The current study demonstrates an important role for Nod1/Nod2 recognition of B. anthracis for induction of both innate and adaptive immune responses. We show that Nod1/Nod2 was involved in detection of B. anthracis after spore infection for the production of proinflammatory mediators, such as IL-1 and CCL5 (Fig. ), as well as IL-12p70 (Fig. ). In addition, Nod1/Nod2 was required for optimal induction of protective immunity after aerosol exposure (Fig. ). Overall, our results illustrate the involvement of Nod1/Nod2 in activation of the immune responses to anthrax infection.
Nod1 and Nod2 are cytosolic host proteins involved in recognition of peptidoglycan substructures of bacteria, including Bacillus
). B. anthracis
has been shown to replicate in the host cytosol, where Nod1 and Nod2 proteins are expressed (9
). After aerosol exposure, B. anthracis
spores are taken up by phagocytes in the lung, which become the first defenders against anthrax infection (4
). The results obtained in the current study show that macrophages and pulmonary cells responded to spore exposure with the production of proinflammatory cytokines, and responses were significantly reduced in Nod1/Nod2-deficient cells (Fig. ). The percentage of AMac in Nod1/Nod2 KO mice is lower in 1-h pulmonary adherent cells than in BL/6 mice; however, the total percentage of phagocytes (interstitial macrophages, DC, and AMac) is not significantly different between WT and KO mice. As phagocytic cells other than AMac have been shown to phagocytose spores, it is likely that all cultured phagocytes produced cytokines in response to spore infection. Thus, the difference in the percentage of AMac between WT and Nod1/Nod2 KO mice was unlikely to affect the observed cellular responses. In addition, after 24 h of culture, there was no significant difference in the phenotype of the pulmonary adherent cells (Fig. ). The differences between the phenotypes observed at 1 h and 24 h is likely due to the proteolytic cleavage of surface proteins during the lung dissociation step.
Spores that were able to germinate induced significantly higher levels of proinflammatory mediators than after stimulation with irradiated spores, indicating that vegetative bacilli contribute significantly to stimulation through Nod1/Nod2. This was also observed with the production of IL-12p70, a cytokine involved in adaptive immune responses (Fig. ). Using a different system, it was previously shown that inactivated spores induce NF-κB activation dependent on Nod1/Nod2, although the relevance of these findings was not explored (18
). Though spore antigens may be able to induce activation through Nod1/Nod2, our results indicate that intracellular vegetative bacilli induce much higher levels of proinflammatory mediators than inactivated spores (Fig. and ). The studies presented here were performed using a nonencapsulated, toxigenic strain, though it is unlikely that the spore coat differs significantly from that of the fully virulent strains of B. anthracis
. As Nod1/Nod2 recognize peptidoglycan motifs on the surface of B. anthracis
, it is possible that the capsule shields these motifs from recognition by host innate receptors. Future work using an encapsulated strain with in vitro assays is aimed at addressing this issue.
Previous work has implicated anthrax toxins in subverting host immune responses. The majority of these studies have been done using purified toxins with in vitro assays (1
). In the current study, in vivo challenges were performed with a toxigenic strain of B. anthracis
. In these experiments WT BL/6 mice were able to mount a protective immune response, but Nod1/Nod2 KO mice were not. This indicates the important contribution of Nod1/Nod2 signaling during the course of infection to the induction of protective immune responses, even in the face of toxin expression. TLR2 expression was not required during priming responses, as TLR2 KO mice were not susceptible to secondary lethal challenge.
It has previously been shown that murine macrophages produce CCL5 in response to Nod1/Nod2 agonists (57
), and in our studies, CCL5 was one of several cytokines affected by the loss of Nod1/Nod2 in response to B. anthracis
spore exposure (Fig. ). CCL5 is involved in recruitment of immune cells, aside from neutrophils, to sites of infection (37
). Although we examined responses of lung antigen-presenting cells to spore exposure, during inhalational anthrax infection it is likely that proinflammatory cytokine secretion occurs primarily in draining LNs, as phagocytes migrate to the draining LNs after uptake and recognition of B. anthracis
). Vegetative bacilli are found in the draining LNs, but not the lungs, early after aerosol spore exposure, and TNF-α levels in draining LNs of infected mice are significantly higher than lung TNF-α levels (34
). Further evidence supporting the unlikelihood of chemokine secretion in the lungs is provided by the lack of inflammation and cell infiltration into the lungs of moribund animals suffering from inhalational anthrax (34
To date, few data are available on the mechanistic involvement of Nod1/Nod2 in induction of adaptive immunity. Previous work by Fritz et al. has shown altered levels of antigen-specific Ig production in Nod1 KO mice after ovalbumin immunization as well as Helicobacter pylori
). Our results indicate an impaired induction of antibody responses following pulmonary infection with B. anthracis
in Nod1/Nod2 KO mice. The susceptibility of Nod1/Nod2 KO mice to secondary challenge was likely due to a failure to mount a rapid anamnestic antibody response (Fig. and ). Cellular expansion in the draining LNs of Nod1/Nod2 KO mice was severely decreased after primary aerosol challenge (Fig. ). This could be due to decreased proliferation of antigen-specific cells in the draining LN, as well as decreased infiltration of other immune cells. Proinflammatory cytokine levels after spore infection were significantly decreased in cells from Nod1/Nod2 KO mice, including the chemokines KC and CCL5 (Fig. , , and ). CCL5 expression, which can recruit macrophages, T cells, and immature DC to sites of infection, occurs after Nod1/Nod2 signaling (37
), and KC is involved in neutrophil recruitment. IL-12p70 induces the differentiation of CD4+
T cells into Th1 cells, and the deficiency of Nod1/Nod2 KO cells in producing IL-12p70 may be involved in the enhanced susceptibility of these mice to secondary challenge. Thus, the decrease in cell numbers in the draining LNs of Nod1/Nod2 KO mice may be attributed to decreased chemokine production and subsequent cell infiltration after aerosol exposure or to the decreased expansion of T cells during the primary stage of infection. Further work is warranted to determine if either of these defective responses is involved in the enhanced susceptibility of Nod1/Nod2 KO mice to secondary anthrax exposure. Additional approaches, such as adoptive transfer of cells, are required due to possible congenital defects in Nod1/Nod2 KO mice, namely, differences in the percentage of basal CD8+
T cells in the cervical LNs.
There is mounting evidence that innate immune recognition of invading pathogens instructs antigen-specific immunity, though the precise mechanisms by which this occurs have not been completely defined (38
). Our results indicate that the susceptibility of Nod1/Nod2 KO mice to rechallenge appears to be due to inadequate activation of adaptive immune defenses after primary aerosol exposure. This is likely due to the impaired activation of the innate response, which typically triggers proper activation of the adaptive response. Future and ongoing work is directed toward identifying and characterizing the mechanisms by which Nod1/Nod2 recognition leads to induction of adaptive immune responses.