CD40 Ligand Trimer Enhances the Response of CD8⁺ T Cells to Mycobacterium tuberculosis ¹

Immunity against Mycobacterium tuberculosis is mediated by CD8⁺ T cells, as animals deficient in CD8⁺ cells as a result of gene deletion show enhanced susceptibility to infection (1). In humans infected with M. tuberculosis, CD8⁺ T cells produce IFN-γ in response to mycobacterial Ag, lyse M. tuberculosis-infected target cells (2, 3), and can directly kill mycobacteria by secreting granulysin (4, 5). Effective immunity against tuberculosis is likely to depend in part on priming and maintenance of Ag-specific CD8⁺ T cells. The maintenance of effective CD8⁺ T cell responses depends in part on CD4⁺ T cells (6, 7). In some experimental systems, CD4⁺ T cell help is mediated primarily through binding of CD40 ligand (CD40L)³ on activated CD4⁺ T cells to CD40 on APCs (8, 9, 10), although CD40-independent mechanisms have also been described (11).

Most humans infected with M. tuberculosis manifest an effective immune response and are healthy tuberculin reactors. However, some infected persons have ineffective immunity and develop active tuberculosis. We previously reported that M. tuberculosis-stimulated peripheral blood CD4⁺ T cells from tuberculosis patients show reduced expression of CD40L, compared with CD4⁺ T cells from healthy tuberculin reactors (12). Furthermore, the number of M. tuberculosis-responsive IFN-γ-producing CD8⁺ T cells is reduced in tuberculosis patients, compared with healthy tuberculin reactors, and M. tuberculosis-induced IFN-γ production by CD8⁺ T cells requires the presence of CD4⁺ T cells (13). We hypothesized that CD4⁺ T cells enhance M. tuberculosis-reactive CD8⁺ T cell function through CD40/CD40L interactions. To evaluate this hypothesis, we investigated the effect of recombinant CD40L trimer (CD40LT) on IFN-γ production and CTL activity of M. tuberculosis-reactive CD8⁺ T cells isolated from the blood of healthy tuberculin reactors.

Peripheral blood was obtained from 20 healthy tuberculin reactors and from three tuberculin-negative persons. All subjects provided informed consent, and the work was approved by the Institutional Review Boards of the University of Texas Health Center (Tyler, TX) and the University of North Texas Health Science Center (Fort Worth, TX).

PBMC were isolated by differential centrifugation over Ficoll-Paque (Pharmacia Fine Chemicals, Piscataway, NJ), and monocytes were isolated by adherence, as previously described (14). Adherent cells were 90–95% CD14⁺, as measured by cytofluorometric analysis, using an EPICS C flow cytometer (Beckman-Coulter, Hialeah, FL). Nonadherent PBMC were maintained in RPMI 1640 (Life Technologies, Grand Island, NY) and 10% heat-inactivated pooled human serum (Interstate Blood Bank, Memphis, TN) overnight. The next day, CD8⁺ T cells were purified by positive selection with magnetic beads conjugated with anti-human CD8 mAb (Miltenyi Biotec, Auburn, CA). The purity of CD8⁺ T cells was 95–99%, as measured by flow cytometry.

Purified monocytes were cultured in 1 ml of RPMI 1640 with 10% heat-inactivated pooled human serum and 100 μg/ml penicillin (Life Technologies, Grand Island, NY) in 24-well plates at 5 × 10⁵ cells/well. Monocytes were cultured with 10 μg/ml heat-killed M. tuberculosis Erdman strain or infected with single cell suspensions of M. tuberculosis H37Ra at a multiplicity of infection of 1.5:1, as described previously (15). After overnight incubation, the cells were washed several times with RPMI 1640 to remove extracellular bacilli. CD8⁺ T cells were then added to the wells at a ratio of five cells to one monocyte. Cells were cultured for 4–6 days, with or without 5 μg/ml recombinant human CD40LT (Amgen, Seattle, WA), a concentration that was found to yield optimal results in preliminary experiments. After 6 days, supernatants were collected in some cases and frozen at −70°C until measurement of IFN-γ concentrations.

In some experiments, 10⁵ CD8⁺ T cells and 2 × 10⁴ autologous M. tuberculosis-infected monocytes were cultured in triplicate in each well of a 96-well flat-bottom plate (Corning, Corning, NY) for 4 days, in the presence or absence of anti-IL-12 (1 μg/ml), anti-IL-15 (1 μg/ml), both from R&D Systems (Minneapolis, MN), and anti-IL-18 (1 μg/ml; MBL International, Nagoya, Japan). Cell culture supernatants were collected and frozen at −70°C until measurement of cytokine concentrations.

The concentrations of IFN-γ, IL-12 p70, IL-15, and IL-18 in supernatants were measured by ELISA, using paired Abs from BD PharMingen (San Diego, CA) (IFN-γ), R&D Systems (IL-12 p70 and IL-15), and MBL International (IL-18). The lower limits of detection were 5 pg/ml for IFN-γ and IL-12, 8 pg/ml for IL-15, and 13 pg/ml for IL-18.

The frequency of CD8⁺ T cells that produced IFN-γ in response to M. tuberculosis was determined by the ELISPOT assay, as previously described (13). Briefly, triplicate wells of 10⁵ purified freshly isolated CD8⁺ T cells were cultured with 2 × 10⁴ autologous M. tuberculosis-infected monocytes, with or without CD40LT, in 96-well nitrocellulose-backed plates for 18 h, using the anti-human IFN-γ mAbs 1-DIK and 7-B6–1 (Mabtech, Nacka, Sweden) as coating and detection Abs, respectively. The wells were developed, according to the manufacturer’s instructions, and the spots in the air-dried plates were counted with a stereomicroscope.

CTL assays were performed by standard methods (16). Target cells were monocytes that had been infected with M. tuberculosis at a multiplicity of infection of 1.5:1 overnight, as outlined above, and labeled with 40 μCi ⁵¹Cr Na₂CrO₄ (ICN Pharmaceuticals, Costa Mesa, CA) for the last 16 h in 12-well plates (BD Biosciences, Franklin Lakes, NJ), then washed, counted and plated in 96-well round-bottom plates (Corning) at a final concentration of 10⁴ target cells per well in 100 μl of RPMI 1640 with 10% human serum. To obtain effector cells, CD8⁺ T cells from healthy tuberculin reactors were cultured with M. tuberculosis-infected autologous monocytes in the presence or absence of 5 μg/ml CD40LT for 6 days. The nonadherent CD8⁺ T cells were collected and added to the target cells in triplicate at an E:T ratio of 40:1. The effector cells were 95% CD8⁺, as determined by flow cytometric analysis. Effector and target cells were incubated for 6 h, 100 μl of supernatant was collected from each well, and radioactivity was measured in a gamma counter. The percent lysis was calculated as 100 × (experimental release − spontaneous release)/(maximum release − spontaneous release). The results are shown as percent net specific lysis, calculated by subtracting the percent lysis of uninfected target cells from the percent lysis of M. tuberculosis-infected target cells.

CD8⁺ T cells from healthy tuberculin reactors were cultured with autologous M. tuberculosis-infected monocytes for 6 days, as outlined above, with or without 5 μg/ml CD40LT. Nonadherent cells (95% CD8⁺) were then collected, washed, and stained for granulysin and perforin as described (17). Briefly, cells were permeabilized with permeabilization buffer (BD PharMingen) for 30 min on ice, then washed with cell wash buffer (BD PharMingen) and incubated with anti-granulysin mAb, DH4 (18), anti-perforin mAb, δG9 (Kamiya Biomedical, Seattle, WA) or isotype control mouse IgG1 (BD PharMingen), for 30 min, and then washed three times with cell wash buffer. The cells were then incubated with FITC goat F(ab′)₂ anti-mouse IgG (Southern Biotechnology Associates, Birmingham, AL) for 30 min on ice. To detect Fas ligand (FasL; CD95), CD8⁺ T cells were incubated with biotin-labeled anti-CD95 mAb, then with streptavidin-PE (both from BD PharMingen). The cells were gated on lymphocytes, and the percentage of positive cells was determined by flow cytometry.

CD8⁺ T cells were cultured with M. tuberculosis-infected monocytes as outlined above, or were stimulated with PMA (50 ng/ml) and ionomycin (1 μg/ml), both from Sigma-Aldrich (St. Louis, MO) for 15–60 min. CD8⁺ T cells were collected, washed and suspended in cell lysis buffer containing 1% Nonidet P-40, 50 mM Tris-HCl (pH 7.4), 150 mM NaCl, 1 mM PMSF, 2 μg/ml leupeptin, 2 μg/ml aprotinin, 10 μg/ml chymostatin, 10 μg/ml trypsin-chymotrypsin, 1 mM DTT, 20 mM β-glycerophosphate, 5 mM sodium orthovanadate, 10 mM sodium fluoride, and 5 mM sodium pyrophosphate (all from Sigma-Aldrich). The samples were then frozen and thawed twice, using dry ice and a 37°C water bath, incubated with gentle agitation at 4°C for 25 min, and centrifuged at 18,000 × g for 10 min. The clear supernatants were collected as cellular protein extracts, aliquoted, and stored at −70°C. The amount of protein was quantified by the bicinchoninic acid method (Pierce, Rockford, IL).

Reducing SDS-PAGE (10%) was performed as previously described (19), using 20 μg of cellular protein extract in each sample. After electroblotting, the nitrocellulose membrane was blocked with 5% fat-free milk in TBS for 2 h at room temperature, followed by overnight incubation at 4°C with primary Ab in 5% BSA in TBS with 0.05% Tween. The Abs used were anti-CREB mAb, (400 ng/ml), anti-phosphorylated CREB Ab (100 ng/ml), anti-c-Jun Ab (100 ng/ml), and anti-c-Fos Ab (100 ng/ml) (all from Santa Cruz Biotechnology, Santa Cruz, CA). The membrane was washed three times with TBS/Tween and incubated with a 1/10,000 dilution of secondary Ab (goat anti-mouse IgG-HRP, rabbit anti-goat IgG-HRP, or goat anti-rabbit IgG-HRP; all from Santa Cruz Biotechnology) in blocking buffer for 1 h at room temperature. After washing four times with TBS/Tween and once with TBS, the membrane was dried and the protein bands were visualized by ECL (Amersham Pharmacia Biotech, Buckinghamshire, U.K.). In some cases, the same blot was stripped with Western blot stripping buffer (62.5 mM Tris-Cl, pH 6.8, containing 2% SDS and 100 mM 2-ME), washed, blocked, and reblotted with a different Ab.

To investigate phosphorylation of CREB, immunoprecipitation with anti-CREB was performed before Western blotting. Eighty micrograms of CD8⁺ T cell lysates were prepared as outlined above, and precleared by incubation with protein G-Sepharose beads (Sigma-Aldrich) at 4°C for 1 h with rotation. After centrifugation at 1800 × g for 3 min, the supernatants were collected and incubated with 2 μg/ml anti-CREB mAb (Santa Cruz Biotechnology) for 2 h. Fifteen microliters of protein G-Sepharose beads were added and incubated overnight at 4°C with rotation. The beads were washed with 1% Nonidet P-40 cell lysis buffer three times. Supernatants were removed, SDS-PAGE loading buffer was added, and the samples were boiled for 5 min. SDS-PAGE and Western blotting were then performed, as outlined above.

Because CD8⁺ T cells are thought to contribute to the elimination of M. tuberculosis through production of IFN-γ (20, 21), we first evaluated the effect of CD40LT on IFN-γ production by M. tuberculosis-reactive CD8⁺ T cells. CD8⁺ T cells purified from PBL of four healthy tuberculin reactors were cultured with autologous monocytes and 10 μg/ml heat-killed M. tuberculosis for 6 days, with or without 5 μg/ml CD40LT. IFN-γ concentrations were then measured in cell culture supernatants. CD8⁺ T cells cultured with monocytes and heat-killed M. tuberculosis secreted low levels of IFN-γ (343 ± 209 pg/ml), but the addition of CD40LT increased IFN-γ levels 5-fold to 1733 ± 936 pg/ml (Fig. 1). This increase was not due to contaminating CD4⁺ T cells, because no CD4⁺ T cells were detectable by flow cytometry. CD8⁺ T cells generally recognize Ags that are secreted into the host cell cytoplasm. The mechanisms for recognition of Ags on heat-killed M. tuberculosis are uncertain, but may involve cross-priming of mycobacterial Ags by macrophages or dendritic cells in the culture.

To confirm that the findings above were relevant to CD8⁺ T cells exposed to live M. tuberculosis, we next cultured CD8⁺ cells from 14 healthy tuberculin reactors with autologous monocytes that were infected with live M. tuberculosis. The addition of CD40LT enhanced IFN-γ production by CD8⁺ T cells (1821 ± 588 pg/ml vs 404 ± 155 pg/ml, p < 0.01; Fig. 1). The addition of CD40LT to CD8⁺ T cells cultured with uninfected autologous monocytes yielded only 22–30 pg/ml IFN-γ. To determine whether the effect of CD40LT was Ag-specific, CD8⁺ cells isolated from PBMC of three healthy tuberculin-negative donors were cultured with autologous monocytes that were infected with live M. tuberculosis. The addition of CD40LT did not significantly enhance IFN-γ production (69 ± 52 pg/ml vs 28 ± 17 pg/ml, p = 0.49).

In four experiments, the addition of 10 μg/ml of a blocking anti-CD40 mAb (M2; Amgen) completely abrogated the capacity of CD40LT to enhance IFN-γ production by CD8⁺ cells from healthy tuberculin reactors that were exposed to M. tuberculosis-infected monocytes, demonstrating that the effects of CD40LT were mediated through CD40/CD40L interactions (Fig. 2).

To determine whether CD40LT enhanced IFN-γ production by CD8⁺ cells by increasing the percentage of IFN-γ-producing cells, we used the ELISPOT assay to measure the frequency of CD8⁺IFN-γ⁺ cells. In experiments performed on 10 healthy tuberculin reactors, the addition of CD40LT to CD8⁺ T cells and M. tuberculosis-infected autologous monocytes increased the frequency of IFN-γ-producing cells from 7 ± 2 cells per 10⁵ cells to 17 ± 4 cells per 10⁵ cells (p = 0.01, Fig. 3,A). CD40LT also increased the size of the spots formed by IFN-γ⁺ cells (Fig. 3 B), suggesting that CD40LT increased the amount of IFN-γ produced per CD8⁺ T cell.

Binding of CD40L on T cells to CD40 on APCs elicits production of IL-12 and IL-18 (22, 23, 24), and IL-12 and IL-18 enhance production of IFN-γ by CD4⁺ T cells (25, 26). IL-15 also contributes to IFN-γ production by CD8⁺ cells in response to mycobacteria (27). Therefore, we hypothesized that CD40LT enhances IFN-γ production by CD8⁺ T cells through production of IL-12, IL-15, and/or IL-18. Concentrations of these cytokines were measured in supernatants of CD8⁺ T cells from nine healthy tuberculin reactors, after culture with M. tuberculosis-infected monocytes for 6 days, with or without CD40LT. IL-12 and IL-15 concentrations were always <1 pg/ml in the absence of CD40LT, and these values did not change significantly with the addition of CD40LT. IL-18 concentrations also did not increase with addition of CD40LT.

Cytokine concentrations in supernatants may not reflect their biologic activity as the majority of the cytokines may be bound to cellular receptors. As an alternative means to determine whether the effects of CD40LT are mediated through these cytokines, we added different combinations of neutralizing Abs to IL-12, IL-15, and IL-18 to CD8⁺ T cells from five healthy tuberculin reactors, cultured with M. tuberculosis-infected monocytes and CD40LT (Fig. 4). Anti-IL-12 had the greatest effect, reducing mean IFN-γ concentrations elicited by CD40LT by 65%. Anti-IL-18 reduced IFN-γ concentrations by 40%, and the combination of both Abs inhibited 82% of IFN-γ production induced by CD40LT. In contrast, anti-IL-15 did not affect IFN-γ production, either alone, or in combination with anti-IL-12 or anti-IL-18. The addition of IgG to CD40LT did not reduce IFN-γ production (data not shown).

The experiments above demonstrate that CD40LT enhances IFN-γ production by M. tuberculosis-reactive CD8⁺ T cells. To investigate the mechanism for this effect, we studied transcription factors that bind the IFN-γ promoter and enhance IFN-γ mRNA expression in T cells. One of these transcription factors is CREB (28), and reduced CREB expression is associated with decreased IFN-γ production in human tuberculosis (29). Phosphorylation of CREB enhances its capacity to up-regulate gene transcription (30) and we used Western blotting with a mAb specific for phosphorylated CREB to show that stimulation of CD8⁺ T cells from three healthy donors with PMA and ionomycin enhanced phosphorylation of CREB within 15 min (Fig. 5,A). Next, purified CD8⁺ T cells from five healthy tuberculin reactors were cultured with M. tuberculosis-infected monocytes for 72 h, total CREB was immunoprecipitated with anti-CREB mAb, and Western blotting was performed with anti-phosphorylated CREB Ab. CD8⁺ T cells cultured with uninfected macrophages had minimal amounts of phosphorylated CREB. Exposure of CD8⁺ T cells to M. tuberculosis-infected monocytes increased expression of phosphorylated CREB, and the addition of CD40LT further enhanced this expression (Fig. 5 B). This finding was not due to differences in the efficiency of immunoprecipitation between samples, as stripping of the blot and reprobing with a polyclonal anti-CREB Ab revealed equal amounts of CREB protein in each lane.

The transcription factor AP-1 is composed of dimers of members of the Jun and Fos families, and IL-12 and IL-18 markedly increase production and phosphorylation of c-Jun, enhancing binding to the IFN-γ promoter (25). Because the experiments above suggested that CD40LT enhanced IFN-γ production through production of IL-12 and IL-18, we studied the effect of CD40LT on expression of c-Jun in CD8⁺ T cells from five healthy tuberculin reactors cultured with M. tuberculosis-infected monocytes. CD40LT increased expression of c-Jun but not c-Fos in CD8⁺ T cells (Fig. 6).

Lysis of infected cells is a central function of CD8⁺ T cells. Therefore, we cultured CD8⁺ T cells from three healthy tuberculin reactors with autologous M. tuberculosis-infected monocytes for 6 days, with or without CD40LT. CD8⁺ cells were then used as effectors against M. tuberculosis-infected monocyte targets. The addition of CD40LT increased mean lytic activity from 25 ± 4% to 62 ± 11% (p = 0.04).

CD8⁺ T cells can lyse infected cells through production of cytotoxic granules that contain lytic proteins such as perforin and granulysin, or through the Fas/FasL pathway. We used immunolabeling to evaluate the effects of CD40LT on expression of these CTL effector molecules by CD8⁺ T cells from six to nine healthy tuberculin reactors cultured with M. tuberculosis-infected monocytes. CD40LT significantly increased the mean percentages of CD8⁺ T cells expressing granulysin and perforin (p = 0.04 and 0.01, respectively, Fig. 7). In contrast, CD40LT did not increase FasL expression.

The present study demonstrated that CD40LT enhanced the capacity of M. tuberculosis-responsive CD8⁺ T cells to produce IFN-γ and to lyse infected autologous mononuclear phagocytes. CD40LT increased the number of IFN-γ-producing CD8⁺ T cells, as well as the amount of IFN-γ produced per cell. Enhanced IFN-γ production was dependent on production of IL-12 and IL-18, and was mediated through up-regulation of the transcription factors CREB and c-Jun, both of which stimulate IFN-γ mRNA transcription by binding to the IFN-γ promoter (28, 31, 32). CD40LT increased the capacity of CD8⁺ T cells to lyse M. tuberculosis-infected monocytes, and enhanced CTL activity was associated with higher expression of perforin and granulysin, but not of FasL. It has recently been shown that under some circumstances, CD40 on CD8⁺ T cells is essential for CD4⁺ T cells to contribute to generation of CD8⁺ memory T cells (33). Therefore, the effects of CD40LT on CD8⁺ CTL function that we observed may be due to binding of CD40LT to CD40 either on APCs or on CD8⁺ T cells.

Most studies of the effects of CD40/CD40L interactions on CD8⁺ CTL responses have been performed in animals, using model Ags such as OVA (8, 9, 10, 11). These studies demonstrated that binding of CD40L on T cells to CD40 on APCs elicits production of IL-12, which in turn stimulates IFN-γ production by T cells (22, 23). However, the transcription factors that mediate IFN-γ production by CD8⁺ T cells have not been studied. In the current report, we provide evidence that CD40LT enhances the functional capacity of CD8⁺ CTL to respond to a bacterial pathogen in humans, and that this is mediated not only through IL-12, but also through IL-18. In addition, we delineated an intracellular signaling pathway through which CD40/CD40L interactions may elicit production of IFN-γ.

Although CD4⁺ T cells play a dominant role in immune defenses against M. tuberculosis in animal models (34), CD8⁺ T cell-deficient mice also have increased susceptibility to tuberculosis (1). CD8⁺ T cells are a major source of IFN-γ in the early pulmonary response to infection in mice (35) and production of IFN-γ is essential for CD8⁺ T cells to exhibit antimycobacterial effects (21). In humans, CD8⁺ T cells produce IFN-γ in response to mycobacterial Ags and lyse M. tuberculosis-infected monocyte-derived macrophages and alveolar macrophages (2, 3, 36, 37, 38, 39). Furthermore, in PBMC of tuberculosis patients with ineffective immunity, the frequency of CD8⁺IFN-γ⁺ cells and the capacity of CD8⁺ T cells to lyse M. tuberculosis-infected monocytes are reduced, compared with findings in healthy persons infected with M. tuberculosis (13, 40).

Because M. tuberculosis-induced IFN-γ production by human CD8⁺ T cells requires the presence of CD4⁺ T cells (13), we hypothesized that interactions between CD40L on activated CD4⁺ cells and CD40 on APCs enhanced IFN-γ production by M. tuberculosis-reactive CD8⁺ T cells. CD40/CD40L interactions can prime CD8⁺ CTL to peptide-pulsed cells in the absence of CD4⁺ cells (8, 9, 10), and contribute to generation of CD8⁺ T cell-mediated alloimmunity (41). However, the contribution of CD40/CD40L interactions to the capacity of CD8⁺ T cells to respond to microbial Ags depends on the pathogen and the experimental system used. CTL activity against adenovirus is abrogated in CD40L-deficient mice (42). In contrast, the primary CD8⁺ T cell response to lymphocytic choriomeningitis virus does not require CD40L (43), and the primary and memory CD8 cell recognition of Listeria peptides is unaffected by anti-CD40L Abs (44). CD4⁺ T cell help mediated by CD40L may be essential for infections in which the precursor frequency of CD8⁺ CTL is relatively low, whereas CD4⁺ T cell help may be dispensable when the precursor frequency is high. In support of this hypothesis, the CD8⁺ CTL response to OVA is dependent on CD4⁺ T cell help only when the frequency of CD8⁺ T cell precursors is low (45). A high precursor frequency of CD8⁺ CTL may produce enough of a critical cytokine such as IL-2 or IL-12 to overcome the need for CD4⁺ T cell help (45).

We found that CD40LT increased IFN-γ production by CD8⁺ T cells, and this effect was abrogated by neutralization of IL-12 and IL-18. Binding of CD40 on APCs to CD40L on T cells elicits production of IL-12 and IL-18 by APCs (22, 23, 24). IL-12 facilitates maturation of naive CD8⁺ cells to become Tc1 cells that lyse allogeneic targets (46, 47). Compared with CD4⁺ T cells, CD8⁺ T cells express a higher density of IL-18R and produce higher concentrations of IFN-γ when stimulated with IL-18 (48). IL-15 enhances the CD8⁺ CTL response against a wide variety of pathogens (27, 49) and is essential for maintenance of CD8⁺ memory cells (50). However, the capacity of CD40 ligation to elicit IL-15 production is controversial (51, 52). We found that IL-15 did not contribute to the capacity of CD40LT to enhance IFN-γ production by M. tuberculosis-reactive CD8⁺ T cells.

CD4⁺ and CD8⁺ T cells differ in the transcriptional pathways that control IFN-γ production (53), and these pathways have not been extensively investigated in CD8⁺ T cells that respond to a microbial pathogen. The control of IFN-γ gene expression is complex and is mediated by binding of multiple transcription factors to several regulatory elements in the promoter region. Activation-specific expression of IFN-γ by T cells is dependent on the proximal (−73 to −48 bp) element of the IFN-γ promoter, the activity of which is controlled by binding of CREB/activating transcription factor (ATF)-1 and AP-1/ATF-2 (31, 54). AP-1 is composed of dimers of the Jun and Fos family proteins, and c-Jun is essential for activation-induced transcription of IFN-γ by T cells in several experimental systems (28, 31). The effect of CREB-ATF on IFN-γ transcription is more controversial. Some investigators found that CREB inhibited IFN-γ transcription (31, 55). In contrast, others found that CREB-enhanced IFN-γ transcription (28, 29), and reduced CREB expression, paralleled decreased IFN-γ promoter activity in tuberculosis patients (29). In view of these findings, we determined whether CD40LT enhanced IFN-γ production by M. tuberculosis-reactive CD8⁺ T cells through the transcription factors CREB and AP-1. CD40LT markedly increased production of phosphorylated CREB by CD8⁺ T cells, and phosphorylation of CREB enhances binding to promoters with imperfect CREB response elements such as the IFN-γ promoter (56). Phosphorylated CREB also stimulates recruitment of the coactivator CREB-binding protein, which associates with RNA polymerase II complexes and facilitates transcription (30). CD40LT elicited production of the c-Jun, but not the c-Fos, component of AP-1. The combination of IL-12 and IL-18 markedly increases production and phosphorylation of c-Jun, enhancing binding to AP-1 sites of the IFN-γ promoter (25). Because we found that the capacity of CD40LT to enhance IFN-γ production by CD8⁺ T cells was dependent on the presence of IL-12 and IL-18, we speculate that this effect is mediated in part through c-Jun.

A critical function of CD8⁺ T cells is lysis of infected targets, and this capacity is markedly reduced in CD4-deficient mice that are infected with M. tuberculosis (57). We found that CD40/CD40L interactions enhanced the CTL activity of CD8⁺ T cells, providing a potential mechanism for this interaction between CD4⁺ and CD8⁺ T cells. CD40LT greatly increased expression of perforin and granulysin, but not that of FasL, by CD8⁺ T cells, consistent with prior reports that human CD8⁺ T cells lyse M. tuberculosis-infected monocytes primarily through the granule exocytosis pathway, whereas apoptosis through Fas and FasL interactions plays a minor role (17). This pathway is likely to be important in immune defenses against tuberculosis as granulysin has direct antimicrobial activity against M. tuberculosis, and the combination of perforin and granulysin results in killing of intracellular M. tuberculosis (5). Furthermore, although there is no known murine homologue of granulysin, T cells in the lungs of M. tuberculosis-infected mice express perforin in vivo and lyse M. tuberculosis-infected macrophages in a perforin-dependent manner (20).

CD40LT has the potential to act as an adjuvant for vaccines that stimulate CD8⁺ CTL responses. This strategy is particularly attractive for patients with reduced numbers of CD4⁺ T cells and/or defects in CD4⁺ T cell function, such as those with HIV infection. Stimulation through CD40/CD40L interactions can enhance the CD8⁺ CTL response and provide substantial protection against tumors, viruses, bacteria, and parasites in animal models (58, 59, 60, 61). In addition, the current and prior studies show that CD40LT enhances CD8⁺ CTL activity against M. tuberculosis-infected monocytes and HIV peptides, respectively (62). Further studies are warranted to evaluate the feasibility of using CD40LT to enhance CD8⁺ T cell responses.