Infiltrated IL-17A-producing gamma delta T cells play a protective role in sepsis-induced liver injury and are regulated by CCR6 and gut commensal microbes

Introduction Sepsis is a common but serious disease in intensive care units, which may induce multiple organ dysfunctions such as liver injury. Previous studies have demonstrated that gamma delta (γδ) T cells play a protective role in sepsis. However, the function and mechanism of γδ T cells in sepsis-induced liver injury have not been fully elucidated. IL-17A-producing γδ T cells are a newly identified cell subtype. Methods We utilized IL-17A-deficient mice to investigate the role of IL-17A-producing γδ T cells in sepsis using the cecum ligation and puncture (CLP) model. Results Our findings suggested that these cells were the major source of IL-17A and protected against sepsis-induced liver injury. Flow cytometry analysis revealed that these γδ T cells expressed Vγ4 TCR and migrated into liver from peripheral post CLP, in a CCR6-dependent manner. When CLP mice were treated with anti-CCR6 antibody to block CCR6-CCL20 axis, the recruitment of Vγ4+ γδ T cells was abolished, indicating a CCR6-dependent manner of migration. Interestingly, pseudo germ-free CLP mice treated with antibiotics showed that hepatic IL-17A+ γδ T cells were regulated by gut commensal microbes. E. coli alone were able to restore the protective effect in pseudo germ-free mice by rescuing hepatic IL-17A+ γδ T cell population. Conclusion Our research has shown that Vγ4+ IL-17A+ γδ T cells infiltrating into the liver play a crucial role in protecting against sepsis-induced liver injury. This protection was contingent upon the recruitment of CCR6 and regulated by gut commensal microbes.


Introduction
Sepsis is an uncontrolled inflammation caused by infection with organ dysfunction (Cheng et al., 2017). More than 1.5 million patients with sepsis are annually admitted to the intensive care unit (ICU) in the US, leading to 30% mortality annually (Seymour et al., 2016). Although a lot of progress in the pathology of the disease has been made, the role of sepsis in multiple organ failure remains elusive. The liver plays a crucial role in the pathology of sepsis (Strnad et al., 2017), and sepsis-related liver injury is associated with high mortality (Baggio Savio et al., 2017). Therefore, it is necessary to explore the pathogenesis of sepsis-induced liver injury and to develop effective therapies for this disease.
Gamma delta (gd) T cells are enriched in the liver, with a frequency of 3%-5% of the total liver lymphocytes (Racanelli and Rehermann, 2006). These cells bridge the innate and adaptive immune responses and secrete pro-inflammatory cytokines, including interleukin 17A (IL-17A) and interferon gamma (IFNg), upon stimulation (Bonneville et al., 2010). IL-17A-producing gd T cells can be considered as a separate cell type, and they play a crucial role in host defense against infections such as bacteria, fungi, and viruses (Hayday, 2009). Moreover, IL-17A + gd T cells have been identified as playing a protective role in polymicrobial sepsis (Ogiku et al., 2012), particularly in organ failure of the lung (de Souza Costa et al., 2015). Although it has been reported that gd T cells are involved in sepsis-related liver injury, their physiological characteristics are largely unclarified.
The thymus is the primary organ for gd T-cell maturation and differentiation. Unlike conventional alpha beta (ab) T cells, the majority of gd T cells acquire their effector function in the thymus. These cells predominantly reside in the mucosal and epithelial barriers and produce cytokines rapidly, although a small fraction still circulate in the periphery (Chien et al., 2014;Nielsen et al., 2017;Ma et al., 2019). According to their function potential and localization, gd T cells can be further divided into effector subsets with bias on the use of T-cell receptor (TCR) gene segments. However, the role of the different gd T-cell subtypes in sepsis has not been clarified. The origin of these gd T cells is also unclear.
CCR6 is an important receptor in mucosal immunity, with CCL20 as the chemokine ligand (Cook et al., 2000). Among the CD4 T cells, CCR6 is specifically expressed on T helper 17 (Th17) cells and regulatory T cells (Tregs). It can mediate distinct immune diseases through the chemotaxis of T cells. Recently, researchers have identified that CCR6 is also expressed in gd T cells, functionally associated with subsets of IL-17A production (Oo et al., 2012). However, the function of CCR6 expressed in gd T cells in liver disease is largely unclear.
In the past decade, the gut microbiota was considered as a major modulator of the injury responses in organs remote from the gastrointestinal tract, including the liver (Llorente and Schnabl, 2015;Schuijt et al., 2016;Hu et al., 2017;Li et al., 2018;Tripathi et al., 2018). It has also been reported that dysbiosis could increase the risk of sepsis  and death (Salio et al., 2014). Germ-free (GF) mice with cecum ligation and puncture (CLP) showed higher pathogen load and mortality (Luoma et al., 2014). The gut microbiota has been shown to be involved in sepsis progression  and reported to regulate the host immune responses in disease pathogenesis (Chassaing et al., 2014). Thus, the gut microbiota could be an important mediator of sepsisrelated organ failure, which still needs to be further elucidated.
In this study, Vg4 + gd T cells were revealed to play a protective role in sepsis-induced liver injury through the production of IL-17A. The CCR6-CCL20 axis was investigated in sepsis-induced liver injury, which demonstrated a CCR6-dependent manner of recruitment of gd T cells into the liver. The gut microbiota was also revealed to be positively correlated with sepsis-induced liver injury.

Materials and method
Mice C57BL/6 mice were purchased from Shanghai SLAC Laboratory Animal Co., Ltd. (Shanghai, China). IL-17A knockout (KO), TCRab −/− , TCRgd −/− , CD45.1, and Rag1 mice were obtained from the Model Animal Research Center (Nanjing University, Nanjing, China). All mice were housed in a specific pathogen-free facility at the National Center for Protein Science Shanghai. All the mice in this study were males, 8-12 weeks old.

Ethical statement
All animal protocols were approved by the Ethics Committee for Animal Care and Use at Shanghai Pudong New Area People's Hospital.

Antibody treatment
The anti-mouse Vg4 antibody (eBioscience, San Diego, CA, USA) was intraperitoneally (i.p.) administered at 500 mg, 3 days before CLP (de Souza Costa et al., 2015). Blockade of the CCR6 + population was conducted through i.p. administration of an antimouse CCR6 antibody (BioLegend, San Diego, CA, USA) at 500 mg, 3 days before CLP (Hammerich et al., 2014), which could block the function of the CCR6-CCL20 axis by binding CCR6 on the cell surface. Control animals received an i.p. injection of isotype immunoglobulin G (IgG; 500 mg, 3 days before CLP) (BioLegend).

Sepsis by CLP
Sepsis was induced by CLP (Rittirsch et al., 2009). Briefly, C57BL/6 mice were anesthetized with Avertin (300 mg/kg, i.p) and a 1-cm midline incision made on the anterior abdomen. The cecum was exposed and ligated below the ileocecal junction without causing bowel obstruction. Puncture was performed with an 18gauge needle five times to induce sepsis. The cecum was squeezed to allow the contents to pass through the punctures. The cecum was then placed back into the abdominal cavity and the peritoneal wall and skin incision closed. All animals were resuscitated with 1 m of serial saline subcutaneously.

Survival analysis
After surgery, the mice were monitored for survival status every 24 h for 7 days. The overall survival analysis was performed using the Mantel-Cox test with Prism v.5 software.

Bacterial load assessment
Serum samples were serially diluted with sterile phosphatebuffered saline (PBS) and cultured on agar plates. After incubation at 37°C overnight, the colony was counted. According to the dilution of the samples, the limit detection for the colony-forming unit (CFU) assay was 1 × 10 4 CFU/ml.

Cell preparation
The mice were euthanized and the thymus, spleen, and liver removed for cell collection. For the thymus, the tissue was ground and passed through a 70-mm mesh to obtain single-cell suspension. The spleen was first ground and passed through a 70-mm mesh, and then red blood cells (RBCs) were lysed and a single-cell suspension in PBS obtained. Similarly, the liver was ground and passed through a 70-mm mesh. The cell suspension was resuspended in 40% Percoll (GE Healthcare, Chicago, IL, USA) and centrifuged for 20 min at 2,800 rpm with breaking at 1. Cells were then collected from precipitates and washed in PBS twice.
For cytokine staining, the cell suspensions were stimulated with phorbol 12-myristate 13-acetate (PMA; 50 ng/ml), ionomycin (500 ng/ml) (both from Sigma, St. Louis, MO, USA), and GolgiPlug (BD Biosciences, Franklin Lakes, NJ, USA) for 4 h at 37°C in an incubator. After surface marker staining, the cells were fixed and permeabilized (BD Biosciences) and stained for cytokine IL-17A or IFN-g at 4°C for 30 min. The samples were analyzed with Fortessa X-20 (BD Biosciences).

Adenoviral knockdown
To knock down the expression of CCR6, a specific RNAi vector of CCR6 was designed using RNAi Designer, with the sequence CCGATAACATCAATGTCCAAGTGAA. Subsequently, 293T cells were transfected with this vector to produce adenovirus. The BLOCK-iT vector was used as a control. The gd T cells were isolated and stimulated with anti-CD3 and anti-CD28 for 24 h. Adenoviral supernatants were then added into the cell culture, and the efficiency of RNAi was evaluated using flow cytometry before adoptive transfer.

Adoptive transfer
Recipient mice (6-8 weeks old) were sublethally irradiated with 5 Gy a day prior to adoptive transfer and were maintained with antibiotics containing water for 2 weeks. On the second day, 0.1 million gd T cells from the thymus or the liver were adoptively transferred into recipients by i.p. injection.

Real-time PCR
RNA was isolated from the liver using the TRIzol reagent (Invitrogen, Carlsbad, CA, USA) according to the manufacturer's instructions. Complementary DNA (cDNA) was synthesized using reverse transcriptase (Promega, Madison, WI, USA). A real-time PCR kit (Takara, San Jose, CA, USA) was used according to the manufacturer's instructions. The reaction was performed on an ABI 7500 Real-Time PCR system. The primer sequences used in this work are listed in Table 1.

Serum alanine transaminase measurement
The serum alanine transaminase (ALT) levels were measured using ALT assay kits (Nanjing Jiancheng Bioengineering Institute) according to the manufacturer's instructions. According to the TTTCTGTGAGAGTGGTCTTACC AGTCTGAGAGATGACATTGCAA sample dilution, the limit of detection for ALT measurement was 1 U/L.

Enzyme-linked immunosorbent assay
The IL-17A levels in the serum and liver were determined using IL-17A enzyme-linked immunosorbent assay (ELISA) kits (ThermoFisher, Waltham, MA, USA) according to the manufacturer's instructions. According to the sample dilution, the limit of detection of the sample was 1.6 pg/ml.

Depletion of microbiota with broadspectrum antibiotics
To deplete the gut microbiota and generate pseudo-GF mice, a broad-spectrum antibiotic cocktail (100 mg/kg vancomycin, 200 mg/kg neomycin sulfate, 200 mg/kg metronidazole, and 200 mg/kg ampicillin) was intragastrically administered once a day for 5 days to 8-to 12-week-old mice.

Fecal microbiota transplantation assay
Fecal microbiota transplantation (FMT) was performed according to a modified method previously described (DeFronzo et al., 2016;Buford, 2017). Briefly, recipient mice were administered the suspension via oral gavage once a day for 3 days with 200 ml fecal dilution in PBS. To prepare this dilution, fecal samples were collected from donors. Stool pellets were pooled and vortexed for 5 min in 0.125 g/L PBS for homogenization. Thereafter, the samples were gently centrifuged for 5 min at 350 × g and the supernatant collected. Subsequently, 200 ml of this suspension was administered by oral gavage to recipient mice.

Statistical analysis
Statistical analysis was performed with GraphPad Prism 5.0 (GraphPad Software, La Jolla, CA, USA). Data were presented as the mean ± SEM. Differences between two groups were assessed using Student's t-test, whereas those among more than two groups were determined using ANOVA. A p-value <0.05 was considered significant (*p < 0.05, **p < 0.01, ***p < 0.001).

IL-17A protected against sepsis-induced liver injury in cecum ligation and puncture mice
To assess the effects of IL-17A on sepsis-induced liver injury, IL-17A-deficient mice were subjected to sepsis using CLP. As shown in Figure 1A, all animals in the IL-17A-deficient group died within 72 h post-CLP. In contrast, 30% of the wild-type (WT) mice survived 7 days post-CLP ( Figure 1A). The serum level of ALT was increased in the WT group 12 h post-CLP, whereas it was even higher in the IL-17A KO group ( Figure 1B). In addition, the bacterial load in the blood of IL-17A-deficient mice was significantly higher than that of control animals ( Figure 1C). As expected, the serum levels of IL-17A were below the detection levels in the deficient group ( Figure 1D). Moreover, the serum IL-6 level was much higher in the IL-17A KO group than that in the WT group ( Figure 1E). However, in the spleen, the results revealed little differences in the IFN-g + CD4 T cells between the control and IL-17A-deficient mice ( Figures 1F, G). Together, these marked differences between the control and KO groups indicate a protective role of IL-17A in liver injury in CLP-induced sepsis.
gd T cells were the major source of IL-17A in sepsis-induced liver injury As multiple populations could secrete IL-17A, its major source was further examined in the liver post-CLP. Single-cell suspensions of the liver were prepared for flow cytometry analysis 12 h after the CLP procedure. More CD3 + T lymphocytes were detected in sepsisinduced livers compared to the control group (Figures 2A, B). To better determine the subtypes, the expression of ab and gd TCRs in these T lymphocytes was analyzed. It was found that, although there was little difference in the expression of ab TCRs ( Figures 2C, D), there were more gd TCR + T lymphocytes in CLP livers than in controls ( Figures 2E, F). In addition, IL-17A was not detected among the ab T cells in either the CLP or the control group ( Figures 2G, H), while 30% of the gd T cells were determined as IL-17A-producing cells in CLP mice ( Figures 2I, J). Thus, more IL-17A + gd TCR cells were detected in the CLP group.
Moreover, hepatic IL-17A levels were significantly higher in CLP livers than in controls, as determined using both quantitative PCR (qPCR) and ELISA ( Figures 2K, L). The fluorescence-activated cell sorting (FACS) plots showed that IL-17A + cells accounted for approximately 20% of the hepatic live cells from CLP mice (Figures 2M,N). To further address which subset produced more IL-17A in CLP livers, the expression of the ab and gd TCRs in these cells was examined. It was discovered that more than 40% of the IL-17A + lymphocytes were TCRd + , but only about 10% were TCRb + in CLP mice ( Figures 2O, P), indicating that hepatic gd T cells comprise the major subset for IL-17A secretion in sepsis-induced liver injury.

IL-17A-producing gd T cells played a major protective role in sepsis-induced liver injury
To better clarify the function of gd T cells expressing IL-17A in sepsis-induced liver injury, CLP was performed on ab and gd TCR KO mice, which were deficient in ab and gd T cells, respectively. The data showed that there was little difference in the serum ALT and IL-17A between the WT and ab T-cell KO mice at 12 h post-CLP ( Figures 3A, B). However, the knockout of gd T cells led to a markedly elevated plasma ALT but a reduced IL-17A level, compared to WT littermates ( Figures 3C, D). On the other hand, adoptive transfer of gd T cells, but not ab T cells, could rescue the levels of IL-17A in both serum and liver (Figures 3E, F) and suppress plasma ALT ( Figure 3G). Similar results were observed in Rag1 mice with exogenous IL-17A ( Figure 3H). The administration of IL-17A in gd T-cell KO mice could alter the plasma ALT level, but not in ab T-cell KO mice ( Figure 3I). However, Rag1 mice that had adoptive transfer of IL-17A D gd T cells showed elevated levels of ALT to vehicle gd T cells, suggesting that the protective role of gd T cells in liver injury was IL-17A-dependent ( Figure 3J). Thus, these data together suggest a major protective role of gd T cells expressing IL-17A in sepsisinduced liver injury in CLP mice.  (Cheng et al., 2017). Data are from one intact experiment and representative of three independent experiments (five mice per group). Error bars denote the mean ± SEM. Statistical analysis was performed using Student's t-test. n.s., not significant. ***p < 0.001.
Vg4 + but not Vg6 + gd T cells were the major protectors in sepsis-induced liver injury IL-17A-producing gd T cells are mainly composed of two subsets: Vg4 and Vg6. These two subtypes of gd T cells have been reported in many disease models and have presented distinguished functions (Ribot et al., 2009;Wu et al., 2014). Thus, the subsets of IL-17A-producing hepatic gd T cells in mice subjected to CLP were further investigated. The FACS data showed that more IL-17A + gd T cells expressed the Vg4 chain ( Figures 4A, B) compared to the Vg6 chain ( Figures 4C, D). Adoptive transfer of Vg4 + gd T cells into Rag1 mice could suppress the plasma ALT and elevate the hepatic Error bars denote the mean ± SEM. Statistical analysis was performed using Student's t-test. n.s., not significant. *p < 0.05, **p < 0.01, ***p < 0.001.
IL-17A levels post-CLP compared to control animals ( Figures 4E,  F). Subsequently, Vg4 + gd T cells were abolished in C57BL/6 mice using an anti-Vg4 depletion antibody and the mice subjected to CLP. After 12 h, these mice presented altered liver IL-17A levels and severe liver injury ( Figures 4G, H). Coordinately, Rag1 mice transferred with IL-17A D Vg4 + gd T cells had decreased levels of hepatic IL-17A and increased plasma ALT compared to vehicle Vg4 + gd T cells (Figures 4I, J). Therefore, these results determined a critical role of Vg4 + gd T cells in protecting against sepsis-induced liver injury through the secretion of IL-17A. Statistical analysis was performed using Student's t-test. n.s., not significant. *p < 0.05, **p < 0.01, ***p < 0.001.

Protective gd T cells migrated into the liver after cecum ligation and puncture
As the development of gd T cells originates from both the thymus and the liver, the origin of protective IL-17 + gd T cells was subsequently explored in sepsis-induced liver injury. The gd T cells from the liver and thymus were isolated and were intravenously injected into sublethally irradiated gd T-cell KO recipients to determine their origin in the liver of septic mice ( Figure 5A). The plasma ALT level was measured 12 h post-CLP and was found to significantly decrease in thymic gd T-cell transfer recipients ( Figure 5B). In addition, the hepatic and serum levels of IL-17A were both higher in thymic gd T-cell transfer recipients compared to liver gd T-cell transfer recipients ( Figures 5C, D). FACS analysis showed that the proportion of hepatic IL-17A + gd T cells was much higher in mice that received thymic gd T cells compared to those that received liver gd T cells ( Figures 5E, F). In contrast, the levels of splenic IL-17A + gd T cells in mice transferred with thymic gd T cells and liver gd T cells were comparable ( Figures 5G, H). These results suggest that the thymic gd T cells, but not the liver gd T cells, played the protective role, and these cells migrated into the liver to protect against sepsisinduced liver injury.

Infiltration of IL-17A + gd T cells into the liver was CCR6-dependent
As IL-17A + gd T cells were recruited into the liver post-CLP, the manner of infiltration during liver injury was investigated next. The accumulation of gd T cells in the liver in CCL4-induced hepatitis has been reported to be CCR6-dependent (Hammerich et al., 2014). We therefore examined whether this approach is also functional in sepsis-induced liver injury. Mice that had the CLP procedure showed significantly higher hepatic gene expression level of Ccl20 compared to the controls ( Figure 6A). To elucidate the functional relevance of CCR6 in sepsis-induced liver injury, CLP mice were treated with an anti-CCR6 blocking antibody, which could suppress the CCL20-CCR6 axis pathway. Compared to mice injected the rabbit IgG, those given the anti-CCR6 antibody showed elevated plasma ALT and decreased hepatic IL-17A levels ( Figures 6B, C). Moreover, there was a significant decrease in the fraction of IL-17A + gd T cells among hepatic lymphocytes ( Figures 6D, E). The expression of CCR6 was knocked out using the RNAi system, and then CD45.1 + CCR6 D and CD45.2 + CCR6 WT gd T cells were mixed in a 1:1 ratio and injected intravenously into sublethally irradiated gd T-cell KO mice ( Figure 6F). At 12 h post-CLP, an overwhelming number of CD45.2 (i.e., CCR6 WT ) gd T cells was found, but only a few CD45.1 + (i.e., CCR6 D ) gd T cells in the liver (Figures 6G, H). Consistently, although only a few cells were detected in the spleen, the frequency of CD45.2 + gd T cells was much higher than that of CD45.1 + gd T cells (Figures 6G, H). These data together indicate that gd T cells migrate into the liver during sepsis through a CCR6-dependent manner.

Hepatic IL-17A + gd T cells were regulated by the microbiota in sepsis-induced liver injury
The gut microbiota has been considered as an important mediator of liver disease; thus, their function in hepatic IL-17A + during sepsis was further investigated. Firstly, sepsis-induced liver injury in the absence of microbiota was analyzed. The absence of the microbiota led to increased plasma ALT and decreased hepatic IL-17A levels ( Figures 7A, B). In addition, the frequency of IL-17A + gd T cells was much lower in pseudo-GF mice (Figures 7C, D).
To further investigate the function of bacterial species in sepsisinduced liver injury, mice were treated using different combinations of antibiotics (ampicillin, vancomycin, neomycin, and metronidazole). Subsequently, these mice were subjected to CLP. However, there was almost no difference among the antibiotic combinations ( Figure 7F). In contrast, the serum level of ALT was positively correlated with the global microbial DNA load ( Figure 7E). These results suggest that the global bacterial load has more significance in liver injury during sepsis.
Similar to the transfer of feces, the transfer of Escherichia coli alone could suppress the plasma ALT ( Figure 7G) and recover the level of hepatic IL-17A ( Figure 7H). Furthermore, in single E. colitransferred mice, the fraction of hepatic IL-17A + gd T cells was strongly elevated (Figures 7I, J). These results indicate that the global microbiota, regardless of species, is crucial in sepsis-induced liver injury.

Discussion
This study first clarified the role of IL-17A-producing gd T cells in sepsis-induced liver injury, in which the Vg4 + gd T cell subtype played a protective role against liver injury during sepsis through the secretion of IL-17A. These cells originate in the thymus and migrate into the liver dependent on CCR6, and gut commensal microbes regulate this subset of gd T cells.
Sepsis triggers systemic uncontrolled immune responses and could lead to organ dysfunction or even death. Over the last decade, research work on sepsis has demonstrated the protective role of gd T cells (Flierl et al., 2008) and IL-17A (Ogiku et al., 2012), but the pathophysiology of sepsis-induced liver injury has not been clearly elucidated. Studies have shown that gd T cells are the major source of IL-17A in sepsis (Kasten et al., 2010;Xu et al., 2010), indicating a potential correlation between IL-17A-producing gd T cells and sepsis-induced liver injury. Consistent with previous findings, our data revealed that, aside from Th17 cells, IL-17A + gd T cells are a major source of IL-17A and that they play a crucial role in protecting against sepsis-induced liver injury.
Similar to ab CD4 + T lymphocytes, subsets of gd T cells may present distinct cytokine profiles, such as IFN-g + and IL-17A + gd T cells. Regarding IL-17A + gd T cells, it has been well established that IL-17A production is restricted to Vg4 + and Vg6 + gd T cells among the total gd T cells in mice (Bonneville et al., 2010;Rei et al., 2014). Consistent with these results, the data in this study revealed that Vg4 + gd T cells comprise the predominant subset among the IL-17A-producing cells in the liver of CLP mice. In line with these results, IL-17A + gd T cells comprise the major subset infiltrating into the liver with IL-17A production in different mouse models (Zhao et al., 2011;Hou et al., 2013). In addition, our data demonstrated that IL-17A predominated over IFN-g from the gd T cells in CLP mouse livers. Therefore, these data further supported that IL-17A + Vg4 + gd T cells and not IFN-g + Vd1 + gd T cells predominated in CLP mouse liver injury.
The liver is the major target organ in sepsis. While this organ is metabolically active, it is immunologically quiescent. It has been shown that hepatocytes are capable of producing IL-7 to regulate the production of IL-17A in gd T cells in situ during viral hepatitis infection (Hou et al., 2013). Hepatic gd T cells originate from fetal gd thymocytes and then migrate into the liver. Moreover, IL-17A +  (Cheng et al., 2017). Data are from one intact experiment and representative of three independent experiments (five mice per group). Error bars denote the mean ± SEM. Statistical analysis was performed using Student's t-test. **p < 0.01, ***p < 0.001.
gd T cells could migrate into the liver from the periphery, similarly to other organs. It has been reported that IL-17 + gd T cells in the skin (Gray et al., 2011;Naik et al., 2012), IL-17A + gd T cells residing in the lungs (Cheng et al., 2014), and IL-17A + gd T cells in colonic lamina propria (Lee et al., 2015) are all recruited from the periphery, but with different pathways to induce the production of IL-17A. In sepsis, protection was conferred by hepatic gd T cells. Furthermore, our data revealed that the thymic and not the liver gd T cells played CD45.1 + gd T cells were transfected with BL-CCR6 D , while CD45.2 + gd T cells were transfected with BL-vehicle; subsequently, these two populations were mixed at a ratio of 1:1 and adoptive transferred into 5 Gy-irradiated gd T-cell knockout (KO) mice as recipients, which were subjected to CLP 3 weeks later. (G) Representative FACS plots of CD45.1 + and CD45.2 + cells from the spleen and liver of recipients. (H) Statistics of CD45.1 + and CD45.2 + cells in (Cheng et al., 2017). Data are from one intact experiment and representative of three independent experiments (five mice per group). Error bars denote the mean ± SEM. Statistical analysis was performed using Student's t-test. n.s., not significant. **p < 0.01, ***p < 0.001. Effect of commensal microbes on cecum ligation and puncture (CLP)-induced liver injury. C57BL/6 mice were fed antibiotics for 5 days to deplete the gut microbiota, received fecal suspension for 3 days, and then subjected to CLP.  (Cheng et al., 2017). Data are from one intact experiment and representative of three independent experiments (five mice per group). Error bars denote the mean ± SEM. Statistical analysis was performed using Pearson's correlation or Student's t-test. *p < 0.05, **p < 0.01, ***p < 0.001. the protective role in CLP, indicating that these cells were infiltrated from the periphery and are not resident cells, further clarifying the function of gd T cells at different barrier sites with a tissuespecific approach. The characteristics of liver injury include complex inflammatory responses between parenchymal and immune cells, no matter whether these immune cells are resident cells or were infiltrated. Recently, increased studies on chemokines and their receptors have revealed their functions in liver injury (Wasmuth et al., 2009;Karlmark et al., 2010). Our work demonstrated that the CCL20-CCR6 axis was functionally involved in sepsis-induced liver injury. Consistent with the findings in a mouse model, patients with cholestatic liver disease showed elevated CCR6 expression, which linked CCL20 to biliary epithelial cells (BECs) (Oo et al., 2012); however, these were considered as Th17-dominated. In this study, CCR6 was mainly expressed on hepatic gd T cells with much higher levels than CD4 T cells in injured livers. Until now, little has been known about the function of CCR6 + gd T cells in vivo, although these cells have also been detected in a colitis model (Mielke et al., 2013). Further functional studies of these subsets are needed.
Commensal microbes could drive the secretion of IL-17A in the gastrointestinal tract, which has been reported to maintain intestinal homeostasis (Ajuebor et al., 2008;Chen et al., 2013). The potential mechanism could be attributed to common components of the intestinal microbiota, such as lipid antigens, which could activate hepatic gd T cells and induce IL-17A production. As indicated by our data, there was little difference between the antibiotic treatments, and E. coli alone could rescue the protection against sepsis in pseudo-GF mice. Reports have shown that IL-17A + gd T cells could protect against infection and promote inflammation in the liver . A study on hepatocellular carcinoma demonstrated that the absolute number of gd T cells in the liver decreased in the absence of the gut microbiota, indicating a correlation between commensal microbiota and the number of hepatic gd T cells (Schneider et al., 2022). These results were consistent with our findings regarding sepsis-induced liver injury. After the depletion of commensal microbes, the number of hepatic IL-17A + gd T cells decreased post-CLP ( Figures 7C, D). It was also found that E. coli alone could restore hepatic IL-17A + gd T cells in CLP mice, which could be supported by the findings of Li et al. (2017). These findings all identified gut microbes as important regulators of hepatic IL-17A + gd T cells both in cancer and during infection.
In conclusion, this study clarified the function of Vg4 + IL-17Aproducing gd T cells in sepsis-induced liver injury. These cells originate from the periphery and migrate into the liver in a CCR6dependent manner, with the gut commensal microbes being the upstream regulators for this population.

Data availability statement
The original contributions presented in the study are included in the article/supplementary material. Further inquiries can be directed to the corresponding author.

Ethics statement
The animal study was reviewed and approved by the Ethics Committee for Animal Care and Use at Shanghai Pudong New Area People's Hospital.