An Anti-Inflammatory Role for Vα14 NK T cells in Mycobacterium bovis Bacillus Calmette-Guérin-Infected Mice

Francesco Dieli, Masaru Taniguchi, Mitchell Kronenberg, Stephane Sidobre, Juraj Ivanyi, Lanfranco Fattorini, Elisabetta Iona, Graziella Orefici, Giacomo De Leo, Domenica Russo, Nadia Caccamo, Guido Sireci, Caterina Di Sano and Alfredo Salerno
J Immunol August 15, 2003, 171 (4) 1961-1968; DOI: https://doi.org/10.4049/jimmunol.171.4.1961

Abstract

The possible contribution of NKT cells to resistance to Mycobacterium tuberculosis infection remains unclear. In this paper we characterized the Vα14 NKT cell population following infection with Mycobacterium bovis bacillus Calmette-Guérin (BCG). BCG infection determined an early expansion of Vα14 NKT cells in liver, lungs, and spleen, which peaked on day 8 and was sustained until day 30. However, an NK1.1+ Vα14 NKT population preferentially producing IFN-γ predominated at an early stage (day 8), which was substituted by an NK1.1− population preferentially producing IL-4 at later stages (day 30). Despite the fact that Vα14 NKT cell-deficient mice eliminated BCG as did control mice, they had significantly higher numbers of granulomas in liver and lungs. Additionally, while control mice developed organized small granulomas, those in Vα14 NKT-deficient mice had signs of caseation, large cellular infiltrates, and some multinucleated macrophages, suggesting that Vα14 NKT cells may actually work as anti-inflammatory cells by limiting excessive lymphocyte influx and tissue pathology. In agreement, we found an increased spontaneous production and mRNA expression of TNF-α in liver and lungs of Vα14 NKT-deficient mice, whose neutralization in vivo by anti-TNF-α mAbs consistently reduced the number of granulomas in liver and lungs. Together, our results support a regulatory role for Vα14 NKT cells in the course of BCG infection through their ability to limit the extent of inflammatory response and point to an important role for this cell subset as a regulator of the balance between protective responses and immunopathology.

Protective immunity against mycobacterial infections is mediated by interactions between specific T cells and activated macrophage effector cells (1). These populations interact with each other by means of a complex network of chemokines, cytokines, and their receptors. The exact role and relative importance of each of these mediators, however, are not completely understood and are even somewhat controversial, although IFN-γ is thought to play a crucial role (2, 3, 4).

Mycobacterium bovis bacillus Calmette-Guérin (BCG),3 widely used as a human vaccine against tuberculosis, gives a poor or ambiguous protection to infection by Mycobacterium tuberculosis. The murine αβ and γδ T cell responses induced by BCG have been well defined (1, 5), but there is less knowledge about the response of smaller T cell subsets.

NK T (NKT) cells are a heterogeneous population of T lymphocytes that express the NK1.1 molecule and comprise at least four different categories of T cells (6, 7). Within NKT cells, the most abundant population expresses an invariant TCR α-chain encoded by the Vα14 and Jα281 gene segments and is usually double (i.e., CD4 and CD8) negative, although Vα14 NKT cells residing in the liver coexpress CD4 (8). Stimulation of Vα14 NKT cells by anti-CD3 mAbs or the synthetic ligand α-galactosylceramide (α-GalCer) in the context of CD1d promptly induces both IFN-γ and IL-4 production (9, 10, 11).

Vα14 NKT cells in the liver of BCG-infected mice rapidly produce IFN-γ (12) and play an important role in granuloma formation in response to phosphatidylinositol mannosides (PIMs) deproteinized cell wall from M. tuberculosis (13). However, this does not seem to involve direct recognition of PIMs by the Vα14 TCR (14). Furthermore, studies in CD1d- and Jα281-deficient mice, which selectively lack Vα14 NKT cells, failed to show any role for protection against M. tuberculosis or BCG (15, 16, 17, 18, 19), although a recent study has reported that specific activation of NKT cells by α-GalCer protects mice from tuberculosis (20).

In view of the lack of knowledge about the significance of the several possible actions of Vα14 NKT cells, we examined their functions during BCG infection using Vα14 NKT-deficient mice.

Materials and Methods

Mice

BALB/c and C57BL/6 mice were purchased from OLAC through Nossan (Correzzana, Italy). Generation of Jα281-deficient mice has been previously described (21). Mice that lack the Jα281 gene segment are devoid of Vα14 NKT cells, but the other lymphoid cell lineages are intact (21). Populations that are more heterogeneous at the genetic level were established by backcrossing heterozygotes to C57BL/6 or BALB/c mice for more than five generations. The resultant heterozygous mice were mated to obtain homozygotes (18). Mice were fed and kept under specific pathogen-free conditions and were used at 8–12 wk of age. In each experiment age- and sex-matched mice were used.

Bacteria and infection

BCG (strain Pasteur) was grown in Middlebrook 7H9 broth base (Difco, Detroit, MI) supplemented with 10% Bacto Middlebrook OADC enrichment (Difco) for 2 wk at 37°C, and aliquots were frozen at −70°C until used. The final concentration of viable bacteria was enumerated by plate counts of CFU with Middlebrook 7H11 agar (Difco) supplemented with 0.5% glycerol (Difco) and 10% Bacto Middlebrook OADC enrichment (Difco). Mice were infected i.v. with 5 × 106 viable bacteria.

Preparation of mononuclear cells and flow cytometry

Single-cell suspensions were prepared from liver, lungs, and spleen according to previously published methods (22, 23). Livers were minced, passed through a mesh, and washed in complete RPMI 1640 medium (Life Technologies, Grand Island, NY). Cells were resuspended in 30% Percoll (Amersham Bioscience Europe, Milan, Italy) containing 100 U/ml heparin and were centrifuged at 2000 rpm for 20 min at room temperature. The cell pellet was washed once in medium and then resuspended in cold distilled water for 20 s to lyse RBC by osmotic shock. After lysis, the cell suspension was diluted with NaCl solution and washed with complete RPMI 1640 medium containing 5% heat-inactivated FCS (Life Technologies). Lungs were removed and digested in the presence of collagenase (200 U/ml; Sigma-Aldrich, St. Louis, MO), and mononuclear cell suspensions were obtained through Lympholyte M (Cedarlane Laboratories, Ontario, Canada) gradient centrifugation. For the spleen, mononuclear cells were obtained by separation of total spleen cells on a Lympholyte M gradient. The viability of all mononuclear cell preparations, as determined by Trypan Blue exclusion, was >90%. Mononuclear cells (5 × 105) obtained from various organs were incubated for 10 min on ice with PBS containing 5% BSA and 0.02% NaN3. After washing, cells were incubated with FITC-conjugated anti-TCRαβ mAb (H57-597; Sigma), PerCP-Cy5.5-conjugated anti NK1.1 mAb (PK136; BD PharMingen, San Diego, CA), and PE-conjugated CD1d/α-GalCer tetramers (24) or PE-conjugated unloaded CD1d as a control for the specificity of tetramer staining (24). The specificity of tetramer staining was further verified by examining mononuclear cells from Vα14 NKT-deficient mice. Viable lymphocytes were gated by forward and side scatter, and analysis was performed on 100,000 acquired events for each sample. Data were acquired on a FACSCalibur instrument (BD Biosciences, Mountain View, CA) and were analyzed using CellQuest software (BD Immunocytometry Systems, San Jose, CA).

Immunomagnetic cell sorting

Enrichment for CD1d/α-GalCer tetramer-positive cells was performed using anti-PE beads (Miltenyi, Bologna, Italy) as previously described (24). In brief, nylon wool-enriched T cells were incubated with the tetramer for 15 min at 23°C, washed, and incubated with the appropriate amount of anti-PE microbeads. Magnetic separation was performed according to the manufacturer’s instructions. The bead-nonadherent population contained <2% tetramer-positive cells, while the adherent population contained >90% positive cells, as determined by FACS staining with PE-labeled-GalCer/CD1d tetramer and FITC-labeled anti-TCRαβ mAbs (24). The viability of the cell population exceeded 90% as determined by trypan blue exclusion.

Cytokine production by ELISA

Immunomagnetically enriched, CD1d/αGalCer tetramer-positive T cells (105) were cultured in a 96-well, flat-bottomed plate (Nunc, Copenhagen, Denmark) that had been precoated overnight with 10 μg/ml of anti-TCRαβ mAb H57-597. After 48 h of culture, supernatants were harvested, and IL-4 and IFN-γ concentrations were determined by a two-mAb sandwich ELISA (BD PharMingen). The lower limit of detection for each cytokine was 15 pg/ml. SD values were always <10% of the mean values. To assess spontaneous cytokine production, mononuclear cells obtained from liver and lungs as above described were cultured at 106/ml in complete medium for 48 h at 37°C. Supernatants were then harvested, and IFN-γ and TNF-α levels were assessed by ELISA (BD PharMingen).

Cytokine mRNA analysis by RT-PCR

Total RNA was extracted using the guanidinium thiocyanate/cesium chloride gradient centrifugation method. cDNA was synthesized with oligo(dT) (Amersham Pharmacia Biotech, Uppsala, Sweden) with reverse transcriptase using 10 μg of RNA according to manufacturer’s instructions. PCR was performed using a GeneAmp PCR system 9600 (PerkinElmer, Rome, Italy), using the following oligonucleotide primers: TNF-α sense, 5′-TCTCATCAGTTCTATGGCCC-3′; TNF-α antisense, 5′-GGGAGTAGACAAGGTACAAC-3′; IFN-γ sense, 5′-GCTCTGAGACAATGAACGCT-3′; IFN-γ antisense, 5′-AAAGAGATAATCTGGCTCTGC-3′; β2-microglobulin sense, 5′-TGACCGGCTTGTATGCTATC-3′; and β2-microglobulinantisense, 5′-CAGTGTGAGCCAGGATATAG-3′.

Each cycle (21) consisted of incubation at 92°C for 45 s, followed by 55°C for 30 s and 72°C for 30 s. Before the first cycle, a 2-min 94°C denaturation step was included, and after the 30th cycle the extension at 72°C was prolonged for 4 min. Aliquots (20 μl) of PCR products were electrophoresed in 2% agarose and visualized using ethidium bromide staining as described. All gels were photographed similarly.

IL-4 detection by intracellular staining

Intracellular staining was used to determine IL-4 production at the single-cell level. Briefly, immunomagnetically enriched-CD1d/αGalCer tetramer-positive T cells were stimulated with anti-TCRαβ or isotype-matched control mAb for 8 h at 37°C and were cultured for 5 h with brefeldin A (Sigma) to accumulate intracellular newly synthesized protein. Cells were harvested and fixed with 4% (w/v) paraformaldehyde in PBS for 10 min at room temperature. Fixed cells were suspended and washed twice with permeabilization buffer containing 0.1% saponin (Sigma), 1% heat-inactivated FCS, and 0.1% NaN3 in PBS. The permeabilized cells were then incubated in the presence of saponin with FITC-conjugated anti-mouse IL-4 mAb (11B11, rat IgG1; BD PharMingen) or an FITC-conjugated isotype control mAb (R3-34, rat IgG1; BD PharMingen) for 30 min at room temperature. After being washed at room temperature, the cells were analyzed by FACS as above described.

To identify the phenotype of the IL-4-producing cells, surface marker analysis was performed by staining the cells with PerCP-Cy5.5-conjugated anti-NK1.1 mAb.

RT-PCR analysis of TCR Vα14 expression

Mononuclear cells from different organs were obtained 30 days after BCG infection, stained with PE-labeled GalCer/CD1d tetramer, PerCP-Cy5.5-conjugated anti-NK1.1 mAb, and FITC-labeled anti-TCRαβ mAbs, and purified by cell sorting with a FACSVantage flow cytometer (BD Biosciences). The CD1d/α-GalCer tetramer-positive TCRαβ+ NK1.1− population was harvested, RNA was extracted using the guanidinium thiocyanate/cesium chloride gradient centrifugation method, and cDNA was synthesized as described above. PCR was performed with a GeneAmp PCR system 9600 (PerkinElmer) using Vα14 (5′-CTAAGCACAGCACGCTGCACA-3′) and Cα (5′-GAAGCTTGTCTGGTTGCTCCA-3′) oligonucleotide primers. β2-Microglobulin sense and antisense primers were used as controls. Each cycle (25) consisted of incubation at 94°C for 60 s, followed by 54°C for 60 s and 72°C for 30 s. Before the first cycle, a 2-min 94°C denaturation step was included. After 30 cycles, 20-μl aliquots of each PCR product were electrophoresed in 2% agarose and visualized using ethidium bromide staining as described above. All gels were photographed similarly.

Determination of CFUs in organs of mice

Mice were infected i.v. with BCG and were killed by cervical dislocation 1, 15, 30, or 45 days after infection. CFU counts were determined by plating serially diluted homogenates on Middlebrook 7H10 agar plates supplemented with 0.5% glycerol and 10% Bacto Middlebrook OADC enrichment. CFUs were determined after 4 wk of incubation at 37°C.

Histology

BCG-infected mice were killed by cervical dislocation 40 days after infection. The livers and lungs were removed immediately, fixed with 10% formalin for 12 h, and embedded in paraffin. Several tissue sections were deparaffinized and stained with H&E. In some experiments mice were injected i.v. with 0.25 mg of neutralizing anti-TNF-α mAb (MP6-XT3, rat, IgG1; BD PharMingen) or rat IgG1 as a control on the same day of BCG infection and again every week.

Statistics

Student’s t tests were used to compare the significance of differences between groups.

Results

Phenotypic and functional properties of Vα14 NKT cells during BCG infection

In the first set of experiments we investigated functional modifications of Vα14 NKT cells during i.v. infection with BCG. To assess this, livers, lungs, and spleens of C57BL/6 mice were harvested 8 and 30 days after BCG infection, mononuclear cells were separated, and the relative percentages and absolute numbers of Vα14 NKT cells were calculated by FACS staining with CD1d/α-GalCer tetramers and anti-TCRαβ mAb. Despite a decrease in the percentages of Vα14 NKT cells in liver, lungs, and spleen after BCG infection (Fig. 1⇓a), their absolute number consistently increased, with a peak 8 days after infection. Although a very low number of Vα14 NKT cells was detected in the lungs of control uninfected mice (0.3 ± 0.1 × 105), Vα14 NKT cell number had increased ∼6-fold (1.7 ± 0.4 × 105) 8 days after infection. The number of Vα14 NKT cells then declined, although values remained higher than controls (Fig. 1⇓b).

FIGURE 1.
FIGURE 1.

Expansion of NKT cells in organs of mice infected with BCG. C57BL/6 mice were infected i.v. with BCG, and liver, lungs, and spleen mononuclear cells were obtained 8 and 30 days later. a, The percentages of CD1d/α-GalCer tetramer-positive cells in the lymphocyte gate was determined by two-color FACS analysis. The percentage of tetramer-positive cells is shown in the upper right corner. b, Absolute numbers of NKT cells within mononuclear cells in liver (▵), spleen (□), and lungs (○) were calculated based on the percentages of NKT cells. Day 0 refers to mice that had not been infected with BCG. Data are representative of three different experiments, each consisting of five mice per group.

Changes in the percentage and absolute numbers of Vα14 NKT cells during BCG infection were paralleled by modifications in their phenotype and functional properties. As shown in Fig. 2⇓a, expression of NK1.1 by Vα14 NKT cells was strongly modified during BCG infection. While the vast majority of Vα14 NKT cells in control mice are NK1.1+, only 25% of Vα14 NKT cells tested at 30 days after BCG infection expressed NK1.1. To confirm that the NK1.1− cells appearing in the late phase of infection were indeed Vα14 NKT cells, the tetramer-positive TCRαβ+ NK1.1− population was sorted, and Vα14 expression was assessed by RT-PCR. As shown in Fig. 2⇓b, expression of the Vα14 gene was detected in the CD1d/α-GalCer tetramer-positive TCRαβ+ NK1.1− population, appearing 30 days after BCG infection, indicating that these cells are Vα14 NKT cells. Additionally, a modification in the pattern of cytokine production was also found (Fig. 2⇓c). Upon in vitro stimulation with immobilized anti-TCRαβ mAb, enriched Vα14 NKT cells from control mice produced much more IFN-γ than IL-4, and Vα14 NKT cells from mice infected 8 days earlier with BCG showed increased production of IFN-γ. Conversely, Vα14 NKT cells from mice that had been infected with BCG 30 days earlier showed increased production of IL-4, which was particularly evident for Vα14 NKT cells taken from liver and lungs, and they produced as much IFN-γ as that produced by Vα14 NKT cells from control mice.

FIGURE 2.
FIGURE 2.

Phenotype and cytokine production by NKT cells in organs of mice infected with BCG. C57BL/6 mice were infected i.v. with BCG, and liver, lungs, and spleen mononuclear cells were obtained 8 and 30 days later. a, NK1.1 expression by NKT cells was determined by triple FACS staining and was calculated on the gated TCRαβ+ CD1d/α-GalCer tetramer-positive population. ▪, The percentage of NKT cells expressing NK1.1; □, the percentage of NKT cells that are NK1.1−. b, Vα14 mRNA expression on the sorted TCRαβ+ CD1d/α-GalCer tetramer-positive NK1.1− population. TCRαβ+ CD1d/α-GalCer tetramer-positive NK1.1− cells were sorted from the liver (line 1), spleen (line 2), and lungs (line 3) of mice that had been infected 30 days earlier with BCG. Vα14 mRNA expression was determined by PCR analysis as described in Materials and Methods. c, CD1d/α-GalCer tetramer-positive NKT cells were sorted as described in Materials and Methods and stimulated with immobilized anti-TCRαβ mAb. Supernatants were collected, and IFN-γ (▪) and IL-4 (□) contents were determined by ELISA. d, CD1d/α-GalCer tetramer-positive NKT cells were sorted as described in Materials and Methods and stimulated with immobilized anti-TCRαβ mAb (upper panels) or with an isotype-matched control mAb (lower panels). Cells were then stained for intracellular IL-4 and surface NK1.1 expression. Similar results were obtained in two additional experiments with five mice per group.

However, because the late-appearing Vα14 NKT cell population consisted of both NK1.1+ and NK1.1− cells, we used intracellular FACS staining to demonstrate that the latter were producing IL-4. Fig. 2⇑d shows that within CD1d/α-GalCer tetramer-positive cells obtained from liver, spleen, and lungs 30 days after BCG infection, the vast majority of NK1.1− cells stained positively for IL-4 after 8-h stimulation with anti-TCRαβ mAb in vitro, while virtually no staining for this cytokine was detected within the NK1.1+ population.

Together, these results clearly indicate that infection with BCG causes an early (8 day) expansion of NK1.1+ Vα14 NKT cells with a predominant production of IFN-γ, while at later stages of infection (day 30) Vα14 NKT cell expansion involves an NK1.1− population that produces IL-4.

BCG CFU in Vα14 NKT-deficient and control mice

To determine the impact of Vα14 NKT cells on the resistance of mice to BCG infection, we compared the viable bacterial counts in livers, lungs, and spleens between Vα14 NKT-deficient and wild-type C57BL/6 mice. As shown in Fig. 3⇓, the trend of BCG infection was similar in the lungs, but slightly different in the liver and spleen. However, no statistically significant differences in CFU counts were apparent in any organ between the Vα14 NKT-deficient and wild-type mice. These results suggest that Vα14 NKT cells do not play any role in the outcome of BCG infection in C57BL/6 mice.

FIGURE 3.
FIGURE 3.

Growth of BCG in organs of NKT-deficient and control mice. NKT-deficient or control C57BL/6 mice were infected i.v. with BCG. Organs were harvested at the indicated time points, and the numbers of CFU were assessed. Results are given as the mean ± SD of five mice per group of one experiment repeated twice. Time zero refers to organs taken 24 h after BCG infection. ▪, Control mice; □, NKT-deficient mice.

Granuloma formation in Vα14 NKT-deficient and control mice infected with BCG

We examined the development of granulomas in livers and lungs of BCG-infected, NKT-deficient and control mice, since their presence at the site of microbial growth is considered essential for local containment and elimination of pathogenic mycobacteria. The numbers of granulomas detected in liver and lungs of Vα14 NKT-deficient mice were significantly higher than those in control mice (Fig. 4⇓). Additionally, while in livers (Fig. 5⇓) of control mice, organized small granulomas were observed on day 40 after BCG infection, granulomas developing in Vα14 NKT-deficient mice contained a necrotic center with signs of caseation, large cellular infiltrates surrounding granulomas, and some multinucleated macrophages. As shown in Figs. 4⇓ and 5⇓, similar data were obtained in Vα14 NKT-deficient C57BL/6 and BALB/c mice, although control BALB/c mice developed granulomas with perigranulomatous infiltration and some necrotic centers, which were not seen in control C57BL/6 mice. Similar differences between Vα14 NKT-deficient and control mice were obtained in granulomas developing in the lungs upon BCG infection (data not shown). Thus, our results clearly indicate that the absence of Vα14 NKT cells influences the number, size, and composition of granulomas developing in liver and lungs upon BCG infection.

FIGURE 4.
FIGURE 4.

Enumeration of granulomas in the livers and lungs of NKT-deficient and control mice. NKT-deficient or control C57BL/6 and BALB/c mice were infected with BCG; livers and lungs were collected 40 days later and stained with H&E. Results represent the mean granuloma number from a total of 15 microscopic fields from three different mice per group. Similar results were obtained in three additional experiments. ▪, Control mice; □, NKT-deficient mice.

FIGURE 5.
FIGURE 5.

Histopathological findings of the livers 40 days after infection with BCG (see Fig. 3⇑). Original magnification, ×400. The arrow indicates the presence of a multinucleated giant cell (MN), which is better depicted in the small box.

Increased spontaneous IFN-γ and TNF-α production in tissues of BCG-infected Vα14 NKT-deficient mice

IFN-γ and TNF-α are two important cytokines for granuloma formation. We therefore assessed the production of both cytokines in liver and lungs of mice infected with BCG. To this end, liver and lungs were removed 40 days after BCG infection, and mononuclear cell suspensions were prepared and cultured in the absence of any stimulation. Supernatants were then collected, and IFN-γ and TNF-α levels were measured by ELISA. Additionally, IFN-γ and TNF-α mRNA levels were determined by semiquantitative PCR analysis.

Fig. 6⇓a shows that mononuclear cells obtained from liver and lungs of BCG-infected Vα14 NKT-deficient mice spontaneously produced ∼3-fold more TNF-α than BCG-infected control mice, with TNF-α levels being higher in lungs than in liver. Although IFN-γ levels were also increased in the liver and lungs of BCG-infected Vα14 NKT-deficient mice, differences from the values detected in control mice were not statistically significant.

FIGURE 6.
FIGURE 6.

IFN-γ and TNF-α production in tissues of BCG-infected NKT-deficient mice. NKT-deficient and control C57BL/6 mice were infected i.v. with BCG, and liver and lungs were removed 40 days later. a, Mononuclear cell suspensions were prepared and cultured in the absence of any stimulation. Supernatants were then collected, and IFN-γ and TNF-α levels were measured by ELISA. ▪, Control mice; □, NKT-deficient mice. b, IFN-γ and TNF-α mRNA levels were also determined by semiquantitative PCR analysis as described in Materials and Methods. Lines 1 and 3, Livers and lungs from control mice, respectively; lines 2 and 4, livers and lungs from NKT-deficient mice, respectively. Similar results were obtained in two additional experiments with five mice per group.

ELISA data were confirmed by IFN-γ and TNF-α mRNA analysis by semiquantitative PCR analysis (Fig. 6⇑b), showing increased expression of TNF-α mRNA in liver and lungs of BCG-infected Vα14 NKT-deficient mice.

To prove that the exaggerated production of TNF-α in NKT-deficient mice was causative of granuloma formation, we treated NKT-deficient and control C57BL/6 mice with neutralizing Abs against TNF-α and measured the number of granulomas in liver and lungs 40 days after BCG infection. Fig. 7⇓ shows that treatment with anti-TNF-α mAb significantly reduced the number of granulomas in liver and lungs of both NKT-deficient and control mice, strongly indicating that this cytokine was responsible for granuloma formation in NKT-deficient and control mice.

FIGURE 7.
FIGURE 7.

Effect of anti-TNF-α neutralizing mAb on granuloma development in tissues of BCG-infected mice. NKT-deficient or control C57BL/6 mice were infected with BCG and treated with anti-TNF-α neutralizing or isotype-matched control mAbs, as indicated in Materials and Methods. Livers and lungs were collected 40 days later and stained with H&E. Results represent mean granuloma number from a total of 15 microscopic fields from seven different mice per group. Similar results were obtained in two additional experiments with five mice per group. ▪, Control mice; □, NKT-deficient mice.

Together, these results clearly indicate that the absence of Vα14 NKT cells induces a stronger inflammatory response in the liver and lungs upon BCG infection, as demonstrated by increased granuloma formation and TNF-α production.

Discussion

In this paper we have studied the role played by Vα14 NKT cells during BCG infection in mice. The results reported here clearly show that the absolute number of Vα14 NKT cells increased in liver, lungs, and spleen and reached a peak 8 days after infection, which was sustained until 30 days after infection. Although a very low number and percentage of Vα14 NKT cells were detected in the lungs of uninfected mice, kinetics similar to those observed in liver and lungs were detected after BCG infection.

Several studies have reported similar early expansion of Vα14 NKT cells in different experimental infection models, such as Leishmania major (26), Plasmodium yoelii (27), Cryptococcus neoformans (28), and Toxoplasma gondii (29). Additionally, early Vα14 NKT cell recruitment at sites of inflammation was caused by injection of glycolipid extracts from M. tuberculosis (13) and Yersinia pseudotuberculosis (30). These observations suggested a regulatory role for Vα14 NKT cells in addition to other subsets of NKT cells in the early host defense against infection, although their contributions might be distinct among infectious pathogens.

An increase in the number of Vα14 NKT cells in the lungs, liver, and spleen of BCG-infected mice might be due to the expansion of resident Vα14 NKΤ cells, the recruitment of Vα14 NKΤ cells from other anatomical locations, or both. IL-15 is known to act as a major growth factor for Vα14 NKT cells, because mice deficient in IL-15Rα or IL-2/IL-15Rβ lack such cells (31). In the present study we did not examine the expression of this cytokine at the site of infection and its contribution to the increase in Vα14 NKT cells. Thus, the first mechanism remains open for further study. In contrast, migration of Vα14 NKT cells into the lungs is a plausible mechanism, as revealed by the pivotal role played by MCP-1 in the recruitment of NK cells to the lungs infected with C. neoformans (28). In addition, macrophage inflammatory protein-2 has been recently demonstrated as a chemoattractant for these cells (32). Finally, it is possible that other chemokines, including monocyte chemoattractant proteins-2 and -3, macrophage inflammatory protein-1α, RANTES, inducing protein-10, and lymphotactin, which are important in the trafficking of NK cells (33), may also act on Vα14 NKT cells. Accordingly, in humans most Vα14 NKT cells express receptors for inflammatory chemokines, such as CCR2, CCR5, and CXCR3 (34).

A major difference between our findings and the results discussed above is that in most infection models, early Vα14 NKT cell expansion declines sharply, usually within 10 days after infection, while after BCG infection, the expansion of Vα14 NKT cells in different organs is sustained, and Vα14 NKT cell numbers higher than those in control mice are detectable 30 days after infection. Moreover, a dramatic change in both the phenotype and the pattern of cytokine production is observed at later stages after BCG infection. As shown in Fig. 2⇑, the vast majority of Vα14 NKT cells at 30 days after infection are NK1.1− and preferentially produce IL-4, while those at early points after infection (day 8) express NK1.1 and preferentially produce IFN-γ. It is possible that the 30-day expanding Vα14 NKT population has lost NK1.1 expression because of activation (24). Alternatively, BCG could induce the expansion of an NK1.1− population of Vα14 NKT cells, which has been shown to preferentially produce IL-4 (35). Finally, the NK1.1− Vα14 NKT population expanding 30 days after BCG infection might represent recent thymic emigrants, which are known to acquire NK1.1 expression once recruited in tissues and are biased to IL-4 production (36).

Despite the lack of Vα14 NKT cells, mice were as resistant to BCG as their control littermates, thus confirming previous studies of primary M. tuberculosis infection in CD1d-deficient mice (15). However, the numbers of granulomas detected in the liver and lungs of Vα14 NKT-deficient mice were significantly higher than those in control mice (Fig. 4⇑). In addition, while in control mice, organized small granulomas were observed on day 40 after BCG infection, granulomas developing in Vα14 NKT-deficient mice contained a necrotic center with signs of caseation, large cellular infiltrates surrounding granulomas, and some multinucleated macrophages. Similar findings in Vα14 NKT-deficient C57BL/6 and BALB/c mice exclude the possibility that the reported differences might be due to background genes.

We interpret these findings as indicating that an exaggerated inflammatory response develops in the organs of Vα14 NKT-deficient mice, and consequently, we suggest the possibility that Vα14 NKT cells may actually work as anti-inflammatory cells by limiting excessive lymphocyte influx and tissue pathology. In agreement with this possibility, we found increased spontaneous production and mRNA expression of TNF-α in liver and lungs of Vα14 NKT-deficient mice, while the production of IFN-γ was not substantially different from that in control mice. Moreover, treatment of NKT-deficient mice with a neutralizing mAb against TNF-α consistently reduced the number of granulomas in liver and lungs, strongly indicating that this cytokine is the cause of the observed exaggerated inflammatory response.

The protective role of TNF-α in mycobacterial infections has been well established in experimental models (37, 38). The cytokine can exert a number of effects. TNF-α facilitates extravasation of monocytes and T cells from the blood and directs their migration to the infected site (39). TNF-α is necessary for the accumulation and organization of monocytes, macrophages, and lymphocytes into well-differentiated granulomas (37, 38, 39). In addition, TNF-α activates T cells and fosters differentiation of dendritic cells for enhanced IL-12 production and Ag presentation (40). Finally, TNF-α may also directly activate macrophages to control the growth of and/or kill the intracellular mycobacteria (40).

In addition to these protective effects, TNF-α has detrimental effects (41). With regard to our results, the extents of inflammation and cellular recruitment to a site of mycobacterial infection are determined by the level of TNF-α, with a high dose of this cytokine determining a more severe inflammatory response. Particularly in the presence of low amounts of TNF-α, the granulomas remain small and clearly distinct from the rest of the lung. These granulomas consist of lymphocytes and macrophages. In contrast, in the presence of high amounts of TNF-α, the granulomas enlarge extensively and contain large macrophages and many lymphocytes (41).

Thus, TNF-α appears to regulate the granulomatous response, resulting in smaller and better differentiated granulomas at low doses, while high levels of TNF-α lead to an overwhelming inflammatory response despite successful control of the growth of infecting mycobacteria.

In a recent paper (20) treatment of mice with the Vα14 NKT ligand αGalCer was shown to improve protection from M. tuberculosis. Interestingly and in agreement with the results reported here, the inflammatory lesions in the lungs of αGalCer-injected mice contained larger macrophage infiltrates with greater numbers of lymphocytes, without any significant modification of IFN-γ production.

What is the reason for the increased inflammatory response in the absence of Vα14 NKT cells? One possibility, and the one we favor, is that Vα14 NKT cells accumulating at sites of infection with BCG at later stages produce IL-4, which, in turn, may limit TNF-α production and the consequent cell influx. Thus, in the absence of Vα14 NKT cells, TNF-α production and cell influx at sites of infection are consistently increased. Although we did not measure IL-10 production by NKT cells, we cannot actually exclude a possible role played by NKT-derived IL-10. Therefore, further study is needed to define the molecular mechanisms of the anti-inflammatory activity of NKT cells during BCG infection.

Vα14 NKT cells have been implicated in the suppression of immune responses directed toward autoantigens (42), foreign Ags introduced in immune-privileged sites (43) and tumor Ags (44). Moreover, specific stimulation of Vα14 NKT cells with α-GalCer suppresses Th1-mediated autoimmune diseases such as type 1 diabetes (45, 46) and experimental autoimmune encephalomyelitis (47) in mice. Our results also support a regulatory role for Vα14 NKT cells in the course of BCG infection through their ability to limit the extent of the inflammatory response. Therefore, although our data tend to exclude an involvement of Vα14 NKT cells in the elimination of mycobacteria, they rather point to an important role for this cell subset as a regulator of the balance between protective responses and immunopathology.

Acknowledgments

We thank Dr. Joan Pere Cardona for helpful advice with the histology analysis, and Dr. Steven Porcelli for critically revising the manuscript.

Footnotes

  • ↵1 This work was supported by grants from the Italian National Research Council (to F.D.), the Ministry for Education, University and Research (MIUR 60%, to A.S. and F.D.), the European Commission (Fifth Framework Program, Contract QLK2-1999-00367), and National Institutes of Health Grant RO1KA52511 (to M.K.).

  • ↵2 Address correspondence and reprint requests to Dr. Francesco Dieli, Dipartimento di Biopatologia, Università di Palermo, Corso Tukory 211, I-90134 Palermo, Italy. E-mail address: dieli{at}unipa.it

  • ↵3 Abbreviations used in this paper: BCG, Mycobacterium bovis bacillus Calmette-Guérin; α-GalCer, α-galactosylceramide; PIM, phosphatidylinositol mannoside.

  • Received January 6, 2003.
  • Accepted June 5, 2003.

References

  1. ↵
    Flynn, J. L., J. Chan. 2001. Immunology of tuberculosis. Annu. Rev. Immunol. 19:93.
    OpenUrlCrossRefPubMed
  2. ↵
    Cooper, A. M., D. K. Dalton, T. A. Stewart, J. P. Griffin, D. G. Russell, I. M. Orme. 1993. Disseminated tuberculosis in interferon γ gene-disrupted mice. J. Exp. Med. 178:2243.
    OpenUrlAbstract/FREE Full Text
  3. ↵
    Flynn, J. L., J. Chan, K. J. Triebold, D. K. Dalton, T. A. Stewart, B. R. Bloom. 1993. An essential role for interferon γ in resistance to Mycobacterium tuberculosis infection. J. Exp. Med. 178:2249.
    OpenUrlAbstract/FREE Full Text
  4. ↵
    Jouanguy, E., F. Altare, S. Lamhamedi, P. Revy, J. F. Emile, M. Newport, M. Levin, S. Blanche, E. Seboun, A. Fischer, et al 1996. Interferon-γ-receptor deficiency in an infant with fatal bacille Calmette-Guérin infection. N. Engl. J. Med. 335:1956.
    OpenUrlCrossRefPubMed
  5. ↵
    Ladel, C. H., J. Hess, S. Daugelat, P. Mombaerts, S. Tonegawa, S. H. R. Kaufmann. 1995. Contribution of αβ and γδ T lymphocytes to immunity against Mycobacterium bovis bacillus Calmette Guérin: studies with T cell receptor-deficient mutant mice. Eur. J. Immunol. 25:838.
    OpenUrlCrossRefPubMed
  6. ↵
    MacDonald, H. R.. 1995. NK1.1+ T cell receptor-αβ+ cells: new clues to their origin, specificity and function. J. Exp. Med. 181:633.
    OpenUrlPubMed
  7. ↵
    Kronenberg, M., L. Gapin. 2002. The unconventional lifestyle of NKT cells. Nat. Rev. Immunol. 2:557.
    OpenUrlPubMed
  8. ↵
    Lantz, O., A. Bendelac. 1994. An invariant T cell receptor α chain is used by a unique subset of major histocompatibility complex class I-specific CD4+ and CD4−8− T cells in mice and humans. J. Exp. Med. 180:1097.
    OpenUrlAbstract/FREE Full Text
  9. ↵
    Arase, H., N. Arase, T. Saito. 1996. Interferon γ production by natural killer (NK) cells and NK1.1+ T cells upon NKR-P1 cross-linking. J. Exp. Med. 183:2391.
    OpenUrlAbstract/FREE Full Text
  10. ↵
    Kawamura, T., K. Takeda, S. K. Mendiratta, H. Kawamura, L. Van Kaer, H. Yagita, T. Abo, K. Okumura. 1998. Critical role of NK1+ T cells in IL-12-induced immune responses in vivo. J. Immunol. 160:16.
    OpenUrlAbstract/FREE Full Text
  11. ↵
    Burdin, N., L. Brossay, M. Kronenberg. 1999. Immunization with α-galactosylceramide polarizes CD1-reactive NK T cells toward Th2 cytokine synthesis. Eur. J. Immunol. 29:2014.
    OpenUrlCrossRefPubMed
  12. ↵
    Emoto, M., Y. Emoto, I. B. Buchwalow, S. H. E. Kaufmann. 1999. Induction of IFN-γ-producing CD4+ natural killer T cells by Mycobacterium bovis bacillus Calmette Guérin. Eur. J. Immunol. 29:650.
    OpenUrlCrossRefPubMed
  13. ↵
    Apostolou, I., Y. Takahama, C. Belmant, T. Kawano, M. Huerre, G. Marchal, J. Cui, M. Taniguchi, H. Nakauchi, J. J. Fourniè, et al 1999. Murine natural killer cells contribute to the granulomatous reaction caused by mycobacterial cell walls. Proc. Natl. Acad. Sci. USA 96:5141.
    OpenUrlAbstract/FREE Full Text
  14. ↵
    Mempel, M., C. Ronet, F. Suarez, M. Gilleron, G. Puzo, L. Van Kaer, A. Lehuen, P. Kourilsky, G. Gachelin. 2002. Natural killer T cells restricted by the monomorphic MHC class 1b CD1d molecules behave like inflammatory cells. J. Immunol. 168:365.
    OpenUrlAbstract/FREE Full Text
  15. ↵
    Behar, S. M., C. C. Dascher, M. J. Grusby, C. R. Wang, M. B. Brenner. 1999. Susceptibility of mice deficient in CD1D or TAP1 to infection with Mycobacterium tuberculosis. J. Exp. Med. 189:1973.
    OpenUrlAbstract/FREE Full Text
  16. ↵
    D’Souza, C. D., A. M. Cooper, A. A. Frank, S. Ehlers, J. Turner, A. Bendelac, I. M. Orme. 2000. A novel nonclassic β2-microglobulin-restricted mechanism influencing early lymphocyte accumulation and subsequent resistance to tuberculosis in the lung. Am. J. Respir. Cell Mol. Biol. 23:188.
    OpenUrlCrossRefPubMed
  17. ↵
    Kawakami, K., Y. Kinjo, K. Uezu, S. Yara, K. Miyagi, Y. Koguchi, T. Nakayama, M. Taniguchi, A. Saito. 2002. Minimal contribution of Vα14 natural killer T cells to Th1 response and host resistance against mycobacterial infection in mice. Microbiol. Immunol. 46:207.
    OpenUrlCrossRefPubMed
  18. ↵
    Sugawara, I., H. Yamada, S. Mizuno, C. Li, T. Nakayama, M. Taniguchi. 2002. Mycobacterial infection in natural killer T cell knockout mice. Tuberculosis 82:97.
    OpenUrlCrossRefPubMed
  19. ↵
    Szalay, G., U. Zugel, C. H. Ladel, S. H. E. Kaufmann. 1999. Participation of group 2 CD1 molecules in the control of murine tuberculosis. Microbes Infect. 14:1153.
    OpenUrl
  20. ↵
    Chackerian, A., J. Alt, V. Perera, S. M. Behar. 2002. Activation of NKT cells protects mice from tuberculosis. Infect. Immun. 70:6302.
    OpenUrlAbstract/FREE Full Text
  21. ↵
    Cui, J., T. Shin, T. Kawano, H. Sato, E. Kondo, I. Toura, Y. Kaneko, H. Koseki, M. Kanno, M. Taniguchi. 1997. Requirement for Vα14 NKT cells in IL-12-mediated rejection of tumors. Science 278:1623.
    OpenUrlAbstract/FREE Full Text
  22. ↵
    Dieli, F., G. Sireci, D. Russo, M. Taniguchi, J. Ivanyi, C. Fernandez, M. Troye-Blomberg, G. De Leo, A. Salerno. 2000. Resistance of natural killer T cell-deficient mice to systemic Shwartzman reaction. J. Exp. Med. 192:1645.
    OpenUrlAbstract/FREE Full Text
  23. ↵
    Dieli, F., J. Ivanyi, P. Marsh, A. Williams, I. Naylor, G. Sireci, N. Caccamo, C. Di Sano, A. Salerno. 2003. Characterization of lung γδ T cells following intranasal infection with Mycobacterium bovis bacillus Calmette-Guérin. J. Immunol. 170:463.
    OpenUrlAbstract/FREE Full Text
  24. ↵
    Matsuda, J. L., O. V. Naidenko, L. Gapin, T. Nakayama, M. Taniguchi, R. Wang, Y. Koezuka, M. Kronenberg. 2000. Tracking the response of natural killer T cells to a glycolipid antigen using CD1d tetramers. J. Exp. Med. 192:741.
    OpenUrlAbstract/FREE Full Text
  25. ↵
    Matsuzaki, G., X. Y. Li, T. Kadena, F. Song, K. Hiromatsu, H. Yoshisa, K. Nomato. 1995. Early appearance of T cell receptor αβ+ CD4− CD8− T cells with a skewed variable region repertoire after infection with Listeria monocytogenes. Eur. J. Immunol. 25:1985.
    OpenUrlCrossRefPubMed
  26. ↵
    Ishikawa, H., H. Hisaeda, M. Taniguchi, T. Nakayama, T. Sakai, Y. Maekawa, Y. Nakano, M. Zhang, M. Nishitani, M. Takashima, et al 2000. CD4+ Vα14 NKT cells play a crucial role in an early stage of protective immunity against infection with Leishmania major. Int. Immunol. 12:1267.
    OpenUrlAbstract/FREE Full Text
  27. ↵
    Pied, S., J. Roland, A. Louise, D. Voegtle, V. Soulard, D. Mazier, P. A. Cazenave. 2000. Liver CD4− CD8− NK1.1+ TCRαβ intermediate cells increase during experimental malaria infection and are able to exhibit inhibitory activity against the parasite liver stage in vitro. J. Immunol. 164:1463.
    OpenUrlAbstract/FREE Full Text
  28. ↵
    Kawakami, K., Y. Kinio, K. Uezu, S. Yara, K. Miyagi, Y. Koguchi, T. Nakayama, M. Taniguchi, A. Saito. 2001. Monocyte chemoattractant protein-1-dependent increase of Vα14 NKT cells in lungs and their roles in Th1 response and host defense in cryptococcal infection. J. Immunol. 167:6525.
    OpenUrlAbstract/FREE Full Text
  29. ↵
    Nakano, Y., H. Hisaeda, T. Sakai, H. Ishiwaka, M. Zhang, Y. Maekawa, T. Zhang, M. Takashima, M. Nishitami, R. A. Good, et al 2002. Roles of NKT cells in resistance against infection with Toxoplasma gondii and expression of heat shock protein 65 in host macrophages. Microbes Infect. 4:1.
    OpenUrlCrossRefPubMed
  30. ↵
    Guinet, F., Ronet C., Mempel M., Huerre M., Carniel E., Gachelin G.. NKT cells-containing inflammatory lesions inducedby Yersinia paratuberculosis glycolipids. Immunol. Lett. 80:113.
  31. ↵
    Liu, C. C., B. Perussia, J. D. E. Young. 2000. The emerging role of IL-15 in NK cell development. Immunol. Today 21:113.
    OpenUrlCrossRefPubMed
  32. ↵
    Faunce, D. E., K. Sonoda, J. Stein-Streilein. 2001. MIP-2 recruits NKT cells to the spleen during tolerance induction. J. Immunol. 166:313.
    OpenUrlAbstract/FREE Full Text
  33. ↵
    Moretta, A.. 2002. Natural killer cells and dendritic cells: rendezvous in abused tissues. Nat. Rev. Immunol. 2:957.
    OpenUrlCrossRefPubMed
  34. ↵
    Kim, C. H., B. Johnston, E. C. Butcher. 2002. Trafficking machinery of NKT cells: shared and differential chemokine receptor expression among Vα24+ Vβ11+ NKT cell subsets with distinct cytokine-producing capacity. Blood 100:11.
    OpenUrlAbstract/FREE Full Text
  35. ↵
    Hameg, A., I. Apostolou, M. Leite-de-Moraes, J. M. Gombert, C. Garcia, Y. Koezuka, J. F. Bach, A. Herbelin. 2000. A subset of NKT cells that lacks NK1.1 marker, expresses CD1d molecules and autopresents the α-galactosylceramide antigen. J. Immunol. 165:4917.
    OpenUrlAbstract/FREE Full Text
  36. ↵
    Benlagha, K., T. Kyin, A. Beavis, L. Teyton, A. Bendelac. 2002. A thymic precursor of the NKT cell lineage. Science 296:553.
    OpenUrlAbstract/FREE Full Text
  37. ↵
    Flynn, J. L., M. M. Goldstein, J. Chan, K. J. Triebold, K. Pfeffer, C. J. Lowenstein, R. Schreiber, T. W. Mak, B. R. Bloom. 1995. Tumor necrosis factor-α is required in the protective immune response against Mycobacterium tuberculosis in mice. Immunity 2:561.
    OpenUrlCrossRefPubMed
  38. ↵
    Kindler, V., A. P. Sappino, G. E. Grau, P. F. Pignet, P. Vassalli. 1989. The inducing role of tumour necrosis factor in the development of bactericidal granulomas during BCG infection. Cell 56:731.
    OpenUrlCrossRefPubMed
  39. ↵
    Ming, W. J., L. Bersani, A. Mantovani. 1987. Tumor necrosis factor is chemotactic for monocyte and polymorphonuclear leukocytes. J. Immunol. 138:1469.
    OpenUrlAbstract
  40. ↵
    Josien, R., B. R. Wong, H. L. Li, R. M. Steinman, Y. Choi. 1999. TRANCE, a TNF family member, is differentially expressed on T cell subsets and induces cytokine production in dendritic cells. J. Immunol. 162:2562.
    OpenUrlAbstract/FREE Full Text
  41. ↵
    Bekker, L. G., A. L. Moreira, A. Bergtold, S. Freeman, B. Ryffel, G. Kaplan. 2000. Immunopathologic effects of tumor necrosis factor α in murine mycobacterial infection are dose dependent. Infect. Immun. 68:6954.
    OpenUrlAbstract/FREE Full Text
  42. ↵
    Godfrey, D. J., K. J. Hammond, L. D. Poulton, M. J. Smyth, A. G. Naxter. 2000. NKT cells: facts, functions and fallacies. Immunol. Today 21:573.
    OpenUrlCrossRefPubMed
  43. ↵
    Sonoda, K., M. Exley, S. Snapper, S. P. Balk, J. Stein-Streilein. 1999. CD1-reactive natural killer T cells are required for development of systemic tolerance through an immune-privileged site. J. Exp. Med. 190:1215.
    OpenUrlAbstract/FREE Full Text
  44. ↵
    Moodycliffe, A. M., D. Nghiem, C. Clydesdale, S. E. Ullrich. 2000. Immune suppression and skin cancer development: regulation by NKT cells. Nat. Immunol. 1:521.
    OpenUrlCrossRefPubMed
  45. ↵
    Sharif, S., G. A. Arreaza, P. Zucker, Q. S. Mi, J. Sondhi, O. V. Naidenko, M. Kronenberg, Y. Koezuka, T. Delovitch, J. M. Gombert, et al 2001. Activation of natural killer T cells by α-galactosylceramide treatment prevents the onset and recurrence of autoimmune type 1 diabetes. Nat. Med. 7:1057.
    OpenUrlCrossRefPubMed
  46. ↵
    Hong, S., M. T. Wilson, I. Serizawa, L. Wu, N. Singh, O. V. Naidenko, T. Miura, T. Haba, D. C. Scherer, J. Wei, et al 2001. The natural killer T-cell ligand α-galactosylceramide prevents autoimmune diabetes in non-obese diabetic mice. Nat. Med. 7:1052.
    OpenUrlCrossRefPubMed
  47. ↵
    Singh, A. K., M. T. Wilson, S. Hong, D. Olivares-Villagomez, C. Du, A. K. Stanic, S. Joyce, S. Sriram, Y. Koezuka, L. Van Kaer. 2001. Natural killer T cell activation protects mice against autoimmune experimental encephalomyelitis. J. Exp. Med. 194:1801.
    OpenUrlAbstract/FREE Full Text
Back to top

In this issue

Print
Download PDF
Article Alerts
Email Article
Citation Tools

Jump to section