The analysis of bacteria-host cell interactions at the molecular and cellular levels has become a major research area in recent years 
. Host defenses against intracellular pathogens such Lm
require the coordinated interactions of a number of innate and adaptive immune system components to clear the infection 
. Early activation of innate immune cells is important for host survival and bacterial clearance 
whereas development of adaptive immunity is crucial for long-term protection and for providing sterilizing immunity. However, it is becoming clear that in the context of Lm
challenge, innate immunity alone is unable to completely control the infection and the development of adaptive immunity is necessary to fully eliminate the pathogen.
The general requirement of DCs during Lm
infection has been studied by using different experimental models but the exact role of DCs in combating infection remains unclear 
. A subpopulation of CD11c+
cells has been implicated in various stages of Lm
immunity, including splenic sequestration 
, early activation of CD8+
Expression profiles have been widely used to determine the molecules involved in complex host-pathogen interactions 
Post-infection gene expression has been evaluated at short/middle time points (at 0 hr and at 1−8 hr p.i.) 
and with end-point assays 
. Nevertheless, no previous study has addressed the genome-wide DC-specific response during the first 24 hours of Lm
infection. Microarray analyses have shown that the IFNβ gene and interferon-stimulated genes (ISGs) are among the genes most highly induced by cytosolic Lm
in macrophages or in Lm
-infected mice 
Typically, bacteria activate type I IFN signaling through TLR-dependent mechanisms, which involve recognition of LPS from Gram-negative organisms, or via TLR-independent cytosolic receptors that respond to bacterial nucleic acids either endocytosed or secreted directly into host cells 
activates cell surface receptor and cytosolic receptor signaling pathways 
. Specifically, cytosolic bacteria trigger a unique cytokine response that includes production of type I IFNs 
Although several cytokines are important for immunity to Lm
infection, type I IFNs appear to be deleterious to the host 
however, opposing roles have been described 
. Whereas type I IFNs provide protection against viruses, their role in bacterial infection is less clear 
. In addition, the diverse effects of type I IFNs include increasing sensitivity to Lm
-induced cell death 
; inducing the downregulation of the IFNγ receptor gene (IFNGR) and thus rendering the host cells refractory to IFNγ, which is crucial to achieving host resistance to Lm
; and regulating the expression of chemokines important for leukocyte recruitment 
. And indeed, monocytes recruitment to Lm infected spleen is maintained in the absence of either MyD88 or IFNAR signaling. However, deficiency of both pathways impair cell migration and increased rather than decreased susceptibility to infection indicating that type I IFN has both positive and negative effects on the resistance to Lm infection 
In this study, we confirmed that IFNβ gene expression is strongly induced in DCs in response to Lm infection, and that the peak of this production (24 hr p.i.) is consistent with that described previously 
. Our data indicated that type I IFNs are not induced by Lm with the same early kinetics observed in vi
rus infections, which cause these cytokines to be produced within a few hours after infection.
This observation is compatible with the cytosolic detection of Lm bacteria by an as yet unknown intracellular receptor. We postulated that intracellular recognition as opposed to recognition by a cell surface receptor could lead to specific pathogen adaptations.
Thus, we sought to determine whether the delay in IFNβ production could provide a selective advantage for Lm
spreading during the very early phase of an activated innate immune response. We hypothesized that the timing of IFNβ production could alter the nascent host innate response, and for this reason, we studied the effect of the lack of IFNβ production at 5 hr p.i. To determine the impact of delayed IFNβ production in Lm
infection, we first performed a survival assay in which the mice were infected with Lm
with or without the administration of exogenous IFNβ. We demonstrated that at the very early phase of Lm
infection, treatment with IFNβ as early as 1 hr after bacteria inoculation allowed the mice to survive the infection (). These results suggest that delayed IFNβ production early during the initial phase of Lm
infection may provide a selective advantage for the bacterium because the innate immune system is not properly activated and/or regulated. Our findings are consistent with those of previous studies showing that cytokine delays indeed affect the outcome of both innate and adaptive immune responses 
. We propose that the presence of an early source of exogenous IFNβ exert an important immuno-regulatory role in that it enhances cellular recruitment (Tip-DC), activation/regulation (NK cells) whereas inhibits other cellular types that could cause excessive tissue damage 
The large amount of IFNβ induced by a lethal dose of Lm at 24 h p.i. when the inflammatory responses are fully activated and the bacterial burden has probably exceeded above the threshold level for survival, the effect of IFNβ on cellular apoptosis may conceal the beneficial effect of IFNβ induction. For this reason, we believe that our data are not in contradiction with previous reports on detrimental effect of type I IFNs because we examined the regulatory role of exogenous IFNβ at 5 h p.i. At this early time point of Lm infection the inflammatory response is not yet fully activated and the positive effect of IFNβ can be appreciated. And indeed, in these conditions the mice early treated with IFNβ are able to survive lethal infection.
We also examined IFNβ induction in DCs infected in vitro
by other Gram-positive bacteria. We found that, for example, Lactococcus lactis
is able to induce IFNβ production with a peak at 8 hr p.i. rather than at the 24 hr p.i. time point, as occurs in Lm
-infected DCs (Figure S5
). This data suggested that IFNβ production is differentially modulated in DCs during infections by different Gram-positive bacteria.
Mechanistically, we found that Lm
-infected DCs were not able to activate NK cells either in vitro
or in vivo
, both in terms of inducing IFNγ production and in stimulating the lysis of target cells, suggesting that innate immunity mediated by NK cells is unpaired at 5 hr p.i. in our model. NK cell activation could be reconstituted by adding IFNβ in the co-cultured system, suggesting that IFNβ is indeed required for early NK cell activation. Nevertheless, bacterial counts in the spleen of infected mice were only moderately reduced indicating that other mechanisms were operating in our model. Recent reports suggest that NK cells and a cell population that is CD11b+Ly6G+ exhibit tissue protective properties in the context of innate immunity, a function that could be enhanced by the presence of IFNβ 
Our results imply that IFNβ maybe able to induce disease tolerance by inducing tissue protection instead of disease resistance by reducing bacterial burden, a mechanism that has been recently suggested by Medzhitov et al. 
. The activation of the above mechanism should explain the observed increase in mice survival in the group of IFNβ treated animals. However, the exact mechanisms remain unknown and it is the focus of our ongoing studies.
In conclusion, the innate immune response is a complex interplay between cells and soluble factors that, in the right context and environment, together provide a framework to combat infection. However, successful pathogens have evolved mechanisms that subvert these ancient controls, allowing them to counteract the innate immune responses of their hosts.
Here we suggest that a transient impaired of NK cell activation/regulation may be an additional strategy employed by Lm to avoid innate immune activation and that IFNβ can regulate antibacterial activity in the very early phases of acute infection acting as a positive regulatory molecule.