Endotoxin-Induced Maturation of MyD88-Deficient Dendritic Cells

Mice

C57BL/6, BALB/c, C3H/HeN, and C3H/HeJ mice were purchased from Japan SLC (Shizuoka, Japan). TLR4- and MyD88-deficient mice were established as described previously (8). The BALB/c-background DO11.10 TCR-transgenic mice (23), which recognize the 323–339 peptide of chicken OVA in the context of I-A^d, were kindly provided by Dr. K. M. Murphy. MyD88-deficient mice with a BALB/c background were obtained by backcrossing them to BALB/c mice for five to six generations.

Reagents

LPS derived from Escherichia coli O55:B5 was purchased from Sigma (St. Louis, MO) (16). Synthetic E. coli-type lipid A (compound 506) was purchased from Dai-ichi Pure Chemicals (Tokyo, Japan) (16). Phosphorothioate-stabilized oligodeoxynucleotides containing unmethylated CpG motifs were purchased from TIB MOLBIOL (Berlin, Germany) (24).

Bone marrow (BM) DC culture

BM cells from femurs and tibias were plated at 1 × 10⁶ cells/ml in 24-well plates with 10% FCS-RPMI 1640 supplemented with 10 ng/ml murine GM-CSF (PeproTech, London, U.K.) as described previously (25). At day 6, loosely adherent cells were harvested by gentle pipeting, plated at 5 × 10⁵ cells/ml in 24-well plates, and cultured with or without 100 ng/ml LPS, 1 μg/ml synthetic lipid A, or 0.1 μM CpG DNA in a fresh medium for 2 more days.

ELISA

At day 8 of the DC culture, cell supernatants were harvested for measuring TNF-α and IL-12 p40 production by ELISA kits (Genzyme, Minneapolis, MN).

Flow cytometry

DCs were incubated with anti-CD16/32 (2.4G2; BD PharMingen, San Diego, CA) to minimize nonspecific staining. They were then stained with biotinylated mAbs for 20 min at 4°C and developed with streptavidine-PE (BD PharMingen). The following biotinylated mAbs were also purchased from BD PharMingen: CD3 (145-2C11), CD11c (HL3), CD40 (3/23), CD80 (16-10A1), and CD86 (GL1). Flow cytometric analysis was performed using a FACSCalibur (BD Biosciences, Mountain View, CA) with CellQuest software (BD Biosciences).

MLR assay

Unstimulated or stimulated BM DCs were harvested at day 8, washed, irradiated at a dose of 30 Gy, and plated at 3-fold serial dilutions in 96-well round-bottom plates. These stimulator DCs were incubated for 3 days with 5 × 10⁴/well of splenic CD4⁺ T cells, which were isolated by using MACS with CD4 microbeads (Miltenyi Biotec, Bergisch Gladbach, Germany). The resulting population of CD4⁺ T cells after MACS was >95% positive for CD4, as determined by flow cytometry. [³H]Thymidine was added for the last 15 h.

Ag presentation assay

Unstimulated or LPS-stimulated BM DCs were generated from wild-type BALB/c or MyD88-deficient mice with a BALB/c background. The DCs at day 8 were irradiated and cocultured with CD4⁺ T cells from DO11.10 TCR-transgenic mice (23) in the presence of 1 μM of chicken OVA peptide (residues 323–339) for 3 days. [³H]Thymidine was added for the last 15 h.

In vitro kinase assay and EMSA

At day 6 of the DC culture, harvested cells were stimulated with 100 ng/ml LPS for indicated periods, lysed, and immunoprecipitated with anti-IL-1R-associated kinase (IRAK) Ab or rabbit anti-c-Jun N-terminal kinase (JNK)1 Ab (Santa Cruz Biotechnology, Santa Cruz, CA). The in vitro kinase assay and EMSA were performed as described previously (16).

LPS injection and splenic DC analysis

LPS (200 ng/head) derived from E. coli O55:B5 was i.v. injected into wild-type, TLR4-deficient, and MyD88-deficient mice. Control mice were injected with PBS. Six hours later, spleens were dissected and incubated with RPMI 1640 medium containing 400 U/ml collagenase (Wako, Osaka, Japan) and 15 μg/ml DNase I (Sigma) for 20 min at 37°C. For the last 5 min of the incubation period, 5 mM EDTA was added. Single-cell suspensions were prepared and stained with biotinylated anti-CD11c followed by streptavidin-PE and FITC-conjugated anti-CD40 or anti-CD86 (BD PharMingen). Stained cells were subjected to flow cytometric analysis.

MyD88 is essential for LPS-induced cytokine induction from BM DC

To assess the effects of LPS on DCs, immature DCs were generated in vitro by culturing BM cells with GM-CSF. First, we measured proinflammatory cytokine production in response to LPS. When wild-type immature BM DCs were stimulated with LPS, they produced more amounts of TNF-α and IL-12 p40 than those without LPS (Fig. 1⇓). Meanwhile, the cytokine induction was abolished in TLR4- or MyD88-deficient BM DCs. The results suggest that LPS stimulates cytokine production in BM DCs through TLR4 in a MyD88-dependent manner.

• Download figure
• Open in new tab
• Download powerpoint

FIGURE 1.

LPS-induced cytokine production in BM DCs is dependent on both TLR4 and MyD88. BM cells from wild-type, TLR4-deficient, or MyD88-deficient mice were cultured with GM-CSF. At day 6, cells were incubated with (▪) or without (□) 100 ng/ml LPS for 2 more days. Amounts of TNF-α and IL-12 p40 in the cell culture supernatants were measured by ELISA. The data indicate means ± SD of triplicate samples of one representative experiment.

LPS-induced DC maturation is independent of MyD88

Next, we analyzed surface expression of costimulatory molecules on LPS-stimulated and unstimulated BM DCs. LPS could enhance surface expression of CD40, CD80, and CD86 in wild-type BM DCs (Fig. 2⇓). Meanwhile, CD11c expression was significantly decreased in response to LPS. This phenotype typically represents DC maturation (21). TLR4-deficient BM DCs showed no modulation of these surface molecules in response to LPS. However, upon LPS stimulation, MyD88-deficient DCs showed up-regulation of CD40, CD80, and CD86 and down-regulation of CD11c (Fig. 2⇓). These results clearly indicate that LPS-induced DC maturation is dependent on TLR4 but can proceed in the absence of MyD88.

• Download figure
• Open in new tab
• Download powerpoint

FIGURE 2.

Flow cytometric analysis of LPS-stimulated and unstimulated BM DCs. At day 6 of the DC culture, cells were harvested and cultured with (solid lines) or without (dotted lines) LPS for another 48 h. Then cells were stained with biotinylated mAbs followed by PE-conjugated streptavidin and analyzed with FACS.

Another hallmark of DC maturation is an increased capacity for T cell stimulation. To analyze allostimulatory activity of BM DCs, wild-type, TLR4-deficient, and MyD88-deficient BM DCs (H-2^b) were irradiated and cocultured with allogeneic BALB/c CD4⁺ T cells (H-2^d). Wild-type BM DCs showed augmented ability to stimulate allogeneic T cells in response to LPS (Fig. 3⇓A). Similarly, the allostimulatory activity of MyD88-deficient BM DCs was also enhanced with LPS stimulation. However, LPS did not enhance the ability of TLR4-deficient BM DCs to stimulate allogeneic T cells. Next, we investigated whether LPS could enhance APC activity of BM DCs. BM DCs from wild-type or MyD88-deficient mice were irradiated and cocultured with CD4⁺ T cells from cOVA-specific TCR-transgenic mice (23) in the presence of cOVA peptide. Both wild-type and MyD88-deficient BM DCs without LPS stimulation could augment cOVA-specific T cell proliferation in a DC number-dependent manner (Fig. 3⇓B, ○). When stimulated with LPS, both BM DCs showed enhanced ability to stimulate T cell proliferation (Fig. 3⇓B, •). Taken together, LPS could not only modulate costimulatory molecule expression but also enhance the T cell stimulatory activity of MyD88-deficient BM DCs.

• Download figure
• Open in new tab
• Download powerpoint

FIGURE 3.

Allostimulatory and APC activity of BM DCs. A, To measure the allostimulatory activity of DCs, LPS-stimulated (•) or unstimulated (○) DCs (H-2^b) were irradiated and incubated with allogeneic BALB/c CD4⁺ T cells (H-2^d) at various cell concentrations as indicated. B, To measure the APC activity of DCs, LPS-stimulated (•) or unstimulated (○) DCs were irradiated and incubated at 1 μM of a cOVA peptide (residues 323–339) with CD4⁺ T cells derived from DO11.10 TCR-transgenic mice. The data from the culture containing T cells and a cOVA peptide without DCs are also shown (▪). T cell responses were evaluated by [³H]thymidine incorporation of CD4⁺ T cells. The data indicate means ± SD of triplicate samples of one representative experiment.

To exclude the possibility that some contaminants were responsible for these LPS effects, we also tested the effects of synthetic lipid A, a biologically active moiety of LPS, in a similar set of experiments. As shown in Fig. 4⇓A, lipid A-induced cytokine production was dependent on MyD88. However, lipid A could significantly enhance not only CD40 expression but also allostimulatory activity of MyD88-deficient BM DCs (Fig. 4⇓, B and C). All of these effects were dependent on TLR4, because lipid A showed no effects on TLR4-deficient BM DCs.

• Download figure
• Open in new tab
• Download powerpoint

FIGURE 4.

Lipid A can enhance CD40 expression and allostimulatory activity of MyD88-deficient BM DCs. At day 6 of the DC culture, cells were harvested and incubated with or without 1 μg/ml of lipid A for 2 more days. Cytokine production (A), CD40 expression (B), and allostimulatory activity (C) of BM DCs were analyzed as described in the legends of Figs. 1–3⇑⇑⇑.

TLR9-mediated signaling is dependent on MyD88

Bacteria-derived DNA containing CpG motifs can also induce cytokine production and DC maturation through another TLR family member, TLR9 (24). CpG DNA-induced cytokine induction from macrophages and DCs is dependent on MyD88 (26, 27). Therefore, we tested the effects of CpG DNA on wild-type and MyD88-deficient DCs. CpG DNA could induce cytokine production, surface expression of costimulatory molecules such as CD40, CD80, and CD86, and allostimulatory activity of wild-type DCs (Fig. 5⇓ and data not shown). However, all of these effects were abolished in not only TLR9-deficient (24) but also in MyD88-deficient DCs (Fig. 5⇓). The results clearly suggest that TLR9-mediated signaling can induce DC maturation in a MyD88-dependent manner.

• Download figure
• Open in new tab
• Download powerpoint

FIGURE 5.

CpG DNA-induced cytokine production and maturation of BM DCs are dependent on MyD88. At day 6 of the DC culture, cells were harvested and incubated with or without 0.1 μM CpG DNA for 2 more days. Cytokine production (A), CD40 expression (B), and allostimulatory activity (C) of BM DCs were analyzed as described in the legends of Figs. 1–3⇑⇑⇑.

Both cytokine induction and DC maturation in response to LPS were abolished in C3H/HeJ-derived BM DCs

Mice of the C3H/HeJ strain show defective responses to LPS and were found to carry the point mutation in the Tlr4 gene locus, which results in the replacement of a conserved proline residue with a histidine residue in the cytoplasmic region of TLR4 (6, 7). This point mutation is responsible for diminished responses to LPS (8). We have investigated whether C3H/HeJ DCs can mature in response to LPS. LPS-induced cytokine production was severely impaired in C3H/HeJ DCs as in TLR4- and MyD88-deficient DCs (Fig. 1⇑ and 6⇓A). Unlike MyD88-deficient DCs, C3H/HeJ DCs did not show either up-regulation of CD40, CD80, and CD86, nor down-regulation of CD11c in response to LPS (Fig. 6⇓B and data not shown). C3H/HeJ DCs also showed severe impairment of LPS-induced allostimulatory activity (Fig. 6⇓C). Taken together, LPS could induce neither cytokine induction nor DC maturation in C3H/HeJ DCs.

• Download figure
• Open in new tab
• Download powerpoint

FIGURE 6.

Impairment of both cytokine induction and maturation of C3H/HeJ DCs in response to LPS. At day 6 of the DC culture, cells were harvested incubated with or without 100 ng/ml LPS for 2 more days. Cytokine production (A) and CD40 expression (B) were analyzed as described in the legends of Figs. 1⇑ and 2⇑. C, Allostimulatory activity of LPS-stimulated (•) or unstimulated (○) C3H/HeN or C3H/HeJ BM DCs (H-2^k) were evaluated by coculture with allogeneic BALB/c CD4⁺ T cells (H-2^d).

Signaling analysis in LPS-stimulated BM DCs

We next examined the LPS signaling pathways of wild-type and mutant DCs. MyD88 interacts with TLR4 through a TIR domain and recruits IRAK to the receptor complex through a death domain, and then IRAK is sequentially activated (14, 15). LPS could not induce IRAK activation in either TLR4- or MyD88-deficient DCs, indicating that MyD88 is essential for LPS-induced IRAK activation (Fig. 7⇓A). Activated IRAK subsequently activates NF-κB via TNFR-associated factor (TRAF)6 (14, 15). Next, LPS-induced NF-κB activation in DCs was examined by EMSA (Fig. 7⇓B). NF-κB DNA binding activity was detected in wild-type DC nuclear extracts at 10 min after LPS stimulation and gradually increased thereafter (Fig. 7⇓B). This NF-κB activation is dependent on TLR4, because it was not observed in TLR4-deficient DCs. NF-κB activation was detected at 60 min after LPS stimulation in MyD88-deficient DCs in a similar level to that observed in wild-type DCs. However, kinetics of NF-κB activation showed delayed induction in MyD88-deficient DCs as compared with that in wildtype DCs. Furthermore, LPS-induced JNK activation, which is dependent on TLR4, was also detected in MyD88-deficient DCs with delayed kinetics (Fig. 7⇓C). Meanwhile, LPS-induced activation of IRAK, NF-κB, and JNK was abolished in both TLR4-deficient and C3H/HeJ DCs (Fig. 7⇓). The results show a close correlation of NF-κB and JNK activation with DC maturation in response to LPS and strongly suggest that residual biochemical responses in MyD88-deficient DCs are involved in LPS-induced DC maturation.

• Download figure
• Open in new tab
• Download powerpoint

FIGURE 7.

Analysis of LPS signaling pathways in BM DCs. BM DCs were stimulated for indicated periods with 100 ng/ml LPS and analyzed for (A) IRAK, (B) NF-κB, or (C) JNK activation. The cell lysates were immunoprecipitated with anti-IRAK or anti-JNK1 Abs and measured for their kinase activities by in vitro kinase assay (A and C, upper panels). For IRAK kinase activity, autophosphorylating activity was measured (KA). For JNK kinase activity, GST-c-jun was used as substrate. The same lysates were also blotted with anti-IRAK or anti-JNK1 Ab (A and C, lower panels). B, LPS-induced NF-κB activation in BM DCs was analyzed as described previously (16 ). Experiments were independently performed three times with similar results.

LPS can enhance surface expression of costimulatory molecules on MyD88-deficient splenic DCs in vivo

We next investigated in vivo effects of LPS on splenic DCs from wild-type and mutant mice. Intravenous injection of LPS can induce up-regulation of costimulatory molecule expression on splenic DCs (28). Splenic CD11c⁺ cells from LPS-injected wild-type mice showed increased expression of CD40 and CD86 compared with those from PBS-injected wild-type mice (Fig. 8⇓A). LPS-induced up-regulation of costimulatory molecules was abolished in TLR4-deficient mice (Fig. 8⇓B). However, in MyD88-deficient mice, splenic DCs significantly increased their surface expression levels of CD40 and CD86 in response to LPS injection (Fig. 8⇓C). Serum IL-12 p40 levels reached 718 ± 27 pg/ml in LPS-injected wild-type mice, whereas levels were <10 pg/ml not only in PBS-injected wild-type mice, but also in LPS-injected TLR4- and MyD88-deficient mice. Thus, these in vivo experiments clearly indicate that LPS can induce up-regulation of costimulatory molecules on splenic DCs in a TLR4-dependent and MyD88-independent manner.

• Download figure
• Open in new tab
• Download powerpoint

FIGURE 8.

In vivo effects of LPS on splenic DCs. LPS was injected i.v. into (A) wild-type, (B) TLR4-, and (C) MyD88-deficient mice. Six hours later, splenic cells were analyzed for their CD40 vs CD11c or CD86 vs CD11c profiles. CD11c⁺ cells were gated, as shown by the rectangles, and their histograms are also shown. Solid and dotted lines in the histograms represent the data from LPS- and PBS-injected mice, respectively. Experiments were independently performed three times with similar results.