Dr K. Onozaki, Department of Molecular Health Sciences, Graduate School of Pharmaceutical Sciences, Nagoya City University, Mizuho, Nagoya 467–8603, Japan. E-mail: address: firstname.lastname@example.org
We have previously reported that the expressions of TLR2 and TLR4 mRNA are differentially regulated in mouse liver and in the parenchymal cells. In the present study, we investigated the mechanism of the up-regulatory effects of interleukin-1α (IL-1α), tumour necrosis factor-α (TNF-α), lipopolysaccharide (LPS), or bacterial lipoprotein (BLP) on TLR2 mRNA expression in primary cultured murine hepatocytes. Although TLR2 mRNA stability was not affected, these treatments enhanced NF-κB activity and TLR2 gene transcription simultaneously. The up-regulation of TLR2 transcription in response to these reagents was completely inhibited by blocking the NF-κB activation pathway, demonstrating a pivotal role of NF-κB activation in the regulation of hepatocyte TLR2 transcription. The expression of TLR2 protein by hepatocytes was also remarkably up-regulated by IL-1α and, to a lesser extent, by TNF-α as well, but not by LPS or BLP. In addition, pretreatment of mice with IL-1α markedly increased the BLP (a ligand for TLR2)-induced serum level of serum amyloid A (SAA), an acute-phase protein predominantly produced by hepatocytes, indicating that IL-1α may also up-regulate functional TLR2 in vivo. These results demonstrate that IL-1α, through activating the TRAF6-NF-κB pathway, serves as the most potent inducer for TLR2 up-regulation, and plays an important role in the regulation of hepatocyte functions by augmenting the hepatocyte response to bacteria or bacterial products.
Toll is the type I transmembrane protein that controls the dorsoventral pattern formation during embryogenesis of Drosophila melanogaster and the antifungal immune response in the adult fly.1 Recently, mammalian homologues of Toll (TLR) have been identified and 10 members have been cloned to date. TLRs, interleukin-1 receptor (IL-1R), and IL-18R share a common signal transduction pathway through their Toll/IL-1R homology region (TIR) domains. In mammals, stimulation of TLR or IL-1R leads to the sequential activation of the adapter protein myeloid differentiation factor 88 (MyD88), the IL-1 receptor-associated kinases (IRAKs), TRAF6, and eventually, the IκB kinase complex (IKKα, -β and -γ). The NF-κB/Rel family of transcription factors is maintained in the cytoplasm as inactive complexes with inhibitory proteins, called IκBs. The IΚΚ complex phosphorylates the IκBs, targeting them for ubiquitination and degradation by the proteasome. Degradation of IκB liberates NF-κB/Rel dimmer which translocates to the nucleus and augment the expression of NF-κB-responsive genes such as defence-related and anti-apoptotic genes.1–3
TLR2 has been shown to function as a pattern recognition receptor for diverse bacteria and their products, including the mycobacterial arabinose-capped lipoarabinomannan (ara-LAM), mannosylated phosphatidylinositol, peptidogycan, lipopolysaccharide (LPS) from Leptospira interrogans and Porphyromonas gingivalis, and lipoproteins from various sources.1 Bacterial lipoproteins (BLPs), which are expressed by all bacteria, are potent activators of TLR2.4–6
We have studied the regulation of TLR2 and TLR4 gene expression in murine tissues, especially in the liver, and found that TLR2, but not TLR4, mRNA was up-regulated by pro-inflammatory cytokines [IL-1α or tumour necrosis factor-α (TNF-α)] or TLR4 agonist (LPS) but not by IL-6.7 In the present study, we showed that IL-1α, as well as TNF-α, LPS and BLP, one of the TLR2-ligands, up-regulate TLR2 mRNA expression in primary mouse hepatocytes, and this up-regulation is regulated at the transcriptional level through the TRAF6-NF-κB pathway. We also analysed the physiological role of the TLR2 up-regulation in vivo.
Materials and methods
Human recombinant IL-1α (2·0 × 107 U/mg) and TNF-α (1·9 × 106 U/mg) were kindly provided by Dainippon Pharmaceutical Co. Ltd. (Osaka, Japan). Escherichia coli LPS (E. coli 026: B6) was purchased from Difco Laboratories (Detroit, MI). Actinomycin D and polymyxin B were purchased from Sigma Chemical Co. (St Louis, MO). Synthetic bacterial lipopeptide Pam3-Cys-Ala-Gly-OH (BLP), corresponding to the N-terminal region of a bacterial lipoprotein, was purchased from Bachem (Torrance, CA). Fetal bovine serum (FBS) was purchased from HyClone (Logan, UT). PD98059 and wortmannin were purchased from Calbiochem (San Diego, CA). SB203580 was synthesized by Dr T. Chiba (Nagoya City University, Japan) according to the method of Gallagher et al.8
Female ICR mice (7 weeks of age) were purchased from Charles River (Yokohama, Japan). Animals were maintained under specific pathogen-free conditions. Food and water were given ad libitum. They were used in experiments after 1 week of acclimatization.
Mouse macrophage cell line RAW264.7 (ATCC TIB-71) was obtained from the American Type Culture Collection (ATCC, Manassas, VA). The cells were maintained in high glucose Dulbecco's modified Eagle's medium (DMEM) supplemented with 10% heat-inactivated FBS, 2 mm glutamine, 100 U/ml of penicillin G and 100 μg/ml of streptomycin.
Mouse hepatocyte cultures
Mouse hepatocytes were isolated using a modification of the collagenase method as described previously.7
RNA extraction and Northern blot analysis
Total RNA was extracted from hepatocytes seeded in 100-mm plates by following the guanidinium thiocyanate method.9 Thirty micrograms of total RNA was separated in a 1% agarose gel containing 2% formaldehyde and transferred onto a filter, Hybond-N+ (Amersham Pharmacia Biotech, Buckinghamshire, UK) with 20 × sodium saline citrate (SSC). Northern blot analysis was performed as described previously.10 A 548-base-pair (bp) murine TLR2 fragment and a 392-bp murine glyceraldehyde 3-phosphate dehydrogenase (GAPDH) fragment were obtained by reverse transcriptase–polymerase chain reaction (RT-PCR) using mRNA derived from RAW264.7 and EL-4.6.1 C10 cells, respectively, as templates based on published sequences.11,12 Primers used for TLR2 were 5′-GGC CAG GTT CCA GTT TTC AC-3′ (1017–1036) and 5′-GGA ACA ACG AAG CAT CTG GG-3′ (1791–1811), and those for GAPDH (392-bp) were 5′-TGG TCT ACA GGA TCC AGT ATG ACT CC-3′ and 5′-TGA TGG CAT GGA TCC TGG TCA TGA GC-3′. The TLR2 fragment (548-bp, 1108–1655) was cloned into HindIII and BamHI-digested pcDNA3.1 vector. GAPDH fragment was cloned into the BamHI site of pGEM-3Z vector.
Measurement of murine TLR2 promoter activity, NF-κB activation and SAA promoter activity
Mouse TLR2 promoter13 corresponding to a 1545-bp (from − 1486 to + 59) BglII–HindIII fragment prepared from genomic DNA of mouse hepatocytes by PCR was cloned into BglII and HindIII-digested pGL3-basic luciferase reporter vector (Promega, Madison, WI) to generate pGL3-TLR2 promoter, whose sequence was confirmed by automated sequencer. Primary cultured mouse hepatocytes were transiently transfected with 17·4 ng/well pGL3-TLR2 promoter (for TLR2 promoter reporter gene assay) or pGL3–4 κB wt (for NF-κB reporter gene assay), and 8·6 ng/well pCMV-β-gal in the presence of 174·0 ng/well pcDNA3.1, pcDNA3.1-TRAF6-DN,14 pcDNA3.1-TRAF2-DN,15 or pcDNA3.1-IκBα (S32/36 A), carrying alanines instead of serines 32 and 36,16 or in the presence of 174·0 ng/well pCMV and/or pCMV-p65 in 24-well plates by lipofection using effectene (QIAGEN GmbH, Hilden, Germany) according to the manufacturer's instructions. For SAA promoter reporter gene assay, the cells were transiently transfected with 50 ng/well pGL3-Type A SAA2 promoter,17 kindly provided by Drs A. Whitehead and C. Thorn (University of Pennsylvania School of Medicine, Philadelphia, PA), 10 ng/well pCMV-β-gal and 140 ng/well pCMV. After 24 hr incubation, the hepatocytes were washed three times and then stimulated with various reagents in fresh medium. The cells were then washed with PBS and lysed with 100 μl lysis buffer. Luciferase assay was performed with the Luciferase reporter gene assay kit (Roche, Germany) according to the manufacturer's instructions. The light emission was measured by multilabel counter 1420 ARVO (Pharmacia, San Francisco, CA). Luciferase activity was determined by using the β-galactosidase value as a basis for normalization.
Observation of NF-κB translocation by confocal laser scanning microscopy
To observe the translocation of NF-κB p65 to the nucleus, primary cultured mouse hepatocytes were cultured on collagen-coated OG chamber ZOG-3 (Elecon, Chiba, Japan). 30 min prior to stimulation as detailed in the figure legends, medium was removed and the cells were washed with PBS. After the cells were fixed with 3% formaldehyde and washed, the cells permeabilized by 0·2% Triton X-100 were stained with anti-NF-κB p65 rabbit polyclonal immunoglobulin G (IgG; 1 μg/ml; Santa Cruz Biotechnology, Santa Cruz, CA) or normal rabbit (Santa Cruz Biotechnology) followed by fluorescein isothiocyanate-conjugated goat anti-rabbit IgG (5 μg/ml) (Santa Cruz Biotechnology). Samples were viewed and photographed by using a Carl-Zeiss LSM-410 confocal laser scanning microscope (Oberkochen, Germany).
Preparation of cell extract
After primary cultured hepatocytes were treated for the indicated periods of time with IL-1α, TNF-α, LPS, or BLP on 60-mm dishes, the medium was removed. The cells were washed three times with ice-cold PBS, and the plates were chilled on ice immediately. An amount equal to 0·4 ml of ice-cold lysis buffer (20 mm Tris–HCl (pH 7·5), 1 mm EDTA, 120 mm NaCl, 0·5% Triton-X, 50 mm NaF, 10 mm sodium pyrophosphate, 10 mm sodium orthovanadate, 10 μg/ml leupeptin, 10 μg/ml antipain, 100 μg/ml benzamidine hydrochloride, 50 μg/ml aprotinin, 100 μg/ml soybean trypsin inhibitor, 10 μg/ml pepstatin, 1 mm phenylmethylsulphonyl fluoride) was added into the plates, and stirred for 15 min. The cells lysates were collected using a rubber policeman and sonicated. The cell debris was pelleted by 5 min centrifugation in a microcentrifuge, and the supernatants were collected.
Western blot analysis
For Western blots, equal amount of proteins were suspended in sodium dodecyl sulphate (SDS)–sample buffer. After boiling for 5 min, proteins were stored at − 80° until they were used. The proteins were separated by 8% SDS–polyacrylamide gel electrophoresis. The separated proteins were transferred to a polyvinylidene difluoride microporous membrane, Immobilon™ PVDF (Millipore, Bedford, MA). After blocking with 5% non-fat dry milk in Tris-buffered saline with 0·1% Tween-20 (TBST), membranes were incubated with rabbit serum anti-mouse TLR2 (eBioscience, San Diego, CA) or anti-β-actin mouse monoclonal antibody (Sigma) as a loading control at 1 : 1000 dilution, and then with horseradish peroxidase-conjugated anti-rabbit IgG (1 : 10000) (Jackson ImmunoReseach, West Grove, PA) or anti-mouse immunoglobulin, horseradish peroxidase-linked whole antibody (1 : 5000) (Amersham Pharmacia Biotech), respectively. The reactive proteins were detected with enhanced chemiluminescence reagents, ECL (Amersham Pharmacia Biotech) and analysed by a chemiluminescence image analyser, LAS-1000 (Fuji Photo Film).
Measurement of SAA concentration
The serum level of serum amyloid A (SAA) was measured by enzyme-linked immunosorbent assay (ELISA) as described previously.10 Differences between group means were assessed by unpaired t-test.
IL-1, TNF, LPS and BLP up-regulate TLR2 mRNA expression in primary cultured murine hepatocytes
We have previously shown that IL-1α, TNF-α and LPS up-regulate TLR2 but not TLR4 mRNA expression in murine hepatocytes as determined by the RT-PCR method.7 To confirm the observations and extend the study, we determined the expression levels of TLR2 mRNA in primary cultured murine hepatocytes treated with IL-1α, TNF-α, LPS, or bacterial lipoprotein (BLP) by Northern blot analysis. First, we treated the cells for 3 hr with IL-1α (1–100 U/ml), TNF-α (1–1000 U/ml), or bacterial component (1–1000 ng/ml). For IL-1α and TNF-α, doses as low as 1 U/ml were able to significantly increase TLR2 mRNA expression, while high doses of LPS and BLP (more than 100 ng/ml) were required to enhance the mRNA expression by murine hepatocytes (Fig. 1a). The mRNA was up-regulated by all stimulations at 1·5 or 3 hr, and the levels induced by cytokines and LPS were sustained for up to 24 hr. However, biphasic induction of TLR2 mRNA was observed in the BLP-treated hepatocytes (Fig. 1b). The augmenting effect of LPS was completely inhibited by polymyxin B (data not shown), indicating that the effect of LPS was not the result of lipoproteins which might be contaminated in the LPS preparation.
To analyse the signalling pathway involved in the TLR2 mRNA up-regulation, we pretreated mouse hepatocytes with PD98059 [MAPK ERK kinase (MEK) inhibitor that blocks ERK activation], SB208530 (p38 MAPK inhibitor), or wortmannin [phosphatidylinositol 3-kinase (PI3K) inhibitor] and then examined TLR2 mRNA up-regulation in hepatocytes in response to various stimuli by Northern blot analysis. PD98059, SB208530 and wortmannin were used at 10 μm, 10 μm and 100 nm, respectively, the doses were sufficient to inhibit each kinase. None of these inhibitors affected the up-regulatory effects of IL-1α, TNF-α, LPS, or BLP on TLR2 mRNA expression (data not shown).
The stability of TLR2 mRNA is not affected by IL-1, TNF, LPS, or BLP treatments in murine hepatocytes
We determined whether TLR2 mRNA stability in hepatocytes was affected by cytokines or bacterial components. Since the basal level of TLR2 mRNA in primary hepatocytes was too low to be demonstrated by Northern blot (data not shown), the cells were treated for 3 hr with IL-1α, TNF-α, LPS, or BLP to increase TLR2 mRNA. After washing with PBS, the cells were further incubated with actinomycin D in the presence or absence of cytokines or bacterial components. The half-life of TLR2 mRNA was in the range of 1·5–3·0 hr irrespective of the presence of various stimuli (Fig. 2), suggesting that the stability of TLR2 mRNA in hepatocytes was not affected by the treatment with IL-1α, TNF-α, LPS, or BLP.
NF-κB activation induced by IL-1, TNF, LPS, or BLP is necessary for the up-regulation of TLR2 mRNA expression in murine hepatocytes
Several transcription factors, NF-κB, STAT and Sp1, have been implicated in the gene activation of TLR2. NF-κB is especially important in the regulation of TLR2 gene expression in macrophages and T cells in response to LPS or TNF-α13 and in J774A.1 cells in response to infection with M. avium.18 TLR2 promoter activities in response to IL-1 or BLP treatments have not yet been reported with hepatocytes, macrophages, or T cells. Since NF-κB activation is common to IL-1α, TNF-α, LPS, or BLP stimulation, we compared the levels of murine TLR2 promoter (−1486) activity and NF-κB activation in mouse primary hepatocytes upon these stimulations by using each reporter gene. We used macrophage cell line RAW264.7 as a positive control. As shown in Fig. 3(a), TLR2 promoter activity and NF-κB activation were augmented by the treatment with TNF-α or LPS. In addition, the augmentation of TLR2 promoter activity in response to TNF-α or LPS was inhibited by overexpression of IκBα (S32/36A), the super suppressor of NF-κB, which lacks the phosphorylation sites for IKK and is resistant to signal-induced degradation. The RAW264.7 cells did not respond to IL-1α (data not shown). Next, we examined mouse hepatocytes. Treatments with IL-1α, TNF-α, LPS, or BLP augmented TLR2 promoter activity and NF-κB activation, both of which were comparably augmented in response to each stimulation. More importantly, the activation of both TLR2 promoter and NF-κB in hepatocytes in response to IL-1α, TNF-α, LPS, or BLP was completely inhibited by overexpression of IκBα (S32/36A) (Fig. 3b). Moreover, overexpression of TRAF6-DN in hepatocytes completely suppressed the activation of both TLR2 promoter and NF-κB induced by IL-1α, LPS or BLP, but not by TNF-α, consistent with the specific involvement of TRAF6 in IL-1α, LPS and BLP, but not in TNF-α, signalling pathways. In contrast, overexpression of TRAF2-DN completely suppressed the activation by TNF-α. Overexpression of NF-κB p65 resulted in the activation of both TLR2 promoter and NF-κB in mouse hepatocytes at a level comparable to IL-1α stimulation (Fig. 3c). Overexpression of NF-κB p50 did not affect the promoter activity (data not shown). Since the basal levels of TLR2 promoter and NF-κB activities were also reduced by overexpression of IκBα (S32/36A), TRAF6-DN and TRAF2-DN, NF-κB is presumably constitutively activated to some extent in mouse hepatocytes without stimulation.
We further examined whether NF-κB translocates to the nucleus in response to IL-1α, TNF-α, LPS, or BLP by confocal laser scanning microscopy. All of the reagents induced NF-κB translocation into the nucleus of hepatocytes (Fig. 4). Similar to the data from luciferase assay, IL-1α or TNF-α induced NF-κB p65 translocation in more cells than LPS or BLP, suggesting that mouse hepatocytes are perhaps more sensitive to cytokines than to bacterial products. Collectively, these results indicate that the increase in TLR2 mRNA level is regulated at the transcriptional level, in which NF-κB activation is important.
IL-1 is a potent inducer of TLR2 protein expression in murine hepatocytes
We next studied whether cytokines and bacterial components actually increase TLR2 at the protein level in primary cultured murine hepatocytes. Hepatocytes were treated with cytokines and bacterial components for the time as specified, and the TLR2 protein level was analysed by Western blot analysis (Fig. 5). TLR2 protein was expressed at a very low level in unstimulated hepatocytes at any time-point examined. Its expression was remarkably up-regulated in a time-dependent manner upon treatment with IL-1α or TNF-α. In contrast, LPS or BLP had little affect on the TLR2 protein expression. Since LPS and BLP can induce TLR2 mRNA expression, albeit less strongly than IL-1α and TNF-α, it is possible that differences in the activation of other signalling pathways by these receptors underlie the lower levels of TLR2 protein production. These agonists also activate MAPK signalling pathways. ERK and p38 MAPK pathways are implicated in regulation of mRNA stability and translation.19–21 These other signalling pathways may also regulate the levels of TLR2 protein.
IL-1 potentiates SAA production in vivo in response to BLP
It is known that hepatocytes produce serum amyloid A protein (SAA), one of the acute-phase proteins, in response to IL-1.22 Since TLR2 and IL-1RI share a common signal transduction pathway through their TIR domains1 and IL-1α induces TLR2 protein up-regulation in hepatocytes (Fig. 5), we speculated that BLP, a TLR2 ligand, might also induce SAA production by hepatocytes and BLP-induced SAA production might be augmented by pretreatment with IL-1α. First, we investigated the SAA promoter activity in hepatocytes cultured in vitro by BLP treatment. BLP-induced SAA promoter activity was augmented by pretreatment with IL-1α, but not TNF-α, LPS, or BLP compared to control treatment (Fig. 6a). Next, we examined if BLP and/or IL-1α could induce SAA production in vivo. To this end, IL-1α was intraperitoneally administered into mice, and BLP was injected 6 hr later. We have observed that the administration of IL-1α into mouse up-regulated TLR2 mRNA level in the liver after 3 and 6 hr treatment.7 The serum level of SAA was determined after 6 hr of BLP injection (Fig. 6b). Whereas BLP alone did not significantly induce SAA production, IL-1α did induce SAA. In addition, BLP induced a marked augmentation of SAA production in IL-1α pretreated mice, suggesting that (i) IL-1α can up-regulate functional TLR2 expression and (ii) BLP is capable of signalling through TLR2 to induce hepatocytes to produce SAA.
TLR2 mRNA expression is up-regulated by various cytokines and stimulants in T cells and macrophages. TLR2 mRNA expression is up-regulated by IL-2 or IL-15 in the mouse T-cell line CTLL-2, by phorbol 12-myristate 13-acetate (PMA) plus ionomycin in the mouse T-cell line S49.1,11 by LPS, IL-1β, IL-2, IL-15, TNF-α or IFN-γ in mouse macrophage cell line RAW264.7,23 and by IL-1α, TNF-α, GM-CSF, or M. avium in mouse macrophages.24 Perhaps, sustained expression of TLR2 mRNA and biphasic increase in mouse hepatocytes result from other factors induced by IL-1α, TNF-α, LPS, or BLP.
TLR2 mRNA up-regulation by PMA plus ionomycin is dependent on both ERK and p38 MAPK pathways in the mouse T-cell line S49.1,11 but in mouse macrophage cell line RAW264.7, ERK and p38 MAPK are not essential for LPS-mediated TLR2 mRNA induction.23 On the other hand, Staphylococcus aureus stimulation causes the recruitment of active Rac1 and PI3K to the TLR2 cytosolic domain, and tyrosine phosphorylation of TLR2 is required for the assembly of a multiprotein complex that is necessary for subsequent NF-κB transcriptional activity.1 Activation of PI3K in response to IL-1β and TNF-α leads to activation of NF-κB,25–27 and PI3K contributes to LPS-induced NO production.28,29 However, in our study, ERK, p38 MAPK and PI3K did not appear to participate in TLR2 mRNA induction by IL-1α, TNF-α, LPS, or BLP. These indicate that the regulation of TLR2 transcription by cytokines and bacterial products is different in distinct cell types.
In our study, the TLR2 mRNA was up-regulated by IL-1α, TNF-α, LPS, or BLP at transcription level. Liu et al. reported that IL-1β does not significantly affect the half-life of TLR2 mRNA in rat hepatocytes, and that the increased transcription rates for TLR2 were observed at 3 hr in total liver from LPS-treated rats by using nuclear run-on assay.30 However, it has not been examined whether the increase in hepatic TLR2 mRNA level is regulated at the transcriptional level. In response to LPS, BLP, or IL-1α, stimulation of TLR or IL-1R leads to the sequential activation of the adapter protein MyD88, IRAK, TRAF6, and eventually, IΚΚ complex.1 In contrast, in response to TNF-α treatment, IKK is recruited to the TNF-RI complex, and this recruitment is accomplished through TRAF2 but not through TRAF6.31 The reporter gene assay indicated that NF-κB activation induced by IL-1α, TNF-α, LPS, or BLP is required for up-regulation of TLR2 mRNA expression in mouse hepatocytes, because TLR2 promoter activity and NF-κB activation were correlated and were simultaneously inhibited by overexpression of IκBα (S32/36A). Moreover, TLR2 promoter activity and NF-κB activation, induced by IL-1α, LPS, or BLP but not by TNF-α treatment, were also completely inhibited by overexpression of TRAF6-DN (Fig. 3). The activation of NF-κB by IL-1α, TNF-α, LPS, or BLP was confirmed by translocation of p65 NF-κB subunit into the nucleus (Fig. 4). The extent of the translocation was well correlated to the activation of NF-κB and TLR2 promoter. Collectively, our study suggests that the up-regulation of TLR2 mRNA in response to IL-1α, TNF-α, LPS and BLP in hepatocytes is the result of transcriptional activation of the TLR2 gene in which NF-κB activation is essential.
Mouse TLR2 synthesis is strongly induced in the adipocyte by LPS, TNF-α and zymosan,32 and expression of TLR2 on the surface of mouse macrophages increases following the addition of M. avium.24 In our study, the expression of TLR2 protein was up-regulated remarkably by IL-1α, and significantly by TNF-α, but not by LPS and BLP (Fig. 5). In the liver, immune cells such as Kupffer cells rather than hepatocytes probably recognize bacteria and bacterial components, and then produce cytokines like IL-1α and TNF-α that, in turn, induce hepatic TLR2. Hepatocytes may have little capacity to recognize bacteria and bacterial components, and may not be able to exert physiological functions via TLR2 until hepatic TLR2 expression is augmented by cytokines. Northern blot analysis (Fig. 1) and reporter gene assay (Fig. 3), as well as observation by confocal laser scanning microscopy (Fig. 4), demonstrate that mouse hepatocytes are more sensitive to cytokines than to bacterial products. These findings are in agreement with a study carried out in rat.33
We showed that IL-1α pretreatment markedly augmented BLP-induced production of SAA (Fig. 6). SAA can act as a chemoattractant for immune cells such as monocytes, polymorphonuclear cells (PMN), mast cells and T lymphocytes.22 SAA is also an activator of PMN antimicrobial functions, such as induction of degranulation, phagocytosis and enhancement of anti-Candida activity.34 Our findings therefore suggest that augmentation of hepatic TLR2 expression by IL-1α is one means for the immune systems to mobilize the host defence mechanisms against pathogens through augmenting the hepatocyte response to bacteria or bacterial products.
In conclusion, we have demonstrated that the expression of TLR2 by hepatocytes can be up-regulated by bacterial products and cytokines, IL-1α in particular, through the activation of NF-κB signalling pathway. Hepatocytes may exert few pathophysiological functions via TLR2 signal pathway until TLR2 expression is up-regulated in response to pro-inflammatory cytokines that, in vivo, may be produced by neighbouring Kupffer cells. It is possible that TLR2 plays various important roles in the liver, especially during systemic inflammation and infection.
This work was supported in part by Grant-in Aid for Scientific Research (B) from the Japan Society for the Promotion of Science. We thank Drs Alexander S. Whitehead, and Caroline F. Thorn (Department of Pharmacology and Center for Pharmacogenetics, University of Pennsylvania School of Medicine, USA) for providing mouse SAA promoter plasmids. We also thank Drs Tadahide Furuno, and Taku Chiba (Nagoya City University, Japan) for their help with the analysis by CLSM, and the synthesis of SB203580, respectively. We also acknowledge Dr De Yang (National Institutes of Health, USA) for reviewing the manuscript.