| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
Published online before print March 14, 2005
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
* Department of Microbiology, Immunology and Pathology, Colorado State University, Fort Collins;
Division of Pulmonary Biology, Children’s Hospital Research Foundation, University of Cincinnati, Ohio; and
Trudeau Institute, Saranac Lake, New York
1Correspondence: Mycobacteria Research Laboratories, Department of Microbiology, Immunology and Pathology, Colorado State University, 200 Lake Street, Fort Collins, CO 80523. E-mail: malba{at}lamar.colostate.edu
 |
ABSTRACT |
|---|
 TOP ABSTRACT INTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
(TNF-) nor the chemokines, regulated on activation, normal T expressed and secreted (RANTES), macrophage-inflammatory protein-1ß (MIP-1ß), MIP-1, and lymphotactin, which are required for recruitment of lymphocytes and expression of a T helper cell type 1 (TH1) response within the lungs. In contrast, transgenic mice overexpressing GM-CSF in the lungs but with a lack of GM-CSF in other organs (GM+) are able to recruit lymphocytes and to express a TH1 response with production of TNF- and interferon-
in the lungs. However, GM+ mice succumb to infection between 60 and 90 days post-challenge, as they are unable to develop a normal granulomatous response. Although GM+ mice are able to express the chemokine RANTES, they lack the ability to express other inflammatory chemokines such as lymphotactin and MIP-1ß. We conclude that GM-CSF is essential to the recruitment of lymphocytes and expression of a TH1 response in the lung, to the generation of a normal mononuclear granuloma, and most importantly, to the containment of M. tuberculosis bacterial growth.
Key Words: GM-CSF • cytokines • chemokines • granuloma
|
INTRODUCTION |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
In mice infected in the lungs with the intracellular pathogen M. tuberculosis, macrophages play central roles [11
12
13
]. Alveolar macrophages encounter the bacillus in the alveolar spaces, dendritic cells almost certainly carry the organism to local lymphoid tissues [14
, 15
], and epithelioid macrophages and incoming monocytes form the basis of the tissue granuloma designed to wall-off the sites of infection and prevent further dissemination of the disease [16
, 17
]. These cells interact throughout this process with T lymphocytes of the CD4 T helper cell type 1 (TH1), in which a cytokine pathway involving interleukin (IL)-12 and interferon- (IFN-) mediates the protective host response [18
19
20
]. Disruption of this pathway leads to uncontrolled bacterial growth in the lungs, giving rise to severe granulomatous inflammation and fatal necrosis [18
].
Given the importance of macrophage functions in controlling pulmonary tuberculosis, we examined the course of M. tuberculosis infection in the lungs of mice lacking GM-CSF (GM-CSF KO) [1 ] and in mice wherein organ-specific promoter activity leads to high expression of GM-CSF in the lungs but not other organs {surfactant protein C (SPC)-GM+/+/GM–/– mice or GM+ mice [6 , 10 ]}. The results of this study show that increased expression of GM-CSF in the lungs does not translate into increased resistance to M. tuberculosis infection and that GM+ and GM-CSF KO mice succumbed to the disease. In the latter, the disease process was rapid and characterized by severe necrotizing pneumonia, whereas the events in the GM+ mice were much slower and more complex. A prominent feature in the GM+ mice was the accumulation of mononuclear cells around blood vessels (perivascular cuffing), suggesting disruption of the normal chemokine responses needed to focus T cells and macrophages into lesions, a hypothesis further supported by our demonstration of low levels of expression of "type 1" chemokines, macrophage-inflammatory protein-1ß (MIP-1ß) and activation-induced, T cell-derived, and chemokine-related cytokine (ATAC)/lymphotactin, in the lungs of these mice. As a result of the absence of a granulomatous response in the lung tissues, bacteria formed large rafts promoting local tissue necrosis. These data suggest that in addition to promoting macrophage function in the lungs, GM-CSF also plays a role in the granulomatous response during pulmonary M. tuberculosis infection.
|
MATERIALS AND METHODS |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
Bacteria and infection
Mice were infected via the aerosol route with a low dose [
200 colony-forming units (CFU)/lung] of bacteria M. tuberculosis strain Erdman. The number of viable bacteria in the lungs and spleen was determined by plating serial dilutions of individual whole lung homogenates onto nutrient Middlebrook 7H11 agar and counting bacterial colony formation after 3 weeks incubation at 37°C. The data were expressed as the log10 value of the mean number of colonies counted (n=4 animals). Mice were monitored on a daily basis and were euthanized when they displayed symptoms associated with disease.
Histology
To study the gross histology of the lung during pulmonary tuberculosis, mice were killed at 35 days post-challenge by CO2 asphyxiation, and the lungs were infused with 10% neutral-buffered formalin prior to removing them from the thoracic cavity. The lungs were fixed in 10% neutral-buffered formalin during 72 h inside the BSL-3 laboratory.
Histological analysis of lung sections was done by hematoxylin and eosin (H&E) and acid fast staining. The accessory lobe from each mouse was also fixed in 10% neutral-buffered formalin and embedded in paraffin. Sections from these samples were made and stained for H&E or Kinyoun methods for acid fast staining.
Isolation of lung cells
Isolation of lung cells was performed by protocols already described [11
]. Briefly, mice were killed, and the pulmonary cavity was opened. The blood circulatory system in the lungs was cleared by perfusing through the pulmonary artery with 3 ml saline containing 50 units/ml heparin (Sigma-Aldrich, St. Louis, MO). Lungs were removed aseptically and cut into small pieces in cold RPMI (Life Technologies, Grand Island, NY). The dissected tissue was then incubated in RPMI containing collagenase XI (0.7 mg/ml, Sigma-Aldrich) and type IV bovine pancreatic DNAse (30 µg/ml, Sigma-Aldrich) for 30–45 min at 37°C. The action of the enzymes was stopped by adding 10 ml completed RPMI (cRPMI) consisting of RPMI-1640 medium (Life Technologies), supplemented with 1% glutamine (Sigma-Aldrich), 0.1 mM nonessential amino acids (Life Technologies), 50 µM 2-mercaptoethanol (Sigma-Aldrich), 1% penicillin-streptomycin (Sigma-Aldrich), and 10% fetal bovine serum (FBS), and digested lungs were further disrupted by gently pushing the tissue through a nylon screen. The single-cell suspension was then washed and centrifuged at 200 g. To lyse contaminating red blood cells, the cell pellet was incubated for 5 min at room temperature with 5 ml Gey’s solution (NH4Cl and KHC03).
Tissue culture
Lung or splenocyte cell suspensions (5x106 cells/ml) from three groups of mice that were previously challenged by the aerosol route with a low dose of M. tuberculosis were cultured in the presence of M. tuberculosis culture filtrate proteins (CFP), 1 µg/ml, or media. Cells were cultured during 72 h at 37°C and 5% CO2, and the supernatants from those cultures were assayed using the inflammatory cytometric bead array (CBA) kit (BD Biosciences, San Diego, CA). Data were analyzed using CBA software. Graphs represent the average concentration for each cytokine in the supernatants from cell culture from four mice per group (n=4 for infected groups, and n=2 for naïve groups) per time-point after infection.
Analysis of cell populations in the lungs
Lung cells were suspended in deficient RPMI (Irvine Scientific, Santa Ana, CA), which was supplemented with 1% L-glutamine (Sigma-Aldrich), 1% Hepes (Sigma-Aldrich), 0.1% sodium azide (Sigma-Aldrich), and 2% FBS for flow cytometric studies.
Analysis of T cell populations was done using monoclonal antibodies (mAb) specific for mouse CD3 [145-2C11 hamster immunoglobulin G (IgG), group 1], CD4 (L3T4 clone RM4-5, rat IgG2a), CD8 (53-6.7, rat IgG2a), rat IgG2a (R35-95), rat IgG 2b (A95-1), or hamster IgG (Ha4/8), purchased from BD PharMingen (San Diego, CA) as direct conjugates to fluorescein isothyocyanate (FITC), peridinin chlorophyll-a protein (PerCP), or allophycocyanin (APC) in a purified form. Cell suspensions for each individual mouse were stained with specific mAb against murine CD3 and CD4 or CD8. Lung cells were washed in deficient RPMI (dRPMI; Irvine Scientific), which was supplemented with 1% L-glutamine, 1% Hepes, 0.1% sodium azide, and 2% FBS media, stained for 30 min at 4°C with direct conjugated antibodies, and washed twice with dRPMI. In addition, some cells were prepared for IFN- intracellular staining by incubation at 37°C and 5% CO2 with anti-CD3 (145-2C11, 0.1 µg/ml), anti-CD28 (37.51, 1 µg/ml), and monensin (30 µM) for 4 h. After washing the cells with cell-permeabilizing solution, the presence of intracellular IFN- was determined using a mAb against murine IFN- (XMG1-2) conjugated to phycoerythrin (PE). Acquisition was performed on a FACSCalibur (Becton Dickinson, Mountain View, CA), and data were analyzed using CellQuest software (Becton Dickinson). Cells were gated for lymphocytes by their characteristic scatter profile, and 50,000 events in the lymphocyte gate per sample were counted.
Analysis of macrophage and dendritic cell populations was done using mAb specific for CD11c (clone HL3, hamster IgG1), CD11b (membrane-activated complex-1, clone M1/70, rat IgG2a), Gr-1 (clone RB6-8C5, rat IgG2b), CD40 (clone 3-23, rat IgG2a), major histocompatibility complex (MHC) class II antigen (clone 2G9, rat IgG2a), rat IgG2a, rat IgG2b, rat IgG1, mouse IgG2a, and hamster IgG, purchased from BD PharMingen or eBioscience (San Diego, CA) as direct conjugates to FITC, PE, PerCP, PerCP-cyanine 5.5 (PerCP-Cy5.5), or APC. Cell acquisition was performed with a dual-laser flow cytometer, FACSCalibur (Becton Dickinson). Compensation of the spectral overlap for each fluorochrome was done gating in the R1 region (see Fig. 4A ) and using CD11b antigen. The data were analyzed using CellQuest software (Becton Dickinson).
|
RNAse protection assay (RPA)
A multiprobe RPA system mCK-5c from BD PharMingen was used to determine mRNA levels from genes of interest in the lungs of WT, GM-CSF KO, or GM+ mice after 0 or 21 days post-challenge with M. tuberculosis. The relative gene expression was determined and compared with the expression of housekeeping gene glyceraldehyde 3-phosphate dehydrogenase (GAPDH). The relative gene expression from each chemokine with housekeeping gene was analyzed using a densitometer, ImageQuant 1999 software (Molecular Dynamics, Sunnyvale, CA).
Statistical analysis
Student’s t-test was used for comparisons of means, and values of P < 0.05 were considered statistically significant.
|
RESULTS |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
|
Differences observed in lung pathology in the three strains
Lymphocytic peribronchiolar and perivascular aggregation followed by formation of granuloma structures in the lungs are the hallmark of containment of M. tuberculosis bacterial growth in WT mice [16
, 21
]. As shown in Figure 2
, although WT mice exhibit this normal progression of events, the gene-altered mice differ significantly from the norm.
|
When tissues were sectioned and analyzed for immunopathologic consequences of infection, the differences observed at the gross level were recapitulated at the microscopic level (Fig. 2B) . Specifically, lesions in WT mice showed the characteristic mononuclear granulomatous formation, and this level of organization was absent in the lungs of GM-CSF KO mice in which much of the lung tissues contained large areas of severe necrosis and tissue damage. As suggested by the gross examination of the tissue, the lungs of GM+ mice showed little evidence of granuloma formation, and instead, the primary facet of the lung sections from these mice was prominent perivascular and peribronchiolar cuffs of mononuclear cells.
The location of bacteria within the lesions was also altered between the strains of mice. As expected, the WT mice had low numbers of bacteria that were located within the granulomata; in contrast, there were prominent clumps of bacteria in the lungs of GM-CSF KO mice, and bacteria were scattered widely in the GM+ mice (Fig. 2C) .
Cellular influxes in the lungs
An effective acquired immune response against pulmonary M. tuberculosis infection is associated with recruitment of lymphocytes and the presence of IFN--producing cells [18
, 20
] in the lungs. To determine whether the reduced resistance of the gene-altered mice was associated with the inability to accumulate lymphocytes in the lung upon infection, the number of potentially protective lymphocytes within the lungs of infected animals was determined. As shown in Figure 3A
, mice unable to produce GM-CSF did not increase the numbers of lymphocytes in the lungs; however, mice overexpressing GM-CSF in the lungs, as in GM+ mice, had similar numbers of lymphocytes than WT animals, but as shown above, these were not reaching the sites of infection in the lung tissues. Similarly, after 21 days of infection, there was a significant accumulation of IFN--producing CD4+ and CD8+ lymphocytes in the lungs of infected WT mice and GM+ mice. In sharp contrast, mice unable to produce GM-CSF were unable to accumulate this type of cell into the lungs (Fig. 3B)
.
|
To address the activation state of the macrophages and dendritic cells within the infected lungs, the levels of expression of activation markers CD40 and MHC II in each cell population were determined in the WT and GM+. (GM-CSF KO mice were not analyzed, as there were not CD11c-positive cells.) Figure 4C shows that the expression of these antigens did not differ within the alveolar macrophage population; however, the expression of CD40 antigens but not MHC II antigens was higher on the dendritic cells from infected lung. (The difference in MFC for CD40 expression between WT and GM+ after infection was not statistically significant.)
The profile of inflammatory cytokines produced after infection differed among the mouse strains
It is apparent from the above data that the level of GM-CSF within the lung has an important effect on granuloma formation and accumulation of lymphocytes within the lung. Key mediators of granuloma formation and effector T cell accumulation are the inflammatory cytokines that are expressed in the lung upon infection.
To determine whether the level of GM-CSF alters levels of inflammatory cytokines, we analyzed the production of these cytokines by lung cells from infected mice in response to CFP from M. tuberculosis. In the absence of GM-CSF, lung cells were less able to produce tumor necrosis factor- (TNF-) and IFN- in response to antigen when compared with WT or GM+ mice. In contrast, the ability of lung cells from GM-CSF KO mice to produce IL-6 in response to antigen was far greater than with WT or GM+ lung cells (Fig. 5A
).
|
-producing antigen-specific response, the ability of lung cells and splenocytes to produce IFN- in response to antigen was assessed. As expected, cells from the lungs of infected WT mice made more IFN- than cells from the spleen. In contrast, in the absence of GM-CSF, as there was no antigen-specific response in the lung, there was a clear response in the spleen (P<0.043). In the presence of excess GM-CSF in the lung, there was a decreased but significant antigen-specific response in the lung (P<0.031) and a significant and increased response in the spleen (P<0.036; Fig. 5B
).
Changes in resistance and lung histology correlate with decreased production of inflammatory chemokines including ATAC/lymphotactin and MIP-ß
In addition to a role for inflammatory cytokines, the chemokines are crucially important in the development of granulomas and the accumulation of cells within tissues. The failure of GM+ mice to adequately focus lymphocytes and macrophages into the lung tissues suggested a potential defect in the appropriate production of chemokine gradients. To determine whether chemokine expression was indeed defective in the presence of excess GM-CSF, we determined the mRNA levels for chemokines within the lungs of infected mice. Although the pattern of chemokine expression was different for each experimental group, it was apparent that the absence of GM-CSF resulted in defective expression of mRNA for the T cell-recruiting chemokines RANTES, ATAC/lymphotactin, and MIP-1ß, with increased mRNA expression of MIP- when compared with the expression of mRNA in the WT. In contrast, the overexpression of GM-CSF resulted in similar expression of mRNA for RANTES as the WT but reduced expression of MIP-ß and ATAC/lymphotactin (Fig. 6
).
|
|
DISCUSSION |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
The failure of GM+ mice to focus T cells and macrophages into sites of infection in the lungs suggests a defect in cytokine and chemokine control of this event. This hypothesis is supported by our observation that the GM+ mice had normal levels of proinflammatory (TNF-) and chemokine RANTES but decreased levels of expression of MIP-1ß and ATAC/lymphotactin.
The dependence of antibacterial activity and granuloma formation on TNF- and the reduced levels of this cytokine in the GM-CSF KO mice are suggestive of why these mice are highly susceptible to M. tuberculosis infection. Defective expression of TNF- and innate responses to other microbial infections in GM-CSF KO were reported previously [5
]. Similarly, the absence of RANTES, MIP-ß, and ATAC/lymphotactin in the GM-CSF KO mice may explain their lack of ability to focus what is apparently a normal, splenic, antigen-specific response to the lung.
The lack of granulomatous response in the GM+ mice appeared to be the reason for the lack of containment of bacterial growth during the chronic stage of infection. GM+ mice have similar activation capacity of their dendritic and alveolar macrophage cell populations as the WT and were able to recruit lymphocytes into the lungs. It is interesting that the cytokine TNF-, which is essential for leukocyte recruitment into the lung, had similar levels of expression in GM+ and WT. However, leukocyte recruitment is also dependent on the expression of inflammatory chemokines type 1, such as RANTES, MIP-1, and MIP-1ß, and the expression of these chemokines differed between both groups of mice. These chemokines are usually produced de novo in response to infection and are known to be important in recruiting effector cells to sites of infection [23
], primarily via the CC chemokine receptor 5, found on macrophages, dendritic cells, and activated CD4 T cells [24
]. ATAC/lymphotactin is more selectively secreted by natural killer (NK) cells and CD8 cells [25
], both of which are known to respond to M. tuberculosis infection. We hypothesize therefore that overproduction of GM-CSF in the lungs depresses the production of the type 1 chemokines MIP-1ß and ATAC/lymphotactin but not RANTES, possibly by a physiological feedback inhibition mechanism. It is likely that cell extravasation occurs through the blood vessel mediated by the presence of RANTES, but once inside the lungs, leukocytes are prevented from further movement out into the lung tissues as a result of the lack of ATAC/lymphotactin and MIP-ß type 1 chemokines. Conversely, outside the lungs, events appear to be proceeding normally, and dendritic cells carry bacteria to lymphoid tissues or to the spleen, and the TH1 response emerges as expected.
These observations have similarities to other models. The progressive disease in the lungs of GM-CSF KO mice is similar to that in IFN- KO mice and was associated in both cases with severe necrosis and rapid death in the absence of any intact granuloma formation [18
]. In the case of the GM+ mice, the perivascular cuffing and lack of further cellular influx are similar to that in M. tuberculosis-infected mice lacking the ß2-microglobulin molecule [26
]. This may suggest that lack of production of type 1 chemokines (specifically, ATAC/lymphotactin) by MHC I-restricted CD8 or NK cells in the lungs may underlie this event, and this should be investigated further.
These findings indicate that the cytokine GM-CSF plays a selective and specific role in the lungs, controlling the maturation and differentiation of macrophage populations as well as affecting the homeostatic, local production of chemokine molecules that influence the influx of macrophages and activated T cells from the blood across the pulmonary vasculature. When GM-CSF is overexpressed, this homeostasis is disrupted, depressing specific type 1 chemokine production and preventing proper granuloma formation in the lung tissues.
|
ACKNOWLEDGEMENTS |
|---|
Received December 13, 2004; revised January 25, 2005; accepted February 3, 2005.
|
REFERENCES |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
gene-disrupted mice J. Exp. Med. 178,2243-2247 in resistance to Mycobacterium tuberculosis infection J. Exp. Med. 178,2249-2254, MIP-1ß, RANTES, and ATAC/lymphotactin function together with IFN- as type 1 cytokines Proc. Natl. Acad. Sci. USA 99,6181-6186This article has been cited by other articles:
|
|
 |
|
  D. M. Higgins, J. Sanchez-Campillo, A. G. Rosas-Taraco, J. R. Higgins, E. J. Lee, I. M. Orme, and M. Gonzalez-Juarrero Relative Levels of M-CSF and GM-CSF Influence the Specific Generation of Macrophage Populations during Infection with Mycobacterium tuberculosis J. Immunol., April 1, 2008; 180(7): 4892 - 4900. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
A. A. Ryan, T. M. Wozniak, E. Shklovskaya, M. A. O'Donnell, B. Fazekas de St. Groth, W. J. Britton, and J. A. Triccas Improved Protection against Disseminated Tuberculosis by Mycobacterium bovis Bacillus Calmette-Guerin Secreting Murine GM-CSF Is Associated with Expansion and Activation of APCs J. Immunol., December 15, 2007; 179(12): 8418 - 8424. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 D. Ordway, D. M. Higgins, J. Sanchez-Campillo, J. S. Spencer, M. Henao-Tamayo, M. Harton, I. M. Orme, and M. Gonzalez Juarrero XCL1 (lymphotactin) chemokine produced by activated CD8 T cells during the chronic stage of infection with Mycobacterium tuberculosis negatively affects production of IFN-{gamma} by CD4 T cells and participates in granuloma stability J. Leukoc. Biol., November 1, 2007; 82(5): 1221 - 1229. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 G.-H. Chen, M. A. Olszewski, R. A. McDonald, J. C. Wells, R. Paine III, G. B. Huffnagle, and G. B. Toews Role of Granulocyte Macrophage Colony-Stimulating Factor in Host Defense Against Pulmonary Cryptococcus neoformans Infection during Murine Allergic Bronchopulmonary Mycosis Am. J. Pathol., March 1, 2007; 170(3): 1028 - 1040. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
B. Carey, M. K. Staudt, D. Bonaminio, J. C. M. van der Loo, and B. C. Trapnell PU.1 Redirects Adenovirus to Lysosomes in Alveolar Macrophages, Uncoupling Internalization from Infection J. Immunol., February 15, 2007; 178(4): 2440 - 2447. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 K. Uchida, D. C. Beck, T. Yamamoto, P.-Y. Berclaz, S. Abe, M. K. Staudt, B. C. Carey, M.-D. Filippi, S. E. Wert, L. A. Denson, et al. GM-CSF Autoantibodies and Neutrophil Dysfunction in Pulmonary Alveolar Proteinosis N. Engl. J. Med., February 8, 2007; 356(6): 567 - 579. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 P.-Y. Berclaz, B. Carey, M.-D. Fillipi, K. Wernke-Dollries, N. Geraci, S. Cush, T. Richardson, J. Kitzmiller, M. O'Connor, C. Hermoyian, et al. GM-CSF Regulates a PU.1-Dependent Transcriptional Program Determining the Pulmonary Response to LPS Am. J. Respir. Cell Mol. Biol., January 1, 2007; 36(1): 114 - 121. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 T. C. Thacker, M. V. Palmer, and W. R. Waters Correlation of Cytokine Gene Expression with Pathology in White-Tailed Deer (Odocoileus virginianus) Infected with Mycobacterium bovis Clin. Vaccine Immunol., June 1, 2006; 13(6): 640 - 647. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
D. Ordway, M. Harton, M. Henao-Tamayo, R. Montoya, I. M. Orme, and M. Gonzalez-Juarrero Enhanced Macrophage Activity in Granulomatous Lesions of Immune Mice Challenged with Mycobacterium tuberculosis. J. Immunol., April 15, 2006; 176(8): 4931 - 4939. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 J. A. Tree, M. J. Elmore, S. Javed, A. Williams, and P. D. Marsh Development of a Guinea Pig Immune Response-Related Microarray and Its Use To Define the Host Response following Mycobacterium bovis BCG Vaccination Infect. Immun., February 1, 2006; 74(2): 1436 - 1441. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
), and GM+ (
) mice after infection with M. tuberculosis. Mice were infected with 
