Genetically Determined Disparate Innate and Adaptive Cell-Mediated ...

Download PDF
2 downloads 0 Views 2MB Size Report
Comment
The magnitude of cellular responses was also much lower in the lung of BALB/c mice ...... holds an MRC-Canadian Lung Association fellowship. Z.X. holds an.
INFECTION AND IMMUNITY, Dec. 2000, p. 6946–6953 0019-9567/00/$04.00⫹0 Copyright © 2000, American Society for Microbiology. All Rights Reserved.

Vol. 68, No. 12

Genetically Determined Disparate Innate and Adaptive Cell-Mediated Immune Responses to Pulmonary Mycobacterium bovis BCG Infection in C57BL/6 and BALB/c Mice JULIA WAKEHAM, JUN WANG,

AND

ZHOU XING*

Department of Pathology and Molecular Medicine and Division of Infectious Diseases, Center for Gene Therapeutics, McMaster University, Hamilton, Ontario L8N 3Z5, Canada Received 13 April 2000/Returned for modification 28 May 2000/Accepted 18 August 2000

The current study was designed to investigate the impact of genetic heterogeneity on host immune responses to pulmonary intracellular infection by using two mouse strains of distinct genetic background, C57BL/6 and BALB/c mice, and a model intracellular pathogen, Mycobacterium bovis BCG. Upon infection, compared to C57BL/6 mice, BALB/c mice developed an earlier response of interleukin 12 (IL-12), gamma interferon (IFN-␥), tumor necrosis factor alpha, and macrophage chemoattractive protein 1, and greater neutrophilic influx to the lung by days 7 and 14. However, the level of these cytokines at days 27, 43, and 71 was much lower in BALB/c mice than in C57BL/6 mice. The magnitude of cellular responses was also much lower in the lung of BALB/c mice around day 27. Histologically, while C57BL/6 mice developed lymphocytic granulomas, BALB/c mice displayed atypical granulomas in the lung. Of importance, the level of type 2 cytokines IL-4 and IL-10 remained low and similar in the lung of both C57BL/6 and BALB/c mice throughout. Furthermore, lymphocytes isolated from systemic and local lymphoid tissues of infected BALB/c mice demonstrated a markedly lower antigen-specific IFN-␥ recall response. While the number of mycobacterial bacilli recovered from both the lung and spleen of BALB/c mice was similar to that in C57BL/6 mice at day 14, it was higher than that in C57BL/6 mice at day 43. However, it was eventually leveled off to that in C57BL/6 counterparts later. These results suggest the following: (i) genetic heterogeneity can lead to differential innate and adaptive cell-mediated immune responses to primary pulmonary mycobacterial infection; (ii) it is the level of adaptive, but not innate, immune response that is critical to host resistance; and (iii) a lower type 1 immune response in BALB/c mice is not accompanied by a heightened type 2 response during pulmonary mycobacterial infection. However, while two strains of mice, C57BL/6 (H-2b) and BALB/c (H-2d), bear the same susceptible allele of the Bcg gene (13, 20), they demonstrate contrasting susceptibilities to certain intracellular pathogens. In this regard, C57BL/6 mice were found to be resistant to Leishmania major or Yersinia enterocolitica infection, whereas BALB/c mice were susceptible (1, 7, 14). Such contrasting nature of host defense determined by genetic background is attributable to a distinct type of cytokine responses during leishmaniasis. Leishmania infection elicits a type 1 cytokine response in C57BL/6 but a type 2 cytokine response in BALB/c mice, characterized by increased IL-12 and IFN-␥ and IL-4, IL-10, and IL-5, respectively. A biased type 2 immune response was also found in BALB/c mice susceptible to Chlamydia trachomatis mouse pneumonitis infection, different from a type 1 profile in resistant C57BL/6 mice (25). The propensity of BALB/c mice to develop a type 2 immune response may be accounted for in part by the requirement of additional cofactors for Th1-type differentiation, different from C57BL/6 hosts (19). Differences have also been noticed between C57BL/6 and BALB/c mice in their immune responses to mycobacterial vaccination. Following intravenous or subcutaneous immunization with Mycobacterium bovis BCG, C57BL/6 mice developed a stronger immune response to intravenous rechallenge with BCG than BALB/c mice (9, 26). Such an enhanced protective immune response in C57BL/6 mice was associated with a type 1 cytokine response more pronounced than that in BALB/c mice. In this regard, treatment with recombinant type 1 cytokine IL-12 of BALB/c mice enhanced host defense against intravenous tuberculous infection (6). However, it has remained controversial whether the suppressed type 1 immune

Host defense to intracellular infections caused by pathogens such as mycobacteria, salmonella, and leishmania involves both innate and adaptive cell-mediated immune responses. It is believed that the innate immunity provides the initial resistance in the first two to three weeks after infection before the adaptive type 1 cell-mediated immunity fully develops. The major cellular components involved in innate immunity include neutrophils, macrophages, and NK cells, whereas lymphocytes and macrophages are the major effector cells in cellmediated immunity against intracellular infection. Innate immune components serve as a linker to cell-mediated immunity in part by releasing soluble signals such as interleukin 12 (IL-12). Cell-mediated immunity plays an essential role in conferring the ultimate protection against intracellular infection (2, 16, 18). Compelling evidence by us and others indicates that type 1 cytokines, including IL-12, gamma interferon (IFN-␥), and tumor necrosis factor alpha (TNF-␣) play a critical role in the development of type 1 cell immunity against intracellular infections (4, 5, 12, 21, 23, 24). Increasing evidence from both human and experimental studies suggests that host genetic heterogeneity affects the nature and/or the level of immune responses to intracellular infections by virus, bacteria, and parasites (8, 13). One of the genetic loci that affects the innate immunity is the Bcg gene.

* Corresponding author. Mailing address: Rm. 4H19, Health Science Center, Department of Pathology and Molecular Medicine, McMaster University, 1200 Main St. West, Hamilton, Ontario L8N 3Z5, Canada. Phone: (905) 525-9140, ext. 22471. Fax: (905) 522-6750. E-mail: [email protected]. 6946

VOL. 68, 2000

IMMUNE RESPONSES TO BCG IN C57BL/6 AND BALB/c MICE

response to mycobacterial infection in BALB/c mice is a result of enhanced counteracting type 2 cytokine response (9, 26), as demonstrated in BALB/c mice infected with other types of intracellular pathogens (1, 7, 14, 25). Tuberculous mycobacterial infection is primarily a pulmonary infectious disease, and increasing evidence has suggested that host responses to mycobacterial infection originating in the lung are quite different from those to systemic mycobacterial infection (3, 15). However, the immune profile and the nature of host responses in the lung to primary pulmonary mycobacterial infection in hosts of distinct genetic background has remained to be determined. In the present study, we examined (i) whether the levels of both innate and adaptive type 1 immune responses to primary pulmonary mycobacterial infection were different between C57BL/6 and BALB/c mice; (ii) whether the level of type 1 cytokines was associated with tissue immune responses in the lung; and (iii) whether type 2 cytokines played a role in regulating the level of type 1 immune responses. We found that compared with C57BL/6 mice, while BALB/c mice developed a greater early innate immune response, they had an impaired ability to develop a vigorous adaptive type 1 cell-mediated immune response to mycobacterial infection in the lung. Of importance, BALB/c mice, unlike their response to other intracellular pathogens, did not develop a polarized type 2 immune response to pulmonary mycobacterial infection.

MATERIALS AND METHODS Mice. Female C57BL/6 and BALB/c mice 10 to 14 weeks old were used (Harlan, Indianapolis, Ind.). All mice were housed in autoclaved cages with autoclaved bedding, food, water, and microfilter lids in a pathogen-free level B facility. All experiments performed were in accordance with the guidelines of the Animal Research Ethics Board of McMaster University. Preparation of M. bovis BCG. M. bovis BCG (Connaught Laboratories Limited, North York, Ontario, Canada) was grown in Middlebrook 7H9 broth (Difco, Detroit, Mich.) supplemented with Middlebrook OADC enrichment (Gibco-BRL, Gaithersburg, Md.), 0.002% glycerol, and 0.05% Tween 80. Lyophilized BCG was reconstituted in saline–0.05% Tween 80 and used to inoculate a small (20-ml) starter culture. The culture was incubated at 37°C for 7 days with gentle aeration. The starter culture was subinoculated with 0.1 ml of fresh broth. After 4 days of incubation, the culture was harvested by centrifugation, and the cell pellet was resuspended in 7H9 broth to an absorbance of 2.8 at 600 nm. This cell suspension was aliquoted and stored at ⫺70°C until needed. After thawing, viable cell counts were determined by plating serial dilutions of the suspension on Middlebrook 7H11 agar plates (Gibco-BRL) and incubating at 37°C. Pulmonary mycobacterial infection. Pulmonary mycobacterial infection was established via the airway as previously described (4, 5). Prior to infection, BCG stock solution was diluted in phosphate-buffered saline (PBS), and the preparation was sonicated to ensure proper dispersion of mycobacteria. Mice were infected by intratracheal instillation of live BCG at a dose of 5 ⫻ 105 CFU in a total volume of 40 ␮l/mouse. Groups of four to five mice per time point per mouse strain were set up for each experiment. At days 7, 14, 27, 43, 57, and 71 postinfection, mice were anesthetized, bled retro-orbitally for serum preparation, and then exsanguinated by bleeding of the abdominal vessels. Lungs were removed and subjected to bronchoalveolar lavage (BAL), followed by perfusion with 10% formalin. At days 43 and 71, mouse lungs and spleens were also used for the colony enumeration assay. Fixed lungs and spleens were further processed for histologic analysis. BAL and cytologic analysis. After retro-orbital bleeding, anesthetized mice were exsanguinated via the abdominal vessels, followed by removal of the lungs, with the heart and a portion of the trachea intact. To collect BAL fluid (21, 24), a polyethylene tube (Becton Dickinson, Sparks, Md.) was used to cannulate the trachea. Lungs were lavaged twice with PBS (0.25 and 0.20 ml), and approximately 0.4 ml of BAL fluid was consistently retrieved. All BAL samples were kept on ice until processing. BAL samples were spun in a microcentrifuge (Hermle-Z180M) at 4,000 rpm for 1 min at 4°C, and supernatants were removed and stored at ⫺20°C for cytokine analysis. Cell pellets were resuspended in 300 to 500 ␮l of PBS, and total cell counts were determined on a hemacytometer. Cytospins were made in a cytospin machine (Shandon Inc., Pittsburgh, Pa.) and stained using Diff-Quick stain (Baxter, McGaw Park, Ill.) for differential cell counting. Routinely, 300 to 500 cells/cytospin were differentiated in a random fashion.

6947

Processing and histologic assessment of lung and spleen tissues. Lungs and spleens were fixed in 10% formalin as described above. Both left and right lungs were sectioned from top to bottom, resulting in four to five cross-sectional pieces of tissue from each side. Tissues were then embedded in paraffin, cut into 4- to 5-␮m-thick sections, and stained with hematoxylin and eosin. Other tissue sections were subjected to Ziehl-Neelsen staining, which is specific for mycobacteria. Measurement of cytokines in BAL and sera. Cytokines and chemokines were measured in BAL fluid and sera by specific enzyme-linked immunosorbent assay (ELISA). All ELISA kits were purchased from either R & D Systems, Minneapolis, Minn. (IFN-␥, TNF-␣, IL-4, IL-10, macrophage chemoattractive protein 1 [MCP-1]) or Biosource, Montreal, Quebec, Canada (IL-12). The sensitivity of detection for all of these ELISA kits was ⱕ5 to 10 pg/ml. Mycobacterial colony enumeration. At days 43 and 71 postinfection, lungs were removed and aseptically placed in 4.5 ml of PBS–0.05% Tween 80 on ice. Spleens were snap-frozen in liquid nitrogen and stored at ⫺70°C for later use. Lungs were cut into small pieces (2 to 3 mm thick) under sterile conditions. Lung pieces or spleens (each in 4.5 ml of PBS–0.05% Tween 80 buffer) were homogenized with a tissue homogenizer. Homogenates were allowed to settle on ice for 30 min, and 200 ␮l of properly diluted homogenates was plated onto each plate of Middlebrook 7H10 agar containing OADC enrichment (Difco). Plates were incubated inside semisealed plastic bags at 37°C. Colonies were counted using a dissecting microscope, at day 14 for spleens and at day 11 for lungs (21, 24). Alveolar macrophage culture and in vitro stimulation. BAL was carried out in three to four naive noninfected C57BL/6 and BALB/c mice as described above, and alveolar macrophages were plated into 96-well plates at a density of 0.1 million cells/well in 300 ␮l of culture medium for 3 days under different conditions. Supernatants were measured for TNF-␣ by ELISA. Isolation of splenocytes and pulmonary lymph node lymphocytes. Spleens were removed from the mice after bleeding and stored in the prepared buffer described for the lung cell isolation procedure. Spleens were then mashed through sterile metal screens immersed in culture medium. Cell suspensions were filtered through two layers of nylon membrane (55-␮m pore size), collected in a 50-ml tube containing 25 ml of PBS, and centrifuged at 1,000 rpm (Beckman TJ-6 instrument) for 10 min at 4°C. Supernatant was removed and the red blood cells in the pellet were lysed using 1 ml of ACK buffer (23). After 1 min, PBS was poured into the tube to build the volume to 50 ml, the suspension was centrifuged again, and the resultant pellet was resuspended in RPMI culture medium. Pulmonary draining mediastinal lymph nodes (MLN) were also removed and lymphocytes were isolated as previously described (23). Briefly, the thoracic cavity was opened and MLN were removed. MLN pooled from several mice of the same strain were ground between two microscopic slides, and lymphocytes were released into culture medium containing 10% fetal calf serum and 1% penicillin and streptomycin. The resultant cell suspension was filtered through two layers of nylon membrane. Cells from individual mice of the same strain were pooled, total cell counts and viability were determined, and cytospins were prepared from pooled groups of cells. More than 97% of these cells were found to be lymphocytes. Cells were resuspended in RPMI and added to 96-well plates at a concentration of 0.5 million cells/well. Cells were cultured without any stimulation, or with 10 ␮g of PPD (M. tuberculosis-derived purified protein derivative; Connaught Laboratories) per ml for 72 h at 37°C. Supernatants were collected and stored at ⫺20°C until the cytokine assay. Statistical analysis. Data were subjected to Student’s t test for analysis of statistical significance, and a P value of ⱕ0.05 was considered to be significant.

RESULTS Cellular profiles in the BAL fluid. To compare the cellular responses to pulmonary mycobacterial infection in C57BL/6 and BALB/c mice, a quantitative evaluation was carried out by examining the differential cell types in BAL fluids recovered from the lungs of both C57BL/6 and BALB/c mice at days 14, 27, 43, and 71 postinfection (Fig. 1). At day 14, while the number of macrophages remained similar between C57BL/6 and BALB/c mice, the number of neutrophils in the lung of BALB/c mice was much greater than that in C57BL/6 mice (Fig. 1B). However, while cellular responses reached a peak around day 27 in both strains of mice, the total cell recovery in BAL fluids was much higher from C57BL/6 mice than from BALB/c mice. This difference was also seen in the differential counts of cells retrieved from BAL fluid, including macrophages/monocytes (P ⫽ 0.0026), lymphocytes (P ⫽ 0.00007), and neutrophils (P ⫽ 0.00017). A comparison of baseline levels of cells obtained from the BAL fluids of noninfected C57BL/6 and BALB/c mice indicates that numbers of cells retrievable from BAL fluid did not differ significantly between the two strains (C57BL/6, 1.23 ⫻ 105 cells, versus BALB/c, 0.91 ⫻ 105

6948

WAKEHAM ET AL.

INFECT. IMMUN.

FIG. 1. Cellular responses in the lung of C57BL/6 and BALB/c mice. Total leukocytes were recovered from BAL fluids collected at various time points postinfection, and differentials were determined on cytospins. Results are expressed as means ⫾ standard errors of the mean (error bars) from four mice per strain per time. The number of neutrophils on day 14 between C57BL/6 and BALB/c mice is significantly different (P ⫽ 0.0133). The differences at day 27 between C57BL/6 and BALB/c mice in the number of macrophages (A), neutrophils (B), and lymphocytes (C) are all statistically significant (P ⫽ 0.0026, 0.00017, and 0.00007, respectively) and is representative of at least two independent experiments. The basal numbers of total cells in BAL fluids from naive noninfected C57BL/6 and BALB/c mice are 1.23 ⫻ 105 and 0.91 ⫻ 105, respectively, and more than 95% of these cells are alveolar macrophages.

cells). At days 43 and 71 postinfection, the total cell counts and differentials were similar between the two strains of mice. The number of total leukocytes markedly decreased in the lung of both strains of mice by day 71 (Fig. 1). Thus, evaluation of the cellular profiles in BAL fluid between the two strains of infected mice indicates an earlier higher innate neutrophilic response but a much lower cell-mediated immune response later during the course of pulmonary mycobacterial infection in BALB/c mice, compared to C57BL/6 mice. Histopathology of lung tissues. To examine tissue responses, histologic assessment of lung tissues from mice sacrificed at days 7, 14, 27, 43, 57, and 71 was carried out. At day 7 postinfection there were few signs of inflammation evident in the lung tissue of C57BL/6 or BALB/c mice. At day 14 postinfection, the lungs of C57BL/6 mice exhibited distinct and wellformed granulomas comprising macrophages, epithelioid cells, and a small number of lymphocytes. Perivascular and peribronchiolar lymphocytic accumulation was evident. By days 27 and 43, the inflammatory tissue response in the C57BL/6 mouse lungs became intensified and diffuse, resulting in many granulomas densely packed with macrophages, epithelioid cells, and infiltrating lymphocytes and neutrophils (referred to as lymphocytic granuloma) (Fig. 2A and C). From day 57 onwards, resolution of inflammation had begun and by day 71 there was primarily loose, inflammatory accumulation in the peribronchial and perivascular areas, and some foamy macrophages were present in alveolar spaces. In comparison, the lungs of BALB/c mice from the same time points postinfection exhibited different granulomatous responses. By day 14 postinfection, there was definite perivascular and peribronchial inflammatory response, with lymphocytes, neutrophils, and macrophages. The overall extent of inflammation was greater in the lung tissue of BALB/c mice than that in C57BL/6 mice by day 14 postinfection. At day 27, there was a much less rigorous granulomatous response in the lung of BALB/c mice than in C57BL/6 mice. Furthermore, at both days 27 and 43, in contrast to the lymphocytic granuloma observed in the lung of C57BL/6 mice, BALB/c mice displayed atypical granuloma formation in the lung (Fig. 2C and D). Such atypical granulomas were composed of fewer macrophages and epithelioid cells and lacked lymphocytic infiltra-

tion. Instead, there was a rather separated lymphocytic accumulation surrounding granulomas. Resolution of inflammation had not begun by day 57 postinfection in the lung of BALB/c mice, and by day 71, a significantly greater remaining inflammatory response was noted compared to that in the lungs of C57BL/6 mice. Type 1 and type 2 cytokine responses in the lung. To investigate the potential molecular mechanisms underlying differential innate and adaptive cell immune responses in C57BL/6 and BALB/c mice, we next characterized the in vivo cytokine responses locally in the lung during pulmonary mycobacterial infection. Cytokine contents in the BAL fluids collected at days 7, 14, 27, 43, and 71 postinfection were assessed by ELISA. A panel of cytokines was examined, including the type 1 cytokines IL-12, IFN-␥, and TNF-␣, and chemokine MCP-1. We and others have previously found that IL-12, IFN-␥, and TNF-␣ are all important mediators of protective immune responses against mycobacterial infection (5, 12, 21, 23, 24). MCP-1 is a potent chemotactic factor for monocytes and also regulates monocyte cytokine production (11). Of interest, the level of IL-12, IFN-␥, and TNF-␣ in the lung increased at day 7 and peaked at day 14 in the lung of BALB/c mice, whereas these cytokines were hardly detectable in the lung of C57BL/6 mice (Fig. 3A to C). However, the level of these type 1 cytokines peaked around day 27 in the lung of C57BL/6 mice and was far higher than that detected in BALB/c mice thereafter. The difference between C57BL/6 and BALB/c mice at day 27 was found to be highly statistically significant (P ⫽ 0.003, P ⫽ 0.017, and P ⫽ 0.004 for IL-12, IFN-␥, and TNF-␣, respectively). Furthermore, chemokine MCP-1 levels followed a similar pattern. While it peaked at day 14 postinfection in the lung of BALB/c mice, it was undetectable in the lung of C57BL/6 mice at this time. However, by day 27 postinfection C57BL/6 lungs contained much higher levels of MCP-1 than BALB/c lungs (P ⫽ 0.042) (Fig. 3D). The level of MCP-1 remained higher in the lung of C57BL/6 mice at days 43 and 71 postinfection, although the difference was not statistically significant. Thus, the difference in cytokine levels correlated with the difference in the level of cellular responses seen in the lungs of these two strains of mice (Fig. 1 to 3). To investigate whether the lower type 1 immune responses in the

VOL. 68, 2000

IMMUNE RESPONSES TO BCG IN C57BL/6 AND BALB/c MICE

6949

FIG. 2. Histopathology of lungs of C57BL/6 (A and C) and BALB/c (B and D) mice. Lungs collected at days 27 (A and B) and 43 (C and D) postinfection were processed. Abbreviations: b, bronchus; LG, lymphocytic granuloma; AL, atypical granuloma with thick lymphocytic ring but little lymphocytic infiltration. Magnification, ⫻380.

lung of BALB/c mice was associated with a heightened type 2 cytokine response, we measured the level of type 2 cytokines IL-4 and IL-10 in the lung. IL-4 and IL-10 are known for their inhibitory effects on type 1 cytokine expression and type 1 immune responses (17). We found that not only were the levels of these cytokines not higher in BALB/c than in C57BL/6 mice but they remained constantly minimal in the lung of BALB/c mice (Table 1). Cytokine response by naive alveolar macrophages. Since we detected a higher level of cytokines in the lung of BALB/c mice than in C57BL/6 mice in the initial stage of infection, it was possible that the innate response of alveolar macrophages in BALB/c mice was different from that in C57BL/6 mice. To this end, we isolated macrophages from the lung of naive C57BL/6 and BALB/c mice and cultured these cells under different conditions. Indeed, we found that macrophages from BALB/c mice released significantly more TNF-␣ upon stimulation by live BCG organisms and IFN-␥ than those of C57BL/6 mice (Fig. 4). Most prominently, BALB/c mouse macrophages released far more TNF-␣ than those from C57BL/6 counterparts upon stimulation with LPS or BCG plus IFN-␥ (Fig. 4). Mycobacterial burden in tissues assessed by colony enumeration assay. To examine the level of immune protection against pulmonary mycobacterial infection, we compared the levels of mycobacterial burden in the lung and spleen of C57BL/6 and BALB/c mice. We chose to focus on days 14, 43, and 71 postinfection. The number of bacilli was slightly, but

not significantly, higher in the lungs of BALB/c mice than in C57BL/6 mice at day 14 (Fig. 5A). However, by day 43, it was significantly greater in the lungs of BALB/c mice than in C57BL/6 mice (68,962 ⫾ 19,977 versus 15,570 ⫾ 8666; P ⫽ 0.025) (Fig. 5A). However, the level of infection decreased significantly in the lung of both C57BL/6 and BALB/c mice by

TABLE 1. Type 2 cytokines in the lung of C57BL/6 and BALB/c mice Cytokine

Day

Cytokine level (pg/ml of BAL fluid) in mouse straina C57BL/6

BALB/c

IL-4

14 27 43 71

8.8 ⫾ 2 10.5 ⫾ 1.3 9.4 ⫾ 4.8 3.0 ⫾ 0.5

0.9 ⫾ 0.6 0.6 ⫾ 0.3 4.8 ⫾ 0.6 0

IL-10

14 27 43 71

0.9 ⫾ 0.9 5.1 ⫾ 4.2 7.2 ⫾ 3.9 NDb

0.5 ⫾ 0.5 5.3 ⫾ 3.1 2.1 ⫾ 1.5 5.8 ⫾ 2.9

a Cytokines were measured with BAL fluids collected at times indicated. Samples from four mice per time per strain were measured, and results are expressed as means ⫾ standard errors of the means. b ND, not done.

6950

WAKEHAM ET AL.

INFECT. IMMUN.

FIG. 3. Level of type 1 cytokines and chemokine MCP-1 in the lung of C57BL/6 and BALB/c mice. The level of type 1 cytokines in the lung was determined by measuring (by ELISA) the level of IL-12 (A), IFN-␥ (B), TNF-␣ (C), and MCP-1 (D) in BAL fluids collected at various times. Results are expressed as means ⫾ standard errors of the means (error bars) from four mice per strain per time. The differences between C57BL/6 and BALB/c mice in IL-12, IFN-␥, TNF-␣, and MCP-1 at day 14 are all statistically significant (P ⫽ 0.050, 0.033, 0.0009, and 0.009, respectively). The differences at day 27 are also all statistically significant (P ⫽ 0.031, 0.017, 0.004, and 0.021, respectively) and are representative of at least two independent experiments.

day 71 postinfection while BALB/c mice still had a greater number of CFU in their lungs than did C57BL/6 mice. While the infection was undetectable at day 14 in the spleen in both BALB/c and C57BL/6 mice, the number of bacilli in the spleen of BALB/c mice was higher than that in C57BL/6 mice at day 43 (Fig. 5B). However, by day 71, the levels of mycobacterial burden were similar in the spleens of both C57BL/6 and BALB/c mice. Antigen-stimulated cytokine recall responses by spleen and lung draining lymph node lymphocytes. To examine whether there was a difference in Th1-type differentiation between C57BL/6 and BALB/c, we examined type 1 cytokine recall release by lymphocytes from the spleen stimulated by mycobacterial antigen. Splenocytes purified at day 27 postinfection from both C57BL/6 and BALB/c mice released little IFN-␥ in the absence of antigenic stimulation. However, upon PPD stimulation, C57BL/6 splenocytes released much more IFN-␥ than those from BALB/c mice (Fig. 6A). Furthermore, splenocytes from C57BL/6 mice also released more TNF-␣ upon antigen stimulation (Fig. 6B). In contrast, cells from either C57BL/6 or BALB/c mice released no IL-4 (Fig. 6C).

To examine whether there was also a difference between these two strains of mice in Th1-type differentiation in lung lymphocytes, we also examined antigen-specific IFN-␥ recall response by lymphocytes isolated from lung draining lymph nodes. We found that the MLN in BALB/c mice at day 27 postinfection were much smaller than those in C57BL/6 counterparts. As a result, these lymph nodes in each BALB/c mouse contained six times fewer lymphocytes than those in each C57BL/6 mouse (Table 2). Furthermore, when the same number of cells was cultured, BALB/c cells produced 50% less IFN-␥ upon mycobacterial antigen stimulation than C57BL/6 cells (Table 2). Thus, when the difference in the number of lymphocytes per mouse was taken into account, BALB/c lung lymph node cells produced 12 times less IFN-␥ than those in C57BL/6 mice (Table 2). DISCUSSION The objective of our current study was to investigate the level and the nature of immune responses in the lung to primary pulmonary intracellular infection elicited by mycobacteria in hosts with distinct genetic backgrounds. We chose to compare C57BL/6 (H-2b) and BALB/c (H-2d) mouse strains. These two mouse strains have genetic differences not only in the H-2 locus but in other H-2-associated genes, mimicking the differences among humans. We found that compared to C57BL/6 mice, while BALB/c mice were able to mount a stronger early innate response (the response before 2 to 3 weeks postinfection) to pulmonary mycobacterial infection, the magnitude of their adaptive type 1 immune response (the response after 2 to 3 weeks postinfection) was much lower, likely as a result of their poor ability to undergo Th1-type differentiation. However, BALB/c mice did not demonstrate a skewed type 2

VOL. 68, 2000

IMMUNE RESPONSES TO BCG IN C57BL/6 AND BALB/c MICE

FIG. 4. Innate cytokine release by naive alveolar macrophages. Alveolar macrophages were isolated from noninfected C57BL/6 and BALB/c mice and cultured for a period of 3 days without or with BCG (50 CFU/cell), IFN-␥ (400 pg/ml), lipopolysaccharide (LPS) (1 ␮g/ml), or both BCG and IFN-␥. Supernatants were measured for TNF-␣ content by ELISA. Results are expressed as means from duplicate wells. Unsti, unstimulated.

immune response in this model of primary pulmonary mycobacterial infection. Our findings suggest that compared to C57BL/6 mice, BALB/c mice are relatively susceptible to primary pulmonary mycobacterial infection. Such greater susceptibility is attributed to a poor ability of BALB/c hosts to develop an adaptive type 1 immune response at both cytokine and cellular levels in the lung and is not a result of heightened type 2 responses. Lack of type 2 immune responses in BALB/c mice during primary pulmonary mycobacterial infection contrasts with the heightened type 2 response in this strain of mice during primary pulmonary infection by other types of intracellular pathogens (1, 7, 14, 25) or systemic mycobacterial infection (9). Our observations thus suggest that differential type 1

6951

cytokine responses in humans of distinct genetic background may underlie the variable susceptibility of humans to primary aerogenic mycobacterial infections, including tuberculous infection. Of particular note are the disparate innate and adaptive immune responses in BALB/c mice. While these mice failed to mount significant adaptive immune responses, they had a greater ability to mount an early innate response characterized by increased cytokines, including IL-12, IFN-␥, and TNF-␣, and neutrophilia in the lung. Macrophages were likely among the cellular sources of these cytokines. We have recently shown that macrophages are a significant source of these cytokines, including IFN-␥, during pulmonary mycobacterial infection (22). In comparison, resistant C57BL/6 mice had little innate responses in the lung within the initial 2 weeks postinfection before the onset of a full-blown adaptive immune response around day 27. Indeed, we found that naive alveolar macrophages from BALB/c mice demonstrated a greater cytokine response to various stimuli, including mycobacteria, than those from C57BL/6 mice. In addition to macrophages, increased neutrophils in the lung of BALB/c mice in earlier stages of infection cannot be ruled out at this point as a source of cytokines. Nevertheless, the fact that the level of mycobacterial infection was similar between C57BL/6 and BALB/c mice at earlier times but was significantly higher around day 43 strongly suggests that the adaptive immune response, but not the early innate host response, is critical to host resistance to pulmonary mycobacterial infection. Likewise, we have previously shown that immunocompromised IL-12-deficient mice did not demonstrate markedly increased mycobacterial counts in their lungs until day 43 (21). Nevertheless, BALB/c mice could eventually control the infection. Apparently, the effector mechanisms responsible for the eradication of bacilli were sufficiently active by the later stages of infection to reduce bacterial burdens in BALB/c mice. This may also reflect the relatively low pathogenicity of the M. bovis BCG strain used in this study. Since C57BL/6 and BALB/c mice all carry the susceptible allele Bcg, the difference in the level of innate responses between these strains likely results from the differences in H-2

FIG. 5. Mycobacterial burden in the lung and spleen of C57BL/6 and BALB/c mice. Lungs (A) and spleens (B) were collected at days 14, 43, and 71 postinfection and subjected to colony assay. Results are representative of two independent experiments and are expressed as means ⫾ standard errors of the mean (error bars) from five mice per strain per time. The difference between C57BL/6 and BALB/c mice is significant at day 43 in the lung and spleen (P ⫽ 0.025 and P ⫽ 0.023, respectively).

6952

WAKEHAM ET AL.

INFECT. IMMUN.

FIG. 6. Antigen-stimulated cytokine recall release by splenocytes from C57BL/6 and BALB/c mice. Splenocytes were purified from the spleens collected from three mice per strain 27 days postinfection and cultured for 3 days. Supernatants were assessed for IFN-␥ (A), TNF-␣ (B), and IL-4 (C) by ELISA. Results are representative of two independent experiments and are expressed as means ⫾ standard errors of the means (error bars) from three wells per condition. The differences between C57BL/6 and BALB/c in PPD-stimulated IFN-␥ and TNF-␣ are all statistically significant. Stimulation by an irrelevant adenoviral antigen released little IFN-␥. No Ag, no antigen.

and perhaps other non-H-2 genes. Apparently, such genetic differences have also dictated a much higher level of adaptive type 1 immune responses in C57BL/6 mice. Although how exactly genetic heterogeneity determines the level of type 1 immune responses remains to be understood at this point, a recent in vitro study by Shibuya and colleagues suggests that unlike C57BL/6 mice, BALB/c hosts need additional signals for Th1-type differentiation (19). Of note, in our study, BALB/c mice mounted a much more vigorous cytokine response in the initial stage of infection, which apparently did not facilitate a Th1 differentiation. With respect to tissue immune-inflammatory responses to infection, C57BL/6 mouse lungs exhibited lymphocytic granulomas consisting of large collections of macrophages, epithelioid cells, and heavily infiltrating lymphocytes, whereas BALB/c mouse lungs had atypical granulomas marked by small collections of macrophages and a clear separation of macrophage granuloma from surrounding dense lymphocytic accumulation. Possible reasons for these differences in granuloma structure can be suggested based on the differential cytokine profile. TNF-␣ has proven to be an essential cytokine involved in granulomatous inflammation (10). Although BALB/c mice had a higher level of TNF-␣ at days 7 and 14 than C57BL/6 mice, the levels of TNF-␣ never reached maximum levels seen in the C57BL/6 mouse lungs. The downstream events of significant TNF-␣ production may include the recruitment of monocytes from peripheral blood and subsequently granuloma formation. Indeed, the level of a C-C monocyte chemotactic cytokine MCP-1 was much lower in later stages of infection in BALB/c mice. Huygen and colleagues have previously demonstrated that following intravenous infection with a high dose of M. bovis BCG, lymphocytes from C57BL/6 mice released much more IFN-␥ whereas those from BALB/c mice released much less IFN-␥ and more IL-4 (9). As a result, these infected BALB/c hosts were less protected from secondary intravenous challenge with BCG. Our findings that BALB/c mice had a much lower adaptive type 1 immune response in the lung than their C57BL/6 counterparts to pulmonary M. bovis BCG infection

lend support to a generally greater susceptibility of this strain of mice to BCG infection. The lack of Th2-type deviation of immune responses in BALB/c mice in our study, however, suggests again that the nature of immune responses in the lung during pulmonary mycobacterial infection may not exactly be the same as that seen in other tissue sites during systemic infection. The lack of Th2-like responses in the lung of BALB/c mice is unlikely a mycobacterial burden-associated phenomenon, since we also found no evidence of Th2 responses in BALB/c mice infected with a much lower dose of mycobacteria (data not shown). In further support of this notion, we have previously found that in a complete absence of Th1-differentiating cytokine IL-12 and subsequently of type 1 immune responses, mice with pulmonary mycobacte-

TABLE 2. Comparison in antigen-specific IFN-␥ recall response by lung draining lymph node cellsa Parameter

Total no. of LN cells/mouse IFN-␥ release (pg/ml) Unstimulated PPD AdAg IFN-␥ release capacity (pg/ml) a

Mouse strain C57BL/6

BALB/c 6

12.6 ⫻ 10

2.1 ⫻ 106

6.2 ⫾ 0.2 3,166.6 ⫾ 306 7.4 ⫾ 0

10.3 ⫾ 3.4 1,685.6 ⫾ 221b 18.4 ⫾ 0.1

79,783

6,740

Lymphocytes were pooled from thoracic MLN of five C57BL/6 and five BALB/c mice. One half million cells per well were cultured without or with mycobacterial antigen PPD or a control adenoviral antigen (AdAg) for 3 days, and IFN-␥ levels in supernatants were measured by ELISA. Also presented in this table are data for the IFN-␥ release capacity, which was defined as the total IFN-␥ that can be released by all MLN cells per mouse based upon the amount of IFN-␥ released by 0.5 million lymphocytes from each strain and the number of total lymph node cells per mouse. IFN-␥ release is based on 0.5 million cells per well. b The level of IFN-␥ was statistically significantly lower than that produced by C57BL/6 cells (P ⫽ 0.03).

VOL. 68, 2000

IMMUNE RESPONSES TO BCG IN C57BL/6 AND BALB/c MICE

rial BCG infection do not develop a type 2 immune response in the lung and spleen (21, 23). 13.

ACKNOWLEDGMENTS We are grateful for the technical assistance of Anna Zganiacz and Micheal Santosuosso and the provision of BCG bacilli and PPD antigens by Robin Harkness. This study was supported by funds from the Medical Research Council (MRC) of Canada and the Ontario Thoracic Society. J.W. holds an MRC-Canadian Lung Association fellowship. Z.X. holds an MRC scholarship, an Ontario Premier’s Research Excellence Award, and a Canadian Foundation for Innovation New Opportunities Award. REFERENCES 1. Bohn, E., J. J. Heeseman, S. Ehler, and I. B. Autenrieth. 1994. Early ␥-interferon mRNA expression is associated with resistance of mice against Yersinia enterocolitica. Infect. Immun. 62:3027–3032. 2. Cooper, A. M., and J. L. Flynn. 1995. The protective immune response to Mycobacterium tuberculosis. Curr. Opin. Immunol. 7:512–516. 3. Cox, J. S., B. Chen, M. NcNeil, and W. R. Jacobs, Jr. 1999. Complex lipid determines tissue-specific replication of Mycobacterium tuberculosis in mice. Nature 402:79–83. 4. Flesch, I. E. A., J. H. Hess, S. Huang, M. Aguet, J. Rothe, H. Bleuthmann, and S. H. E. Kaufmann. 1995. Early IL-12 production by macrophages in response to mycobacterial infection depends on interferon-gamma and TNF alpha. J. Exp. Med. 181:1615–1622. 5. Flynn, J. L., M. M. Goldstein, J. Chan, K. J. Triebold, K. Pfeffer, C. J. Lowenstein, R. Schreiber, T. W. Mak, and B. R. Bloom. 1995. TNF alpha is required in the protective immune response against Mycobacterium tuberculosis in mice. Immunity 2:561–572. 6. Flynn, J. L., M. M. Goldstein, K. J. Triebold, J. Sypek, S. Wold, and B. R. Bloom. 1995. IL-12 increases resistance of Balb/c mice to Mycobacterium tuberculosis infection. J. Immunol. 155:2515–2524. 7. Heinzel, F. P., M. D. Sadick, B. J. Holaday, H. L. Coffman, and R. M. Locksley. 1989. Reciprocal expression of interferon-␥ or IL-4 during the resolution or progression of murine leishmaniasis. J. Exp. Med. 169:59–72. 8. Hill, A. V. 1998. The immunogenetics of human infectious diseases. Annu. Rev. Immunol. 16:593–617. 9. Huygen, K., D. Abramowicz, P. Vandenbussche, F. Jacobs, J. De Bruyn, A. Kentos, A. Drowart, J.-P. Van Vooren, and M. Goldman. 1992. Spleen cell cytokine secretion in Mycobacterium bovis BCG-infected mice. Infect. Immun. 60:2880–2886. 10. Kindler, V., A.-P. Sappino, G. E. Grau, P.-F. Piguet, and P. Vassalli. 1989. The inducing role of tumor necrosis factor in the development of bactericidal granuloma during BCG infection. Cell 56:731–740. 11. Liles, W. C., and W. C. Van Voorhis. 1995. Review: nomenclature and biologic significance of cytokines involved in inflammation and the host immune response. J. Infect. Dis. 172:1573–1580. 12. Magram, J., S. E. Connaughton, R. R. Warrier, D. M. Carvajal, C.-Y. Wu, J.

Editor: S. H. E. Kaufmann

14. 15. 16. 17. 18. 19.

20.

21.

22. 23.

24. 25.

26.

6953

Ferrante, C. Stewart, U. Sarmiento, D. A. Faherty, and M. K. Gately. 1996. IL-12 deficient mice are defective in IFN-␥ production and type 1 cytokine responses. Immunity 4:471–481. McLeod, R., E. Buschman, L. D. Arbuckle, and E. Skamene. 1995. Immunogenetics in the analysis of resistance to intracellular pathogens. Curr. Opin. Immunol. 7:539–552. Miralles, G. D., M. Y. Stoeckel, D. F. McDernott, F. D. Finkelman, and H. W. Murray. 1994. Th1 and Th2 cell-associated cytokines in experimental visceral leishmaniasis. Infect. Immun. 62:1058–1063. North, R. J. 1995. Mycobacterium tuberculosis is strikingly more virulent for mice when given via the respiratory than via the intravenous route. J. Infect. Dis. 172:1550–1553. Orme, I. M., P. Andersen, and W. H. Boom. 1993. T cell response to Mycobacterium tuberculosis. J. Infect. Dis. 167:1481–1497. Powrie, F., S. Menon, and R. L. Coffman. 1993. Interleukin-4 and interleukin-10 synergize to inhibit cell-mediated immunity in vivo. Eur. J. Immunol. 23:2223–2229. Schluger, N. W., and W. N. Rom. 1998. The host immune response to tuberculosis. Am. J. Respir. Crit. Care Med. 157:679–691. Shibuya, K., D. Robinson, F. Zonin, S. B. Hartley, S. E. Macatonia, C. Somoza, C. A. Hunter, K. M. Murphy, and A. O’Garra. 1998. IL-1␣ and TNF␣ are required for IL-12-induced development of Th1 cells producing high levels of IFN-␥ in BALB/c but not C57BL/6 mice. J. Immunol. 160: 1708–1716. Thompson-Snipes, L., E. Skamene, and D. Radzioch. 1998. Acquired resistance but not innate resistance to Mycobacterium bovis bacillus CalmetteGue´rin is compromised by interleukin-12 ablation. Infect. Immun. 66:5268– 5274. Wakeham, J., J. Wang, J. Magram, K. Croitoru, R. Harkness, P. Dunn, A. Zganiacz, and Z. Xing. 1998. Lack of both types 1 and 2 cytokines, tissue inflammatory responses, and immune protection during pulmonary infection by Mycobacterium bovis bacille Calmette-Gue´rin in IL-12 deficient mice. J. Immunol. 160:6101–6111. Wang, J., J. Wakeham, R. Harkness, and Z. Xing. 1999. Macrophages are a significant source of type 1 cytokines during mycobacterial infection. J. Clin. Investig. 103:1023–1029. Xing, Z., A. Zganiacz, J. Wang, M. Divangahi, and F. Nawaz. 2000. IL-12independent Th1-type immune responses to respiratory viral infection: requirement of IL-18 for IFN-␥ release in the lung but not for the differentiation of viral-reactive Th1-type lymphocytes. J. Immunol. 164:2575–2584. Xing, Z., J. Wang, K. Croitoru, and J. Wakeham. 1998. Protection by CD4 or CD8 T cells against pulmonary Mycobacterium bovis bacillus CalmetteGue´rin infection. Infect. Immun. 66:5537–5542. Yang, X., K. T. HayGlass, and R. C. Brunham. 1996. Genetically determined differences in IL-10 and IFN-gamma responses correlate with clearance of Chlamydia trachomatis mouse pneumonitis infection. J. Immunol. 156:4338– 4344. Yoshida, A., Y. Koide, M. Uchijima, and T. O. Yoshida. 1995. Dissection of strain difference in acquired protective immunity against Mycobacterium bovis Calmette-Gue´rin bacillus (BCG). Macrophages regulate the susceptibility through cytokine network and the induction of nitric oxide synthase. J. Immunol. 155:2057–2066.