Maturation of dendritic cells (DC) is crucial for their ability to induce adaptive immunity. Although several mediators of DC maturation have been found, their contributions to DC maturation during infection are poorly understood. In this study we show that murine conventional (CD11chigh) DC up-regulate costimulatory molecules in a subset-specific manner after oral Salmonella infection. Although both CD8α+ and CD8α subsets increase CD86 expression, CD40 was preferentially up-regulated on CD8α+ DC, and CD80 was preferentially increased on CD8α DC. In addition, high levels of CD80 and CD86 were found on CD11cintCD11b+ cells that accumulated in infected organs. Costimulatory molecules were simultaneously induced on CD11chigh and CD11cintCD11b+ cells in Peyer’s patches, mesenteric lymph nodes and spleen 5 days after infection despite different kinetics of peak bacterial burden in these organs. Up-regulation of costimulatory molecules occurred on all DC within the respective subset. Moreover, <1% of CD11c-expressing cells associated with Salmonella expressing enhanced GFP in vivo. Thus, DC maturation did not depend on bacterial uptake. Rather, infection-induced up-regulation of CD80, CD86, and CD40 on CD11c-expressing cells of mesenteric lymph nodes was dependent on TNFR type I (TNFRI) signaling. Although indirect up-regulation of costimulatory molecules on DC and CD11cintCD11b+ cells was TNFRI dependent, cells directly associated with Salmonella were able to mature independently of TNFRI signaling. Thus, Salmonella-induced TNF-α is an important mediator of indirect DC maturation during infection, whereas a TNF-α-independent maturation pathway contributes to direct maturation of bacteria-associated DC.

Dendritic cells (DC)3 are professional APC with the capacity to activate naive T cells. Most DC present in secondary lymphoid organs during steady state are immature and cannot induce optimal T cell responses (1, 2, 3, 4). However, upon activation, DC mature into potent immunostimulatory APC that can drive clonal expansion of T cells and direct differentiation of Th effectors (2, 3, 4). Because the ability of DC to activate naive T cells and induce adaptive immune responses is tightly linked to their maturation state, elucidating the factors that regulate DC maturation has been an important topic of investigation.

Maturation of DC can be induced after recognition of microbial components via pattern recognition receptors such as the TLRs (5) and upon CD40-CD40L engagement (2, 3, 4, 5, 6, 7). In addition, proinflammatory cytokines, such as TNF-α and IL-1β, can induce DC maturation (8, 9, 10, 11). In vivo, microbial infection represents a primary stimulus for DC maturation (5, 7, 12). However, the pathways that lead to DC maturation in the complex environment in vivo during an infection remain largely undefined.

DC exist as phenotypically distinct subsets that are localized in different microenvironments and may be specialized for different functions. Two major DC subsets can be defined in murine lymphoid organs based on expression of the CD8 α-chain (4). Considering that CD8α+ and CD8α DC differ in, for example, TLR expression, ability to cross-prime CD8+ T cells, capacity to internalize apoptotic cells, and cytokine production (2, 4, 5, 12), it is important to understand the relative functions of the DC subsets in the immune system.

In this study we use oral infection with the enteric pathogen Salmonella enterica serovar Typhimurium (S. typhimurium) to study the mechanism of DC maturation during bacterial infection in vivo. S. typhimurium is a Gram-negative facultative intracellular bacterium that infects a host by the oral route, penetrates the intestinal epithelium, and spreads to internal organs (13, 14). Thus, after oral infection, bacterial replication occurs in the Peyer’s patches (PP), mesenteric lymph nodes (MLN), spleen, and liver. Splenic DC subsets are differentially modulated after oral Salmonella infection (15). In addition, both CD8α+ and CD8α DC of spleen and liver associate with Salmonella after i.v. infection, and CD11c+ DC from these organs can present Salmonella Ag ex vivo (16, 17). It is not known where initial Ag presentation and T cell activation occur in vivo. However, T cells specific for Salmonella flagellin proliferate in PP and MLN, but not in spleen, 2–5 days after oral Salmonella infection (18).

Salmonella contains several microbial patterns that are recognized by TLRs, including LPS and flagellin. The recognition of Salmonella LPS is important for controlling bacterial replication in vivo (19, 20) and for eliciting TNF-α and IL-6 in mice injected with killed Salmonella (21). In addition, TLR ligands such as LPS, flagellin, and CpG-DNA can induce DC maturation (5, 22, 23, 24, 25, 26). TLR-mediated DC maturation occurs via two pathways, either by direct signaling into DC or via the induction of proinflammatory cytokines that, in turn, can mediate indirect DC maturation (5, 22). During Salmonella infection, proinflammatory cytokines, such as TNF-α, are produced systemically (27). In addition, Salmonella-associated TLR agonists, such as LPS and flagellin, are available for direct recognition by DC. The influence of proinflammatory cytokines or direct bacterial association on DC maturation in vivo during bacterial infection is presently not known.

In this study we investigated the kinetics of costimulatory molecule up-regulation on CD8α+ and CD8α DC in PP, MLN, and spleen after oral infection with Salmonella. We also investigated the contribution of proinflammatory cytokines to DC maturation during oral Salmonella infection and how bacterial association affects DC maturation. We found that costimulatory molecules were simultaneously up-regulated on conventional DC in PP, MLN, and spleen despite different kinetics of peak bacterial burden in these organs. In addition, high levels of costimulatory molecules were induced on CD11cintCD11b+ cells that accumulated in infected organs. Infection studies in wild-type and cytokine-deficient mice using Salmonella expressing enhanced GFP (eGFP) revealed that wild-type mice are not dependent on bacterial uptake for costimulatory molecule up-regulation. Instead, signaling via TNFR type I (TNFRI) was important in mediating indirect maturation of the entire population of CD11c+ cells in a subset-specific manner. However, DC and CD11cintCD11b+ cells that were directly associated with eGFP-expressing Salmonella were able to up-regulate CD80 and CD86 independently of TNFRI. These data indicate the existence of both direct and indirect pathways of DC maturation during Salmonella infection.

C57BL/6 mice were purchased from Charles River Laboratories. TNFRI−/− mice on the C57BL/6 background were bred at the experimental biomedicine animal facility of Goteborg University. Caspase-1−/− mice from BASF Bioresearch on a mixed background (28) were backcrossed to C57BL/6 mice at the experimental biomedicine animal facility. Caspase-1−/− mice were backcrossed five generations or more on a C57BL/6 background, and heterozygous littermates were used as controls. Mice were used between 8 and 12 wk of age and were provided food and water ad libitum. All mouse experiments were performed following protocols approved by the government animal ethical committee and institutional animal use and care guidelines.

The S. typhimurium strains used in this study are the χ3181 SR11 derivatives χ4666 (15) and χ4550 (29). The wild-type strain SL1344 and its eGFP-expressing derivative SMO22 (30) were also used. Briefly, χ4666 and SL1344 are fully virulent strains, and after an oral dose of 108 bacteria, mice succumb to the infection in ∼6–10 days. χ4550 contains deletions in the genes encoding adenylate cyclase and the cAMP receptor protein that reduce its virulence relative to the wild-type strain. In all experiments, mice given an oral dose of 109 χ4550 survive, clear the bacteria ∼4 wk after infection, and generate lasting CD4+ and CD8+ T cell immunity (29).

Bacteria were cultured from frozen glycerol stocks overnight in Miller’s Luria-Bertoni broth at 37°C. χ4666 and χ4550 were cultured with constant agitation, whereas SMO22 and SL1344 were grown in static cultures supplemented with antibiotics (SMO22, 50 μg/ml kanamycin and 100 μg/ml streptomycin; SL1344, 100 μg/ml streptomycin). The concentration of bacteria was estimated spectrophotometrically. Mice were inoculated intragastrically with 0.1 ml of 1% NaHCO3, followed 10–15 min later with bacteria in a volume of 0.1–0.2 ml. The actual bacterial dose was determined by plating serial dilutions of the bacterial suspension on Luria-Bertoni agar plates.

C57BL/6 mice received 108 χ4666 or 109 χ4550, whereas TNFRI−/− mice received 2 × 107 to 1 × 108 χ4666 orally. These doses were used to obtain TNFRI−/− mice with a similar bacterial burden as C57BL/6 mice for the experiments shown in Figs. 7 and 8. When using SL1344 and SMO22, mice were given 2–4 × 109 bacteria orally to increase the possibility of detecting a sufficient number of eGFP+ events. Using this bacterial dose, the optimal day for examining maturation of CD11c-expressing cells containing bacteria was determined to be 3 days after infection (data not shown).

FIGURE 7.

DC maturation is impaired in infected TNFRI−/− mice. A and E, Expression of CD80, CD86, and CD40 on CD11chighCD8α DC, CD11chighCD8α+ DC, and CD11cintCD11b+ cells from MLN (A) or spleen (E) of naive (○) or day 5-infected (•) C57BL/6 (B6) or TNFRI−/− mice as indicated. Cells were stained for flow cytometry as described in Fig. 3. Data were obtained in six experiments and are expressed as fold up-regulation compared with the mean MFI of CD80, CD86, or CD40 in naive TNFRI−/− mice. Each symbol represents an individual mouse, and horizontal bars indicate the mean value. Each group contains six to 11 individual mice. The p values obtained from the Mann-Whitney U test for infected C57BL/6 compared with infected TNFRI−/− mice are shown. B and F, Bacterial load in MLN (B) and spleen (F) of the infected C57BL/6 and TNFRI−/− mice shown in A and D. C and G, Absolute number of CD11cintCD11b+ cells in MLN (C) and spleen (G) of the C57BL/6 and TNFRI−/− mice shown in A and E. D and H, The level of IL-1β in lysates of MLN (D) and spleen (H) of naive (○) or day 5-infected (•) C57BL/6 and TNFRI−/− mice. Each symbol represents an individual mouse, and horizontal bars indicate the mean value. Organ lysates were prepared directly ex vivo, and IL-1β was quantitated by ELISA.

FIGURE 7.

DC maturation is impaired in infected TNFRI−/− mice. A and E, Expression of CD80, CD86, and CD40 on CD11chighCD8α DC, CD11chighCD8α+ DC, and CD11cintCD11b+ cells from MLN (A) or spleen (E) of naive (○) or day 5-infected (•) C57BL/6 (B6) or TNFRI−/− mice as indicated. Cells were stained for flow cytometry as described in Fig. 3. Data were obtained in six experiments and are expressed as fold up-regulation compared with the mean MFI of CD80, CD86, or CD40 in naive TNFRI−/− mice. Each symbol represents an individual mouse, and horizontal bars indicate the mean value. Each group contains six to 11 individual mice. The p values obtained from the Mann-Whitney U test for infected C57BL/6 compared with infected TNFRI−/− mice are shown. B and F, Bacterial load in MLN (B) and spleen (F) of the infected C57BL/6 and TNFRI−/− mice shown in A and D. C and G, Absolute number of CD11cintCD11b+ cells in MLN (C) and spleen (G) of the C57BL/6 and TNFRI−/− mice shown in A and E. D and H, The level of IL-1β in lysates of MLN (D) and spleen (H) of naive (○) or day 5-infected (•) C57BL/6 and TNFRI−/− mice. Each symbol represents an individual mouse, and horizontal bars indicate the mean value. Organ lysates were prepared directly ex vivo, and IL-1β was quantitated by ELISA.

Close modal
FIGURE 8.

TNFRI-independent maturation of CD11c-expressing cells containing Salmonella. C57BL/6 (A) or TNFRI−/− (B) mice were orally infected with eGFP-expressing Salmonella, and after 3 days CD11c+ cells were purified from PP and MLN and stained with anti-CD11c, CD8α, CD11b, 7AAD, and either CD80, CD86, CD40, or appropriate isotype control mAb and analyzed by six-color flow cytometry. Because TNFRI−/− mice lack visible PP, only MLN could be analyzed in these animals. Histograms show the expression of CD80, CD86, and CD40 on the indicated CD11c-expressing populations in the indicated organs on gated live eGFP+ (open histogram) and eGFP (filled histogram) cells. Cells were gated as described in Fig. 5. The upper number in each box shows the fold up-regulation of costimulatory molecules on eGFP+ cells compared with the corresponding total cell population from naive mice. The lower number in each box shows the fold up-regulation of costimulatory molecules on eGFP cells compared with the corresponding total cell population from naive mice. Cells were pooled from 14–15 mice, and 100–800 eGFP+ events were acquired in each sample. Data are representative of four independent experiments. ∗, For CD11chighCD8α DC in PP of C57BL/6 mice, data from naive animals are missing, and the fold up-regulation of eGFP+ and eGFP cells compared with the naive population could not be calculated. However, the fold differences in MFI for CD80, CD86, and CD40 on eGFP+ compared with eGFP CD8α DC are 1.4, 0.5, and 1.4, respectively.

FIGURE 8.

TNFRI-independent maturation of CD11c-expressing cells containing Salmonella. C57BL/6 (A) or TNFRI−/− (B) mice were orally infected with eGFP-expressing Salmonella, and after 3 days CD11c+ cells were purified from PP and MLN and stained with anti-CD11c, CD8α, CD11b, 7AAD, and either CD80, CD86, CD40, or appropriate isotype control mAb and analyzed by six-color flow cytometry. Because TNFRI−/− mice lack visible PP, only MLN could be analyzed in these animals. Histograms show the expression of CD80, CD86, and CD40 on the indicated CD11c-expressing populations in the indicated organs on gated live eGFP+ (open histogram) and eGFP (filled histogram) cells. Cells were gated as described in Fig. 5. The upper number in each box shows the fold up-regulation of costimulatory molecules on eGFP+ cells compared with the corresponding total cell population from naive mice. The lower number in each box shows the fold up-regulation of costimulatory molecules on eGFP cells compared with the corresponding total cell population from naive mice. Cells were pooled from 14–15 mice, and 100–800 eGFP+ events were acquired in each sample. Data are representative of four independent experiments. ∗, For CD11chighCD8α DC in PP of C57BL/6 mice, data from naive animals are missing, and the fold up-regulation of eGFP+ and eGFP cells compared with the naive population could not be calculated. However, the fold differences in MFI for CD80, CD86, and CD40 on eGFP+ compared with eGFP CD8α DC are 1.4, 0.5, and 1.4, respectively.

Close modal

The bacterial load in spleen, MLN, and PP was assessed after plating serial dilutions of the cell suspension obtained from the organ on Luria-Bertoni agar plates. In long-term kinetic experiments, mice were infected at 8 wk of age.

Spleen, MLN, and PP were incubated in 1.6 mg/ml collagenase type IV (Sigma-Aldrich) and 1 mg/ml DNase I (Sigma-Aldrich) in HBSS (Invitrogen Life Technologies) at 37°C for 45 min, after which a single-cell suspension was made by pipetting the digested organs. The cells were washed in HBSS, and erythrocytes were lysed in splenic cell suspensions by incubating with 0.14 M NH4Cl for 5 min at room temperature. All preparations were filtered to remove debris and were resuspended in RPMI 1640 (Invitrogen Life Technologies) containing 10% heat-inactivated FBS (PAA Laboratories). The total number of cells per organ was determined by trypan blue exclusion. Cells from spleen, MLN, and PP were prepared from individual mice for all experiments except those in Figs. 5 and 8. In these experiment, CD11c-expressing cells were enriched from single-cell suspensions of MLN and PP pooled from 14 or 15 mice using N418 magnetic beads (Miltenyi Biotec) and AutoMACS (Miltenyi Biotec) following the manufacturer’s protocol.

FIGURE 5.

In vivo uptake of Salmonella by CD11c-expressing cells. Mice were orally infected with Salmonella expressing eGFP (Sal-eGFP) or the parental strain lacking eGFP (Sal). Three days after infection, CD11c-expressing cells were purified by MACS from PP (A–C) or MLN (D–F) pooled from 15 mice. Cells were then stained with anti-CD11c, CD8α, CD11b, and 7AAD and analyzed by five-color flow cytometry. Dot plots show fluorescence of eGFP and CD11b on gated CD11cintCD11b+ cells (A and D) or fluorescence of eGFP and CD8α on gated CD11chighCD8α (B and E) or CD11chighCD8α+ (C and F) cells. CD11cintCD11b+ cells were gated as described in Fig. 2 (R1), and CD11chighCD8α and CD11chighCD8α+ cells were gated as described in Fig. 1. Data are representative of five independent experiments.

FIGURE 5.

In vivo uptake of Salmonella by CD11c-expressing cells. Mice were orally infected with Salmonella expressing eGFP (Sal-eGFP) or the parental strain lacking eGFP (Sal). Three days after infection, CD11c-expressing cells were purified by MACS from PP (A–C) or MLN (D–F) pooled from 15 mice. Cells were then stained with anti-CD11c, CD8α, CD11b, and 7AAD and analyzed by five-color flow cytometry. Dot plots show fluorescence of eGFP and CD11b on gated CD11cintCD11b+ cells (A and D) or fluorescence of eGFP and CD8α on gated CD11chighCD8α (B and E) or CD11chighCD8α+ (C and F) cells. CD11cintCD11b+ cells were gated as described in Fig. 2 (R1), and CD11chighCD8α and CD11chighCD8α+ cells were gated as described in Fig. 1. Data are representative of five independent experiments.

Close modal

FACS buffer in HBSS (Invitrogen Life Technologies) containing 3% FBS (PAA Laboratories), 2 mM EDTA (Sigma-Aldrich), 20 mM HEPES (Invitrogen Life Technologies), and 0.01% NaN3 (Sigma-Aldrich) was used for all incubations and washing steps. Before staining, samples were incubated with anti-FcγRII/III mAb (2.4G2; purified as described in Ref.16) for 15 min at 4°C. Cells were then stained with FITC, PE, allophycocyanin, PE-Cy7, allophycocyanin-Cy7, or biotinylated conjugates of anti-CD11c (HL3), CD8α (53-6.7), CD80 (16-10A1), CD86 (GL1), CD40 (3/23), CD11b (M1/70), Gr-1 (RB6-8C5), B220 (RA3-6B2), MHC-II (M5/114.15.2), Mac-3 (M3/84), Ly-6C (AL-21), DEC-205 (NLDC-145), or F4/80 for 20 min at 4°C. All mAb were purchased from BD Biosciences, except F4/80 and DEC-205, which were obtained from Caltag Laboratories or from our own production, respectively. 7-Aminoactinomycin D (7AAD; Sigma-Aldrich) was always used to exclude dead cells. Isotype control mAb (BD Biosciences) with the appropriate conjugates were used. When biotinylated mAb were used, cells were stained a second time with streptavidin-allophycocyanin or streptavidin-PE-Cy7 (BD Biosciences). To detect intracellular production of TNF-α, cells were seeded at 5 × 106 cells/well in low adherence, 24-well plates (Corning) and incubated for 4 h with brefeldin A (Sigma-Aldrich) at a final concentration of 5 μg/ml. Cells were then stained for their surface phenotype, fixed, permeabilized, and stained intracellularly with anti-TNF-α (MP6-XT22; BD Biosciences) as described previously (15). Appropriate isotype control mAb were used.

Samples were acquired for four-color analysis using an LSR-I flow cytometer or for five- or six-color analysis on an LSR-II flow cytometer (BD Biosciences) using DIVA software (BD Biosciences) and were analyzed using FlowJo software (Tree Star). A total of 400,000–1,200,000 cells were acquired from splenic or MLN samples to ensure the analysis of at least 8,000 viable splenic DC and 3,000–10,000 viable MLN DC in each sample from individual mice. From PP samples of individual mice we were able to acquire 200,000–900,000 total cells and thus could analyze 1,000–3,000 viable DC in samples from naive mice and 500–1,000 viable DC in samples from infected mice. In studies of bacterial uptake and maturation of Salmonella-containing cells, 100–800 CD11c+eGFP+ cells were acquired for each subset shown.

Spleens and MLN from naive or infected mice were lysed in ice-cold PBS containing 0.01% Triton X-100 (Sigma-Aldrich). Lysates were centrifuged at 10,000 × g and frozen at −20°C until assayed for cytokine content or total protein content. The presence of IL-1β and GM-CSF in organ lysates was determined using Luminex 100 technology and Bio-Plex kits (Bio-Rad) according to the manufacturer’s protocols. Results were analyzed using Bio-Plex Manager software (Bio-Rad). For Fig. 7, D and H, the presence of IL-1β in organ lysates was determined by ELISA following the manufacturer’s protocol (BD Biosciences). Total protein content in organ lysates was assayed using the DC protein assay (Bio-Rad) with bovine plasma γ-globulin as standard (Bio-Rad). The level of cytokines in organ lysates is expressed as picograms per milligram of protein.

To determine the capacity of MLN and splenic cells to produce TNF-α, these organs from naive mice or mice infected 3 or 5 days earlier with χ4666 were meshed through cell strainers. The cells were washed and resuspended in complete medium (RPMI 1640 Glutamax-1 containing 10% FBS (PAA Laboratories), 2 mM MEM sodium pyruvate, 20 mM HEPES, 0.05 mM 2-ME, 100 U/ml penicillin/streptomycin, 2.5 μg/ml fungizone, and 25 μg/ml gentamicin (all from Invitrogen Life Technologies)). In splenic samples, erythrocytes were lysed as described above. Cells were seeded in 24-well plates (2 × 106/well) and cultured for 24 h with or without Salmonella lysate (1/200) prepared as described previously (31). The presence of TNF-α in supernatants was determined by ELISA following the manufacturer’s protocol (BD Biosciences).

During oral infection with Salmonella, bacteria replicate in PP, MLN, and spleen. To analyze the maturation status of DC subsets within these organs during infection, the expression of CD80, CD86, and CD40 on CD8α+ and CD8α DC from spleen, MLN, and PP of orally infected mice was assessed directly ex vivo. In the steady state, CD8α+ and CD8α DC from each organ expressed a similar, low level of CD80 and CD40. The steady state level of CD86, however, was higher on CD8α+ DC relative to CD8α DC in each of the three organs (Fig. 1 and Table I).

FIGURE 1.

Maturation of DC subsets in vivo after Salmonella infection. Splenic, MLN, and PP cells from naive mice (filled histogram) or from mice orally infected with χ4666 5 days earlier (thick line, open histogram) were stained directly ex vivo with anti-CD11c, CD8α, 7AAD, and either CD80, CD86, CD40, or appropriate isotype control mAb and analyzed by flow cytometry. Viable (7AAD) CD8α+ DC (upper histograms) and CD8α DC (lower histograms) were gated as shown in the dot plots and analyzed for the expression of CD80, CD86, and CD40. The histograms with a dotted line represent staining with appropriate isotype control mAb. The data shown are representative for 13–31 mice.

FIGURE 1.

Maturation of DC subsets in vivo after Salmonella infection. Splenic, MLN, and PP cells from naive mice (filled histogram) or from mice orally infected with χ4666 5 days earlier (thick line, open histogram) were stained directly ex vivo with anti-CD11c, CD8α, 7AAD, and either CD80, CD86, CD40, or appropriate isotype control mAb and analyzed by flow cytometry. Viable (7AAD) CD8α+ DC (upper histograms) and CD8α DC (lower histograms) were gated as shown in the dot plots and analyzed for the expression of CD80, CD86, and CD40. The histograms with a dotted line represent staining with appropriate isotype control mAb. The data shown are representative for 13–31 mice.

Close modal
Table I.

Expression of costimulatory molecules on DC subsets of naive or Salmonella-infected mice

OrganDC SubsetNaiveDay 5
CD80CD86CD40CD80CD86CD40
Spleen CD8α+ 2800 ± 200a 3200 ± 200 600 ± 50 2900 ± 700 5300 ± 700 2000 ± 500 
 CD8α 2600 ± 200 1000 ± 200 400 ± 30 5200 ± 800 5300 ± 1100 900 ± 80 
MLN CD8α+ 2800 ± 200 3400 ± 300 1300 ± 300 3400 ± 400 4400 ± 1000 2400 ± 180 
 CD8α 2100 ± 300 2600 ± 200 800 ± 200 5700 ± 1200 6100 ± 2500 1600 ± 400 
PP CD8α+ 700 ± 600 800 ± 600 200 ± 200 1400 ± 600 1300 ± 400 800 ± 400 
 CD8α 3100 ± 600 800 ± 200 500 ± 70 5500 ± 800 1900 ± 100 1000 ± 300 
OrganDC SubsetNaiveDay 5
CD80CD86CD40CD80CD86CD40
Spleen CD8α+ 2800 ± 200a 3200 ± 200 600 ± 50 2900 ± 700 5300 ± 700 2000 ± 500 
 CD8α 2600 ± 200 1000 ± 200 400 ± 30 5200 ± 800 5300 ± 1100 900 ± 80 
MLN CD8α+ 2800 ± 200 3400 ± 300 1300 ± 300 3400 ± 400 4400 ± 1000 2400 ± 180 
 CD8α 2100 ± 300 2600 ± 200 800 ± 200 5700 ± 1200 6100 ± 2500 1600 ± 400 
PP CD8α+ 700 ± 600 800 ± 600 200 ± 200 1400 ± 600 1300 ± 400 800 ± 400 
 CD8α 3100 ± 600 800 ± 200 500 ± 70 5500 ± 800 1900 ± 100 1000 ± 300 
a

Values represent the mean of the MFIs for CD80, CD86, or CD40 of four to six mice ± SD. The number of events collected is described in Materials and Methods.

In response to oral Salmonella infection, CD80, CD86, and CD40 were up-regulated on DC in a subset-specific fashion. For instance, CD80 was clearly up-regulated on CD8α DC 5 days after infection, whereas little, if any, up-regulation of this molecule was apparent on CD8α+ DC in spleen, MLN, and PP (Fig. 1). CD8α DC showed the greatest up-regulation of CD86 compared with naive mice. However, given the high basal level of CD86 on CD8α+ DC, the median fluorescence intensity (MFI) for CD86 was similar on CD8α+ and CD8α DC 5 days after infection. In contrast to CD80 and CD86, CD40 was up-regulated the most and was expressed at the highest level on CD8α+ DC after oral Salmonella infection (Fig. 1 and Table I).

In naive mice, CD80 was only expressed on conventional DC (Fig. 2,A). However, 5 days after infection, a high level of CD80 was also found on a cell population that expressed an intermediate level of CD11c and a high level of CD11b (Fig. 2,A). These CD11cintCD11b+ cells accumulated in spleen (Fig. 2, B and C) and MLN (Fig. 2, D and E) as well as PP (data not shown) of infected mice. In addition to CD11cintCD11b+ cells, CD11b+ cells lacking CD11c expression accumulated in infected organs (Fig. 2, B and D, gate R3). However, only the CD11cintCD11b+, not the CD11clowCD11b+, cells expressed costimulatory molecules and MHC-II at a level similar to conventional DC from infected mice (Table II). The surface phenotype of CD11cintCD11b+ cells was different from those of both conventional DC and CD11clowCD11b+ cells, because they expressed a higher level of F4/80 and Mac-3 as well as an intermediate level of Gr-1 (Fig. 2,F). CD11cintCD11b+ cells lacked expression of CD8α and DEC-205 and expressed Ly6/C at a level similar to conventional DC. The majority of cells within the CD11cint population (gate R1) were negative for B220 and expressed a high level of CD11b, suggesting that they are not plasmacytoid DC (Fig. 2, A and F) (32). Furthermore, a large proportion of the CD11cintCD11b+ cells produced TNF-α in response to Salmonella infection (Fig. 2 G). Thus, infection with Salmonella induces up-regulation of CD80, CD86, and CD40 on conventional DC of PP, MLN, and spleen in a subset-specific manner. In addition, CD11cintCD11b+ cells, which have a mixed monocyte/DC phenotype and a level of costimulatory molecules similar to conventional DC in infected mice, are recruited to infected organs.

FIGURE 2.

Characterization of CD11cintCD11b+ cells with high expression of costimulatory molecules that appear after Salmonella infection. Splenic or MLN cells from naive mice or mice orally infected with χ4666 5 days earlier were stained with anti-CD11c, CD11b, 7AAD, and either F4/80, Mac-3, Gr-1, CD8α, DEC-205, Ly6/C, B220, TNF-α, or appropriate isotype control mAb. A, Expression of CD11c and CD11b on gated CD80+ live splenic cells in naive and infected mice. The numbers in the histograms show the percentage of CD80+ cells among total live cells. Staining with isotype control mAb for CD80 resulted in <0.03% positive cells (not shown). Gate R1 contains CD11cintCD11b+ cells, and gate R2 contains CD11chigh conventional DC. Data are representative of 31 mice. B and D, CD11c and CD11b expression on gated live cells from spleen (B) or MLN (D) in naive and infected mice. The R1 and R2 gates are the same as in A. The R3 gate depicts CD11clowCD11b+ cells lacking high CD80 expression. For clarity, cells within gates R1 and R2 are gray. C and E, Absolute number of CD11cintCD11b+ cells (R1) in spleen (C) and MLN (E) of naive or infected mice. Data are mean values of six to 10 mice, and error bars show the SEM. F, Histograms show the expression of the indicated markers on gated R1 (CD11cintCD11b+), R2 (CD11chigh DC), or R3 (CD11clowCD11b+) cells among total live cells from spleen on day 5 after infection. Values in histograms depict the percentage of cells within the indicated gate. Histograms with a dotted line represent staining with appropriate isotype control mAb. Data are representative of 10 mice. G, TNF-α production by gated CD11cintCD11b+ cells (R1) in spleen of naive or infected mice. Values in dot plots show the percentage of cells within the indicated gate. Staining with isotype control resulted in <2% positive cells (not shown). Data are representative of 10 mice.

FIGURE 2.

Characterization of CD11cintCD11b+ cells with high expression of costimulatory molecules that appear after Salmonella infection. Splenic or MLN cells from naive mice or mice orally infected with χ4666 5 days earlier were stained with anti-CD11c, CD11b, 7AAD, and either F4/80, Mac-3, Gr-1, CD8α, DEC-205, Ly6/C, B220, TNF-α, or appropriate isotype control mAb. A, Expression of CD11c and CD11b on gated CD80+ live splenic cells in naive and infected mice. The numbers in the histograms show the percentage of CD80+ cells among total live cells. Staining with isotype control mAb for CD80 resulted in <0.03% positive cells (not shown). Gate R1 contains CD11cintCD11b+ cells, and gate R2 contains CD11chigh conventional DC. Data are representative of 31 mice. B and D, CD11c and CD11b expression on gated live cells from spleen (B) or MLN (D) in naive and infected mice. The R1 and R2 gates are the same as in A. The R3 gate depicts CD11clowCD11b+ cells lacking high CD80 expression. For clarity, cells within gates R1 and R2 are gray. C and E, Absolute number of CD11cintCD11b+ cells (R1) in spleen (C) and MLN (E) of naive or infected mice. Data are mean values of six to 10 mice, and error bars show the SEM. F, Histograms show the expression of the indicated markers on gated R1 (CD11cintCD11b+), R2 (CD11chigh DC), or R3 (CD11clowCD11b+) cells among total live cells from spleen on day 5 after infection. Values in histograms depict the percentage of cells within the indicated gate. Histograms with a dotted line represent staining with appropriate isotype control mAb. Data are representative of 10 mice. G, TNF-α production by gated CD11cintCD11b+ cells (R1) in spleen of naive or infected mice. Values in dot plots show the percentage of cells within the indicated gate. Staining with isotype control resulted in <2% positive cells (not shown). Data are representative of 10 mice.

Close modal
Table II.

Expression of CD80, CD86, CD40, and MHC-II on CD11cintCD11b+ cells, CD11clowCD11b+ cells, and conventional DC of Salmonella-infected micea

CellsbCD80CD86CD40MHC-II
CD11cintCD11b+ (R1) 4,400 ± 900c 6,400 ± 1,000 1,211 ± 200 25,300 ± 8,500 
Conventional DC (R2) 4,200 ± 900 4,800 ± 600 1,000 ± 100 25,000 ± 7,900 
CD11clowCD11b+ (R3) 900 ± 200 1,600 ± 200 200 ± 100 3,900 ± 1,400 
CellsbCD80CD86CD40MHC-II
CD11cintCD11b+ (R1) 4,400 ± 900c 6,400 ± 1,000 1,211 ± 200 25,300 ± 8,500 
Conventional DC (R2) 4,200 ± 900 4,800 ± 600 1,000 ± 100 25,000 ± 7,900 
CD11clowCD11b+ (R3) 900 ± 200 1,600 ± 200 200 ± 100 3,900 ± 1,400 
a

Mice were orally infected with Salmonella and after 5 days splenic cells were analyzed.

b

Cells were gated as shown in Fig. 2 B.

c

Values represent the mean MFI for CD80, CD86, CD40, or MHC-II of 7-11 mice ± SD.

Next, the kinetics of costimulatory molecule up-regulation on CD11c-expressing populations were determined. We hypothesized that costimulatory molecules would be up-regulated earlier in PP and MLN than in spleen because bacterial numbers peak in PP and MLN before those in spleen after oral infection (18). To this end, mice were orally infected with Salmonella, and after 2, 3, and 5 days the maturation status of conventional DC subsets and CD11cintCD11b+ cells from spleen, MLN, and PP was determined. For conventional DC, little, if any, change in surface expression of CD80, CD86, and CD40 was apparent relative to the steady-state level on either CD8α+ or CD8α DC in any organ until 5 days after infection (Fig. 3,A). This simultaneous up-regulation on day 5 occurred despite differential kinetics of peak bacterial burden in the PP, MLN, and spleen (Fig. 3,B) and thus was not consistent with our hypothesis. Similarly, little increase in surface expression of CD80, CD86, and CD40 before day 5 after infection was detected on the CD11cintCD11b+ population in any organ. In general, this population had the lowest steady state level of costimulatory molecules and the greatest fold increase in MFI on day 5 after infection (Fig. 3 A). The up-regulation of costimulatory molecules on CD11cintCD11b+ cells was most apparent for CD80 expression and was seen in all three organs examined.

FIGURE 3.

Short-term kinetics of costimulatory molecule up-regulation and accumulation of CD11c-expressing cells after infection with χ4666. Two, 3, and 5 days after infection, cells from the spleen, MLN, and PP were stained with anti-CD11c, CD8α, CD11b, 7AAD, and either CD80, CD86, CD40, or appropriate isotype control mAb and were analyzed by five-color flow cytometry. A, Data show mean MFI for CD80, CD86, and CD40 on gated live (7AAD) CD11chighCD8α+, CD11chighCD8α, or CD11cintCD11b+ cells. Each time point represents the mean of four to six mice, and error bars are the SEM. Data are representative of two such independent experiments. B, Bacterial load in PP, MLN, and spleen of the mice shown in A at the indicated time points after infection. Data are the mean values of four infected mice, and error bars are the SEM. C and D, Absolute number of CD11chighCD8α+, CD11chighCD8α, and CD11cintCD11b+ cells in MLN (C) or spleen (D) of the mice shown in A at the indicated time points after infection. CD11chigh conventional DC were defined as in Fig. 1, and CD11cintCD11b+ cells as in Fig. 2, gate R1. Data are the mean of four to six mice, and error bars are the SEM.

FIGURE 3.

Short-term kinetics of costimulatory molecule up-regulation and accumulation of CD11c-expressing cells after infection with χ4666. Two, 3, and 5 days after infection, cells from the spleen, MLN, and PP were stained with anti-CD11c, CD8α, CD11b, 7AAD, and either CD80, CD86, CD40, or appropriate isotype control mAb and were analyzed by five-color flow cytometry. A, Data show mean MFI for CD80, CD86, and CD40 on gated live (7AAD) CD11chighCD8α+, CD11chighCD8α, or CD11cintCD11b+ cells. Each time point represents the mean of four to six mice, and error bars are the SEM. Data are representative of two such independent experiments. B, Bacterial load in PP, MLN, and spleen of the mice shown in A at the indicated time points after infection. Data are the mean values of four infected mice, and error bars are the SEM. C and D, Absolute number of CD11chighCD8α+, CD11chighCD8α, and CD11cintCD11b+ cells in MLN (C) or spleen (D) of the mice shown in A at the indicated time points after infection. CD11chigh conventional DC were defined as in Fig. 1, and CD11cintCD11b+ cells as in Fig. 2, gate R1. Data are the mean of four to six mice, and error bars are the SEM.

Close modal

Regarding the absolute number of conventional DC subsets and CD11cintCD11b+ cells after infection, CD11chighCD8α DC of MLN showed no significant change in number during the first 5 days of infection (Fig. 3,C). CD8α+ DC in this organ remained unchanged until day 3 and showed an ∼50% reduction on day 5 after infection (Fig. 3,C). In contrast, CD11cintCD11b+ cells in MLN increased 3-fold in number 5 days after infection. In the spleen, the number of CD8α+ and CD8α DC showed no significant change compared with those in naive mice, whereas the number of CD11cintCD11b+ cells increased 2-fold 5 days after infection (Fig. 3,D). Thus, the total number of CD11chigh DC was ∼4-fold higher than that of CD11cintCD11b+ cells in the spleen at 5 days after infection, whereas the MLN contained approximately equal numbers of CD11chigh DC and CD11cintCD11b+ cells at this time (Fig. 3, C and D).

Infection with virulent Salmonella has a fatal ending, and on day 5 after infection, the bacterial load in spleen and MLN is relatively high (Fig. 3,B). To analyze the kinetics of costimulatory molecule up- and down-regulation under conditions where mice survive long-term and build protective immunity (29), mice were infected with a less virulent strain of Salmonella, χ4550. With this strain, bacterial load peaked in MLN and spleen on days 5 and 23, respectively, before bacterial numbers declined and the host cleared the infection (Fig. 4,B). Infection with χ4550 induced changes in CD80 and CD86 expression on conventional DC and CD11cintCD11b+ cells similar to those observed in the short-term kinetics study (Fig. 3,A) with some slight exceptions. For example, the kinetics of costimulatory molecule up-regulation were slower, and the level of CD80 on CD11cintCD11b+ cells did not reach that on CD8α conventional DC (Figs. 3,A and 4,A). Furthermore, the apparent reduction in CD86 MFI for CD11cintCD11b+ cells in MLN after infection was more pronounced in the long-term kinetics study compared with the short-term study (Figs. 3,A and 4,A). This reduction in CD86 MFI was consistently seen in CD11cintCD11b+ cells of infected MLN and is probably due to the paucity of these cells in naive MLN, which makes the determination of MFI easily influenced by other cells that may fall into this gate (see Fig. 2 D).

FIGURE 4.

Long-term kinetics of costimulatory molecule up-regulation and DC accumulation after infection with χ4550. At the indicated time points after infection, cells from spleen and MLN were stained as described in Fig. 3. A, Data show mean MFI for CD80 and CD86 on gated live (7AAD) CD11chighCD8α+, CD11chighCD8α, or CD11cintCD11b+ cells. B, Bacterial load in MLN and spleen of the mice analyzed in A at the indicated time points after infection. C and D, Absolute number of CD11chighCD8α+, CD11chighCD8α, and CD11cintCD11b+ cells in MLN (C) and spleen (D) of the mice analyzed in A at the indicated times after infection. CD11chigh conventional DC are defined in Fig. 1, and CD11cintCD11b+ cells in Fig. 2, gate R1. Data represent the mean of four infected mice per time point and a total of 10 naive mice in each of two independent experiments. Error bars are the SEM.

FIGURE 4.

Long-term kinetics of costimulatory molecule up-regulation and DC accumulation after infection with χ4550. At the indicated time points after infection, cells from spleen and MLN were stained as described in Fig. 3. A, Data show mean MFI for CD80 and CD86 on gated live (7AAD) CD11chighCD8α+, CD11chighCD8α, or CD11cintCD11b+ cells. B, Bacterial load in MLN and spleen of the mice analyzed in A at the indicated time points after infection. C and D, Absolute number of CD11chighCD8α+, CD11chighCD8α, and CD11cintCD11b+ cells in MLN (C) and spleen (D) of the mice analyzed in A at the indicated times after infection. CD11chigh conventional DC are defined in Fig. 1, and CD11cintCD11b+ cells in Fig. 2, gate R1. Data represent the mean of four infected mice per time point and a total of 10 naive mice in each of two independent experiments. Error bars are the SEM.

Close modal

After infection with the less virulent strain, the peak MFI for CD80 and CD86 was observed on day 13 in both spleen and MLN despite the fact that the bacterial load peaked at different times in these organs (Fig. 4, A and B). Although the up-regulation of costimulatory molecules occurred with a similar kinetics in spleen and MLN, the kinetics with which costimulatory molecules were down-regulated differed between the two organs. In MLN the expression of CD80 had subsided >50% after 23 days and returned to baseline levels on day 47. In contrast, the levels of CD80 and CD86 on splenic CD11c+ cells remained at a peak 13–23 days after infection. At 47 days after infection, however, the expression of costimulatory molecules in the spleen was back to baseline levels (Fig. 4 A).

The transient pattern of costimulatory molecule up- and down-regulation in MLN was also reflected in the accumulation and contraction of CD11c-expressing cells in this organ. The three CD11c-expressing populations expanded and contracted rapidly in MLN, with the highest cell number occurring 13 days after infection, coincident with the peak expression of CD80 in this organ (Fig. 4,C). The greatest increase in number was observed for CD11cintCD11b+ cells, which showed a 22-fold increase in the MLN on day 13 after infection. Among the conventional CD11chigh subsets, CD8α DC increased in number 7 times compared with that in naive mice, whereas CD8α+ DC showed only a 2-fold increase in MLN (Fig. 4 C). On day 23 after infection, the number of cells in the three CD11c-expressing populations had dropped to naive levels.

In contrast to that in MLN, the increase in number of conventional DC subsets and CD11cintCD11b+ cells in spleen was prolonged, peaking 13–23 days after infection (Fig. 4,D). This may reflect the slower seeding of bacteria to the spleen and/or the lower bacterial burden in this organ relative to that reached after infection with the fully virulent strain (Figs. 3,B and 4,B). Unlike that in MLN, the greatest number of CD11c+ cells in infected spleens was accounted for by the conventional CD8α DC, showing a 4-fold increase on day 23 after infection (Fig. 4,D). CD11cintCD11b+ cells had the greatest expansion in the spleen, increasing 5 times, whereas CD8α+ DC increased in number only 2-fold compared with naive levels (Fig. 4 D).

Thus, during a long-term infection that induces protective immunity, costimulatory molecules are up-regulated in a subset-specific fashion. Similar to infection with fully virulent Salmonella, the up-regulation of costimulatory molecules and accumulation of CD11c-expressing cells occurred with similar kinetics in MLN and spleen. However, the down-regulation of costimulatory molecules and contraction of CD11c+ cells required a longer time period in spleen than in MLN.

The data obtained showed that the three CD11c-expressing populations differentially up-regulated costimulatory molecules in response to oral Salmonella. We thus examined the capacities of these populations to associate with Salmonella in vivo as a means to understand the mechanism of costimulatory molecule up-regulation during infection. Therefore, mice were orally infected with Salmonella expressing eGFP or the parental strain lacking eGFP, and the uptake of Salmonella by CD11c-expressing cells was determined by flow cytometry after 3 days. In PP, the greatest fraction of eGFP+ cells was observed in the CD11cintCD11b+ population (0.7–1.6%; Fig. 5,A). Also, conventional CD8α DC associated with eGFP-expressing Salmonella (0.4–1.0%), although at a reduced frequency compared with CD11cintCD11b+ cells (Fig. 5,B). In contrast, few, if any, CD8α+ DC in PP were directly associated with Salmonella, as ≤0.05% of the CD8α+ DC were eGFP+, a value close to that detected in mice infected with the parental strain lacking eGFP (Fig. 5 C). Likewise, CD8α+ DC in PP did not associate with eGFP-expressing Salmonella at 1 or 2 days after infection (data not shown).

In MLN, Salmonella were associated with CD11cintCD11b+ cells, with the frequency of eGFP+ cells being 0.1–0.2% (Fig. 5,D). The frequency of eGFP+ events among CD8α+ and CD8α conventional DC was <0.05% in this organ at 1, 2, and 3 days after infection (Fig. 5, E and F, and data not shown). Thus, a very low proportion of CD11c-expressing cells associated with Salmonella after oral infection, and the CD11cintCD11b+ cells constituted the main subset of CD11c-expressing cells that take up Salmonella in vivo with some contribution by conventional CD8α DC.

The low frequency of CD11c+ cells directly associated with Salmonella (Fig. 5); the simultaneous up-regulation of costimulatory molecules in PP, MLN, and spleen (Fig. 3,A); and the homogenous up-regulation of costimulatory molecules on all cells within a given subset (Fig. 1 and Table II), suggest that up-regulation of costimulatory molecules on DC and CD11cintCD11b+ cells occurs independently of bacterial uptake. We thus hypothesized that a soluble mediator was responsible for the observed up-regulation of CD80, CD86, and CD40 and searched for the presence of cytokines that could influence the expression of costimulatory molecules in lysates of spleen and MLN from Salmonella-infected mice (8, 9, 10, 11, 33).

Both the spleen and MLN of mice infected 5 days previously with Salmonella contained IL-1β and low levels of GM-CSF (Fig. 6,A). However, although significantly higher levels of TNF-α were found in lysates of spleen and PP from infected compared with naive mice (data not shown), MLN lysates appeared to contain a factor(s) that interfered with the detection of TNF-α. We thus determined the ability of splenic and MLN cells from naive or Salmonella-infected mice to secrete TNF-α after a 24-h ex vivo culture in medium alone or in medium supplemented with Salmonella lysate. Both splenic and MLN cells from mice infected 5 days previously secreted TNF-α, whereas naive mice did not produce this cytokine (Fig. 6,B). Moreover, TNF-α secretion was enhanced in naive and infected mice when the cultures were supplemented with Salmonella lysate (Fig. 6 B). Splenic cells appeared to be more efficient at secreting TNF-α compared with MLN cells, which is partially explained by the higher frequency of TNF-α-secreting cells in the spleen (A. Rydström and M. J. Wick, manuscript in preparation). Together these data show that IL-1β, TNF-α, and GM-CSF, which can induce DC maturation, are present in spleen and MLN after Salmonella infection.

FIGURE 6.

TNF-α, IL-1β, and GM-CSF are present in spleen and MLN of Salmonella-infected mice. A, Levels of IL-1β and GM-CSF in splenic and MLN lysates of naive mice (□) or mice infected 5 days earlier with χ4666 (▪). Organ lysates were prepared directly ex vivo, and cytokines were quantitated by Bio-Plex. Each bar represents the mean value of seven mice, and error bars are the SEM. B, Splenic and MLN cells from naive or 5-day-infected mice were cultured for 24 h in the presence or the absence of Salmonella lysate. Secretion of TNF-α in supernatants was measured by ELISA. Mean values from six or seven mice are shown. Error bars are the SEM. The values of p were obtained by Student’s t test. ∗, p < 0.05; ∗∗, p < 0.01. Data are representative of two independent experiments.

FIGURE 6.

TNF-α, IL-1β, and GM-CSF are present in spleen and MLN of Salmonella-infected mice. A, Levels of IL-1β and GM-CSF in splenic and MLN lysates of naive mice (□) or mice infected 5 days earlier with χ4666 (▪). Organ lysates were prepared directly ex vivo, and cytokines were quantitated by Bio-Plex. Each bar represents the mean value of seven mice, and error bars are the SEM. B, Splenic and MLN cells from naive or 5-day-infected mice were cultured for 24 h in the presence or the absence of Salmonella lysate. Secretion of TNF-α in supernatants was measured by ELISA. Mean values from six or seven mice are shown. Error bars are the SEM. The values of p were obtained by Student’s t test. ∗, p < 0.05; ∗∗, p < 0.01. Data are representative of two independent experiments.

Close modal

The high levels of TNF-α and IL-1β in spleen and MLN of Salmonella-infected mice prompted us to investigate the contributions of these cytokines to the up-regulation of costimulatory molecules during infection. To this end, TNFRI−/− mice and caspase-1−/− mice that cannot respond to TNF-α-mediated signaling via TNFRI or synthesize active IL-1β, respectively, were orally infected with Salmonella. After 5 days, the maturation status of the conventional DC subsets and CD11cintCD11b+ cells was determined by flow cytometry. Both TNFRI−/− and caspase-1−/− mice were more susceptible to Salmonella infection than wild-type animals, and bacterial numbers increased to a greater extent in TNF-α-deficient hosts (27) and in caspase-1−/− mice compared with caspase-1+/− controls (data not shown). To examine the roles of TNF-α and IL-1β in costimulatory molecule up-regulation without the confounding factor of higher bacterial burden in the knockout mice, only TNFRI−/− or caspase-1−/− mice with a bacterial load similar to that of wild-type mice were assessed (Fig. 7, B and F, and data not shown).

Infected TNFRI−/− mice showed a significant inhibition of the up-regulation of CD80, CD86, and CD40 on conventional CD8α DC of MLN (Fig. 7,A). Furthermore, the up-regulation of CD80 on conventional CD8α+ DC was significantly reduced in MLN of infected TNFRI−/− mice (Fig. 7,A). The CD11cintCD11b+ cells of MLN appeared less dependent on TNFRI for up-regulation of costimulatory molecules than conventional DC. In the absence of TNFRI, CD11cintCD11b+ cells showed a reduced up-regulation of CD80, but it was not statistically significant (Fig. 7,A). Likewise, the up-regulation of CD40 on CD11cintCD11b+ cells was not significantly inhibited in infected TNFRI−/−. However, the apparent reduction in the level of CD86 on these cells in the MLN (Figs. 3,A and 4,A) was less dramatic in infected TNFRI−/− mice compared with infected C57BL/6 mice. This could imply a defective recruitment of CD11cintCD11b+ cells to MLN in the absence of TNFRI signaling, because the reduced level of CD86 is probably due to dilution of CD86high cells in the CD11cintCD11b+ gate in naive mice by CD11cintCD11b+ cells recruited during infection. Indeed, the absolute number of CD11cintCD11b+ cells in MLN was significantly lower in infected TNFRI−/− mice compared with C57BL/6 mice (Fig. 7 C).

In contrast to MLN, splenic conventional DC showed less dependence on TNFRI for up-regulation of CD80, CD86, and CD40 (Fig. 7,E). On the contrary, CD8α+ DC from TNFRI−/− expressed a higher level of CD86 than C57BL/6 mice despite a similar bacterial load (Fig. 7, E and F). However, the up-regulation of CD80 on CD11cintCD11b+ cells was significantly inhibited in the spleens of infected TNFRI−/− mice compared with C57BL/6 mice (Fig. 7,E). As in MLN, the recruitment of splenic CD11cintCD11b+ cells was defective in infected TNFRI−/− mice, although not to the same extent (Fig. 7 G).

Infected TNFRI−/− mice showed a defective up-regulation of costimulatory molecules, particularly in MLN. However, caspase-1−/− mice showed no significant reduction in costimulatory molecule up-regulation on conventional CD8α+ and CD8α DC subsets or CD11cintCD11b+ cells in both spleen and MLN compared with caspase-1+/− mice with a similar bacterial load (n ≥ 12; data not shown). This suggested a predominant role of TNF-α, rather than IL-1β, in mediating costimulatory molecule up-regulation in MLN. However, because TNF-α is also involved in the recruitment of inflammatory cells to sites of infection (34) (Fig. 7, C and G), there was a possibility that other cytokines capable of inducing DC maturation failed to be induced in TNFRI−/− mice. Considering the high level of IL-1β in infected wild-type mice (Fig. 6,A), we determined whether the production of IL-1β was affected in the absence of TNFRI. We found that IL-1β was produced at an unaltered level in the spleen (Fig. 7,H), whereas five of six TNFRI−/− mice contained naive levels of this cytokine in the MLN (Fig. 7,D). In contrast, TNFRI−/− mice with a bacterial load higher than wild-type mice produced substantial amounts of this cytokine (mean, 2200 pg/mg protein) and up-regulated CD80 and CD40 on CD8α+ and CD8α DC in MLN to a similar extent as wild-type mice (Table III). The up-regulation of CD86 on CD8α DC of MLN was still defective, however, in the TNFRI−/− mice with a high bacterial load despite the presence of IL-1β (Table III).

Table III.

Fold up-regulation of costimulatory molecules on DC subsets in MLN of TNFRI−/− mice with a bacterial load higher than wild-type micea

MiceCD8α+ DCCD8α DC
CD80CD86CD40CD80CD86CD40
C57BL/6 1.4 ± 0.08b 1.4 ± 0.1 2.1 ± 0.1 1.7 ± 0.2 3.6 ± 0.2 1.9 ± 0.09 
TNFRI−/− 1.4 ± 0.07 1.4 ± 0.07 2.4 ± 0.2 2.1 ± 0.3 2.8 ± 0.3 1.7 ± 0.04 
MiceCD8α+ DCCD8α DC
CD80CD86CD40CD80CD86CD40
C57BL/6 1.4 ± 0.08b 1.4 ± 0.1 2.1 ± 0.1 1.7 ± 0.2 3.6 ± 0.2 1.9 ± 0.09 
TNFRI−/− 1.4 ± 0.07 1.4 ± 0.07 2.4 ± 0.2 2.1 ± 0.3 2.8 ± 0.3 1.7 ± 0.04 
a

Mice were orally infected 5 days earlier with Salmonella. C57BL/6 mice (data are from Fig. 7 A) had a bacterial burden of 3.5–4.5 log10 in MLN. TNFRI−/− mice (n = 4) had a bacterial burden of 5.2–6.9 log10 in MLN.

b

Values show fold up-regulation of CD80, CD86 or CD40 on DC from infected mice compared to naive mice ± SEM.

Together, these data show that the up-regulation of CD80 and CD86 on conventional DC subsets in MLN is TNFRI dependent, whereas splenic DC mature predominantly independently of TNFRI. In addition, production of IL-1β is defective in MLN, but not spleen, of TNFRI−/− mice, which might explain the predominant role of TNFRI in mediating costimulatory molecule up-regulation in MLN. Moreover, recruitment of CD11cintCD11b+ cells to both MLN and spleen is compromised in the absence of TNFRI signaling.

The data obtained strongly suggest that proinflammatory cytokines produced during infection mediate up-regulation of costimulatory molecules on conventional DC and CD11cintCD11b+ cells regardless of whether they contain bacteria. To directly address the influence of bacterial association on costimulatory molecule up-regulation, C57BL/6 mice were infected with eGFP-expressing Salmonella, and the expression of CD80, CD86, and CD40 were compared on eGFP+ vs eGFP CD11c-expressing cells. After oral infection with Salmonella, bacteria mainly associated with CD11cintCD11b+ cells in PP and MLN as well as with CD11chighCD8α DC in PP (Fig. 5). Therefore, a sufficient number of eGFP+ events for reliable analysis could be acquired only from these populations. Analysis of costimulatory molecule expression on conventional CD8α DC from PP as well as CD11cintCD11b+ cells from both PP and MLN revealed no difference in CD80, CD86, or CD40 expression on eGFP+ compared with eGFP cells (Fig. 8 A). The fold difference in costimulatory molecule expression between eGFP+ and eGFP cells was <1.4 for all markers in wild-type mice. Thus, the uptake of bacteria per se by CD11cintCD11b+ cells or CD11chighCD8α DC in infected immunocompetent mice does not have a dramatic effect on costimulatory molecule expression.

To be able to discriminate up-regulation of costimulatory molecules induced indirectly by proinflammatory cytokines from that induced directly by bacterial association, we performed a similar analysis with eGFP-expressing Salmonella in TNFRI−/− mice. Based on the reduced indirect maturation observed in TNFRI−/− mice (Fig. 7,A), we reasoned that a mechanism of maturation dependent on bacterial association might become apparent in these animals. Sufficient numbers of eGFP+ events could be acquired from all three CD11c-expressing populations in the MLN of infected TNFRI−/− mice. In contrast to C57BL/6 mice, eGFP+ cells from TNFRI−/− mice expressed 3- to 7-fold higher levels of CD80 and CD86 than their eGFP counterparts in all three CD11c+ populations (Fig. 8,B). The level of CD40 was the least affected by bacterial association, because eGFP+ cells expressed a ≤2-fold higher level of CD40 than their eGFP counterparts (Fig. 8,B). Two factors contributed to the higher fold difference in CD80 and CD86 expression between eGFP+ and eGFP populations in TNFRI−/− mice. First, eGFP cells of infected TNFRI−/− mice showed a defective up-regulation of costimulatory molecules compared with eGFP cells of infected wild-type mice (Fig. 8). Second, eGFP+ cells of infected TNFRI−/− mice showed a greater up-regulation of costimulatory molecules compared with eGFP+ cells of infected wild-type mice (Fig. 8). This raises the possibility that TNFRI signaling might retard the up-regulation of costimulatory molecules on Salmonella-containing cells in wild-type mice.

Thus, bacteria-containing CD11c-expressing cells can up-regulate costimulatory molecules independently of TNFRI. This suggests that during Salmonella infection there exist both direct (dependent on bacterial uptake) and indirect (TNFRI-dependent) pathways of mediating costimulatory molecule up-regulation on DC and CD11cintCD11b+ cells.

Our results show that costimulatory molecules are up-regulated in a subset-specific manner in PP, MLN, and spleen after oral Salmonella infection. Thus, CD80 was greatly up-regulated on CD8α DC, whereas little, if any, up-regulation of this surface molecule occurred on CD8α+ DC. Similarly, CD8α DC showed the greatest infection-induced change in CD86 expression. However, CD8α+ DC had a higher steady-state level of CD86, so the net result was similar MFIs for this costimulatory molecule on CD8α+ and CD8α DC after infection. Both DC subsets in the three organs examined also up-regulated CD40 in response to Salmonella infection, with the greatest fold increase and the highest MFI occurring in the CD8α+ subset.

Subset-specific up-regulation of costimulatory molecules has been observed in other in vivo models, including after administration of α-galactoceramide or TLR ligands and during infection with certain parasites or viruses (11, 23, 25, 35, 36, 37). The differential capacities of DC subsets to up-regulate CD80, CD86, or CD40 in response to microbial infection could be due to differential expression of TLRs (5, 38, 39). However, TLR-deficient DC up-regulate costimulatory molecules as well as TLR-expressing DC in mixed bone marrow chimeras after administration of the TLR agonist (22). Similarly, i.v. administration of a TLR7 agonist induces up-regulation of costimulatory molecules on both CD8α+ and CD8α DC (23, 38) despite differential TLR7 expression on these subsets detected at the mRNA level (38, 39). Thus, the subset-specific up-regulation of costimulatory molecules could be dependent on a differential capacity of DC subsets to respond to proinflammatory cytokines (11), rather than a differential expression of TLRs.

Expression of costimulatory molecules is required for T cell activation. However, it has recently been shown that up-regulation of costimulatory molecules does not necessarily result in DC capable of eliciting T cell immunity in vivo (11, 22). Thus, maturation of DC into fully immunogenic cells may require CD40 engagement and/or distinct chemokine and cytokine production in addition to costimulatory molecules (2, 3, 11, 22). Despite the multiple requirements to generate immunogenic DC, the subset-specific up-regulation of costimulatory molecules and CD40 may modulate the antibacterial immune response. Indeed, subset-specific differences in costimulatory molecule function have been reported. For example, blocking CD40 signaling selectively inhibits the ability of CD8α+ DC to induce Th1 responses, but does not influence a CD8α DC-driven Th2 response to protein Ags (40). The higher steady-state level and the greatest up-regulation of CD40 on CD8α+ DC after Salmonella infection may influence the capacity of this subset to direct the immune response to infection, particularly in light of the importance of CD40 ligation to modulating DC functions such as IL-12p70 production, CD80 and CD86 up-regulation, and cross-priming of CD8+ T cells (6, 40). Furthermore, the striking difference in CD80 up-regulation on CD8α vs CD8α+ DC raises the possibility of subset-specific modulation of T cell activation. Interestingly, the expression of CD80 on APC preferentially concentrates the inhibitory receptor CTLA-4 at the immunological synapse, whereas CD86 preferentially recruits CD28 (41). Furthermore, CD80 has a higher affinity for CTLA-4 compared with CD86 and favors binding to CTLA-4 over CD28, whereas CD86 displays much less bias (42). Moreover, interaction of CD80 or CD86 with CTLA-4 has immunosuppressive effects on the ensuing immune response (43), whereas their interaction with CD28 results in activated and immunostimulatory DC (44). Although the functional significance of differential expression of costimulatory molecules on DC subsets after Salmonella infection is presently not known, the high level of CD80 on CD8α DC and CD11cintCD11b+ cells may serve to down-modulate the antibacterial immune response by engaging CTLA-4 after initial T cell expansion has occurred.

Oral Salmonella infection also recruited CD11cintCD11b+ cells expressing CD80, CD86, CD40, and MHC-II at levels similar to those on conventional (CD11chigh) DC to PP, MLN, and spleen. No other cell type expressed such high levels of costimulatory molecules (Fig. 2,A and Table II). We also found the highest fraction of bacteria-containing CD11c-expressing cells in the CD11cintCD11b+ population. The high expression of CD80, CD86, and MHC-II on CD11cintCD11b+ cells as well as their efficient uptake of Salmonella in vivo suggest that they might function as APCs during infection. Indeed, cells expressing CD11b and intermediate levels of CD11c have been described in other infection models and after the onset of diabetes (45, 46, 47, 48, 49). CD11cintCD11b+ cells appear to be recruited from the blood (48), have a strong phagocytic ability (48, 49), and, in Listeria-infected mice, produce TNF-α and iNOS and can activate T cells (45). During Salmonella infection, the CD11cintCD11b+ cells constitute, together with macrophages, neutrophils, and DC, a source of TNF-α (15, 50) (Fig. 2 G). The collective features of the CD11cintCD11b+ cells suggest that they are recruited monocytes differentiating along a DC-macrophage axis.

Our data showed that <1% of CD11c+ cells in the PP and MLN associated with eGFP-expressing Salmonella in the first few days after oral infection, and that CD11cintCD11b+ cells contained the highest frequency of eGFP+ events. CD8α DC also contributed to bacterial uptake in PP, whereas in MLN, Salmonella association with the conventional DC subsets was not detected. The inability of CD8α+ DC in PP to associate with Salmonella after oral infection is not due to an inability of this subset to phagocytose the bacteria, because both CD8α+ and CD8α DC associate with Salmonella after i.v. infection (16, 17). It is possible that the location of CD8α+ DC in T cell areas of PP (51) reduces their access to orally acquired bacteria.

Despite the low number of CD11c-expressing cells associated with Salmonella and the subset-specific difference in acquiring bacteria in infected organs, up-regulation of costimulatory molecules occurred on the entire population of DC subsets or CD11cintCD11b+ cells. Our data show that TNFRI signaling is important for mediating costimulatory molecule up-regulation after oral infection. TNFRI-mediated costimulatory molecule up-regulation was most apparent on conventional DC, particularly the CD8α subset, and to a lesser extent on the recruited CD11cintCD11b+ cells. The TNFRI-mediated effects were also most striking in MLN compared with spleen of infected TNFRI−/− mice. The organ-specific dependence on TNFRI for costimulatory molecule up-regulation is probably explained by the reduced IL-1β production in MLN, but not spleen, of infected TNFRI−/− mice. Keeping in mind the unimpeded up-regulation of costimulatory molecules in caspase−/− mice, together the data suggest that the absence of either TNF-α- or IL-1β-mediated signaling had no major impact on costimulatory molecule expression. Only when both cytokines were absent, which was the case in MLN, but not spleen, of infected TNFRI−/− mice, was costimulatory molecule up-regulation inhibited. The lower number of CD11clowCD11b+ neutrophils and macrophages in naive MLN compared with naive spleens (Fig. 2, B and D) might make this organ more dependent on the TNFRI-mediated recruitment of these cells, which are potential sources of IL-1β (52).

The individual as well as synergistic effects of IL-1β and TNF-α on maturation of DC subsets and CD11cintCD11b+ cells during Salmonella infection require additional studies. However, mice with a bacterial load higher than wild-type mice (Table III) were dependent on TNFRI, but not IL-1β, for up-regulation of CD86 on CD8α DC of MLN, indicating the existence of direct TNF-α-mediated effects. This is consistent with data showing a greater influence of TNF-α on CD86 expression on CD8α relative to CD8α+ DC in vivo (11). In addition to TNF-α and IL-1β, low levels of GM-CSF were produced after oral Salmonella infection. Although the level of GM-CSF was significantly elevated compared with that in naive mice, the total amount of GM-CSF in spleen and MLN was low (<10 pg/spleen and <5 pg/MLN) compared with the levels used for in vitro culture or activation of DC (53). Thus, the local cytokine milieu influences DC maturation during microbial infection, and a differential cytokine content in different organs adds to the complexity of the analysis.

The TNFRI-mediated signaling and production of proinflammatory cytokines indirectly induced costimulatory molecule up-regulation on DC and CD11cintCD11b+ cells of MLN, regardless of whether they were directly associated with bacteria. To study how direct bacterial association affects costimulatory molecules, we used six-color flow cytometric analysis of CD11c-expressing cells that either were or were not associated with Salmonella in vivo. We found that bacterial association did not lead to differential expression of costimulatory molecules in wild-type mice. However, the reduced indirect maturation in TNFRI−/− mice allowed us to discriminate between cytokine-mediated costimulatory molecule up-regulation and that induced directly by bacterial association. These studies revealed that Salmonella-containing cells could mature independently of TNFRI. Thus, although TNFRI signaling indirectly induces costimulatory molecule up-regulation during Salmonella infection, DC or CD11cintCD11b+ cells directly associated with bacteria can up-regulate CD80 and CD86 via a TNFRI-independent pathway. The mechanism of direct maturation of Salmonella-containing cells remains to be elucidated, although it probably involves direct TLR signaling.

Recent studies suggest that there are functional differences between DC that have been induced to mature indirectly via proinflammatory cytokines and DC that, in addition to proinflammatory cytokines, have received a direct TLR-mediated maturation stimulus (22). Although both the indirect and direct pathways lead to up-regulation of costimulatory molecules, direct TLR triggering is required for IL-12 production by DC and eliciting DC that induce differentiation of CD4+ T cells into IFN-γ-producing Th1 effectors (22). In addition, it has been shown that the up-regulation of costimulatory molecules on DC after NKT cell activation by α-galactoceramide is mediated by proinflammatory cytokines (11). However, despite costimulatory molecule up-regulation, Ag presentation and production of proinflammatory cytokines, DC lacked immunogenicity in the absence of CD40 triggering (11). This suggests that inflammation per se cannot induce complete DC maturation, and indirect maturation in the absence of direct TLR signals or CD40 might produce DC that induce tolerance rather than immunity. Accordingly, bone marrow-derived DC treated with TNF-α induce Ag-specific protection from autoimmune disease despite high expression of costimulatory molecules (54). Thus, full T cell activation requires aspects of DC maturation other than high levels of costimulatory molecules, such as the CD40-CD40L pathway or DC cytokine production induced by direct TLR triggering (11, 22). The influence of direct bacterial association on DC cytokine production during in vivo Salmonella infection is presently not known. However, in vitro, only a fraction of Salmonella-containing DC produces TNF-α or IL-12, and DC not directly associated with bacteria contribute substantially to cytokine production (16, 17).

The results presented in this study suggest that during Salmonella infection, exposure of DC or CD11cintCD11b+ cells to proinflammatory cytokines indirectly induces up-regulation of costimulatory molecules regardless of whether the cells are directly associated with bacteria. Moreover, a TNFRI-independent pathway for direct maturation of Salmonella-containing cells occurs in vivo. Dissecting the relative function of cytokine-matured vs directly matured DC and CD11cintCD11b+ cells during the initial stages of Salmonella infection will provide insight into the complex interplay of both bacteria-derived signaling molecules and the host factors they elicit in steering adaptive immunity to bacteria.

We gratefully acknowledge the technical assistance of Kristina Lindgren. We are grateful to Arturo Zychlinsky for providing caspase-1−/− mice and S. typhimurium SL1344 and SMO22.

The authors have no financial conflict of interest.

The costs of publication of this article were defrayed in part by the payment of page charges. This article must therefore be hereby marked advertisement in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.

1

This work was supported by the Swedish Research Council (621-2004-1378) and the Sahlgrenska Academy at Goteborg University.

3

Abbreviations used in this paper: DC, dendritic cell; 7AAD, 7-aminoactinomycin D; eGFP, enhanced GFP; MFI, median fluorescence intensity; MLN, mesenteric lymph node; PP, Peyer’s patch; TNFRI, TNFR type I.

1
Wilson, N. S., D. El-Sukkari, G. T. Belz, C. M. Smith, R. J. Steptoe, W. R. Heath, K. Shortman, J. A. Villadangos.
2003
. Most lymphoid organ dendritic cell types are phenotypically and functionally immature.
Blood
102
:
2187
.-2195.
2
Steinman, R. M., D. Hawiger, M. C. Nussenzweig.
2003
. Tolerogenic dendritic cells.
Annu. Rev. Immunol.
21
:
685
.-711.
3
Banchereau, J., R. M. Steinman.
1998
. Dendritic cells and the control of immunity.
Nature
392
:
245
.-252.
4
Shortman, K., Y. J. Liu.
2002
. Mouse and human dendritic cell subtypes.
Nat. Rev. Immunol.
2
:
151
.-161.
5
Reis e Sousa, C..
2004
. Toll-like receptors and dendritic cells: for whom the bug tolls.
Semin. Immunol.
16
:
27
.-34.
6
Quezada, S. A., L. Z. Jarvinen, E. F. Lind, R. J. Noelle.
2004
. CD40/CD154 interactions at the interface of tolerance and immunity.
Annu. Rev. Immunol.
22
:
307
.-328.
7
Sundquist, M., C. Johansson, M. J. Wick.
2003
. Dendritic cells as inducers of antimicrobial immunity in vivo.
APMIS
111
:
715
.-724.
8
Winzler, C., P. Rovere, M. Rescigno, F. Granucci, G. Penna, L. Adorini, V. S. Zimmermann, J. Davoust, P. Ricciardi-Castagnoli.
1997
. Maturation stages of mouse dendritic cells in growth factor-dependent long-term cultures.
J. Exp. Med.
185
:
317
.-328.
9
Sallusto, F., A. Lanzavecchia.
1994
. Efficient presentation of soluble antigen by cultured human dendritic cells is maintained by granulocyte/macrophage colony-stimulating factor plus interleukin 4 and downregulated by tumor necrosis factor α.
J. Exp. Med.
179
:
1109
.-1118.
10
Trevejo, J. M., M. W. Marino, N. Philpott, R. Josien, E. C. Richards, K. B. Elkon, E. Falck-Pedersen.
2001
. TNF-α-dependent maturation of local dendritic cells is critical for activating the adaptive immune response to virus infection.
Proc. Natl. Acad. Sci. USA
98
:
12162
.-12167.
11
Fujii, S., K. Liu, C. Smith, A. J. Bonito, R. M. Steinman.
2004
. The linkage of innate to adaptive immunity via maturing dendritic cells in vivo requires CD40 ligation in addition to antigen presentation and CD80/86 costimulation.
J. Exp. Med.
199
:
1607
.-1618.
12
Pulendran, B..
2005
. Variegation of the immune response with dendritic cells and pathogen recognition receptors.
J. Immunol.
174
:
2457
.-2465.
13
Niess, J. H., S. Brand, X. Gu, L. Landsman, S. Jung, B. A. McCormick, J. M. Vyas, M. Boes, H. L. Ploegh, J. G. Fox, et al
2005
. CX3CR1-mediated dendritic cell access to the intestinal lumen and bacterial clearance.
Science
307
:
254
.-258.
14
Sundquist, M., A. Rydstrom, M. J. Wick.
2004
. Immunity to Salmonella from a dendritic point of view.
Cell. Microbiol.
6
:
1
.-11.
15
Kirby, A. C., U. Yrlid, M. Svensson, M. J. Wick.
2001
. Differential involvement of dendritic cell subsets during acute Salmonella infection.
J. Immunol.
166
:
6802
.-6811.
16
Yrlid, U., M. J. Wick.
2002
. Antigen presentation capacity and cytokine production by murine splenic dendritic cell subsets upon Salmonella encounter.
J. Immunol.
169
:
108
.-116.
17
Johansson, C., M. J. Wick.
2004
. Liver dendritic cells present bacterial antigens and produce cytokines upon Salmonella encounter.
J. Immunol.
172
:
2496
.-2503.
18
McSorley, S. J., S. Asch, M. Costalonga, R. L. Reinhardt, M. K. Jenkins.
2002
. Tracking Salmonella-specific CD4 T cells in vivo reveals a local mucosal response to a disseminated infection.
Immunity
16
:
365
.-377.
19
Weiss, D. S., B. Raupach, K. Takeda, S. Akira, A. Zychlinsky.
2004
. Toll-like receptors are temporally involved in host defense.
J. Immunol.
172
:
4463
.-4469.
20
Vazquez-Torres, A., B. A. Vallance, M. A. Bergman, B. B. Finlay, B. T. Cookson, J. Jones-Carson, F. C. Fang.
2004
. Toll-like receptor 4 dependence of innate and adaptive immunity to Salmonella: importance of the Kupffer cell network.
J. Immunol.
172
:
6202
.-6208.
21
Lembo, A., C. Kalis, C. J. Kirschning, V. Mitolo, E. Jirillo, H. Wagner, C. Galanos, M. A. Freudenberg.
2003
. Differential contribution of Toll-like receptors 4 and 2 to the cytokine response to Salmonella enterica serovar Typhimurium and Staphylococcus aureus in mice.
Infect. Immun.
71
:
6058
.-6062.
22
Spörri, R., C. Reis e Sousa.
2005
. Inflammatory mediators are insufficient for full dendritic cell activation and promote expansion of CD4+ T cell populations lacking helper function.
Nat. Immunol.
6
:
163
.-170.
23
Asselin-Paturel, C., G. Brizard, K. Chemin, A. Boonstra, A. O’Garra, A. Vicari, G. Trinchieri.
2005
. Type I interferon dependence of plasmacytoid dendritic cell activation and migration.
J. Exp. Med.
201
:
1157
.-1167.
24
De Smedt, T., B. Pajak, E. Muraille, L. Lespagnard, E. Heinen, P. De Baetselier, J. Urbain, O. Leo, M. Moser.
1996
. Regulation of dendritic cell numbers and maturation by lipopolysaccharide in vivo.
J. Exp. Med.
184
:
1413
.-1424.
25
Pulendran, B., P. Kumar, C. W. Cutler, M. Mohamadzadeh, T. Van Dyke, J. Banchereau.
2001
. Lipopolysaccharides from distinct pathogens induce different classes of immune responses in vivo.
J. Immunol.
167
:
5067
.-5076.
26
Didierlaurent, A., I. Ferrero, L. A. Otten, B. Dubois, M. Reinhardt, H. Carlsen, R. Blomhoff, S. Akira, J. P. Kraehenbuhl, J. C. Sirard.
2004
. Flagellin promotes myeloid differentiation factor 88-dependent development of Th2-type response.
J. Immunol.
172
:
6922
.-6930.
27
Mastroeni, P..
2002
. Immunity to systemic Salmonella infections.
Curr. Mol. Med.
2
:
393
.-406.
28
Li, P., H. Allen, S. Banerjee, S. Franklin, L. Herzog, C. Johnston, J. McDowell, M. Paskind, L. Rodman, J. Salfeld, et al
1995
. Mice deficient in IL-1β-converting enzyme are defective in production of mature IL-1β and resistant to endotoxic shock.
Cell
80
:
401
.-411.
29
Kirby, A. C., M. Sundquist, M. J. Wick.
2004
. In vivo compartmentalization of functionally distinct, rapidly responsive antigen-specific T-cell populations in DNA-immunized or Salmonella enterica serovar Typhimurium-infected mice.
Infect. Immun.
72
:
6390
.-6400.
30
Valdivia, R. H., S. Falkow.
1996
. Bacterial genetics by flow cytometry: rapid isolation of Salmonella typhimurium acid-inducible promoters by differential fluorescence induction.
Mol. Microbiol.
22
:
367
.-378.
31
Svensson, M., B. Stockinger, M. J. Wick.
1997
. Bone marrow-derived dendritic cells can process bacteria for MHC-I and MHC-II presentation to T cells.
J. Immunol.
158
:
4229
.-4236.
32
Colonna, M., G. Trinchieri, Y. J. Liu.
2004
. Plasmacytoid dendritic cells in immunity.
Nat. Immunol.
5
:
1219
.-1226.
33
Larsen, C. P., S. C. Ritchie, R. Hendrix, P. S. Linsley, K. S. Hathcock, R. J. Hodes, R. P. Lowry, T. C. Pearson.
1994
. Regulation of immunostimulatory function and costimulatory molecule (B7-1 and B7-2) expression on murine dendritic cells.
J. Immunol.
152
:
5208
.-5219.
34
Sedgwick, J. D., D. S. Riminton, J. G. Cyster, H. Korner.
2000
. Tumor necrosis factor: a master-regulator of leukocyte movement.
Immunol. Today
21
:
110
.-113.
35
Straw, A. D., A. S. MacDonald, E. Y. Denkers, E. J. Pearce.
2003
. CD154 plays a central role in regulating dendritic cell activation during infections that induce Th1 or Th2 responses.
J. Immunol.
170
:
727
.-734.
36
Montoya, M., G. Schiavoni, F. Mattei, I. Gresser, F. Belardelli, P. Borrow, D. F. Tough.
2002
. Type I interferons produced by dendritic cells promote their phenotypic and functional activation.
Blood
99
:
3263
.-3271.
37
Dalod, M., T. Hamilton, R. Salomon, T. P. Salazar-Mather, S. C. Henry, J. D. Hamilton, C. A. Biron.
2003
. Dendritic cell responses to early murine cytomegalovirus infection: subset functional specialization and differential regulation by interferon α/β.
J. Exp. Med.
197
:
885
.-898.
38
Doxsee, C. L., T. R. Riter, M. J. Reiter, S. J. Gibson, J. P. Vasilakos, R. M. Kedl.
2003
. The immune response modifier and Toll-like receptor 7 agonist S-27609 selectively induces IL-12 and TNF-α production in CD11c+CD11b+CD8 dendritic cells.
J. Immunol.
171
:
1156
.-1163.
39
Edwards, A. D., S. S. Diebold, E. M. Slack, H. Tomizawa, H. Hemmi, T. Kaisho, S. Akira, C. Reis e Sousa.
2003
. Toll-like receptor expression in murine DC subsets: lack of TLR7 expression by CD8α+ DC correlates with unresponsiveness to imidazoquinolines.
Eur. J. Immunol.
33
:
827
.-833.
40
Yasumi, T., K. Katamura, T. Yoshioka, T. A. Meguro, R. Nishikomori, T. Heike, M. Inobe, S. Kon, T. Uede, T. Nakahata.
2004
. Differential requirement for the CD40-CD154 costimulatory pathway during Th cell priming by CD8α+ and CD8α murine dendritic cell subsets.
J. Immunol.
172
:
4826
.-4833.
41
Pentcheva-Hoang, T., J. G. Egen, K. Wojnoonski, J. P. Allison.
2004
. B7-1 and B7-2 selectively recruit CTLA-4 and CD28 to the immunological synapse.
Immunity
21
:
401
.-413.
42
Collins, A. V., D. W. Brodie, R. J. Gilbert, A. Iaboni, R. Manso-Sancho, B. Walse, D. I. Stuart, P. A. van der Merwe, S. J. Davis.
2002
. The interaction properties of costimulatory molecules revisited.
Immunity
17
:
201
.-210.
43
Grohmann, U., C. Orabona, F. Fallarino, C. Vacca, F. Calcinaro, A. Falorni, P. Candeloro, M. L. Belladonna, R. Bianchi, M. C. Fioretti, et al
2002
. CTLA-4-Ig regulates tryptophan catabolism in vivo.
Nat. Immunol.
3
:
1097
.-1101.
44
Orabona, C., U. Grohmann, M. L. Belladonna, F. Fallarino, C. Vacca, R. Bianchi, S. Bozza, C. Volpi, B. L. Salomon, M. C. Fioretti, et al
2004
. CD28 induces immunostimulatory signals in dendritic cells via CD80 and CD86.
Nat. Immunol.
5
:
1134
.-1142.
45
Serbina, N. V., T. P. Salazar-Mather, C. A. Biron, W. A. Kuziel, E. G. Pamer.
2003
. TNF/iNOS-producing dendritic cells mediate innate immune defense against bacterial infection.
Immunity
19
:
59
.-70.
46
Gonzalez-Juarrero, M., T. S. Shim, A. Kipnis, A. P. Junqueira-Kipnis, I. M. Orme.
2003
. Dynamics of macrophage cell populations during murine pulmonary tuberculosis.
J. Immunol.
171
:
3128
.-3135.
47
Homann, D., D. B. McGavern, M. B. Oldstone.
2004
. Visualizing the viral burden: phenotypic and functional alterations of T cells and APCs during persistent infection.
J. Immunol.
172
:
6239
.-6250.
48
Balazs, M., F. Martin, T. Zhou, J. Kearney.
2002
. Blood dendritic cells interact with splenic marginal zone B cells to initiate T-independent immune responses.
Immunity
17
:
341
.-352.
49
Turley, S., L. Poirot, M. Hattori, C. Benoist, D. Mathis.
2003
. Physiological β cell death triggers priming of self-reactive T cells by dendritic cells in a type-1 diabetes model.
J. Exp. Med.
198
:
1527
.-1537.
50
Kirby, A. C., U. Yrlid, M. J. Wick.
2002
. The innate immune response differs in primary and secondary Salmonella infection.
J. Immunol.
169
:
4450
.-4459.
51
Iwasaki, A., B. L. Kelsall.
2000
. Localization of distinct Peyer’s patch dendritic cell subsets and their recruitment by chemokines macrophage inflammatory protein (MIP)-3α, MIP-3β, and secondary lymphoid organ chemokine.
J. Exp. Med.
191
:
1381
.-1394.
52
Duncan, L. M., L. S. Meegan, E. R. Unanue.
1991
. IL-1 gene expression in lymphoid tissues.
J. Immunol.
146
:
565
.-571.
53
Chang, D. Z., W. Lomazow, C. Joy Somberg, R. Stan, M. A. Perales.
2004
. Granulocyte-macrophage colony stimulating factor: an adjuvant for cancer vaccines.
Hematology
9
:
207
.-215.
54
Menges, M., S. Rossner, C. Voigtlander, H. Schindler, N. A. Kukutsch, C. Bogdan, K. Erb, G. Schuler, M. B. Lutz.
2002
. Repetitive injections of dendritic cells matured with tumor necrosis factor α induce antigen-specific protection of mice from autoimmunity.
J. Exp. Med.
195
:
15
.-21.