, Maitland ML, Freitag NE, Condeelis J, Bloom BR. Mycobacterial infection of macrophages results in membrane-permeable phagosomes. Proc Natl Acad Sci USA 1999;96:15190–15195.[Abstract/Free Full Text]
  51. Fratazzi C, Arbeit RD, Carini C, Balcewicz-Sablinska MK, Keane J, Kornfeld H, Remold HG. Macrophage apoptosis in mycobacterial infections. J Leukoc Biol 1999;66:763–764.[Abstract]
  52. Bermudez LE, Sangari FJ, Kolonoski P, Petrofsky M, Goodman J. The efficiency of the translocation of Mycobacterium tuberculosis across a bilayer of epithelial and endothelial cells as a model of the alveolar wall is a consequence of transport within mononuclear phagocytes and invasion of alveolar epithelial cells. Infect Immun 2002;70:140–146.[Abstract/Free Full Text]
  53. Hernandez-Pando R, Jeyanathan M, Mengistu G, Aguilar D, Orozco H, Harboe M, Rook GA, Bjune G. Persistence of DNA from Mycobacterium tuberculosis in superficially normal lung tissue during latent infection. Lancet 2000;356:2133–2138.[CrossRef][Medline]
  54. Van Pinxteren LAH, Cassidy JP, Smedegaard BHC, Agger EM, Andersen P. Control of latent Mycobacterium tuberculosis infection is dependent on CD8 T cells. Eur J Immunol 2000;30:3689–3698.[CrossRef][Medline]
  55. Gonzalez-Juarrero M, Turner OC, Turner J, Marietta P, Brooks JV, Orme IM. Temporal and spatial arrangement of lymphocytes within lung granulomas induced by aerosol infection with Mycobacterium tuberculosis. Infect Immun 2001;69:1722–1728.[Abstract/Free Full Text]



This article has been cited by other articles:


Home page
 ThoraxHome page
I Petrache, K Diab, K S Knox, H L Twigg III, R S Stephens, S Flores, and R M Tuder
HIV associated pulmonary emphysema: a review of the literature and inquiry into its mechanism
Thorax, May 1, 2008; 63(5): 463 - 469.
[Abstract] [Full Text] [PDF]

Home page
 J. Immunol.Home page
R. Billeskov, C. Vingsbo-Lundberg, P. Andersen, and J. Dietrich
Induction of CD8 T Cells against a Novel Epitope in TB10.4: Correlation with Mycobacterial Virulence and the Presence of a Functional Region of Difference-1
J. Immunol., September 15, 2007; 179(6): 3973 - 3981.
[Abstract] [Full Text] [PDF]

Home page
 J. Immunol.Home page
D. M. Lewinsohn, J. E. Grotzke, A. S. Heinzel, L. Zhu, P. J. Ovendale, M. Johnson, and M. R. Alderson
Secreted Proteins from Mycobacterium tuberculosis Gain Access to the Cytosolic MHC Class-I Antigen-Processing Pathway
J. Immunol., July 1, 2006; 177(1): 437 - 442.
[Abstract] [Full Text] [PDF]

Home page
 Am. J. Respir. Crit. Care Med.Home page
C. Carranza, E. Juarez, M. Torres, J. J. Ellner, E. Sada, and S. K. Schwander
Mycobacterium tuberculosis Growth Control by Lung Macrophages and CD8 Cells from Patient Contacts
Am. J. Respir. Crit. Care Med., January 15, 2006; 173(2): 238 - 245.
[Abstract] [Full Text] [PDF]

Home page
 Am. J. Respir. Crit. Care Med.Home page
I. Smith, C. Nathan, and H. H. Peavy
Progress and New Directions in Genetics of Tuberculosis: An NHLBI Working Group Report
Am. J. Respir. Crit. Care Med., December 15, 2005; 172(12): 1491 - 1496.
[Abstract] [Full Text] [PDF]

Home page
 J. Immunol.Home page
V. Lazarevic, D. Nolt, and J. L. Flynn
Long-Term Control of Mycobacterium tuberculosis Infection Is Mediated by Dynamic Immune Responses
J. Immunol., July 15, 2005; 175(2): 1107 - 1117.
[Abstract] [Full Text] [PDF]

Home page
 J. Immunol.Home page
H. Shams, P. Klucar, S. E. Weis, A. Lalvani, P. K. Moonan, H. Safi, B. Wizel, K. Ewer, G. T. Nepom, D. M. Lewinsohn, et al.
Characterization of a Mycobacterium tuberculosis Peptide That Is Recognized by Human CD4+ and CD8+ T Cells in the Context of Multiple HLA Alleles
J. Immunol., August 1, 2004; 173(3): 1966 - 1977.
[Abstract] [Full Text] [PDF]

Home page
 Am. J. Respir. Crit. Care Med.Home page
M. J. Tobin
Tuberculosis, Lung Infections, Interstitial Lung Disease, Social Issues and Journalology in AJRCCM 2003
Am. J. Respir. Crit. Care Med., January 15, 2004; 169(2): 288 - 300.
[Full Text] [PDF]

This Article
Right arrow Abstract Freely available
Right arrow Full Text (PDF)
Right arrow Online Supplement
Right arrow All Versions of this Article:
200306-837OCv1
168/11/1346    most recent
Right arrow Alert me when this article is cited
Right arrow Alert me if a correction is posted
Services
Right arrow Similar articles in this journal
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Download to citation manager
Right arrow reprints & permissions
Citing Articles
Right arrow Citing Articles via HighWire
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by Lewinsohn, D. A.
Right arrow Articles by Lewinsohn, D. M.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Lewinsohn, D. A.
Right arrow Articles by Lewinsohn, D. M.

HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS
Proc. Am. Thorac. Soc. Am. J. Respir. Cell Mol. Biol.
Copyright © 2003 American Thoracic Society