Intrahepatic Innate Lymphoid Cells Secrete IL-17A and IL-17F That Are Crucial for T Cell Priming in Viral Infection

Viral hepatitis is one of the most important public health problems globally. Many viruses can cause acute or chronic hepatitis, typically hepatitis A through E, as well as liver infections caused by adenovirus (Ad) and several other viruses. In most cases, patients’ immune responses to a virus or a viral strain vary greatly, which leads to a wide range of clinical manifestations and prognosis, from disease resolution to fulminant hepatitis, viral persistence, and liver failure. In previous studies, virus-specific CD8⁺ and CD4⁺ T cell functions have been found to be critical in viral clearance and disease resolution (1). More recently, IL-17 production has been reported in hepatitis B and C infections (2, 3). In these cases, hepatic levels of IL-17 are significantly elevated in viral hepatitis, alcoholic liver disease, autoimmune hepatitis, and hepatocellular carcinoma, and correlate with the severity of disease (4). In previous work, we found that hepatic IL-17 produced early in Ad infection played a critical role in initiating successful antiviral CD8⁺ and CD4⁺ T cell responses (5). To date, the source of the liver-derived IL-17 species are not well understood, and their immune functions remain debatable (6–10).

IL-17 belongs to a family of cytokines consisting of IL-17A, B, C, D, E, and F (11). As a prototypical member of the family, IL-17A has been found to act as a potent inducer in T cell–mediated immune responses by activating and recruiting DCs, monocytes, and neutrophils in various tissues including the liver (5, 12, 13). It is known to be produced by the Th17 cells and several other cell types (14). Although IL-17 is typically associated with destructive tissue damage in autoimmune diseases and bacterial infections (15–17), more recently, it was found to promote Th1 and CTL responses in antitumor immunity (18), inflammatory bowel disease (19), and antiviral immune responses (5). Among IL-17 family members, IL-17A and IL-17F share the highest sequence homology and similar biological functions. Both cytokines bind to the same heterodimeric receptor molecule comprised by the IL-17RA and IL-17RC chains (11). Although IL-17F is known to be a weaker inducer of proinflammatory responses and is produced by a wider range of cell types (20), its cellular source, functions, and gene regulation in the liver, particularly during the T cell priming phase of viral hepatitis, are not well understood.

Ad is an important pathogen and one of the preferred vectors for gene and cancer therapy, and experimental vaccines for human immunodeficiency and hepatitis C viruses (21, 22). Lymphocytic choriomeningitis virus (LCMV) is a prototypical virus used in animal models of acute and persistent hepatitis (23). These viruses target the liver when given i.v. and can induce strong innate immune responses, Th, CTL, and B cell responses against viruses (24–27). They are eliminated by innate immune mechanisms initially (28–31). In subsequent periods, virus elimination and liver pathology are mediated by cytotoxic and Th cells (1, 27). In previous work, we found that IL-17 was produced in the liver within the first day of Ad infection (5). This brief surge played a critical role in initiating full CD8⁺ and CD4⁺ T cell responses. Furthermore, this early IL-17A was produced by γδ T cells, along with another yet-to-be characterized population in the liver.

In this work, we found novel innate lymphoid cells (ILCs) as a major source of hepatic IL-17A and IL-17F production in addition to γδ T cells. In the last several years, ILCs have been found to be involved in innate immunity, as well as tissue remodeling (32). In the liver, ILCs have been shown to protect against acute hepatitis (33, 34) but can also mediate hepatic fibrosis (35). In addition, we found that IL-17 signaling was important for adaptive T cell responses in viral hepatitis. IL-17F engagement was crucial for effector T cell functions and antiviral responses. Moreover, IL-17A could negatively regulate IL-17F production. Collectively, this study unveiled a previously unknown source and cross talk between IL-17A and IL-17F in the liver, and may provide potential therapeutic approaches to target ILCs and IL-17 species in viral hepatitis.

Female C57BL/6 (B6) and Rag2^−/− mice were purchased from The Jackson Laboratory. Mice deficient in IL-17RA (IL-17R^−/− mice) were provided by Amgen. Mice deficient in γ δ T cells (γδ^−/−) were kindly provided by Dr. Tian Wang (University of Texas Medical Branch, Galveston, TX). IL-17A^−/− and IL-17F^−/− mice were reported previously (16, 36). Naive IL-17R^−/−, IL-17A^−/−, and IL-17F^−/− mice displayed normal liver function as wild-type mice. All mice were maintained and bred under specific pathogen-free conditions in the animal facility at the University of Texas Medical Branch. Eight- to 12-wk-old mice were used for all the experiments. All experiments were reviewed and approved by the Institutional Animal Care and Use Committees of the University of Texas Medical Branch. To induce hepatitis, we i.v. injected mice with 3 × 10⁹ PFU replication-deficient recombinant Ad carrying the LacZ gene (AdLacZ, purchased from Vector Development Laboratory of Baylor College of Medicine) as described previously (25). Mice were i.v. injected with 2 × 10⁶ PFU LCMV Clone 13 (a kind gift from Dr. Maria Salvato, University of Maryland) (37). Titration of LCMV was performed on Vero cell monolayers plated on 24-well plates, followed by the viral quantification of immunological focus assay (38). The Ab of LCMV was kindly provided by Dr. Robert Tesh from the University of Texas Medical Branch.

Recombinant mouse IL-17A, mouse IL-17F, mouse IL-6, and human TGF-β were purchased from Peprotech. All fluorochrome-labeled mAbs and their corresponding isotype controls were anti-mouse Abs and purchased from eBioscience (San Diego, CA), BD Pharmingen (San Diego, CA), and Biolegend (San Diego, CA). The following were purchased from eBioscience: PE-conjugated anti–IL-17A (eBio17B7), allophycocyanin-conjugated anti–IL-17A (eBio17B7), FITC-conjugated anti–IFN-γ (XMG1.2), allophycocyanin-conjugated anti–IFN-γ (XMG1.2), allophycocyanin-conjugated anti-TCRγδ (eBioGL3), PE-Cy7–conjugated anti-CD3 (17A2), Pacific blue–conjugated anti-CD4 (GK1.5), PerCp-Cy5.5–conjugated anti-CD8 (53-6.7), allophycocyanin-conjugated anti-NK1.1 (PK136), Alexa Fluor 488–conjugated anti-CD107a (1D4B), Alexa Fluor 488–conjugated anti-CD107b (ABL-93), FITC-conjugated anti-B220 (RA3-6B2), FITC-conjugated anti-CD11b (M1/70), FITC-conjugated anti-CD11c (N418), FITC-conjugated anti–Gr-1 (RB6-8C5), FITC-conjugated anti–Ter-119 (TER-119), FITC-conjugated anti-NK1.1 (PK136), FITC-conjugated anti-CD90.2 (30-H12), allophycocyanin-conjugated anti-RORγt (B2D), and biotin-conjugated anti–c-kit (2B8). Allophycocyanin-conjugated anti–IL-17F (O79-289) and purified anti-CD16/32 (2.4G2) were purchased from BD Pharmingen. PerCp-Cy5.5–conjugated anti-NKp46 (29A1.4), allophycocyanin-Cy7–conjugated anti-CD3 (17A2), allophycocyanin-Cy7–conjugated anti–Sca-1 (D7), and PerCp-Cy5.5–conjugated anti-ICOS (C398.4A) were purchased from Biolegend.

Liver specimens were fixed in 10% buffered formalin. Paraffin-embedded sections were stained with H&E for histological evaluation by using a modified Knodell scoring system (39). In brief, normal liver architecture without remarkable injury or cellular infiltration was scored as 0. A score of 1 represented limited infiltration of inflammatory cells in the portal triad without significant involvement in the lobular and pericentral regions. In addition to these pathological changes, a score of 2 reflected a moderate involvement in the portal areas, accompanied by isolated apoptosis and necrosis in the lobular and pericentral areas. A score of 3 involved extensive lymphocyte infiltration in the portal area with widespread apoptosis and bridging necrosis throughout the liver.

Intrahepatic lymphocytes (IHLs) were isolated according to our previous method with slight modifications (25). In brief, liver tissue was pressed and collected in complete RPMI 1640. After washing (300 × g, 10 min), cell suspensions were resuspended in complete RPMI 1640 containing collagenase IV (0.05%; Roche Applied Science, Indianapolis, IN) at 37°C for 30 min. After digestion, cell suspensions were passed through 70-μm nylon cell strainers to yield single-cell suspensions. Intrahepatic mononuclear cells were purified by centrifugation (400 × g) at room temperature for 30 min over a 30/70% discontinuous Percoll gradient (Sigma). The cells were collected from the interphase, thoroughly washed, and resuspended in complete RPMI 1640 containing 10% FBS (Hyclone, Logan, UT). The total numbers of IHLs per liver were counted. The relative percentages of CD4⁺, CD8⁺, and γδ T cells were measured by flow cytometry, and the absolute numbers of these lymphocyte subpopulations per liver were calculated according to their percentages and the total IHL numbers in each liver.

Intracellular staining was performed according to our previous methods (5). In brief, cells were incubated for 4 h with PMA (50 ng/ml) and ionomycin (750 ng/ml). For the simultaneous detection of surface CD107a/b (lysosomal-associated membrane proteins 1/2) and intracellular cytokines, cells were stimulated by plate-coated anti-CD3 mAb (145-2C11, 10 μg/ml; eBioscience) for 4 h, in the presence of GolgiStop (BD Biosciences). After incubation, cells were collected and blocked with FcγR blocker (CD16/32) and stained for specific surface molecules. After surface staining, cells were fixed, permeabilized, and stained for intracellular cytokines by using a fixation/permeabilization kit (eBioscience).

Murine lymphocytes were blocked with anti-CD16/CD32 (eBioscience) and stained with fluorochrome-labeled Abs, and then processed on an LSRII FACSFortessa (Becton Dickinson, San Jose, CA) and analyzed by using FlowJo software (Tree Star, Ashland, OR). All fluorochrome-labeled mAbs and their corresponding isotype controls were purchased from BD Pharmingen (San Diego, CA) and eBioscience (San Diego, CA).

To polarize IL-17–producing cells, we cultured total splenocytes on the precoated anti-CD3 Ab (5 μg/ml) with 10 ng/ml TGF-β, 20 ng/ml IL-6, 10 μg/ml anti–IFN-γ, and 10 μg/ml anti–IL-4. After 4 d of culture, cells were rested for 1 d and then stimulated with PMA (50 ng/ml) and ionomycin (750 ng/ml). Cells were stained with fluorescence-labeled anti-mouse surface Abs, including anti-CD3, anti-CD4, anti-CD8, and anti-TCRγδ. After surface staining, cells were fixed, permeabilized, and counterstained with fluorescence-labeled Abs for IL-17A and IL-17F.

Frozen liver tissues were used to extract genomic DNA and total RNA. DNA was extracted with a DNeasy blood and tissue kit (Qiagen), and total RNA was extracted with an RNeasy Mini kit (Qiagen) and digested with DNase I (Ambion). The concentrations of DNA and RNA were measured by using a spectrophotometer (Eppendorf). For relative quantitation of the cytokine and chemokine mRNA levels, cDNA was prepared from 1 μg RNA by using an iScript Reverse Transcription Kit (Bio-Rad), and 4 μl of the cDNA was amplified in a 25-μl reaction mixture containing 12.5 μl iQ SYBR Green Supermix (Bio-Rad) and 0.9 μM each of gene-specific forward and reverse primers. The PCR assays were denatured for 10 min at 95°C, followed by 40 cycles of 15 s at 95°C and 60 s at 60°C. The PCR was performed with the CFX96 Touch real-time PCR detection system (Bio-Rad). Relative quantitation of mRNA expression was calculated as the fold increase in expression by using the 2^−ΔΔCt method. Meanwhile, melting curve analysis was used to check the specificity of the amplification reaction. The sequences of the forward and reverse gene-specific primers used are as follows: GAPDH—forward 5′-TGGAAAGCTGTGGCGTGAT-3′ and reverse 5′-TGCTTCACCACCTTCTTGAT-3′; IFN-γ—forward 5′-ATGAACGCTACACACTGCATC-3′ and reverse 5′-CCATCCTTTTGCCAGTTCCTC-3′; TNF-α—forward 5′-CCCTCACACTCAGATCATCTTCT-3′ and reverse 5′-CTTTGAGATCCATGCCGTTG-3′; CXCL9—forward 5′-GGAGTTCGAGGAACCCTAGTG-3′ and reverse 5′-GGGATTTGTAGTGGATCGTGC-3′; CXCL10—forward 5′-CCAAGTGCTGCCGTCATTTTC-3′ and reverse 5′-GGCTCGCAGGGATGATTTCAA-3′; Hexon—forward 5′-GAGCCAGCATTAAGTTTGATAGCA-3′ and reverse 5′-AGATAGTCGTTAAAGGACTGGTCGTT-3′.

Liver proteins were extracted from frozen liver tissues by homogenization on ice in the RIPA buffer (Cell Signaling) with a protease inhibitor mixture (Sigma). After centrifugation at 20,000 × g for 15 min, the supernatant was collected and protein concentration was measured with a protein assay kit (Bio-Rad). Equal amounts of the liver proteins (100 μg) were loaded for ELISA assays. The levels of IL-17A and IL-17F in the liver proteins were measured by using the ELISA kits (eBioscience) according to the manufacturer’s instructions. Detection limits were 4 pg/ml for IL-17A and 15 pg/ml for IL-17F, respectively.

The difference between the two different groups was determined by using Student t test. One-way ANOVA was used for multiple-group comparisons (GraphPad Software v4.0). The p values <0.05 were considered significant and <0.01 as highly significant.

Early IL-17A production by intrahepatic γδ T cells is known to be important for adaptive immune responses in Ad-induced hepatitis (5). IL-17F, the closest homolog to IL-17A among members of the IL-17 cytokine family, has partially concordant expression with and shares the same receptor with IL-17A (11). However, the precise role of IL-17F in viral hepatitis is still not well understood. To define the dynamics of IL-17A and IL-17F production in the course of Ad infection, we i.v. injected B6 mice with 3 × 10⁹ PFU AdLacZ. The animals were sacrificed at 0 h, 12 h, 24 h, 3 d, and 6 d postinfection. ELISA analysis of liver lysates revealed a significant accumulation of IL-17A and IL-17F during the first 24 h postinfection (Fig. 1A). Meanwhile, we isolated the IHLs and analyzed their intracellular levels of IL-17A and IL-17F by flow cytometry. We found that the IL-17A⁺ IL-17F⁻ cells expanded from 1.0% at 0 h to 2.0% at 24 h postinfection (Fig. 1B). The absolute cell number of IL-17A⁺ IL-17F⁻ cells expanded from 2.2 × 10⁴ cells at 0 h to 8.9 × 10⁴ cells at 24 h postinfection (Fig. 1C). In addition, the IL-17A⁺ IL-17F⁺ cells expanded robustly, from 0.2% at 0 h to 0.5% at 24 h postinfection (Fig. 1B). The absolute cell number of IL-17A⁺ IL-17F⁺ cells expanded from 0.4 × 10⁴ cells at 0 h to 2.0 × 10⁴ cells at 24 h postinfection (Fig. 1C). IL-17A– or IL-17F–producing cells did not expand in the spleen (Supplemental Fig. 1), which suggested to us that this surge of IL-17A⁺ or IL-17F⁺ cells was liver specific. In addition, the increases of IL-17⁺ cells in the liver were accompanied by a surge of IL-17 in the serum as we reported previously (5).

Both innate and acquired T cells were reported to produce IL-17A and IL-17F (14). Although most IL-17⁺ T cells in the liver were indeed γδ T cells at 24 h postinfection (Fig. 1D), small populations of IL-17⁺ γδ⁻ IHLs were heterogeneous. We therefore characterized the remaining IL-17 producers among IHLs (Fig. 1D). To our surprise, in the CD3⁻ population, the IL-17A producers did not express lineage markers, such as CD8, CD11b, CD11c, NK1.1, B220, Gr-1, and Ter-119. Further study showed that similar cells from a lineage-negative population in the liver produced IL-17F (Supplemental Fig. 2A). Collectively, these data suggested that early surges of IL-17A/F were produced by both classical γδ T cells and nonclassical ILCs.

To further characterize these lineage-negative cells that produced IL-17A and IL-17F, we analyzed their surface markers. Interestingly, these cells expressed high levels of CD90, Sca-1, ICOS, c-Kit, RORγt, and IL-7Rα, but low levels of NKp46 and CD4 (Fig. 2A). The phenotypical and functional characteristics of these cells were consistent with group 3 ILCs (ILC3s), originally described in the gastrointestinal tract (40). Based on the surface markers of these innate cells in the liver, we further concluded that most of these ILC3s belonged to the NKp46⁻ ILC3 population. In addition, they expanded robustly and peaked within the first 24 h post Ad-infection and waned in the next few days (Fig. 2B). The remaining IL-17A⁺ cells were comprised of double-negative, CD4⁺ and CD8⁺ T cells (Supplemental Fig. 2C). Th17 cells contributed only small amounts of IL-17 in this model (Supplemental Fig. 2C). Interestingly, similar populations of IHLs produced IL-17F post Ad infection (Supplemental Fig. 2A–C). To examine whether intrahepatic ILC3s are present in a persistent viral infection model, we i.v. injected B6 mice with 2 × 10⁶ PFU LCMV Clone 13. Similar to our previous findings in an Ad model, nearly all CD3⁻ IL-17⁺ cells did not express lineage markers, including CD8, CD11b, CD11c, NK1.1, B220, Gr-1, and Ter-119 (Fig. 2C). These results indicated that, in addition to Ad infection, ILC3 secreting IL-17 was also present in the liver after LCMV infection.

To further confirm that ILC3s in the liver can produce the IL-17 species, we examined the IL-17A/F levels in the mice deficient of T and B lymphocytes (41). IHLs from uninfected Rag2^−/− and wild-type mice were analyzed by flow cytometry. In the wild-type animals, there were ∼0.8% IL-17A⁺ cells, 0.2% IL-17F⁺ cells, and 0.2% IL-17A⁺ IL-17F⁺ cells (Fig. 2D). In Rag2^−/− mice, however, there were >2-fold and 10-fold increases of IL-17A⁺ and IL-17A⁺ IL-17F⁺ cells, respectively, among the total IHLs. These results were confirmed in the lineage-negative IHLs as well (Fig. 2D). Given that ILC3s and γδ T cells were important sources of early IL-17A/F production in the liver, we investigated the role of ILC3-derived IL-17A/F in the outcome of Ad-induced hepatitis using γδ^−/− mice. Surprisingly, IL-17A production in γδ^−/− mice was comparable with that in the wild-type animals at day 1 postinfection (Fig. 3A). The γδ⁻ IL-17A⁺ cells in γδ^−/− mice increased >3-fold compared with those in controls (Fig. 3A). Further study revealed that these γδ⁻ IL-17A⁺ cells were composed by ILC3s and double-negative T cells (data not shown). At day 6 postinfection, we found Ad-infected γδ^−/− mice displayed comparable IHL infiltration and Th1/CTL functions (Fig. 3B, 3C). In addition, γδ^−/− mice presented similar liver inflammation and pathological scores compared with the control animals (Fig. 3B). Taken together, these results indicated that ILC3 in the liver can secrete the IL-17 species, and hence is an important source of IL-17 in the liver.

Ad infection induced strong CD4⁺ T cell and CD8⁺ T cell recruitment into the liver. To determine whether this early surge of IL-17A and IL-17F signals through IL-17R and mediates T cell responses, we injected the wild-type and IL-17R^−/− mice with 3 × 10⁹ PFU AdLacZ. At day 6 postinfection, we examined the accumulation of total IHLs, CD4⁺ T cells, and CD8⁺ T cells in these animals. Compared with the control animals, IL-17R^−/− mice developed significantly less total lymphocyte infiltration, with reduced numbers of CD4⁺ and CD8⁺ T cells (Fig. 4A). A lack of IL-17 signaling significantly reduced IFN-γ–producing ability among CD8⁺ T cells in the liver (Fig. 4B). In addition, considerable IFN-γ⁺ cells (14 ± 0.5%) in the control mice expressed lysosomal-associated membrane proteins 1/2 (CD107a/b), indicative of their ability to degranulate cytolytic vesicles. In the IL-17R^−/− animals, because of the reduced IFN-γ–producing ability of intrahepatic CD8⁺ T cells, significantly lower percentages (9 ± 2%) and fewer numbers of these cells are IFN-γ⁺ CD107a/b⁺ ones (Fig. 4B). As found with CD8⁺ T cells, lower percentages and fewer numbers of CD4⁺ T cells expressed IFN-γ (Fig. 4C). The presence or absence of IL-17 signaling did not significantly change the percentages of regulatory T cells in the liver (Fig. 4C). Because of the decrease in the total number of IHLs in IL-17R^−/− mice, there were fewer numbers of regulatory T cells in these animals (Fig. 4C).

To determine the effect of IL-17 signaling on disease outcomes, we examined the pathology of Ad-infected IL-17R^−/− mice. The control animals experienced development of hepatitis characterized by inflammatory infiltration, hepatocytes with megaloblastic changes, and single-cell necrosis at 6 d postinfection (Fig. 5A). Compared with the wild-type mice, the Ad-infected IL-17R^−/− mice displayed much milder inflammation and lower pathological scores (Fig. 5), as well as lower serum alanine aminotransferase (ALT) and aspartate aminotransferase (AST) levels (Fig. 5B). Taken together, the lack of IL-17 signaling impaired the recruitment and functions of intrahepatic Th1/CTL responses and ameliorated Ad-induced hepatitis.

IL-17A and IL-17F bind to the same heterodimeric receptor composed of IL-17RA and IL-17RC (11). Having demonstrated that IL-17R^−/− mice displayed greatly reduced hepatic inflammation after Ad inoculation, we investigated which ligand was responsible for IL-17R signaling and immune-mediated liver injury. We injected IL-17A^−/−, IL-17F^−/−, and control mice with 3 × 10⁹ PFU AdLacZ. Surprisingly, IL-17A^−/− mice developed serum ALT and IHL infiltration comparable with those in the wild-type mice (Fig. 6A, 6B). However, the IL-17F^−/− mice displayed significantly reduced liver injury and lower pathological scores (Fig. 6A, 6B). Likewise, they presented lower serum ALT levels and fewer numbers of infiltrating IHLs (Fig. 6B). Consistent with the liver inflammation and infiltrated lymphocytes, lower percentages of CD8⁺ T cells in IL-17F^−/− mice expressed IFN-γ compared with the findings in the control animals (Fig. 6C). However, those in the IL-17A^−/− mice produced similar levels of IFN-γ compared with the controls (Fig. 6C). Furthermore, liver proinflammatory cytokines and chemokines, including IFN-γ, TNF-α, CXCL9, and CXCL10, were significantly decreased in IL-17F^−/− mice, but not in IL-17A^−/− mice, compared with those in wild-type animals (Fig. 6D).

By using quantitative real-time PCR analysis, we found no significant difference in the viral copy numbers among the wild-type, IL-17R^−/−, IL-17A^−/−, and IL-17F^−/− groups on day 7 post Ad infection (p > 0.05; Supplemental Fig. 3). Although there was a steady reduction of the viral genome in all groups, no statistical difference was found among these mice on day 14, as well as on day 21. After i.v. injection of Ad in mice, a majority of the viruses was eliminated quickly by the innate immune mechanisms within 24 h (28). However, overzealous T cell responses may result in increased necroinflammatory hepatitis without accelerating viral elimination in vivo (5, 25, 31, 33). These results suggested to us that IL-17A/F signaling affected lymphocyte infiltration and hepatic inflammation, rather than viral clearance in the liver in Ad-induced acute hepatitis. Thus, further investigations in chronic infection models, such as LCMV infection, are needed to define the role of IL-17 in virus clearance.

IL-17A^−/− mice developed Ad-induced hepatitis similar to that of the wild-type animals (Fig. 6). We speculated that IL-17F could compensate for IL-17A deficiency in these animals. To test this hypothesis, we infected the IL-17A^−/−, IL-17F^−/−, and control mice with AdLacZ, and compared their IL-17 levels at 24 h postinfection. Interestingly, in IL-17A^−/− mice, IL-17F⁺ γδ T cells increased >3-fold compared with those in controls (Fig. 7A). In IL-17F^−/− mice, however, there was no such increase among IL-17A⁺ cells (Fig. 7B). This IL-17F increase was also observed in the uninfected IL-17A^−/− mice (Supplemental Fig. 4). Furthermore, IL-17R^−/− mice seemed to produce more IL-17A and IL-17F than wild-type animals in both uninfected and infected animals (data not shown). Consistent with an earlier report (42, 43), our results suggest that IL-17A has a strong negative feedback loop, repressing its own production and that of IL-17F through IL-17R.

To test whether IL-17A can inhibit IL-17F production in vitro, we isolated the naive splenocytes and cultured them under the Th17 differentiation conditions in the presence of rIL-17A and rIL-17F, respectively. We found that IL-17A significantly suppressed IL-17F production in vitro in CD4⁺, CD8⁺, γδ⁺ T cells, and total splenocytes in a dose-dependent manner (Fig. 7C). However, IL-17F did not affect IL-17A production from these cells (Fig. 7D).

ILCs are essential effectors of innate immunity and have an important role in tissue remodeling (32). They are characterized by the absence of lineage markers, as well as lymphoid morphology. Recently, ILCs have been categorized into three groups based on their cytokines and transcriptional factors (32). Group 1 comprises ILCs that produce IFN-γ. Group 2 consists of ILCs that produce type 2 cytokines (including IL-5 and IL-13). Group 3 includes ILC subsets that produce IL-17 and/or IL-22 and depend on the transcriptional factor RORγt for their development and function. ILC3s also have been shown to play important roles in intestinal immunity and homeostasis (44, 45). However, the phenotype and role of ILC3s in the liver is unclear. In this study, we found that several intrahepatic cell populations secreted IL-17A and IL-17F locally shortly after Ad infection (Fig. 1). Among these cells, γδ T cells were highest in number, as we reported previously (5). Among γδ⁻ cells, we revealed that a previously uncharacterized population constituted a major group of IL-17 producers. These lineage marker–negative, stem cell marker–positive cells belong to ILC3s (Figs. 1, 2). Based on their CD4 and NKp46 expression profiles, we further speculate that most of these ILC3s belonged to the NKp46⁻ ILC3 population, and a small population belonged to the CD4⁺ lymphoid-tissue inducer cells in the liver (Fig. 2). Using Rag2^−/− and γδ^−/− mice, we further confirmed that ILC3s are indeed potent IL-17 producers and an integral part of the immune defense system in the liver (Figs. 2, 3). Moreover, in addition to the Ad model, we also observed that ILC3s were present in mouse liver after LCMV infection (Fig. 2C). To the best of our knowledge, this is the first report for detailed study of ILC3 subsets in the liver post viral infection.

IL-17A and IL-17F belong to the IL-17 superfamily. They were originally reported to be predominantly produced by activated Th17 along with several other cell types (e.g., CD8⁺ T cells, γδ cells, NKT cells, double-negative T cells) (14, 20). They were typically linked to destructive tissue damage in autoimmune diseases and bacterial infections (15–17). More recent evidence has pointed to their involvement in promoting Th1 and CTL responses in antitumor immunity (18), inflammatory bowel disease (19), and antiviral immune responses (5). Both IL-17A and IL-17F have been found to be required for immune responses against extracellular bacterium such as Staphylococcus aureus infection (17, 20). Presently, the immunoregulatory effects of the IL-17 species on the Ag presentation process, CTL, and Th responses in virus-infected liver are not well understood. In contrast, in the presence of IL-17A, IL-17F was thought to be dispensable for disease progression in experimental allergic encephalomyelitis and collagen-induced arthritis (16, 46). The roles of IL-17A and IL-17F appear to be more controversial as either proinflammatory or anti-inflammatory mediators in inflammatory bowel disease (47, 48).

IL-17–producing cells are known to play a role in autoimmune and viral hepatitis (3, 49, 50). Several studies showed that Th17 cells could promote the activation of stellate cells and Kupffer cells, which, in turn, may aggravate liver fibrosis and the inflammatory response in chronic hepatitis (51, 52). There is a paucity of data addressing the possible involvement in DC activation and T cell priming. In a Con A–induced hepatitis model, overexpression of IL-17A resulted in massive hepatocyte necrosis, and anti–IL-17A blockage significantly ameliorated the disease (10). In addition, Lafdil et al. (8) showed that liver injury was alleviated in Con A–induced hepatitis among IL-17–deficient mice. In a separate study, however, IL-17A deficiency did not appear to thwart T cell activation and liver inflammation (6). We speculate that these discrepancies are attributable to the compensatory IL-17F production, as we showed in this report (Figs. 6, 7). Second, Con A–induced liver injury is an extremely acute hepatitis model, in which lectin-activated NKT cells play a critical role (53). Also, the dose of Con A injection and the timing of liver injury assessment may also contribute to the discrepancies among these studies. In this study, we report that adaptive T cell responses and associated liver injury were dependent on IL-17 signaling (Figs. 4, 5). Surprisingly, these clinical parameters were not affected by the lack of IL-17A in the gene knockout animals in this study (Fig. 6). However, in our previous observation, IL-17A neutralization by mAb clearly hampered DC activation and alleviated liver inflammation upon viral infection (5). Additional experiments revealed that IL-17A^−/− mice displayed a >3-fold increase in IL-17F⁺ cells post viral infection (Fig. 7A). Interestingly, this compensatory IL-17F increase was also observed in uninfected IL-17A^−/− animals (Supplemental Fig. 4). Finally, in vitro experiments directly confirmed that IL-17A represses IL-17F secretion in γδ, CD4⁺, and CD8⁺ T cells (Fig. 7C). These results are consistent with previous reports that IL-17A controls IL-17F production through an IL-17R–dependent, short-loop inhibition mechanism (42, 43). In the absence of IL-17A, however, IL-17F could compensate and maintain baseline neutrophil counts in mice (42). In a recent commentary, it was proposed that IL-17A and IL-17F can cause negative feedback of their own and each other’s synthesis via IL-17R (54). However, we did not observe IL-17F–mediated IL-17A inhibition in our studies (Fig. 7D). In addition, in IL-17F^−/− animals, there was no compensatory IL-17A increase or rescue of T cell functions (Figs. 6, 7B). Although IL-17A and IL-17F have some functional redundancy in viral infection, our results unveil a mechanism underlying the seeming discrepancies between IL-17A– and IL-17F–deficient mice, and underscore the unique functions of IL-17F in T cell responses to viral infection in the liver.

Recombinant Ad is one of the preferred vectors for gene therapy, cancer therapy, and experimental vaccines (21, 22). However, it can also induce strong Th, CTL, and B cell responses against the viral vector and the transgene (24, 25). At day 1 post Ad infection, we observed an elevation of IFN-β, IL-7, IL-23, and TNF-α in addition to IL-17 (5). Although the IL-17–producing cell expansion and IL-17 level increase were relatively brief (Fig. 1A, 1B), the IFN-γ, TNF-α, IL-1β, CXCL9, and CXCL10 levels continued to persist into day 6 postinfection (Fig. 6D) (5). In this report, we found that IL-17 signaling blockage in IL-17RA knockout mice developed less CD4⁺ and CD8⁺ T cell infiltration and displayed much milder liver inflammation and ALT and AST elevations (Figs. 4, 5). Our results indicate that blockade of IL-17/IL-17RA signaling pathway may represent a novel therapeutic intervention to constrain liver inflammation when using Ad for gene therapy.

In summary, we have defined the early source and function of hepatic IL-17, which are important for DC activation and T cell priming in viral hepatitis. We have provided evidence that besides γδ T cell, ILC3 seemed a significant source of IL-17A/F in the liver within the first few hours and days of viral infection using Ad and LCMV models. Furthermore, this surge of IL-17 mediated DC licensing and adaptive immune responses through binding to IL-17R. Also, although we know that IL-17A is a dominant species constituting the cytokine microenvironment and regulating IL-17F production, in its absence, IL-17F⁺ IHLs expanded significantly and compensated for IL-17A deficiency in an IL-17 signaling pathway-dependent fashion. In contrast, IL-17F deficiency resulted in compromised T cell priming and tissue infiltration. Collectively, this study indicates that innate IL-17A/F signaling is important for adaptive immune responses in viral hepatitis. Our study unveiled a previously unknown source and cross talk between IL-17A and IL-17F, and may provide potentially important information aimed at targeting ILCs and IL-17 species in acute and chronic viral hepatitis.

We thank Amgen, Inc. and Dr. Tian Wang for providing the IL-17R^−/− and γδ^−/− breeding mouse pairs, respectively. We thank Drs. Maria Salvato and Robert Tesh for providing LCMV and anti-LCMV Ab, respectively. We thank Dr. Hui Wang, Dr. Yiqun Xiong, Houpu Liu, David Vu, and Yixiao Sun for excellent technical support and Mardelle Susman for assistance with manuscript preparation.

The authors have no financial conflicts of interest.