Potential conflict of interest: Nothing to report.
Supported by the Natural Science Foundation of China (#91029303, #30911120480, and #31021061).
Acetaminophen overdose causes acute liver inflammation with neutrophil infiltration; however, the mechanism of damage-associated inflammation has not been elucidated. In this study we found that the HMGB1-TLR4-IL-23-IL-17A axis played a crucial role in acetaminophen-induced infiltration of neutrophils and liver injury. Notably, interleukin (IL)-17A and IL-23 significantly increased after acetaminophen challenge. A neutralizing antibody against IL-17A attenuated the recruitment of neutrophils, accompanied by reduced liver injury. Only IL-17A+CD3+γδ T cell receptor (TCR)+ cells were significantly increased in the liver, and depletion of γδ T cells, but not CD4+ T cells or natural killer (NK)T cells significantly reduced IL-17A production, attenuated liver injury, and decreased the number of neutrophils in the liver. Furthermore, a neutralizing IL-23 p19 antibody or p40-deficiency significantly decreased the levels of IL-17A and infiltration of neutrophils. After in vitro stimulation, the percentage of IL-17A-producing γδ T cells and the levels of supernatant IL-17A from total hepatic lymphocytes or purified γδ T cells markedly increased in the presence with IL-23. Importantly, IL-23 and IL-17A were reduced after inhibition of macrophages and could not be induced in Toll-like receptor TLR4−/− mice after acetaminophen challenge. Meanwhile, serum high-mobility group box 1 (HMGB1), a damage-associated molecule released from necrotic hepatocytes, increased after acetaminophen challenge, and the HMGB1 inhibitor glycyrrhizin markedly reduced the production of IL-23 and IL-17A and the recruitment of hepatic neutrophils. HMGB1 stimulated the production of IL-23 by TLR4+/+ but not by TLR4−/− macrophages. Conclusion: The HMGB1-TLR4-IL-23 pathway in macrophages makes the generation of IL-17-producing γδ T cells, which mediates neutrophil infiltration and damage-induced liver inflammation. (HEPATOLOGY 2013)
Acetaminophen is usually used as an over- the-counter analgesic and antipyretic drug. However, acetaminophen overdose has become a frequent cause of intentional or accidental death in many countries.1, 2 Acetaminophen is metabolized by hepatic CYP2E1 into the toxic intermediate N-acetyl-p-benzoquinone-imine, which is then detoxified by hepatic glutathione. However, excessive N-acetyl-p-benzoquinone-imine consumes hepatic glutathione and covalently binds cellular proteins, resulting in hepatocyte necrosis.3, 4 Because the innate immune response following hepatocyte necrosis has been noted to cause a second wave of liver destruction,5, 6 the overall progression is now described by a “two-hit” model.7 Natural killer (NK) and natural killer T (NKT) cells have been reported to play a pathogenic role in the progression of acetaminophen-induced liver injury by up-regulating Fas ligand and secreting interferon (IFN)-γ. Depletion of NK/NKT cells significantly ameliorates liver injury.8 However, Masson et al.9 revealed that the role of NK and NKT cells in these studies was dependent on dimethyl sulfoxide (DMSO), the solvent used to dissolve acetaminophen in the experiments. That group found that low levels of DMSO could recruit NKT cells to the liver and activate NK and NKT cells. In the absence of DMSO, NK and NKT cells did not produce IFN-γ after acetaminophen challenge, and depletion of these cells did not protect mice from acetaminophen-induced liver injury. However, increasing evidence has demonstrated that the innate immune response does participate in the pathogenesis of acetaminophen-induced injury, even in the absence of DMSO. Thus, understanding the critical immune cells and cytokines that mediate acetaminophen-induced liver injury is important. Imaeda et al.10 demonstrated that DNA from apoptotic hepatocytes can up-regulate the transcription of pro-interleukin (IL)-1β and pro-IL-18 in liver sinusoidal endothelium in a Toll-like receptor (TLR)9- and Nalp3 inflammasome-dependent manner. However, further investigation of the immune cells involved in this type of liver injury following elevated production of IL-1β and IL-18 has not been documented.
A substantial number of neutrophils infiltrate the liver after acetaminophen challenge.11, 12 Neutrophils play an important role in acetaminophen-induced liver injury, although controversy exists regarding their precise contributions.13 In addition, IL-1β has been reported to be dispensable in the recruitment of neutrophils into the liver and to play a protective role in the liver injury.14, 15 The mechanism by which neutrophils infiltrate the liver remains unclear. IL-17A was discovered by Rouvier et al.16 and was named cytotoxic T lymphocyte-associated serine esterase-8. T helper (Th)17 cells are recognized as the primary source of IL-17A.17 However, additional innate immune cell populations have been shown to secrete IL-17A, including γδ T cells, NK cells, NKT cells, and neutrophils.18 The receptor for IL-17A is expressed on various types of cells, such as endothelial cells, macrophages, and stromal cells. These cells produce diverse proinflammatory cytokines and chemokines in response to IL-17 to mediate inflammation and induce granulopoiesis and neutrophil recruitment to inflammatory sites.19 γδ T cells are a component of the innate immune cell population and play important roles during physiological processes, such as defense against pathogens, tumor surveillance, and regulation of immune responses through cytokine production (IFN-γ, IL-4, IL-10, TGF-β, or IL-17A).20 Unlike conventional αβ T cells, IFN-γ- or IL-17A-producing-γδ T cells are stably divided into two subsets during development in the thymus.21 Recent studies have demonstrated that γδ T cells play an important role in infectious and autoimmune diseases in an IL-17A-dependent manner. IL-17A-producing γδ T cells protect against Listeria monocytogenes infection in the murine liver22 and are pathogenic in collagen-induced arthritis.23 However, in the progression of acetaminophen-induced liver injury, whether γδ T cells produce IL-17A, how γδ T cells would produce IL-17A, and whether IL-17A induces neutrophil recruitment and expansion have not been investigated.
Necrotic hepatocytes release many types of damage-associated molecular pattern molecules (DAMPs), such as high-mobility group box 1 (HMGB1), heat shock proteins, DNA, and cyclophilin A.10, 24, 25 Extracellular HMGB1 acts through multiple receptors, including TLR2, TLR4, TLR9, and the receptor for advanced glycation end products.26 Many cell populations, such as macrophages and endothelial cells, can respond to stimulation with HMGB1.27 HMGB1 has been shown to play an important role in acetaminophen-induced liver injury.28 Blocking HMGB1 with monoclonal antibodies (mAbs) attenuates liver injury.29 In addition to acetaminophen-induced liver injury, HMGB1 also contributes to other liver diseases. HMGB1 mediates hepatic ischemia-reperfusion injury through a TLR4-dependent pathway. Inhibition of HMGB1 or TLR4 deficiency can protect mice from ischemia-reperfusion-caused liver injury.30 HMGB1 released from acetaminophen-induced necrotic hepatocytes has been demonstrated to be able to induce macrophage activation and cytokine release.28, 31 However, the cellular and molecular immune mechanisms of HMGB1 in liver injury, particularly tissue damage-induced pathogenic inflammation, remain elusive. How the specific innate immune receptor for HMGB1 activates macrophages and induces proinflammatory cytokines and how these cytokines prime the subsequent innate immune response and mediate inflammation are completely unclear.
In this study, we demonstrated that the interaction between macrophages and γδ T cells plays an important role in acetaminophen-induced liver inflammation. Serum HMGB1 released from necrotic hepatocytes stimulates the production of IL-23 by hepatic macrophages in a TLR-4-dependent manner, and IL-23 aids in the generation of IL-17A-producing γδ T cells in the liver. IL-17A secreted by γδ T cells then recruits hepatic neutrophils. Thus, the HMGB-TLR4-IL-23-IL17A axis between macrophages and γδ T cells contributes to the accumulation of neutrophils and liver inflammation.
C57BL/6 males aged 6-8 weeks were purchased from the Shanghai Laboratory Animal Center, Chinese Academy Sciences (Shanghai, China). T cell receptor (TCR)δ−/−mice, IL-23p40−/− mice, and TLR4−/− mice were kindly provided by professors Z.N. Yin (Nankai University), Z.X. Lian (University of Science and Technology of China), and S.B. Su (Sun Yat-Sen University), respectively. All mice were housed in microisolator cages under humidity- and temperature-controlled specific pathogen-free conditions in the animal facility of the School of Life of the University of Science and Technology of China. The mice were maintained on an irradiated sterile diet and provided autoclaved water. All experiments were performed according to the guidelines outlined in the Guide for the Care and Use of Laboratory Animals (NIH, Bethesda, MD).
Treatment of Mice.
Fresh acetaminophen (Sigma-Aldrich, USA) solution was prepared for each experiment by dissolving acetaminophen to a concentration of 10 mg/mL in phosphate-buffered saline (PBS) warmed to 40°C. Mice were fasted for 16 hours and then injected intraperitoneally with PBS or acetaminophen at 400 mg/kg (body weight). At the indicated timepoints, sera were collected and stored at −20°C for measuring alanine aminotransferase (ALT), total bilirubin, and cytokines. At the end of the experiment, the mice were anesthetized, bled, and euthanized. The livers were excised and portions were snap-frozen in liquid nitrogen and stored at −80°C for RNA isolation, portions fixed in 4% paraformaldehyde for tissue sections stained with hematoxylin/eosin (H&E). For neutralization of endogenous IL-17A or IL-23, 0.2 mg of neutralizing rabbit antimouse IL-17A (Clone TC11-18H10.1, BioLegend, USA) or neutralizing rabbit antimouse IL-123p19 (Clone G23-8, eBioscience, USA) was administered intravenously at the time of acetaminophen treatment. Control rabbit IgG was used as an isotype control. For deletion of γδ T cells, NK1.1+ cells, or CD4+ cells, the mice were injected intravenously with 0.5 mg of an anti-γδ TCR mAb (clone TIB-207, ATCC, Manassas, VA), anti-NK1.1 mAb (clone HB191, ATCC), or anti-CD4 mAb (clone TIB-207, ATCC), respectively, 48 hours before acetaminophen treatment. For inhibition of macrophages, mice were injected intravenously with GdCl3 at 20 mg/kg (body weight, Sigma-Aldrich) at 24 hours before acetaminophen treatment. For inhibition of HMGB1, mice were treated with glycyrrhizin (TCI, Shanghai, China) at 5 mg/mouse 1 hour before acetaminophen treatment.
Assessment of Liver Injury.
Acute liver injury was evaluated by serum levels of ALT and total bilirubin. They were measured using diagnostic kits (Rongsheng, Shanghai, China).
RNA Isolation and Quantitative Reverse-Transcription Polymerase Chain Reaction (RT-PCR) Analysis.
Total RNA was isolated from frozen liver tissue using total RNA purification solutions (Invitrogen, USA). Two μg of total RNA was reverse-transcribed at 25°C for 15 minutes, 42°C for 50 minutes, and 70°C for 10 minutes using reverse transcription kits (Sangon Biotech, Shanghai, China). Complementary DNA (cDNA) fragments were amplified using the following gene-specific primers: IL-17A (sense 5-GCTCCAGA AGGCCCTCAG-3; antisense 5-CTTTCCCTCGCA TTGACA-3); IL-23p19 (sense 5-AGCGGGACATAT GAATCTACTAAGAGA-3; antisense 5-TCCTAGT AGG GAGGTGTGAAGTTG-3); IL-23p40 (sense 5-TCCACCAAACTCCCCAGACA-3; antisense 5-CTG TGCATGCTCTTTGGTTGAT-3); and β-actin (sense 5-TGGAATCCTGTGGCATCCATGAAA-3; antisense 5-TAAAACGCAGCTCAGTAACAGTCC-3). Quantitative RT-PCR was performed to measure the messenger RNA (mRNA) expression of IL-17A, IL-23p19, and IL-23p40 using commercially available SYBR Premix Ex Taq (TaKaRa Biotechnology, Dalian, China) and specific primers in a reaction with an optimal number of cycles at 95°C for 10 seconds, then 60°C for 30 seconds in a Corbett Rotor-Gene 3000 real-time PCR system (Corbett Research). The gene expression levels were calculated relative to the housekeeping gene β-actin.
Liver specimens from mice exposed to different treatments were fixed in 4% paraformaldehyde, dehydrated with a graded series of alcohol, and embedded in paraffin. Six-micron tissue sections were prepared and stained with H&E.
Measurement of IL-17A, IL-23, and HMGB1 in Sera or Supernatant.
At each indicated timepoint, sera were harvested for measurement of IL-17A, IL-23, IL-23p40, and HMGB1. Hepatic mononuclear cells were stimulated in vitro with IL-1β (50 ng/mL, PeproTech, USA), IL-23 (50 ng/mL, Miltenyi Biotec, USA) or the combination for 48 hours. To purify γδ T cells, hepatic mononuclear cells were incubated with an FITC-conjugated anti-γδ TCR antibody (Clone GL3, eBioscience). Then the labeled cells were washed and incubated with anti-FITC-conjugated magnetic beads (Miltenyi Biotec). Positive cells were sorted using columns and a MACS kit (Miltenyi Biotec). Finally, the purities were tested (>90%). The purified γδ T cells were stimulated with either IL-1β or IL-23 or the combination for 48 hours. The supernatants were collected for measurement of IL-17A. The remaining cells were directly stained for intracellular IL-17A either without additional stimulation or with phorbol-12-myristate-13-acetate (PMA, 50 ng/mL; Sigma-Aldrich), ionomycin (1 μg/mL; Sigma-Aldrich), and monensin (5 μg/mL; Sigma-Aldrich) for 5 hours. To measure IL-23 secretion by macrophages stimulated with HMGB1, peritoneal macrophages were harvested from TLR4+/+ mice or TLR4−/− mice 3 days after treatment with 3% sodium thioglycolate. The cells were stimulated with HMGB1 (20 ng/mL, eBioscience) for 18 hours and the supernatant was collected for IL-23 measurement. The concentrations of IL-17A, IL-23, IL-23p40, and HMGB1 were measured by a standard enzyme-linked immunosorbent assay (ELISA). The following ELISA kits were used: IL-17A and IL-23p40 (Dakewe Biotech, Shenzhen, China); IL-23 (Biolegend, USA); and HMGB1 (Yanhui Biotech, Shanghai, China).
Flow Cytometric Analysis.
To isolate hepatic leukocytes, livers were pressed through a 200G stainless steel mesh and suspended in PBS. The suspension was centrifuged at 50g for 1 minute. The supernatant was then transferred into a new tube and centrifuged at 800g for 10 minutes. The pellets were resuspended in 40% Percoll and centrifuged at 1,260g for 15 minutes at room temperature. The pellets were resuspended and the cell number was determined. To detect hepatic neutrophils, 1 × 106 cells were stained with specific mAb against mouse FITC-CD11b (M1/70, BD Bioscience, USA), PE-Ly6G (1A8, BD Bioscience), Percp-Cy5.5-CD45.2 (104, BD Bioscience), and APC-Gr-1 (RB6-8C5, BD Bioscience). To detect γδ T cells, 1 × 106 cells were stained with specific mAb against mouse FITC-γδTCR (GL3, eBioscience), PE-CD3 (145-2C11, BD Bioscience), and Percp-Cy5.5-CD45.2 (104, BD Bioscience). To detect IL-17A+ cells, 1 × 106 cells were stimulated with PMA (50 ng/mL), ionomycin (1 μg/mL), and monensin (5 μg/mL) for 4 hours. The cells were stained with FITC-CD4 (RM4-5, BD Bioscience), Percp-Cy5.5-CD3 (145-2C11, BD Bioscience), APC-γδTCR (GL3, eBioscience), and PE-CY7-NK1.1 (PK136, BD Bioscience), and then intracellularly stained with PE-IL-17A (BD Bioscience) after fixation and permeabilization. Finally, the stained cells were analyzed using a FACSCalibur (BD Biosciences) or BD LSR II (BD Biosciences) flow cytometer. The acquired data were analyzed using FlowJo software.
Data are presented as the mean ± standard error of the mean (SEM). The significance of differences was determined using a two-tailed unpaired t test; the significance levels are marked *P < 0.05; **P < 0.01; ***P < 0.005.
Acetaminophen Induces Damage-Associated Liver Inflammation by Way of the Production of IL-17A.
As previously reported, acetaminophen induces damage-associated liver injury.1, 2, 24 As shown in Fig. 1A, serum ALT started to elevate early and peaked at 24 hours after acetaminophen challenge. Accordingly, H&E staining demonstrated the presence of many necrotic areas around the central port veins in the liver (Fig. 1B). The number of total hepatic leukocytes was 2-fold greater than that in control mice (Fig. 1C), and neutrophils (but not lymphocytes) were the major constituent of the increased leukocyte population (Fig. 1D). Both the percentage and number of neutrophils in the liver were significantly increased (Fig. 1E).
IL-17A has been reported to play an important role in inducing granulopoiesis and chemotaxis through the stimulation of endothelial and epithelial cells to produce granulocyte-colony stimulating factor, macrophage inflammatory protein-2, and keratinocyte cytokine.19 To investigate the role of IL-17A in the accumulation of neutrophils in the liver, we measured serum and hepatic IL-17A levels. The concentration of IL-17A in the serum gradually increased and peaked at 24 hours after acetaminophen challenge (Fig. 2A), which was consistent with a clinical report of acetaminophen patients.32 Importantly, the mRNA level of IL-17A in acetaminophen-treated livers was much higher than that in control livers (Fig. 2A). To understand the effect of IL-17A on neutrophil accumulation in the liver, a neutralizing antibody was used to inhibit the function of IL-17A. The percentage and number of neutrophils in the murine liver were reduced to almost baseline levels (Fig. 2B,C). The serum ALT level in anti-IL-17A-treated mice (4,313 ± 264.7 IU/L) was less than that in the control group (9,062 ± 716.7 IU/L, Fig. 2D). Accordingly, the survival rate of mice pretreated with the neutralizing antibody was better than that of the control mice (Fig. 2D). Therefore, our data demonstrate that IL-17A is required for the accumulation of neutrophils in the liver during acetaminophen-induced liver inflammation.
γδ T Cells Are the Major Producers of IL-17A.
αβTh17 cells, NKT cells, NK cells, and γδ T cells have been reported to mediate liver disease in an IL-17A-dependent manner.18, 33 To determine which population of lymphocytes produces IL-17A in the acute liver inflammation induced by acetaminophen, we examined the generation of IL-17A from hepatic lymphocytes. Hepatic lymphocytes were isolated and stimulated with PMA and ionomycin. Only IL-17A+CD3+CD4-NK1.1−γδ TCR+ cells significantly increased after acetaminophen challenge (Fig. 3A). After depletion of γδ T cells (Fig. 3B), but not CD4+ T cells (Fig. 3C) or NK/NKT cells (Fig. 3D), the concentration of IL-17A in the serum was significantly reduced. After acetaminophen challenge, the percentage of hepatic γδ T cells slightly decreased in all hepatic leukocytes due to the increasing neutrophils in the liver (Fig. 3E). However, the absolute number of hepatic γδ T cells significantly increased (Fig. 3F), whereas the percentage and absolute number of γδ T cells and the percentage of IL-17A+ γδ T cells in the spleen did not increase (Supporting Fig. 1A-D) after acetaminophen challenge. Moreover, few splenic γδ T cells expressed IL-23 receptor but most of them expressed CD27, which were prone to producing IFN-γ (Supporting Fig. 1E,F).20 Together, our results demonstrate that hepatic γδ T cells are the major producers of IL-17A during acetaminophen-induced liver inflammation.
Meanwhile, after depletion of γδ T cells, liver injury was attenuated (Fig. 4; Supporting Fig. 2). ALT and bilirubin levels were reduced (Fig. 4A,B). The necrotic hepatic areas were also reduced (Fig. 4C). The survival rate of γδ T cell-depleted mice was markedly improved (Fig. 4D), with a decreased content and total number of CD11bhiLy-6G+ neutrophils in the liver (Fig. 4E,F). The attenuated liver injury, decreased neutrophils in the liver, and improved survival ratio were also observed in TCRδ−/− mice compared to that of age-matched control mice (Supporting Fig. 2). Thus, hepatic γδ T cells are critical during acetaminophen-induced, damage-associated IL-17A-mediated liver inflammation.
IL-23 Is Required for the Generation of IL-17A-Producing γδ T Cells.
To investigate the role of IL-23 in the production of IL-17A by hepatic γδ T cells, IL-23 in the sera and liver was measured. Serum IL-23 significantly increased and peaked at 12 hours after acetaminophen challenge (Fig. 5A), and p40, one subunit of IL-23, also increased and peaked at 12 hours (Fig. 5A). In the liver, p19 and p40, two subunits of IL-23, also increased (Fig. 5B). To further determine whether IL-23 is required for the production of IL-17A, we neutralized its function using an anti-IL-23p19 antibody or p40-deficient mice (Fig. 5C). Serum IL-17A significantly decreased after neutralizing IL-23 or using p40-deficient mice. Moreover, infiltration of neutrophils into the liver was significantly ameliorated (Fig. 5D-F). Meanwhile, the liver injury was also reduced in the p40-deficient mice compared to the aged-matched control mice (Supporting Fig. 3). Taken together, these results show that IL-23 is important for the production IL-17A by γδ T cells after acetaminophen challenge.
To confirm the role of IL-23 in the generation of IL-17A-producing γδ T cells, we stimulated hepatic γδ T cells with exogenous IL-23 in vitro. After stimulation for 48 hours, the percentage of IL-17A-producing γδ T cells was significantly increased (Fig. 6A,B). After a second cycle of stimulation with PMA for 5 hours, the percentage of IL-17A-producing γδ T cells further increased to 32.3% (Fig. 6A). Supernatant IL-17A from total hepatic lymphocytes or purified γδ T cells was increased after stimulation with IL-23, which was further enhanced by IL-23+IL-1β stimulation (Fig. 6C,D). Therefore, the in vitro experimental data demonstrate that IL-23 is required for the production of IL-17A from γδ T cells.
HMGB-1-TLR4 Mediates the Production of IL-23 by Macrophages.
To understand whether macrophages mediate the production of IL-23, we inhibited macrophages, including Kupffer cells, with GdCl3. The concentrations of serum IL-23 and IL-17A were reduced after depletion of macrophages (Fig. 7A) and could not be induced in TLR4−/− mice (but was induced in TLR2−/− and TLR9−/− mice, data not shown) (Fig. 7B) after acetaminophen challenge. Furthermore, the inhibition of macrophage with GdCl3 also reduced the liver injury (Supporting Fig. 4). Concurrently, serum HMGB1, a damage-associated molecule released from necrotic hepatocytes, increased after treatment with acetaminophen (Fig. 7C), and use of the HMGB1 inhibitor glycyrrhizin markedly reduced the production of IL-23 and IL-17A (Fig. 7D) and hepatic neutrophil recruitment (Fig.7E). To confirm the role of HMGB-TLR4 pathway in the generation of IL-23 from macrophages, we stimulated macrophages from TLR4+/+ or TLR4−/− mice with soluble HMGB1. Soluble HMGB1 enhanced the production of IL-23 by TLR4+/+ macrophages but not by TLR4−/− macrophages (Fig. 7F). Thus, the HMGB1-TLR4-IL-23 pathway in macrophages determines the generation IL-17-producing γδ T cells, which mediate neutrophil infiltration and liver inflammation.
This study revealed a crucial role for the HMGB1-TLR4-IL-23-IL-17A axis in drug-induced liver inflammation. HMGB1, a damage-associated molecule from necrotic hepatocytes, stimulates the production of IL-23 by hepatic macrophages in a TLR-4-dependent manner, and macrophage-derived IL-23 aids in the generation of IL-17A-producing γδ T cells in the liver. IL-17A secreted by γδ T cells then recruits neutrophils into the liver. Thus, the interaction between macrophages and γδ T cells contributes to tissue damage-induced liver inflammation following the accumulation of neutrophils (Fig. 8). This study provides new insight into the role of IL-17-producing γδ T cells during sterile inflammation and sheds light on how drug-induced liver diseases may be controlled.
Although controversies exist regarding the precise role of each component, DAMPs released from necrotic hepatocytes have been well established to mediate the second wave of inflammation by activating the innate immune response.5 We found that serum HMGB1 significantly increased after acetaminophen treatment (Fig. 7C), and inhibition of HMGB1 with the specific inhibitor glycyrrhizin protects mice from neutrophil infiltration and liver injury (Fig. 7E), which is consistent with the effect of anti-HMGB1 antibodies shown in a previous report.26 Meanwhile, blockade of HMGB1 markedly reduced the production of IL-23 and IL-17A (Fig. 7D). In vivo inactivation of macrophages attenuated liver injury and decreased the concentration of IL-23 and IL-17A in the murine sera (Fig. 7A). Moreover, when acetaminophen was administered to TLR2−/−, TLR4−/−, and TLR9−/− mice, only TLR4−/− mice exhibited a reduced production of IL-23 and IL-17A (Fig. 7B), and TLR4−/− macrophages lacked the ability to produce IL-23 in vitro (Fig. 7F); these data are in agreement with a previous report that showed that TLR4 is involved in acetaminophen pathogenesis.34 These data suggest that the HMGB1-TLR4-IL-23-IL-17A axis plays a pivotal role during drug-mediated, tissue damage-induced liver injury. Our results also suggest that blocking any portion of this axis will attenuate liver injury and neutrophil infiltration. However, our research cannot exclude the possibility that HMGB1 directly induces IL-17A production independent of IL-23. In addition to HMGB1, other DAMPs (such as DNA and cyclophilin A) have been reported to participate in activating the innate immune response.7, 21, 22 Except for TLR4, other receptors for HMGB1 may also stimulate the release of inflammatory cytokines and should be further investigated.
Macrophages can quickly respond to endogenous stimulating factors after tissue injury.35 However, the role of macrophages in the acetaminophen-induced liver injury is controversial. Hepatic macrophages have been demonstrated to play a pathogenic role through their secretion of proinflammatory factors, such as tumor necrosis factor alpha (TNF-α), IL-1β, and NO.36 However, hepatic macrophages have also been reported to play a protective role through their secretion of regulatory factors, such as IL-10.37 This controversy stems from the effects of compounds used to inactivate (GdCl3) and deplete macrophages (clodronate/liposome). Macrophages are heterogeneous and plastic, and at least two major macrophage populations exist, including classically activated macrophages (M1) and alternatively activated macrophages (M2).35 An induced macrophage (IM) population that differs from resident hepatic macrophages has been reported in acetaminophen-induced liver injury. IMs are formed from circulating monocytes infiltrating the liver after acetaminophen treatment and exhibit phenotypes of alternatively activated macrophages. The absence of IMs delays the recovery of liver injury.38 However, resident hepatic macrophages isolated from normal livers have enhanced mRNA expression of IL-1β and TNF-α after stimulation with DAMPs in vitro.24 Thus, these studies demonstrate that hepatic resident macrophages are classically activated macrophages, which are prone to generating proinflammatory cytokines during acetaminophen-induced liver injury. In our study, macrophages also produced IL-23 after HMGB1 stimulation. γδ T cells were also able to produce IL-17A rapidly in response to DAMPs,18 and naïve γδ T cells produced IL-17 in response to IL-23 in the absence of TCR engagement,39 which was enhanced by the addition of IL-1β.40 In this study, IL-17 was dramatically elevated after acetaminophen treatment. Although NK and NKT cells are the dominant innate immune cells in murine liver,41 they did not produce IL-17A, which was confirmed by depleting NK and NKT cells with mAb (Fig. 3D). In our study, hepatic CD4+ T cells were not the major source of IL-17A, and CD4+ T cell depletion did not influence IL-17A production (Fig. 3C). Surprisingly, deletion of γδ T cells significantly reduced IL-17A production. The percentage of IL-17A+ γδ T cells rapidly increased after acetaminophen challenge, indicating that hepatic γδ T cells are the primary source of IL-17A. IL-23 induces γδ T cells to secrete IL-17A in vitro, and blocking IL-23 or IL-23 deficiency decreases the IL-17A levels in vivo (Figs. 5C, 6). Although acetaminophen increased IL-1β production, blocking IL-1β with IL-1RA had no significant effect on neutrophil infiltration (data not shown). IL-1β alone did not induce γδ T cell production of IL-17A in vitro; however, IL-1β synergized with IL-23 to further increase IL-17A production, implying that IL-1β also plays a role in IL-17A production by γδ T cells. Because other studies have shown that IL-17A can stimulate macrophages to produce the inflammatory cytokines and chemokines,42, 43 further research on the interaction between macrophages and γδ T cells is required. Although γδ T cells dominantly produce IL-17A in this study, other immune cells, such as CD8+T cells, neutrophils, and lymphoid tissue inducer-like cells, also can produce IL-17A.18 Their roles in pathogenesis need to be further investigated. Meanwhile, whether other cell types are involved in liver injury in other ways also needs to be studied.
In summary, our study provides evidence that the macrophage-γδ T-neutrophil cascading response is involved in acetaminophen-induced liver inflammation by way of an HMGB1-TLR4-IL-23-IL-17A axis. Whether this mechanism extends to sterile inflammation other than drug-induced liver injury requires further study. The development of new therapeutic approaches that control DAMP-induced liver injury is important.
The authors thank professors Zhexiong Lian, Zhinan Yin, and Shaobo Su for providing gene-deficient mice.