A Critical Role for p38 Mitogen-Activated Protein Kinase in the Maturation of Human Blood-Derived Dendritic Cells Induced by Lipopolysaccharide, TNF-α, and Contact Sensitizers

DC generation from buffy coats

Buffy coats of healthy donors were obtained according to institutional guidelines. PBMC were prepared by density centrifugation using Ficoll-Paque (Pharmacia, Uppsala, Sweden). PBMC were seeded (50 × 10⁶ cells) in 100-mm tissue culture plates (Nunc, Roskilde, Denmark) in 10 ml of RPMI 1640 medium supplemented with 10% heat-inactivated FCS, 2 mM l-glutamine, 100 U/ml penicillin, 100 μg/ml streptomycin (all supplied by Life Technologies, Paisley, U.K.), and incubated for 1 h at 37°C in a humidified 5% CO₂ atmosphere. Nonadherent cells were removed by extensive washing, and the remaining adherent cells were recovered by scraping. In some instances, residual contaminating cells were depleted using anti-CD3 and anti-CD19 mIgG-coated M450 Dynabeads obtained from Dynal (Oslo, Norway). The mean purity of purified CD14⁺ cells was greater than 90%. Cells were subsequently cultured in six-well plates (3 × 10⁶ cells/well) in RPMI 1640 medium supplemented with 10% heat-inactivated FCS, 10 mM HEPES, 1% nonessential amino acids, 1 mM sodium pyruvate, 2 mM l-glutamine, 100 U/ml penicillin, 100 μg/ml streptomycin, 50 μM 2-ME (Life Technologies), 10³ U/ml GM-CSF, and 10³ U/ml IL-4 (Stratagen, Hanover, Germany). Cells were refed with 1 ml of fresh medium containing 3 × 10³ U of GM-CSF and 3 × 10³ U of IL-4 on days 2, 4, and 6. To get mature DC, the nonadherent cells (immature DC) were harvested on day 7 and stimulated with 100 U/ml recombinant human TNF-α (a gift of Glaxo Wellcome, Geneva, Switzerland), 10 ng/ml LPS (Escherichia coli, strain 055:B5; Difco, Detroit, MI), or with different chemical reagents as described below. For all of the following assays, 5 × 10⁵ immature DC were cultured in 1 ml of medium containing 10³ U/ml GM-CSF and 10³ U/ml IL-4 in 24-well plates.

Chemical reagents

Nickel sulfate (NiSO₄), 2,4-dinitrofluorobenzene (DNFB), benzalkonium chloride (BC), SDS, and DMSO were obtained from Sigma (Buchs, Switzerland). SB203580 and PD98059 were purchased from Alexis (Läufelfingen, Switzerland). Treatment of immature DC with SB203580 or PD98059 before stimulation was performed for 30 and 75 min, respectively. Because SB203580 and PD98059 were dissolved in DMSO, a 0.1% (v/v) concentration of DMSO was used as a negative control where indicated.

Abs and immunoreactants

Flow cytometric analysis

Cultured DC were washed, resuspended at 0.5–1 × 10⁵ cells in 50 μl of cold PBS containing 0.1% sodium azide, 10 mg/ml BSA, and 200 μg/ml mouse IgG (Sigma), and incubated for 15 min on ice. Subsequent staining with labeled mAb or appropriate isotypic controls was performed for 30 min on ice. Cells were then washed and resuspended in 300 μl of cold PBS containing 0.1% sodium azide, 10 mg/ml BSA, and 10 μg/ml 7-amino-actinomycin D (Sigma). Stained cells were analyzed for three-color immunofluorescence with a FACSCalibur cell analyzer (Becton Dickinson). Cellular debris were eliminated from the analysis using a gate on forward and side scatter. A life gate was set using 7-amino-actinomycin D, which allows discrimination between viable, necrotic, and apoptotic cells (35). A minimum of 10⁴ cells were analyzed for each sample. Results were processed using CellQuest software (Becton Dickinson).

FITC-dextran uptake

Cells were incubated with FITC-dextran (0.1 mg/ml) either at 4°C (internalization control) or at 37°C for 1 h. Then, cells were washed twice with a cold buffer consisting of PBS containing 0.1% sodium azide and 10 mg/ml BSA and analyzed with a FACSCalibur cell analyzer as described above.

Allogenic MLR

Allogenic T cells were obtained from peripheral blood of healthy adults after Ficoll-Paque gradient (Pharmacia), adherence to plastic for 1 h at 37°C, and passage over a nylon wool column (Biotest, Dreieich, Germany). Cells recovered after purification were on the mean greater than 90% CD3⁺, and were distributed at 5 × 10⁴ cells per well into round-bottom 96-well microplates (Nunc). Cells were incubated for 5 days in the presence of graded numbers of irradiated DC stimulators (3000 rad, ¹³⁷Cs source) in 200 μl of medium containing 10% FCS. T cell proliferation was assessed after 8–14 h incorporation of [³H]thymidine (1 μCi/well; New England Nuclear, Boston, MA) by using standard procedures. The results are expressed as the mean of quadruplicate cultures. The SEM of the results never exceeded 15%.

Western blot analysis of cellular MAPKs

Immature DC were exposed to various agonists for the indicated periods of time. Thereafter, cells were washed twice with cold PBS and incubated with 100 μl of lysis buffer (50 mM Tris-HCl, 1% Triton X-100, 150 mM NaCl, 10% glycerol, 2 mM EDTA, 2 mM EGTA, 40 mM β-glycerophosphate, 50 mM NaF, 10 mM sodium pyrophosphate, 200 μM Na₃VO₄, 0.3 mM leupeptin, 1 μM pepstatin A, 1 mM PMSF, and 100 nM okadaic acid, pH 7.4). The homogenates were centrifuged at 14,000 rpm for 10 min at 4°C. Cell lysates (10–20 μg) were electrophoresed on 12% SDS-PAGE gels, and transferred to nitrocellulose membranes for Western blot analysis. Briefly, nitrocellulose membranes were incubated in a blocking buffer (50 mM Tris-HCl, 200 mM NaCl, 0.2% Tween 20, and 5% nonfat dried milk) for 1 h at room temperature, then incubated for 2 h with Abs raised against phosphorylated p42/44 ERK, p38 MAPK (New England Biolabs, Beverly, MA), or p46/54 JNK (Santa Cruz Biotechnology, Santa Cruz, CA). The membranes were washed and incubated for 1 h with HRP-labeled goat anti-rabbit (CovalAb, Oullins, France) or rabbit anti-goat (Sigma) Abs. Immunoreactive bands were visualized by ECL detection reagent (Amersham, Zürich, Switzerland) and quantified by scanning densitometry (Molecular Dynamics, Sunnyvale, CA). Abs to total p38 MAPK or ERK (Santa Cruz Biotechnology) were used to analyze p38 MAPK and ERK expression as loading controls.

Bioassay for TNF quantification

A subclone (WEHI 1.14) of the TNF-sensitive WEHI 164 clone 13 was used as previously described (36) with the following modifications. Fifty microliters of graded dilutions from culture supernatant were added to 50 μl (2 × 10⁴) WEHI 1.14 cells in flat-bottom 96-well plates (Nunc) in duplicates and incubated for 24 h at 37°C. Twenty microliters of MTS (333 μg/ml; Promega, Madison, WI) and 1 μl of the electron-coupling reagent PMS (25 μM; Sigma) were subsequently added to each well. After 2 h of incubation, the resulting intensity of the coloration was measured at 490 nm in a Thermomax microplate reader (Molecular Devices, Menlo Park, CA) and analyzed using Softmax software from the same company. Recombinant human TNF-α (provided by J.-M. Dayer, University of Geneva, Geneva, Switzerland) was used as a standard. The sensitivity of this assay was 0.1 pg/ml. This assay does not distinguish TNF-α from lymphotoxin. Thus, the resulting activity is referred to as TNF. SB203580 had no effect on the TNF bioassay.

Multiprobe RNase protection assay

Total RNA was extracted from DC using TRIzol reagent (Life Technologies). Multiprobe template sets hCK2 and hCK3 were purchased from PharMingen. hCK2 and hCK3 DNA templates were used to synthesize the [α³²P]UTP (800 Ci/mmol, 5 mCi/ml; Hartmann, Braunschweig, Germany)-labeled riboprobes in the presence of rNTPs using T7 RNA polymerase (Promega). Hybridization with 1 μg of each target RNA was performed overnight at 56°C followed by digestion with RNases A and T1 at 37°C for 30 min. The samples were then treated by a proteinase K/SDS mixture, extracted with phenol/chloroform/isoamylalcohol (25:24:1), and precipitated in the presence of carrier E. coli MRE 600 tRNA (Boehringer Mannheim, Mannheim, Germany). The samples were loaded onto an acrylamide-urea sequencing gel next to the labeled control probes treated or not with RNases A and T1, and run at 50 W for 1.5 h. The gel was absorbed to filter paper, dried under vacuum, and exposed on Kodak X-AR film with intensifying screens at −80°C.

The relative amounts of proinflammatory cytokine encoding mRNA were measured by scanning densitometry (Molecular Dynamics) on subexposed autoradiograms and further normalized using both housekeeping gene values (L32 and GAPDH).

TNF-α and LPS induce phosphorylation of members of the three families of MAPK in immature DC

To investigate the intracellular MAPK signaling pathways involved in DC maturation by two potent DC maturation factors, TNF-α and LPS, we performed Western blot analysis to detect the phosphorylated forms of ERK1/2, p46/54 JNK, and p38 MAPK. Stimulation of immature DC with 100 U/ml TNF-α or 10 ng/ml LPS induced the phosphorylation of all of the MAPK tested (Fig. 1⇓, A and B). Upon TNF-α stimulation, maximal MAPK phosphorylation was observed between 5 and 10 min, whereas the LPS-induced peak response occurred after 15–30 min. Densitometric quantification revealed a 3.5 ± 1.7-fold and a 4.8 ± 1.2-fold induction of p38 MAPK phosphorylation by TNF-α and LPS, respectively (mean ± SD of eight experiments for TNF-α, and of three experiments for LPS). In addition, TNF-α induced the phosphorylation of ERK1 (2.9 ± 0.4), ERK2 (2.5 ± 0.3), p46 JNK (1.6 ± 0.2), and p54 JNK (2.8 ± 0.2) (n = 3). LPS induced also the phosphorylation of ERK1 (3.8 ± 0.4), ERK2 (4.2 ± 0.6), p46 JNK (1.7 ± 0.3), and p54 JNK (4.8 ± 0.9) (n = 3). To exclude any LPS contamination in the TNF-α preparation, we used neutralizing polyclonal Ab raised against TNF-α. This Ab effectively blocked TNF-α-induced p38 MAPK phosphorylation (Fig. 1⇓C) as well as TNF-α-stimulated ERK1/2 and p46/54 JNK phosphorylation (data not shown). Thus, both TNF-α and LPS induce early phosphorylation of ERK1/2, p46/54 JNK, and p38 MAPK in immature DC.

• Download figure
• Open in new tab
• Download powerpoint

FIGURE 1.

Detection of phosphorylated MAPK induced by TNF-α and LPS in immature DC. A, Immature monocyte-derived DC were stimulated with TNF-α (100 U/ml) for the indicated periods of time or left untreated (c). Cell lysates were subjected to Western blotting. Membranes were incubated with Abs raised against the phosphorylated form of p38 MAPK (p38-P), of p42/44 MAPK (ERK1/2-P), or of p46/54 JNK (p46/54-P). Membranes were then reprobed with an Ab against total p38 MAPK (total p38) for loading control. B, Monocyte-derived DC were stimulated or not (c) with LPS (10 ng/ml) as described in A. C, Monocyte-derived DC were stimulated or not (c) with TNF-α (100 U/ml) for 5 min. TNF-α was preincubated for 30 min on ice with 20 μg of either neutralizing rabbit polyclonal Ab against TNF-α or control rabbit polyclonal Ab before DC stimulation. Cell lysates were then subjected to Western blot analysis. Membranes were incubated with Abs raised against the phosphorylated form of p38 MAPK (p38-P), and reprobed with an Ab against total p38 MAPK (total p38) for loading control. Blots are representative of results obtained in three separate experiments (A and B) or in two independent experiments (C).

Inhibition of p38 MAPK activity prevents the up-regulation of costimulatory molecules induced by TNF-α and LPS

To study the roles of the MAPK signaling pathways in DC maturation, we treated immature DC with PD98059, an inhibitor of the ERK1/2 pathway (25), and SB203580, a specific blocker of p38 MAPK (26), and analyzed the surface Ag expression profile of DC after 48 h of culture in the presence of 100 U/ml TNF-α or 10 ng/ml LPS. The up-regulation of CD1a, CD40, CD80, CD86, HLA-DR, and CD83 induced by TNF-α or LPS was prevented by preincubating immature DC with SB203580 (Fig. 2⇓, A–C and Table I⇓). The mean fluorescence intensity (MFI) rather than the percentage of HLA-DR expression was affected by SB203580 treatment. SB203580 inhibition on CD1a, CD80, CD86, and CD83 expression was concentration-dependent, and a significant inhibitory effect was still observed using doses between 1 and 10 μM. Indeed, in these experimental conditions both the percentage of positive cells and the MFI values were lower compared with nontreated cells (Fig. 2⇓ and Table I⇓). The expression of CD40 was inhibited using high doses of SB203580 (10–50 μM). In contrast, a high concentration of PD98059 (50 μM) did not prevent up-regulation of CD86 and HLA-DR expression induced by TNF-α or LPS (Fig. 3⇓A). The expression of other surface molecules including CD1a, CD40, CD80, and CD83 were not modified by PD98059 treatment (data not shown). To confirm that PD98059 was active, cell lysates of DC pretreated or not with 50 μM PD98059 and subsequently stimulated with TNF-α or LPS were subjected to Western blot analysis to detect phosphorylated forms of ERK1/2 and p38 MAPK. Both TNF-α and LPS-induced ERK1/2 phosphorylation were abrogated by 50 μM PD98059, whereas the level of activated p38 MAPK was not affected (Fig. 3⇓B). Conversely, 50 μM SB203580 neither affected the phosphorylation levels of ERK1/2 nor of p46/54 JNK induced after TNF-α or LPS stimulation (data not shown).

• Download figure
• Open in new tab
• Download powerpoint

FIGURE 2.

Effect of SB203580 on TNF-α- and LPS-induced up-regulation of cell surface molecules in monocyte-derived DC. Peripheral blood monocytes were differentiated into immature DC for 7 days with GM-CSF and IL-4. Cells were left untreated in the presence of GM-CSF and IL-4 for 48 h (A). At day 7, DC were pretreated or not for 30 min with increasing concentrations of SB203580 as indicated and then stimulated with 100 U/ml TNF-α (B) or 10 ng/ml LPS (C) for 48 h. Two-dimensional plots show the surface expression profile of DC for the indicated markers evaluated by flow cytometric analysis as described in Materials and Methods. These results are representative of four independent experiments.

• Download figure
• Open in new tab
• Download powerpoint

FIGURE 3.

Effect of PD98059 on TNF-α- and LPS-induced DC maturation. A, Peripheral blood monocytes were differentiated into immature DC for 7 days with GM-CSF and IL-4. At day 7, DC were pretreated for 75 min with 50 μM PD98059 and then stimulated with 100 U/ml TNF-α or 10 ng/ml LPS for 48 h. Two-dimensional plots show the surface expression profile of DC for CD86 and HLA-DR evaluated by flow cytometric analysis as described in Materials and Methods. Numbers in each quadrant represent cell percentages. One representative experiment of three is shown. B, Immature monocyte-derived DC were stimulated or not (medium) with 100 U/ml TNF-α for 5 min or with 10 ng/ml LPS for 15 min. Before stimulation, cells were treated or not (control) for 75 min with 0.1% DMSO or 50 μM PD98059. Cell lysates were subjected to Western blotting. Membranes were incubated with Abs raised against the phosphorylated forms of p42/44 MAPK (ERK1/2-P) and of p38 MAPK (p38-P). Membranes were then reprobed with an Ab against total ERK1/2 for loading control. One experiment of two is shown.

Taken together, these results suggest that in TNF-α- and LPS-induced DC maturation early activation of p38 MAPK, but not of ERK1/2, correlated with the up-regulation of the T cell-costimulatory molecules CD80 and CD86, and with the expression of CD1a, CD83, and HLA-DR.

SB203580 prevents the up-regulation of the allostimulatory capacity and the down-regulation of endocytosis in LPS- or TNF-α-stimulated DC

Fully mature DC elicit higher levels of allogenic T cell proliferation than immature DC (5). Thus, we tested whether SB203580 had an effect on the allostimulatory capacity of immature DC stimulated for 48 h with TNF-α or LPS (Fig. 4⇓, A and B). SB203580 added to immature DC before addition of LPS or TNF-α inhibited the allostimulatory capacity of DC in a concentration-dependent manner. DC treated with a high concentration of SB203580 (50 μM) before TNF-α or LPS stimulation showed an allostimulatory capacity equivalent to that of immature DC. An inhibitory effect was still observed using 10 μM of SB203580. During their maturation process, DC lose their high endocytic activity. This function can be measured by the uptake of FITC-dextran, which is mediated by the mannose receptor (6). Thus, we investigated the effect of SB203580 on the uptake of FITC-dextran by immature DC stimulated with LPS or TNF-α. As shown in Fig. 4⇓C, SB203580 treatment partially preserved the FITC-dextran uptake lost in the absence of the drug. In the presence of 50 μM SB203580, approximately twice more of the cells could still endocytose FITC-dextran. These results show that SB203580 inhibited the LPS and TNF-α-induced up-regulation of the allostimulatory capacity of DC and partially interfered with the down-regulation of FITC-dextran uptake.

• Download figure
• Open in new tab
• Download powerpoint

FIGURE 4.

SB203580 treatment decreases the T cell stimulatory capacity and prevents the down-regulation of FITC-dextran uptake in TNF-α- or LPS-activated DC. DC were treated or not with increasing concentrations of SB203580 as indicated, further stimulated with TNF-α (100 U/ml) (A) or LPS (10 ng/ml) (B), and allogenic T cell proliferation was evaluated. Means of quadruplicates obtained after 5 days of culture are shown. These results are representative of three independent experiments. C, At day 7, DC were treated or not for 30 min with 50 μM SB203580 before stimulation or not for 48 h with TNF-α (100 U/ml) or LPS (10 ng/ml). Cells were then incubated with FITC-dextran for 1 h at 4°C (thin lines) or 37°C (bold lines). The percentage of cells positive for FITC-dextran uptake is indicated. These results are representative of two independent experiments with similar results.

SB203580 treatment abrogates TNF secretion by LPS-stimulated DC

DC maturation is a process during which DC secrete many inflammatory cytokines such as IL-1α, IL-1β, IL-6, IL8, and TNF-α. Thus, we studied TNF secretion by DC stimulated with LPS, and tested the effect of SB203580 on this biological response. Immature DC were treated or not with increasing concentrations of SB203580, and then stimulated for 24 h with LPS. Bioactive TNF was measured after 24 h in the culture supernatants using a bioassay. As shown in Fig. 5⇓A, SB203580 abrogated LPS-induced TNF secretion in a concentration-dependent manner with an ID₅₀ ranging between 0.1 and 1 μM.

• Download figure
• Open in new tab
• Download powerpoint

FIGURE 5.

Effect of SB203580 on LPS-induced proinflammatory cytokine expression in immature DC. A, Seven-day cultured immature DC were pretreated or not for 30 min with increasing concentrations of SB203580 as indicated, and further stimulated with LPS (10 ng/ml) for 24 h. Bioactive TNF was determined in 24-h culture supernatants using a TNF-sensitive cell line as described in Materials and Methods. These data represent means ± SD of triplicates, and are representative of three independent experiments. B and C, Effect of SB203580 on mRNA levels encoding for proinflammatory cytokines induced in LPS-stimulated DC was evaluated by RNase protection assay as described in Materials and Methods. Upper left autoradiograms show kinetic studies of induction of proinflammatory cytokines in LPS-treated cells using multiprobe template set hCK3 (B) and hCK2 (C). Upper right autoradiograms show the effect of increasing concentrations of SB203580 on mRNA levels encoding for proinflammatory cytokines. B and C (lower panels), mRNA levels for two distinct housekeeping genes (L32 and GAPDH). These results are representative of three independent experiments.

SB203580 treatment affects mRNA levels encoding proinflammatory cytokines induced by LPS

To extend our study to other proinflammatory cytokines that are induced during maturation of DC stimulated by LPS and TNF-α, we performed multiplex RNase protection assay allowing the determination of mRNA levels encoding various cytokines. Kinetic studies showed that LPS-induced TNF-α mRNA expression peaked at 1 h (Fig. 5⇑B, upper left panel), whereas LPS-induced IL-1α, IL-1β, and IL-1Ra mRNA expression peaked at 2 h (Fig. 5⇑C, upper left panel). SB203580 inhibited in a concentration-dependent manner the LPS-induced increase of mRNA encoding for all the cytokines described above (Fig. 5⇑, B and C, upper right panels). The induction of TNF-α, Il-1β, and IL-1Ra mRNA was only partially inhibited using high doses of SB203580 (10–50 μM). SB203580 had the same effect in TNF-α-stimulated immature DC (data not shown). These results indicate that inhibition of p38 MAPK activity partially prevents the increase of proinflammatory cytokines mRNA induced by LPS and TNF-α in immature DC.

Contact sensitizers but not irritants induce p38 MAPK phosphorylation

Recent reports demonstrated that tyrosine phosphorylation is an early molecular event in the activation of APCs by skin contact sensitizers (37, 38). This was not observed with irritants. Furthermore, Aiba et al. showed that human monocyte-derived DC responded to contact sensitizers but not to irritants by increased expression of the costimulatory molecule CD86 and of HLA-DR (15). In this context, we first asked whether contact sensitizers and irritants could differentially activate MAPK signaling pathways. To achieve this, we chose two potent and chemically unrelated sensitizers, NiSO₄ and DNFB, as well as two irritants, SDS and BC. We stimulated immature DC with the above chemicals and performed Western blot analysis to detect phosphorylated forms of MAPK. As shown in Fig. 6⇓, A–C, NiSO₄ and DNFB, but not the two irritants, led to p38 MAPK phosphorylation. The phosphorylation was already detectable after 5 min and remained during 1 h of stimulation. Maximal responses were obtained using 500 μM and 2.5 μg/ml of NiSO₄ and DNFB, respectively. Densitometric quantification revealed a 4.3 ± 0.4-fold and a 6.5 ± 0.7-fold induction of p38 MAPK phosphorylation by NiSO₄ and DNFB, respectively (mean ± SD of three experiments). A slight increase in ERK1/2 phosphorylation level was also observed after 15–30 min following NiSO₄ but not DNFB stimulation, whereas p46/54 JNK phosphorylation was not affected by either of the two contact sensitizers (data not shown). Moreover, the two irritants tested did not induce ERK1/2 and p46/54 JNK phosphorylation (data not shown). These results show that p38 MAPK is the MAPK pathway activated by both contact sensitizers tested. In contrast, none of the MAPKs studied were phosphorylated by irritants in our cellular model. Next, we analyzed the surface Ag profile of DC after a 48-h NiSO₄ stimulation. As shown in Fig. 6⇓D, we observed a strong up-regulation of CD40, CD80, CD86, CD83, and HLA-DR. To determine whether the p38 MAPK pathway was involved in the up-regulation of these surface molecules induced by NiSO₄ in DC, cells were pretreated with the p38 MAPK inhibitor. SB203580 blocked almost completely the increase in CD80 and CD83 and, to a lesser extent, the increase of CD86 (Fig. 6⇓D). The surface densities of CD1a, CD40, and HLA-DR were not affected by SB203580. Altogether, these results indicate that p38 MAPK phosphorylation also plays a role in the up-regulation of the surface molecules CD80, CD86, and CD83 induced by contact sensitizers in maturating DC.

• Download figure
• Open in new tab
• Download powerpoint

FIGURE 6.

Contact sensitizers and irritants differentially activate immature DC. A-C, Immature monocyte-derived DC were stimulated or not with the indicated concentrations of contact sensitizers (NiSO₄ and DNFB) or irritants (SDS and BC). Upper panels from (A) and (B) and panels from (C) show kinetic studies of induction of p38 MAPK phosphorylation analyzed by Western blotting. A and B (lower panels), Dose response of contact sensitizers on the activation of p38 MAPK. c, Unstimulated cells; TNF, 5-min stimulation of cells with 100 U/ml TNF-α. Blots showing kinetic studies are representative of results obtained in three separate experiments. Blots dealing with contact sensitizer dose-responses are representative of results obtained in two independent experiments. D, The effect of SB203580 (50 μM) on NiSO₄ (500 μM)-induced cell surface molecule expression in immature DC was evaluated by flow cytometric analysis as described in Materials and Methods. Numbers in each quadrant represent cell percentages. These results are representative of results obtained in two separate experiments with similar results.