LX-2 cells (Millipore Cat# SCC064, RRID: CVCL_5792) were cultured in 100 mm-cell culture dish or seeded at 10⁵ cells per ml in 6-Well dishes or 10⁴ cells per ml in glass coverslips placed in 24-well dishes, containing DMEM medium supplemented with 10% fetal bovine serum (FBS), 1% penicillin/streptomycin and grown in a 5% CO₂ atmosphere at 37°C.

To determine whether GTIN can rescue LX-2 cells activation, we set up different processing groups:

H₂O₂ group: LX-2 cells were treated with 75 μmol H₂O₂ for 2 h (Chengdu Kelong Chemical Co., Ltd. 7722-84-1, H₂O₂ diluted with serum-free medium). Replaced fresh medium and continued to culture for 22 h (Cheng et al., 2015; Chen et al., 2022).

H₂O₂+GTIN group: LX-2 cells were treated with 75 μmol H₂O₂ (H₂O₂ diluted with serum-free medium) with 40 μmol GTIN for 2 h. Replaced fresh medium with 40 μmol GTIN and continued to culture for 22 h.

Control group: LX-2 cells were cultured with equal-volume DMEM for 2 h. Replaced fresh medium and continued to culture for 22 h.

The LX-2 cells were seeded at 10⁴ cells in 96-Well dishes. The H₂O₂ group was treated with different concentration gradients of H₂O₂ solution (0, 50, 75, 100 μmol). The H₂O₂+GTIN group was treated in an equally graded H₂O₂ solution containing 40 μmol GTIN. The control group was treated with equal-volume DMEM. Two hours later, the H₂O₂ group and the control group were replaced with fresh medium. The H₂O₂+GTIN group was replaced with fresh medium containing 40 μmol GTIN. After 24 h, the proliferation-toxicity test of LX-2 cells was detected by Cell Counting Kit-8 (CCK-8, BS350A, Biosharp) according to the manufacturer’s instructions.

Samples were lysed with ice-cold NP-40 lysis Buffer containing Phosphatase inhibitor cocktail A. After high-speed centrifugation, the protein supernatant was obtained, which was separated by SDS-PAGE polyacrylamide gel electrophoresis, transferred to polyvinylidene fluoride (PVDF) membrane, blocked by 5% nonfat-milk powder in Tris Buffered Saline with Tween 20 (TBST), incubated with primary antibody at 4°C overnight and the coupled secondary antibody at room temperature for 2 h. Finally, the image was developed by Enhanced chemiluminescence kit. The primary antibodies were α-Smooth Muscle Actin (α-SMA, Cell Signaling Technology Cat# 19245, RRID: AB_2734735), Phospho-p38 MAPK (Thr180/Tyr182) (P-p38 MAPK, Cell Signaling Technology Cat# 4511 (also 4511S), RRID: AB_2139682), Phospho-MKK3 (Ser189)/MKK6 (Ser207) (P-MKK3/MKK6, Cell Signaling Technology Cat# 12280, RRID: AB_2797868), Tumor protein 53 (p53, Cell Signaling Technology Cat# 2524, RRID:AB_331743) and glyceraldehyde-3-phosphate dehydrogenase (GAPDH, Abcam Cat# ab9485, RRID: AB_307275).

Cells growing on glass coverslips were fixed in 4% PFA solution at room temperature for 20 min, 0.1% Triton X-100 Phosphate Buffer Saline solution inducing cell membrane permeabilization, 1% Bovine Serum Albumin (BSA) Phosphate Buffer Saline with Tween 20 (PBST) solution blocked at room temperature. The primary antibody was incubated at 4°C overnight, abstersion. The coupled secondary antibody was incubated at room temperature for 1 h kept in dark place. Antifade Mounting Medium with 4′,6-diamidino-2-phenylindole (DAPI, P0131, Beyotime Biotechnology) incubated cell for 1 min. The primary antibody was Collagen type I alpha 1 chain (Col1a1, Cell Signaling Technology Cat# 72026, RRID: AB_2904565). Fluorescent secondary antibody was Goat Anti-Rabbit IgG H&L (Alexa Fluor ^® 488, Abcam Cat# ab150077, RRID: AB_2630356).

Eighteen SPF grade 5-week healthy wild-type male C57BL/6J mice, 25 ± 2 g, purchased from Gempharmatech, Jiangsu, without abnormal breathing, coarse hair, abnormal behavior or hunchback posture. They were kept in a clean room with 12 h sunshine, a temperature of about 25°C, and a relative humidity of about 40%–70%. They were free to eat water and forage for food. Follow-up experiments were conducted after 1 week of adaptation. After passing the experimental animal welfare ethical review of Chengdu Medical College, the mice were raised in the clean Laboratory Animal Room.

The mice were randomly divided into 3 groups (n = 6 in each group). CCl₄ (C805332, Macklin) was dissolved in Olive oil (O815210, Macklin), and mixed evenly. GTIN powder (B29179, Yuanyebio) was dissolved in Dimethyl sulfoxide (DMSO, D8371, Solarbio). Both CCl₄ and GTIN were administered by intraperitoneal (i.p.) injection.

CCl₄ group: Each mouse was injected with CCl₄ solution for 6 weeks (CCl₄: Oil = 1:4, 0.75 μ L/g, twice a week). In the last week, each mouse was injected with 300 mL of PBS daily. (Masuda et al., 2020; Guo et al., 2023; Xia et al., 2023).

CCl₄+GTIN group: The injection time and method of CCl₄ were the same as those of the CCl₄ group. In the last week, each animal received a daily i.p. injection of GTIN (10 mg/kg). (Samant and Gupta, 2022; Oh et al., 2023).

Control group: The control group was injected with equal volume PBS.

The mice were anesthetized with 4% chloral hydrate (100 mL/kg i.p.). After cutting off the whiskers, the eyeballs were pressed, tweezed, quickly removed and the blood was collected into 15 mL centrifuge tubes, placed at room temperature for 30 min, and centrifuged at 4°C (2000 rpm, 20 min). The serum was carefully pipetted into 1.5 mL tubes, marked, and stored at −20°C for subsequent serological assays.

After blood collection, each mouse was immediately injected with pre-cooled PBS solution through the heart and continued with pre-cooled 4% paraformaldehyde (PFA) until liver discoloration. The brain and liver were gently extracted and fixed in PFA overnight and were washed with water for 6 h and stored in 75% ethanol.

Serum alanine aminotransferase (ALT) and aspartate transaminase (AST) activity was detected using Mouse ALT ELISA Kit (MM-0260M1, Meimian) and Mouse AST ELISA Kit (MM-44384M1, Meimian) according to the standardized procedures suggested by the manufacturer. Both ALT and AST were shown in units per liter (U/L).

Total RNA was extracted from fresh liver tissues by Trizol Reagent (T9424, Sigma). Reverse transcription was performed using PrimeScript™ RT reagent Kit with gDNA Eraser (RR047A, Takara), according to the manufacturer’s instructions. qPCR with TB Green^® Premix Ex Taq™ II (RR820A, Takara) and primers, according to the manufacturer’s instructions. GAPDH was used as the internal reference. 2^ (- ∆ ∆ Cq) was used for statistical analysis. The sequences of primers are shown in Table 1.

Quantitative data were statistically analyzed by GraphPad prime (v9.1.1, RRID:SCR_002798, United States). All numerical variables are expressed as (Means ± SEM). Group differences were analyzed with One-way ANOVA. p < 0.05 was considered a statistical difference.

As oxidative stress plays an important role in the development of liver fibrosis (Singal et al., 2011), we used hydrogen peroxide to stimulate LX-2 cells to mimic liver injuries caused by oxidative stress. Increased H₂O₂ concentrations lead to decreased LX-2 viability, which can be rescued by 40 μmol GTIN (Figure 1A). The expression of α-SMA protein was increased upon H₂O₂ treatment (Figures 1B,C), indicating that LX-2 cells were activated by oxidative stress. The use of GTIN restored α-SMA protein expression (Figures 1B,C). Similarly, immunofluorescence revealed that increased Col1a1 expression in H₂O₂-stressed LX-2 cells was abolished by GTIN treatment (Figures 1D,E). These results indicated that GTIN could suppress the LX-2 activation induced by oxidative stress.

To clarify the effect of GTIN on pre-existing liver fibrosis, we established a liver fibrosis model by intraperitoneal injection CCl₄ in male C57BL/6J mice for 6 weeks (Figure 2A). The CCl₄+GTIN group received intraperitoneal injections of GTIN once a day in the last week of CCl₄ injection. Mice injected with CCl₄ became lethargic, and their hair gradually began to lose its luster. The average weight gain of mice in CCl₄ group was lower than those in the control group from the fifth week to the end of the experiment (Figure 2B). At the end of the 6-week course of the experiment, behavioral tests were performed on the three groups of mice. In elevated plus maze, CCl₄ decreased the percentage of visits to the open arm and GTIN ameliorated this (Figures 2C,D). Tail suspension experiment showed a significantly increased immobile time of mice treated with CCl₄, and GTIN rescued this increase (Figure 2E). These results suggest that anxiety and depression like behavior induced by CCl₄ could be rescued by short-term GTIN treatment. In the new object recognition experiment, mice in the CCl₄ group spent less time exploring new objects and had lower recognitive indexes, which were improved by GTIN (Figure 2F). The lack of statistical significance may be because cognition-related experiments have greater variations within the group and require more mice in the experiments. Likewise, the spontaneous alternating rate of Y-mazes test in CCl₄ and CCl₄+GTIN mice decreased compared to the control group (Figure 2G). On the contrary, CCl₄ does not influence motor performance according to open-field and pole test (Figures 2H–J). After the behavioral experiment, mice were euthanized, serum, liver tissues and brain tissues were collected for subsequent analysis (Figure 2A). Compared with the control group, serum aspartate aminotransferase (AST) and alanine aminotransferase (ALT) enzyme activities in CCl₄ group dramatically increased, indicating impairment of liver function (Figures 2K,L). Short-term GTIN treatment reduced both AST and ALT enzyme activity, though the decrease of ALT is not statistically significant (Figures 2K,L). These results suggested that GTIN could alleviate CCl₄-induced behavioral change and liver function decline in mice.

We examined the appearance of the whole liver. The livers from the control group were smooth with soft texture and reddish-brown color, while the livers from CCl₄ and GTIN groups were enlarged with mottled surface and granulated texture. Hematoxylin and Eosin (HE) staining (Figure 3A) showed that the control group has normal liver lobular structure. In CCl₄ group, the structure of hepatic lobules was preserved, but the volume of hepatic cells in the lobules was increased, hepatocytes were crowded, hepatic sinuses were narrow, red fine particles could be seen in the cytoplasm, and cell edema appeared. There was more inflammatory cell infiltration in the sink area and a small amount of fibrous tissue hyperplasia in the interlobular. After short-term treatment with GTIN, the degree of hepatocyte edema and inflammatory cell infiltration was reduced. Sirius red staining showed that these proliferating collagen fibers were type 1 collagen fibers with a strong orange-yellow color (Figure 3B). GTIN treatment decreased Sirius red positive collagen fiber content but not statistically significant (Figure 3B). We also use Col1a1 as another marker of liver fibrosis and observed solid alleviation of liver fibrosis upon GTIN treatment (Figure 3C). IHC of α-smooth muscle actin (α-SMA), a marker of HSCs activation, got similar results, indicating that GTIN effectively suppressed the activation of HSCs (Figure 3D). Kupffer cells (KCs) directly regulate the activation of HSCs and promote the development of liver fibrosis. KCs differentiate into different subtypes under different stimuli and have different roles (Wang et al., 2021). The two most common subtypes are CD68-positive KCs with phagocytic function, and CD11b-positive with cytokine secretion function (Kinoshita et al., 2010; Wen et al., 2021; Kulle et al., 2022). Therefore, we examined these two types of KCs in different groups. In CCl₄ group, both were significantly increased, located around the hepatic lobules (Figures 3E, F). GTIN reduced both CD68 and CD11b positive KCs (Figures 3E, F).

Collectively, our results suggest that short-term of GTIN treatment is sufficient to alleviate pre-existing liver fibrosis induced by CCl₄.

p38 MAPK signaling plays an important role in the development of liver diseases (Cicuéndez et al., 2021). Upon activation by Ser189/Thr222 phosphorylation of MKK3 and Ser207/Thr211 of MKK6, they activate downstream p38 MAPK by phosphorylation of Thr180 and Tyr182 of p38α (Han et al., 1996; Raingeaud et al., 1996; Ge et al., 2002). As GTIN was shown to be an inhibitor of MKK3/6 signaling in esophageal cancer (Xie et al., 2019), we investigated whether GTIN could regulate MKK3/6-p38 MAPK signaling in our mice model. The expression of P-MKK3/MKK6 and P-p38 MAPK in CCl₄ group was significantly increased compared to control, especially in the cytoplasm of hepatocytes surrounding the hepatic central vein (Figures 4A,B). GTIN treatment decreased CCl₄-induced activation of P-MKK3/MKK6 and P-p38 MAPK (Figures 4A,B). Since p38 MAPK could regulate cell survival and cell death through p53, we detected the expression of p53 in the liver of our mice. As shown in Figure 4C, the expression of p53 protein was significantly increased in CCl₄ group, indicating that apoptosis occurred, whereas GTIN could suppress p53 activation by CCl₄. WB results showed that the expression of α-SMA, P-MKK3/MKK6, P-p38 MAPK, p53 in liver was significantly upregulated by CCl₄, and the expression of α-SMA, P-MKK3/MKK6, P-p38 MAPK could be rescued by GTIN (Figure 4D). During the immune response, p38α regulates the phosphorylation of various enzymes, transcription factors, regulatory proteins, and DNA/RNA binding proteins, which can directly lead to the production and secretion of inflammatory factors (Canovas and Nebreda, 2021), thus we examined the expression of cytokines in liver tissue. IL-1α, IL-6, IL-10, and TNF-α were all upregulated due to CCl₄ stimulation, while the IL-1α and TNF-α mRNA level were significantly reduced after GTIN treatment (Figure 4E).

These results suggest that GTIN suppresses CCl₄-induced activation of liver MKK3/6-p38 MAPK signaling pathway, leading to ameliorated p53 overactivation and reduced hepatic inflammation.

As an organ that plays important roles in metabolism and detoxification, chronic liver disease often influences other systems including the nervous system. A plethora of clinical studies revealed that chronic liver diseases were associated with neurological symptoms including anxiety, depression, and accelerated cognitive decline (Filipović et al., 2018; Labenz et al., 2020; Jiang et al., 2023). Thus, we investigated whether there were pathological alterations in the brain of hepatic fibrosis mice. IHC staining of Iba1 showed that compared with the control group, the number and area of Iba1-positive microglia in hippocampal (CA1) region and amygdala region in CCl₄ group was significantly increased (Figures 5A,B). CCl₄ induction led to the increase in the volume of microglia cells in the hippocampus and the decrease in the number of processes, called reactive microglia. After one-week of GTIN administration, microglia returned to a resting state (Figure 5A). The number and morphology of microglia in striatum and cortex did not change significantly (Figures 5A,B). Astrocytes in the same brain area were detected by GFAP IHC staining. Compared with the control group, the number and area of GFAP-positive astrocytes in hippocampal (CA1) and striatum of mice in CCl₄ group increased (Figures 5C,D). Increased and elongated astrocyte branches were also observed in the hippocampus, indicating occurrence of astrocytes in this area (Figure 5C). GTIN treatment not only reduced the number of astrocytes but also rescued the morphological change of astrocytes in the hippocampus (Figures 5C,D). In the amygdala and cortex, the number and morphology of astrocytes did not change significantly (Figures 5C,D).

Collectively, our results show that short-term GTIN treatment ameliorates neuroinflammation in the hippocampus of CCl₄-treated mice, as indicated by decreased microglial and astrocyte activation.

We also detected the number of neurons in brain regions including hippocampus, amygdala, striatum, and cortex using Nissl staining. No difference of Nissl bodies in these brain regions of control and CCl₄-treated mice was observed (Figures 6A,B). Thus, CCl₄ treatment did not cause neuronal loss in the above brain area.