Abstract
It has been proposed that the induction of cellular immunity and resistance to intracellular pathogens is dependent upon CD40 ligand (CD40L). In the present study we show that this proposal is not ubiquitously supported. Mice genetically deficient in CD40L (CD40LKO) were resistant to i.v. infection with Mycobacterium tuberculosis when assessed by survival and bacteriologic burden in the spleen, liver, and lungs. Infected CD40LKO mice developed granulomas that lacked epithelioid cells and were less numerous and markedly smaller than those observed in control mice. Upon stimulation with purified protein derivative of M. tuberculosis,CD4+ T cells from infected CD40LKO mice proliferated and produced high levels of IFN-γ but not IL-4. Finally, spleen cells from CD40LKO mice stimulated with M. tuberculosis produced IL-12, TNF, and nitric oxide levels comparable to those produced by control cells. In contrast to original proposals, these data clearly show that protective Th1 immunity can be achieved against intracellular pathogens (e.g., Mycobacterium) independently of CD40L.
Among the infectious diseases afflicting humans, tuberculosis is the leading killer of adults, with a mortality greater than that of all other infectious diseases combined (1). The lack of an effective vaccine is at least in part responsible for this high mortality rate. However, the development of an efficacious vaccine against Mycobacterium tuberculosis has been hampered by a poor understanding of the immunologic mechanisms of protection and the pathogenesis of this disease. Nonetheless, it is widely accepted that both CD4+ and CD8+ T cells are essential for the control of tuberculosis infection. The precise mechanisms by which the cell-mediated immune response operates to effectively contain the infectious process are not fully understood, although recent studies in gene knockout mice have begun to shed some light on this topic. Using this approach, several groups have recently shown that the cytokines IFN-γ, TNF, and IL-12 are essential for the development of resistance to M. tuberculosis (2, 3, 4, 5, 6, 7). In addition, humans who have a defect in their IFN-γ receptor gene are more susceptible to mycobacterial infections and develop severe disease after vaccination with Calmette-Guérin bacillus, the attenuated strain of Mycobacterium bovis (8, 9). The convergent mechanism by which these cytokines mediate anti-mycobacterial activity appears to be the production of nitric oxide (NO)2 by the macrophages. This metabolite is particularly induced by IFN-γ and TNF and is crucial for protection against tuberculosis (10).
CD40 ligand (CD40L) is a type II membrane protein preferentially expressed by activated T cells that binds to its counter-receptor CD40 present on the surface of resting B cells, macrophages, and other APC (11). CD40-CD40L interactions are critical not only for the synthesis of IgG, IgA, and IgE (12, 13, 14, 15), but also for the activation of CD4+ T cell-dependent effector functions, including the killing of intracellular pathogens such as the protozoan Leishmania (16, 17, 18). Moreover, humans with a defective CD40L gene develop hyper-IgM syndrome and display increased susceptibility to infection with Cryptococcus, Pneumocystis, and Histoplasma (19). Because resistances to tuberculosis (also an intracellular pathogen) and leishmaniasis are associated with the production of similar CD40-CD40L-associated mediators of immunity, namely IFN-γ, TNF, IL-12, and NO, the aim of this study was to assess the role of CD40L in the induction of cell-mediated immunity and control of M. tuberculosis infection. Using CD40L-deficient (CD40LKO) mice, the results show, in contrast to what is observed in leishmaniasis, that both cell-mediated immunity and resistance to M. tuberculosis can be achieved independently of CD40L.
Materials and Methods
Mice
C57BL/6 mice were obtained from Charles River Laboratories (Wilmington, MA). CD40LKO mice and mice lacking the TNFp55 receptor (TNFRKO) were generated at Immunex Corp. (Seattle, WA) by homologous gene recombination as described previously (13). The CD40LKO mice used in these studies were backcrossed for seven or eight generations onto the C57BL/6 background. All mice were maintained under specific pathogen-free conditions and used at 8 to 12 wk of age.
Bacteria and mouse infections
Virulent M. tuberculosis H37Rv strain (American Type Culture Collection, Rockville, MD) bacteria were stored as frozen aliquots at −70°C. For infection of mice, bacteria were thawed, resuspended in PBS/Tween-80 (0.05%), and pushed through a 26-gauge needle six times in preparation for i.v. injection.
Histology
Tissue sections were fixed in 10% formalin and embedded in paraffin blocks. Sections (5 μm) were stained with hematoxylin and eosin or by the Ziehl-Neelsen method for acid fast bacilli or were impregnated with silver for demonstration of reticulum fibers.
Colony-forming units
Homogenates of spleen, liver, and lung were prepared in PBS/Tween-80 (0.05%) and plated at a 5- or 10-fold serial dilution on BBL Middlebrook 7H10 agar plates (Becton Dickinson Microbiology Systems, Cockeysville, MD). CFU were enumerated 2 to 3 wk later.
Proliferation and cytokine assays
Spleen cells were obtained by conventional procedures and centrifuged over Ficoll-Hypaque followed by passage through a Sephadex G-10 column. Proliferative responses of mononuclear cells stimulated with anti-CD3 mAb or purified protein derivative of M. tuberculosis (PPD) were measured by [3H]thymidine incorporation on day 6. For cytokine analysis, spleen cells were cultured at 2 × 106 cells/well (24-well plate) for 72 h, and in some experiments blocking rat anti-mouse CD4 and CD8 mAbs (clones GK1.5 and 53-6.7 respectively, PharMingen, San Diego, CA) were added to the cultures. Supernatants were harvested and analyzed for IFN-γ, TNF, and IL-4 by double sandwich ELISAs following the manufacturer’s protocol (PharMingen). IL-12 (p70) was measured using a biologic assay, essentially as previously described (20). NO concentrations were evaluated using the Griess Reagent System (Promega, Madison, WI) according to the manufacturer’s protocol.
Results and Discussion
To investigate the role of CD40L in resistance to tuberculosis, we initially compared the course of infection of M. tuberculosis in CD40LKO mice with that in sex- and age-matched C57BL/6 mice. The CD40LKO mice used in these studies had been backcrossed onto the C57BL/6 background for at least seven generations, which made them >99% genetically identical with C57BL/6 mice. Mice were infected i.v. with 2 × 105 CFU of M. tuberculosis H37Rv strain, and two parameters of resistance were analyzed, namely survival and bacterial load in the spleen, liver, and lungs. As an internal control for susceptibility, mice lacking the p55 TNF receptor (TNFRKO) were also challenged with M. tuberculosis. The results (Fig. 1⇓) show that all the CD40LKO and the C57BL/6 mice (resistant strain) survived for the entire 30-wk observation period. In contrast, the TNFRKO mice died within 4 wk of infection, as expected (4).
Survival of CD40LKO mice infected with M. tuberculosis. CD40LKO, C57BL/6, and TNFRKO mice were infected i.v. with 2 × 105 M. tuberculosis H37Rv, and the course of infection (survival) was monitored for 30 wk. The C57BL/6 and TNFRKO mice were included as polar controls for resistance and susceptibility to M. tuberculosis infection, respectively. Data shown are from one of two experiments with identical results, each with five mice per group.
The bacterial load and the specific Ab response to tubercle bacilli were next determined in mice 3 and 6 wk after infection. As indicated in Figure 2⇓, the number of M. tuberculosis CFU recovered from the lungs of CD40LKO and C57BL/6 mice was virtually indistinguishable at either time point after infection, and there was less than a 0.2-log difference for the spleen and liver. To test the resistance of the CD40LKO mice to a larger M. tuberculosis challenge, the animals were infected with 2 × 106 CFU. Ten weeks after infection, survival was again 100% in both the C57BL/6 and CD40LKO mice; however, the CD40LKO mice presented with 0.5 to 1 log more CFU in the three organs than did the C57BL/6 mice (data not shown). Nevertheless, CD40LKO mice are highly resistant to infection with M. tuberculosis when compared with susceptible strains of mice such as BALB/c, CB.10-H-2, and KSJ, which have an average survival of <70 days after infection (6). This resistance is even more striking when compared with that in IFN-γKO, TNFRKO, and IL-12KO mice, which are unable to control the infection and die 15 to 45 days postchallenge (2, 3, 4, 7).
Bacterial burden in M. tuberculosis infected CD40LKO and C57BL/6 mice. Mice were infected with M. tuberculosis, and the course of infection (colony-forming units) was determined 3 wk (A) and 6 wk (B) later in the lungs, spleen, and liver. Data shown are the mean ± SEM of the CFU from five mice.
The Ab response to M. tuberculosis was evaluated in infected animals by ELISA using M. tuberculosis lysate as Ag. Sera from C57BL/6 mice contained high titers of both IgM and IgG anti-mycobacterial Abs. In contrast, the sera from CD40LKO mice had low titer of IgM, and no anti-mycobacteria IgG Abs could be detected (data not shown). These results confirm the previous observation that CD40L is essential for Ig isotype switching (12, 13, 14, 15) and further support the concept that resistance to tuberculosis is independent of the Ab-mediated immune response.
Histopathologic examination of tissue sections revealed marked differences between the granulomas developed by CD40LKO mice compared with those in C57BL/6 mice following infection with M. tuberculosis. Histologic sections of the liver, spleen, and lung showed that granulomas developed in parallel in these organs in both groups of mice; only liver lesions were analyzed morphometrically (Fig. 3⇓). The number of granulomas observed in the livers of CD40LKO mice was reduced compared with that in C57BL/6 mice (7.1 ± 0.78 vs 11.3 ± 0.55/×100 microscopic field, respectively). Moreover, the granulomas developed by the CD40LKO mice appeared disintegrated (Fig. 3⇓B), lacked epithelioid cells (Fig. 3⇓D), and were markedly reduced in size, measuring 3.2 ± 0.25 × 103 μm2 in comparison with the 10.27 ± 1.20 × 103 μm2 found for the C57BL/6 mice. In agreement with the CFU counts, the numbers of M. tuberculosis bacilli in both groups were remarkably similar (0.35 bacilli/×500 microscopic field in CD40LKO mice vs 0.36 bacilli in controls). Since the survival and bacterial burden were similar in these mice, the results indicate that organized granulomas are not essential for the prevention of dissemination of the tubercle bacilli.
Histopathology of liver sections from M. tuberculosis-infected CD40LKO and C57BL/6 mice. Mice were infected with M. tuberculosis and killed 6 wk later. Liver sections were stained with hematoxylin and eosin (A andB) or impregnated with silver for reticulum fibers (C and D). A, Typical epithelioid granuloma in C57BL/6-infected mice. B, Small granuloma devoid of epithelioid cells in CD40LKO-infected mice. C, Normal pattern of reticulum fibers present in granulomas of C57BL/6-infected mice. D, The reticulum fiber pattern appears to disintegrate in small granulomas of CD40LKO-infected mice. Original magnification, ×160.
CD40-CD40L interactions have been directly implicated in the metabolic pathways for the production of IL-12, TNF, and NO (20, 21, 22, 23, 24). In addition, the enhanced susceptibility of CD40LKO mice to leishmaniasis is clearly associated with impaired Ag-specific generation of Th1 responses, which is reflected by the production of low levels of IFN-γ, TNF, IL-12, and NO (16, 17, 18). Moreover, in the murine model of experimental allergic encephalomyelitis, the onset of the disease in mice is dependent upon the signals provided by the CD40-CD40L interactions for the up-regulation of B7 expression, T cell activation, and IFN-γ production (25). In view of these observations, it became important to investigate the association between these mediators of immunity and the observed resistance of the CD40LKO mice to tuberculosis. To this end, CD40LKO and C57BL/6 mice were infected with M. tuberculosis, and PPD-induced T cell proliferation and IFN-γ production were evaluated in vitro. Figure 4⇓ demonstrates that nonadherent spleen cells from CD40LKO mice at 3 wk postinfection develop a strong proliferative response and IFN-γ production to the same extent or greater than that observed in C57BL/6 mice. In addition, the production of IFN-γ was totally abrogated by an anti-CD4 mAb, but not by an anti-CD8 mAb. These Abs individually inhibited 50% of the IFN-γ production by anti-CD3-stimulated cells (data not shown), thus confirming their specificity. These experiments clearly demonstrate that M. tuberculosis Ags induce the production of IFN-γ by CD4+ T cells independently of CD40L. No IL-4 could be detected in the supernatants of any of the cultures stimulated with PPD. In addition, spleen cells from uninfected CD40LKO or C57BL/6 mice failed to proliferate or produce IFN-γ upon stimulation with PPD (data not shown).
In vitro proliferative response and IFN-γ production by nonadherent spleen cells of M. tuberculosis-infected CD40LKO and C57BL/6 mice. Three weeks after infection with M. tuberculosis, spleens were removed, and nonadherent mononuclear cells were stimulated with PPD (25 μg/ml) for 6 days or with anti-CD3 mAb (1 μg/ml) for 3 days. The proliferative response (A) was measured by [3H]thymidine incorporation during the last 8 h of culture. IFN-γ production (B) was measured by ELISA in the culture supernatants harvested 72 h after the initiation of the cultures. C, Inhibition of IFN-γ production by anti-CD4 mAb. Nonadherent spleen cells from M. tuberculosis-infected CD4LKO mice were either nonstimulated (medium) or stimulated with various concentrations of PPD in the absence (control) or the presence of anti-CD4 or anti-CD8 mAbs. IFN-γ was measured by ELISA in the culture supernatants harvested 72 h after the initiation of the cultures. Data shown are the mean ± SEM of the results obtained from five mice.
The production of IL-12 and TNF was measured in vitro after stimulation of whole spleen cells with either LPS or M. tuberculosis. Cells (5 × 106/ml) were primed with IFN-γ for 24 h and then stimulated with either LPS (5 μg/ml) or M. tuberculosis (50 viable bacteria/spleen cell). Supernatants were collected and assayed for IL-12 p70 (by bioassay) and TNF (by ELISA). Similar levels of IL-12 and TNF were present in the supernatants of cultures from both CD40L/KO and C57BL/6 mice stimulated with either M. tuberculosis (Table I⇓) or LPS (data not shown). NO production was measured in the culture supernatants of spleen cells obtained 14 days after infection with M. tuberculosis (4). Spleen cells were cultured in the absence of any stimulus for 24 h, and the NO production was assayed in the culture supernatants using the Griess reagent. Table I⇓ shows that cells from M. tuberculosis-infected CD40LKO mice produce significant levels of NO, albeit to a slightly lesser extent than spleen cells from C57BL/6 mice.
Cytokine and NO production by CD40LKO and C57BL/6 spleen cellsa
Collectively, these results indicate that while the CD40-CD40L interactions are critical for the generation of protective immune response against some intracellular pathogens such as Leishmania, the immune system has alternative or redundant mechanisms to impair the infectious process caused by other intracellular pathogens such as M. tuberculosis. The question arises as to why CD40LKO mice are highly susceptible to infection with an intracellular pathogen such as Leishmania but are able to control infection by M. tuberculosis. The clear difference is that there is no apparent T cell response following infection of CD40LKO mice with Leishmania (16, 17, 18), whereas there is clear priming for T cell proliferation and IFN-γ production following M. tuberculosis infection. The answer may lie in the distinct effects of these two pathogens on the APCs. Activation of macrophages for the production of IL-12, for the up-regulation of B7-1 (CD80), and for the production of NO can occur in either a T cell-independent (CD40L-independent) or a T cell-dependent (CD40L-dependent) manner. The CD40L-independent pathway can be mediated by bacterial products such as LPS that act directly on APCs. Interestingly, M. tuberculosis has been reported to have a number of direct activation effects on macrophages, including induction of IL-12, TNF, and NO and up-regulation of B7-1; thus, it appears to bypass the requirement for the CD40L-dependent pathway (10, 26, 27). In addition, it has recently been reported that T. gondii another intracellular pathogen, also induces in vivo IL-12 production by dendritic cells independently of CD40L (28). The reasons why intramacrophage pathogens induce the production of different patterns of mediators of immunity is not understood, but may be related to a particular macrophage invasion mechanism present among pathogenic mycobacteria, including M. tuberculosis, M. leprae, and M. avium, and not in other nonmycobacterial pathogens, such as Leishmania (29). The utilization of this unique pathway by pathogenic mycobacteria and not by Leishmania could result in different signaling for the production of the different macrophages mediators of immunity.
In conclusion, our results indicate that pathogens such as M. tuberculosis, in contrast with other organisms such as Leishmania, activate macrophages and dendritic cells in a T cell-independent manner and thus do not require CD40L-CD40 interactions for the development of a successful cell-mediated immune response.
Acknowledgments
The authors thank Drs. Kim Campbell and Jacques Peschon for providing KO mice, and Drs. Teri Foy and Fiona Durie for guidance and discussions.
Footnotes
-
↵1 Address correspondence and reprint requests to Dr. Antonio Campos-Neto, Infectious Disease Research Institute, 1124 Columbia St., Suite 464, Seattle, WA 98104. E-mail address: acampos{at}corixa.com
-
↵2 Abbreviations used in this paper: NO, nitric oxide; CD40L, CD40 ligand; CD40LKO, mice genetically deficient in CD40 ligand; TNFRKO, mice lacking the TNFp55 receptor; PPD, purified protein derivative of Mycobacterium tuberculosis.
- Received October 28, 1997.
- Accepted December 30, 1997.
- Copyright © 1998 by The American Association of Immunologists
References
In this issue
