BBR were purchased from Coolaber Reagent (Beijing, China). GEF was purchased from MACKLIN Reagent (Shanghai, China). TUNEL Apoptosis Detection Kit (FITC) was purchased from YEASEN (Shanghai, China), and immunostaining permeabilization solution (Triton X-100) were purchased from Beyotime Reagent (Shanghai, China). Human and rat tumor necrosis factor-alpha (TNF-α), interleukin-1 beta (IL-1β), and interleukin-6 (IL-6) enzyme-linked immunosorbent assay (ELISA) kits were purchased from Thermo Fisher Scientific (Massachusetts, United States) and Elabscience Biotechnology (Wuhan, China), respectively. The primers for RT-qPCR were synthesized by Qingke Biotechnology (TSINGKE, Shenzhen, China). The specific inhibitors for NF-κB (SC75741) and TLR4 (TLR4-IN-C34) were purchased from Selleck. cn (Shanghai, China). GAPDH antibody, rabbit anti-HMGB1 antibody, rabbit anti-TLR-4 antibody were purchased from Proteintech (Wuhan, China). Rabbit anti-NF-κB (p65) and rabbit anti-p-NF-κB [p-p65(Ser536)] antibody were purchased from Affinity Biosciences (Jiangsu, China). HRP-Goat Anti-Rabbit IgG (H + L) and Goat-Anti-Rabbit IgG/Alexa Fluor 488 was purchased from ZSGB. BIO (Beijing, China). (Specific antibody information is provided in Supplementary Table S1).

The THLE-2 cells were obtained from Xiamen Immocell Biotechnology Co.,Ltd. (Xiamen, China), and the THLE-3 cells were provided by Professor Wang Ling’s team (Sichuan University, Chengdu, China). All cells were cultured in basic DMEM (Thermo Fisher Scientific, American) supplemented with 10% fetal bovine serum, 100 μg/mL of penicillin/streptomycin mixture at 37 °C in a 5% of CO₂ atmosphere.

Based on preliminary CCK-8 cytotoxicity assays and referenced literature reports (Chen et al., 2019; Xu et al., 2018), we selected appropriate drug concentrations for the study: gefitinib (10 μM) for hepatotoxicity induction and berberine (5, 10, 20 μM) for evaluating protection against gefitinib-induced injury.

Male SD rats (4–5 weeks old, 100–150 g) were purchased from Chongqing Tengxin biotechnology co., LTD (Chongqing, China). The animals were kept in polycarbonate cages (6/cage) under 12 on/off light cycles, feed and water ad libitum. All animal experiments were conducted in accordance with the principles outlined in the Guide for the Care and Use of Laboratory Animals by the National Institutes of Health and were approved by the Animal Ethics Committee of Chongqing University Cancer Hospital (Chongqing, China), in compliance with the ARRIVE guidelines.

Following a 7-day quarantine period, rats were randomly assigned to one of six groups (n = 8 per group): (1) Blank control (no treatment); (2) Vehicle control (0.5% sodium carboxymethyl cellulose, CMC-Na with 0.04% Tween 80); (3) GEF (100 mg/kg); (4) GEF + low-dose berberine (BBR-L, 25 mg/kg); (5) GEF + medium-dose berberine (BBR-M, 50 mg/kg); and (6) GEF + high-dose berberine (BBR-H, 100 mg/kg). The gefitinib-induced rat liver injury model and berberine co-administration dosing regimens were based on results from preliminary in vivo experiments and relevant published studies (Chen et al., 2021; Mehrzadi et al., 2018; Mahmoud et al., 2017). Due to limited solubility, GEF and BBR were administered orally by gavage as suspensions in 0.5% CMC-Na with 0.04% Tween 80. The GEF group received GEF and the CMC-Na/Tween 80 vehicle. In the GEF + BBR groups, BBR was administered via gavage, followed 4 h later by gefitinib (100 mg/kg). All treatments were administered daily for 21 days. Body weight was measured daily throughout the treatment period. At the conclusion of the 21-day treatment, rats were anesthetized via pentobarbital injection, and blood samples were collected via cardiac puncture. Livers were then excised, weighed, and snap-frozen in liquid nitrogen before being stored at −80 °C until further analysis. A schematic representation of the experimental design is provided in Figure 5A.

The alanine aminotransferase (ALT) and aspartate aminotransferase (AST) levels in rat serum were determined by Scientist BIOTECHNOLOGY (Sichuan, China). For cell supernatants, ALT and AST levels were measured using Alanine Aminotransferase (ALT/GPT) Activity Assay Kit and Aspartate Aminotransferase (AST/GOT) Activity Assay Kit which were purchased from Elabscience Biotechnology (Wuhan, China).

The liver tissues were fixed in neutral formalin solution for at least 48 h, then embedded in paraffin, and sectioned at a thickness of 4 μm for histopathological evaluation. The tissue sections were stained with hematoxylin and eosin (HE), and the histopathological changes were observed under a light microscope (Precipoint-M8 Digital Scanning Microscopic Imaging System, Germany, 200×).

The THLE-2 and THLE-3 cells were transfected with 50 nM of HMGB1-targeting siRNAs (5′-GGA​GAG​AUG​UGG​AAU​AAC​ATT-3′, 5′-UGU​UAU​UCC​ACA​UCU​CUC​CTT -3′) (HanBio, Shanghai, China) using Lipofectamine™ 3000 (Thermo Fisher Scientific, United States) following the manufacturer’s instructions. Cells were collected for subsequent analysis 48 h post-transfection.

Total RNA was isolated from rat liver and THLE-2/THLE-3 cells using TRNzol reagent (TIANGEN, Beijing, China), and cDNA was synthesized using the HifairⅢfirst Strand cDNA Synthesis SuperMix for qPCR (gDNA digester plus) (YEASEN, Shanghai, China). RT-qPCR was performed using Hieff^® qPCR SYBR Green Master Mix (No Rox) (YEASEN, Shanghai, China) and primers specific for human/rat HMGB1/Hmgb1, TLR4/Tlr4, and GAPDH/Gapdh on the QX-Real-time PCR System (JLM, Sichuan, China). GAPDH/Gapdh served as the internal control, and relative expression was calculated using 2^−ΔΔCt method. The primer sequences employed throughout this study are listed in Table 1.

Total proteins were extracted from lysate of rat liver tissues and THLE-2/THLE-3 cells using RIPA buffer supplemented with PMSF. An appropriate amount of protein was loaded onto a 10% SDS-PAGE gel. Subsequently, the proteins were transferred to PVDF membranes. The membranes were then incubated with primary antibodies at 4 °C overnight. After incubation with HRP-Goat Anti-Rabbit IgG (H + L) secondary antibody, protein bands were visualized using an ECL detection kit (YEASEN, Shanghai, China) and imaged with a ChemiDoc MP imaging system (Bio-Rad, United States). The intensity of the bands was quantified by means of ImageJ software.

The levels of TNF-α, IL-1β, and IL-6 in the cell supernatants and rat serum were quantified using commercially available ELISA kits, according to the manufacturer’s instructions.

Cell apoptosis was detected using the TUNEL Apoptosis Detection Kit (FITC) (YEASEN, Shanghai, China). Cells were seeded on coverslips in 24-well plates and cultured for 48 h at 37 °C in a 5% CO₂ incubator. After fixing with 4% paraformaldehyde for 30 min and permeabilizing with Triton X-100, the cells were incubated with 1×Equilibration buffer for 20 min at room temperature. Next, TdT incubation solution was applied for 60 min at 37 °C in the dark. Following PBS washes, DAPI staining was performed for 2 min, and anti-fluorescence quencher (Beyotime, Shanghai, China) was added. TUNEL-positive cells were imaged using an Olympus IX73 microscope (400×).

HMGB1 nuclear-to-cytoplasmic translocation were assessed by immunofluorescence in THLE-2 and THLE-3 cells. Cells were seeded on coverslips in 24-well plates and cultured for 48 h at 37 °C in a 5% CO₂ incubator, then fixed with 4% paraformaldehyde for 30 min, permeabilized with Triton X-100, and blocked with 1% BSA. After overnight incubation with primary HMGB1 antibody (1:200, Proteintech, 10829-1-AP) at 4 °C, slices were treated with Alexa Fluor 488-conjugated secondary antibody (1:50, ZSGB-BIO, ZF-0511). HMGB1 localization was imaged using an Olympus IX73 microscope (200×, 400×).

Data analysis in this study was performed using GraphPad (version 9.5) and Origin (version 2021), and results are presented as mean ± standard deviation. The differences in means between the groups were assessed by Student’s t-test based on the data analysis. Statistical significance was established when the p-value was less than 0.05.

To initially investigate the mechanism of GEF-induced hepatotoxicity and the role of the HMGB1/TLR4/NF-κB pathway, THLE-2 and THLE-3 cells were exposed to GEF (10 μM) for 48 h. The mRNA expression of HMGB1 and TLR4 was measured by quantitative PCR (qPCR), and Western blot was used to assess the protein levels of HMGB1, TLR4, and NF-κB (including p65 and its phosphorylated form, p-p65). GEF treatment resulted in a significant upregulation of HMGB1 and TLR4 mRNA expression (Figures 1A,B). Additionally, the protein levels of HMGB1, TLR4, and NF-κB (p65 and p-p65) were markedly elevated (Figures 1C,D). Notably, the increase in p-p65 protein expression exceeded that of p65, indicating significant activation of NF-κB. These findings indicate that GEF activates these key molecular components.

Subsequently, THLE-2 and THLE-3 cells were co-treated with GEF and either HMGB1 small interfering RNA (siRNA), a TLR4 inhibitor (TLR4-IN-C34), or an NF-κB inhibitor (SC75741) for 48 h. The levels of ALT, AST, and inflammatory cytokines (TNF-α, IL-1β, and IL-6) in the cell supernatant were measured by ELISA. The results revealed that each co-treatment significantly reduced the elevation of ALT, AST, and pro-inflammatory cytokines induced by GEF (Figures 1E–H). These findings indicate that inhibition of HMGB1, TLR4, or NF-κB mitigates GEF-induced hepatocyte injury and inflammatory responses.

To clarify the upstream-downstream regulatory relationships among HMGB1, TLR4, and NF-κB in GEF-induced liver injury, THLE-2 and THLE-3 cells were pretreated with HMGB1 siRNA and subsequently exposed to 10 μM GEF for 48 h to assess the expression of downstream pathways, with a particular focus on TLR4 and NF-κB. The mRNA and protein expressions of the relevant molecules were measured using the methods described previously. The results revealed that following HMGB1 downregulation by siRNA, GEF treatment could not increase HMGB1 and TLR4 mRNA expression anymore (Figures 2A,B). Moreover, the protein levels of HMGB1, TLR4 and NF-κB (p65 and p-p65) upregulation induced by GEF were not observed (Figures 2C,D). This finding demonstrates that HMGB1 plays a crucial role in mediating GEF-induced activation of TLR4 and NF-κB, suggesting that the activation of this signaling pathway is a key mechanism driving the hepatotoxic effects of GEF.

To preliminarily explore the protective effect of BBR on GEF-induced liver injury in vitro, THLE-2 and THLE-3 cells were treated with GEF (10 μM) in combination with BBR at low, medium, and high concentrations (5 μM, 10 μM, and 20 μM, respectively) for 48 h. The levels of ALT, AST, and inflammatory cytokines (including TNF-α, IL-1β, and IL-6) in the cell supernatant were measured by ELISA, and cell apoptosis was assessed using the TUNEL assay.

The results demonstrated that co-treatment with BBR led to a dose-dependent decrease in the levels of ALT and AST in both cell lines (Figures 3A,B). Additionally, the levels of pro-inflammatory cytokines, including TNF-α, IL-1β, and IL-6 in the cell supernatants, were markedly reduced in a dose-dependent manner (Figures 3C,D). Higher concentrations of BBR leading to more significant reductions inflammatory responses, indicating that the protective effects of BBR are concentration-dependent. Furthermore, a significant reduction in cell apoptosis was observed with BBR treatment (Figures 3E,F; Supplementary Figures S1A,S1B), especially at high concentrations, which supports its role in mitigating GEF-induced hepatotoxicity.

Next, the molecular mechanism of BBR’s protection against GEF-induced liver injury was investigated. Accordingly, THLE-2 and THLE-3 cells were co-treated with 20 μM gefitinib and BBR at concentrations of 5, 10, and 20 μM. The mRNA expression levels of HMGB1 and TLR4, as well as the protein levels of HMGB1, TLR4, and NF-κB, were analyzed using the methods described previously. Moreover, the nuclear-to-cytoplasmic translocation of HMGB1 was evaluated via immunofluorescence. The results showed that, compared to GEF treatment alone, co-treatment with BBR significantly downregulated the mRNA levels of HMGB1 and NF-κB (Figures 4A,B), as well as the protein expression of HMGB1, NF-κB, and TLR4 (Figures 4C,D). Furthermore, BBR significantly reduced the nuclear-to-cytoplasmic translocation of HMGB1 in both THLE-2 and THLE-3 cells, a key event associated with inflammation and liver injury (Figures 4E,F; Supplementary Figures S1C,S1D). Therefore, BBR alleviates GEF-induced hepatotoxicity by inhibiting the HMGB1/TLR4/NF-κB signaling pathway, reducing inflammation, and limiting the nuclear-to-cytoplasmic translocation of HMGB1.

To further investigate the protective effect of BBR against GEF-induced liver injury in vivo, we established a rat model of GEF-induced liver injury by intragastric administration of 100 mg/kg GEF for 3 weeks. Following daily co-administration of BBR at low, medium, and high doses (25 mg/kg, 50 mg/kg, and 100 mg/kg, respectively), serum samples were collected for biochemical analysis of ALT and AST. Inflammatory markers, including TNF-α, IL-1β, and IL-6, were quantified using ELISA kits. Liver tissues were harvested for histopathological evaluation through HE staining. Additionally, mRNA expression levels of HMGB1 and TLR4, as well as protein levels of HMGB1, TLR4, and NF-κB, were measured as previously described (Figure 5A).

The results indicated that GEF treatment led to a reduction in body weight in rats (Figure 5B) and an increase in the liver index (liver-to-body weight ratio, Figure 5C). Pathological analysis through HE staining revealed hepatocellular injury, eosinophilic change, along with inflammatory cell infiltration (Figure 5D). Serum levels of transaminases (ALT, AST) and inflammatory cytokines (TNF-α, IL-1β, and IL-6) were also elevated (Figures 5E,F), indicating GEF-induced liver injury. These findings align with clinical reports of GEF-induced hepatotoxicity, which primarily manifests as hepatocellular injury (Zhou et al., 2024).

As depicted in Figures 5B–D, BBR co-treatment partially restored body weight and reduced the liver index and liver damage in a dose-dependent manner, with higher doses preserving liver architecture and reducing tissue injury more effectively. Serum ALT and AST levels were significantly lower in BBR-treated rats, with the greatest reductions observed at higher doses (Figure 5E). Furthermore, levels of pro-inflammatory cytokines, such as TNF-α, IL-1β, and IL-6, were reduced dose-dependently (Figure 5F), indicating that BBR mitigated the inflammatory response induced by GEF.

Mechanistically, BBR co-treatment effectively inhibited the activation of the HMGB1/TLR4/NF-κB pathway in rat liver. The mRNA and protein expression of HMGB1 and TLR4 was significantly reduced in the co-treated rats, with greater downregulation at higher doses (Figures 5G,H). Moreover, the protein expression of NF-κB was also significantly inhibited, particularly at the highest dose of BBR (Figure 5H). Consistent with in vitro data, BBR dose-dependently suppressed the HMGB1/TLR4/NF-κB axis, thereby attenuating liver inflammation and damage, and offering a promising therapeutic for GEF-induced hepatotoxicity.