Dr Xuetao Cao, Institute of Immunology, Second Military Medical University, 800 Xiangyin Road, Shanghai 200433, China. E-mail: firstname.lastname@example.org
Toll-like receptors (TLR) are sentinel receptors capable of recognizing pathogen-associated molecule patterns (PAMP) such as lipopolysaccharide (LPS) and CpG-containing oligonucleotides (CpG ODN). TLR2 and TLR4 are major receptors for Gram-positive and Gram-negative bacterial cell wall components, respectively. TLR9 is necessary for CpG signalling. LPS or CpG ODN can activate immature dendritic cells (DC) and induce DC maturation characterized by production of cytokines, up-regulation of co-stimulatory molecules, and increased ability to activate T cells. However, little is known regarding the regulation of TLR gene expression in mouse DC. In this study, we investigated the regulation of TLR2, TLR4 and TLR9 gene expression by LPS in murine immature DC. TLR2, TLR4 and TLR9 mRNA were up-regulated following LPS stimulation. The up-regulation of TLR9 expression coincided with significantly increased production of tumour necrosis factor-α induced by LPS plus CpG ODN. While inhibition of extracellular signal-related kinase and NF-κB activation suppressed the up-regulation of the expression of TLR2, TLR4 and TLR9 mRNA, inhibition of p38 kinase prevented the up-regulation of TLR2 and TLR4 mRNA expression but enhanced the up-regulation of TLR9 expression. These results demonstrated that TLR2, TLR4 and TLR9 gene expression was differently regulated by LPS in mouse immature DC. Up-regulation of TLR2, TLR4 and TLR9 expression by LPS might promote the overall responses of DC to bacteria and help to explain the synergy between LPS and other bacterial products in the induction of cytokine production.
Dendritic cells (DC), as the most potent professional antigen-presenting cells, are critical sentinels in antimicrobial immune responses. Microbial products, such as lipopolysaccharide (LPS) and bacterial DNA, can activate immature DC and induce DC maturation, characterized by production of cytokines, up-regulation of co-stimulatory molecules and increased ability to activate T cells.1 Cells of the innate immune system use a variety of pathogen-associated molecule pattern (PAMP) recognition receptors to recognize the patterns shared between pathogens.1,2 However, the mechanisms by which the microbial products are recognized by immune cells and the signals are transmitted to induce downstream events have not been fully understood.
Recently, Toll-like receptors (TLR) were identified as major recognition receptors for PAMP such as LPS, peptidoglycan, lipoteichoic acid and CpG-containing oligonucleotides (CpG ODN).3–5 Toll, first identified as a key protein controlling dorsoventral pattern formation during the early development of Drosophila melanogaster, is a transmembrane protein whose intracellular portion is homologous to that of interleukin-1 receptor (IL-1R) family proteins.6 It has been shown that Toll participates in antimicrobial immune responses in adult Drosophila.7 Activation of Toll results in the initiation of a signalling cascade homologous to that activated by IL-1R in mammalian cells and the production of the antifungal peptide, drosomycin.7 Toll is conserved throughout various species. In addition to cytoplasmic portion homologues to IL-1R, all mammalian TLR family members share repeating leucine-rich motifs in their extracellular portions, which interact with and recognize PAMP.2,6 Up to now, 10 TLRs have been identified in the human.8 Among them, TLR2 and TLR4 have been studied intensively. Genetic analysis of hyposensitive mouse C3H/HeJ identified TLR4 as a primary receptor for LPS.4,9 On the other hand, TLR2-deficient mice failed to respond to Gram-positive cell wall components but remained responsive to LPS.10,11 However, these data do not rule out a possible role of TLR2 as a helper receptor for TLR4 in signalling to LPS in addition to its critical role in recognizing Gram-positive cell wall components.12,13 TLR2 is capable of mediating LPS responsiveness in TLR2-transfected cell lines,14 suggesting that TLR2 may recognize and mediate LPS signalling when over-expressed. Regulation of TLR2 expression, as well as of other members of the TLR family, by LPS may influence the overall responses of immune cells.
There are three members of the TLR family that have been demonstrated to be critical for PAMP signalling. Besides TLR2 and TLR4, TLR9 was shown to be necessary for CpG ODN signalling.3 Mouse immature DC are sensitive to LPS and CpG ODN stimulation. However, little is known regarding the regulation of TLR gene expression in mouse DC. In this study, we examined the regulation of the gene expression of TLR2, TLR4 and TLR9 by LPS in mouse immature DC as well as the roles of NF-κB and mitogen-activated protein kinase (MAPK) signal pathways in the regulation of TLR gene expression.
Materials and methods
Reagents and antibodies
LPS (Escherichia coli, O26:B6) was purchased from Sigma (St Louis, MO) and repurified by phenol extraction as described by Hirschfeld et al.15 Pyrrolidinecarbodithoic acid (PDTC), an inhibitor of NF-κB, PD98059, a specific inhibitor of extracellular signal-regulated kinase (ERK) (MEK1), and SB203580, a specific inhibitor of p38 kinase, were purchased from Calbiochem (San Diego, CA). Murine granulocyte–macrophage colony-stimulating factor (mGM-CSF) and mIL-4 were purchased from Sigma. RPMI-1640 and fetal bovine serum were from Hyclone (Loga, UT). Anti-phospho-ERK monoclonal antibody (mAb), anti-ERK polyclonal antibody, anti-phospho-p38 MAPK mAb, anti-NF-κB p65 subunit mAb, horseradish peroxide (HRP)-conjugated anti-goat immunoglobulin G (IgG) and HRP-conjugated anti-mouse IgG were obtained from Santa Cruz Biotechnology (Santa Cruz, CA).
Culture of mouse immature DC
Mouse immature DC were generated as described previously.16 Briefly, bone marrow suspensions from C57BL/6J mice were cultured in complete medium supplemented with 3·3 ng/ml mGM-CSF and 5 ng/ml mIL-4 in 5% CO2 at 37° after depleting red cells with ammonium chloride and removing T cells, B cells, granulocytes and Ia+ cells using a specific mAb cocktail and rabbit complement. The mAb cocktail comprised 2·43 anti-CD8, GK1.5 anti-CD4, B21-2 anti-Ia, 2.4G2 anti-FcRII, and RA3-3A1/6.1 anti-B220/CD45R (TIB210, TIB207, TIB229, HB197 and TIB146, American Type Culture Collection, Rockville, MD). On the 3rd day of culture, non-adherent cells were removed by mild pipetting, and the remaining adherent cells were cultured in the complete medium containing mGM-CSF and mIL-4 at the same concentration as indicated above. On the 7th day of culture, DC suspensions were harvested as immature DC and used in the following experiments.
Reverse transcription polymerase chain reaction (RT-PCR) analysis of TLR2, TLR4 and TLR9 expression
Total RNA was isolated from resting and LPS-treated DC with TRIzol reagent following the manufacturer's instructions. The cDNA was synthesized from 1 µg of the total RNA by extension of the oligo(dT)18 primer with 200 units of SuperScript II (Gibco BRL, Rockville, MD). PCR amplification of the cDNA was performed in a final volume of 50 µl containing 2·5 mmol/l magnesium dicholoride, 1·25 units Ex Taq polymerase (TaKaRa, Dalian, China), and 1 µmol/l specific primers. Cycling conditions were 94° for 30 seconds, 56° for 30 seconds and 72° for 45 seconds (geneAmp 9600 PCR system, Perkin-Elmer, Norwalk, CT). The optimum numbers of cycles were 32 cycles for TLR2, 28 cycles for TLR4, 32 cycles for TLR9 and 25 cycles for β-actin. The sequences of the specific primers used in this study were: TLR2 upstream primer 5′-GTC TCT GCG ACC TAG AAG TGG A; TLR2 downstream primer 5′-CGG AGG GAA TAG AGG TGA AAG A; TLR4 upstream primer 5′-AGC AGA GGA GAA AGC ATC TAT GAT GC; TLR4 downstream primer 5′-GGT TTA GGC CCC AGA GTT TTT CTC C; TLR9 upstream primer GCA CAG GAG CGG TGA AGG T; TLR9 downstream primer GCA GGG GTG CTC AGT GGA G; mouse β-actin upstream primer 5′-TGG AAT CCT GTG GCA TCC A; mouse β-actin downstream primer 5′-TAA CAG TCC GCC TAG AAG CA. All PCR products were resolved by 2% agarose gel electrophoresis and visualized by staining the gel with ethidium bromide. The RT-PCR products were purified and sequenced to confirm the identities of the DNA bands.
Northern blot analysis of TLR2, TLR4 and TLR9 expression
RT-PCR products of TLR2, TLR4 and TLR9 were used as probe templates and labelled with a random primer DNA labelling kit (TaKaRa). Then, 30 µg total RNA, isolated as described above, was loaded and analysed by Northern blot using NorthernMaxTM-Gly kit (Ambion, Austin, TX) according to the manufacturer's instructions.
Assessment of ERK and p38 kinase activation by Western blotting
Cells were lysed in 1× sodium dodecyl sulphate (SDS) sample buffer containing dithiothreitol, sonicated for 10 seconds and boiled at 95° for 5 min. Proteins were separated by SDS–polyacrylamide gel electrophoresis in Tris/glycine/SDS buffer (25 mm Tris, 250 mm glycine, 0·1% SDS), and electroblotted onto Protran nitrocellulose transfer membranes (Schleicher & Schuell Inc., Keene, NH) (100 V, 1·5 hr, 4°). After blocking for 2 hr in TBST (20 mm Tris–HCl, 150 mm NaCl, 0·1% Tween) containing 10% non-fat milk, the blots were probed for 1 hr with anti-phospho-ERK mAb or anti-phospho-p38 MAPK mAb as described by the manufacturer of the antibodies. Following washing three times in TBST, membranes were incubated with secondary HRP-conjugated anti-mouse IgG for 1 hr. Following washing for 10 min four times, proteins were visualized using SuperSignal West Femto Maximum Sensitivity Substrate as described by the manufacturer (Pierce, Rockford, IL). Membranes that were probed with phospho-specific antibodies were stripped as recommended by the manufacturer of the membranes, blocked and probed with anti-ERK polyclonal antibody as described above.
Preparation of nuclear extracts and assessment of NF-κB activation by Western blotting
Nuclear extracts were prepared with NE-PER nuclear and cytoplasmic extraction reagents (Pierce) and the protein concentration of the nuclear extracts was determined with a BCA-200 protein assay kit (Pierce) following the manufacturer's instructions. Nuclear content of NF-κB p65 subunit was determined by Western blot analysis using anti-NF-κB p65 subunit mAb as described above.
Assay of tumour necrosis factor-α (TNF-α)
Mouse immature DC were seeded at a density of 5 × 105/ml in 24-well culture plates and treated for 24 hr. TNF-α production in the supernatant was measured by enzyme-linked immunosorbent assay (ELISA) using mouse TNF-α Minikit (Endogen, Woburn, MA) according to the manufacturer's instructions.
Up-regulation of TLR2, TLR4 and TLR9 gene expression in mouse dendritic cells by LPS
Stimulation of immature DC by LPS leads to the induction of cytokine gene expression. To determine whether TLR2, TLR4 and TLR9 expressions in DC are regulated by LPS stimulation, mouse immature DC were stimulated with 10 ng/ml LPS for various time periods. Semiquantitative RT-PCR was performed to determine TLR mRNA expression by using mTLR2, mTLR4 and mTLR9 primers. TLR2, TLR4 and TLR9 gene expressions were up-regulated after LPS stimulation (Fig. 1). The increases were transient and reached peak levels 1 hr following LPS stimulation. At 5 hr after LPS stimulation, mRNA of TLR2, TLR4 and TLR9 were down-regulated to a similar or even lower level compared with that in unstimulated DC.
Synergistic induction of TNF-α by CpG ODN and LPS
CpG ODN has been reported to induce production of TNF-α by murine DC. Recently, TLR9 was shown to play an important role in bacterial DNA signalling.3 It is intriguing to speculate that the up-regulation of TLR9 expression in mouse DC by LPS may affect the responses of mouse DC to CpG-ODN. To investigate whether CpG-ODN and LPS are synergistic in the induction of TNF-α production, mouse immature DC were stimulated with CpG-ODN and LPS, respectively, or in combination. As shown in Fig. 2, LPS or CpG alone could induce production of TNF-α by DC. However, stimulation of DC with CpG plus LPS resulted in significantly enhanced production of TNF-α. Our data suggested that synergistic induction of TNF-α by LPS and CpG might be related to the up-regulation of TLR9 gene expression on DC.
Involvement of ERK and p38 MAPK pathways in the regulation of TLR2, TLR4, and TLR9 mRNA expression
LPS stimulation has been shown to activate MAPK signal pathways in DC.17 To evaluate the effect of ERK and p38 activation on TLRs gene expressions in DC, mouse immature DC were treated with a specific MEK1 inhibitor or p38 inhibitor before LPS stimulation. Pretreatment with 30 µm of PD98059, a specific inhibitor of MEK1, remarkably inhibited LPS-induced up-regulation of TLR2, TLR4 and TLR9 mRNA expressions. In contrast, pretreatment with SB203580, a specific inhibitor of p38 kinase, inhibited LPS-induced up-regulation of TLR2 and TLR4 mRNA expressions but enhanced the up-regulation of TLR9 mRNA expression (Fig. 3a). Northern blot was also performed to confirm the results of RT-PCR concerning LPS-induced regulation of TLR gene expressions and similar results were obtained as shown in Fig. 3(b). To verify whether the inhibitors were active in inhibiting the activity of the p38 and ERK pathway, immature DC were treated with PD98059 or SB203580 before LPS stimulation, and the activation of the kinases was measured as the phosphorylation of ERK and p38 kinase. As shown in Fig. 3(c), PD98059 and SB203580 were effective in inhibiting LPS-induced activation of ERK and p38 kinase, respectively, at the dose of 30 µm. These results indicate that both ERK and p38 kinase are involved in LPS-induced regulation of TLR2, TLR4 and TLR9 expression in mouse DC.
Up-regulation of TLR2, TLR4 and TLR9 mRNA expression via NF-κB pathway
LPS signal transduction has been shown to activate a variety of signal pathways, including the NF-κB pathway, which plays a critical role in gene expression regulation. To study the role of NF-κB activation in the regulation of TLR gene expression, mouse immature DC were stimulated with LPS in the presence or absence of 15 µm PDTC, an inhibitor of NF-κB. As determined by semiquantitative RT-PCR as well as Northern blot, pretreatment with PDTC suppressed LPS-induced increase of TLR2, TLR4 and TLR9 mRNA in mouse immature DC (Fig. 3a and b). To determine whether PDTC blocked LPS-induced activation of NF-κB at 15 µm, a nuclear extract was prepared from DC treated with LPS and PDTC, nuclear translocation of NF-κB p65 subunit was detected by Western blot. LPS induced nuclear translocation of NF-κB p65 subunit within 30 min. Pretreatment with 15 µm PDTC suppressed NF-κB p65 nuclear translocation induced by LPS stimulation (Fig. 3d). These results suggest that NF-κB activation plays an important role in LPS-induced up-regulation of TLR2, TLR4 and TLR9 in mouse immature DC.
TLRs are sentinel receptors capable of recognizing PAMP and initiating innate immune responses. The members of the TLR family play different roles in PAMP signalling and are not equally expressed on immune cells. Investigation of TLR gene expression helps to understand how immune cell responses to bacteria are controlled.
In this study, LPS-induced regulation of TLR2, TLR4 and TLR9 gene expression in mouse immature DC was investigated. After LPS stimulation, TLR2, TLR4 and TLR9 mRNA expression in DC increased significantly. While inhibition of ERK or NF-κB activation suppressed the up-regulation of TLR2, TLR4 and TLR9 gene expression by LPS, inhibition of p38 kinase prevented the up-regulation of TLR2 and TLR4 mRNA expression but enhanced the up-regulation of TLR9 mRNA expression.
Comparing the gene expression of TLR2 and TLR4 at base line revealed much lower expression of TLR2 in unstimulated cells. Although TLR2 is involved in the recognition of LPS purified from Porphyromonas gingivalis and Leptospira interrogans, it is generally accepted that TLR2 cannot recognize purified Gram-negative bacterial LPS.13–15 Recently, Dziarski et al. reported that over-expression of CD14, TLR2 and MD-2 enabled HEK293 cells to respond to purified Gram-negative bacterial LPS.18 But another report showed that coexpression of MD-2 could not enhance binding of radiolabelled LPS to TLR2 in HEK293 cells, arguing against involvement of the TLR2/MD-2 complex in LPS recognition.19 While these conflicting results remain to be confirmed, it should also be investigated whether MD-2 can be up-regulated by LPS and enable the up-regulated TLR2 to play a role in the recognition of Gram-negative bacterial LPS. Despite the undetermined role of TLR2 in LPS signalling, up-regulated TLR2 may enhance the responses of immune cells to bacteria by recognizing other cell wall components.
Recently, several reports showed that TLR2 mRNA, but not TLR4 mRNA, was up-regulated in LPS-stimulated murine cells, including macrophages, hepatocytes and adipocytes.12,20,21 Our observations showed that both TLR2 and TLR4 mRNA expression was up-regulated in mouse immature DC by LPS. Up-regulation of TLR4 expression by LPS was also observed in mouse cardiac myocytes, coronary microvascular endothelial cells and human monocytes.22,23 These results suggest that TLR4 expression is differentially regulated in a cell-specific fashion.
In addition to TLR2 and TLR4, our results demonstrated that TLR9 mRNA was also increased in mouse immature DC in response to LPS. TLR9 mediates the cellular response to CpG ODN. Up-regulation of TLR9 may boost responses of DC to CpG ODN. As shown in the study, up-regulation of TLR9 does coincide with increased production of TNF-α upon stimulation with CpG ODN and LPS. Synergy of LPS and bacterial DNA in inducing NO production was also observed in macrophages in previous studies,24 but the molecule mechanism remained unclear. Interestingly, LPS stimulation also up-regulates TLR9 gene expression in mouse macrophages RAW264.7 (data not shown). Our results provide a possible explanation for the synergy.
LPS has been shown to activate MAPK pathways in DC.17 The roles of ERK and p38 pathways in the regulation of TLR expression induced by LPS were investigated. Both ERK and p38 kinase were activated in DC by LPS. Pretreatment with MEK1 inhibitor suppressed the LPS-induced increase of TLR2, TLR4 and TLR9 mRNA, whereas an inhibitor of p38 kinase prevented the increase of TLR2 and TLR4 mRNA but enhanced TLR9 mRNA increase. These results suggest that TLR2, TLR4 and TLR9 gene expression is regulated by different mechanisms in mouse DC and that ERK and p38 kinase play different roles in the LPS response of DC.
In a recent study, Matsuguchi et al. showed that ERK pathway inhibitor enhanced LPS-induced up-regulation of TLR2 gene expression in mouse macrophage cell line RAW264.712. No similar phenomenon was observed in our experiments. Gene expression of TLR2 may be differently regulated in macrophages and DC. Alternatively, the different effects of the ERK pathway inhibitor may be attributed to the time for which the cells were treated in the two studies.
Inhibition of p38 kinase has been shown to prevent LPS-induced production of a variety of cytokines in DC17. As anticipated, the inhibitor of p38 kinase suppressed LPS-induced TLR2 and TLR4 mRNA increases. Unexpectedly, the TLR9 mRNA increase was enhanced upon pretreatment with p38 kinase inhibitor. It is well known that both CpG ODN and non-CpG ODN can enter cells by CpG sequence non-specific endocytosis.25 However, recent studies showed that over-expressed TLR9 could enhance vesicular uptake of CpG ODN but not control ODN despite the fact that both of them could enter cells lacking TLR9 in a sequence-independent manner.26 Immature DC are highly endocytic. During maturation induced by LPS, DC lose their highly endocytic activity, which can be measured by the uptake of FITC-dextran.17 Interestingly, the inhibitor of p38 kinase SB203580 prevents the down-regulation of FITC-dextran uptake induced by LPS.17 Whether SB203580 pretreatment affects CpG ODN endocytosis remains to be demonstrated.
NF-κB activation is essential for the expression of a variety of cytokines in the LPS response. The reactive oxygen intermediate scavenger PDTC is a potent NF-κB inhibitor. When used at a dose of 15 µm, PDTC inhibited LPS-induced NF-κB p65 subunit nuclear translocation as well as the up-regulation of TLR2, TLR4 and TLR9 mRNA, suggesting that the NF-κB pathway is important in the regulation of TLR expression in DC.
In this study, differential expression and regulation of TLRs was observed. The implication of the phenomenon is not very clear. The differentiation might be related to the differential functions between TLRs. TLRs are activated not only by different agonists but also at different concentrations of the respective agonists. In macrophages, TLR2 is fully activated by MALP-2 at about 0·3 ng/ml whereas TLR4 is fully activated by LPS at about 10 ng/ml.27 These results suggest that TLRs and their corresponding agonists may interact with different affinity. Alternatively, TLRs are saturated by agonists at different concentrations. Recently, several reports have indicated that LPS signalling differed from other TLR-mediated signalling pathways. In addition to the commonly used MyD88-dependent signalling pathways, LPS, but not CpG, could activate NF-κB and mitogen-activated protein kinases through a Toll–IL-1 receptor (TIR) domain-containing adapter protein (TIRAP).28,29 Moreover, it was reported that LPS could also be recognized by a CD14-independent receptor cluster.30 Differential expression and signal pathways of individual TLR may result in different biological responses. Fully understanding the signalling pathways of LPS as well as of other TLR agonists may help us to gain an insight into the differential expression and regulation of TLRs.
In summary, we investigated LPS-induced regulation of TLR2, TLR4 and TLR9 expression in mouse immature DC. Following LPS stimulation, TLR2, TLR4 and TLR9 mRNA expression was up-regulated. The regulation of TLR expression by LPS may influence the overall responses of immune cells to bacteria. Inhibition of ERK or NF-κB activation suppressed the up-regulation of TLR2, TLR4 and TLR9 gene expressions by LPS. In contrast, inhibition of p38 kinase prevented up-regulation of TLR2 and TLR4 mRNA expressions but enhanced the up-regulation of TLR9 mRNA expression in DC. Both MAPK and NF-κB signal pathways participate in the regulation of TLR gene expression. TLR2, TLR4 and TLR9 gene expressions are differently regulated via the p38 kinase pathway in DC.
This work was supported by grants 30028022 and 39970689 from the National Natural Science Foundation of China and by a grant from the National Key Basic Research Program of China (2001CB510002). We thank Dr W. Zhang, Dr T. Wan, Dr L. He, Dr N. Li, Dr X. Huang and Dr G. Chen for their excellent technical assistance and discussion of this work.