Hepatitis C virus promotes t-helper (Th)17 responses through thymic stromal lymphopoietin production by infected hepatocytes


  • Potential conflict of interest: Dr. Ziegler owns stock in Amgen.

  • Supported by NIH grants AI098126 to Y.S.H.; U19AI066328 to H.R., Y.S.H.; U19AI083024 to S.J.S.


Hepatitis C virus (HCV) is a major cause of liver cirrhosis and hepatocellular carcinoma. Here we report that infection of hepatic cells by HCV stimulates nuclear factor kappa B (NFκB)-dependent production of thymic stromal lymphopoietin (TSLP). Hepatocyte-derived TSLP in turn conditions dendritic cells (DCs) to drive T-helper (Th)17 differentiation. The TSLP secreted by HCV-infected hepatoma cells is capable of activating human monocyte-derived DCs by up-regulating the expression of CD40, CD86, CCL17, CCL22, and CCL20 which are activating markers of DCs. In addition, the production of key cytokines for Th17 differentiation, transforming growth factor beta (TGF-β), interleukin (IL)-6, and IL-21, is enhanced by human monocytes upon coculture with HCV-infected cells. Importantly, the blockade of TSLP using neutralizing antibody prevented the activation and maturation of DCs as well as the production of Th17 differentiation cytokines. DC conditioning by TSLP secreted from HCV-infected cells activated naïve CD4+ T lymphocytes, resulting in Th17 differentiation. Furthermore, we can detect substantial levels of hepatocyte TSLP in fibrotic liver tissue from chronic HCV patients. Thus, blockade of TSLP released by HCV-infected hepatocytes may suppress the induction/maintenance of hepatic Th17 responses and halt the progression of chronic liver disease to fibrosis and liver failure. Conclusion: Hepatocyte-derived TSLP conditions DCs to drive Th17 differentiation. Treatment of TSLP neutralizing antibody in HCV-infected hepatocyte/DC coculture abrogates DC conditioning and thereby inhibits Th17 differentiation. (HEPATOLOGY 2013)

Hepatitis C virus (HCV) is a serious worldwide health problem, with more than 170 million people infected globally. HCV establishes persistent infection in 70% of infected individuals, leading to chronic liver inflammation, fibrosis, and cirrhosis.1 The outcome of HCV infection is primarily dictated by the magnitude and character of the T-cell response to infection. CD4+ T-cell responses play a critical role in the resolution of infection2, 3 and impaired HCV-specific CD4+ T-cell responses are observed in chronic HCV.3, 4 However, it is not known how HCV impairs CD4+ T-cell responses regarding the magnitude or alteration of differentiation of T cells and effector activity in the infected liver. Because of fenestrations in the liver sinusoidal endothelial cells, liver parenchymal cells (hepatocytes) are not separated from the vascular compartment by a basal membrane, and consequently HCV-infected hepatocytes have the potential to directly interact with innate immune cells such as liver resident dendritic cells (DCs). As cells of the innate immune system play a pivotal role in inducing and shaping the character of adaptive immune responses, the encounter of HCV-infected hepatocytes with liver DCs are likely to affect the activation state and properties of DCs and thereby influence the quality and effector function of T-cell responses to HCV.

Recently, interleukin (IL)-17-producing T-helper (Th)17 cells have been reported to trigger tissue inflammation and damage5 and there is accumulating evidence that Th17 cells are important contributors to hepatic inflammation and liver cirrhosis.6, 7 IL-17 is produced by monocytes/DCs through recognition of viral pathogen-associated molecular pattern (PAMP) such as Toll-like receptor (TLR)3 ligands.8 In addition to the ability of HCV to trigger the TLR3 pathway,9, 10 the increased number of Th17 cells appears to be associated with the severity of liver inflammation in chronic HCV patients and treatment of infected patients with pegylated interferon (IFN)-α and ribavirin reduced the level of Th17-related cytokines.11 As one of the crucial factors for Th17 differentiation, thymic stromal lymphopoietin (TSLP), a member of the common γ-chain cytokine, is capable of activating (conditioning) DCs, thereby stimulating naïve T cells to differentiate into Th2 cells.12 In addition, DCs treated with both TSLP and poly (I:C) activate naïve T cells and differentiate into Th2 and Th17 cells.8, 13 Thus, TSLP-activated DCs, which are known to be strong inducers of Th2 responses, can simultaneously induce Th17 cells under certain pathological conditions.

In this report we demonstrate that the infection of hepatic cells in vitro by HCV triggers robust TSLP production and this HCV-induced production of TSLP is regulated in an nuclear factor kappa B (NFκB)-dependent manner. TSLP secreted by HCV-infected cells activates and conditions human monocyte-derived DCs to enhance the production of Th17 differentiating cytokines, TGF-β, IL-6, and IL-21 by the DCs. Moreover, the addition of TSLP neutralizing antibody to the coculture of monocytes/DCs with HCV-infected hepatocytes blocks the production of these cytokines. Consistent with these data, we find that the hepatocyte-derived TSLP is readily detected in liver biopsies from chronic HCV patients. Our studies suggest a novel role for the hepatocyte-derived TSLP in the generation of CD4+ Th17 effector T cells through its ability to condition DCs to drive CD4+ Th17 differentiation. The potential implications of these findings in the development of HCV-induced chronic progressive liver disease are discussed.


DC, dendritic cell; HCV, hepatitis C virus; TSLP, thymic stromal lymphopoietin.

Materials and Methods

Subject and Cell Preparation.

Human hepatoma cell lines, Huh 7.5.1, were maintained in Dulbecco's modified Eagle's medium (DMEM) with 10% fetal bovine serum (FBS) and penicillin/streptomycin (100 μg/mL). THP-1 cells purchased from the American Tissue Culture Collection (ATCC) were cultured in RPMI 1640 and supplements as recommended by ATCC. Liver biopsies and peripheral blood samples from chronic HCV or control patients were obtained from Dr. Hugo Rosen (University of Colorado). Blood samples were also obtained from the Virginia Blood Services. All information of age, gender, and HCV genotype were previously described.14, 15 For infection of cells with secreted HCV, Huh 7.5.1 permissive cells were seeded at 3 × 106 cells in a T75 plate for 24 hours. Cells were infected with 4 × 104 FFU (multiplicity of infection [MOI] of 0.01) of JFH-1 producing cell supernatant and cultured for 10 days in DMEM-10% FCS media. Determination of infectivity was performed as described.16 To serve as a control, supernatant from JFH-1-infected hepatocytes (UV-JFH-1sup) were irradiated with 3.6 J/cm2 for 25 minutes at 254 nm using an ultraviolet (UV) crosslinker MT-10 (UVP, Upland, CA). JFH-1 was kindly provided by Dr. Wakita (Tokyo Metropolitan Institute).

Real-Time Quantitative Polymerase Chain Reaction (PCR).

Total RNA was extracted with the RNeasy Mini kit (Sigma-Aldrich) and reverse-transcribed to complementary DNA (cDNA) using the SuperScript III (Invitrogen). The real-time reverse-transcribed (RT)-PCR was performed on an ABI Prism Sequence Detection System (Applied Biosystems). The primer sequences were as follows: hIL-17A (sense, 5′-GGTCCTCAGATTACTACAACCG-3′; antisense, 5′-GTTCCCATCAGCGTTGATGCAG-3′), hIL-17F (sense, 5′-CCATGTCACGTAACATCGAG-3′; antisense, 5′-GTTCCTACACTGGGCCTGTA-3′). The sense and antisense primer sequence for RORc, CCR6, TSLP, CCL17, CCL22, CCL20, and GAPDH were synthesized as described.8, 17

Western Blot Analysis.

Cells were cultured in 6-well plates and treated with JFH-1 as described in the figure legends. Western blot analysis was performed as described.17

Enzyme-Linked Immunosorbent Assay (ELISA).

The level of IL-17, TGF-β, IL-6, IL-21, and IL-12 (eBiosciences), and TSLP (R&D Systems) in culture supernatants were quantified by ELISA according to the manufacturer's instructions.

Transfection and Luciferase Assay.

Huh 7.5.1 cells (5 × 105) were seeded into 6-well plates and transfected using the Mirus transfection reagent (Mirus Bio) with various plasmids (see figure legends for details) and pRL-TK was used as the normalization control. The cells were cultured for 30 hours and then infected with JFH-1 for 11 hours. Cells were harvested and lysed in 100 μL lysis buffer. Luciferase activity was measured as described.18

Generation and Activation of Human Monocyte-Derived DCs.

To obtain DCs, peripheral blood mononuclear cells (PBMCs) were allowed to adhere at 5 × 106 cells/mL onto 6-well plates in RPMI-10 culture medium. After 24 hours of incubation, nonadherent cells were removed by stringent rinsing and adherent cells were cultured in DC culture medium supplemented with granulocyte-macrophage colony-stimulating factor (GM-CSF; 50 ng/mL) and IL-4 (400 U/mL). On day 6, the culture consisted uniformly of CD11c+ CD14 HLA-DR+ CD86+ CD83 immature DCs. Cell differentiation was monitored by light microscopy and the expression of cell surface molecules was analyzed by flow cytometry after 6 days of culture. To determine the activation status, DCs were cultured in medium, supernatant from JFH-1-infected hepatocytes (JFH-1sup), rTSLP, or JFH-1sup plus anti-TSLP monoclonal antibody (mAb) for 24 hours and stained with PE-conjugated mouse antibodies to CD40, CD80 and CD86. The activation was analyzed by flow cytometry.

DCs-CD4+ T Cell Coculture and Intracellular Staining.

Purified DCs were plated at 2 × 104 cells/well in 96-well plates and cultured in the presence of IL-1 (10 ng/mL), IL-6 (20 ng/mL), IL-23 (20 ng/mL), JFH-1sup, or JFH-1sup plus anti-TSLP mAb (10 μg/mL) for 24 hours. These DCs were cocultured with freshly isolated allogeneic CD4+ T cells (4 × 104 cells/well). After 7 days of culture, cells were harvested and stimulated with 0.1 μg/mL phorbol 12-myristate 13-acetate (PMA) and 0.5 μg/mL ionomycin (Sigma-Aldrich) for 6 hours, with the presence of GolgiStop in the last 4 hours (BD Bioscience). Cells were stained for surface antigen CD4, fixed, and permeabilized using Cytofix/Cytoperm (BD Bioscience), followed by intracellular IL-17 and IFN-γ staining. Flow cytometry was performed as described.19

Confocal Microscope Analysis.

Liver tissues from normal and chronic HCV patient were kindly provided by Dr. Hugo R. Rosen. Frozen sections (4 um thickness) of OCT-embedded tissues were treated with 0.3% Triton X-100, blocked in horse and donkey serum, and incubated with anti-TSLP Ab followed by donkey Alexa Fluor-546-antigoat IgG Ab and Alexa Fluor-488-cytokeratin mAb. Goat IgG and Alexa Fluor 488-mIgG1 were used for control staining. Sheep anti-TSLP Ab was purchased from R&D Systems. For fibrosis staining, mouse anti-cytokeratin pan mAb (Clone C-11; Sigma) and biotinylated mouse antihuman collagen IV mAb (Cedarlane) were used and followed by Alexa Fluor-555-conjugated streptavidin. A-546-conjugated donkey antigoat Ab (Invitrogen) was absorbed with mouse and rat serum before use. Confocal images were captured on a Zeiss LSM-700 confocal microscope and analyzed by ZEN software.

Statistical Analysis.

Student t test (two-tailed) was used for statistical analysis of differences between two groups. P values are depicted as *P ≤ 0.05. All data were analyzed using Prism software (GraphPad Prism4).


TSLP Secretion by HCV-Infected Hepatoma Cells and Detection of TSLP-Expressing Hepatocytes from Chronic HCV Patients.

Because TSLP plays a critical role in triggering inflammatory responses and promotes Th2 and Th17 differentiation in response to microbial infection,12 we examined whether HCV infection of hepatic cells stimulates TSLP production. To this end we analyzed the impact of in vitro infection of Huh 7.5.1-derived cell lines with a replicating JFH-1 HCV strain. We first infected Huh 7.5.1 cells with HCV (JFH-1) virus for 10 days and infection was then confirmed by immunofluorescence through the expression of HCV core protein (Fig. 1A). We next examined the tempo of TSLP messenger RNA (mRNA) induction in Huh 7.5.1 cells following JFH-1sup (supernatant from JFH-1-infected hepatocytes) infection. The TSLP signal was first detected early in infection, from about 4 to 8 hours, and reached maximal levels at 12 hours postinfection (Fig. 1B). The TSLP signal also enhanced TSLP protein release at 24 hours, which showed a significantly higher fold increase of TSLP production by HCV-infected cells compared to control cells (Fig. 1C). In contrast, TSLP induction was significantly decreased in cells infected with UV-irradiated JFH-1sup (Fig. 1B,C). These results demonstrate that human TSLP is induced in hepatocytes by HCV infection.

Figure 1.

HCV increases TSLP production in JFH-1-infected human hepatoma cells. (A) Huh7.5.1 cells infected with JFH-1 (MOI of 0.1) for 10 days and confirmed by immunofluorescence using antibodies; primary antibody was mouse anti-HCV core mAb, and secondary antibody was goat antimouse Alexa Fluor 488 antibody. DAPI was added to the staining to monitor the nuclear structure. (B) Huh7.5.1 cells were inoculated by UV-irradiated JFH-1sup or JFH-1 (MOI of 0.1) for the indicated time course, and human TSLP mRNA levels were quantified by real-time PCR. TSLP mRNA levels were normalized with GAPDH (B) and TSLP protein levels were analyzed by ELISA (C). Data represent means ± SD of three independent experiments. Similar results were obtained for five independent experiments (B,C). n.d.: not detected.

To determine if HCV infection of hepatocytes in situ within the infected liver stimulated TSLP production, we analyzed TSLP expression in liver tissues from chronic HCV patients. In keeping with our in vitro data on TSLP expression by HCV-infected hepatocytes (Fig. 1B,C), we detected significant up-regulation of TSLP mRNA level in liver tissues from chronic HCV patients relative to those in healthy individuals (Fig. 2A). Of particular interest, we found that the expression of RORc, a key transcription factor in the Th17 cell lineage, was also significantly up-regulated in liver tissue from chronic HCV patients (Fig. 2B). To precisely define the cellular source of TSLP, we carried out immunofluorescence staining of liver tissues from chronic HCV patients. As shown in Fig. 2C (panels P1 to P4), there was extensive fibrosis (indicated by collagen-red staining) in the interlobular regions of the liver biopsies from HCV patients as well as intense cellular infiltration in the areas of fibrosis. As expected, in biopsies from individuals without chronic disease there was no staining of collagen reticulin fibers and minimal collagen deposition in liver stromal elements (Fig. 2C, N1 and N2). Cytokeratins are proteins of keratin-containing intermediate filaments found in the intracytoplasmic cytoskeleton of epithelial tissue. Human TSLP was found to be expressed by epithelial cells in the peripheral mucosal-associated lymphoid tissue, where it activates myeloid dendritic cells to induce a strong TH2 response in vivo and in vitro.20 Strikingly, a significant production of TSLP was found in the liver of HCV patients but not in those of normal individuals (Fig. 2D). TSLP staining was largely, if not exclusively, localized to hepatocytes because TSLP staining was found within hepatic lobules and colocalized with cytokeratin, a hepatocyte marker (arrows). Minimal TSLP staining was found in the fibrotic interlobular septa. In contrast, staining of normal and patient samples with control Ab showed little staining, indicating that the staining is specific for TSLP. Taken together, these results indicate that TSLP is indeed produced by hepatocytes in patients with chronic HCV infection. Furthermore, TSLP production in this tissue might be responsible for inducing the expression of Th17 differentiating cytokines and a transcription regulator, RORc, associated with CD4+ Th17 differentiation.

Figure 2.

Increased TSLP is located in hepatocytes of liver tissues of HCV patients. (A,B) Total RNA was extracted from liver tissues of healthy and chronic HCV patients. The mRNA expression for TSLP or RORc was quantified by real-time PCR. (C) Fibrosis in HCV patient livers. Tissue sections (4 μm thickness) of normals (N1 and N2) and HCV patients (P1 to P4) were stained with Alexa Fluor 488-anti-cytokeratin plus biotinylated mouse anti-collagen IV followed by Alexa Fluor 555-streptavidin. Regions with collagen deposition in patients (red) are marked by asterisks (*). Hp marks hepatocyte regions in the lobules. Liver histology from two normals and four patients are presented. Micrographs were captured at 200× on a LSM700 confocal microscope and bars = 200 μm. *Fibrotic areas; Hp, hepatocytes. (D) The confocal images of normal and chronic HCV patient samples are shown. Sections stained with control Ab showed low background staining (A). Low-magnification images show staining of TSLP in hepatocytes in patients but not in normal hepatocytes (B). Higher magnification shows the presence of TSLP staining of HCV-infected liver but minimal staining in normal liver (C). TSLP staining colocalizes with cytokeratin which stains hepatocytes (C). Arrows show the colocalized staining. The widths of the bars are in μm. A total of three normal and five HCV patient samples were analyzed and the results are similar. Insets show magnified images of boxed areas.

NFκB Activation Mediates TSLP Gene Expression in HCV-Infected Hepatoma Cells.

In the host response to HCV infection IPS-1 has been reported to localize in the mitochondria and plays a critical role in the activation of IFN regulatory factor-3 (IRF-3) and NFκB. To understand the mechanism of TSLP induction by JFH-1-infected cells, we first assessed the ability of IPS-1 to trigger the TSLP promoter, which is located 4.0 kb upstream at the start of transcription (pGL3-b+hTSLP-full) in human Huh 7.5.1 cells. Overexpression of wildtype IPS-1 led to enhanced TSLP promoter activity following JFH-1sup infection (Fig. 3A). We next examined TSLP promoter induction by overexpression of wildtype IRF-3. No difference was found in the ability of IRF-3 to express TSLP in response to JFH-1sup (Fig. 3B). We also investigated the effect of overexpression of wildtype NFκB or a dominant-negative mutant of IκB kinase (IKKβ). Overexpression of the NFκB p65 subunit resulted in increased expression of the TSLP promoter in the absence of JFH-1sup infection. In addition, infection of the cells with JFH-1sup further enhanced expression (Fig. 3C). Conversely, overexpression of cDNA encoding a dominant-negative mutant of IKKβ inhibited JFH-1-mediated transcription in a dose-dependent fashion (Fig. 3D). The human TSLP gene promoter contains two NFκB binding sites, 3.8 and 0.2 kb upstream of the start of transcription.18 A single mutation in any of the two motifs caused notable decreases in TSLP expression. Furthermore, double mutations led to more profound decreases than a single mutation (Fig. 3E). We confirmed that NFκB was activated following infection with JFH-1sup. As expected, JFH-1 induced phosphorylation of NFκB in hepatoma-derived cells (Fig. 3F). NFκB is a ubiquitously expressed transcription factor known to mediate the expression of many inflammatory mediators including cytokines, adhesion molecules, chemokines, and growth factors. These results indicate that HCV-induced TSLP production occurs through the activation of NFκB, a crucial mediator in the innate immune pathway.

Figure 3.

NFκB-dependent transcriptional activation of human TSLP by HCV-infected hepatoma cells. (A-E) Induction of human TSLP promoter in cells cotransfected with 1 μg of wildtype human TSLP promoter plasmid or the indicated plasmids. At 30 hours after transfection, cells were infected with JFH-1 virus (MOI of 0.1) for 11 hours. Cell lysates were prepared and promoter activity was measured by luciferase activity. Huh7.5.1 cells were transiently transfected with control plasmid pGL3-basic, a mixture of luciferase constructs containing only the human TSLP promoter plasmid (hTSLP-full), or containing a combination of hTSLP-full and expression vectors for IPS-1 expression plasmid (IPS-1) (0.05 or 0.5 μg) (A), IRF-3 expression plasmid (IRF-3) (10 or 50 ng) (B), NFκB expression plasmid (NFκB) (10 or 50 ng) (C), dominant-negative mutant of IκB plasmid (DN-IκB) (50 or 100 ng) (D), or promoter NFκB-site mutants (m1 = deletion of −3.2 kb site, m2 = deletion of −0.2 kb site) (1 μg) (E). Luciferase activity in the whole cell lysate was normalized to Renilla luciferase activity. Low doses are represented as + and high doses as ++. Data represent mean ± SD of triplicate data points of one experiment and are representative of five independent experiments. (F) Huh7.5.1 cells were infected with JFH-1sup (MOI of 0.1) for the indicated time course and the cell lysates were subject to immunoblotting by using phosphorylated NFκB or NFκB antibody.

TSLP Released by HCV-Infected Cells Activates DCs and Produce Factors Crucial for Th17 Differentiation.

TSLP is a potent stimulator of DC activation to increase expression of MHC II, costimulatory molecules (i.e., CD40, CD80, and CD86), cytokines, and chemokines (i.e., CCL17, CCL22 CCR4+ T-cell recruitment)20 and CCL20 (i.e., recruitment for CCR6+ Th17 cells).21 Therefore we examined the effect of TSLP secreted from JFH-1-infected hepatocytes on activating DCs by coculturing monocyte-derived DCs with recombinant TSLP, JFH-1sup, JFH-1sup plus neutralizing anti-TSLP antibodies, or untreated control culture media. As shown in Fig. 4A, the expression of costimulatory molecules (i.e., CD40, CD86) was increased in DCs after exposure to JFH-1sup to a level comparable to that of rTSLP-treated DCs. Addition of anti-TSLP neutralizing antibodies to JFH-1sup inhibited their ability to activate costimulatory molecules on DCs. We also wanted to know whether TSLP receptor is expressed in these DCs and if DCs activation is blockable by anti-TSLP antibody. TSLP receptor is expressed by DCs, and was further up-regulated following culture in JFH-1sup. In contrast, TSLP receptor expression was decreased by neutralization of TSLP receptor (data not shown). In addition, CCL17, CCL22, and CCL20 were expressed at higher levels in JFH-1sup-stimulated DCs than that in media control-treated DCs, and neutralization of TSLP suppressed chemokine production by these DCs (Fig. 4B). Taken together, these data demonstrated that TSLP produced by hepatocytes infected with HCV can support DC activation/maturation.

Figure 4.

TSLP released by JFH-1-infected cells activates human monocyte-derived DCs by inducing the production of TGF-β, IL-6, and IL-21 cytokines. (A) FACS analysis of DCs after 24 hours incubation with the indicated stimuli: rTSLP, JFH-1sup, JFH-1sup plus anti-TSLP mAb. Filled histogram represents staining for DCs cultured with media alone and open histogram represents staining for DCs cultured with indicated stimuli. The expression levels of DC activation markers CD40, CD80, and CD86 were determined by flow cytometry. (B) The expression level of CCL17, CCL22, and CCL20 chemokines relative to GAPDH was determined by real-time PCR. Data shown are the mean ± SD of three independent experiments (B) or one experiment representative of three independent experiments with similar results (A). Both the JFH-1-infected and non-infected HepG2 cells were cocultured with THP-1 cells in the transwell system at a 1:1 ratio for the indicated time course in the presence or absence of anti-TSLP antibody (15 ng/mL). The levels of TGF-β, IL-6, IL-21, and IL-12 production were measured in the culture supernatant by ELISA (C). Data represent mean ± SD of three independent experiments. These results were reproducible from five independent experiments.

To determine whether HCV-infected hepatocyte-derived TSLP might also condition antigen-presenting cells (e.g., monocytes, macrophages, DCs) to stimulate Th17 differentiation, we examined the capacity of mononuclear cells exposed to HCV-infected hepatic cells or their supernatants to release inflammatory cytokines, in particular, TGF-β, IL-6, and IL-21, which supports the differentiation and expansion of human Th17 cells. The production of these cytokines was determined from the coculture of the human monocytic cell line, THP-1, with either JFH-1-infected or uninfected hepatoma cells using the transwell system. Notably, JFH-1-infected HepG2 cells stimulated a statistically significant increase in the secretion of TGF-β, IL-6, IL-21, but not IL-12, by THP-1 cells in a transwell membrane system (Fig. 4C). However, cultures of THP-1 cells alone produced low levels of TGF-β, IL-6, and IL-21 cytokines regardless of the presence of JFH-1sup (data not shown). Importantly, the addition of anti-TSLP-neutralizing antibodies led to a decrease of Th17-specific cytokines (Fig. 4C). These results suggest that monocytes/DCs conditioned by TSLP secreted from HCV-infected hepatocytes produce Th17 differentiating cytokines which could support the induction of CD4+ Th17 responses.

DCs Conditioned by TSLP Released from HCV-Infected Cells Promote CD4+ T-Cell Differentiation into Th17 Cells.

Based on the role of IL-1, IL-6, and IL-21 production in Th17 polarization, we evaluated the effect of hepatocyte-derived TSLP on Th17 differentiation in coculture of naïve T cells with DCs activated by IL-1/IL-6/IL-23, JFH-1sup, or JFH-1sup plus anti-TSLP antibodies. Following stimulation with PMA/ionomycin, the production of intracellular cytokines (IFN-γ, IL-17A) by CD4+ T cells was assessed using flow cytometry. As expected, IL-1/IL-6/IL-23-treated DCs, used as positive control, produced more IL-17 cells compared to control cells (5.09 ± 0.6% versus 0.91 ± 0.08%). Notably, the percentage of IL-17-producing cells increased after coculture of CD4+ T cells with JFH-1sup-treated DCs (4.65 ± 0.55%), which then significantly decreased upon the addition of anti-TSLP mAbs (1.21 ± 0.1%) (Fig. 5A,B). There was no significant difference in the percentage of IFN-γ production from JFH-1sup-treated DCs in the presence or absence of anti-TSLP antibody (Fig. 5A,B). This result was further verified by the detection of IL-17 release using ELISA. The enhancement of IL-17-producing T cells by JFH-1sup-treated DCs was significantly inhibited by neutralization of TSLP (Fig. 5C). This suggests that TSLP released from infected hepatocytes activates and conditions DCs to drive the differentiation of activated CD4+ T cells into Th17 cells.

Figure 5.

DC conditioned by TSLP released from HCV-infected cells contributes to Th17 differentiation. (A) CD4+ T cells were cocultured with monocyte-derived DCs in the presence or absence of JFH-1sup for 7 days. After 7 days of coculture, T cells were restimulated with PMA and ionomycin for 6 hours, with the last 4 hours in the presence of GolgiStop, and production of IL-17 and IFN-γ was determined by intracellular cytokine staining (A). (B) Bar graphs show percentage of IL-17, IFN-γ-producing cells, measured by flow cytometry. (C) IL-17A production was measured in the culture supernatant by ELISA. Data shown are the mean ± SD of three independent experiments (B,C) or one experiment representative of three independent experiments with similar results (A).

Addition of Both TSLP and NS3/5 Enhances the Production of Th17 Cytokines by CD4+ T Cells from Chronic HCV Patients.

To further examine the effect of TSLP on promoting Th17 differentiation during HCV infection, we assessed the capacity of Th17 cell generation by CD4+ T cells from PBMC in a chronic HCV patient. As shown in Fig. 6A, there is a significant increase of Th17 lineage-specific transcription factor (i.e., RORc) and markers (i.e., CCR6 and CD161) from chronic HCV patients as compared to those in healthy individuals. We next determined the role of HCV-specific antigen in induction of Th17 CD4+ T cells. HCV NS3/5 proteins have been reported to induce a Th17 response.22 Th1/Th17 cells differentiations were compared using intracellular staining of IFN-γ and IL-17, respectively. The results indicated that Th17 cells were significantly increased in response to NS3/NS5 compared to normal control (5.18 ± 1.09% versus 1.21 ± 0.90%) (Fig. 6B,C). We also observed a higher percentage of IL-4-producing cells in addition to IL-17-producing cells, but not IL-10-producing cells (data not shown). In addition, NS3/5 treatment also resulted in a significant induction of IL-17 production, while they reduced the production of IFN-γ (Fig. 6D). Collectively, these data indicate that IL-17-producing CD4+ T cells are induced during HCV infection.

Figure 6.

Direct action of TSLP enhance IL-17 production by PBMC from HCV-infected patients. (A) CD4+ T cells were isolated from PBMC of healthy and chronic HCV patients. Expression levels of mRNA encoding IL-17A, IL-17F, RORc, and CCR6 were quantified using real-time PCR. (B-D) CD4+ T cells were cocultured for 7 days by DCs activated with 5 μg/mL mixed-NS3/NS5 or medium alone. T cells were restimulated with PMA plus ionomycin after 7 days of coculture. Production of IL-17 and IFN-γ was determined by intracellular cytokine staining (B). (C) Bar graph show the percentage of IL-17-producing cells in total CD4+ cells. (D) Cytokine production was measured in the culture supernatant by ELISA. Data shown are the mean ± SD of three independent experiments (C,D) or one experiment representative of three independent experiments with similar results (B). Purified CD4+ T cells from PBMC of HCV-infected patients were stimulated with 5 μg/mL mixed-NS3/NS5 or medium alone, either in the presence or absence of recombinant TSLP antibody (5 ng/mL). Cells were harvested after 72 hours culture. Expression levels of mRNA encoding IL-17A, IL-17F were quantified using real-time PCR (E,F). Production of IL-17 was measured in the culture supernatant using ELISA (G). Data represent mean ± SD of three independent experiments.

We next examined the possibility that the direct action of TSLP on CD4+ T cells is involved the differentiation of Th17 cells. CD4+ T cells from PBMC of HCV-infected patients were stimulated with NS3/NS5 in the presence or absence of TSLP for 3 days. Stimulation with TSLP or NS3/5 significantly enhanced IL-17 mRNA and protein compared to control cells cultured medium only (Fig. 6E,F,G). Moreover, differentiation of IL-17 cells was increased following combined stimulation of TSLP and NS3/5 compared to either TSLP or NS3/5 alone. These results clearly indicate that TSLP is an important factor in the differentiation of IL-17-producing CD4+ T cells during HCV infection and might play a role in the development of chronic liver diseases.


In this report we demonstrate that HCV infection of hepatocytes induces NFκB-dependent TSLP gene expression and protein production. Furthermore, TSLP mRNA and protein was increased selectively in liver tissues from chronic HCV patients. Intriguingly, TSLP released from HCV-infected hepatocytes activates human monocyte-derived DCs and conditions DCs to support the polarization of CD4+ T cells toward Th17 cells. The blockade of TSLP action by neutralizing antibody suppresses differentiation of Th17 cells. These results suggest a novel mechanism to account for the infiltration of TH17 cells in the liver as likely a result of the role of TSLP in promoting Th17 differentiation. However, it remains unclear whether hepatic TSLP accounts for facilitating the recruitment of Th17 infiltration in the infected liver. These results also raise the intriguing possibility that the crosstalk between HCV-infected hepatocytes and local (liver and/or liver draining lymph node) DCs may be a pivotal mechanism both in a defective antiviral (Th1) CD4+ T-cell response and also in an enhanced expression of an injury-provoking (Th17) CD4+ T-cell response. To our knowledge, our findings represent the first report identifying hepatic-secreted TSLP as a regulator of Th17 differentiation.

Although CD4+ T cells have been reported to be critical to the antiviral function of CD8+ T cells in chronic infection, failure of CD4+ T cell help is associated with the inability to clear HCV infection and CD4+ T-cell responses in chronically infected HCV patients leans toward Th2 deviation and T regulatory (Treg) cells. Based on the report that the frequency of IFN-γ-producing CD4+ T cells in the liver was significantly reduced compared to those in peripheral blood, it is likely that the function of cells within the liver may be subject to be influenced by local immunoregulatory factor(s). Our studies in this report implicate that circulating CD4+ T cells from HCV-infected individuals are skewed toward Th2 and Th17 cell differentiation via secretion of TSLP from HCV-infected hepatocytes (Fig. 1). In addition, a combination of TSLP with HCV proteins (i.e., NS3/5) increases Th17 differentiation (Fig. 6). Interestingly, another group showed that TGF-β and IL-10, which are induced by the HCV NS4 protein, suppress Th1 and Th17 responses in HCV-infected patients.23 Given that the above study used the NS4 protein alone, it is possible that this viral protein alone is able to dampen the immune response by way of production of IL-10, which, in turn counteracts Th17 responses. However, our study is based on the whole virus, JFH-1, containing all HCV proteins and thereby DCs in exposure to JFH-1 HCV virus produce TSLP. Moreover, the production of TSLP depends on TGF-β and has been shown to antagonize IL-10 production. Recent studies have reported that HCV-specific IL-17-producing CD8+ T cells are detectable in blood and liver of chronically HCV-infected patients.8, 24 In addition, IL-17-producing CD4+ T cells have also been shown to be present in chronic HCV-infected patients.22, 23, 25 These results suggest that HCV-specific IL-17-producing T cells are not limited to the CD4+ T cells alone, but also CD8+ T cells. Nevertheless, the molecular and cellular mechanisms underlying the generation of these Th17 cell responses in HCV infection were not elucidated. Our results suggest a mechanistic link between TSLP derived from HCV-infected hepatocytes and the infiltration of IL-17-producing Th17 T-cells into the HCV-infected liver. HCV-derived proteins play a role in inducing TGF-β, IL-6, and IL-21 production from monocyte-differentiated DCs.26 In terms of a potential relationship between TSLP and these cytokines, there is no existing report for a direct effect of these cytokines in TSLP induction. However, it is worthwhile to point that IL-6 is well known to activate STAT3 and TSLP is also able to induce STAT3 activation.27 Thus, there is the convergence of intracellular pathways downstream of TSLP and IL-6 and thereby these cytokines act in similar ways leading to skew T-cell responses towards Th17 cells. Our studies in this report demonstrate that HCV infection of hepatocytes induce TSLP production by these cells, resulting in the activation and maturation of DCs with increased expression of CCL17, CCL22, and CCL20 chemokines. Importantly, the blockade of TSLP action by neutralizing antibodies prevents DC activation/maturation conditioned by HCV-infected hepatocytes (Fig. 4). Thus, there may be at least two different pathways for triggering DC activation and maturation during viral infection: (1) direct sensing of viral PAMP by DCs, and (2) crosstalk between virus-infected cells and DCs. Although plasmacytoid DCs are activated by direct sensing of HCV by way of TLR7 and produce Type 1 IFN,28 TSLP acts as a key molecule in linking HIV-exposed epithelial cells to activation of myeloid DCs and these activated myeloid DCs, in turn, promote CD4+ T cell proliferation and increase HIV replication.29

Although Th17 cells participate in inducing proinflammatory responses during viral infection, the antiviral activity of Th17 cells to control HCV replication appears to be limited. Rather, Th17 cells appear to play a role in the progression of liver fibrosis and cirrhosis.22, 25 Indeed, histological studies show extensive infiltration of Th17 cells in the area of severe hepatic damage. Based on the role of TSLP in inducing differentiation of CD4+ Th17 effector T cells, we speculate a potential mechanism(s) for HCV-induced Th17 differentiation through several stepwise processes of crosstalk between virus-infected cells and local DCs. Liver-resident DCs are conditioned by TSLP released from HCV-infected hepatocytes to favor, either at the site of viral infection and/or in the draining lymph node, the differentiation of Th17 cells, which then home to liver damage sites and possibly contribute to disease progression by interacting with other immune cells and nonimmune cells. This points out a critical biological role of TSLP secreted by HCV-infected hepatocytes in activating DCs which result in Th17 differentiation.

Future studies are necessary to elucidate whether Th17 cells promote liver damage and are directly involved in enhanced virus survival or if they are simply less capable than Th1 cells in the induction of antiviral responses. Nevertheless, our studies point out that blockade of TSLP might be a new strategy for the treatment of chronic HCV patients with severe liver diseases. As TSLP is known for its role in inducing inflammation, efforts to block TSLP is being investigated for its potential as a treatment for asthma. The availability of therapeutic agent(s) targeting TSLP could accelerate the translation of these findings into clinical practices such as treatment of HCV-associated chronic liver diseases.

In summary, our findings suggest that TSLP produced by HCV-infected hepatocytes can enhance the development of potential injury-provoking CD4+ Th17 effector T cells through the ability of cytokines to condition DCs to drive the differentiation of CD4+ T cells towards Th17 effector status. If, as we believe, CD4+ Th17 effector T-cell responses in the HCV-infected liver are important regulators of liver injury, then TSLP might represent a novel therapeutic target in HCV-infected liver with the potential to limit tissue injury and possibly promote virus clearance.


We thank Susan Landes for excellent technical assistance, Dr. Thomas J. Braciale and Dr. Taeg S. Kim for critical discussion of the article prior to submission, and members of the Hahn laboratory for helpful discussions throughout the course of this work.