key: cord-0000541-1xacvavs
authors: Solodova, Evgenia; Jablonska, Jadwiga; Weiss, Siegfried; Lienenklaus, Stefan
title: Production of IFN-β during Listeria monocytogenes Infection Is Restricted to Monocyte/Macrophage Lineage
date: 2011-04-11
journal: PLoS One
DOI: 10.1371/journal.pone.0018543
sha: 27548472c288a185911a8907c927d18f3c9b0abe
doc_id: 541
cord_uid: 1xacvavs

The family of type I interferons (IFN), which consists of several IFN-α and one IFN-β, are produced not only after stimulation by viruses, but also after infection with non-viral pathogens. In the course of bacterial infections, these cytokines could be beneficial or detrimental. IFN-β is the primary member of type I IFN that initiates a cascade of IFN-α production. Here we addressed the question which cells are responsible for IFN-β expression after infection with the intracellular pathogen Listeria monocytogenes by using a genetic approach. By means of newly established reporter mice, maximum of IFN-β expression was observed at 24 hours post infection in spleen and, surprisingly, 48 hours post infection in colonized cervical and inguinal lymph nodes. Colonization of lymph nodes was independent of the type I IFN signaling, as well as bacterial dose and strain. Using cell specific reporter function and conditional deletions we could define cells expressing LysM as the major IFN-β producers, with cells formerly defined as Tip-DCs being the highest. Neutrophilic granulocytes, dendritic cells and plasmacytoid dendritic cells did not significantly contribute to type I IFN production.

Interferons (IFN) were first discovered over 50 years ago by Isaacs and Lindenmann [1] . Due to its complexity, the IFN system is still little understood and remains a subject of intensive research. Currently, three types of IFNs are distinguished -type I, type II consisting of only IFN-c, and the recently discovered type III also called IFN-l. Here, we will focus on type I IFN. This family of cytokines comprises a single IFN-b, more than 13 IFN-a, and several other less well characterized members [2, 3] . All these subtypes signal through a common cell surface receptor (IFNAR) to activate IFN-inducible genes that exert a wide range of effects central to innate and adaptive immunity [2] . As a rule, IFN-b is produced as the earliest of type I IFN initiating a cascade of IFN-a via autocrine and paracrine loops [4, 5] . In addition, during viral infections type I IFN may be induced in plasmacytoid dendritic cells (pDCs) also named ''natural interferon producing cells''. Due to constitutive expression of the transcription factor IRF-7, pDCs are able to immediately produce large amounts of IFN-a [6] . Nonviral pathogens (i.e., bacteria, protozoa, fungi and helminthes) may also induce type I IFN. Interestingly, in contrast to viral infections where IFNs are normally protective, in non-viral infections IFN production might be defensive or deleterious [7, 8] .

The prototype of a deleterious type I IFN response during bacterial infection is experimental listeriosis elicited by the grampositive rod-shaped intracellular bacterium Listeria monocytogenes that was discovered in 1926 (reviewed in [9] ). L. monocytogenes induces type I IFN synthesis via triggering a still uncharacterized pattern recognition receptor in the cytosol of host cells [10, 11] . Cyclic diadenosine monophosphate (c-di-AMP), secreted by L. monocytogenes, might be one of the cytosolic activators of type I IFN [12] . In addition, lymphotoxin-a might be involved in the triggering of type I IFN responses [13, 14] and lymphotoxin b receptor was shown to be crucially involved in the controlling of Listeria monocytogenes infection [15] .

Secretion of type I IFN increases susceptibility to Listeria by inducing apoptosis of T cells and macrophages [16] [17] [18] . In accordance, mice deficient of IFNAR are more resistant to Listeria and show decreased T cell apoptosis in the spleen during infection [17, 19, 20] . Moreover, type I IFN are able to down regulate IFN-c receptor thereby suppressing macrophage activation and increasing host susceptibility to Listeria infection [21, 22] .

Intravenously inoculated L. monocytogenes is taken up by the spleen and removed from the blood predominantly by mononuclear phagocytes in the marginal zone of the white pulp [23] . Myeloid cells, especially monocytes and macrophages are rapidly recruited to sites of bacterial infection and are required for initial control of L. monocytogenes infection [11, [24] [25] [26] . During bacterial infections, circulating monocytes are known to differentiate into tissue macrophages and dendritic cells (DCs) [27] . Additionally, inflammatory Gr1 + /Ly6C high monocytes are able to differentiate into tumor necrosis factor-a (TNF-a) and inducible nitric oxide synthase (iNOS) producing so-called Tip-DCs at sites of infection [11] .

According to recent publications, different myeloid cell populations were shown to be responsible for IFN-b production during Listeria infection. Stockinger et al. defined CD11b + CD11c 2 PDCA1 2 B220 2 macrophages to be the major IFN-b producers and pDCs were claimed to make no contribution [28] . On the other hand, Dresing et al. observed IFN-b exclusively in the myeloid cell population called Tip-DCs [29] . In the present study, we used a genetic approach to address the question which cells produce IFN-b during experimental listeriosis.

Previously, we generated a conditional mouse line, in which the IFN-b coding sequence is replaced by the reporter luciferase upon cell specific Cre expression [30] . Here we explored CD19cre, CD4cre and LysMcre mice to activate the reporter function in B cells, T cells, monocyte/macrophages and neutrophils. As reporter mice, we used mice that were heterozygous for the targeted mutation. This allowed IFN-b production from the functional wild type allele. To demonstrate the physiological relevance of IFN-b produced by the cell types mentioned above, we included mice homozygous for the conditional deletion and compared them with mice in which IFN-b was deleted in germ line. Overview of transgenic mice used in this paper is presented in Figure S1 .

Using these genetically modified mice, we highlight the importance of IFN-b during the deleterious action of type I IFN in the course of Listeria infection. In addition, the novel reporter mice revealed a maximum of IFN-b induction 24 hours post infection (p.i.) in spleen and, surprisingly, 48 hours p.i. in cervical and inguinal lymph nodes after high dose intravenous infection. Low dose infection followed the same pattern of IFN-b production, although with delayed kinetics. The luciferase signal in spleen and lymph nodes mirrored the presence of Listeria in these organs. Moreover, our results showed that LysM-expressing cells are the main producers of IFN-b during murine listeriosis, especially the cells previously defined as Tip-DCs. Finally, neutrophils, DCs and pDCs and did not significantly contribute to type I IFN production under our conditions of L. monocytogenes infection.

During Listeria monocytogenes infection, IFN-b is produced in spleen and lymph nodes Different strains of Listeria monocytogenes vary in their ability to activate type I IFN production [31] . Here we compared LO28 and EGDe, two commonly used strains of Listeria monocytogenes. First, we asked when and where type I IFN is induced. To address this question, we used a previously described IFN-b reporter mouse, which allows whole body in vivo imaging of IFN-b induction using firefly luciferase as a reporter [30] . Figure 1A shows induction of IFN-b in albino IFN-b +/Db-luc reporter mice after intravenous (i.v.) injection of Listeria monocytogenes LO28 and EGDe strains. At 24 hours post infection (p.i.), IFN-b induction occurred almost exclusively in the spleen. Interestingly, 48 hours p.i. a bright luminescent signal appeared in the cervical and in the inguinal lymph nodes, especially when using the LO28 strain. At this time, no production of IFN-b was detectable in the spleen any longer. Loss of the signal from the spleen 48 hours p.i. could be due to complete destruction of spleen together with the massive apoptosis of lymphocytes [32] . Quantitation of luciferase activity in the selected regions of interest (Fig. 1A , circles 1 and 2 for spleen and cervical lymph nodes, respectively) showed that LO28, induced a stronger signal compared to EGDe (Fig. 1B) , although colonization by bacteria was comparable (Fig. 1C) . This confirmed the findings by Reutterer et al. that LO28 is the most potent type I IFN stimulator amongst the commonly used Listeria strains [31] . Of note, there was no significant signal from the liver at any time point, although this organ is highly colonized by Listeria.

Systemic application of L. monocytogenes is used in many studies related to host-pathogen interaction [9, 33, 34] , and spleen and liver were defined as the target organs of these bacteria [23] . Little attention has been paid to lymph nodes until now. Our finding that IFN-b is produced in the lymph nodes lead us to test whether lymph nodes are also colonized by Listeria after i.v. infection. Indeed, already 24 hours p.i. we could observe bacteria in cervical and inguinal lymph nodes independent of the bacterial strain used (Fig. 1C) . Between 24 and 48 hours p.i., bacterial burdens in lymph nodes increased about 100 fold. Colony forming units of Listeria in spleens and livers also increased although not as dramatic.

To investigate the influence of type I IFN on lymph node colonization, we compared wild type C57BL/6, IFN-b 2/2 and IFNAR 2/2 mice. Like in spleen, a defect in the type I IFN system resulted in lower bacterial numbers (Fig. 2) confirming a detrimental role of these cytokines also in lymph nodes. These differences were only detectable at 48 hours p.i., consistent with the peak of IFN-b production at this time point.

To rule out whether lymph node colonization is specific and is not due to the high dose of infection, we first monitored albino IFN-b +/Db-luc reporter mice for IFN-b production after low dose Listeria monocytogenes infection (Fig. 3A) . Luciferase expression at 24 hours p.i. peaked in spleen. At 48 and 72 hours p.i. we observed the signal both in spleen and in cervical lymph nodes. At 96 hours p.i. the IFN-b production pattern was almost identical to the 48 hour time point after high dose infection (Fig. 1A) , although the overall luminescent signal was lower (note the different scales). Quantification of luminescence intensity in regions of interest ( Fig. 3B ) in spleen and cervical lymph nodes (Fig. 3A , selected circled areas 1 and 2, respectively) confirmed qualitative in vivo data presented in Fig. 3A .

As expected, colonization of lymph nodes paralleled IFN-b production (Fig. 3C) . These observations suggested that lymph node colonization by L. monocytogenes with subsequent IFN-b production is general phenomenon and does not depend on the bacterial dose, strain or type I IFN signaling.

pDCs do not contribute to the production of type I IFN during Listeria monocytogenes infection In response to a wide variety of enveloped DNA and RNA viruses (e.g., Influenza virus, vesicular stomatitis virus) as well as parasites (Plasmodium falciparum) and CpG oligonucleotides, IFN type I is produced by pDCs [35] . L. monocytogenes was shown after intragastric administration to activate pDCs in spleen and in mesenteric lymph nodes as detected by an increased expression of MHC II and CD86 [36] . By extrapolation, a contribution of pDC to the type I IFN response after systemic Listeria infection could be expected, although in vitro stimulation of splenic pDC with Listeria did not induce type I IFN [28] . To evaluate the contribution of pDC to the overall IFN production in vivo after Listeria monocytogenes infection, we performed depletion experiments. To this end, 24 hours prior to infection mice were intravenously injected with the depleting antibody anti-mPDCA-1. 24 hours p.i. spleens were analyzed for the presence of pDCs. Figure 4A demonstrates gating strategy for pDCs and depletion efficiency.

RT-PCR of spleen cells 24 hours p.i. showed that mRNA expression of IFN-b and total IFN-a were not altered regardless of pDC depletion (Fig. 4B) . Similarly, serum levels of these cytokines did not reveal any significant differences between mice depleted of pDCs or control mice (Fig. 4C) . Moreover, bacterial loads in spleens and livers of such mice were also similar (Fig. 4C) . Hence, we conclude that pDCs do not significantly contribute to type I IFN production and also do not have a physiological relevance with regard to colonization of spleen and liver during murine listeriosis.

To determine the contribution of other myeloid or lymphoid cell populations to the IFN-b production after L. monocytogenes infection, we employed tissue specific reporter mice since these mice allow tissue specific replacement of the IFN-b gene by the luciferase reporter in the IFN-b +/floxb-luc mice [30] . These mice were crossed with mice expressing Cre recombinase under various tissue specific promoters. CD19cre mice were used to implement the reporter activity in B cells, CD4cre mice in T cells and LysMcre mice in monocytes/macrophages and granulocytes. All reporter mice were heterozygous for both markers to allow normal cellular development as well as potential production of IFN-b in all cells. Comparison of such tissue specific reporter mice after Listeria infection with the ''global'' IFN-b reporter mouse (IFN-b +/Db-luc ) by whole body in vivo imaging demonstrated that T-and B-cells did not have any apparent impact on the IFN-b production after infection by L. monocytogenes (Fig. 5A) .

Moreover, about 80% of pDCs are supposed to exhibit recombination in CD4cre mice [37] . Nevertheless, these mice did not show any luciferase production, confirming our data from above. Thus, we conclude that pDCs do not contribute to the type I IFN production during Listeria monocytogenes infection. Only cells, in which the LysM-promoter is active, significantly contribute to the IFN-b signal in spleen at 24 and in lymph nodes at 48 hours p.i. (Fig. 5A ). Of note, the bioluminescence signal in such mice is lower than in the mice displayed in Fig. 1A . Obviously, the signal is quenched by the black skin and fur of the recombinant C57BL/6 mice.

To overcome the restrictions of in vivo imaging with black mice, the results were corroborated by ex vivo quantification of luciferase activity (Fig. 5B ). Using this assay, we could confirm that T-and Bcells did not contribute to luciferase activity in spleen and lymph nodes at both time points. In contrast, LysMcre specific IFN-b reporter mice contributed with similar levels of luciferase activity compared to IFN-b +/Db-luc mice. Hence, LysM-expressing cells almost exclusively produce IFN-b in both spleen and lymph nodes upon infection with L. monocytogenes. Confirming the role of LysM-expressing cells during Listeria monocytogenes infection

To confirm the contribution of LysM positive cells to the type I IFN response against L. monocytogenes as well as to characterize their role during infection with this pathogen, we analyzed homozygous mice, in which the gene encoding IFN-b was cell-specifically deleted on both alleles. As a control, we included IFN-b floxb-luc/floxb-luc mice to demonstrate that the genetic changes in the targeted locus have no impact on their own on the course of infection. C57BL/6, IFNb floxb-luc/floxb-luc , IFN-b floxb-luc/floxb-luc x LysMcre and IFN-b 2/2 mice were infected with Listeria monocytogenes and 24 hours p.i. serum levels of type I IFN were determined by ELISA (Fig. 6A) . Importantly, IFN-b production was identical in C57BL/6 and IFN-b floxb-luc/floxb-luc mice, while the extent of IFN-b expression was dramatically reduced in IFN-b floxb-luc/floxb-luc x LysMcre mice. As a consequence, since type I IFN induction in macrophages depends on feedback signaling, total IFN-a levels were also markedly reduced in IFN-b floxb-luc/floxb-luc x LysMcre mice and they were similar to the levels in IFN-b 2/2 mice (Fig. 6A) .

Consistent with these results, the CFUs isolated from spleen and liver of C57BL/6 and IFN-b floxb-luc/floxb-luc mice did not differ significantly, while IFN-b floxb-luc/floxb-luc x LysMcre and IFN-b 2/2 mice showed strongly reduced bacterial numbers in the organs 24 hours after infection (Fig. 6B) . Moreover, comparison of IFNb floxb-luc/floxb-luc mice with IFN-b floxb-luc/floxb-luc x CD11ccre, where IFN-b production is ablated in dendritic cells, showed no significant differences in bacterial loads in spleen and liver, as well as in serum levels of type I IFN ( Figure S2 ).

Taken together, our results showed that preventing LysMexpressing cells from IFN-b production is sufficient to avoid the detrimental effects of type I IFN during L. monocytogenes infection. Thus, such cells are apparently the major source of IFN-b during murine listeriosis.

The LysM gene in mice is active mostly in monocyte/ macrophages and neutrophils [38, 39] . Neutrophils are essential during the early innate response to Listeria monocytogenes and are known to be capable to produce various cytokines, such as IL-1b, IL-6, TNF-a, IL-12, MIP-1a, MIP-1b, IL-18 and reactive oxygen intermediates in response to different stimuli [40] [41] [42] [43] . However, the contribution of neutrophils to type I IFN production is still unclear. Therefore, we wanted to determine whether neutrophils are involved in IFN-b production after L. monocytogenes infection.

To answer this question, we depleted granulocytes 24 hours prior infection after carefully titrating the antibody. Depletion of neutrophils was close to complete (Fig. 7A) , while changes in other cell populations were negligible ( Figure S3 ). Ex vivo analysis of luciferase activity in IFN-b +/Db-luc mice depleted of neutrophils revealed, that these cells did not have a significant impact on luciferase expression (Fig. 7B ). In addition, similar amounts of IFN-b and total IFN-a in serum were found in C57BL/6 mice despite of the neutrophil depletion (Fig. 7C) . Hence, LysM-positive monocyte/macrophages are the population that is responsible to almost entirely produce type I IFN after Listeria monocytogenes infection.

It is known, that the LysM promoter is active not only in monocyte/macrophages and neutrophils, but also in part of the CD11c + DCs, while no expression is observed in T and B cells [44] . Since it was recently shown that CD11b + CD11c int Ly6C + cells previously defined as Tip-DCs are the major IFN-b producers upon Listeria infection [29] , we wanted to know whether Tip-DCs belong to LysM-expressing monocyte /macrophage population and whether we could confirm the production of IFN-b by such cells. Therefore, we sorted CD11b + CD11c 2 macrophages, CD11c + CD11b 2 dendritic cells (DCs), CD11b + CD11c int Ly6C + Tip-DCs and B220 + CD11c 2 B cells as a negative control from IFN-b +/floxb-luc x LysMcre mice 24 hours after Listeria infection and tested for LysM, LysMcre as well as IFN-b expression by RT-PCR. A strong signal was observed for LysM in Tip-DCs and macrophages and, to a lesser extent, in dendritic cells (Fig. 8) . Cre recombinase was expressed in macrophages and in Tip-DCs, but not in DCs. IFN-b mRNA was highly pronounced in Tip-DCs and weakly in DCs. Faint bands were also detectible for IFN-a in these cells. In contrast, CD11b + CD11c 2 macrophages failed to produce any detectable amount of IFN-b or IFN-a. Taking together, our results indicate that cells defined as Tip-DCs are within LysM-expressing cells and are responsible for the major IFN-b production upon Listeria monocytogenes infection.

Type I IFN are extremely pleiotropic cytokines. For instance, they are involved in defense against many viruses [45] [46] [47] , could be either protective or detrimental during non-viral pathogens [7, 8] , act in cancer surveillance [48, 49] , but they are also effector molecules in toxic shock elicited by LPS [50] and TNF [51] . In addition, recombinant type I IFN is used in the clinics as therapy against certain cancers as well as for treatment of chronic infections caused by HBV [52] , HCV [53] and multiple sclerosis [54] . Thus, these molecules could be of great health benefit but could also elicit serious detrimental effects. Therefore, knowledge about the cells involved in production of type I IFN and their regulation is of utmost importance for the understanding of the IFN system.

Listeria monocytogenes is an intracellular pathogen, broadly used as a model to investigate host-pathogen interactions. Different groups have shown the production of type I IFN in vitro and in vivo upon infection with Listeria monocytogenes [10, 11] . We followed IFN-b induction in the mouse using non-invasive whole body in vivo imaging. This way we derived a spatiotemporal resolved picture of the early IFN-b response during Listeria infection. We focused on production of IFN-b, because IFN-b appears to be the primary member of type I IFN family that is induced after infection with Listeria [16, 17] and initiates the cascade of type I IFN in most cell types [4, 5] .

By a genetic approach we investigated the cells that are responsible for IFN-b production during murine listeriosis. We could show that IFN-b is almost exclusively produced by LysMexpressing cells in the spleen as well as in colonized lymph nodes. LysM promoter is highly expressed in monocytes/macrophages and neutrophils [44] . However, the use of the LysMcre mouse does not allow distinction between myeloid cell populations, although lower expression in CD11c + cells was reported [55] .

Macrophages have been identified as one of the target cell population in the murine spleen [26] . Besides, macrophages play a major role in the early innate defense against Listeria monocytogenes via secretion of different cytokines [20, 56] . However, DCs were also observed to be directly infected by L. monocytogenes [57] . Since IFN induction requires intracellular Listeria, we would have expected that DCs, in addition to monocytes/macrophages, would be involved in IFN-b production. This is apparently not the case, because i) only negligible amounts of IFN-b are produced in infected mice, in which IFN-b is conditionally deleted by LysMcre. This also excludes a major contribution of a non-lymphoid tissue associated cell. ii) Similar amounts of reporter luciferase are produced in the global reporter mice and in the mice reporting production of IFN-b by monocytes/macrophages and neutrophilic granulocytes. Although LysMcre is supposed to be expressed by a small population of DCs [44] , CD11ccre mice experiments confirm that conventional DCs do not significantly contribute to IFN-b production. We also could exclude the contribution of pDC. The depletion of this cell population had no effect on the overall production of type I IFN. This extends the in vitro data of Stockinger et al. [28] to the in vivo situation. Finally, the involvement of neutrophils that are expressing LysMcre was also excluded by a depletion experiment.

Stockinger et al. [28] defined cells with surface antigens characteristic of macrophages as major producers of type I IFN during Listeria infection. According to the recent publication of Dresing et al. [29] , TNF-a and iNOS producing Tip-DCs, that share markers with macrophages, were the main cellular source of IFN-b in the course of Listeria infection. Our data showed that a LysM-expressing cell population is responsible for IFN-b production in murine listeriosis. Testing sorted macrophages, DCs and Tip-DCs for type I IFN and LysM expression revealed that the LysM promoter was active in all three cell populations. However, only Tip-DCs were able to express vast amounts of IFN-b in agreement with Dresing et al. [29] . Interestingly, such Tip-DCs showed strong expression of LysM. This sheds some doubts on the DC nature of Tip-DCs, which was originally defined by their T cell priming capacity in mixed lymphocyte reaction [29] . Thus, Tip-DCs might as well be defined as inflammatory monocytes.

DCs might be involved only at the very early time points after infection and IFN-b production might be below our detection limit. We think this is unlikely given the extreme sensitivity of our reporter system. Alternatively, infected DCs might undergo apoptosis before they are able to significantly contribute to IFN production or IFN production in DCs might be inhibited by L. monocytogenes. The latter notion would be in agreement with the fact that pDCs also do not produce IFN-b.

A similar argument we would use for the absence of IFN-b production by hepatocytes. Such cells are known to be severely infected by L. monocytogenes during the complete course of infection. Nevertheless, no significant signal was observed in the reporter mice from liver. The absence of a contribution of Kupffer cells was expected since such cells are known to only bind Listeria but are not directly infected [58] . IFN-b induction requires cytosolic residence of Listeria.

The colonization of lymph nodes after i.v. infection with L. monocytogenes has been ignored so far, although it is known that such bacteria are able to infect lymph nodes after subcutaneous [59] or oral application [36] . Here we have shown that Listeria monocytogenes successfully colonizes cervical and inguinal lymph nodes also after i.v. infection. We noticed lymph node colonization due to the activation of the IFN-b reporter in such tissues. This impressively demonstrates the importance of using reporter systems that allow holistic unbiased observations. Colonization of lymph nodes could have taken place via dissemination of bacteria from the ''classical'' target organs spleen and liver. The delayed production of IFN-b in such organs would be in agreement with this notion. However, production of IFN-b correlated with the number of bacteria found in the colonized lymph nodes, which also showed a delayed increase. Thus, lymph nodes obviously are also primary target organs for L. monocytogenes, although they do not show the same penetrance as spleen and liver.

Together, our work compellingly shows that LysM-expressing Tip-DCs are the major producers of IFN-b during murine listeriosis, which is responsible for the induction of the type I IFN cascade in most cell types. The expression of LysM sheds some doubt on the DC nature of such cells. We also show in vivo that pDCs, the cell type that is able to produce type I IFN independent of IFN-b and neutrophils do not contribute at all to the overall production of type I IFN during murine listeriosis. Having settled this, the question now arises: why cDCs, that are infected at least early after i.v. application or hepatocytes that are known to be heavily infected by L. monocytogenes, do not produce type I IFN.

Mouse care and experimental procedures were performed under the approval of local authority Niedersä chsisches Landesamt für Verbraucherschutz und Lebensmittelsicherheit (LAVES). Permit numbers for this study are 33.11.42502-04-067/07 and 33.42502-071/06.

All mice were bred at the animal facility of the Helmholtz Centre for Infection Research (HZI) and maintained under specific pathogen-free conditions. For experiments, female mice 8 to 12 weeks of age were used. All mice used in this study were on the C57BL/6 background. IFN-b 2/2 [60] mice were backcrossed onto C57BL/6 for more than 15 generations. Conditional deletion/reporter mice IFN-b floxb-luc were generated using C57BL/6 ES-cells (Bruce4). To replace the IFN-b CDS by luciferase in germ line (IFN-b Db-luc ), IFN-b floxb-luc mice were crossed with K14cre mice [61] . To receive tissue specific IFN-b deletion as well as reporter expression, IFN-b floxb-luc were crossed to CD19cre [62] , CD4cre [63] and LysMcre [44] mice respectively (kindly provided by Dr. Angela Schippers, HZI). CD11ccre mice [64] were kindly provided by Prof. Dr. Ulrich Kalinke, Twincore. Additionally IFN-b +/Db-luc mice were crossed with albino C57BL/6 (C57BL/6-Tyr,c-2J.) kindly provided by Thomas Blankenstein (MDC, Berlin) to improve in vivo imaging.

Spleen cells were prepared by gently flushing the spleen with IMDM supplemented with antibiotics (100 U/ml penicillin and 100 mg/ml streptomycin), 10% FCS, 50 mM b-ME, and 2 mM Lglutamine. Erythrocytes were lysed for 2 min in ACK buffer (0.15 M NH 4 Cl, 10 mM KHCO 3 , and 0.1 mM EDTA) and cells were washed two times with PBS. Cell clumps were removed by passaging through a 50 mm nylon filter. Preparation was conducted strictly at 0uC.

Single cell suspensions were treated with anti-mouse CD16/ CD32 BD Fc Block (2.4G2, Becton Dickinson, NJ, USA) for 10 min followed by staining with appropriate mAbs for 20 minutes on ice. Abs used in this work included PE anti-mPDCA-1clone JF05-1C2.4.1 (Miltenyi Biotec), APC CD11c clone N418 (eBioscience, San Diego, USA), Pacific Blue CD11b clone M1/ 70.15 (Invitrogen), APC-Cy7 B220 clone RA3-6B2 (BD Pharmingen, New York, USA), anti-mouse Ly6G PE-Cy7 (Gr1, eBioscience, San Diego, USA), biotynylated anti-mouse Ly6C (Pharmingen, New York, USA), streptavidin APC-Cy7 (Pharmingen, New York, USA). Flow cytometric analysis and sorting was performed using LSRII (Becton Dickinson, NJ, USA. The data were analyzed using FACSDiva (Becton Dickinson) software.

Total RNA was extracted from spleen cells using RNeasy mini kit (Qiagen) according to the manufacturer's instructions. DNA contamination in the total RNA preparation was eliminated using DNase I (Qiagen). RevertAid First Strand cDNA Synthesis Kit (Fermentas) with oligo(dT) primers was used for reverse transcription of purified RNA. PCR was performed using the Promega kit according to the instructions of the supplier. 

Listeria monocytogenes strain LO28 serotype 1/2c [65] (kindly provided by Thomas Decker, Vienna, Austria) and EGDe serotype 1/2a [66] was grown in Brain Heart Infusion (BHI) infusion or on BHI-plates (Difco, Detroit, MI) at 37uC overnight. The next day, suspensions were diluted and grown until reaching log-phase. Bacteria were then centrifuged, washed several times and resuspended in sterile PBS. Concentrations of bacteria were determined by measurement at OD 600 and confirmed by plating serial dilutions on appropriate agar plates.

For the determination of the enzymatic activity of luciferase, cells were lysed in Reporter Lysis Buffer (Promega). For luciferase activity assays from tissue, weight of tissue fragments was determined and fragments were homogenized in proportional volumes of Reporter Lysis Buffer using Lysing Matrix A on a FastPrep-24 (MP Biomedicals). Lysates were mixed with LARII (Promega) and measured in a luminometer (Berthold). For in vivo imaging, mice were injected i.v. with 150 mg/kg of D-luciferin (Synchem) in PBS, anesthetized using Isofluran (Baxter) and monitored using an IVIS 200 imaging system (CaliperLS). Photon flux was quantified using the Living Image 3.2 software (CaliperLS).

Sera of infected mice were collected and levels of IFN-b and total IFN-a were measured by ELISA according to manufacturer's protocols (PBL InterferonSource).

Mice were infected intravenously (i.v.) with Listeria monocytogenes LO28 or EGDe strains. 5610 5 CFU/mouse is lethal dose (256LD 50 ), mice start dying at day 2-3 after infection. 2610 3 CFU/mouse is a sublethal dose (0.16LD 50 ), all mice are able to clear the bacterial around day 7 after infection. For determination of bacterial loads, spleens, livers and lymph nodes of sacrificed mice were removed and homogenized in 1 ml PBS supplemented with 0.2% NP-40. Serial dilutions of homogenates were plated on BHI agar plates and colonies were counted after overnight incubation at 37uC.

Plasmacytoid dendritic cells and granulocytes were depleted in vivo by intravenous injection of 100 mg anti-mPDCA-1 mAb (clone JF05-1C2.4.1, functional grade; Miltenyi Biotec) or 10 mg anti-Gr1 mAb (clone RB6-8C5) 24 hours prior Listeria monocytogenes infection. As an isotype control, rat IgG (Jackson ImmunoResearch Laboratories, Inc.) was used. Successful depletion of pDCs and granulocytes was determined by FACS analysis 24 hours after infection. Figure S1 Overview of transgenic mice used in this work. The generation of the global IFN-b reporter mouse has been previously described [30] and is since then maintained independent of cre expression. In brief, the targeted locus contains a luciferase gene (dark blue arrow) with a preceding polyA signal (black box) to avoid unspecific reporter activity. Two loxP sites (black arrowheads) allow cre-dependent replacement of the ifnb coding sequence by the luciferase reporter (light blue arrow) which is then driven by the endogenous ifnb promoter. In all reporter mice we keep one wt ifnb allel to allow IFN-b expression. Tissue specific reporter mice were obtained from breeding IFN-b floxb-luc/floxb-luc mice with mice expressing cre (red arrow) in a cell type specific manner. Cre activity (light red) in the given cell population (within the red circle) then allows ifnb promoter dependent reporter activity while the reporter cannot be activated in cells without cre expression (dark red). Tissue specific knockout mice carry two cre dependent alleles. Therefore in cells expressing cre both ifnb coding sequences are deleted. However, in cells without cre expression the IFN-b production is normal despite the genomic alterations introduced into the locus. 

Virus interference. I. The interferon. By A. Isaacs and J. Lindenmann, 1957

Interferons, interferon-like cytokines, and their receptors

Type I interferons (alpha/beta) in immunity and autoimmunity

A weak signal for strong responses: interferonalpha/beta revisited

Transformation of mouse fibroblasts alters the induction pattern of type I IFNs after virus infection

Plasmacytoid dendritic cells and type I IFN: 50 years of convergent history

The role of type I interferons in nonviral infections

The yin and yang of type I interferon activity in bacterial infection

Listeria monocytogenes, a unique model in infection biology: an overview

Innate recognition of bacteria by a macrophage cytosolic surveillance pathway

TNF/iNOS-producing dendritic cells mediate innate immune defense against bacterial infection

) c-di-AMP secreted by intracellular Listeria monocytogenes activates a host type I interferon response

Lymphotoxin-mediated crosstalk between B cells and splenic stroma promotes the initial type I interferon response to cytomegalovirus

A lymphotoxin-IFN-beta axis essential for lymphocyte survival revealed during cytomegalovirus infection

The lymphotoxin beta receptor is critically involved in controlling infections with the intracellular pathogens Mycobacterium tuberculosis and Listeria monocytogenes

Immune activation of type I IFNs by Listeria monocytogenes occurs independently of TLR4, TLR2, and receptor interacting protein 2 but involves TNFRassociated NF kappa B kinase-binding kinase 1

Type I interferon production enhances susceptibility to Listeria monocytogenes infection

IFN-beta increases listeriolysin O-induced membrane permeabilization and death of macrophages

Type I interferon sensitizes lymphocytes to apoptosis and reduces resistance to Listeria infection

Mice lacking the type I interferon receptor are resistant to Listeria monocytogenes

Induction of IFN-alphabeta enables Listeria monocytogenes to suppress macrophage activation by IFN-gamma

Antagonistic crosstalk between type I and II interferons and increased host susceptibility to bacterial infections

Early pathogenesis of Listeria monocytogenes infection in the mouse spleen

Neutrophils are essential for early anti-Listeria defense in the liver, but not in the spleen or peritoneal cavity, as revealed by a granulocyte-depleting monoclonal antibody

Listeria monocytogenes induces apoptosis of infected hepatocytes

Essential role of CCL2 in clustering of splenic ERTR-9+ macrophages during infection of BALB/c mice by Listeria monocytogenes

Development of monocytes, macrophages, and dendritic cells

Characterization of the interferon-producing cell in mice infected with Listeria monocytogenes

A fluorescence reporter model defines ''Tip-DCs'' as the cellular source of interferon beta in murine listeriosis

Novel reporter mouse reveals constitutive and inflammatory expression of IFNbeta in vivo

Type I IFN are host modulators of strain-specific Listeria monocytogenes virulence

Lymphocyte apoptosis during early phase of Listeria infection in mice

Innate and adaptive immune responses to Listeria monocytogenes: a short overview

Listeria monocytogenes infection in the face of innate immunity

Plasmacytoid dendritic cells: linking innate and adaptive immunity

Differential expansion, activation and effector functions of conventional and plasmacytoid dendritic cells in mouse tissues transiently infected with Listeria monocytogenes

Type I IFN-mediated protection of macrophages and dendritic cells secures control of murine coronavirus infection

Mouse lysozyme M gene: isolation, characterization, and expression studies

Repetitive sequence involvement in the duplication and divergence of mouse lysozyme genes

Cytokine expression by neutrophils and macrophages in vivo: endotoxin induces tumor necrosis factor-alpha, macrophage inflammatory protein-2, interleukin-1 beta, and interleukin-6 but not RANTES or transforming growth factor-beta 1 mRNA expression in acute lung inflammation

Experimental Listeria meningoencephalitis. Macrophage inflammatory protein-1 alpha and -2 are produced intrathecally and mediate chemotactic activity in cerebrospinal fluid of infected mice

Expression and regulation of human neutrophil-derived macrophage inflammatory protein 1 alpha

Phagocytes, O2 reduction, and hydroxyl radical

Conditional gene targeting in macrophages and granulocytes using LysMcre mice

Interferons and viruses: an interplay between induction, signalling, antiviral responses and virus countermeasures

Interferon type I responses in primary and secondary infections

Type 1 interferons and the virus-host relationship: a lesson in detente

Interferons, immunity and cancer immunoediting

Neutrophils responsive to endogenous IFN-beta regulate tumor angiogenesis and growth in a mouse tumor model

Central role for type I interferons and Tyk2 in lipopolysaccharideinduced endotoxin shock

Type I interferon drives tumor necrosis factor-induced lethal shock

Hepatitis B virus: from diagnosis to treatment

Intracellular innate immune cascades and interferon defenses that control hepatitis C virus

Interferon-beta therapy for multiple sclerosis-is the injection site the relevant action site?

Applications of myeloid-specific promoters in transgenic mice support in vivo imaging and functional genomics but do not support the concept of distinct macrophage and dendritic cell lineages or roles in immunity

Immunity to Listeria infection

Unique functions of splenic CD8alpha+ dendritic cells during infection with intracellular pathogens

Bacteria in the bloodstream are trapped in the liver and killed by immigrating neutrophils

Experimental validation of low virulence in field strains of Listeria monocytogenes

Interferon-beta is required for interferon-alpha production in mouse fibroblasts

Keratin 14 Cre transgenic mice authenticate keratin 14 as an oocyteexpressed protein

Impairment of T-cell-dependent B-cell responses and B-1 cell development in CD19-deficient mice

A critical role for Dnmt1 and DNA methylation in T cell development, function, and survival

Notch-RBP-J signaling controls the homeostasis of CD8-dendritic cells in the spleen

L. monocytogenes-induced actin assembly requires the actA gene product, a surface protein

Detection of listeriolysin, the thiol-dependent hemolysin in Listeria monocytogenes, Listeria ivanovii, and Listeria seeligeri

The authors wish to thank Martina Krey, Susanne zur Lage and Regina Lesch for excellent technical assistance.