TLR 9 Activation in Dendritic Cells Enhances Salmonella Killing and Antigen Presentation via Involvement of the Reactive Oxygen Species

Synthetic CpG containing oligodeoxynucleotide Toll like receptor-9 agonist (CpG DNA) activates innate immunity and can stimulate antigen presentation against numerous intracellular pathogens. It was observed that Salmonella Typhimurium growth can be inhibited by the CpG DNA treatment in the murine dendritic cells. This inhibitory effect was mediated by an increased reactive oxygen species production. In addition, it was noted that CpG DNA treatment of dendritic cells during Salmonella infection leads to an increased antigen presentation. Further this increased antigen presentation was dependent on the enhanced reactive oxygen species production elicited by Toll like receptor-9 activation. With the help of an exogenous antigen it was shown that Salmonella antigen could also be cross-presented in a better way by CpG induction. These data collectively indicate that CpG DNA enhance the ability of murine dendritic cells to contain the growth of virulent Salmonella through reactive oxygen species dependent killing.


Introduction
Dendritic cells (DCs) are specialized antigen presenting cells which constantly screen the environment for foreign particles [1] and engulf them via a variety of ways like phagocytosis, macropinocytosis, caveolin-mediated or clathrin-dependent endocytosis [2,3]. DCs function at the dividing line of innate and adaptive immunity [4] and regulate the T cell response [5,6]. DCs capture antigens in the peripheral tissues and present the processed antigen via the major histocompatibility complex (MHC) I and II receptors [7,8]. Salmonella in its turn is a very successful pathogen. It is a facultative intracellular pathogen and resides in macrophages and DCs by virtue of its pathogenicity island encoded virulence factors which are required for intracellular survival, replication and for the efficient colonization of deeper tissues. Salmonella is capable of causing symptoms ranging from self limiting diarrhea and localized gastrointestinal inflammation to the systemic typhoid fever. The mice model of infection mimics the pathogenesis of human typhoid fever.
Toll like receptors (TLRs) are germ line receptors expressed on DCs and recognize infectious agents through the various moieties present on them and act as a bridge between innate and adaptive immunity [9]. Their localization is determined by the nature of the ligand that they bind to. For instance, Toll like receptor-9 (TLR-9) is localized in the late endosomes or lysosomes, where it detects unmethylated CpG motifs in double stranded DNA. Ligand receptor engagement leads to the docking of adaptor molecules like MyD88 to TLRs and recruitment of proteins belonging to the IRAK family [10,11]. This ultimately leads to the NFkb activation and gene expression for the production of inflammatory cytokines like IL-6, IL-12 and TNF-a which lead to further recruitment of successive waves of immature DCs and monocytes to the portals of pathogen entry [12]. Therefore, TLR activation helps in mounting a more prominent T cell response, better killing of the pathogen and thus utilizing TLR signaling could be an effective strategy to clear the invading pathogen. The same has been shown in cultured hepatocytes where CpG treatment led to an increased TLR-9 expression and peroxide formation resulting in an enhanced killing of Salmonella [13].
Peroxide in a cell is formed by the NADPH oxidase, a multicomponent enzyme system responsible for releasing reactive oxygen species (ROS) in a process known as the respiratory burst [14,15]. A voluminous literature exists pertaining to the role of ROS in regulating antigen presentation. ROS mediated maintenance of the intracellular redox potential of macrophages and DCs is crucial for their antigen presenting capability. In addition, it is also reported that antigen specific bidirectional dendritic cell and T cell interaction is blocked by inhibiting NADPH oxidase due to inhibition of ROS productio. In addition, published data suggest that phagocyte NADPH oxidase prevents acidification of the phagosomes and thereby promote antigen cross presentation [6,[16][17][18][19][20][21]. Thus, it can be clearly concluded that by modulating the ROS profile of an infected host, its antigen presentation capacity can be altered.
In this study, we sought to examine whether CpG-DNA treatment can bring about the growth attenuation of Salmonella in DCs and to identify the mechanisms involved in the antibacterial response of CpG-DNA treated DCs. We further hypothesized that CpG induced ROS can enhance the antigen presentation capacity of the Salmonella infected DCs and can therefore alter the fate of the infection.

TLR-9 activation decrease Salmonella burden in DCs
To understand the role of TLR-9 in Salmonella infection, we infected dendritic cells with Salmonella in the presence of the TLR-9 ligand, CpG DNA. Previous report suggests that CpG DNA can change the conformation of TLR-9 and activate the downstream signaling [22]. In accordance with previous reports, we also found that in the untreated DCs, Salmonella load from 2 to 12 h of infection remained static. However, the data presented in Fig. 1A clearly indicate that the CpG DNA mediated activation of the TLR-9 post infection led to a significantly diminished Salmonella burden in DCs. The essential role of TLR-9 was further supported by the observation that the significantly reduced proliferation of Salmonella was not observed in the DCs wherein the TLR-9 signaling was inhibited by a specific TLR-9 antagonist. The antagonist treatment does not significantly increase the bacterial load even at 2 or 12 h of growth indicating Salmonella infection does not itself activate TLR-9 signaling. We further measured Salmonella proliferation in TLR-9 ligand pre-treated cells and also when TLR-9 ligand induction was performed along with the infection. In both conditions we observed a diminished burden as shown in Fig. 1B indicating that TLR-9 treatment at pre, post or along with the Salmonella infection leads to an enhanced killing of the bacteria. The ligand and the antagonist treatment do not lead to the killing of the dendritic cells. No decrease in cell number or viability was observed in any of the experimental conditions as indicated by MTT assay ( Figure S1). When we checked for the bacterial invasion in the treated cells, there was no difference in the uptake upon ligand or antagonist treatment (data not shown).

TLR-9 protein level is enhanced upon CpG treatment
To determine the effect of Salmonella infection or the CpG ligand treatment on TLR-9 expression, we performed western blot to determine the protein level of TLR-9 in DCs. CpG treatment has earlier been reported to increase the transcription of TLR-9 [23]. Further, it has been shown that DNA isolated from pathogenic Salmonella also leads to the upregulation of the TLR-9 protein expression [24]. Interestingly, in the same study it was observed that live Salmonella infection did not increase TLR-9 expression as the bacterial DNA might not have been accessible to the TLR-9 receptor. The data shown in Fig. 2 represent TLR9 expression. In accordance with previous reports; we also observed that Salmonella infection does not significantly increase the TLR-9 level in the infected DCs compared to the control cells. However, with the cognate ligand treatment alone, there was a 4-fold increase in the TLR-9. Also in the case of Salmonella infection along with the CpG treatment there was a significant increase in the TLR-9 level. These results clearly prove that CpG treatment allows Salmonella infected DCs to induce an enhanced level of TLR-9 expression, thus enabling a better clearance of the pathogen.
The mechanism of TLR-9 mediated anti Salmonella response In order to better understand the mechanism of CpG mediated Salmonella growth attenuation, we went ahead to measure the amount of ROS and nitric oxide (NO) production in the CpG treated DCs. Our results showed that there was no difference in the NO production between CpG treated and untreated Salmonella infected DCs (data not shown). We next went ahead to check whether the CpG induced antibacterial activity could be due to an enhanced ROS production. In accordance with our hypothesis, we indeed observed an enhanced ROS production in the infected DCs after CpG treatment (Fig. 3A). In the infected cells treated with CpG, the MFI was 2-fold more than the untreated cells for the DAF2DA staining (Fig. 3B). The TLR-9 antagonist treatment showed a reduced ROS production. To further evaluate our result, we next blocked the ROS production in the CpG treated DCs and examined Salmonella proliferation. Data presented in Fig. 4 indicate that the Salmonella attenuation with CpG treatment is no longer observed in presence of the ROS inhibitor PAO. Thus, our results clearly suggest that CpG mediated killing of Salmonella is mediated by an enhanced ROS production.
In resting DCs the ROS generating complex is disassembled and inactive. However, in mature DCs in response to various stimulations, the cytosolic (p47phox, p67phox, p40phox and Rac2) and the membrane (gp91 phox and gp22 phox ) components are assembled together to form a functional enzyme and release oxygen radicals [25][26][27]. As it was evident that ROS production in the CpG treated DCs was significantly higher than the control cells, we next examined whether there is any difference in the NADPH phagocytic oxidase subunit localization between the two conditions. After infecting DC with the WT-GFP bacteria for 12 h, cells were stained for gp91phox subunit of NADPH phagocytic oxidase. Interestingly, increased co-localization of the WT and gp91phox was observed in the CpG treated cell (Fig. 5B) as compared to the control (Fig. 5A). In the ligand treated cells, the gp91 phox or the ROS complex staining was distinctly located near the bacteria containing regions. There were enhanced colocalization with the bacteria and also the ROS complex was found to be in close viscinity to the bacteria. This difference in the distribution and colocalization of the complex itself suggests that CpG treatment leads to direct bacterial killing by inducing ROS and not by any other mechanism. Antagonist treatment produced similar staining pattern like the control cells (Fig. 5C). When various fields were counted, the percentage co-localization of gp91phox with the WT was 4% in the control DCs in contrast to an enhanced 19% in the CpG treated DCs (data not shown).
CpG mediated TLR-9 activation increases the capacity of DCs to process and/present Salmonella antigen in a ROS dependent manner There are contradictory reports of the effect of Salmonella infection in the antigen presentation by the dendritic cells. It has been previously reported that intracellular Salmonella affect the antigen presentation capacity of DCs. Salmonella infection in DCs inhibit T cell proliferation in a Salmonella pathogenicity island 2 and NO dependent manner. Interestingly, a recent report suggests that if Salmonella antigen is carried to the DCs by extracellular vesicles from infected macrophages a specific T cell response is generated [28]. In addition, injection of Salmonella-loaded DCs into mice result in both the CD4 + and CD8 + T cell activation specific for the model antigen used [29]. Thus, to better understand the effect of CpG on Salmonella antigen presentation in DCs, we performed mixed lymphocyte reaction (MLR). Our data revealed that Salmonella infected DCs from Balb/c mice do not effectively present antigen to the allogenic mixed T cell population from C57BL/6 mice spleen (Fig. 6A). There is actually a drop in the T cell proliferation in the Salmonella treated cells (Bar 6) than even the control cells (Bar 1). However, in the CpG DNA treated condition the antigen presentation and the T cell proliferation of the Salmonella infected DCs was significantly increased. It is also observed that only CpG treated cells (Bar 2) showed high antigen presentation. This finding indicates wild type Salmonella could inhibit antigen presentation from the infected DCs, but not from the CpG treated cells. TLR-9 antagonist treated DCs showed no difference in antigen presentation. We next substantiated our finding by measuring the IFN-c production of the proliferating T cells. The T cells added to Salmonella infected DCs along with TLR-9 ligand treatment produced significantly higher amount of IFN-c than without treatment (Fig. 6B). Since it is well known that ROS production affects antigen presentation in DCs we examined the effect of PAO during Salmonella infection. PAO treatment has no other adverse effect on DC death/survival ( Figure S1) or maturation ( Figure S2). Under PAO treatment (Bar 9), the CpG DNA mediated enhanced antigen presentation was no longer observed. Hence, it can be concluded that the CpG mediated enhanced ROS production allows better presentation of Salmonella antigen like any other antigen.

TLR-9 activation enhances the capacity of DCs to process and/present exogenous antigen in a ROS dependent manner
We further questioned whether the enhanced antigen presentation occurs with a purified exogenous antigen as well. For this purpose, we used the purified envelope glycoprotein haemagglutinin (H-antigen) of rinder-pest virus and specific CD8 + T cells that were generated earlier from H-2d restricted mice [30,31]. The process by which exogenous antigen get processed by the MHC I pathway and are presented to the CD8 + T cells is known as crosspresentation [32]. Antigens from intracellular pathogens like Salmonella have been shown to elicit a MHC class I dependent CD8 + T cell response and this process is also termed as crosspresentation. There are various explanations for cross-presentation. However, it has been convincingly shown that phagosomes are self sufficient organelles for the cross-presentation of any exogenous antigen [33]. In addition, it has also been reported that in the dendritic cells, endoplasmic reticulum and phagosome fusion leads to the generation of a specialized compartment capable of cross-presentation [8]. It is further known that TLR-3 and TLR-9 ligands induce cross-presentation in the dendritic cells [34]. Therefore, we asked whether CpG mediated TLR-9 activation could alter the cross-presentation capacity of the Salmonella infected DCs. Thus, we examined whether the exogenous antigen is taken up by the DCs and get presented MHC-I pathway. For this purpose, we treated the infected DCs with H-antigen and checked for the specific CD8 + T cell proliferation. In these sets of experiments a different trend of result was observed when compared to our MLR assay. Fig. 7A clearly indicates that H antigen cross-presentation gets significantly enhanced upon CpG DNA treatment. Although, Salmonella infection (Bar 7) did not enhance the antigen presentation capacity of the control cells (Bar 3), we did not observe a decrease in the antigen presentation after Salmonella infection like the MLR assay. However, interestingly we noticed that the H-antigen cross presentation is increased in the infected CpG treated cells (Bar 8). There was an enhanced antigen presentation in the control cells treated with CpG (Bar 4) but after Salmonella infection this was further enhanced. Salmonella could not downregulate the enhanced antigen presentation. Blocking the ROS production with PAO prevented this enhanced cross presentation clearly indicating the fact that the enhanced antigen presentation is actually mediated by CpG induced ROS production. These results indicate that if we use CpG it might lead to enhanced presentation of Salmonella antigen also. However, to confirm this finding, we need to perform the experiment with an assay system to detect specific Salmonella antigen cross-presentation. Further, analysis of the cytokine profile showed a positive correlation with the T cell proliferation assay. As shown in Fig. 7B, the IFN-c production was significantly low in case of only Salmonella infection without CpG DNA treatment. In case of Salmonella infection with CpG DNA treatment anti-H T cells produced significantly higher amount of IFN-c. This enhanced IFN-c production was abrogated with PAO treatment.

Lower burden of Salmonella in the organs of CpG induced DC treated mice
DCs are the principal cells wherein Salmonella reside after host invasion in the sub epithelial dome of the Peyer's patches [35]. In fact, Salmonella use DCs as a means of infecting M cells and employ the high infiltration capacity of the dendritic cells to help in its dissemination to other organs [29,36]. To extrapolate our findings in an in vivo context, we intraperitoneally injected control DCs or CpG treated DCs to mice and after 7 days infected them with WT Salmonella. We analyzed the bacterial load after 5 days in the spleen and liver after infecting mice orally at a dose of 10 6 bacteria per mouse. Data presented in Fig. 8A-B clearly indicates that there was a significant decrease in the bacterial burden in all the organs of mice treated with the CpG induced DCs when compared to the mice treated with control DCs or PBS treated mice.

Discussion
We observed that CpG DNA treatment inhibits Salmonella Typhimurium growth in murine DCs in a ROS dependent manner. Further, our study unravels the role of ROS in promoting antigen presentation in Salmonella infected DCs. An enhanced antigen presentation was seen in the Salmonella infected dendritic cells upon CpG treatment. Salmonella infected cells showed a decrease in the allogenic T cell proliferation than the control cells. However, Salmonella could not decrease the CpG induced enhanced T cell proliferation. Hence, it can be concluded that Salmonella utilizes the unique strategy of inhibiting the DC mediated antigen presentation. However, activating the host innate receptor TLR-9 allows an efficient presentation of Salmonella. Further, CpG treated DCs when injected in the mice system could reduce the Salmonella load from various organs suggesting this result can be extrapolated in vivo as well. Thus, our data identifies a strategy to increase the efficiency of antigen presentation by Salmonella infected DCs.
The intracellular life of Salmonella in the DC is very interesting. Salmonella does not get killed or proliferate in the DC population. Intracellular survival of S. Typhimurium in murine DC was independent from the function of virulence factors known to be important for survival in macrophages, such as the SPI2-T3SS. Surprisingly, the bacterial LPS O antigen is found to be very crucial to maintain bacterial steady burden in the DCs [37]. As known from previous studies, the effect of CpG mediated TLR-9 activation on Salmonella proliferation is a cell line dependent phenomenon. In the case of cultured hepatocytes, Salmonella growth is inhibited by the CpG DNA treatment [23], whereas in the case of human monocyte-derived macrophages [38] it has no effect. In hepatocytes, an enhanced ROS production after the CpG DNA treatment has been attributed to be the cause of decreased Salmonella burden [23]. However, in bone marrow derived macrophages CpG-antagonist act in a TLR-9 independent manner to inhibit Salmonella killing [39]. Our study is the first report which indicates that in the dendritic cells CpG DNA enhances Salmonella killing. We went ahead to check the exact mechanism behind, and noticed that CpG induction leads to an enhanced generation of ROS from the infected cells. To look into the mechanism behind this phenomenon, we blocked the ROS production in the ligand treated infected dendritic cells. Under that condition this enhanced killing was no more observed. This finding clearly demonstrates that the CpG treatment allows better killing of Salmonella in a ROS dependent manner.
Several report have documented that Salmonella inhibits antigen presentation from DC to MHC class II molecules. This effect was dependent on the induction of inducible NO synthase by DC and on the function of virulence genes in SPI2, as SPI2 deficient bacterial infection led to an enhanced amount of both CD4 and CD8 cells [40]. SPI-2 encoded proteins like SifA, SspH2, SlrP, PipB2, and SopD2 were found to be important for the interference with antigen presentation, whereas SseF and SseG contributed to a lesser extent to this phenotype [41]. Published data clearly outline the pivotal role of ROS in the antigen presenting cells [20]. Maintaining the redox potential is extremely important for macrophages and DCs to present antigen in the context of MHC molecules [16]. Oxidative stress has been reported to modulate the signal transduction pathways in lymphocytes [17]. It    has been further observed that antioxidants like NAC, quercetin and polyphenolic compounds directly inhibit B cell mediated antigen presentation by inhibiting NFkb activation [19]. Moreover, the antigen specific DC-T cell interaction can be specifically blocked by modulating the redox potential of DCs [42]. These findings led us to look into the antigen presentation capacity of the CpG treated infected cells as the cells were producing higher amount of ROS. So, we went ahead to measure the antigen presentation by the help of a mixed lymphocytic reaction. The role of CpG DNA has been well documented as an adjuvant in multiple experimental and clinical vaccines [43]. Therefore, our study indicates that CpG DNA can be used as an adjuvant to design new vaccines with increased efficiency against Salmonella. Our next aim was to check the role of CpG in aiding cross presentation in the Salmonella infected DC. We observed that unlike MLR assay Salmonella does not downregulate CD8 T cell proliferation than the control cells which was also observed by other groups previously [44]. However, CpG treatment leads to enhanced cross presentation from the DC and this was also dependent on the ROS production. When we allowed Salmonella infection in the CpG treated DCs, the same trend was maintained. Salmonella could not alter the enhanced antigen presentation and thus we hypothesize that like H-antigen any other Salmonella antigen could also be cross-presented in a better way by CpG induction. This idea is further supported by the animal experiments performed with the CpG treated DCs. The injected DCs could reduce the bacterial load from infected mice from both liver and spleen. This might be due to enhanced antigen presentation and killing of the bacteria by the anti-Salmonella T cells.
To be precise, this paper deals with two novel findings. In one hand, CpG treatment leads to killing of Salmonella in a ROS dependent manner. On the other, CpG induction leads to better antigen presentation from the Salmonella infected DCs which is also dependent on ROS. These two findings might be related in a way as enhanced killing of the pathogen might allow better presentation as well which is again supported by the in vivo data. In conclusion, our finding is the first to indicate that activating host innate immune receptor TLR-9 can improve Salmonella killing and antigen presentation by DCs. We speculate that our results will have important implications in the development of novel Salmonella vaccines utilizing CpG as an adjuvant. Future studies should focus on the exact mechanism by which ROS enhances the antigen presentation in the Salmonella infected DCs.

Ethics Statement
All the work with animals has been done with Institution approved ethics protocol. The ethics number being CAF/ ETHICS/189/2010.

Preparation of dendritic cell cultures
Murine bone marrow derived dendritic cells were cultured from the Balb/C mice as described before [40]. Briefly, the femurs were collected aseptically from mice and the marrow was flushed out. The cells were spun at 1,000 rpm for 10 min and resuspended in RPMI containing 10 ng/ml GMCSF (Peprotech), 10% heatinactivated FBS, 2 mM l-glutamine, 100 U/ml penicillin, 100 mg/ ml streptomycin and plated in 75-cm 2 tissue culture flasks. The flasks were placed in a tissue culture incubator in 10% CO 2 at 37uC and after 3 days GMCSF supplementation was done. Finally, after 6 days of culture the medium was removed, and the cells were washed once and detached by pipetting. The cells were observed to be .90% CD11c positive by fluorescence-activated cell sorting analysis. The above mentioned cell suspension was further purified using CD11c + magnetic beads (MACS, Miltenyi Biotec). The purified DCs were plated at 1610 6 cells per ml in 6well tissue culture plates 6 h prior to use.

Cell culture and bacterial infection
For the infection of cells, the wild type (WT) Salmonella Typhimurium strain was grown to stationary phase in LB in the presence of carbenicillin. The OD (600) of the culture was adjusted with Luria broth (LB) to 0.3 and the bacteria were washed once with phosphate-buffered saline (PBS). Appropriate dilution of the bacterial culture was made in cell culture medium and added to the dendritic cells growing in 24-well tissue culture plates at a multiplicity of infection (MOI) of about 10. Bacteria were centrifuged onto the cells at 5006g for 5 min. After infection for 25 min, cells were washed thrice with PBS and incubated for 1 h in cell culture medium containing 50 mg/ml gentamicin (Sigma). The cells were maintained in 10 mg/ml gentamicin containing medium for the rest of the experiment. In some sets of the experiment, 10 mg/ml TLR-9 ligand (CpG-ODN-1826, Invivogen) or 10 mg/ml TLR-9 antagonist (CpG-ODN-2088, Invivogen) was added to the cells along with 10 mg/ml gentamicin. In some experiments, cells were pretreated for 30 min with PAO (phenylarsine oxide, 5 mM) before performing the infection. For the enumeration of the intracellular bacteria, the DCs were washed three times with PBS and lysed with 0.1% Triton X-100 for 10 min at room temperature and serial dilutions were plated onto LB agar plates containing carbenicillin.

Western blot analysis
In order to detect TLR-9, aliquots containing equal amounts of protein (100 mg) were loaded onto 12.5% gel and transferred onto nitrocellulose membranes using mini gel transfer apparatus (Bio-Rad). The membranes were incubated with mouse monoclonal antibody against mouse TLR-9 (BD transduction laboratories; 1:500) for 2 h. The blots were further treated with goat anti mouse IgG-horseradish peroxidase conjugate (AP-Biotech, dilution 1:2000) for 2 h. The immune complex on the blots were detected with enhanced chemiluminescence substrate (PerkinElmer) and exposed to Eastman Kodak Co. XAR x-ray film. All blots were stripped and stained with HRP conjugated anti b-actin antibody (Sigma, 1:1000).

Localization of NADPH subunit, gp91phox by fluorescence microscopy
Localization of gp91phox in the DCs were determined by standard immunocytochemical methods. The WT-GFP infected cells were washed free of medium and the cells were fixed for 10 min in paraformaldehyde (3.5%) at room temperature and washed with PBS. Fixed cells were then incubated with goat antimouse gp91phox immunoglobulin G (BD Transduction) diluted 1:100 in PBS containing 2%BSA, 2% goat serum and 0. 1% saponin. The cells were then washed three times in PBS and incubated in identical conditions but with rabbit anti-goat IgG conjugated to Cy3 (Jackson's lab). Samples were viewed on a confocal laser-scanning microscope equipped with an argon laser (Zeiss, Germany). For determining the % co-localization, 50 microscopic fields were counted.

Splenocyte preparation
Total splenocytes were prepared from C57BL/6 mice by incubating small pieces of spleens in serum-free, calcium-free HBSS at 37uC for 45 min. This preparation was disaggregated by pipetting to produce a single-cell suspension. The cells were then washed in HBSS, and erythrocytes were lysed by hypotonic shock. Preparations were filtered to remove debris and were resuspended in RPMI.

Mixed lymphocyte reaction
Dendritic cells were isolated from the Balb/C mice as described earlier and were plated at a density of 5610 5 cells/well in 96-well flat-bottomed tissue culture plates. Bacterial infections of the DCs were performed as described previously. 10 mg/ml TLR-9 ligand, 10 mg/ml TLR-9 antagonist or PAO (5 mM, phenylarsine oxide) were added to the cells along with 10 mg/ml gentamicin. After 12 h post infection, total T cells isolated from spleens of allogenic C57BL/6 mice (as described above) were added to each well. DC and T cell ratio was maintained at 1:10. Cells were cocultured for 72 hr and then pulsed for 16 hr with [ 3 H] thymidine (0.1 mCi/ well). Cells were harvested on glass fibre filters using a semiautomated cell harvester (Nunc, Roskilde, Denmark). [ 3 H] thymidine was measured in a liquid scintillation spectrophotometer and T-cell proliferation was measured as mean counts/min of triplicate wells.

H antigen presentation assay
Dendritic cells were isolated as described earlier and were plated at a density of 5610 5 cells/well in 96-well flat-bottomed tissue culture plates. Bacterial infection of the DCs was performed as described previously. A soluble form of previously [31] purified envelope glycoprotein haemagglutinin (H-antigen) of rinder-pest virus (50 mg/ml) was added to the cells along with 10 mg/ml gentamicin. After 12 h post infection responder T cells (specific CD8 + T cells that were generated earlier from H-proteinimmunized mice) were added to each well and a proliferation assay was carried out as described for the mixed lymphocytic reaction. The specific CD8 T cell clone was generated earlier.
Briefly, inguinal lymph nodes from H immune mice were collected one week after booster immunization and co-cultured with irradiated syngenic normal mouse splenocytes in 1.5 ml complete medium containing 50 mg/ml H protein. To obtain T cell clones, limiting dilution of cultured cells was performed in presence of irradiated syngenic splenocytes and IL-2. Single cells from a growing colony were selected and expanded for testing antigen specificity and surface markers. To test the antigen specificity and MHC restriction of T cell clones, cells were co-cultured with irradiated P815 cells stably expressing H protein and with or without purified anti-mouse Kd antibody (e-Bioscience, USA). Generated H specific T cell clone on staining for surface CD marker was found to be CD8 + . The antigen specificity of the T cell clone was established using these cells as effectors cells in a CTL assay.

Cytokine ELISA
Splenocytes from C57BL/6 mice were incubated with Salmonella infected DCs from the Balb/C mice for 72 hr as described above for cell proliferation assay. Supernatants from the culture were collected for cytokine assays. Interferon-c (IFNc) was assayed using cytokine ELISA kits (BD-Bioscience).

Animals
Six to eight-week-old Balb/C and C57BL/6 mice were maintained under specific-pathogen-free conditions in Central Animal Facility, Indian Institute of Science, Bangalore, India. All the procedures with animals were carried out as approved by the animal ethic's committee of the Institute.

In vivo transfer of DCs and organ load
Female BALB/c (n = 5) were intra peritoneally injected with 10 8 DCs (untreated or CpG treated) dissolved in 100 ml PBS. After 7 days of treatment, the mice were orally infected with 10 6 CFU of the WT strain. After 5 th day of infection, homogenized samples of liver, spleen and MLN of infected mice were plated on antibiotic plates and the colonies were counted.

Statistical analysis and software
Each assay was repeated at least 3 times. In vitro data were analyzed by paired t test (two sample, equal variance) and P values below 0.05 were considered significant. FACS data were plotted and analyzed using WinMDI 2.9 software. Results of mouse in vivo challenge studies were evaluated by using Mann-Whitney U tests from the GraphPad Prism 4.0 software. Immunoblots were quantified using Multi Gauge V2.3 software. Differences between experimental groups were considered significant for P,0.05. Figure S1 MTT assay. DCs were infected in triplicate with the WT strain with/without different treatments. After 12 h of infection, MTT assay was performed and the values are represented as % cell viability considering the uninfected cell viability as 100%. There was no significant cell death in any of the conditions tested. Found at: doi:10.1371/journal.pone.0013772.s001 (0.06 MB TIF) Figure S2 Activation assay. DCs were infected in triplicate with the WT strain with/without different treatments. After 12 h of infection, FACS was performed for the dendritic cell maturation marker CD80 and the maturation statuses of the DCs are shown as MFI. Only CpG treatment leads to an enhanced maturation of the DCs. The statistics is defined as * P,0.05 (Student 't' test). Found at: doi:10.1371/journal.pone.0013772.s002 (0.06 MB TIF)