Tofacitinib was purchased from Dalian Meilun Biotechnology Co., Ltd. ConA and dimethyl sulfoxide (DMSO) were purchased from Sigma-Aldrich; Merck KGaA. The antibodies used for western blotting and immunohistochemistry in this study were purchased from Santa Cruz Biotechnology, Inc., R&D Systems, and Cell Signaling Technology, Inc., including antibodies against STAT1, phosphorylated STAT1 (p-STAT1), TNF-α, and IFN-γ. The antibodies used for flow cytometry, such as those recognizing CD4, CD25, Foxp3, and IL-17A were purchased from BioLegend, Inc. and BD Pharmingen; BD Biosciences. Fetal bovine serum (FBS) was purchased from Gibco; Thermo Fisher Scientific, Inc.

Plasmid pCYP2D6, the expression vector carrying the cDNA encoding human CYP2D6, was constructed by the insertion of cDNA into plasmid pcDNA3.1 (Invitrogen; Thermo Fisher Scientific, Inc.) in our laboratory. For in vivo gene transfection, pCYP2D6 was injected to mice via the tail vein using the hydrodynamics-based gene delivery technique (32).

Specific pathogen-free (SPF) male C57BL/6 mice (6–8 weeks old; 18–20 g) were purchased from Beijing Vital River Laboratory Animal Technology Co., Ltd. The mice were housed in an SPF environment at 24±2°C with an alternating 12-h light/dark cycle at the Experimental Animal Center of the Tongji Medical College. A total of 120 mice were used in the whole experiment and 6 mice were assigned to each experimental group except when otherwise indicated in the figure legends. Ninety-six of those mice were used to study the effect of tofacitinib in the ConA-induced immune-mediated liver injury and the remaining mice were used to confirm the effect of tofacitinib on liver fibrosis in an AIH mouse model. For all mouse experiments, the method of euthanasia was cervical dislocation. The protocol was approved by the Ethics Committee of Animal Experiments of Tongji Medical College and monitored by the Department of Experimental Animals of Tongji Medical College.

The treatment was as follows for the ConA mouse model: The mice were injected with a single dose (15 mg/kg of body weight) of ConA via the tail vein to induce acute immune-mediated liver injury, as described in a previous study (12). Three days before the injection of ConA, the mice in the treatment groups were administered with tofacitinib (5, 10 and 15 mg/kg/day) by gavage based on the recommended dose described in a previous study (33). The mice were sacrificed at 12, 24 and 48 h post-ConA injection. The blood was collected from the angular vein and the livers were collected for hematoxylin and eosin (H&E) staining, immunohistochemistry (IHC), western blot analysis, and quantitative polymerase chain reaction (qPCR). For the AIH mouse model: To detect the effects of tofacitinib on the mouse model of chronic immune-mediated hepatitis, adenovirus (10⁹ pfu; Viraltherapy Technology) was injected once initially and then pCYP2D6 plasmid was transfected several times (50 µg per injection) into mice to induce the AIH mouse model, as described in our previous studies (34,35). Thirty-six days after the adenovirus injection, the AIH mice were administered with tofacitinib (10 mg/kg/day, 2 days apart) by gavage and were then sacrificed two weeks later to observe the level of liver fibrosis by Sirius red staining.

The entire left lobe of the mouse livers was excised and fixed in 4% paraformaldehyde for at least 24 h, embedded in paraffin, and cut to yield 5-µm-thick sections. Following hydration in a decreasing ethanol gradient, all sections were deparaffinized and stained with Harris hematoxylin solution for 5 min at 37°C. For IHC, paraffin-embedded liver tissue samples were cut into 5-µm-thick consecutive sections, dewaxed in xylene, and rehydrated in graded ethanol solutions. Then, the nonspecific binding sites in the tissues were blocked, and the steam cooking method was used for antigen retrieval. The sections were incubated with the following primary antibodies: Anti-p-STAT1 (1:50; product no. 7649; Cell Signaling Technology, Inc.); anti-IL-6 (1:100; cat. no. GB11117) and anti-IL-10 (1:100; cat. no. GB11108; both from Servicebio Technology Co., Ltd.) overnight at 4°C. Then, they were incubated with horseradish peroxidase-conjugated polyclonal goat anti-rabbit secondary antibodies (1:200; GB23303; Servicebio Technology Co., Ltd.) for 1 h at room temperature. Finally, the sections were counterstained with hematoxylin.

Cell extracts and liver tissue samples were digested in RIPA buffer containing a phosphatase inhibitor cocktail and PMSF (Wuhan Boster Biological Technology, Ltd.). These samples were centrifuged at 12,000 × g for 30 min and the supernatant was retained. The protein concentration was determined using the bicinchoninic acid method. Then, 30 µg of each protein sample was separated on 10% SDS polyacrylamide gels; the protein bands were then transferred onto PVDF membranes. The membranes were blocked with 5% non-fat milk (non-fat dry milk powder dissolved in Tris-buffered saline with Tween-20) for 1 h at room temperature and then incubated overnight at 4°C on shaking tables with the following primary antibodies: Anti-STAT1 (1:1,000; product no. 14994) and anti-p-STAT1 (1:1,000; product no. 7649; both from Cell Signaling Technology, Inc.); anti-IFN-γ (1:2,000; cat. no. AF-485; R&D Systems); anti-TNF-α (1:500; cat. no. sc-12744; Santa Cruz Biotechnology, Inc.); and anti-β-actin (1:2,000; Promoter Biotechnology Ltd.). Next, the membranes were washed and incubated with horseradish peroxidase-conjugated secondary antibodies (1:1,000; Promoter Biotechnology Ltd.) for 1 h at room temperature. β-actin was used as a loading control. The expression of the antibody-linked proteins was visualized by enhanced chemiluminescence using an ECL assay kit (Wuhan Boster Biological Technology, Ltd.) and analyzed using ImageJ V1.48 (National Institutes of Health).

Total RNA was isolated from the liver tissue samples and cells using TRIzol reagent (Takara Bio, Inc.) and transcribed into cDNA using a reverse transcription kit (cat. no. RR036A; Takara Bio, Inc.), according to the manufacturer's protocols. qPCR was performed according to the following steps: 40 cycles at 95°C for 30 sec and 60°C for 30 sec; the PCR analyses were performed using Maxima SYBR-Green qPCR Master Mixes (Takara Bio, Inc.) on an ABI StepOne Real-Time PCR system (Applied Biosystems; Thermo Fisher Scientific, Inc.). Fold significance was determined by the 2^ΔΔCq method, as described in a previous study (36). The primers used for real-time qPCR are listed in Table I.

Venous blood was collected from the angular vein of the mice and then centrifuged at 500 × g for 10 min to obtain serum as the supernatant. The activities of liver enzymes, including aspartate transaminase (AST) and alanine transaminase (ALT), in the collected mouse sera, were determined using an automatic biochemistry analysis apparatus at the Clinical Laboratory of Tongji Hospital. The cytokines (IL-2, IL-4, IL-6, IFN-γ, TNF-α, IL-17A, and IL-10) in the serum were detected by flow cytometry using a mouse cytometric bead array (CBA) kit (560485; BD Biosciences).

To isolate the non-parenchymal cells (NPCs) from the liver, first, in situ perfusion with DMEM/F12 (Thermo Fisher Scientific, Inc.) containing type IV collagenase (Sigma-Aldrich; Merck KGaA) was performed. Mice in each experimental group were anesthetized using a mixture of ketamine (100 mg/kg) and xylazine (10 mg/kg) by intraperitoneal injection. The liver was then perfused, ground, digested, and filtered. The filtrate was centrifuged at 20 × g for 5 min and washed in DMEM/F12. The resuspended cells were processed with 30% Percoll (Sigma-Aldrich; Merck KGaA) and gently overlaid onto 70% Percoll. After density gradient centrifugation at 1,000 × g for 30 min, liver MNCs were harvested from the interface of the Percoll gradient (37). The MNCs collected were used for further fluorescence-activated cell sorting (FACS) analysis.

Mice spleens were ground and filtered with 70 µm cell strainer in PBS. After elimination of erythrocytes by Red Blood Cell Lysis Buffer (BD Biosciences), splenocytes were washed and resuspended in PBS for further FACS analysis.

A single-cell suspension of MNCs (at least 10⁶ cells/tube) was resuspended in phosphate-buffered saline (PBS) containing 1% BSA. To reduce nonspecific fluorescent staining, the cells were incubated with anti-mouse CD16/CD32 (cat. no. 101319; BioLegend, Inc.), which blocked the Fcγ III/II receptor. Then, the surfaces of the cells were stained with fluorochrome-conjugated antibodies for 30 min on ice. The following antibodies were used: FITC-conjugated anti-CD4 (cat. no. 557307), APC-conjugated anti-CD25 (cat. no. 557192), BV421-conjugated anti-Foxp3 (cat. no. 562996), and PE-conjugated anti-IL-17A (cat. no. 559502; all from BD Biosciences). The cell samples were assessed on a FACS Calibur flow cytometer (BD Immunocytometry Systems) and the data were analyzed using the FlowJo software V10 (Tree Star, Inc.).

All data are expressed as the mean ± standard error and all experiments were performed independently in triplicate. One-way analysis of variance (one-way ANOVA) with Tukey's multiple comparisons test or Dunnett's multiple comparisons test if the P-value was significant, were used. Statistical analysis was performed with GraphPad Prism 5.0 (GraphPad Software). P<0.05 was considered to indicate a statistically significant difference.

To establish a mouse model of immune-mediated liver injury, mice were injected with ConA (15 mg/kg) via the tail vein and sacrificed at 0, 6, 12, 24, and 48 h to observe the degree of inflammation in the livers. The liver tissues were harvested, and the levels of inflammation were analyzed by H&E staining (Fig. 1A). As revealed in the representative images, a handful of inflammatory cells began to infiltrate around the portal area 6 h post-injection, and more inflammatory cells had accumulated at 12 h post-injection. In addition, abundant necrotic areas were observed at 24 h after the ConA injection. The plasma ALT and AST levels increased 6 h after the intravenous ConA administration and peaked at 12 h (Fig. 1B and C). Real-time PCR was also used to detect the mRNA levels of the important inflammatory cytokines TNF-α and IFN-γ. Their expression patterns revealed clear tendencies, peaking at 12 h after the ConA injection (Fig. 1D).

To further explore whether the JAK1/STAT1 pathway is involved in ConA-induced acute immune-mediated liver injury, western blotting and immunohistochemical staining were used to detect the expression of STAT1 and p-STAT1 in the mouse livers at different time-points (Fig. 2A and B). The expression of p-STAT1 increased at the 12-h time-point, and further significantly increased at 24 h. However, the p-STAT1 levels revealed a decreasing trend at the 48-h time point, indicating that the JAK1/STAT1 pathway plays an important role in the progression and development of ConA-induced acute liver injury in mice. These results revealed that ConA could in fact induce acute hepatitis in mice and that the JAK1/STAT1 pathway was activated in this immune-mediated hepatitis mouse model, indicating that this pathway may play a key role in ConA-induced mouse hepatitis.

To detect the effects of tofacitinib on ConA-induced hepatitis in vivo, the mice were randomly divided into different experimental groups. The mice in the treatment groups received different doses of tofacitinib (5, 10, and 15 mg/kg/day) three days before the ConA injection to ascertain whether tofacitinib exerts protective effects and whether its effects are dose-dependent. The mice in each group were sacrificed 24 h post-ConA injection. The mice injected with ConA developed hypothermia, becoming unresponsive and inactive. They curled up in the cage and were unusually quiet. However, the body temperature of the mice in the treatment group was much higher and these mice were in better condition. H&E staining revealed that tofacitinib alleviated the histological liver damage (Fig. 3A) and significantly decreased the upregulated serum transaminase expression (Fig. 3B) in the ConA-injected mice. The gross appearance of the liver also provided strong evidence of the inflammation-relieving effect of tofacitinib from another perspective (Fig. 3C). Moreover, liver damage was alleviated to a greater extent in the mice administered with 10 or 15 mg/kg/day of tofacitinib than in those administered with 5 mg/kg/day of tofacitinib (data not shown). However, there was no significant difference between the experimental groups comprised of mice administered with 10 mg/kg/day of tofacitinib and those administered with 15 mg/kg/day of tofacitinib; thus, only the results of the group comprised of the mice that were administered with 10 mg/kg/day of tofacitinib were presented.

A mouse CBA kit was used to detect important cytokines (IL-2, IL-4, IL-6, IFN-γ, TNF-α, IL-17A and IL-10) in the serum of mice from each experimental group; the data are presented in Fig. 4A. As observed in the results, the expression of some pro-inflammatory cytokines, such as IL-2, IL-6, IFN-γ, TNF-α, and IL-17A, was upregulated after the ConA injection and revealed a downward trend in mice from the tofacitinib treatment group, while the expression of the anti-inflammatory cytokines, such as IL-4 and IL-10, increased after the tofacitinib treatment. The protein expression levels of TNF-α and IFN-γ significantly increased in the whole liver extracts from mice in the ConA-treated groups, while the pretreatment with tofacitinib downregulated these expression levels (Fig. 4B). Considering that IL-6 and IL-10 are the representative cytokines contributing to the distinct differentiation of T cells, immunohistochemical staining was further used to confirm the results presented in Fig. 4A (Fig. 4C). These results collectively confirmed that tofacitinib regulated the secretion level of anti- and pro-inflammatory cytokines under conditions of immune-mediated hepatitis.

The imbalance of Tregs and Th17 cells plays a key role in the pathogenesis of many immune-mediated diseases; this imbalance may be regulated by different types of cytokines. ConA can induce severe inflammation in the mouse liver and cause the upregulation of pro-inflammatory cytokine expression in the serum. In addition, many autoimmune diseases are characterized by systemic disorders. Therefore, the ratio of Treg/Th17 cells was detected not only in the mouse liver, but also in the spleen, which is representative of the situation in the peripheral regions. ConA induced an increase in the number of Th17 cells in both the liver and spleen (Fig. 5A and C); the number of Tregs also increased. This increase occurred because Tregs, which are immunoregulatory cells, respond to the acute inflammation caused by ConA. Notably, the number of Tregs in mice from the treatment group exhibited an even greater increase, while the number of Th17 cells in the same group exhibited a downward trend. Although the number of both Tregs and Th17 cells exhibited an increasing tendency, the ratio of Treg/Th17 cells decreased following ConA injection (Fig. 5B and D). However, in the tofacitinib treatment group, the impaired Treg/Th17 cell ratio was significantly recovered.

Although the mouse model of ConA-induced immune-mediated liver injury resembles conditions of AIH in humans (38), chronic inflammation and liver fibrosis are not observed in this model. Evidence has revealed that the balance of Tregs and Th17 cells is involved in the liver fibrosis associated with chronic immune-mediated hepatitis (39,40). Therefore, an AIH mouse model was used to further explore the effects of tofacitinib under conditions of liver fibrosis during immune-mediated hepatitis. Evident liver fibrosis appeared in the livers of mice with AIH and was in fact alleviated following tofacitinib treatment; representative Sirius red staining images are presented in Fig. 6. These data confirmed, from another point of view, that tofacitinib could suppress immune-mediated liver injury.