Murine Plasmacytoid Pre-Dendritic Cells Generated from Flt3 Ligand-Supplemented Bone Marrow Cultures Are Immature APCs

Mice

Female C57BL/6 (B6) and BALB/c mice (8–12 wk of age) were obtained from Taconic Farms (Germantown, NY). FL-KO mice, maintained on a C57BL/6 background, were described previously (24) and bred at Amgen (Seattle, WA). OT-II TCR-transgenic (-Tg) mice specific for chicken OVA peptide 323–339 (OT-IIp) in the context of I-A^b (25) and OT-I TCR-Tg mice specific for chicken OVA peptide 357–364 (OT-Ip) in the context of H-2K^b (26) were bred at Amgen. All mice were housed under specific pathogen-free conditions and according to federal guidelines.

DC cultures

BM cells were isolated by flushing femurs with PBS supplemented with 2% heat-inactivated FBS (Life Technologies, Grand Island, NY). The BM cells were resuspended in Tris-ammonium chloride at room temperature for 1 min to lyse RBC. The cells were then resuspended in culture medium consisting of McCoy’s medium supplemented with essential and nonessential amino acids, 1 mmol/L sodium pyruvate, 2.5 mmol/L HEPES buffer (pH 7.4), vitamins, 5.5 × 10⁻⁵ mol/L 2-ME, 100 U/ml penicillin, 100 μg/ml streptomycin, 0.3 mg/ml l-glutamine (PSG), and 10% FBS (all medium reagents from Life Technologies).

BM cells were cultured as previously described (22) for the time indicated in the figures at 1 × 10⁶ cells/ml, in tissue culture flasks (Falcon; BD Biosciences, Mountain View, CA) or in 24-well plates (Costar Corning, Cambridge, MA) in culture medium supplemented with 200 ng/ml recombinant human FL (Chinese hamster ovary cell-derived), 20 ng/ml murine IL-3, and/or 20 ng/ml murine GM-CSF (all from Amgen). Cultures were incubated at 37°C in a humidified atmosphere containing 10% CO₂. Cells were harvested at different times from the cultures by vigorous pipetting and removal of nonadherent cells. To recover adherent cells, Versene 1:5000 (Life Technologies) was added after two washes and cells were incubated for 10 min at 37°C. The cells were then pooled with the nonadherent fraction.

In vitro activation of the DCs from FL-supplemented cultures was accomplished by the addition of different stimulation mixtures including 2 μg/ml phosphorothioate-modified oligodeoxynucleotides containing CpG motifs 1826 (TCCATGACGTTCCTGACGTT) (Sigma-Genosys, The Woodlands, TX), 5 μg/ml CD40L trimer (Amgen), 20 ng/ml recombinant murine GM-CSF, 20 μg/ml Staphylococcus aureus Cowan I bacteria (SAC; Pansorbin; Calbiochem-Novabiochem, San Diego, CA), 10 μg/ml Poly I:C (Sigma-Aldrich, St. Louis, MO), 100 ng/ml LPS (Sigma-Aldrich), 20 ng/ml IFN-γ (BD PharMingen, San Diego, CA), or HSV-1 (strain F, multiplicity of infection of 10; ATCC VR-33; American Type Culture Collection, Manassas, VA) for 16–24 h unless otherwise noted in figures.

Analysis of spleen DCs

Splenic DCs were purified following a previously described protocol (27). Briefly, spleens were digested with 200 U/ml collagenase (Life Technologies) in HBSS with Ca²⁺ and Mg²⁺ for 30 min at 37°C. The spleens were then mashed and further dissociated in Ca²⁺-free medium in the presence of 10 mmol/L EDTA. The cells were then resuspended in Tris-ammonium chloride at room temperature for 1 min to lyse RBC, resuspended in FACS buffer (see Flow cytometry below), and counted.

Flow cytometry

Before incubation with mAbs, the cells were blocked at 4°C for 20 min in FACS buffer (PBS containing 2% FBS, 2% normal rat serum, 2% normal hamster serum, 2% normal mouse serum, 10 μg/ml CD16/CD32 (2.4G2) anti-FcR mAb (BD PharMingen), and 0.02% sodium azide (Sigma-Aldrich)).

All mAbs were purchased from BD PharMingen except where noted. In addition to isotypes controls, the following mAbs (clone name given in parentheses) were used: CD3ε (145-2C11), CD4 (GK1.5), CD8α (53-6.7), CD11b (M1/70), CD11c (HL3), CD19 (1D3), CD40 (HM40-3), CD45R/B220 (RA3-6B2), CD62L (MEL-14), CD80 (16-10A1), CD86 (GL-1), CD123 (IL3Ra) (5B11), H-2K^b (AF6-88.5), IA^b (AF6-120.1), Gr-1 (RB6-8C5), Ly6c (AL-21), and Pan-NK (DX5). PE-conjugated F4/80 (CI:A3-1) was purchased from Caltag (Burlingame, CA). Biotinylated DEC-205 (NLDC145; Accurate Chemical and Scientific, Westbury, NY) binding was detected with streptavidin-PE (BD PharMingen). Propidium iodide (BD PharMingen) at 2 μg/ml was added in the last wash to exclude dead cells from analysis. Flow cytometric analyses were performed on a FACSCalibur with CellQuest software (both BD Biosciences).

Cell sorting

Cells were sorted by flow cytometry using either a FACSVantage cell sorter (BD Biosciences) or a MoFlo cell sorter (Cytomation, Fort Collins, CO). Cells were stained in FACS buffer (without azide) with anti-CD11c-PE, -B220-APC, and -CD11b-FITC mAbs and sorted into CD11c⁺CD11b⁺B220⁻ and CD11c⁺CD11b⁻B220⁺ populations (purity >95%).

Cytologic assays

Harvested cells were centrifuged at room temperature onto slides at 30,000–40,000 cells/slide, at 500 rpm for 5 min. Slides were air-dried and stained with Diff-Quik (Dade Behring, Newark, NJ) for morphological analysis. Phase-contrast observations of cultures were made by means of an inverted microscope (Leica, Bannockburn, IL) at ×400 magnification.

Chemotaxis assay

Purified CD11c⁺CD11b⁺B220⁻ and CD11c⁺CD11b⁻B220⁺ cells were resuspended in RPMI 1640 without Phenol Red and 10% FBS added. Calcein-AM dye (Molecular Probes, Eugene, OR) was added to a final concentration of 5 μg/ml. Cells were then incubated at 37°C for 30–40 min. Unincorporated calcein dye was washed out three times by centrifugation in PBS plus 0.1% BSA.

Recombinant mouse macrophage-inflammatory protein (MIP)-1α/CC chemokine ligand (CCL)3, MIP-1β/CCL4, RANTES/CCL5, MIP-3β/CCL19, and secondary lymphoid-tissue chemokine/CCL21 (all R&D Systems, Minneapolis, MN) were diluted in PBS 0.1% BSA at a final concentration of 100 ng/ml and were added according to manufacturer’s instructions to the bottom wells (30 μl) of Neuroprobe ChemoTX 96-well plates #101-3 (NeuroProbe, Gaithersburg, MD). Pore size was 3 μm, and well diameter was 3.2 mm. Labeled cells were resuspended in RPMI 1640 10% FBS and added (2 × 10⁴ cells in 25 μl) to the top filter sites of the ChemoTX system. The plates were then incubated at 37° and 5% CO₂ for 60 min. After incubation, the cell droplets on the top of the plate were washed off thoroughly with PBS/0.1% BSA four to five times. Excess liquid was removed from the top of the filter. The plates were read on a Molecular Devices (Sunnyvale, CA) Gemini Spectramax XS reader at excitation 490 nm/emission 528 nm, with a cutoff of 515 nm. Values were expressed as mean fluorescent count fold increase by calculating as follows: experimental fluorescent counts/spontaneous fluorescent counts.

Measurement of CD8⁺ and CD4⁺ T cell priming capacities by BM-derived DCs in vitro

DCs from C57BL/6 mice were generated, sorted, and activated in vitro overnight as described above. DCs were incubated overnight with 0.5 mg/ml OVA-protein (Albumin, Chicken Egg, 5X Crystalline; Calbiochem-Novabiochem). Naive CD8⁺ or CD4⁺ T cells were enriched from spleen and peripheral LNs of OVA-specific TCR-Tg OT-I or OT-II mice, respectively, by immunomagnetic cell separation using negative selection with the StemSep Enrichment mixture kits for CD8⁺ or CD4⁺ T cells (StemCell Technologies, Vancouver, British Columbia, Canada). The percentage of enriched T cells expressing Tg TCR was determined by flow cytometry (>90%), using anti-CD4-PE or anti-CD8-PE and anti-Vα2-FITC mAb (BD PharMingen).

For proliferation and cytokine production assays, naive T cells at 5 × 10⁴ cells/well were seeded into 96-U-well plates (Costar Corning, Cambridge) and cultured with varying numbers of protein-pulsed BM-derived DCs in 200 μl of IMDM (Life Technologies) plus additives per well. Cultures were maintained for 72 h at 37°C. The proliferation was assayed by pulsing the cells with 0.5 μCi of [³H]thymidine overnight and harvesting them the next day.

At 72 h, culture supernatants were collected and tested in triplicate for IFN-γ by ELISA (BD PharMingen) or for IL-2, IL-4, IL-10, IFN-γ, and TNF-α by Luminex using the Beadlyte Mouse MultiCytokine Standard 1 kit (Upstate, Charlottesville, VA), according to manufacturer’s instructions.

For cytotoxic assays, naive OT-I CD8⁺ T cells were stimulated in vitro as described above with varying numbers of protein-pulsed BM-derived DCs. After 72 h, 1 × 10⁴ syngeneic (H-2^b) C1498 target cells, which were pulsed for 1 h at 37°C with or without 1 μM OT-Ip in the presence of radioactive chromium (⁵¹Cr) in IMDM, were added in a total volume of 200 μl to the OT-I culture. After 4 h of incubation at 37°C, 25 μl of supernatant was removed and added to Lumaplate-96 plates (Packard Instrument, Meriden, CT), and radioactivity was counted on a Topcount counter (Packard Instrument). Spontaneous and total release were determined by adding culture medium or detergent, respectively, to target cells. The percentage of specific ⁵¹Cr release was calculated as follows: (experimental release − spontaneous release)/(total release − spontaneous release) × 100.

In vitro IFN-α, IL-10, and IL-12 production assay

DCs from naive mice were generated, sorted, and activated in vitro overnight as described above. Supernatants were then collected and assayed by ELISA for murine IL-10 and IL-12p70 (BD PharMingen) and IFN-α (PBL Biomedical Laboratories, New Brunswick, NJ) according to the manufacturer’s instructions. The limits of detection for IL-10, IL-12p70, and IFN-α were 10, 40, and 20 pg/ml, respectively.

TaqMan real-time PCR

In vitro-derived DCs were sorted by flow cytometry. Total RNA was purified with QIAshredder and Rneasy Mini spin column kits (Qiagen, Valencia, CA) and treated with DNase (Ambion, Austin, TX) as recommended by the manufacturers. Single-stranded cDNA synthesis was conducted on total RNA using TaqMan reverse transcription kits (Applied Biosystems, Foster City, CA) with random hexamer primers.

Mouse gene-specific primers and probe sets for TaqMan PCR were designed using Primer Express software (Applied Biosystems) with additional manual adjustments to minimize oligonucleotide hairpin loops, homo- or hetero-dimerization, and amplicon size. Test gene primer and FAM-labeled probe sets were optimized for concentration, amplification efficiency, and faithful coamplification with one or more housekeeper gene primer and VIC-labeled probe sets—the latter including β-actin, GAPDH, hypoxanthine phosphoribosyltransferase, and porphobilinogen deaminase. Real-time quantitative PCR were set up in 96-well plates using the above reagents and TaqMan master mix, as recommended by the manufacturer and as indicated by the optimization data, and were run on 7700 ABI thermal cyclers (Applied Biosystems). Real-time data were acquired and analyzed using Sequence Detection System software (Applied Biosystems) with manual adjustment of baseline and threshold parameters. Relative expression levels were determined using cycle threshold values and the 2^−ΔΔCt method to adjust for coamplified housekeeper gene levels, 2-fold amplification/cycle rates, and the reference expression level of control samples (28).

The sequences for primers and probe sets were as follows: muTLR2^up, GCC ACC ATT TCC ACG GAC T; muTLR2tqf, CAC GCC CAC ATC ATT CTC AGG TAC CA; muTLR2_low, GCT TCC TCT TGG CCT GGA G; muTLR3^up, CCC AGC TCG ATC TTT CCT ACA; muTLR3tqf, ACA ACC TCC ATG ATG TCG GCA ACG; muTLR3_low, GCT TGG GAG ATA GGA GAA GGA A; muTLR4^up, ACT GGG TGA GAA ATG AGC TGG T; muTLR4tqf, AAT TTA GAA GAA GGA GTG CCC CGC TTT CAC; muTLR4_low, GAA TAA AGT CTC TGT AGT GAA GGC AGA; muTLR7^up, CCA AAA ATG GGC TCA AAT CTT; muTLR7tqf, TCT TTT GGG ACA GAC TCC AGT TAC TGA AGC AT; muTLR7_low, GCT GAG GTC CAA AAT TTC CAA; muTLR9^up, GTA CTT GAT GTG GGT GGG AAT TG; muTLR9tqf, TGC GAC CAT GCC CCC AAT CCC; muTLR9_low, AGG GAC TTT TGG CCA CAT TCT AT; muCCR5^up, CCA TGC AGG CAA CAG AGA CT; muCCR5tqf, CTT GGA ATG ACA CAC TGC TGC CTA AAC C; muCCR5_low, TCC AAC AAA GGC ATA GAT GAC AG; muCCR7^up, TG CTG CGT CAA CCC TTT CT; muCCR7tqf, ATG CCT TCA TCG GCG TCA AGT TCC; and muCCR7_low, AGT CCT TGA AGA GCT TGA AGA GGT.

In vitro generation of pDCs

Our laboratory recently reported that multiple subsets of murine DCs can be generated from FL-supplemented BM culture (22). We examined these cultures for the possible generation of CD11c⁺B220⁺ pDCs. BM cells from naive mice were prepared and cultured at high density (1 × 10⁶ cells/ml) in the presence or absence of FL, GM-CSF, and IL-3. Cells were analyzed periodically over a 15-day period to determine their expansion and phenotype by flow cytometry. Fig. 1⇓A shows the appearance over time of BM-derived CD11c⁺ cells expressing or not expressing B220 under different conditions. Our data show that addition of FL only was sufficient for the in vitro generation of CD11c⁺B220⁺ cells from BM precursors, confirming recently published data from another group (23) using a similar in vitro culture system.

• Download figure
• Open in new tab
• Download powerpoint

FIGURE 1.

In vitro generation of BM-derived CD11c⁺B220⁺ DCs. BM cells were cultured for different times in culture medium alone or supplemented with human FL, murine GM-CSF, or murine IL-3. A, Kinetic analysis of the in vitro generation of CD11c⁺B220⁻ and CD11c⁺B220⁺ DCs in total cell numbers under the different culture conditions is shown. Results are expressed as mean ± SD of duplicate wells. B, CD11c-FITC vs B220-APC FACS profiles at different time points in the FL culture are shown. Numbers represent the percentage of cells in each region. C, Comparative FACS analysis after 9 days in culture of the generation of CD11c⁺B220⁺ DCs among FL-, GM-CSF-, and FL plus GM-CSF-containing medium is shown. An electronic gate was determined on CD11c⁺ cells, and CD11b vs B220 expression was analyzed. Results shown are representative of four experiments.

Indeed, as observed in Fig. 1⇑B, freshly isolated BM cells analyzed ex vivo contained <2% of CD11c⁺B220⁺ cells. Culture in FL-containing medium induced the generation of CD11c⁺B220⁺ and CD11c⁺B220⁻ DC subsets. After 9 days of culture, at the peak of the culture, 15–20% of the cells were CD11c⁺CD11b⁻B220⁺ (Fig. 1⇑, B and C) whereas the remaining CD11c⁺B220⁻ population could be subdivided into CD11b^high and CD11b^int as previously reported (22). Between days 9 and 15, total CD11c⁺CD11b⁺B220⁻ and CD11c⁺CD11b⁻B220⁺ numbers dropped rapidly. On a percentage basis, CD11c⁺CD11b⁻B220⁺ cells disappeared more rapidly than CD11c⁺CD11b⁺B220⁻ cells, arguing in favor of a possible shorter survival in culture for CD11c⁺CD11b⁻B220⁺ cells.

When BM-cells were cultured in the presence of GM-CSF alone, “classical” CD11c⁺B220⁻ but not CD11c⁺B220⁺ DCs were rapidly generated (Fig. 1⇑, A and C). The total CD11c⁺B220⁻ cell number dropped after 5 days due to overgrowth in the culture. GM-CSF antagonized the effect of FL on pDC generation when both cytokines were mixed in the same culture medium (Fig. 1⇑, A and C). Blocking GM-CSF by adding a neutralizing dose of anti-murine GM-CSF in the FL-containing culture did not modify the generation of either CD11c⁺B220⁻ or CD11c⁺B220⁺ cells (data not shown).

Finally, IL-3 had no effect on the generation of CD11c⁺B220⁻ and CD11c⁺B220⁺ DCs (Fig. 1⇑A).

FL is necessary for the optimal generation of CD11c⁺B220⁺ DCs in vivo

Because FL was shown to be a potent inducer of CD11c⁺B220⁺ cells in vitro, we examined its role in vivo. Total spleen cells from FL-KO (24) and wild-type mice were stained with anti-CD11c, -B220 and -CD19 mAbs and the percentages (Fig. 2⇓A) and total numbers (Fig. 2⇓B) of CD11c⁺B220⁻ and CD11c⁺B220⁺ were determined by electronically gating on CD19⁻ cells to exclude B cells. A dramatic reduction in the percentages and numbers of both splenic CD11c⁺B220⁻ and CD11c⁺B220⁺ DCs was observed in FL-KO mice compared with those of wild-type animals with a four- and seven-time reduction in numbers, respectively.

• Download figure
• Open in new tab
• Download powerpoint

FIGURE 2.

FL is necessary for the in vivo generation of CD11c⁺B220⁺ DCs. A, Multicolor fluorocytometric analysis of splenic DCs from wild-type and FL-KO mice. Total spleen cells were stained with anti-CD11c-FITC, -B220-APC, and -CD19-PE mAbs. An electronic gate was performed on CD19⁻ cells to exclude B cells, and CD11c vs B220 expression was analyzed. Numbers indicate the percentage of cells for each region among CD19⁻ cells. B, Shown is a comparative analysis of CD11c⁺B220⁻ (□) and CD11c⁺B220⁺ (▪) total cell numbers in spleen cells from wild-type control and FL-KO mice. The data represent the mean ± SD of three individual mice and are representative of three similar experiments.

BM-derived CD11c⁺B220⁺ cells have an immature phenotype

Morphologic differences distinguish CD11c⁺B220⁺ from CD11c⁺B220⁻ DCs. Fig. 3⇓A shows cytospin cover glasses of both BM-derived DC types. Whereas CD11c⁺B220⁻ DCs showed the classical stellate DC morphology, CD11c⁺B220⁺ cells appeared rounder and had a smooth surface with few dendrites and diffuse nuclei, an appearance similar to that of freshly isolated murine and human pDCs (3, 20, 21).

• Download figure
• Open in new tab
• Download powerpoint

FIGURE 3.

Morphologic and phenotypic analysis of BM-derived CD11c⁺ DCs after 9 days in culture. A, Morphologic comparison between CD11c⁺B220⁻ and CD11c⁺B220⁺ BM-derived DCs is shown. Cells were harvested and sorted after 9 days in culture in FL-containing medium, and cytospins were prepared. B, Phenotypic analysis by multicolor fluorocytometry is shown. Cells were stained with anti-CD11c-FITC and -B220-APC mAbs and with a panel of different FITC-conjugated mAbs. Electronic gates were performed on CD11c⁺ cells, and B220 expression was analyzed against each third marker. Isotype controls were tested (control) and dead cells were excluded by propidium iodide labeling. Numbers in quadrants represent the percentage of cells for each quadrant. Analyses were performed a minimum of three times for each marker.

To better define the phenotype of in vitro BM-derived CD11c⁺B220⁺, we performed multicolor cytometric analyses after 9 days of culture. Fig. 3⇑B shows that CD11c⁺B220⁺ cells did not express T, B, and NK cell markers (e.g., CD3ε, CD4, CD8α, CD19, and DX5) and showed an immature APC phenotype based upon little to no expression on most CD11c⁺B220⁺ DCs of MHC class II (IA^b) and costimulatory molecules CD40, CD80, and CD86. CD11c⁺B220⁺ cells were positively stained for MHC class I (H-2K^b), CD62L, and Ly6C, whereas CD11b and F4/80 were not expressed. GR-1 (also called Ly6G), which was previously used to characterize murine pDCs (19, 20, 21), was expressed by a fraction of CD11c⁺B220⁺ cells. Interestingly, IL-3Rα (CD123) could not be detected on the surface of murine BM-derived CD11c⁺B220⁺ cells whereas it was shown to be expressed by human pDCs (3).

BM-derived CD11c⁺B220⁺ DCs display a different pattern of Toll-like receptor (TLR) expression than CD11c⁺B220⁻ DCs

TLR are pattern-recognition receptors implicated in the recognition of pathogen structures and may link innate and adaptive immunity (reviewed in Ref. 12). CD11c⁺B220⁻ and CD11c⁺B220⁺ DCs were sorted by flow cytometry after 9 days of culture in FL-containing medium, total RNA was extracted, treated with DNase, and single-stranded cDNA synthesis was conducted. The expression of TLR2, TLR3, TLR4, TLR7, and TLR9 was measured by real-time TaqMan PCR. Nonoverlapping patterns of expression were observed for the different TLR tested (Fig. 4⇓A). “Classical” CD11c⁺B220⁻ BM-derived DCs preferentially expressed TRL2, TLR3, and TLR4, whereas the expression of TLR7 and TLR9 was stronger in CD11c⁺B220⁺ cells.

• Download figure
• Open in new tab
• Download powerpoint

FIGURE 4.

BM-derived CD11c⁺B220⁺ DCs can be activated in vitro. In vitro activation of the DCs from FL-supplemented cultures was accomplished by the addition of different stimulation mixtures including CpG, CD40L trimer, recombinant murine GM-CSF, IL-3, SAC, Poly I:C, LPS, IFN-γ, or HSV-1 during the final 24 h of culture. A, TLR mRNA expression analysis by real-time TaqMan PCR on unactivated CD11c⁺B220⁻ (□) and CD11c⁺B220⁺ (▪) DCs is shown. Data are relative to coamplified housekeeper gene levels and to CD11c⁺B220⁻ value. B, IL-10, IL12 p70, and IFN-α production analysis by ELISA on supernatants from CD11c⁺B220⁺ and CD11c⁺B220⁻ sorted cells with or without activation is shown. Results are presented as the mean ± SD of triplicate wells. C, Activation marker up-regulation was analyzed by fluorocytometry after in vitro activation with CpG plus CD40L plus GM-CSF. Level of expression for CD4, CD8α, MHC class II (IA^b), CD80, CD86, and DEC-205 were compared between CD11c⁺B220⁺ and CD11c⁺B220⁻ with (thick line) or without (thin line) in vitro activation with CpG plus CD40L plus GM-CSF. Dotted lines indicate isotype control stainings. Results shown are representative of three similar experiments.

BM-derived CD11c⁺B220⁺ DCs can be activated in vitro, gain CD8α expression, and secrete high levels of IFN-α and low levels of IL-12p70

After 9 days of culture in FL-containing medium, total cells were recovered, purified according to CD11c and B220 vs CD11b expression by flow cytometry (CD11c⁺CD11b⁺B220⁻ vs CD11c⁺CD11b⁻B220⁻, purity > 95%) and cultured with or without different stimulatory reagents for 24 h. Cell culture supernatants were harvested and tested for the production of IFN-α, IL-12p70, and IL-10 by ELISA (Fig. 4⇑B). CpG, SAC, and HSV-1 alone, as well as the CpG plus CD40L plus GM-CSF mixture, each induced secretion of high levels of IFN-α (>500 pg/ml/10⁵ cells) by CD11c⁺B220⁺ DCs. “Classical” CD11c⁺B220⁻ DCs, when activated with CpG plus CD40L plus GM-CSF produced ∼10-fold more IL-12p70 than did CD11c⁺B220⁺ DCs (∼3000 pg/ml/10⁵ cells ± 420 vs ∼270 pg/ml/10⁵ cells ± 62, respectively). Finally, CpG alone was the strongest IL-10 secretion inducer (>50 pg/ml) for both CD11c⁺B220⁻ and CD11c⁺B220⁺ subsets.

Cell phenotype after in vitro stimulation was assessed by flow cytometry. Of the 11 stimulatory conditions we tested, the CpG plus CD40L plus GM-CSF mixture was the strongest activator for pDCs, based upon CD4, CD8α, MHC class II (IA^b), CD80, CD86, and DEC-205 levels of expression. Indeed, as shown in Fig. 4⇑C, a significant up-regulation of CD8α, CD80, CD86, and IA^b molecules was observed among CD11c⁺B220⁺ cells whereas CD4 and DEC-205 were slightly up-regulated. In comparison, CD11c⁺B220⁻ cells underwent a spontaneous maturation after reculture in vitro in the absence of activation. The levels of CD80, CD86, IA^b, and DEC-205 were further up-regulated by CpG plus CD40L plus GM-CSF on these cells, whereas CD4 and CD8α expression were only slightly up-regulated.

BM-derived DCs and migratory capacities in vitro

A key characteristic of DCs appears to be their capacity to migrate from the blood to peripheral tissues and, upon stimulation, to the draining LN where they can activate naive Ag-specific T cells. This migration is driven by different chemokines interacting with chemokine receptors that are differently expressed by DCs, relative to subset and maturation stage. Immature DCs were shown to express receptors such as CCR1, CCR2, CCR5, and CCR6, which are specific for inflammatory chemokines, whereas maturation down-regulates these receptors and up-regulates the expression of CCR7 (29, 30, 31).

We compared CD11c⁺B220⁻ and CD11c⁺B220⁺ DCs for CCR5 and CCR7 mRNA expression and for their attraction toward different chemokines in an in vitro chemotaxis assay. Three different DC maturation stages were analyzed. Immature DCs were harvested after the 9-day culture period, sorted by flow cytometry, and analyzed directly (9-day condition). After flow cytometry sorting, weak and full maturation were induced by cultivating CD11c⁺B220⁻ and CD11c⁺B220⁺ DCs in FL-containing medium in the absence or presence of CpG plus CD40L plus GM-CSF (24 h, no stimulation, and 24 h, plus stimulation, respectively).

Freshly isolated immature CD11c⁺B220⁻ and CD11c⁺B220⁺ DCs showed similar levels of CCR5 mRNA expression (Fig. 5⇓A), whereas CCR5 expression was slighty reduced after overnight culture with CpG plus CD40L plus GM-CSF. CCR7 mRNA expression was low in both immature subsets (9-day condition) but was dramatically augmented in CD11c⁺B220⁻ DCs after reculture in vitro in the absence of stimulation. Activation with CpG plus CD40L plus GM-CSF increased CCR7 mRNA expression in CD11c⁺B220⁺ DCs.

• Download figure
• Open in new tab
• Download powerpoint

FIGURE 5.

BM-derived DCs and migratory capacities. Immature DCs were harvested and sorted into CD11c⁺B220⁻ and CD11c⁺B220⁺ either after the 9-day culture period (9-day condition) or after a subsequent 24-h culture in FL-containing medium in the absence or presence of CpG plus CD40L plus GM-CSF (plus 24 h, no stimulation, and plus 24 h, plus stimulation, respectively). A, CCR5 and CCR7 chemokine receptor mRNA expression analysis by real-time TaqMan PCR on CD11c⁺B220⁻ (□) and CD11c⁺B220⁺ (▪) DCs is shown. Data are relative to coamplified housekeeper gene levels and to CD11c⁺B220⁻ value. B, In vitro DC migration analysis in a chemotaxis assay toward CCL3, CCL4, CCL5, CCL19, and CCL21 chemokines is shown. Values are expressed as follows: experimental fluorescent counts/spontaneous fluorescent counts. Results are presented as the mean ± SD of triplicate wells, and the experiment was performed four times.

Despite similar CCR5 mRNA expression between CD11c⁺B220⁻ and CD11c⁺B220⁺ DCs, these two subsets displayed different migration patterns toward CCL3 (MIP-1α), CCL4 (MIP-1β), and CCL5 (RANTES), three putative CCR5 ligands (32). Immature CD11c⁺B220⁺ DCs were strongly mobilized by these three chemokines, whereas the CD11c⁺B220⁻ fraction showed mobilization only toward CCL3 (Fig. 5⇑B). After overnight culture in the absence of stimulation, CCL3-, CCL4- and CCL5-driven migration was barely detectable with CD11c⁺B220⁻ DCs, but still evident with CD11c⁺B220⁺ DCs. This condition was also sufficient to induce the migration of CD11c⁺B220⁻ cells toward CCR7 ligands (CCL19 (MIP-3β) and CCL21 (6Ckine)) but insufficient for CD11c⁺B220⁺ DCs. However, overnight activation with CpG plus CD40L plus GM-CSF induced CCL19- and CCL21-mediated migration by the CD11c⁺B220⁺ subset and increased CCL21-mediated CD11c⁺B220⁻ DC migration, accompanied by a lack in mobilization toward CCL3, CCL4, and CCL5.

T cell stimulatory capacities of BM-derived CD11c⁺B220⁺ DCs

Another key characteristic of dendritic cells is their ability to prime naive T cells. We analyzed the in vitro T cell stimulatory capacity of both BM-derived CD11c⁺B220⁻ and CD11c⁺B220⁺ DC subsets in two different syngeneic model systems involving either OVA-specific TCR-Tg CD4⁺ (OT-II) or OVA-specific TCR-Tg CD8⁺ (OT-I) naive T cells.

Without any prior activation in vitro, OVA-protein-pulsed CD11c⁺B220⁺ pDCs were poor naive OT-II CD4⁺ T cell stimulators, in terms of proliferation and cytokine secretion, relative to CD11c⁺B220⁻ DCs (Fig. 6⇓). However, in vitro activation with CpG plus CD40L plus GM-CSF increased the ability of BM-derived CD11c⁺B220⁺ cells to stimulate naive OT-II CD4⁺ T cells, although to lower levels than those obtained with the CD11c⁺B220⁻ counterpart. The same effect was observed when the two DC subsets were pulsed with the immunodominant OVA-peptide (data not shown).

• Download figure
• Open in new tab
• Download powerpoint

FIGURE 6.

CD4⁺ T cell stimulatory capacity of BM-derived CD11c⁺B220⁺ DCs. Sorted CD11c⁺B220⁻ (□, ▪) and CD11c⁺B220⁺ (○, •) DCs from C57BL/6 mice were pulsed in vitro for 24 h with 0.5 mg/ml OVA (OVA-protein) and simultaneously activated (□, ○) or not (▪, •) with 2 μg/ml CpG, 5 μg/ml CD40L, and 20 ng/ml recombinant murine GM-CSF. DCs were then used as stimulator cells for syngeneic OVA-specific TCR-Tg (OT-II) CD4⁺ T cells. Proliferation (A) and cytokine secretion (B–F) were quantified by [³H]thymidine incorporation and Luminex, respectively. Results are presented as the mean ± SD of triplicate wells, and the experiment was performed three times.

In addition to proliferation induction, the CD4⁺ T helper response was assayed by collecting supernatants after 72 h of culture to analyze IL-2, IL-4, IL-10, IFN-γ, and TNF-α secretion (Fig. 6⇑). As expected, unstimulated CD11c⁺B220⁺ DCs did not induce strong cytokine secretion by CD4⁺ T cells, reflecting the weak T cell stimulation observed at the level of proliferation. However, after pDC stimulation with CpG plus CD40L plus GM-CSF, a significant increase in cytokine secretion was observed. No clear Th1 or Th2 bias was induced, because both Th1-related (IFN-γ) and Th2-related (IL-4) cytokines were secreted. IL-10 was produced when activated CD11c⁺B220⁻ or CD11c⁺B220⁺ DCs were used as APCs for naive CD4⁺ T cell priming. Intracellular detection by flow cytometry of this cytokine confirmed its production by activated CD4⁺ T cells (data not shown).

Fig. 7⇓ shows the priming capacity of BM-derived DCs for naive CD8⁺ T cells. OT-I CD8⁺ T cells cultured with unstimulated CD11c⁺B220⁺ DCs previously incubated with OVA protein were poorly activated and proliferated weakly. In addition, these T cells inefficiently killed syngeneic (H-2^b) peptide-pulsed C1498 target cells in a cytotoxic assay, and IFN-γ secretion remained under detectable levels (Fig. 7⇓, middle and lower panels). However, stimulation of CD11c⁺B220⁺ DCs with CpG plus CD40L plus GM-CSF was efficient for the induction of cell proliferation, specific cytoxicity, and IFN-γ secretion, to levels comparable to those observed when CD11c⁺B220⁻ cells were used as APCs.

• Download figure
• Open in new tab
• Download powerpoint

FIGURE 7.

CD8⁺ T cell stimulatory capacity of BM-derived CD11c⁺B220⁺ DCs. Sorted CD11c⁺B220⁻ (▪, □) and CD11c⁺B220⁺ (•, ○) DCs from C57BL/6 mice were pulsed in vitro for 24 h with 0.5 mg/ml OVA (OVA-protein) and simultaneously activated (□, ○) or not (▪, •) with 2 μg/ml CpG, 5 μg/ml CD40L, and 20 ng/ml recombinant murine GM-CSF. DCs were then used as stimulator cells for syngeneic OVA-specific TCR-Tg (OT-I) CD8⁺ T cells. Proliferation (A), cytotoxicity against syngeneic (H-2^b) C1498 targets cells (B), and IFN-γ secretion (C) were quantified by [³H]thymidine incorporation, ⁵¹Cr release, and ELISA, respectively. Results are presented as the mean ± SD of triplicate wells, and the experiment was performed three times.