Selective regulation of CD40 expression in murine dendritic cells by thiol antioxidants

Abstract

Interaction of CD40 on dendritic cells (DC) with CD40 ligand induces interleukin-12 (IL-12) production by these DC during the antigen presentation. Thus, the level of CD40 expression appears to influence the capability of DC to induce a T helper 1 (Th1) response. However, it is not fully understood how CD40 expression on DC is regulated. In the present study, we examined the effects of the reducing agents, N-acetyl-l-cysteine (NAC) and reduced glutathione (GSH), on tumour necrosis factor-α (TNF-α)-induced phenotypic changes in murine DC. TNF-α markedly increased the expression on DC of major histocompatibility complex (MHC) and the costimulatory molecules, CD40, CD80 and CD86. Both NAC and GSH completely abolished the TNF-α-induced enhancement of CD40 expression, but had no considerable effect on the expression of CD80, CD86 and MHC. The marked decrease of CD40 protein with NAC was also detected by Western blotting, but was not associated with the expression level of CD40 mRNA in DC. Thus, NAC appears to reduce CD40 expression on DC by regulating a post-transcriptional pathway. The inhibitory effect of NAC or GSH on TNF-α-induced CD40 expression was released by simply removing these agents from the culture. In contrast, culture of TNF-α-treated DC with NAC or GSH markedly decreased the expression of CD40 within 12 hr. These results demonstrate that reducing agents selectively, rapidly and reversibly regulate CD40 expression on DC, which may eventually affect the capability of DC for Th1/Th2 polarization.

Introduction

Dendritic cells (DC) are potent antigen-presenting cells (APC) and play major roles in the regulation of immune responses to various antigens.^1–3 Thus, DC exhibit a unique ability to activate naive T cells depending on their maturational stage. Immature DC that bear low amounts of major histocompatibility complex (MHC) and costimulatory molecules, including CD80, CD86 and CD40, show modest ability to activate T cells. Upon encountering foreign antigen, immature DC are rapidly activated and become mature. These mature DC exhibit a potent ability to activate T cells that corresponds to high surface expression of the MHC and costimulatory molecules.^1,2,4 Thus, the ability of DC to present antigen appears to be, at least in part, controlled by the expression levels of MHC and of costimulatory molecules on the surface.

CD40, a member of the tumour necrosis factor receptor (TNFR) superfamily, is a 45 000–50 000 MW type I phosphoprotein that is expressed on a wide range of cell types such as DC, B cells, macrophages and non-haematopoietic cells.^5,6 It has been shown that interaction of CD40 and CD40 ligand (CD40L) plays a crucial role in immune responses. Thus, interaction of CD40L and CD40 on B cells results in the B-cell proliferation and differentiation to antibody-secreting plasma cells.⁷ During antigen presentation, the interaction of CD40 on DC with CD40L on T cells induces interleukin-12 (IL-12) production by the DC, which leads to T helper type 1 (Th1) differentiation of the T cells.^8–10

CD40 expression on DC is increased by various maturational stimuli, such as inflammatory cytokines and certain bacterial components.^11,12 In addition, it seems that the level of CD40 expression on DC corresponds to their ability to produce IL-12 upon stimulation with CD40L. Recently, Tone et al.¹³ reported that Sp1 and nuclear factor-κB (NF-κB) were key transcription factors in the basal regulation or up-regulation of CD40 expression following lipopolysaccharide (LPS) stimulation in a macrophage cell line and bone marrow-derived DC. However, the precise mechanisms underlying the regulation of CD40 expression in DC have not been fully elucidated.

N-acetyl-l-cysteine (NAC) is an antioxidant drug and acts as a scavenger for reactive oxygen species in mammalian cells. NAC also acts as a precursor for reduced glutathione (GSH), a physiological reducing agent, which plays important roles in various cellular functions.^14,15 It has been reported that NAC exerts immunomodulatory activity, such as enhancement of T-cell proliferation^16,17 and inhibition of IL-4 production by T cells.¹⁸ In humans, NAC inhibits DC function.¹⁹ Furthermore, it has recently been reported that NAC down-regulates the expression on B cells of CD40 and CD27.²⁰ Thus, NAC plays different roles in a variety of cells involved in immune responses. However, the precise mechanism(s) underlying the complex effects of these reducing agents on immunocytes, including DC, have not been fully elucidated.

Winzler and colleagues²¹ established a growth factor-dependent immature DC line from splenocytes of C57BL/6 mice. Although this DC line showed an immature phenotype, various activating signals such as living bacteria or cytokines including TNF-α and IL-1 promoted full maturation of this precursor. Thus far, using the similar in vitro differentiation system of DC, a number of important findings has been reported and verified.^21–26 An immature DC line, BC1 cells, derived from BALB/c splenocytes,²⁷ had been previously established following the method of Winzler et al.²¹ These BC1 cells represented the normal DC population with regard to phenotypic and functional properties.^27–30

In the present study, using BC1 cells as well as spleen-derived primary cultures of DC, the effects of NAC and GSH on the TNF-α-induced phenotypical maturation of DC were examined. It is demonstrated herein that treatment of DC with TNF-α in the presence of a reducing agent generates mature phenotype DC expressing high levels of MHC and costimulatory molecules but lacking CD40 expression. These findings provide a new strategy that may selectively regulate CD40 expression on DC and modify the subsequent immune responses.

Materials and methods

Reagents and antibodies

Recombinant murine granulocyte–macrophage colony-stimulating factor (GM-CSF) and TNF-α were purchased from PeproTech (London, UK). fluorescein isothiocyanate (FITC) -conjugated anti-mouse CD11c monoclonal antibody (mAb; HL3), FITC-conjugated anti-mouse CD40 mAb (3/23), phycoerythrin (PE)-conjugated anti-mouse CD80 mAb (16-10A1), FITC-conjugated anti-mouse CD86 mAb (GL1), PE-conjugated anti-mouse CD86 mAb (GL1), biotin-conjugated anti-mouse H-2K^d mAb (SF1-1.1), biotin-conjugated anti-mouse I-A^d mAb (AMS-32.1) and streptavidin Cy-Chrome™ were obtained from PharMingen (La Jolla, CA). As control immunoglobulin G (IgG), FITC-conjugated rat IgG2aκ, and FITC-conjugated hamster IgG (anti-TNP) were obtained from PharMingen. PE-conjugated rat IgG2a, biotin-conjugated mouse IgG2a, and biotin-conjugated mouse IgG2b were purchased from Immunotech (Marseille, France). PE-conjugated hamster IgG was obtained from Caltag Laboratories (Burlingame, CA). NAC, GSH, LPS, and TRI REAGENT™ were purchased from the Sigma Chemical Co. (St Louis, MO). NAC and GSH were dissolved in Iscove's modified Dulbecco's medium (IMDM) as previously described.¹⁹

DC cultures

A DC line (BC1) was generated from BALB/c mouse spleen as previously described.^27,28 BC1 cells were cultured and expanded in IMDM containing 10% fetal calf serum (FCS),10 ng/ml mouse recombinant GM-CSF, 30% NIH-3T3 supernatant (hereafter referred as R1 medium).^21,27 BC1 cells cultured for 100–180 days were used for experiments.

Spleen-derived DC (SDDC) were generated from C57BL/6 (B6) splenocytes as described elsewhere.^28,29 Briefly, B6 female mice were purchased from Charles River Japan (Atsugi, Japan). Spleens were removed from B6 mice (8 weeks of age) and single-cell suspensions were prepared by passage through a stainless mesh. Erythrocytes were lysed by treatment with ammonium chloride. Remaining unfractionated cell populations were plated at a density of 1 × 10⁶ cells/ml in R1 medium. Cultures were fed with fresh R1 medium every 3–4 days. On the 14th day of culture, suspended and weakly adherent cells (DC-enriched fraction) were collected by treatment with 3 mm ethylenediaminetetraacetic acid (EDTA). DC were then positively selected using anti-CD11c (N418) MicroBeads and a magnetic antibody cell-sorting (MACS) column (Miltenyi Biotec, Auburn, CA). The purified cells were more than 94% CD11c⁺ and were used as SDDC.

Flow cytometry

BC1 cells (2 × 10⁵ cells/ml) were pretreated with or without NAC or GSH for 1 hr and then treated with TNF-α (40 ng/ml) for 24 hr in the presence of each chemical. The cells were detached with 3 mm EDTA for 5 min at 37°. The cells were incubated with 2.4G2 (rat anti-mouse FcII/III receptor, CD32) supernatant to prevent binding to FcRII/III.³¹ These cells were then stained using FITC-, PE-, or biotin-conjugated mAb and streptavidin-Cy-Chrome™. For staining antigen expressed both on the cell surface and in the cytoplasm, cells were fixed with 4% paraformaldehyde in NaH₂PO₄ (pH 7·4) on ice, washed in phosphate-buffered saline (PBS) containing 1% bovine serum albumin (BSA) and 0·1% NaN₃, and then permeabilized with 0·1% saponin buffer (0·1% saponin, 1% BSA, 0·1% NaN₃ in PBS) for 15 min Thereafter, the cells were stained with specific mAb as described above. Flow cytometry was performed on EPICS^® XL (Coulter Co., Miami, FL). Data are presented as mean fluorescence intensity (MFI) subtracted the level of isotype-matched control antibody.

Western blotting

BC1 cells (1 × 10⁶ cells) were pretreated with NAC for 1 hr and then treated with TNF-α for 24 hr in the presence of NAC. These cells were detached with 3 mm EDTA for 5 min at 37°, washed in ice-cold PBS containing 2 mm EDTA, and lysed in an ice-cold lysis buffer containing 0·2 mm EDTA, 20 mm Tris–HCl (pH 8·0), 100 mm NaCl, 3% Nonidet P-40, 50 mm NaF, 10 mm sodium pyrophosphate, 2 mm orthovanadate, 10 μg/ml each of aprotinin and leupeptin. After incubation for 30 min on ice, samples were centrifuged at 13 000 g for 15 min, and supernatants were collected. An aliquot corresponding to 50 μg of total protein of each sample was separated by 7·5% sodium dodecyl sulphate–polyacrylamide gel electrophoresis (SDS–PAGE), and transferred onto a polyvinylidene-difluoride membrane. The membrane was probed with an anti-CD40 mAb (1 : 500; T-20; Santa Cruz Biotechnology, Santa Cruz, CA), and developed with horseradish-peroxidase-conjugated secondary antibody by enhanced chemiluminescence. The sample membrane was reprobed with rabbit anti-actin mAb (Sigma) as described above.^6,32 The intensity of the specific band was analysed with science laboratory 99 image gauge Version 3·4 software (Fuji Film, Tokyo, Japan).

In vitro kinase assays

BC1 cells (2 × 10⁶−5 × 10⁶ cells) were incubated in 5% FCS IMDM [for the extracellular signal-related kinase 1/2 (ERK1/2) and p38 mitogen-activated protein kinase (p38mapk) activity] or R1 medium [for the stress-activated protein kinase/c-Jun N-terminal kinase (SAPK/JNK) activity] for 1 hr at 37°, and then treated with TNF-α (40 ng/ml) or LPS (5 μg/ml) for 15 min. Reactions were stopped by rapid cooling on ice. The cells were washed with ice-cold PBS containing 2 mm EDTA. The cell's kinase activity of ERK1/2, p38mapk, or SAPK/JNK was determined using a commercially available kinase assay kit (New England BioLabs, Beverly, MA) according to the manufacturer's protocol. The cell lysate was subjected to immunoprecipitation with immobilized anti-phospho–ERK1/2 mAb, anti-phospho–p38mapk mAb, or c-Jun fusion protein beads. The complex was then subjected to kinase reaction with a substrate [Ets-like-protein-1 (Elk-1) for ERK1/2, activating transcription factor-2 (ATF-2) for p38mapk, or c-Jun for SAPK/JNK]. After the kinase reaction, the amount of phospho–Elk-1, phospho–ATF-2, or phospho–c-Jun was determined by Western blotting as described above.

Reverse transcriptase–polymerase chain reaction (RT-PCR)

BC1 cells (2 × 10⁵ cells/ml) were pretreated with NAC (20 mm) for 1 hr at 37° and then incubated with TNF-α (40 ng/ml) for 10 hr in the presence of the NAC. Total RNA was isolated from the cells using the RNA isolation reagent, TRI REAGENT™, according to the manufacturer's protocol, and then quantified spectrophotometrically. Total RNA (0·4 μg) was reverse transcribed in a 15-μl volume containing 10 mm Tris–HCl (pH 8·3), 50 mm KCl, 5 mm MgCl₂, 1 mm each dNTP (dATP, dGTP, dTTP and dCTP), 60 U RNase inhibitor (TaKaRa, Kyoto, Japan), 3·75 U avian myeloblastosis virus reverse transcriptase (TaKaRa), and 150 pmol Random 9-mers for 10 min at 30° and for 30 min at 42°.

Five microlitres of cDNA was amplified in a 25-μl volume containing 10 mm Tris–HCl (pH 8·3), 50 mm KCl, 2·5 mm MgCl₂, 200 mm each dNTP (dATP, dGTP, dTTP and dCTP), 200 nm appropriate primer pair, and 0·625 U Taq DNA polymerase (TaKaRa). Primer sequences for CD40 and HPRT genes were obtained from previous reports.^13,33 After an initial denaturation step, the cDNA mixture was subjected to amplification cycles, each cycle consisting of denaturation (94° for 30 seconds), annealing (58° for 30 seconds), and extension (72° for 30 seconds) using a thermal cycler. The number of amplification cycles was 20 (CD40 and HPRT). An aliquot (10 μl) of the PCR product was electrophoresed on 2% agarose gel, and amplified DNA fragments were stained with SYBR Green I (Molecular Probes, Eugene, OR).³⁴ The fluorescence intensity of the specific band was visualized and analysed using FLA-3000 (Fuji Film) with science laboratory 99 image gauge Version 3·4 software (Fuji Film).

Statistical analysis

The statistical significance was assessed using analysis of variance (anova) followed by a Scheffe's test.

Results

Effects of NAC and GSH on the expression of MHC and costimulatory molecules on BC1 cells

It has been demonstrated that unstimulated BC1 cells are phenotypically and functionally immature DC.^27–30 Various activating signals such as TNF-α and LPS promoted the phenotypic and functional maturation of BC1 cells. Figure 1(a) shows the effects of TNF-α on the expression of the MHC and costimulatory molecules on the surface of BC1 cells and the influences of treatment with the reducing agents, NAC or GSH, on this expression. Treatment of BC1 cells with NAC or GSH alone abolished slight but constitutive expressions of CD40 and CD80 (Fig. 1a, lower). In agreement with our previous study,^27,30 TNF-α markedly increased the expression of CD40, CD86, CD80, MHC class II (I-A^d) and MHC class I (H-2K^d) on BC1 cells (Fig. 1a). It should be noted that treatment with either NAC or GSH almost completely inhibited the TNF-α-induced CD40 expression, but only slightly decreased CD86 and CD80 expressions on BC1 cells. Neither NAC nor GSH treatment showed a significant influence on the I-A^d and H-2K^d expression on TNF-α-treated BC1 cells.

Figure 1.

Effects of NAC and GSH on TNF-α-induced surface expressions of MHC and costimulatory molecules on murine DC. BC1 cells (a) and spleen-derived DC (SDDC) (b) were pretreated with NAC (20 mm) or GSH (20 mm) for 1 hr and then treated with TNF-α (40 ng/ml) for 24 hr in the presence of each agent. Cell-surface expressions of MHC and costimulatory molecules were analysed by flow cytometry. Representative FACS profiles [CD40, CD86 and I-A^d (a); CD40, CD86 and CD80 (b)] are shown (upper panel). Each column [CD40, CD86, CD80, I-A^d and H-2K^d (a); CD40, CD86 and CD80 (b)] represents the mean ± SE of at least three independent experiments (lower panel). Statistical significance was calculated by Scheffe's test (*P < 0·05; **P < 0·01; ***P < 0·001).

The effects of NAC and GSH were then analysed using another DC system. Figure 1(b) shows the effects of NAC or GSH on the expression of CD40, CD86 and CD80 on SDDC. The SDDC were generated by culturing B6 splenocytes with GM-CSF and fibroblast supernatant for 14 days.^28,29 SDDC expressed considerable levels of CD40, CD86 and CD80 when cultured with medium alone for 24 hr (Fig. 1b, lower panel). NAC or GSH treatment inhibited this up-regulation of CD40. TNF-α markedly increased CD40 expression on SDDC, and the TNF-α-induced enhancement of CD40 expression was almost completely inhibited by treatment with NAC or GSH (Fig. 1b, lower left). TNF-α slightly increased the expression of both CD86 and CD80 on SDDC. Again, neither NAC nor GSH showed significant effects on these expressions. These findings demonstrate that both NAC and GSH selectively inhibit the surface expression of CD40 on murine DC.

We then examined the effect of NAC on the total amount of CD40 protein expressed on the cell surface and in the cytoplasm of BC1 cells. Figure 2(a) shows flow cytometric analysis on permeabilized BC1 cells stained with specific mAb. As was shown in the analysis of intact cells, NAC completely inhibited TNF-α induced expression of CD40, and slightly decreased that of CD86 in BC1 cells (Fig. 2a, lower). On the other hand, I-A^d expression was unaffected by the NAC treatment.

Figure 2.

Effect of NAC on total amount of CD40 protein expressed on the cell surface and in the cytoplasm of BC1 cells. BC1 cells were pretreated with NAC (20 mm) for 1 hr and then treated with TNF-α (40 ng/ml) for 24 hr in the presence of NAC. (a) Flow cytometric analysis: the cells were fixed, permeabilized and stained with each specific mAb. The cell surface expressions were analysed by flow cytometry. Representative FACS profiles are shown (upper panel). Each column represents the mean ± SE of three independent experiments (lower panel). (b) Western blotting: the total cell lysate was subjected to Western blotting using specific antibody against CD40 or actin. B-cell lymphoma (A20) and human T-cell leukaemia (Jurkat) were used as positive and negative controls, respectively (left). The relative intensity of the specific band represents the ratio to the intensity of A20 (right). Each column represents the mean ± SE of three independent experiments. Statistical significance was calculated by the Scheffe's test (**P < 0·01; ***P < 0·001).

Figure 2(b) shows Western blotting using total cell lysate of BC1 cells. CD40 expression in BC1 cells was markedly increased by treatment with TNF-α (lane 5). The TNF-α-induced enhancement of CD40 expression was significantly decreased by NAC treatment (lane 6). Mean relative intensities of CD40 bands from three independent experiments are shown in the right-hand panel of Fig. 2(b). It was demonstrated that the level of CD40 was significantly lower in both NAC- and TNF-α-treated BC1 cells than in BC1 cells treated with TNF-α alone.

Effect of NAC on MAPK pathways

Recently, it was reported that TNF-α induced tyrosine phosphorylation and activation of ERK, p38mapk and SAPK/JNK in human monocyte-derived DC, and IL-10 suppressed phosphorylation of these MAPK and the maturation of DC.³⁵ Thus, it seems that these MAPK are involved in the signal transduction for DC maturation corresponding to the increase in expression of the MHC and costimulatory molecules, including CD40. We then examined the effects of NAC on ERK, p38mapk and SAPK/JNK activities in BC1 cells treated with TNF-α. As shown in Fig. 3, ERK, p38mapk and SAPK/JNK activities in BC1 cells were increased following treatment with TNF-α. NAC showed no influence on ERK, p38mapk and SAPK/JNK activities either in the presence or absence of TNF-α.

Figure 3.

The effect of NAC on TNF-α-induced activation of ERK 1/2, p38mapk and SAPK/JNK. BC1 cells were pretreated with NAC (20 mm) for 1 hr and then treated with TNF-α (40 ng/ml) for 15 min in the presence of NAC. An in vitro kinase assay was performed using Elk-1, ATF-2, or c-Jun as a substrate for ERK 1/2 (a), p38mapk (b), or SAPK/JNK (c), respectively. Phosphorylation of Elk-1, ATF-2, or c-Jun was detected by Western blotting using antibodies against phospho–Elk-1, phospho–ATF-2, or phospho–c-Jun. Data are representative of three (a, b) or two (c) independent experiments (upper panel). Each column represents the mean ± SE of three (a, b) or two (c) independent experiments (lower panel).

Effect of NAC on CD40 mRNA expression

We then examined the possibility that NAC inhibits CD40 expression at the mRNA level in BC1 cells. For semi-quantitative assessment of the CD40 mRNA levels by RT-PCR, a two-fold dilution of cDNA solution prepared from LPS-treated BC1 cells was amplified with CD40 or HPRT-specific primers. Since LPS-treated BC1 cells expressed higher levels of CD40 mRNA than TNF-α-treated BC1 cells (data not shown), cDNA solution from these LPS-treated BC1 cells was used for the semi-quantitative assessment. The fluorescence intensity of the PCR product of CD40 or HPRT was decreased in a dose-dependent manner linked to the cDNA level in the reaction mixture (Fig. 4a). Thus, our analysis of CD40 messages by the PCR method was semi-quantitative. The amounts of CD40 mRNA in BC1 cells treated with TNF-α and/or NAC were then analysed. As shown in Fig. 4(b), CD40 mRNA was detected at a low level in unstimulated BC1 cells, and TNF-α markedly increased the CD40 mRNA level. NAC showed no considerable effect on the TNF-α-induced expression of CD40 mRNA in BC1 cells (Fig. 4b, lower).

Figure 4.

Effect of NAC on the level of CD40 mRNA expression. (a) Semiquantitative PCR of target cDNA using CD40- or HPRT-specific primer pair. Twofold dilutions of cDNA from BC1 cells treated with LPS (5 μg/ml) were amplified with CD40- or HPRT-specific primer-pair (upper panel). Relative intensity is shown (lower panel). (b) Effect of NAC on TNF-α-induced CD40 mRNA expression. BC1 cells were pretreated with NAC (20 mm) for 1 hr and then treated with TNF-α (40 ng/ml) for 10 hr in the presence of NAC. CD40 mRNA levels were analysed by RT-PCR. Data are representative of three independent experiments (upper panel). Each column represents the mean ± SE of three independent experiments (lower panel). Statistical significance was calculated by the Scheffe's test (***P < 0·001).

The time–course of CD40 expression on TNF-α-treated or TNF-α + NAC-treated BC1 cells in the second culture

To analyse the influence of NAC on the CD40 expression of DC at later stages, CD40 expression was sequentially quantified using a secondary culture system on BC1 cells. BC1 cells were treated with TNF-α in the presence or absence of NAC for 24 hr (first culture). These cells were then washed and cultured again with or without NAC for the indicated time (second culture) (Fig. 5). The level of CD40 expression on BC1 cells pretreated with TNF-α plus NAC in the first culture (TNF + NAC/DC) was markedly low compared to that on BC1 cells pretreated with TNF-α alone (TNF/DC) (Fig. 5a). This finding is consistent with those shown in Fig. 1. The CD40 expression on TNF + NAC-treated DC was further decreased by the second culturing with NAC and became almost negative after 24 hr. In contrast, CD40 expression on TNF + NAC-treated DC was increased by the second culture with medium alone and recovered to the comparative level to that on TNF-treated DC, which showed no significant changes during the second culture with medium alone for 24 hr (Fig. 5a). In contrast, in the presence of NAC the CD40 expression on these TNF-treated DC rapidly decreased during the second culture and became almost the same level as on BC1 cells cultured with medium alone during the first and second cultures (Fig. 5a).

Figure 5.

Time–course of CD40 and CD86 expressions in the presence or absence of NAC in the second culture. BC1 cells were treated with TNF-α (40 ng/ml) in the presence or absence of NAC (20 mm) for 24 hr (first culture). The cells were then washed and cultured with or without NAC for the indicated time (second culture). CD40 (a) and CD86 (b) expressions at the indicated time-point during the second culture were analysed by flow cytometry. Each symbol represents the mean ± SE of three independent experiments.

On the other hand, similar levels of CD86 expression were observed on TNF-treated DC and TNF + NAC-treated DC at the beginning of the second culture (Fig. 5b). The CD86 expression on TNF-treated DC and TNF + NAC-treated DC similarly increased in cells cultured with medium alone. NAC showed no significant effect on CD86 expression on TNF + NAC-treated DC and TNF-treated DC during the second culture for 24 hr (Fig. 5b).

Discussion

Activation and maturational states of DC are regulated by various extracellular stimuli including cytokines and bacterial products.^11,12 These events are closely related to alterations of the morphological, phenotypic and functional properties of DC.^1–4 TNF-α, an inflammatory cytokine, is a potent promoter of the activation and maturation of DC. The TNF-α-induced mature DC in the inflammatory site migrate to the T-cell area of draining lymph nodes and activate the antigen-specific T cells. Thus, the maturational step of DC by TNF-α is an important component for the appropriate antigen presentation.

In the present study, we showed that TNF-α markedly increased the expression of both MHC and costimulatory molecules, including CD40, CD80 and CD86, on murine DC. More importantly, we could demonstrate that the reducing agents NAC and GSH completely inhibited both constitutive and TNF-α-induced expression of CD40. Neither NAC nor GSH exerted considerable influences on the expression of CD80, CD86 and MHC molecules. Thus, in the presence of the reducing agent, TNF-α generated mature-phenotype DC expressing high levels of CD80, CD86 and MHC molecules, but lacking CD40 expression.

During antigen presentation CD40–CD40L interaction induces IL-12 production, which promotes the Th1 response.^9,36 Indeed, Th1 development was abolished by inhibiting the CD40–CD40L interaction during the antigen presentation.³⁷ Thus, the present finding that the reducing agents NAC and GSH selectively and completely inhibited TNF-α-induced CD40 expression on murine DC may provide a new strategy that enables us to drive T-cell responses toward Th2 type responses. It seems that lack of CD40 expression on DC treated with NAC or GSH results in a low production of IL-12 by these DC and the malfunction in their ability to induce Th1 polarization. Indeed, we observed that IL-12 p40 production by CD40 ligation in the presence of NAC was inhibited in BC1 cells pretreated with TNF-α plus NAC compared to that in BC1 cells pretreated with TNF-α alone (data not shown). However, since NAC was present during the whole period in our culture system, it was still unclear whether this inhibitory effect of NAC was through down-regulation of CD40 or whether NAC influenced pathways other than CD40 expression, which resulted in the inhibition of IL-12 production.

Although NAC completely inhibited TNF-α-induced CD40 expression, the inhibitory effect was eradicated by simply removing the agent from the culture. On the contrary, the high-level expression of CD40 on TNF-α-treated DC was markedly decreased within 12 hr by treatment with NAC in the secondary culture. Similar results were obtained with GSH treatment (data not shown). Thus, NAC and GSH appear to regulate the level of CD40 expression on DC both reversibly and rapidly. It seems that these reducing agents can be used to control efficiently the Th1/Th2 polarization during the antigen presentation.

TNF-α activates ERK, p38mapk and SAPK/JNK in DC that may result in DC maturation. Recently, it has been reported that blocking the p38mapk pathway with a specific inhibitor abrogates TNF-α-induced CD40 expression on DC.³⁸ Thus, activation of p38mapk appears to be involved in the enhancement of the CD40 expression in DC. However, NAC, a potent inhibitor of CD40 expression, showed no considerable effects on the p38mapk activation in our study. Thus, it seems that NAC abrogates another pathway that is also involved in the induction of CD40 expression. In this context it was found that NAC reduced the total amount of CD40 protein in BC1 cells but showed no significant effect on the level of CD40 mRNA expression. These findings suggest that NAC affects a post-transcriptional pathway such as translation and/or degradation of CD40 protein.

It seems interesting and important to pursue how reducing agents modulate immunological functions in vivo. In addition, down-regulation of CD40 expression on DC induced by these agents appears to be applicable to the treatment of certain diseases. However, at present it appears to be difficult to use NAC and GSH directly in vivo for the selective regulation of CD40 expression on DC. These reducing agents affect a wide variety of physiological phenomena. Furthermore, since the inhibitory effect of NAC on CD40 expression is released by simply removing these agents from the culture, functional analysis of these NAC-pretreated DC in vivo appears not to be effective. We are currently analysing the mechanisms underlying the NAC-induced selective-regulation of CD40 expression focusing on the post-transcriptional events. We believe that elucidation of the precise molecule(s) involved in the NAC-induced selective regulation of CD40 will lead to identification of a new target(s) that may be available for clinical applications (in vivo).

It has been reported that Th1/Th2 polarization is regulated by microenvironmental conditions such as antigen concentration and/or extracellular stimuli during the antigen presentation.^30,39,40 Our current findings imply that reducing agents such as GSH in the microenvironment affect the capability of DC for Th1/Th2 polarization by selectively, rapidly and reversibly regulating CD40 expression on the DC during the antigen presentation.

Abbreviations

Ab

antibody

Ag

antigen

APC

antigen-presenting cells

ATF-2

activating transcription factor-2

CD40L

CD40 ligand

DC

dendritic cells

Elk-1

Ets-like-protein-1

ERK

extracellular signal-related kinase

FCS

fetal calf serum

GM-CSF

granulocyte–macrophage colony-stimulating factor

HPRT

hypoxanthine phosphoribosyl-transferase

IL

interleukin

IMDM

Iscove's modified Dulbecco's medium

LPS

lipopolysaccharide

MAPK

mitogen-activated protein kinases

MFI

mean fluorescence intensity

NAC

N-acetyl-l-cysteine

NF-κB

nuclear factor-κB

PE

phycoerythrin

RT-PCR

reverse transcriptase–polymerase chain reaction

SDDC

spleen-derived dendritic cells

SAPK/JNK

stress-activated protein kinase/c-Jun N-terminal kinase

Th

T helper

TNF-α

tumour necrosis factor-α