Multiple innate signaling pathways cooperate with CD40 to induce potent, CD70-dependent cellular immunity

Abstract

We have previously shown that Toll-like receptor (TLR) agonists cooperate with CD40 to generate CD8 T cell responses exponentially larger than the responses generated with traditional vaccine formulations. We have also shown that combined TLR agonist/anti-CD40 immunization uniquely induces the upregulation of CD70 on antigen bearing dendritic cells (DCs). In contrast, immunization with either a TLR agonist or a CD40 stimulus alone does not significantly increase CD70 expression on DCs. Furthermore, the CD8(+) T cell response generated by combined TLR agonist/anti-CD40 immunization is dependent on the expression of CD70 by DCs, as CD70 blockade following immunization dramatically decreases the CD8 T cell response. Here we show that other innate pathways, independent of the TLRs, can also cooperate with CD40 to induce potent, CD70 dependent, CD8 T cell responses. These innate stimuli include Type I IFN (IFN) and alpha-galactosylceramide (alphaGalCer) or aC-GalCer, glycolipids that are presented by a nonclassical class I MHC molecule, CD1d, and are able to activate NKT cells. Furthermore, this combined IFN/anti-CD40 immunization generates protective memory against bacterial challenge with Listeria monocytogenes. Together these data indicate the importance of assessing CD70 expression on DCs as a marker for the capacity of a given vaccine formulation to potently activate cellular immunity. Our data indicate that optimal induction of CD70 expression requires a coordinated stimulation of both innate (TLR, IFN, alphaGalCer) and adaptive (CD40) signaling pathways.

Figure 1

Type I IFN and α-C-GalCer work in combination with antiCD40 to generate enhanced CD8 T cell responses. Wild type C57/BL6 mice were immunized with recombinant IFN (4X10⁶ Units), αGalCer (2ug), anti-CD40 (50ug), IFN/anti-CD40, or αGalCer/anti-CD40. Seven days later the specific CD8 T cell response was analyzed by tetramer stain in the spleen for percentage (A, B) and total numbers (C) of CD8+ Tetramer+ T cell responses. Each bar in B and C is the average of 3 mice per treatment, with the error bars indicating the standard deviation, and is representative of at least 4 experiments performed. D) Wild type mice were immunized with IFN, α-GalCer or both. Seven days later the CD8 T cell response was measured in the spleen as in A. The data is representative of 2 independent experiments.

Figure 2

IFN/antiCD40 and αGalCer/antiCD40 and TLR/antiCD40 elicit CD8+ T cell responses via non-overlapping innate signaling pathways. A–C) Wild type B6, MyD88^−/−, IFNαR^−/−, IFNγ^−/−, and Jα18^−/− mice were immunized with the various combinations of IF, αGalCer, or both as described in Figure 1. The T cell responses were measured by tetramer stain of spleen cells, as described in the Materials and Methods, six days after immunization. Total numbers of tetramer staining CD8+ T cells was calculated also as described in the Materials and Methods. Each bar is the average of 2–3 mice with the error bar indicating the range or standard deviation, respectively. The data shown is representative of 2–4 experiments. D) B6 mice were injected with either αGalCer alone or in combination with antiCD40, as marked. These injections were given either without or with 400ug of the antiCD40L antibody MR-1. Six days later, the percent of tetramer staining CD8+ T cells out of total CD8+ T cells in the spleen were calculated by FACS analysis. Each bar is the average of 3 mice with the error bar indicating the standard deviation. The data shown is representative of 2 experiments.

Figure 3

IFN and αGalCer work in combination with anti-CD40 to upregulate CD70 DCs. Mice were immunized with IFN, aGalCer, antiCD40 and their varying combinations as described in Figure 1. The zero time point indicates uninjected controls. At 12, 24, 36 hrs, DCs were harvested as previously described (23) and stained to identify the expression of the given TNF ligands on either CD8+ or CD11b+ DC subsets. Representative FACS plots showing both DC subsets following either kinds of injections are shown in A and B. C and D) The Y axis indicates the fold increase in the MFI of the expression in the given TNF ligand over the uninjected control from 3 independent experiments, with error bars indicating the standard deviation.

Figure 4

CD8 T cell responses generated by IFN/anti-CD40 or NKT glycolipid/CD40 are CD70-dependent. A) B6 mice were immunized with 2ug αGalCer or 2ug of the variant αC-GalCer alone or in combination with antiCD40. Six days later, antigen specific T cell responses in the spleen were measured by tetramer stain. The data shown is representative of independent experiments with 2–3 mice per immunization per experiment. B and C) Mice were immunized as described in Figure 1 with (A) combined IFN/CD40 (4×10⁶ units IFN) or (B) αC-GalCer/anti-CD40. The day before and the day of immunization, the mice were injected IP with 250ug blocking antibodies to the indicated TNF ligand family member. Control mice were injected with PBS (No Ab). Seven days after immunization, CD8 T cell responses in the spleen were analyzed by tetramer stain as described in Figure 1. Bar graphs indicate the average of 2 mice per treatment with the error bars representing the range of responses observed. The data shown is representative of at least 3 experiments performed.

Figure 5

Combined IFN/anti-CD40 and αC-GalCer/anti-CD40 immunization generate superior immune memory. Mice were immunized against ovalbumin with the indicated combinations of IFN, αC-GalCer, and anti-CD40 as described in Figure 1. Seven days later the primary response was analyzed via tetramer stain from peripheral blood as described above. 230 days later, the mice were boosted with 5×10⁶ pfu of Vaccinia-ova and the secondary ova-specific T cell responses in the peripheral blood measured 5 days later via tetramer staining. Graphs indicate the average percent tetramer staining cells out of total CD8+ T cells from 3 mice per immunization. Data is representative of 2 experiments performed.

Figure 6

IFN/anti-CD40 synergy can protect against challenge with Listeria monocytogenes. Mice were immunized against ovalbumin in combination with IFN, antiCD40, or both, as shown. Two doses of antiCD40 were used; a low dose (1ug) that does not synergize to promote exponential CD8+ T cell expansion (not shown), and a high dose (50ug) that does synergize with IFN to promote exponential CD8+ T cell expansion as shown in Figure 1. Mice were challenged 75 days later with 2×10⁵ CFU LMova and three days after LM challenge, the bacteria burden was determined by plating liver homogenates on BHI agar. The data shows the average between 2 mice per treatment, error bars indicating the range between the two mice, and is representative of 2 experiments performed.