Human Dendritic Cells and T Cells Monophosphoryl Lipid A Activates ...

Monophosphoryl Lipid A Activates Both Human Dendritic Cells and T Cells1 Jamila Ismaili,2* Joe¨lle Rennesson,* Ezra Aksoy,* Johan Vekemans,* Benoit Vincart,* Zoulikha Amraoui,* Francois Van Laethem,† Michel Goldman,* and Patrice M. Dubois‡ The induction of dendritic cell (DC) maturation is critical for the induction of Ag-specific T lymphocyte responses and may be essential for the development of human vaccines relying on T cell immunity. In this study, we have investigated the effects of monophosphoryl lipid A (MPL) on human monocyte-derived DC as well as peripheral blood T cells. Calcium mobilization, mitogen-activated protein kinase activation, and the NF-␬B transcription factor were induced after MPL stimulation of DC and required high doses of MPL (100 ␮g/ml). Maturation parameters such as production of IL-12 and increases in cell surface expression of HLA-DR, CD80, CD86, CD40, and CD83 were observed following DC treatment with MPL. However, lower levels of IL-12 were induced by MPL when compared with lipopolysaccharide. This is likely to be related to differences in the kinetics of extracellular signal-related kinase 1/2 and p-38 phosphorylation induced by both molecules. Although maturation induced by MPL was weaker when compared with lipopolysaccharide, it appeared to be sufficient to support optimal activation of allogeneic naive CD45RAⴙ T cell and anti-tetanus toxoid CD4 T cells. MPL at low doses (5 ␮g/ml) had no impact on DC maturation, while its addition to DC-T cell cocultures induced full T cell activation. The observed effect was related to the fact that MPL also acts directly on T cells, likely through their Toll-like receptors, by increasing their intracellular calcium and up-regulating their CD40 ligand expression. Together, these data support a model where MPL enhances T cell responses by having an impact on DC and T cells. The Journal of Immunology, 2002, 168: 926 –932.

D

endritic cells (DC)3 are the most potent APCs. Immature DC reside in peripheral tissues and effectively capture and process native protein Ag and migrate to peripheral lymphoid organs. A maturation process, characterized by IL-12 production and the up-regulation of MHC and costimulatory molecules, is critical for initiation of primary T cell response (1–3). This maturation is driven by inflammatory cytokines, such as TNF-␣, or bacterial products, such as lipopolysaccharide (LPS), encountered in peripheral organs (4). LPS, a constituent of the outer membrane of the cell wall of Gram-negative bacteria, is a complex glycolipid composed of a hydrophilic polysaccharide portion and a hydrophobic domain known as lipid A (5). Recent studies (6 –11) showed that CD14 associates with Toll-like receptor (TLR)2 and TLR4, which are the signaling component of LPS, and consequently triggers its cellular transduction, leading to NF-␬B activation and DC maturation. The adjuvant activity of bacterial products is important not only for antibacterial responses induced by peripheral DC but also for vaccine development. However, LPS is excluded because of its high toxicity, as it is one of the main

causes of septic shock in humans. The adjuvant activity of LPS resides in its lipid moiety, thus lipid A derivatives and analogs have been developed. The removal of an acid labile phosphate group and normal fatty acid groups from diphosphoryl lipid A dramatically reduced the toxicity and pyrogenicity (12). The generated monophosphoryl lipid A (MPL) has been shown to activate APC and to enhance the generation of both Th1- and Th2-specific immune response in mice (13, 14). Its adjuvant effect may be tempting for vaccine development in humans. We tested its effect on maturation and activation of immature DC generated in vitro from adherent monocytes in the presence of GM-CSF and IL-4. Their maturation (surface molecule up-regulation and IL-12 production), their activation (calcium mobilization, mitogen-activated protein kinase (MAPK) activation, and NF-␬B translocation to the nucleus), and their ability to induce T cell responses have been assessed. We also explored the direct effect of MPL on T cells.

Materials and Methods Reagents and medium

*Laboratory of Experimental Immunology, Faculte´ de Medecine, Universite Libre de Bruxelles, Brussels, Belgium; †Laboratory of Animal Physiology, Institut de Biologie et Medecine Moleculaire, Gosselies, Belgium; ‡GlaxoSmithKline, Rixensart, Belgium Received for publication January 16, 2001. Accepted for publication November 6, 2001. 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 has been supported by grants from La Re´gion Wallone, Formation et Impulsion a la Recherche Scientifique et Technologique, and GlaxoSmithKline. 2 Address correspondence and reprint requests to Dr. Jamila Ismaili at the current address: M.R.C. Laboratories, PO Box 273, Fajara, WA, Gambia. E-mail address: [email protected]

LPS from Escherichia coli serotype (0128:B12) or from Salmonella minesota were purchased from Sigma-Aldrich (Bornem, Belgium). The S. minesota LPS derivative MPL (GlaxoSmithKline, Rixensart, Belgium) was prepared as previously described (15). Of note, LPS preparations are unlikely to contain MPL, because extraction of this compound requires both alkaline and acid treatments following organic solvent extraction out of bacterial outer membranes. Cells were grown in RPMI 1640 (Life Technologies, Merelbeke, Belgium) supplemented with 50 ␮M ME, 20 ␮g/ml gentamicin, 2 mM Lglutamine, 1% nonessential amino acids (Life Technologies), and 10% FBS (Perbio, Aalst, Belgium).

DC generation and stimulation

3

Abbreviations used in this paper: DC, dendritic cell; MPL, monophosphoryl lipid A; CD40L, CD40 ligand; TLR, Toll-like receptor; MAPK, mitogen-activated protein kinase; ERK, extracellular signal-related kinase; LPS, lipopolysaccharide; MKK, MAPK kinase. Copyright © 2002 by The American Association of Immunologists

PBMC from healthy volunteers were isolated by density centrifugation of heparinized blood on Lymphoprep (Nycomed, Oslo, Norway), washed with HBSS, resuspended in culture medium, and allowed to adhere in 0022-1767/02/$02.00

The Journal of Immunology culture flasks for 2 h at 37°C. Nonadherent cells were removed by extensive washes and adherent monocytes were cultured for 6 days in the presence of 500 U/ml GM-CSF (Leucomax; Schering-Plough, Kenilworth, NJ) and 800 U/ml IL-4 (Cellgenix, Freiburg, Germany). As assessed by morphology and FACS analysis for the majority of donors, the resulting cell preparation contained ⬎95% DC. No CD14-positive cells and ⬍2% of contaminating B cells were detected in such cell preparations. GM-CSFand IL-4-derived DC were cultured at 106 cells/ml in 24-well plates either in medium alone or in the presence of 1 ␮g/ml LPS or MPL (5, 50, or 100 ␮g/ml) for 24 h. Supernatants were analyzed for IL-12 p-40 production using ELISA kits (BioSource International, Fleurus, Belgium). Detection of the p70 bioactive form of IL-12 was assessed by an indirect test using PBMC production of IFN-␥ as a marker of IL-12 bioactivity. Briefly, 106/ml PBMC were cultured for 48 h in medium alone, MPL (100 ␮g/ml), LPS (1 ␮g/ml), or 1 ml of supernatant from DC treated overnight with either MPL (100 ␮g/ml) or LPS (1 ␮g/ml). To confirm that IFN-␥ production was IL-12 dependent, anti-IL-12 neutralizing Abs or the isotype control (BioSource International) were used at 20 ␮g/ml to block IL-12 activity.

Immunophenotyping of DC Monocyte-derived DC were stained using FITC- or PE-labeled mAb specific for HLA-DR, CD80, CD86, CD83 (BD Biosciences, Mountain View, CA), and CD40 (BioSource International). Briefly, 5 ⫻ 105 cells were incubated with the relevant mAbs or their isotype-matched controls for 20 min at 4°C and washed, and the fluorescence intensity was measured using a FACSCalibur (BD Biosciences).

T cell purification and stimulation Serial dilutions of allogeneic DC (2 ⫻ 104–2 ⫻ 103 DC/well) or autologous DC pulsed with tetanus toxoid (Calbiochem-Novabiochem, La Jolla, CA) at 1 ␮g/ml for 2 h were stimulated with LPS (1 ␮g/ml) or MPL (5 or 50 ␮g/ml). CD4⫹CD45RA⫹ or CD4⫹ T cells and 2 ⫻ 105 cells were added (triplicates) respectively to DC for 5 days. Cytokine release (IFN-␥ and IL-5) was analyzed by ELISA. CD4⫹CD45RA⫹ or CD4⫹ T cells were purified from PBMC using magnetic beads (Miltenyi Biotec, Auburn, CA). To get highly pure T cell preparations, CD4 T cell cell lines were established subsequent to cultures for 3 wk in the presence of 20 U/ml IL-2 (R&D Systems, Oxon, U.K.), PHA (5 ␮g/ml; Life Technologies), and irradiated (9000 rad) allogeneic PBMC. Briefly, 106/ml CD4 T cells were positively selected using Miltenyi Biotec beads (⬎95% purity) and stimulated weekly with irradiated PBMC (105/ml) and PHA. Human rIL-2 was added twice a week in the cultures. Irradiated PBMC were removed using a Ficoll gradient (Nycomed) before each T cell stimulation.

CD40L analysis on activated T cells OKT3 Abs (10 ␮g/ml; Ortho Biotech, Raritan, NJ) or the isotype-matched control Ab (BioSource International) were coated in 96-well flat-bottom plates and 2 ⫻ 105 purified T cells were cultured in triplicate in the presence of MPL (10 ␮g/ml). After 16 h, T cells were analyzed for CD40 ligand (CD40L) expression by intracytoplasmic staining or RT-PCR. For intracytoplasmic staining, T cells were fixed, permeabilized, and stained using PE-conjugated anti-CD40L or the isotypic control (BD Biosciences) and directly analyzed by flow cytometry.

RT-PCR Total RNA was isolated from T cells (unstimulated, OKT3, or MPL stimulated) using TRIzol reagent (Life Technologies) following the instructions of the manufacturer. cDNA was synthesized from mRNA using Moloney murine leukemia virus reverse transcriptase (Life Technologies) and the PCR amplification was performed for ␤-actin and CD40L, TLR2, and TLR 4 in 25 ␮l of reaction mix containing Taq polymerase (Life Technologies) during 40 cycles at 95°C for 40 s, 54°C for 40 s, and 72°C for 2 min. PCR primers for CD40L were 5⬘-TACAACCAAACTTCTCCCCG and 5⬘TAGGCAGTTAACAGGGGGTG. PCR primers for TLR2 were 5⬘-GC CAAAGTCTTGATTGATTGG and 5⬘-TTGAAGTTCTCCAGCTCCTG. PCR primers for TLR4 were 5⬘-TGGATACGTTTCCTTATAAG and 5⬘GAAATGGAGGCACCCCTTC (Life Technologies).

Calcium mobilization assays Human DC, CD4 T cells purified from PBMC, and CD4 T cell lines were washed in calcium/magnesium-free HBSS (Life Technologies) and incubated at 5 ⫻ 106cells/ml with 0.25 mM sulfinpyrazone (Sigma-Aldrich), 100 ␮g/ml pluronic acid F-127 (Molecular Probes, Leiden, The Netherlands), and 5 ␮M Fluo-3 (Molecular Probes). Loading was conducted at

927 37°C for 30 min. Cells were then washed twice in complete medium supplemented with 0.25 mM sulfinpyrazone. Loaded cells were resuspended at a final concentration of 5 ⫻ 105/ml. DC were stimulated with MPL (50 ␮g/ml) or LPS at different doses (1, 5, or 50 ␮g/ml). T cells were stimulated with anti-CD3 mAb (10 ␮g/ml), anti-CD28 mAb (5 ␮g/ml), and rabbit anti-mouse Igs (40 ␮g/ml; Sigma-Aldrich) in the presence or absence of MPL. The FL1 signal for Fluo-3 was calibrated by adding calcium ionophore A23187 (10 ␮g/ml; Calbiochem-Novabiochem) to the reaction buffer containing saturating concentrations of Ca2⫹ to obtain the maximum signal (Fmax) followed by Mg2⫹ (MgCl2, 2 mM) to obtain the minimum signal (Fmin). The intracellular Ca2⫹ concentration (Ca2⫹i) was calculated from the Fluo-3 fluorescence using the following equation: Ca2⫹i ⫽ Kd(F ⫺ Fmin)/(Fmax ⫺ F), where Kd ⫽ 400 nM for the Fluo-3 intracellular dye.

Immunoblot analysis A total of 1 ⫻ 106/ml DC were incubated with or without LPS (10 ␮g/ml) or MPL (100 ␮g/ml) for 2–15 min. Cells were quickly washed twice with cold PBS and lysed in 1% Brij buffer (200 mM boric acid, 150 mM NaCl, pH 8) containing 2 mM PMSF, 5 mM EDTA, 1 mM sodium orthovanadate, and 5 mM NaF). Total cell extracts were resolved by 8% SDS-PAGE, transferred to nitrocellulose membranes (Millipore, Bedford, MA), and incubated with 1/2000 dilution of either phospho-extracellular signal-related kinase (ERK)1/2 and phospho-p38 (New England Biolabs, Leusden, The Netherlands) in 5% BSA, 1⫻ TBS, and 0.1% Tween 20 at 4°C with gentle shaking, overnight. After five washes, membranes were incubated for 1 h at room temperature in a 1/1000 dilution of HRP-conjugated anti-rabbit IgG (Amersham Life Sciences, Little Chalfont, U.K.). Blots were then washed five times and bound Abs were detected using an enzymatic chemiluminescence kit (Amersham Life Sciences). To verify equal loading, membranes were stripped of bound Ab and incubated in 1/1000 dilution Abs to total ERK1/2 and p38 (Santa Cruz Biotechnology, Santa Cruz, CA), washed, and incubated in 1/1000 HRP-conjugated anti-rabbit IgG (Amersham Life Sciences).

EMSA analysis Cells (106/ml) were stimulated with either MPL or LPS for 2 h and lysed, and nuclear extracts were harvested. Cells were washed once with 1⫻ PBS and resuspended in 400 ␮l of buffer A (10 mM HEPES (pH 7.8), 10 mM KCl, 0.1 mM EDTA, 0.5 mM DTT, 1⫻ protease inhibitor mixture (SigmaAldrich), 0.2 mM PMSF (Roche Diagnostic Systems, Somerville, NJ), and 0.5% Triton X-100 (Sigma-Aldrich) for 10 min. Nuclei were pelleted and the cytoplasmic proteins were carefully removed. The nuclei were then resuspended in 30 ␮l of buffer C (50 mM (pH 7.8), 420 mM KCl, 0.1 mM EDTA, 5 mM MgCl2, 10% glycerol, 0.5 mM DTT, 1⫻ protease inhibitor mixture, and 0.2 mM PMSF). Nuclei were then vortexed, incubated on ice for 20 min, and centrifuged at 4°C for 5 min. Protein concentrations were determined by Bio-Rad Protein Assay Reagent (Bio-Rad, Hercules, CA) The double-stranded consensus binding sequences for the appropriate EMSAs comprised the oligonucleotides 5⬘-AGTTGAGGGGACTTTC CCAGG-3⬘ (NF-␬B), and mutant NF-␬B was created with a G3 C substitution (Santa Cruz Biotechnology). Oligonucleotides were end-labeled with [␥-32P]ATP (Amersham Life Sciences) by using T4 polynucleotide kinase (Roche Diagnostic Systems). For the binding reaction, 5–10 ␮g of the extract was added to a reaction mixture containing 2 ␮g of poly(dI-dC) (Pharmacia, Roosendaal, The Netherlands), 4 ␮l of 5⫻ binding buffer (10 mM HEPES, (pH 7.8), 50 mM KCl, 1 mM EDTA, 5 mM MgCl2, 10% glycerol), and 30000 cpm of [32P]-labeled oligonucleotide in a final volume of 20 ␮l and were incubated at room temperature for 15 min. The free and protein-bound oligonucleotides were resolved by electrophoresis on a 5% polyacrylamide gel in a 0.5⫻ Tris-borate EDTA buffer. After electrophoresis, the gel was dried and exposed to autoradiography film (Eastman Kodak, Rochester, NY).

Statistical analysis Data from unstimulated DC were compared with MPL- or LPS-stimulated DC using unpaired (Mann-Whitney) or paired (Wilcoxon) nonparametric tests. A value of p ⬍ 0.05 was accepted as the level of significance.

Results MPL induces maturation of human monocyte-derived DC LPS has been described as an inducer of DC activation and maturation (4). Because MPL is a derivative of LPS we first sought to determine whether MPL also induced DC activation and maturation. DC were cultured overnight in the presence of MPL (5–100

928

␮g/ml), LPS (1 ␮g/ml), or medium only. First, MPL was compared with LPS for its ability to up-regulate HLA-DR, costimulatory molecules such as CD80, CD86, CD40, and the activation marker CD83, on human DC. Up-regulation of DC surface markers required up to 50 –100 ␮g/ml MPL in vitro. Compared with LPS, the induction remained heterogeneous between donors. However, for all donors tested, at least three markers of five were upregulated in response to MPL (Fig. 1). IL-12 production is also an important marker for DC maturation and can be used as a method for selecting Th1-inducing adjuvants. A recurrent and significant enhancement of IL-12 p-40 release was induced by high doses of MPL (Fig. 2). However, IL-12 production remained largely below the amounts released after LPS treatment. As observed for cell surface markers, doses of MPL under 50 ␮g/ml did not induce any significant production of IL-12 p-40. The IL-12 bioactivity was assessed by testing supernatants from unstimulated, MPL-treated, or LPS-treated DC for IL-12 (p-70)dependent IFN-␥ induction in PBMC cultures (Table I). LPS, and to a lesser extend MPL, induced low levels of IFN-␥ by PBMC. Supernatants from DC treated with LPS or MPL increased the IFN-␥ production by PBMC, suggesting the presence of the bioactive IL-12. Anti-IL-12 neutralizing Abs decreased this IFN-␥ release, confirming both the presence of bioactive IL-12 and its direct role in IFN-␥ induction in this culture system. Furthermore, we found that MPL, at a concentration of 100 ␮g/ml, did not require CD14 for DC activation. Indeed, the addition of anti-CD14 Abs in MPL-treated DC cultures failed to diminish DC IL-12 release (Table II) while it inhibited IL-12 production by low doses of LPS (100 ng/ml), which are known to require CD14 to activate DC. This would suggest that MPL signaling to DC is CD14 independent at these concentrations. Together these observations indicate that LPS and its derivative MPL both induce maturation of DC in vitro. MPL induces NF-␬B activation DC maturation driven by LPS has been clearly associated to NF-␬B activation, which is mediated by a member of the Toll-like family of receptors. To determine whether MPL uses similar activation pathways, we monitored its ability to activate NF-␬B translocation into the nucleus. DC were cultured in the presence of MPL for 2 h and nuclear extracts were analyzed for NF-␬B content. As shown in Fig. 3A, MPL was able to induce NF-␬B translocation and activation. Identical results were obtained after treatment of DC with LPS. Using RT-PCR, we also found that DC express significant levels of mRNA coding for both TLR2 and

FIGURE 1. MPL up-regulates cell surface molecules on human DC. Monocyte-derived DC were cultured overnight in the presence of MPL (100 ␮g/ml), LPS (1 ␮g/ml), or medium alone, and surface markers were analyzed by flow cytometry. One representative donor of 10 is shown

MPL ACTIVATES T CELLS AND DC

FIGURE 2. MPL induces IL-12 production by human DC. Human monocyte-derived DC were stimulated by LPS (1 ␮g/ml) or MPL (5, 50, or 100 ␮g/ml) for 24 h and IL-12 p-40 production was subsequently analyzed by ELISA. Data (mean ⫾ SD) are from 15 donors. Statistical analysis concerns unstimulated vs stimulated DC. ⴱⴱ, p ⬍ 0.05.

TLR4 (Fig. 3B). These data are compatible with MPL activation of DC via a TLR. As further evidence of common activation pathways, we found that cells treated with either MPL or LPS upregulated TLR2 mRNA. These treatments had no impact on TLR4 expression. These observations provide indirect evidence in favor of a common activation pathway for both LPS and MPL. MPL induces calcium mobilization and ERK activation in DC To determine whether other signaling pathways are activated by MPL, we monitored calcium mobilization and tyrosine phosphorylation. For calcium mobilization assays, human DC were loaded with Fluo-3 and analyzed by flow cytometry. The addition of MPL (50 ␮g/ml) on DC induced increases in intracellular free calcium while even LPS at 50 ␮g/ml had no impact on these cells (Fig. 4A). LPS used at lower doses (1 or 5 ␮g/ml) also failed to induce calcium signals (data not shown). Phosphorylation patterns and intensities were also found to be slightly different (data not shown). Consequently, we further characterized the MAPK activation pathways involved in MPL signaling and compared them with LPS. This experiment focused on p38 and ERK1/2, which are both involved in IL-12 regulation (16). Results presented in Fig. 4B show that MPL induced a higher

The Journal of Immunology

929

Table I. IL-12 mediated IFN-␥ by PBMCa IFN-␥ (pg/ml)

Culture Conditions

None LPS MPL Supernatant Supernatant Supernatant Supernatant Supernatant

⬍39c 440 ⫾ 345d 110 ⫾ 98 358 ⫾ 102 1998 ⫾ 542 345 ⫾ 425 598 ⫾ 250 116 ⫾ 153

of unstimulated DC of LPS-treated DC ⫹ isotype control of LPS-treated DC ⫹ anti-IL-12 Abs from MPL-treated DC ⫹ isotype control from MPL-treated DC ⫹ anti-IL-21 Abs

a PBMC (106/ml) were cultured in the presence of different stimuli for 48 h and supernatants were tested for nonspecific IFN-␥ induction. LPS (1 ␮g/ml) and to a lesser extent MPL (50 ␮g/ml) induced IFN-␥ production by PBMC. Supernatants from DC stimulated overnight by MPL or LPS induced higher amounts of IFN-␥ ( p ⬍ 0.05). IFN-␥ production was inhibited by anti-IL-12 Abs (20 ␮g/ml) demonstrating the presence of the bioactive form of IL-12 in the supernatants. b Culture condition where anti-IL-12 isotype-matched controls (20 ␮g/ml) were added to PBMC. c Below limits of ELISA detection (39 pg/ml). d Mean ⫾ SD are from seven donors.

ERK1/2 phosphorylation at earlier time points than LPS. However, the kinetics of p-38 phosphorylation was similar for both molecules. This early activation of ERK1/2 proteins might explain why lower levels of IL-12 are induced by MPL. Preliminary results on the pretreatment of DC with PD98059, a specific inhibitor of the MAPK kinase (MKK)/ERK pathway which binds to the inactive forms of MKK1/2 and prevents their activation by upstream regulators, enhanced IL-12 production induced by MPL (data not shown). This would argue in favor of a model in which the impact of MPL on the MKK/ERK pathway leads to partial inhibition of IL-12 production. Differences between LPS and MPL in terms of calcium response and MAPK activation pathways argue that these two molecules induce activation of DC using common as well as distinct pathways. Enhancement of T cell activation by MPL-treated DC In DC MPL induces activation of NF-␬B, up-regulates cell surface markers, and increases IL-12 production. To test whether this maturation is sufficient to promote activation of naive T cells, DC were treated with LPS or MPL. These cells were then used to activate allogeneic naive CD4⫹CD4RA⫹ T cells. Results presented in Fig. 5A show that MPL enhanced T cell activation induced by DC as evidenced by secretion of IFN-␥ and IL-5 in the culture supernatant. Cytokine production induced under these experimental conditions was similar to that seen following LPS treatment of DC. A similar trend was observed when we monitored the impact of MPL on DC-dependent Ag-specific CD4 T cell responses. Tetanus toxoid-pulsed DC treated with MPL significantly enhanced Ag-specific autologous T cell responses (Fig. 5B). Furthermore, both Th1

Table II. MPL requirement for soluble CD14a IL-12 p-40 (pg/ml) Culture Condition

Medium alone LPS MPL

Isotypic control mAb

Anti-CD14 mAb

425 ⫾ 87 25,000 ⫾ 5,400 2,780 ⫾ 273

500 ⫾ 98 2,150 ⫾ 975b 2,845 ⫾ 154

a DC (5 ⫻ 105) were stimulated with LPS (100 ng/ml) or MPL (100 ␮g/ml) for 24 h in the presence of anti-CD14 mAb (20 ␮g/ml) or isotype-matched control (20 ␮g/ml) and IL-12 p40 levels were determined by ELISA. b Value of p ⬍ 0.05.

FIGURE 3. MPL induces NF-␬B activation and modulates TLR2. A, DC were stimulated by LPS (1 ␮g/ml) or MPL (100 ␮g/ml) for 2 h or remained unstimulated, and nuclear extracts were analyzed for their NF-␬B binding activity using EMSA. To assess the specificity of the binding, 100-fold excess of cold NF-␬B probe or irrelevant probe was added to the LPS condition. One of three experiments is shown. B, Unstimulated DC or DC stimulated for 3 h with MPL (100 ␮g/ml) or LPS (1 ␮g/ml) were lysed, their total RNA was extracted, and RT-PCR was performed for TLR2 and TLR4.

and Th2 cytokines were increased under these experimental conditions (Fig. 5B). Somewhat unexpectedly, we also found that low doses of MPL (5 ␮g/ml) that have little impact on DC maturation (see Fig. 2) were able to increase T cell activation in both allo- and Ag-specific stimulations (Fig. 5). These observations show that when used at high doses the impact of MPL can be attributed mainly to its ability to induce maturation of DC, whereas at low doses some other mechanism is likely to be critical for the enhancement of T cell activation. MPL increases calcium mobilization of OKT3- and anti-CD28activated T cells As shown above, low doses of MPL (5 ␮g/ml) are ineffective for induction of DC maturation. However, they are sufficient to induce strong T cell responses. Together these data suggested that MPL might have a direct impact on T cells. Subsequently, we analyzed the expression of TLR2 and TLR4 (Fig. 6A) on purified CD4 T cells or established CD4 T cell lines (see Materials and Methods) as well as their calcium mobilization, which is one of the early events involved in T cell activation (17) (Fig. 6B). Using RT-PCR, we found that resting human T cells also express TLR2 and TLR4 mRNA. Their activation by OKT3 or MPL modulated TLR4 mRNA while it had no impact on the mRNA levels of TLR2. These experiments provide evidence that the bacterial product MPL may use these receptors to act on T cell activation. Data presented in Fig. 6B show that MPL alone increased intracellular free calcium levels in CD4 T cell lines. Despite some differences in the magnitude of the calcium signal, MPL induced this response in both cell lines and peripheral T cells. In contrast, and as reported previously (18), no significant increase of intracellular calcium was observed on T cells stimulated with LPS at the concentrations

930

MPL ACTIVATES T CELLS AND DC

FIGURE 5. MPL enhances T cell responses. Allogeneic DC (A) or autologous DC (B) pulsed with tetanic toxoid (TT) were treated with MPL or LPS and cultured with CD4⫹CD45RA⫹ or CD4⫹ T cells, respectively. Supernatants were analyzed for IFN-␥ and IL-5 produced by activated T cells after 5 days of culture. Data (mean ⫾ SD) are from 10 donors for A and eight donors for B. Statistical analysis concerns unstimulated vs stimulated DC. ⴱⴱ, p ⬍ 0.05.

FIGURE 4. Calcium mobilization and MAPK phosphorylation of MPLtreated DC. A, DC were loaded with Fluo-3, stimulated with MPL (50 ␮g/ml), LPS (50 ␮g/ml), ionomycin (A23187; 10 ␮g/ml), or MgCl2, and analyzed for calcium-dependent Fluo-3 fluorescence by flow cytometry using a FACSCalibur. B, Total cell lysates of DC unstimulated (NS) or stimulated with MPL (100 ␮g/ml) or LPS (10 ␮g/ml) for 2, 5, 10, or 15 min were electrophoresed in 8% acrylamide gels and immunoblotted with antiphospho-ERK1/2 or anti-phospho-p38. Membranes were stripped and analyzed for equal loading using anti-total ERK and anti-total p-38 Abs.

used in our culture system. Besides, subsequent to T cell activation with OKT3 and anti-CD28 Ab, intracellular calcium increased in the presence of low doses of MPL. The impact of MPL on T cells might explain why low doses of this compound which do not induce detectable DC responses enhance T cell responses in MLR reactions. MPL augments CD3-mediated CD40L expression by T cells CD40-CD40L engagement on DC and T cells is known to play a key role in promoting DC maturation and T cell activation. The increase of calcium mobilization of activated T cells in the presence of MPL might act on CD40L expression on T cells (19). Accordingly, we stimulated T cells in the presence of low doses of MPL and monitored CD40L expression using intracytoplasmic staining (Fig. 7A) and RT-PCR (Fig. 7B). Data presented in Fig. 7 show that treatment of purified T cells with low doses of MPL had no impact on CD40L expression. However, MPL was found to increase anti-CD3-induced expression of CD40L by T cells (Fig.

7A). Similar results were obtained using RT-PCR, thus indicating that MPL regulation of CD40L expression occurs at the transcriptional level. Together these data show that MPL has a direct impact on activated T cells in that it enhances their CD40L expression, thus providing maturation and survival signals to DC. This observation is compatible with the observed effects of low doses of MPL on DC-dependent T cell proliferation in vitro.

Discussion This study provides further insight into the mechanisms involved in MPL activation of human DC. Moreover, it provides experimental evidence showing a direct impact of MPL on T cells and highlights some important differences between MPL and its parental molecule, LPS. Both MPL and LPS induced strong activation of the transcription factor NF-␬B, which has been clearly associated to DC maturation. Accordingly, as observed with LPS, MPL induced up-regulation of surface molecules (HLA-DR, CD80, CD86, and CD83) in GM-CSF and IL-4 monocyte-derived DC. This DC response required higher doses of MPL (50 –100 ␮g) than LPS (1 ␮g/ml). The up-regulation of at least three markers of five tested (among them CD86) were consistently increased by MPL in all tested donors. MPL induced a recurrent and significant IL-12 p-40 production compared with unstimulated DC. The maturation induced by MPL was also sufficient for a full activation of naive CD4⫹CD45RA⫹ T cells by allogeneic DC treated with MPL. Moreover, DC pulsed with the soluble protein Ag, tetanus toxoid, also displayed a stronger activation of specific CD4 T cells compared with untreated DC. Both Th1 and Th2 cytokines were induced in MLR by MPL-treated human DC, in contrast to reported data, using murine DC, where MPL was described as a Th1 inducer (13). Our experimental data support previous observations showing that under some experimental conditions MPL induces a mixed Th1/Th2 differentiation which may be optimal for induction of humoral responses (14). Its combination with QS21 has been shown to be 1) necessary to enhance IL-12 production (20), 2) required to

The Journal of Immunology

931

FIGURE 6. MPL increases calcium mobilization of activated T cells. A, CD4 T cells were cultured with OKT3 Ab or its isotype control in the presence or absence of MPL for 2 h and TLR2 and TLR4 mRNA levels were analyzed using and RT-PCR. B, CD4 T cell lines were loaded with Fluo-3 and either stimulated with MPL (50 ␮g/ml), ionomycin (A23187; 10 ␮g/ml), or MgCl2, or stimulated with OKT3, anti-CD28, and rabbit anti-mouse Abs in the presence or absence of MPL. Calcium mobilization over time (sec) was analyzed by flow cytometry.

switch T cell responses induced with a soluble recombinant HIV protein from Th2 to Th1 in mice (14), and 3) required to activate CTL to tumor-specific peptides or HSV proteins in vitro (20, 21). The low amounts of IL-12 produced by MPL matured DC are highly significant when compared with those produced by untreated immature DC. At the same time they are sufficient to induce an efficient activation of T cells. However, these low amounts induced by MPL (100 ␮g/ml) might explain the mixed Th1/Th2 profile displayed by in vitro activated T cells. The understanding of mechanisms controlling IL-12 induction by adjuvants in general and MPL in particular may contribute to improving their impact on cellular immune responses in human therapies. Our experiments show that common signaling pathways are activated by MPL and LPS, as they both induce NF-␬B activation and modulate TLR expression. However, at least at the doses used in our in vitro system, MPL induced rapid intracellular free calcium increases. This was not observed when the same cells were treated with comparable doses of LPS. Likewise, preliminary results indicated that LPS and MPL induced a different pattern of protein phosphorylation (data not shown), suggesting that different signaling pathways may be activated by these two molecules. The increase of intracellular calcium, which follows MPL treatment of DC, may contribute to the differences in the kinetics of mitogenactivated ERK-activating kinase/ERK activation and could explain why low amounts of IL-12 are produced using this compound. Indeed, IL-12 production by LPS-treated DC has been associated to the signaling cascade involving the activated stress kinases c-Jun N-terminal kinase and p-38 activity. In contrast, activation of

the ERK pathway has been reported to decrease IL-12 production and to be preferentially induced by LPS in macrophages (16). We found that MPL induced ERK1/2 phosphorylation with faster kinetics than LPS. In contrast, the kinetics of p-38 phosphorylation was found to be similar for both molecules. Based on current understanding of IL-12 regulation, rapid ERK1/2 phosphorylation induced by high doses of MPL could well contribute to decreasing IL-12 production. In support of this, pretreatment of DC with PD98059, which binds to the inactive forms of MKK1/2 and prevents their activation by upstream regulators, enhanced IL-12 production induced by MPL (data not shown). These results suggest that MPL should be associated with other adjuvants which act on IL-12 induction to improve the Th profile of the cellular immune response. Low doses of MPL were ineffective at inducing DC maturation, while they were sufficient to induce a strong T cell activation in MLR. This observation suggests a direct effect of MPL on T cells. TLR2 and TLR4 expression by T cells (22) and the ability of LPS to induce T cell proliferation (23) support the notion that bacterial compounds or their derivatives can act directly on T cells. In this study we have also detected TLR2 and TLR4 expression in human T lymphocytes and found that MPL acts on T cells directly. In contrast to what is observed for LPS, which failed to induce high calcium signals (17) at doses which are biologically active, MPL increased intracellular free calcium in CD4 T cell lines and, to a lesser extent, in CD4 T cells purified from peripheral blood. As expected, this intracellular calcium increase had no apparent effect on T cell activation in the absence of TCR triggering. The calcium

932

MPL ACTIVATES T CELLS AND DC This is likely to be due to MPL inducing low amounts of IL-12. Thus, combinations of MPL with other compounds which increase IL-12 production should enhance human Th1 cellular immune responses.

Acknowledgments We thank Dr. M. Moser for critical reading of this manuscript.

References

FIGURE 7. MPL increases CD40L expression on T cells. Purified CD4 T cells were cultured with OKT3 Ab or its isotype control (unstimulated cells) with or without MPL (10 ␮g/ml) for 16 h. A, CD40L up-regulation was performed using intracytoplasmic staining (dotted line, isotype controls; solid line, CD40L staining. B, RT-PCR performed on the same cultures confirmed CD40L increased by MPL on OKT3-activated T cells. Fibroblasts transfected with CD40L were used as internal control for specificity of the RT-PCR.

mobilization subsequent to OKT3 and anti-CD28 stimulation was also enhanced, indicating a direct contribution of MPL to T cell activation and suggested an effect on CD40L expression by T cells. MPL alone had no effect on CD40L expression on resting T cells. Rather, TCR engagement was required for CD40L expression and MPL further enhanced CD40L T cell surface expression levels. These elements support a model where low doses of MPL may act first on T cells by increasing their intracellular calcium and cell surface CD40L expression when stimulated by immature DC in MLR. In this case, CD40L-CD40 interaction would induce maturation of DC. This model could explain why low doses of MPL enhance T cell activation in MLR. A high dose of MPL is required for DC activation, while only low doses are sufficient to act on T cells. This discordance might be related to a different signaling threshold between both subsets, because DC were directly stimulated by MPL, whereas T cells were activated trough TCR and CD28 molecules and MPL was only a coactivation molecule. Together, our data show that MPL enhances T cell responses by having an impact on both DC and T cells. At high doses, MPL effects on T cell activation can be attributed to its ability to induce DC maturation, whereas at low doses of MPL effects on DCs and T lymphocytes synergize to achieve T cell activation. The T cell response induced by DC appears to be a mixture of Th1 and Th2.

1. Banchereau, J., and R. M. Steinman. 1998. Dendritic cells and the control of immunity. Nature 392:245. 2. Rescigno, M., C. Winzler, D. Delia, C. Mutini, M. Lutz, and P. Ricciardi-Castagnoli. 1997. Dendritic cell maturation is required for initiation of the immune response. J. Leukocyte Biol. 61:415. 3. Rescigno, M., M. Martino, C. L. Sutherland, M. R. Gold, and P. Ricciardi-Castagnoli. 1998. Dendritic cell survival and maturation are regulated by different signaling pathways. J. Exp. Med. 188:2175. 4. De Smedt, T., B. Pajak, E. Muraille, L. Lespagnard, E. Heinen, P. De Baetselier, J. Urbain, O. Leo, and M. Moser. 1996. Regulation of dendritic cell numbers and maturation by lipopolysaccharide in vivo. J. Exp. Med. 184:1413. 5. Schletter, J., H. Heine, A. J. Ulmer, and E. T. Rietschel. 1995. Molecular mechanisms of endotoxin activity. Arch. Microbiol. 164:383. 6. Cunningham, M. D., R. A. Shapiro, C. Seachord, K. Ratcliffe, L. Cassiano, and R. P. Darveau. 2000. CD14 employs hydrophilic regions to “capture” lipopolysaccharides. J. Immunol. 164:3255. 7. Muzio, M., D. Bosisio, N. Polentarutti, G. D’amico, A. Stoppacciaro, R. Mancinelli, C. van’t Veer, G. Penton-Rol, L. P. Ruco, P. Allavena, and A. Mantovani. 2000. Differential expression and regulation of toll-like receptors (TLR) in human leukocytes: selective expression of TLR3 in dendritic cells. J. Immunol. 164:5998. 8. Golenbock, D. T. 2000. Toll-like receptor 4 imparts ligand-specific recognition of bacterial lipopolysaccharide. J. Clin. Invest. 105:497. 9. Yang, R. B., M. R. Mark, A. L. Gurney, and P. J. Godowski. 1999. Signaling events induced by lipopolysaccharide-activated Toll-like receptor 2. J. Immunol. 163:639. 10. Kirschning, C. J., H. Wesche, T. Merrill Ayres, and M. Rothe. 1998. Human toll-like receptor 2 confers responsiveness to bacterial lipopolysaccharide. J. Exp. Med. 188:2091. 11. May, M. J., and S. Ghosh. 1998. Signal transduction through NF-␬B. Immunol. Today 19:80. 12. Ribi, E. 1986. Structure-function relationship of bacterial adjuvants. In Advances in Carriers and Adjuvants for Veterinary Biologicals. R. M. Nervig, P. M. Gough, M. L. Kaeberle, and C. A. Whestone, eds. Iowa State Univ. Press, Ames, p.35. 13. De Becker, G., V. Moulin, B. Pajak, C. Bruck, M. Francotte, C. Thiriart, J. Urbain, and M. Moser. 2000. The adjuvant monophosphoryl lipid A increases the function of antigen-presenting cells. Int. Immunol. 12:807. 14. Moore, A., L. McCarthy, and K. H. Mills. 1999. The adjuvant combination monophosphoryl lipid A and QS21 switches T cell responses induced with a soluble recombinant HIV protein from Th2 to Th1. Vaccine 17:2517. 15. Ulrich, J. T., and K. R. Myers. 1995. Monophosphoryl lipid A as an adjuvant. Pharm. Biotechnol. 6:495. 16. Hacker, H., H. Mischak, G. Hacker, S. Eser, N. Prenzel, A. Ullrich, and H. Wagner. 1999. Cell type-specific activation of mitogen-activated protein kinases by CpG-DNA controls interleukin-12 release from antigen-presenting cells. EMBO. J. 18:6973. 17. Rudd, C. E., O. Janssen, Y. C. Cai, A. J. Da Silva, M. Raab, and K. V. Prasad. 1994. Two-step TCR ␨/CD3-CD4 and CD28 signaling in T cells: SH2/SH3 domains, protein-tyrosine and lipid kinases. Immunol. Today 15:225. 18. Chien, E. J., C. H. Chien, J. J. Chen, S. W. Wang, and D. J. Hsieh. 2000. Bacterial lipopolysaccharide activates protein kinase C, but not intracellular calcium elevation, in human peripheral T cells. J. Cell. Biochem. 76:404. 19. Nusslein, H. G., K. H. Frosch, W. Woith, P. Lane, J. P. Kalden, and B. Manger. 1996. Increase of intracellular calcium is the essential signal for the expression of CD40 ligand. Eur. J. Immunol. 26:846. 20. Silla, S., F. Fallarino, T. Boon, and C. Uyttenhove. 1999. Enhancement by IL-12 of the cytolytic T lymphocyte (CTL) response of mice immunized with tumorspecific peptides in an adjuvant containing QS21 and MPL. Eur. Cytokine Network 10:181. 21. Mikloska, Z., M. Ruckholdt, I. Ghadiminejad, H. Dunckley, M. Denis, and A. L. Cunningham. 2000. Monophosphoryl lipid A and QS21 increase CD8 T lymphocyte cytotoxicity to herpes simplex virus-2 infected cell proteins 4 and 27 through IFN-␥ and IL-12 production. J. Immunol. 164:5167. 22. Matsuguchi, T., T. Kensuke, M. Tipayaratn, and Y. Yoshikai. 2000. Gene expressions of lipopolysaccharide receptors, Toll-like receptors 2 and 4, are differently regulated in mouse T lymphocytes. Blood 95:1378. 23. Vogel, S. N., M. L. Hilfiker, and M. J. Caulfield. 1983. Endotoxin-induced T lymphocyte proliferation. J. Immunol. 130:1774.