The dendritic cell mannose receptor mediates allergen internalization and maturation involving notch 1 signalling

Abstract

Dendritic cells (DCs) have been shown recently to play a key role in inducing and mediating T helper type 2 (Th2) responses associated with atopic disease. These responses are mediated in part by ligation to different Toll-like receptors (TLRs) and C-type lectins, e.g. the mannose receptor (MR), depending upon the DC subset involved and the respective microenvironments. Because ovalbumin (OVA) (which is structurally related to various allergens) can engage the MR, we can use OVA stimulation as a model for understanding the roles of both TLRs and the MR in allergic inflammatory responses. We examined TLR- and MR-mediated responses from mouse bone marrow-derived DCs in the context of antigen recognition and presentation in addition to examining the relationship between notch 1, TLRs and MR signalling pathways. This work demonstrated that OVA-mediated signalling up-regulated both TLR-2 and MR and that MR RNA interference (RNAi) but not TLR2 RNAi inhibited DC internalization of fluorescein isothiocyanate–OVA. Furthermore, MR RNAi inhibited OVA- and house dust mite allergen extract-induced DC maturation and MR RNAi and TLR2 RNAi influenced DC interleukin-12 production independently. Finally, we demonstrated that blocking notch 1 signalling inhibited both notch 1 and TLR-2 expression but not MR expression levels. However, MR RNAi inhibited the expression of MR, TLR-2 and notch 1. These results indicate that MR is the primary receptor mediating the internalization of environmental allergen glycoproteins. In addition, TLR-2 and notch 1 play important roles in DC maturation and antigen presentation and signals originating from the MR and TLR-2 receptors converge with the notch 1 signalling pathway.

Introduction

Dendritic cells (DCs) play a key role in the initiation of the immune response. Derived from bone marrow precursors, DCs colonize peripheral tissues including the skin, bronchial mucosa and nasal mucosa where they form a tight immune surveillance network. Increased numbers of DCs have been detected in nasal or bronchial epithelium from atopic patients. After allergen capture and processing DCs migrate into the draining lymph nodes to activate and initiate T cell responses [1–3]. Several reports have demonstrated a key role for myeloid DCs (mDC) in the induction and control of both T helper type 2 (Th2) and inflammatory responses observed in asthmatic patients and in experimental mouse allergy models [4]. Studies have indicated that airway DCs are not only crucial for regulating sensitization to inhaled antigens but also for controlling allergic inflammatory reactions. Therefore, developing therapeutics based on modulation of DC function could be used to either prevent sensitization or treat established atopic diseases.

Toll-like receptors (TLRs) are expressed by a variety of cells involved in allergic reactions, including mast cells, T lymphocytes, mononuclear phagocytes and DCs. DCs can express different TLRs, depending upon the DC subset and the nature of TLR engagement that is dependent upon exposure to allergens or to infectious agents that can modulate DC function [5]. Specifically, TLRs have been shown to play a role in the development and control of allergic reactions and some TLRs have been linked with opposing immune responses [6,7]. For example, in vitro stimulation of human monocyte-derived DCs with TLR-2 ligands failed to produce interleukin (IL)-12 p70 and interferon (IFN)-c inducible protein (IP-10) but stimulated the release of the IL-12 inhibitory p40 homodimer, producing conditions that are predicted to favour Th2 development [8]. In a mouse model of ovalbumin (OVA) sensitization, the TLR-2 synthetic ligand Pam3Cys administered at the time of sensitization increased Th2 responses and led to exacerbation of the asthmatic phenotype. These studies highlighted the complexity of TLR-induced regulation of allergic asthma. Because DCs bridge innate and adaptive immune responses, they play a significant role in the initiation and the regulation of allergic diseases. This makes it critical to comprehend how different TLRs affect allergic inflammation.

In addition to TLRs, DCs express C-type lectins which are classical Ca²⁺-dependent carbohydrate-binding molecules [9] also capable of binding OVA and similarly structured proteins. C-type lectins are divided into two groups and the mannose receptor (MR) family are part of the type I transmembrane proteins possessing multiple carbohydrate recognition domains (CRDs). This family includes the MR (CD206), DEC-205 (CD205), the phospholipase A2 receptor and Endo 180 (CD280) [10]. Der p I and Der f I are the primary allergens associated with Dermatophagoides pteronyssinus and D. farinae and possess cysteine proteinase activity and are similar structurally to the OVA glycoprotein, in that they possess a single oligosaccharide [11] that has been implicated in the recognition of soluble OVA in mice [12]. The work described in this study examines the role(s) of TLRs and the MR expressed on mouse bone marrow-derived allergen-stimulated DCs. We examined changes in notch 1 expression in the context of DC activation, as both TLRs and C-type lectin receptors are important mediators of allergic inflammation. The notch family of transcriptional regulators is highly conserved and has been shown to play a critical role in T and B lymphocyte differentiation [13]. However, in recent years a role for notch signalling has also been linked to DC development and differentiation [14,15]. Until now the receptors associated with environmental allergen interaction were unknown. Because DCs capture and internalize antigens (under the appropriate conditions) they have the potential of mediating Th2 responses. Furthermore, signalling via TLRs and C-type lectins could act co-operatively in eliciting immune responses, as described previously [16,17]. However, as the notch 1, TLR and C-type lectin relationship (in the context of DC differentiation and activation) remains unclear, signalling initiated via these respective receptors was examined. This study demonstrated that signalling via the MR mediates uptake of allergen glycoproteins and that both TLR-2 and notch 1 signalling are probably involved in DC maturation and antigen presentation, suggesting that modulation of DC activity by regulating DC expression of TLR-2 or MR may provide a novel strategy for treating allergic inflammatory responses.

Materials and methods

Mice

Male BALB/C mice, 6–7 weeks of age were purchased from the Guangdong Medical Experimental Animal Center of China.

Reagents

All cell lines were grown in complete medium consisting of RPMI -1640, 10% fetal calf serum (Hyclone, Loan, UT, USA), 100 U/ml penicillin (Invitrogen, Carlsbad, CA, USA) and 100 µg/ml streptomycin (Invitrogen) during the first 4 days of cell culture. Red blood cells (RBCs) were lysed using RBC lysis buffer (16 mM Tris, 139 mM NH₄CI, pH 7·2). Recombinant granulocyte–macrophage colony-stimulating factor (GM-CSF) and IL-4 were purchased from PeproTech (Rock Hill, NJ, USA).

Fluorescence-conjugated fluorescein isothoicyanate (FITC)-CD11c, phycoerythrin (PE)-CD80, FITC-CD86 and FITC-major histocompatibility complex (MHC)-II (BD Biosciences, Franklin Lakes, NJ, USA) were used for flow cytometric analyses. Rabbit anti-mouse TLR-2 (Cell Signaling Technology, Danvers, MA, USA), rat anti-mouse MR (R&D Systems, Minneapolis, MN, USA) antibodies and rabbit anti-notch 1 monoclonal antibody (Abcam, Cambridge, MA, USA) were used for Western blot analysis.

Notch 1 activity was inhibited by the γ-secretase inhibitor IX DAPT (Merck Biosciences, Whitehouse Station, NJ, USA).

Reverse transcriptase–polymerase chain reaction (RT–PCR) primer sequences corresponding to TLR-1–9, notch 1 and MR are as follows: TLR-1 forward primer, 5′-GGA CTT CCA CAT GTC TCC ACT ATC C-3′ and reverse primer, 5′-TCC ATG CTT GTT CTT CTC TGT GG-3′; TLR-2 forward primer, 5′-GTG GTA CCT GAG AAT GAT GTG GG-3′ and reverse primer, 5′-GTT AAG GAA GTC AGG AAC TGG GTG-3′; TLR-3 forward primer, 5′-AGG TAC CTG AGT TTG AAG CGA GC-3′ and reverse primer, 5′-GAG CAT CAG TCT TTG AAG GCT GG-3′; TLR-4 forward primer, 5′-CTG GGT GAG AAA TGA GCT GG-3′ and reverse primer, 5′-GAT ACA ATT CCA CCT GCT GCC-3′; TLR-5 forward primer, 5′-TAT CTC CCT GTT CTT CAG ACG GC-3′ and reverse primer, 5′-TGG TTG CCA GAT AGG TCT AAG CG-3′; TLR-6 forward primer, 5′-TTA ACT GAC CTT CCT GGG TGT GG-3′ and reverse primer, 5′-GCA GAA CAG TAT CAC AGG ACA GTG G-3′; TLR-7 forward primer, 5′-CAA ACT TCT GTA GAC CGT CAT GGG-3′ and reverse primer, 5′-AAG TAC CGC AAC TCT CTC AAC GG-3′; TLR-8 forward primer, 5′-GTT ATG TTG GCT GCT CTG GTT CAC-3′ and reverse primer, 5′-TCA CTC TCT TCA AGG TGG TAG C-3′; TLR-9 forward primer, 5′-GAC TTA CTG TTG GAG GTG CAG ACC-3 and reverse primer, 5′-GAA CAC CAC GAA GGC ATC ATA GG-3′; MR forward primer, 5′-TGC ATT GGT TTG TCC TTT TTC-3′ and reverse primer, 5′-GCA GGG TTG ACA TGA GAC CT-3′; β-actin forward primer, 5′-TGG AAT CCT GTG GCA TCC ATG AAA C-3′ and reverse primer, 5′-TAA AAC GCA GCT CAG TAA CAG TCC G-3′; glyceraldehyde-3-phosphate-dehydrogenase (GAPDH) forward primer, 5′-ACC ACA GTC CAT GCC ATC AC-3′ and reverse primer, 5′-TCC ACC ACC CTG TTG CTG TA-3′; notch 1 forward primer, 5′-CAG CTT GCA CAA CCA GAC AGA C-3′ and notch 1 reverse primer, 5′-ACG GAG TAC GGC CCA TGT T-3′.

OVA (grade V) was purchased from Sigma (St Louis, MO, USA) and FITC-conjugated OVA (FITC–OVA) was purchased from Invitrogen. Allergen extract from house dust mites (HDM) was purchased from Allergopharma (Reinbek, Germany). RNA was isolated using Trizol Reagents (Invitrogen) prepared for RT–PCR. Promega RT Kit was used for RT reactions (Promega, Madison, WI, USA). PCR reactions were performed using a Promega GoTaq Green Master Mix containing Taq DNA polymerase, dNTPs, MgCL₂, reaction buffer and blue and yellow dyes. Product sizes were estimated using the DL2000 molecular size marker (BioRule, Guangzhou, China).

The following RNA interference oligonucleotides (RNAi oligos) were synthesized by the Shanghai Genepharma Co. Ltd: (1) TLR-2-255, sense 5′-CAG CAG AAU CAA UAC AAU ATT-3′ and anti-sense 5′-UAU UGU AUU GAU UCU GCU GTT-3; (2) TLR-2-490, sense 5′-GAG AUA AGG AGA AUA GAU UTT-3′ and anti-sense 5′-AAU CUA UUC UCC UUA UCU CTT-3′; (3) GAPDH-positive control, sense 5′-GUA UGA CAA CAG CCU CAA GTT-3′ and anti-sense 5′-CUU GAG GCU GUU GUC AUA CTT-3′; (4) negative control FAM (NC-FAM), sense 5′-UUC UCC GAA CGU GUC ACG UTT-3′ and anti-sense 5′-ACG UGA CAC GUU CGG AGA ATT-3′; (5) negative control, sense 5′-UUC UCC GAA CGU GUC ACG UTT-3′ and anti-sense 5′-ACG UGA CAC GUU CGG AGA ATT-3′; (6) β-actin-positive control (internal parameter), sense 5′-GUA UGA CAA CAG CCU CAA GTT-3′ and anti-sense 5′-CUU GAG GCU GUU GUC AUA CTT-3′; (7) MR-1386, sense 5′-GCG UGG UUA UGA AAG GCA ATT-3′ and anti-sense 5′-UUG CCU UUC AUA ACC ACG CAG T-3′; (8) MR-1968, sense 5′-GCA UGU GUU UCA AAC UGU ATT-3′ and anti-sense 5′-UAC AGU UUG AAA CAC AUG CTG-3′. RNAi oligos were transfected into dendritic cells using Lipofectamin 2000 (Invitrogen). Optimem I (Invitrogen) culture medium was used when transfections were complete.

Dendritic cell culture

Femurs and tibias were harvested from mice and sterilized in 70% ethanol for 1 min. Bone marrow cells were flushed with complete medium using a 0·25 mm needle. Marrow suspensions were pipetted to disintegrate cell clusters in six-well culture dishes and RBCs lysed by incubating in RBC lysing buffer at room temperature for 5 min. The remaining cells were then centrifuged at 1500 g and washed once in complete medium. Bone marrow cells were seeded in 2·5 ml complete media containing GM-CSF (10 ng/ml) and IL-4 (10 ng/ml) at day 0 in six-well plates at a concentration of 2·5 × 10⁶/well. After 48 h, the medium was replaced with fresh complete medium containing GM-CSF and IL-4 at a concentration of 10 ng/ml. At days 4 and 6, half the medium was changed gently and both cytokines were added to maintain a concentration of 10 ng/ml. At day 4, large DC suspensions were visible and some clusters were attached to adherent cells (macrophages and fibroblasts). At day 6, more clusters were visible and the weakly attached clusters became dislodged from the adherent cells. At day 7 numerous typical DCs were observed floating, and after 7 days about 5 × 10⁶ cells were recoverable/well. The cell surface of bone-marrow-derived DCs was characterized by flow cytometry by staining for CD11c, CD80, CD86 and MHC-II. DCs were stimulated with either 0, 10, 100 or 1000 µg/ml OVA. CD11c expression remained constant (75%). The expression levels of CD80, CD86 and MHC-II increased in a dose-dependent manner, suggesting that higher OVA concentrations accelerated DC maturation (data not shown).

Ovalbumin and HDM stimulation

20 mg OVA was added to 2 ml RPMI-1640 culture medium to a final concentration of 10 mg/ml. At day 7, OVA at different concentrations (0, 10, 100 or 1000 µg/ml) or allergen extract from HDM at different concentrations (0, 0·1, 0·2 or 1 µg/ml) was added into the culture medium. Endotoxin-like lipopolysaccharide (LPS) was removed from the OVA solution using a protocol adapted from others [18]. After 24 or 48 h DC suspensions were collected for flow cytometry, total RNA extraction or total protein extraction.

Flow cytometry

DCs were washed twice in RPMI-1640 medium and incubated with FITC-CD11C, FITC-CD86, FITC-MHC-II or PE-CD8 for 30 min and then analysed by flow cytometry. FITC–OVA (20 µg/ml) was incubated with DCs and internalization measured by flow cytometry after 2 h.

RT-PCR

Total RNA was isolated and RT-PCR was performed using a Bio-Rad (Hercules, CA, USA) PCR gene amplification apparatus using the Promega Kit protocol. The RT reaction was carried out in 20 µl under the following conditions: 42°C for 15 min, 95°C for 5 min and 4°C for 5 min. cDNA was prepared for PCR reactions or stored at −20°C until use. PCR reactions were carried out as follows: denaturing at 93°C for 3 min followed by 30 cycles of 93°C for 30 s, 55°C for 30 s, 72°C for 40 s with a final 10 min 72°C step. PCR products were stored at 4°C until further analysis. The Promega GoTaq Green Master Mixture was loaded onto 1·5% agarose gel containing 0·5 µg/ml ethidium bromide. Electrophoresis was carried out at 100 V for 30 min and gels examined by visualization under ultraviolet light.

Western blot analysis

Dendritic cells were lysed with 1× sodium dodecyl sulphate (SDS) sample buffer (62·5 mM Tris-HCl pH 6·8, 2% SDS, 10% glycerol, 50 mM dithiothreitol), sonicated for 15 s, and then heated at 100°C for 5 min. Cell lysates were centrifuged at 15 350 g (4°C) for 5 min and the protein concentration of the supernatants determined using the Micro BCA Protein Assay System (Pierce, Rockford, IL, USA). Western blotting of mouse TLR-2 and mouse MR was performed according to the manufacturer's instructions using the respective antibodies. Briefly, proteins were electrophoresed on 10% SDS-polyacrylamide gel electrophoresis (PAGE) precast gels (Invitrogen) under reducing conditions and transferred onto an ImmunoBlot polyvinylidene membrane (Bio-Rad). The membranes were blocked with 5% non-fat milk and then incubated with primary antibodies overnight at 4°C. After incubation with a horseradish peroxidase-conjugated secondary antibody, the protein bands were detected using a Super Signal Chemiluminescent Substrate (Pierce) and BIOMAX-MR film (Eastman Kodak, Rochester, NY, USA).

RNAi transfection

Transfection was conducted according to the manufacturer's protocol. Briefly, 80 pmol RNAi was mixed in 200 µl of Opti-MEM I (Invitrogen) and 3 µl of Lipofectamine 2000 (Invitrogen) was incubated with 150 µl of Opti-MEM I at room temperature for 5 min. The diluted RNAi and Lipofectamine 2000 were then incubated for an additional 20 min at room temperature for complex formation; 350 µl media was aspirated from each well and the complex added to each well. The final RNAi concentration was 32 nmol/l. Ten per cent fetal calf serum (FCS) was then added to OptiMEM I culture medium after 6 h of transfection. DCs were incubated in a 5% humidified CO₂ incubator at 37°C. DC were transfected with FITC-labelled, scrambled RNAi (Shanghai Gene-Chem Co. Ltd, Shanghai, China) for 36 h. Then the transfection efficiency was assessed using a fluorescence microscope (Olympus, Tokyo, Japan) and gene expression was validated using RT–PCR (24 h after transfection) and Western blot analyses (48 h after transfection).

DC FITC–OVA internalization efficiency

TLR-2 RNAi and MR RNAi were used to inhibit TLR-2 and MR expression in DCs. RNAi transfection was performed at DC culture day 6. After 48 h of RNAi transfection, FITC–OVA (20 µg/ml) was added to the culture media for 1 h followed by flow cytometric analysis of the DCs for detection of internalized FITC–OVA. FITC–OVA internalization was also observed using fluorescence microscopy.

Detection of IL-12 production

Mouse IL-12 present in culture supernatants was measured using enzyme-linked immunosorbent assay (ELISA) kits (Wuhan Boster Biological Technology, Malden, MA, USA), according to the manufacturer's instructions. IL-12 secreted by DC was measured after 24 h of OVA stimulation at a concentration of 100 µg/ml. Twenty-four hours post-RNAi transfection, OVA was added to the culture medium and cytokine supernatants were measured 24 h later to assess the effects of RNAi transfection.

Statistical analysis

All experiments were performed at least three times and shown representatively. The data are expressed as the mean ± standard deviation (s.d.) of three observations. Statistical comparisons between groups were performed using a one-way analysis of variance (anova) test. Differences between groups were considered significant at P< 0·05.

Results

Allergen-mediated up-regulation of DC TLRs and the mannose receptor

The expression pattern of different TLRs has been shown to vary depending upon the stimulation conditions. In our studies, TLR mRNAs (except TLR-1) were detected in all four OVA treatment groups. There were no significant changes in TLR-3, TLR-4, TLR-5, TLR-6 and TLR-9 expression after OVA stimulation. However, TLR-2 expression increased in the presence of increasing OVA concentrations (Fig. 1a and c) and following stimulation with 10 µg OVA the TLR-7 and TLR-8 expression levels were higher than in other treatment groups (Fig. 1a and c). β-actin mRNA expression levels were similar in the four treatment groups.

Fig. 1.

Assessment of Toll-like receptor (TLR) and mannose receptor (MR) expression. TLR-1–9 and MR expression was assessed by reverse transcription–polymerase chain reaction (a,c,d) and Western blot analysis (b,e) following stimulation with different ovalbumin (OVA) concentrations (lane 1, 0 µg/ml OVA; lane 2, 10 µg/ml OVA; lane 3, 100 µg/ml OVA and lane 4, 1000 µg/ml OVA). TLR-2 and MR expression was assessed by Western blot analysis (f,g) following stimulation with allergen extract from house dust mites (HDM) at different concentrations (lane 1, 0 µg/ml HDM; lane 2, 0·1 µg/ml HDM; lane 3, 0·2 µg/ml HDM and lane 4, 1 µg/ml HDM). The data are expressed as the mean ± standard deviation of six observations. **P< 0·01 using analysis of variance.

Expression of TLR-2 was also detected by Western blot analysis and paralleled TLR-2 mRNA expression in DCs (Fig. 1b). The dose-dependent increase in MR protein expression also paralleled the MR mRNA expression profile (Fig. 1d and e).

Expressions of TLR-2 and MR stimulated by allergen extract from HDM were also detected by Western blot analysis and paralleled expressions of TLR-2 and MR stimulated by OVA (Fig. 1f and g).

Effect of TLR-2 and MR RNAi on DC FITC–OVA uptake and IL-12 production

The transfection efficiency for both TLR-2 and MR RNAi was 87·5% determined using fluorescence microscopy and the functional efficacy of TLR-2 and MR RNAi was assessed by RT–PCR and Western blot. Six treatment groups were evaluated, including the untransfected control (no RNAi sequences added), blank (transfection reagents were added minus RNAi sequences), negative control (NC, transfection reagent plus negative sequences), GAPDH (GAPDH was used as a positive control), TLR-2-255 (TLR-2 255 sequence), TLR-2-490 (TLR-2 490 sequence), MR-1386 (MR 1386 sequence), MR-1968 (MR 1968 sequence) and the β-actin internal control (Fig. 2). GAPDH RNAi knocked down GAPDH expression but not expression of TLR-2 (Fig. 2a and b, middle image, P< 0·0001). RT–PCR analysis revealed that the TLR-2-490 and MR-1386 sequences (but not the TLR-2-255 or MR-1968) sequences inhibited TLR-2 and MR expression significantly in DCs (Fig. 2a and b, lower image, P= 0·001 and P< 0·0001, respectively). TLR2 and MR RNAi was assessed by Western blot (Fig. 2c and d). TLR-2-490 and MR-1386 sequences (but not the TLR-2-255 or MR-1968) sequences inhibited TLR-2 and MR expression significantly in DCs (Fig. 2c and d, lower image, P< 0·001 and P< 0·0001, respectively).

Fig. 2.

Effect of RNA interference (RNAi) on Toll-like receptor (TLR)-2 and mannose receptor (MR) mRNA /protein levels. Inhibition of TLR-2 and MR mRNA /protein levels was assessed by reverse transcription–polymerase chain reaction and Western blot [lane 1, untransfected cells; lane 2, transfection reagents only; lane 3, negative control consisting of transfection reagents and non-specific RNAi sequences; lane 4, glyceraldehyde-3-phosphate-dehydrogenase (GAPDH) RNAi transfection; lane 5, TLR2-255 transfection (a,c) or MR-1386 transfection (b) or MR-1968 transfection (d); lane 6, TLR2-490 transfection (a,c) or MR-1968 transfection (b) or MR-1386 transfection (d)]. β-actin expression was assessed as an internal mRNA or protein expression control. The data are expressed as the mean ± standard deviation β-actin/GAPDH and β-actin/TLR-2–MR ratios from six observations. **P< 0·01 using analysis of variance.

Flow cytometric and fluorescence microscopic analyses demonstrated that TLR-2 RNAi did not inhibit DC FITC–OVA internalization but MR RNAi did. Assessment of FITC–OVA internalization using flow cytometry was carried out on DCs that were either untransfected (control group), treated with transfection reagents only (blank group) or transfected with TLR-2 or MR RNAi (RNAi group). No differences in FITC–OVA internalization between the three groups in the TLR-2 RNAi experiment was observed (control, 38·57% ± 0·40%; blank 38·83% ± 0·55% and TLR-2 RNAi, 39·03 ± 0·61%) (data not shown). The percentage of FITC–OVA internalization between groups in the MR RNAi experiment, however, were 46·47% ± 2·96% (control), 45·03% ± 3·79% (blank) and 20·25 ± 1·89% (MR RNAi) (Fig. 3a). Statistical differences between the MR RNAi group and the control and blank groups were observed (P< 0·0001). No statistical differences between the control and blank group were observed.

Fig. 3.

The effect of mannose receptor (MR) RNA interference (RNAi) on fluorescein isothiocyanate–ovalbumin (FITC–OVA) internalization and Toll-like receptor (TLR)-2 or MR RNAi on interleukin (IL)-12 production. Cultured dendritic cells (DCs) were either left untreated (control) or incubated in the presence of transfection reagents only (blank) or transfected with MR RNAi prior to incubation with FITC–OVA (a). FITC–OVA internalization assessed by flow cytometry in the respective groups was 49·3%, 42·0% and 20·2%. Data are representative of six independent experiments. Green FITC–OVA staining was observed under fluorescence microscopy. OVA–FITC internalization in untreated DCs (b) and MR RNAi transfected cells (c). Magnification 200 ×. Cultured DCs were incubated in media alone (no-OVA), cultured with OVA only (control) and cultured with OVA and transfection reagents (blank) or treated with OVA and transfected with either TLR-2 RNAi (d) or MR RNAi (e). After 48 h, supernatants were harvested and examined by enzyme-linked immunosorbent assay for IL-12 production. The data are expressed as the mean pg/ml of IL-12 ± standard deviation of six observations per treatment group. **P< 0·01 using analysis of variance.

Flow cytometry results were confirmed using fluorescence microscopy. Three visual fields were chosen randomly, examined at 200× and mean green fluorescence in all fields was measured using Image Pro Plus 5·0. No differences in OVA–FITC internalization were observed using this analysis; however, significant differences between the MR RNAi group were observed (P= 0·001), confirming the flow cytometry data and demonstrating further that MR RNAi could inhibit DC uptake of FITC–OVA (Fig. 3b and c).

The effect of TLR-2 and MR RNAi on DC IL-12 production was also examined. DCs were either left untreated (no-OVA), treated with OVA only (control), treated with OVA plus transfection reagents (blank group) or treated with OVA and transfected with either TLR-2 RNAi (TLR-2 RNAi group) or MR RNAi (MR RNAi group). IL-12 production was measured by ELISA. Significant differences were observed between the control and blank group DCs and the no-OVA, TLR-2 RNAi (Fig. 3d) and MR RNAi (Fig. 3e) DC groups (P< 0·0001). A significant difference in IL-12 production between the TLR-2 and MR RNAi group DCs and DCs in the no-OVA group were also observed (P= 0·002 and P= 0·008, respectively) (Fig. 3d and e).

Effects of TLR-2 and MR RNAi on DC maturation

In our previous experiments, OVA stimulation up-regulated the expression levels of CD80, CD86 and MHC-II; however, CD11c expression remained constant (75%). DCs were stimulated with either 0, 10, 100 or 1000 µg/ml OVA and the expression levels for respective markers determined by flow cytometry (Fig. 4). The expression levels of CD80, CD86 and MHC-II increased significantly (P ≤ 0·05) in a dose-dependent manner, suggesting that higher OVA concentrations accelerated DC maturation. So we used CD86 as the marker of OVA-induced DC maturation. Because MR RNAi inhibited FITC–OVA uptake, we next examined the effect of MR RNAi on DC maturation by examining CD86 expression by flow cytometry following incubation with 10 µg/ml OVA. DCs were divided into three groups consisting of DCs incubated with OVA only (control), DCs incubated with OVA solution and transfection reagents (blank) or DCs incubated with OVA and transfected with MR RNAi. CD86⁺ cells were distributed as follows: control group (57·28 ± 6·74%), blank (54·75 ± 5·22%) and the MR RNAi group (27·50 ± 0·82%) (Fig. 4a). CD86 expression was significantly different between the MR RNAi group and the control group (P< 0·0001). There were also significant differences between the MR RNAi group and blank group (P< 0·0001), suggesting that MR RNAi inhibited DC maturation. No statistical differences were observed between the blank and control groups.

Fig. 4.

Effect of mannose receptor (MR) RNA interference (RNAi) on dendritic cell (DC) maturation. Cultured DCs were transfected with MR RNAi and incubated with 10 µg/ml ovalbumin (OVA) (a) or transfected with MR RNAi and incubated with either 0·1 or 1·0 µg/ml house dust mites (HDM) (b) and CD86 expression was analysed by flow cytometry. Data are representative of six independent experiments.

As we observed that allergen extract from HDM induced DC maturation based upon CD86 expression levels (Fig. 4b), we examined the effects of MR RNAi on allergen-mediated DC maturation. This analysis was carried out by examining untreated DCs (control) or DCs incubated with either 0·1 or 1·0 µg HDM or DCs treated with 0·1 or 1·0 µg HDM and transfected with MR RNAi. Significant differences in CD86 expression were observed between MR RNAi-treated DCs and untreated DCs stimulated with either 0·1 or 1·0 µg HDM (P< 0·0001, Fig. 4b).

MR influenced TLR-2 expression via the notch 1 signalling pathway

To define the pathways associated with MR signalling, DC notch 1 mRNA expression levels were assessed. Stimulation of DCs with 1000 µg OVA significantly up-regulated notch 1 mRNA levels (Fig. 5a). As both MR and TLR-2 were up-regulated following exposure to OVA, we examined the relationship between MR, TLR-2 and notch 1 by Western blot analysis. DCs were either stimulated with 1000 µg/ml OVA or pretreated with either a notch 1 blocker or MR RNAi prior to incubation with 1000 µg/ml OVA. The notch 1 blocker decreased both notch 1 and TLR-2 expression levels but not MR levels. However, MR, TLR-2 and notch 1 were all inhibited by MR RNAi. These results indicated that MR, TLR-2 and notch 1 all became up-regulated during DC maturation. MR RNAi influenced the DC maturation signalling pathway by inhibiting allergen internalization as both the notch 1 and TLR-2 signalling pathways were both activated following MR-mediated allergen internalization, suggesting that the notch 1 signalling pathway could bridge the MR and TLR-2 signalling pathways (Fig. 5b).

Fig. 5.

Analysis of notch 1 expression by dendritic cells (DCs). The effects of ovalbumin (OVA) on notch 1 expression were assessed by reverse transcription–polymerase chain reaction following incubation with OVA (a: lane 1, 0 µg/ml OVA; lane 2, 1000 µg/ml OVA). Notch 1 expression is shown as the mean ± standard deviation (s.d.) of glyceraldehyde-3-phosphate-dehydrogenase (GAPDH)/notch 1 ratio. DCs were also incubated in the presence of OVA only (lane 1), notch 1 blocker prior to OVA incubation (lane 2) or mannose receptor (MR) RNA interference (RNAi) prior to OVA stimulation (lane 3) (b). The OVA concentration tested was 1000 µg/ml. The notch 1 blocker decreased both notch 1 and Toll-like receptor (TLR)-2 expression levels but not MR levels. However MR, TLR-2 and notch 1 were all inhibited by MR RNAi (b). MR, notch 1 and TLR-2 expression are shown as the mean ± s.d. of MR-notch 1-TLR-2/GAPDH ratios from six observations. *P< 0·05, **P< 0·01 using analysis of variance.

Discussion

This work has demonstrated effects of OVA or mite allergen stimulation on DC maturation in vitro, suggesting that the dose and nature of respective antigens would probably affect the nature of the ensuing T cell response [19]. The DC maturation process involved changes in cell surface marker and receptor expression profiles, including TLRs and C-type lectin receptors. Using the model established in this report, we demonstrated that the expression levels of CD80, CD86 and MHC-II increased post-OVA stimulation. These observations suggest that activation of DCs by glycoproteins, including OVA and various allergens, could result in allergic inflammation.

In this study, we screened for the expression of TLR-1–9 on mouse bone marrow-derived DCs and observed a dose-dependent increase in TLR-2 examined following OVA stimulation; however, the expression levels of TLR-7 and TLR-8 following stimulation with 10 µg/ml OVA were significantly higher than the expression levels of the other TLRs examined. Because OVA stimulation resulted in TLR-2 up-regulation on DCs, this suggested that TLR-2 might play an important role in DC maturation and antigen presentation in addition to allergic inflammation [20]. Activation of TLR-2-related signalling pathways results in the secretion of different cytokines and in the expression of different cell surface molecules (i.e. TLRs) associated with DC function; however, changes in TLR expression can also be affected by exogenous antigens and cytokines in a dose-dependent manner [21,22], e.g. IL-4, IL-1β, IL-15, tumour necrosis factor (TNF)-α and GM-CSF have been shown to up- or down-regulate TLR expression levels [23]. Our data suggest that TLR-2 was the primary contributor to DC maturation. Furthermore, the interaction between TLR-2 and TLR-7 (or TLR-8) could represent a novel mechanism associated with the regulation of allergic immunoreactivity [24].

Most studies have addressed the MR function using macrophages [25], which use this receptor to recognize and take up mannosylated structures [26]. In addition to MR expression by phagocytic cells, the MR can also be expressed by liver endothelial cells, dermal microvascular endothelial cells, monocytes and Langerhans cells [27]. An in vivo study demonstrated that DCs recognized OVA via interactions with the MR [12], but no reports have demonstrated that MR could mediate OVA endocytosis in vitro. As OVA shares structural components (based on its single carbohydrate modification) [28,29] with major mite allergens and pollen [30,31], it has been suggested that the MR could be involved in Der p I uptake [32].

We have demonstrated previously that TLR-2 could play an important role in DC maturation and antigen presentation. Batzer et al. demonstrated that house dust extracts could induce DC maturation dependent upon TLR-2, TLR-4 and TLR-9 signalling [22]. Others also demonstrated that TLR-2 was related to allergic inflammation and that TLR-2 affected DC maturation and antigen presentation but not glycoprotein internalization [7,33].

In this study, we have demonstrated that bone marrow-derived DCs up-regulated MR expression following stimulation with OVA or allergen extract from HDM, suggesting that MR may play a critical role in DC maturation. These observations were substantiated further by demonstrating that inhibition of TLR-2 or MR by specific RNAi sequences affected DC maturation and antigen internalization properties negatively. MR RNAi effectively inhibited FITC–OVA internalization and CD86 up-regulation induced following OVA or mite allergen stimulation. In contrast to the effects of MR RNAi on maturation and CD86 expression, MR RNAi up-regulated DC production of IL-12. Based on the MR ligand specificity, combined with the data presented in this report, we hypothesized that the MR could play a key role in DC-mediated antigen recognition and internalization of environmental allergens.

Inhibition of DC TLR-2 expression using TLR-2 RNAi did not inhibit the internalization of FITC–OVA, in contrast to the effects observed following treatment with MR RNAi; however, TLR-2 RNAi up-regulated IL-12 production similarly to MR RNAi, suggesting that TLR-2 did not participate directly in DC antigen internalization but that TLR-2 could also influence DC function.

Several receptor–ligand pairs and cytokines have been shown to be involved in T cell priming following co-localization to the lymph nodes [34]. However, the role of DCs in Th2 development remains undefined [35,36]. For example, inhibition of IL-12, the primary cytokine associated with Th1 differentiation, could accelerate Th2 response development [37,38]; however, our data demonstrated that OVA-stimulated DCs decreased IL-12 production following OVA stimulation that could be reversed following inhibition of TLR-2 or MR expression using respective RNAi sequences. However, the mechanism behind the respective effects conferred by TLR-2 and MR RNAi sequences, respectively, could be different, as OVA internalization by DCs was inhibited by MR RNAi but not TLR-2 RNAi. TLR-2 may play a role in DC maturation, as inhibition of TLR-2 expression by TLR-2 RNAi impaired DC maturation and antigen presentation. The DC cytokine expression profile (including IL-12) could be impaired following inhibition of OVA-mediated signal transduction, suggesting that cytokine regulation using RNAi methodology could be used to treat allergic inflammation.

Notch signals influence multiple processes that govern normal morphogenesis initiated following binding of the notch extracellular domain to notch ligand and plays a controversial role in DC differentiation [39,40]. The notch and nuclear factor (NF)-κB pathways are evolutionarily conserved pathways that determine the fate of different cell types, and notch signalling may exert effects on DC development via NF-κB. As NF-κB is a major transcription factor involved in DC-TLR signalling it may affect both DC differentiation and function. If allergen-mediated stimulation resulted in notch 1 signalling in DCs it would suggest that notch 1 probably participates in DC activation and maturation. Based on the above data, MR and TLR-2 could affect DC activation and maturation. This was assessed by inhibiting MR or notch 1 signalling prior to allergen stimulation to observe the expression levels of these receptors. This analysis determined that notch 1 and TLR-2 expression were both inhibited by respective RNAi sequences, but only TLR-2 expression was inhibited following notch 1 down-regulation. Finally, the notch 1 signalling pathway is a probable cross-roads for signals originating following ligation of either MR or TLR-2.

Disclosure

None.