Functional polymorphisms of TLR8 are associated with hepatitis C virus infection

Abstract

Chronic hepatitis C virus (HCV) infection is a worldwide threat to public health. Toll-like receptor 8 (TLR8) is critical for eliminating RNA viruses, and variation within the TLR8 gene may alter the function of TLR8 in response to HCV infection. Our previous study demonstrated that the TLR8-129G>C (rs3764879) and TLR8+1G>A (rs3764880) variants were in complete linkage disequilibrium, and that the frequency of TLR8-129C/+1A was significantly higher in male patients with HCV infection compared with the healthy controls. In the present study, we found that the promoter activity of TLR8-129G was higher than that of TLR8-129C in THP-1 cells. Moreover, TLR8-129G mRNA stability and competitive DNA-binding ability were significantly lower than that of TLR8-129C. To investigate the functional effects of TLR8 polymorphisms, we compared the nuclear factor-κB (NF-κB)-driven luciferase activity in HEK293 cells transfected with the TLR8 variants. TLR8+1A plasmids induced less NF-κB signalling than did those transfected with TLR8+1G after 20 μm CL075 (P = 0·011) stimulation. We also analysed the mRNA expression and cytokine production in whole blood and monocytes from people of various genotypes stimulated ex vivo by the interferon-γ and TLR7/8 agonist CL075, R848. TLR8 expression in CD14⁺ cells derived from volunteers with TLR8-129G/+1G was significantly higher than that derived from TLR8-129C/+1A, and interleukin-12p40 production was higher in volunteers with TLR8-129G/+1G after stimulation. The data indicate that variations in TLR8 genes may modulate immune responses during HCV infection.

Introduction

Successful host defence against viral pathogens requires the rapid recognition of virus-specific ‘danger signals’ and the activation of both innate and adaptive immunity. Hepatitis C virus (HCV), a single-stranded (ss) RNA virus, infects more than 170 million people worldwide.^1,2 The clinical outcome of HCV infection is highly variable, and genetic factors involving innate immunity are likely to affect disease susceptibility and progression after infection.^3,4 The role of Toll-like receptor 8 (TLR8) in host defence against viruses suggests that TLR8 is a suitable marker for exploring the linkage between variants in innate-immune genes and susceptibility to viral infections. TLR8, the gene of which is located on the X chromosome, is situated in endosomes where it recognizes non-self nucleic acids (such as viral ssRNA) and subsequently activates downstream signals to induce inflammatory cytokines and type I interferons.^5–8

The genetic variants of TLRs and downstream signalling molecules can influence the ability of the host to respond appropriately to TLR ligands and consequently alter the susceptibility of the affected person to infectious diseases.^9–11 In our previous study, we screened more than 250 Han Chinese people in Taiwan and determined that a TLR8 Met/Val substitution at the start codon and a G>C change (rs3764880) at the −129 position in the promoter region (rs3764879) are in complete linkage disequilibrium, with TLR8-129G/+1G allele frequencies of 84·0%.¹² A single nucleotide polymorphism (SNP) at the start codon (ATG>GTG, Met 1 Val) of TLR8 causes a frame-shift mutation leading to the formation of a truncated form of TLR8 (1038 versus 1041 amino acids) containing a three-amino-acid deletion at the N-terminus, which is also predicted to be a component of the signal peptide.^12,13 Several published reports have demonstrated that TLR8 Met1Val confers statistically significant changes to the host regarding the progression of HIV infection and the susceptibility to Mycobacterium tuberculosis infection, asthma and related atopic disorders, particularly in males.^13–15 In HIV patients of Indonesia, the TLR8-129G/+1G genotype confers a significant protective effect,¹³ and males carrying the TLR8-129C/+1A genotype are more susceptible to tuberculosis infection (P = 0·007, odds ratio = 1·8).¹⁴

HCV carries an ssRNA genome, which might be the natural TLR8 ligand that modulates the immune response to HCV infection. However, only a few studies have investigated the functional effects of genetic TLR8 polymorphisms and the mechanisms responsible for the differences in TLR8 function during activation.^12,13 Therefore, the aim of this study was to understand the genetic mechanism underlying host pathogen-recognition receptors by investigating the functional differences and signal transduction that occur after the ligand-induced stimulation of TLR8 genetic variants. We previously genotyped a population of 187 HCV-positive male adults and 146 age-matched control participants for TLR8 Met1Val SNPs. TLR8-129C/+1A frequency was significantly higher in male hepatitis C patients than in healthy controls (17·6% versus 6·8%; P = 0·004), with an odds ratio of 2·91 (95% CI 1·38–6·13).¹⁶ In the current study, we provide evidence that the innate immune characteristics of TLR8 may depend upon the type of genetic variant present within the cells.

Materials and methods

Cell culture

Human embryonic kidney (HEK293) cells and human hepatoma Huh-7 cells were maintained in Dulbecco's modified Eagle medium (GIBCO-BRL; Life Technologies, Gaithersburg, MD), 10% fetal bovine serum (FBS; Hyclone Laboratories, Waltham, MA), 5 mm HEPES (Invitrogen, Carlsbad, CA), and 1% penicillin/streptomycin (Invitrogen). Human monocytic THP-1 cells and murine macrophage RAW264.7 cells were grown in RPMI-1640 medium (GIBCO-BRL, Life Technologies), supplemented with 10% heat-inactivated FBS (Hyclone Laboratories), 2 mm l-glutamine, 100 U/ml of penicillin and 100 μg/ml of streptomycin.

TLR8 promoter activity analysis

The TLR8 promoter region (−1042 to −68) was amplified from the genomic DNA of THP-1 cells (Table S1) and cloned into the pGL3-basic vector (Promega, Madison, WI). This plasmid served as a wild-type reporter (TLR8-129C) construct. The resulting TLR8-129G variant was derived by facilitating the site-directed mutagenesis of TLR8-129C using a QuikChange Site-Directed Mutagenesis Kit (Stratagene, La Jolla, CA). The cloned plasmids were transformed into Ecos101™ competent cells (DH5α strain) (Yeastern Biotech Co., Ltd., Taipei, Taiwan) and purified using PureLink® HiPure Plasmid Filter Maxiprep Kit (Life Technologies, Grand Island, NY). TLR8-129C or TLR8-129G plasmids were transfected into Huh7 and HeLa cells by using Lipofectamine 2000 (Invitrogen) for 24 hr; pRL-TK (Promega) was used as the internal control. THP-1 cells were transfected through electroporation by using a 4D-Nucleofector™ System according to the manufacturer's instruction (Lonza, Walkersville, MD). After transfection, the TLR8 promoter-transfected THP-1 cells were treated with 100 ng/ml of interferon-γ (IFN-γ) for 6 hr. Luciferase activity was measured using a Berthold Luminometer (LB 9507; Berthold Technologies, Bad Wildbad, Germany). Firefly luciferase (Luc) activity was normalized to TK-renilla luciferase, and the relative activity was calculated based on the TLR8 reporter activity relative to the pGL3-basic plasmid.

Electrophoretic mobility shift assay

Nuclear extracts were harvested in cold PBS from THP-1 cells. Cell pellets were lysed in 400 μl of freshly prepared lysis buffer A containing 1 × protease inhibitor cocktail (Roche Applied Science, Penzberg, Germany) and incubated on ice for 10 min. Afterwards, 40 μl of 10% Nonidet-40 was added to the cell lysate, and the portion containing intact nuclei was obtained by centrifuging the lysate at 13 200 g for 10 seconds at 4°. Intact nuclei were further lysed during a 20-min incubation on ice with 100 μl of freshly prepared lysis buffer B containing 1 × protease inhibitor cocktail, followed by centrifugation at 13 200 g for 20 min at 4°. The THP-1 nuclear extract was stored at −80°, and the protein concentration was determined using BCA assay (Thermo Scientific, Waltham, MA).

The 106-bp TLR8-129C and TLR8-129G double-stranded DNA probes (Table S1) were analysed using electrophoresis and confirmed by sequence analysis. These probes were labelled with digoxigenin (Stratagene). Electrophoretic mobility shift assay (EMSA) was performed using a Gel Shift Assay System (Promega). Briefly, 9 μg of THP-1 nuclear extract was incubated with 0·844 ng of digoxigenin-labelled TLR8-129C probe in binding buffer (0·5 mm EDTA, 0·5 mm dithiothreitol, 4% glycerol, 1 mm MgCl₂, 50 mm NaCl, 10 mm Tris–HCl), 0·005 μg/ml of poly l-lysine, and 0·05 μg/ml of poly[d(I-C)] at room temperature for 15 min. To determine the affinity of the TLR8-129C probes, competition reactions were performed by adding a 5-, 10-, 50-or 100-fold excess of the unlabelled TLR8-129C and TLR8-129G probes. The nuclear extract/DNA probe complexes were then separated in a 4·5% non-denaturing DNA–PAGE gel at 4° in 0·25 TBE buffer at 300 V for 1·5 hr. The gels were electrophoretically transferred to Biodyne® B 0·45-μm membranes at 400 mA for 1 hr. The nuclear extract/DNA probe complexes were detected using an anti-digoxigenin-AP solution and then incubated with CDP-star detection solution (Blossom Biotechnologies, Waltham, MA). The chemiluminescent signal on the nylon membranes was detected using an LAS-3000 imaging system (Fujifilm, Tokyo, Japan).

TLR8 expression and reporter assay plasmids

Human TLR8 (transcript 2; +1G) was generated using PCR from human complementary DNA (cDNA) derived from THP-1 (Table S1) and was cloned into the mammalian expression vector pcDNA4/V5-His B (Invitrogen) to produce the TLR8+1G plasmid. The first nucleotide of the TLR8 gene was mutated from ‘G’ to ‘A’ by using a QuikChange Site-Directed Mutagenesis Kit (Stratagene) to generate the TLR8+1A plasmid.

TLR8 mRNA stability test

HEK293 cells were transiently transfected with either pcDNA4-TLR8+1G or pcDNA4-TLR8+1A. The TLR8 mRNA levels were determined after actinomycin D (5 μg/ml) treatment and harvesting at various time-points. Total RNA was isolated using a Total RNA Extraction Miniprep System (Viogene, Taipei, Taiwan). The relative TLR8 mRNA levels were determined using a CFX96 Touch™ Real-Time PCR Detection System (Bio-Rad Laboratories, Inc., Hercules, CA) and normalized to the amount of β₂-microglobulin expression (Table S1) observed in the transfected HEK293 cells.

Assessment of nuclear factor-κB activation using luciferase assays

HEK293 cells were seeded into a 10-cm culture plate at a density of 1 × 10⁵ cells per well and incubated overnight at 37° and 5% CO₂. The cells were transiently co-transfected using Lipofectamine 2000 (Invitrogen) with 5 μg of various TLR8 expression plasmids (pcDNA4, HEK293-TLR8+1G, or HEK293-TLR8+1A), 0·5 μg of luciferase-containing nuclear factor-κB (NF-κB) receptor plasmids (pNF-κB-Luc), and 0·01 μg of pRL-TK (Promega) as an internal control. After 24 hr of transfection, HEK293 cells were collected and aliquots were plated onto 12-well plates. Eight hours after plating, the cells were treated with the indicated concentration of CL075 (20 μm), R848 (10 μm) (InvivoGen, San Diego, CA), or recombinant HCV core protein (100 pg/ml) (ProSpec, East Brunswick, NJ) for 12 hr. After stimulation, HEK293 cells were lysed with 1 × passive lysis buffer (Promega). The firefly and Renilla luciferase assays using a Dual-Luciferase Reporter Assay System were obtained using an EG&G Berthold Microplate Luminometer (LB 9507) (Berthold Technologies). The relative luciferase activity was calculated as the ratio of NF-κB-firefly luciferase to TK-renilla luciferase and was normalized to the value obtained for the HEK293 cells transfected with pcDNA4 only.

Ex vivo whole blood stimulation

Whole blood was obtained at a fixed time in the morning from healthy male volunteers with the TLR7IVS2-151A genotype (rs179009)¹² who were anti-HCV negative and had provided informed consent before donating. Genomic DNA was extracted from EDTA-containing blood by spin columns with a QIAamp® Blood Mini Kit (Qiagen, Hilden, Germany). TLR8 (−129 G>C) polymorphisms were genotyped using PCR-restriction fragment length polymorphism, as previously described.¹⁶

Peripheral blood mononuclear cells (PBMC) were separated using Ficoll-Hypaque density gradients from heparinized blood. The cells from the interface were washed three times in PBS (without calcium and magnesium) with RPMI-1640 medium supplemented with 10% FBS, 100 U/ml of penicillin, 100 μg/ml streptomycin, and 2 mm glutamine, and allowed to settle at 37°. After 24 hr, the non-adherent cells were removed with warm PBS and the adherent cells were harvested. Freshly heparinized whole blood was diluted twofold using Hanks’ balanced salt solution and 5 × 10⁵ isolated monocytes were incubated with 20 μm CL075, 100 pg/ml recombinant HCV core protein, 10 μm R848 and 100 ng/ml of IFN-γ for either 6 hr or 24 hr. Total RNA was extracted using a Qiagen RNA Blood Isolation Kit (Qiagen) according to the manufacturer's instructions. Complementary DNA was synthesized using reverse transcription with oligo-dT primers. The TLR8 and cytokines [IFN-α, IFN-β, IFN-γ, tumour necrosis factor-α (TNF-α), interleukin-6 (IL-6), IL-10 and IL-12p40] mRNA expressions were analysed using quantitative RT-PCR (qRT-PCR) (Table S1) with an ABI PRISM 7700 Sequence Detection System (Applied Biosystems, Warrington, WA).

The intracellular TLR8 protein levels in monocytes from healthy male volunteers with various haplotypes were analysed using flow cytometric analysis. First, 100 μl of stimulated whole blood was stained with 10 μl FITC-conjugated anti-CD14 antibody (BD Biosciences, San Jose, CA) for 30 min at room temperature. The red blood cells were lysed by adding 2 ml of working 1 × BD FACSTM Lysing Solution into each reaction tube for 5 min at room temperature, and the excess unbound antibodies were then washed away using PBS. To perform intracellular staining, the remaining leucocytes were fixed and permeabilized by adding 250 μl of Fixation and Permeabilization Solution (BD Biosciences) for 20 min at 4°, after which the cells were incubated with 2 μl phycoerythrin-conjugated anti-TLR8 antibody (Catalogue no: IMG-321D; Clone 44C143; Imgenex, San Diego, CA) diluted with 18 μl 10% FBS in PBS containing 0·2% saponin for 1 hr at 4°. The cells were then resuspended in PBS and analysed using a FACSCalibur flow cytometer with WinMDI98 software (BD Biosciences). All of the experiments and measurements were conducted blindly with regard to donor polymorphisms.

Statistical analysis

The results from the luciferase assays and qRT-PCR were presented as the average fold induction over vector-only cells or cells that were not stimulated. All of the values are expressed as mean ± SD. The results were analysed using the Wilcoxon–Mann–Whitney method for unpaired values. P-values of < 0·05 were considered statistically significant. SPSS (version 14.01) was used for data management and statistical analyses, and Graph Pad Prism 5 software was used to generate the figures.

Results

Comparison of TLR8 promoter activity in THP-1

The TLR8 promoter region (−1042 to +68) was amplified from genomic DNA and cloned into the pGL3-basic vector. TLR8-129C or TLR8-129G plasmids were transfected into various cell lines for 24 hr. The results demonstrated that TLR8 promoter activation in THP-1 cells were significantly higher than those in the other cell lines. In addition, the levels of TLR8-129G promoter activity were significantly higher than that of TLR8-129C (P = 0·05), although no significant differences were observed in either the Huh7 or HeLa cell lines (Fig. 1). However, the differences between them were negligible after IFN-γ treatment, although the activities were enhanced in both of the genotypes compared with those at the baseline levels (Fig. 1).

Figure 1.

Toll-like receptor 8 (TLR8) promoter activity in various cell lines. The luciferase activity of TLR8 promoter pGL3-1042–-68 plasmids containing TLR8-129C or TLR8-129G were measured in Huh7, HeLa and THP-1 cells. Transfected cells were harvested 24 hr after transfection. The transfected THP-1 were treated with 100 ng/ml interferon-γ (IFN-γ) for 6 hr or left untreated. The relative luciferase activity is expressed as the TLR8 reporter activity relative to the pGL3-basic plasmid. The values represent the mean ± SD derived from three independent experiments. *P < 0·05 between the plasmids of the two genotypes.

The ability of the THP-1 nuclear extracts to shift the mobility of the digoxigenin-labelled TLR8-129C and TLR8-129G probes was evaluated using EMSA. To compare the relative binding affinities of the TLR8-129C and TLR8-129G probes of the monocyte-derived nuclear extracts, competition experiments were conducted by adding a 5-, 10-, 50-or 100-fold excess of unlabelled TLR8-129C and TLR8-129G probes to the THP-1 nuclear extracts. The EMSA results shown in Fig. 2 reveal that the concentration required for the TLR8-129G probes to bind to the nuclear extracts was significantly higher than that of TLR8-129C, indicating that the THP-1 nuclear extracts might have a higher affinity for TLR8-129C than for TLR8-129G. These results suggest that TLR8-129C>G might modulate transcriptional regulation in monocytes and subsequently lead to various promoter activities between the two genotypes.

Figure 2.

THP-1 extracts bind to the Toll-like receptor 8 (TLR8) promoter probe. A digoxigenin-labelled TLR8-129C promoter probe was used in an electrophoretic mobility shift assay with THP-1 nuclear extracts. To conduct competition analysis, an excess of unlabelled probes was used. Lane 1, probe alone; lane 2, probe and nuclear extract (NE); lanes 3–6, probe, NE, and increasing amounts of unlabelled competitor TLR8-129C probes; lanes 7–10, probe, NE, and increasing amounts of unlabelled competitor TLR8-129G probes.

The TLR8+1G variant exhibits a less stable expression and higher activation of NF-κB compared with that of TLR8+1A

The qRT-PCR data regarding the relative TLR8 mRNA levels in HEK293 cells transiently transfected with either TLR8+1A or TLR8+1G plasmids were determined after administering actinomycin D treatment at various time-points. We determined that the amount of TLR8 mRNA in TLR8+1G-transfected cells was significantly lower than that of the TLR8+1A-transfected cells after 12 hr (Fig. 3). Therefore, these results demonstrate that TLR8+1G mRNA may be less stable than TLR8+1A in HEK293 cells.

Figure 3.

Analysis of Toll-like receptor 8 (TLR8) mRNA stability. Actinomycin D was added to TLR8 variant-transfected HEK293 cells to block mRNA synthesis. Cell lysates were collected using TRIzol at the indicated time-points. TLR8 mRNA expression was determined using quantitative RT-PCR. The value of 0 hr was set as 100%. The bar represents means ± SDs of three independent experiments. *P < 0·05 when comparing TLR8+1A with TLR8+1G genotypes.

To determine the functional activity of each TLR8 variant, we compared the HEK293 cells transiently co-transfected with the TLR8 isoforms and an NF-κB luciferase-reporter plasmid after being stimulated with CL075, R848 or recombinant HCV core protein. The levels of NF-κB-driven luciferase activity were significantly higher in HEK293 cells co-transfected with TLR8+1G compared with those co-transfected with TLR8+1A under basal conditions (P < 0·05) and in response to CL075 treatment (P < 0·01) (Fig. 4). However, HEK293 cells over-expressing TLR8+1G and TLR8+1A exhibited almost the same amount of NF-κB activity after being stimulated with 10 μm R848 and 100 ng/ml of HCV core protein (Fig. 4).

Figure 4.

Differential responsiveness of HEK293 cells transfected with Toll-like receptor 8 (TLR8) variants after undergoing stimulation. HEK293 cells were co-transfected with the nuclear factor-κB (NF-κB) luciferase-reporter plasmid and with plasmids encoding TLR8 variants. The cells were treated with 20 μm CL075, R848 (10 μm), or hepatitis C virus (HCV) core protein (100 ng/ml) for 12 hr at 37°. The indicated fold changes that occurred during relative activation were based on the cells transfected with pcDNA4 from four independent experiments (mean ± SD). *P < 0·05, **P < 0·01 between the plasmids of the two genotypes.

TLR8-129G/+1G carriers increase cytokine production after ex vivo simulation

Figure 5 shows the results from the quantitative determination of TLR8 mRNA expression in the whole blood of healthy male volunteers carrying either of the two TLR8 genotypes. Volunteers with the TLR8-129C/+1A genotype produced significantly higher levels of basal TLR8 mRNA expression (P = 0·015) after CL075 stimulation (P = 0·002) compared with those of the TLR8-129G/+1G genotype. However, after R848, IFN-γ and recombinant HCV core-protein stimulation was performed, TLR8 mRNA expression did not differ significantly between the two genotypes.

Figure 5.

Quantitative analysis of Toll-like receptor 8 (TLR8) mRNA expression in volunteers with various genotypes, using RT-PCR. Total RNA was extracted from the whole blood of male volunteers 24 hr after being stimulated with 20 μm CL075, R848 (10 μm), interferon-γ (IFN-γ; 100 ng/ml), or hepatitis C virus (HCV) core protein (100 ng/ml). Relative TLR8 mRNA expression was determined using RT-PCR and was obtained by the 2^(−△△C_T) method; this was normalized with an endogenous control, glyceraldehyde-3 phosphate dehydrogenase (GAPDH), and compared with the self basal expression. Each dot represents a volunteer: (○) represents volunteers with the TLR8-129C/+1A genotype, and (•) represents volunteers with the TLR8-129G/+1G genotype. The horizontal bars represent the mean values. *P < 0·05 between the two genotypes.

Figure 6 shows the results from the FACS analysis of intracellular TLR8 protein levels in monocytes from individuals carrying various TLR8 genotypes. Those containing the TLR8-129C/+1A genotype possessed significantly fewer TLR8-expressing CD14⁺ cells compared with those containing TLR8-129G/+1G. After CL075 stimulation was performed, significantly more TLR8-expressing CD14⁺ cells were present within each of the TLR8-129C/+1A-and TLR8-129G/+1G-carrying whole blood cells; comparing the two genotypes, the percentage of TLR8-positive CD14⁺ cells was significantly higher in whole blood cells of the TLR8-129G/+1G genotype than those of the TLR8-129C/+1A genotype (Fig. 6a). However, after recombinant HCV core-protein stimulation, the TLR8 protein levels in the monocytes did not differ significantly between the two genotypes. TLR8 protein expression within the monocyte population was not significantly different between the two genotypes under basal conditions or after stimulation, as assessed by comparing the mean fluorescence intensity (MFI) (Fig. 6b). In addition, the TLR8 protein expression in natural killer cells and neutrophils did not differ significantly between the two genotypes under basal conditions or after stimulation (data not shown).

Figure 6.

Toll-like receptor 8 (TLR8) protein expression in monocytes of healthy male volunteers with various genotypes. TLR8 protein expression in whole blood from healthy male volunteers with various genotypes before and 6 hr after undergoing stimulation with 20 μm CL075 or 100 ng/ml of hepatitis C virus (HCV) core protein. Monocytes were identified using FACS analysis. (a) TLR8 expression is represented as the percentage of CD14-positive cells; (b) TLR8 protein expression (mean fluorescence intensity) in CD14-positive cells. (○) represents volunteers with the TLR8-129C/+1A genotype (n = 8), and (•) represents volunteers with the TLR8-129G/+1G genotype (n = 11). The horizontal bars represent the median values. *P < 0·05 between the two genotypes.

To examine whether the TLR8 variants affected cytokine production, we stimulated whole blood and monocytes from healthy donors with 20 μm CL075, 10 μm R848 and 100 ng/ml IFN-γ for 6 hr, and the mRNA expression of cytokines was analysed using qRT-PCR. Ex vivo-stimulated whole blood cells from TLR8-129G/+1G donors produced more IL-12p40 than those from TLR8-129C/+1A donors did (P = 0·03 for CL075) (Fig. 7a). However, the monocytes from TLR8-129G/+1G donors exhibited more TNF-α after IFN-γ stimulation, and produced more IL-12p40 after CL075 stimulation than those from TLR8-129C/+1A donors did (Fig. 7b). The data of stimulation with 10 μM R848 was presented in Fig. S1, in whole blood, R848 induced IL-6, IL-10, and IL-12, with higher level seen at 6 h than 24 h. But the effect of R848 on monocytes was not significant. The expression of other cytokines, including IL-6 and IL-10, was not significantly different between the two genotypes after stimulation (see Supplementary material, Fig. S2).

Figure 7.

Ex vivo cytokine production upon stimulating whole blood cells and monocytes in culture. (a) Fresh whole blood samples and isolated (b) monocytes from male volunteers of various genotypes were stimulated ex vivo, using 20 μm CL075, 10 μm R848, or 100 ng/ml interferon-γ (IFN-γ) for 6 hr. The relative mRNA induction folds of cytokines were determined using RT-PCR and compared with the self basal expression. Each dot represents a volunteer: (○) represents volunteers with the TLR8-129C/+1A genotype (n = 6), and (•) represents volunteers with the TLR8-129G/+1G genotype (n = 5). The horizontal bars represent the median values.

Discussion

Variation within a host genome has been reported to contribute substantially to HCV infection.^17,18 A characteristic feature of the immune status in chronically infected HCV patients is a weak immune response.^19,20 Our previous study demonstrated that TLR8+1A polymorphisms (rs3764880) were significantly more frequent in male participants with chronic HCV infection compared with the healthy controls.¹⁶ TLR8 activation caused by HCV RNA was recently suggested to directly mediate the persistent state of immune activation observed in HCV-infected patients.²¹ This suggests that variations in the TLR8 gene may modulate immune responses during HCV infection. However, the mechanisms by which TLR8 polymorphisms affect HCV infection are not fully understood.

In our previous study,¹⁶ we investigated the functional effects of TLR8 polymorphisms by analysing the mRNA expressions of TLR8 and cytokine production induced ex vivo by TLR8-specific agonists using whole blood from participants of various genotypes. TLR8 mRNA expression for wild-type (-129C/+1A) was significantly higher than for those with the-129G/+1G haplotype. In the present study, we demonstrated that TLR8 polymorphisms altered transcriptional activity. Carriers of the TLR8-129G/+1G allele exhibited high levels of promoter activity but relatively unstable mRNA in vitro. Promoter activity analysis and EMSA analysis were used to demonstrate the effect of TLR8-129 polymorphisms (rs3764879) on promoter basal activity. The TLR8+1 polymorphism (rs3764880) affected mRNA stability and NF-κB activation through CL075 stimulation. When we conducted the ex vivo analysis, the basal levels were compared using methods that were used in the previous study,¹⁶ and the stimulating effects associated with CL075, R848 and IFN-γ were compared. In addition, whole blood and monocyte systems were compared to demonstrate the dependence of cytokine production on TLR8-, TLR7-and TLR7+8 caused by various genotypes. The data demonstrated that basal IL-12p40 expression differed significantly between the two genotypes in the whole blood stimulation system, and that after CL075 and R848 stimulation was performed, participants with TLR8-129G/+1G produced higher levels of pro-inflammatory cytokine expression than those with TLR8-129C/+1A did (Fig. 7a). Consequently, this increased TLR8 expression, and NF-κB activation might be responsible for the enhanced immune activation observed in TLR8-129G/+1G carriers, which explains why the natural course of HCV infection is inhibited in these hosts.

We also investigated the functional effects of the TLR8 polymorphisms in leucocytes from healthy volunteers during stimulation. In ex vivo whole-blood and monocyte assays, we determined that the TLR8-129G/+1G mutation was associated with an increase in TLR8 protein and pro-inflammatory cytokine expression in monocytes after stimulation (Figs 6a and 7b). The low mRNA levels in cells carrying the TLR8-129G/+1G mutation are probably caused by decreased TLR8 mRNA stability. Oh et al. demonstrated that TLR8 polymorphisms modulated cytokine patterns in monocyte-rich PBMCs, resulting in high levels of TNF-α production.¹³ The results from the monocyte stimulation system used in this study were consistent with their findings. Therefore, the altered pro-inflammatory cytokine profile, which produces a beneficial effect on HCV susceptibility, observed in mutation carriers, might be caused by the differential impact of TLR8 polymorphisms on monocyte responses.

Our previous study demonstrated the frequency of complete linkage disequilibrium in TLR8 SNPs,-129C>G (rs3764879) and +1A>G (rs3764880).¹² The NCBI ORF-finder software revealed that when the ATG start codon was altered to GTG, the TLR8 molecule shifted downstream by nine nucleotides and gained a new start codon from the +10 to the +12 position, so deleting three amino acids at the N-terminus.²² According to the GeneBee software analysis results, the secondary RNA structures of the two TLR8 variants were predicted to differ, indicating that they might interact with various proteins and mediators involved in numerous RNA degradation pathways, such as RNA polymerases, helicases and chaperones. Consistent with this prediction, the results of this study confirmed that the low levels of mRNA expression in cells carrying the TLR8-129G/+1G genotype might be caused by unstable TLR8 mRNA. In addition, the TLR8-129C>G SNP in the promoter region might alter the pattern of transcription-factor binding and consequently change its promoter activity. The EMSA analysis results indicated that the concentration of TLR8-129G probes competing to bind to monocyte-derived nuclear extracts was significantly higher than that of TLR8-129C. However, the ‘G’ allele did not cause the TLR8 Kozak context to shift, which was predicted to cause the initiation of TLR8 gene translation to be slightly less favourable.²² TLR8-129G under both of the Kozak contexts controlled by the native TLR8 promoter expression demonstrated a significant decrease in translation efficiency.

The data presented in this study demonstrate the protective effect of the TLR8-129G/+1G variant on HCV susceptibility. The presence of the mutation caused NF-κB activation to increase after CL075 stimulation in vitro and strongly affected the induction of an immune response. The modulation of NF-κB activation prompts us to propose that future studies should address the relevance of TLR8 polymorphisms in the clinical course of other RNA viral diseases. When we studied the production of pro-inflammatory cytokine in whole blood and monocytes from male volunteers with various TLR8 genotypes after being stimulated ex vivo by IFN-γ and TLR7/8 agonists, CL075 and R848, we observed increased levels of TNF-α and IL-12p40 production in TLR8-129G/+1G carriers after stimulation occurred. This suggests that the TLR8 variant plays a vital role in the generation of responsiveness to stimulation. We also demonstrated that whole blood cells and PBMCs have distinct capacities for producing cytokines that are crucial during the early stages of infection. The reason that a low level of pro-inflammation cytokine production was observed in whole blood cells from TLR8-129G/+1G carriers is that whole blood, unlike PBMCs, contains neutrophils, which are poor cytokine producers in short-term cultures. The results of this study also confirm that TLR8 polymorphisms might influence TLR8 expression and NF-κB activation after TLR8 agonist stimulation.

Whether the TLR8 SNPs are associated with HCV infection in several African and Asian countries that have a higher prevalence rate of chronic HCV-induced hepatitis remains unclear. TLR8, located in endosomes, acts as an antiviral receptor for recognizing ssRNA, which is present at various stages of viral infection. After TLR8 is activated by ssRNA, signals are transduced through the myeloid differentiation primary response gene 88 (MyD88) and subsequently enhances NF-κB and IFN regulatory factors 5 and 7 to produce inflammatory cytokines, such as IL-8 and type I IFNs.⁸ Therefore, TLR8-mediated innate immune responses might play a key role in viral pathogenesis. Several clinical studies have demonstrated the antiviral effect of TLR8 in response to ligand stimulation.^23,24 The ssRNA genome of HCV is a well-known natural ligand for TLR8 activation.⁸ HCV core protein was reported to be involved in negatively regulating the dendritic cell-induced T helper type 1 response.²⁵ Therefore, we investigated the effect of HCV core protein stimulation on cells containing the TLR8 variants to clarify the role of TLR8 in the immune response against HCV infection and further understand the mechanism that underlies the interaction between variant TLR8 proteins and HCV core protein and impairs inflammatory responses. To address these research topics, we performed a series of in vitro experiments to determine the effects of TLR8 polymorphisms on immune modulation using HCV core protein. The results demonstrated that TLR8 polymorphisms affected TLR8 expression and signal transduction after CL075, but not HCV core-protein, stimulation. Although monocytes comprise approximately 10% of circulating leucocytes and play crucial roles in inflammatory responses, studies on the immune status of chronic HCV patients have not focused on monocytes. In response to TLR ligation, monocytes can produce large quantities of pro-inflammatory and anti-inflammatory cytokines, which are vital in the eradication of pathogens and might also lead to immunopathology of tissue damage.¹⁹ As valuable IL-10 producers, monocytes can also act as crucial negative regulators, limit excessive immune responses to pathogens, and thereby prevent damage to the host;²⁶ whether this mechanism plays a role in chronic HCV infection is currently unclear, and conflicting data have been reported. Monocytes from HCV patients are generally considered to be more activated than those in healthy individuals, as indicated by the higher TNF, IL-12p40 and IL-10 production that occurs in the absence of activating stimuli.²⁷ In this study, we detected only TNF-α and IL-12p40 expression in monocytes from healthy donors after performing stimulation. The mechanisms underlying this enhanced activation state of monocytes in patients with chronic HCV are not clear, but HCV core protein was detected in the serum of chronic HCV patients and may prime circulating monocytes.²⁸ Currently, the manner in which monocytes respond to pathogen-derived products in chronic HCV patients, compared with the response to these products in healthy individuals, is not entirely understood. Therefore, the HCV core-protein stimulation that we observed in whole blood from healthy volunteers may have occurred through a mechanism other than CL075 stimulation.

In conclusion, variations in the TLR8 gene may modulate immune responses during HCV infection. The results of the present study might have implications for assessing the risk profile of individual patients and for the future use of TLR agonists in the prevention of, or therapy for, HCV infection.

Disclosures

The authors declare no conflict of interest.

Supporting Information

Additional Supporting Information may be found in the online version of this article:

Figure S1

Ex vivo cytokine production upon R848 stimulating whole blood cells and monocytes in culture.

Figure S2

Ex vivo cytokine production upon stimulating whole blood in culture.

Table S1

Primer sequences of this paper used.