Gefitinib, an epidermal growth factor receptor (EGFR) tyrosine kinase inhibitor developed by AstraZeneca, is widely used for treating advanced non-small cell lung cancer (NSCLC) and is clinically approved as a standard first-line therapeutic agent for patients with somatic EGFR mutation-positive advanced NSCLC (Hida et al., 2009). However, drug-induced liver injury of varying severity, including fatal liver failure, is often overlooked when gefitinib during long-term usage (Mok et al., 2009; Mitsudomi et al., 2010; Wang et al., 2016). Elevated levels of alanine aminotransferase (ALT) are more common among patients receiving gefitinib, and meticulous and periodic liver function test monitoring is mandatory to avoid severe hepatic impairment (Chen et al., 2012; Shah et al., 2013; Tan et al., 2021). Gefitinib is associated with apoptosis-induced hepatotoxicity; however, a clinically accessible intervention strategy remains lacking (Luo et al., 2021; Zhang et al., 2021b). Therefore, gefitinib should be administered with great care, especially when combined with other drugs.

Antipyretic analgesics may be used to alleviate cancer pain and cancer-related low-grade fever or even high fever. Recently, acetaminophen (APAP) or ibuprofen was often used clinically to relieve fever and pain associated with the coronavirus disease 2019 (COVID-19) infection in patients susceptible to COVID-19 (Rinott et al., 2020; Yousefifard et al., 2020). APAP is a commonly available safe analgesic and antipyretic agent, which can be easily acquired and is more likely to be used in combination with other drugs (Chapman et al., 2020). The conventional treatment dose is relatively safe in humans, but long-term or overdose APAP can result in severe liver injury or even death. In many Western countries, APAP-induced hepatotoxicity is the primary cause of acute liver failure and is associated with numerous deaths (Larson, 2007; Herndon and Dankenbring, 2014; Ramachandran and Jaeschke, 2019). Excess N-acetyl-p-benzoquinone-imine, an intermediate metabolism product of APAP, induces oxidative stress, dysfunction of mitochondria and DNA damage. The disruption and loss of hepatocyte function may lead to direct cell damage and death (Ramachandran and Jaeschke, 2018). Currently, relevant studies or clinical reports on the combination of gefitinib and APAP at therapeutic doses are lacking, and APAP may exacerbate gefitinib-induced hepatotoxicity.

In this study, we simulated the combination strategy in an animal model at clinical doses and confirmed that gefitinib in combination with APAP exhibits additional hepatotoxicity, suggested that this exacerbating effect is through the large-scale production of ROS, damaging hepatocytes, thereby mediating apoptosis. When the two drugs are taken together, the upregulation of p62 expression brought on by gefitinib is inhibited by APAP intervention, which causes Keap1 to accumulate. In the cytoplasm, free Nrf2 combines with Keap1 and then is degraded. The synthesis of antioxidant factors decreases, and ROS levels rise, ultimately resulting in ROS-mediated cell apoptosis. This study provides a feasible theoretical guidance for the combined use of gefitinib and APAP in clinical practice, and emphasizes the risk of aggravating hepatotoxicity.

Animals

We purchased 6-8 weeks old male Institute of Cancer Research (ICR) mice from Shanghai Experimental Animal Center (Shanghai, China). All experimental procedures and methods were approved by the Center for Drug Safety Evaluation and Research of Zhejiang University and were bred according to the Institutional Animal Care and Use Committee (IACUC) protocols of Zhejiang University (Approval No: IACUC-23-350). The mice were housed in the Zhejiang University animal facility under a 12 h light/dark cycle and were provided food and water ad libitum. Before performing in vivo experiments, the animals were allowed to acclimatize to the laboratory environment for approximately 1 week. The mice were sacrificed after they were subjected to 12 h of fast. Gefitinib (SML1657, Topscience, Shanghai, China) and APAP (T0065, Topscience) were dissolved in 0.5% sodium carboxymethyl cellulose (CMC-Na, 419273, Sigma-Aldrich, St. Louis, MO, USA) to obtain stock solutions. To investigate the hepatoxicity of the drug combination of gefitinib and APAP, the mice were divided into four groups of six animals each: vehicle control with 0.5% CMC-Na, 75 mg/kg of intragastrical gefitinib administered continuously for 5 weeks, 100 mg/kg of intragastrical APAP administered 3 days a week continuously for 5 weeks, and gefitinib plus APAP for 5 weeks.

Cell culture and treatment

Human hepatocyte cell line, HL-7702, was purchased from Jennio Biological Technology (JNO-048, Guangzhou, China). The human hepatoma cell line, HepG2, was obtained from the Cell Bank of China Science Academy (TCHu 72, Shanghai, China). HL-7702 cells were maintained in Roswell Park Memorial Institute (RPMI)-1640 (31800, Gibco, CA, USA), while HepG2 cells were maintained in Dulbecco’s Modified Eagle’s Medium (DMEM, 10569010, Gibco, NY, USA) supplemented with 10% fetal bovine serum (SH30396.03, Hyclone, Logan, UT, USA), 100 U/mL penicillin (Invitrogen, Carlsbad, CA, USA) and 100 μg/mL streptomycin (Invitrogen) in a humidified atmosphere with 5% CO₂ at 37°C.

Gefitinib (SML1657), N-acetyl-L-cysteine (NAC, A7250), and 3-Methyladenine (3-MA, 5142-23-4) were purchased from Sigma-Aldrich. APAP (T0065), necrostatin-1 (T1847) and ferrostatin-1 (T6500) were purchased from Topscience (Shanghai, China). Z-VAD-FMK (C1202) was purchased from Beyotime (Shanghai, China). Z-YVAD-FMK (HY-P1009) was purchased from MedChem Express (Shanghai, China). Throughout the experiments (except otherwise stated), HL-7702 cells and HepG2 cells were treated with 10 μM gefitinib and/or 1.25 mM APAP for 24 h. In selected samples, 20 μM Z-VAD-FMK, 10 μM Z-YVAD-FMK, 1 μM ferrostatin-1, 20 μM necrostatin-1, 1 mM 3-MA and 1 mM NAC were used.

Blood biochemistry analysis

Blood samples were collected and left for more than 2 h, and then centrifuged at 4000 rpm for 15 min to obtain serum for the determining alanine aminotransferase (ALT) and aspartate aminotransferase (AST) levels using an automated chemical analyzer (XN-1000V, Sysmex, Kobe, Japan).

Hematoxylin and eosin (H&E) staining

Liver tissues were fixed in formalin (F8775, Sigma-Aldrich), embedded in paraffin, and cut into 3 μm sections. After dewaxing and rehydration, the sections were stained in hematoxylin (C0105, Beyotime) for 8 min and rinsed with tap water for 5 min and thereafter, stained in eosin (C0105, Beyotime) for 30 s. Finally, the sections were dehydrated and sealed with neutral resin to observe the morphology of the liver using a fluorescent microscope (IX81-FV1000, Olympus, Tokyo, Japan).

Cell survival analysis

A colorimetric assay using sulforhodamine B (SRB, S1402, Sigma-Aldrich) was used to assess cell viability as previously described (Rubinstein et al., 1990). Cells were seeded at a density of 3000 cells per well in 96-well plates and cultured for 24 h to allow free growth. Thereafter, they were exposed to the drugs for 48 h. The absorbance was measured at 515 nm using a Multiskan Spectrum instrument (Thermo Electron Corporation, GA, USA).

Flow cytometric analysis

Apoptosis rates were determined using the Pharmingen™ FITC Annexin V Apoptosis Detection Kit I (556547, BD Biosciences, NJ, USA). The procedure was performed according to the manufacturer’s instructions. Briefly, the cells were processed for the indicated time, harvested, and washed with precooled phosphate buffer saline (PBS) for binding and Annexin V-PI staining. For each sample, 1×10⁴ cells were obtained and analyzed using BD FACSCalibur™ flow cytometry (342973, BD Biosciences).

Intracellular reactive oxygen species (ROS) levels were determined using a ROS Assay Kit (S0033S, Beyotime) according to the manufacturer’s instructions. Cells were harvested and incubated with 10 μM dichlorofluorescin diacetate (DCFH-DA) probes for 20 min at 37°C. The cells were washed with serum-free culture medium and resuspended. The cells were obtained and analyzed using BD FACSCalibur™ flow cytometry (342973, BD Biosciences).

Western blot

Total protein was extracted from cells or liver tissues using a lysis buffer (150 mM NaCl, 50 mM Tris-HCl, 2 mM EGTA, 2 mM EDTA, 25 mM β-sodium glycerophosphate, 25 mM NaF, 0.3% Triton X-100, 0.3% NP-40, 0.3% leupeptin, 0.1% NaVO3 and 0.1% PMSF). Protein lysates (20-50 μg per sample) were separated on 10% sodium dodecyl-sulfate polyacrylamide gel electrophoresis gels, transferred to polyvinylidene difluoride membranes (IPVH00010, Millipore Corporation, Boston, MA, USA) and blocked with a blocking buffer. Blots were cropped according to their molecular weights before probing with primary antibodies. Incubation with primary antibodies, secondary antibodies and the Western Lightning Plus-ECL reagent (NEL105001EA, PerkinElmer, Waltham, MA, USA) was performed for signal detection.

The following antibodies were used: Primary antibody against cleaved PARP (ET1608-10) was purchased from Huabio (Hangzhou, China). Primary antibodies against Nrf2 (12721S) γH2AX (97148SF) and p62 (5114s) were obtained from Cell Signaling Technology (Beverly, MA, USA). Primary antibody against β-actin (ACTB) (db7283) was obtained from Diagbio (Hangzhou, China). Primary antibody against p53 (sc-126) was obtained from Santa Cruz Biotechnology (TX, USA). Primary antibody against Keap1 (10503-2-AP) was obtained from Proteintech (IL, USA).

Terminal deoxynucleotidyl transferase dUTP nick end labeling (TUNEL) staining

A one-step TUNEL Apoptosis Detection Kit (C1088, Beyotime) was used to detect apoptosis in mouse liver sections according to the manufacturer’s instructions. Briefly, tissue sections were pretreated with proteinase K (ST532, Beyotime) working solution after dewaxing and rehydration. Then, TUNEL detection solution was added to the tissue samples and incubated for 60 min at 37°C in a light-proof humidified chamber. Finally, the nuclei were stained with 4’,6-diamidino-2-phenylindole (DAPI, D212, Dojindo, Kumamoto, Japan) and the TUNEL signals were observed and captured with a fluorescent microscope (IX81-FV1000, Olympus).

Comet assay

The comet assay was performed as described previously (Olive and Banáth, 2006). Firstly, a single-cell suspension was prepared in PBS at a density of approximately 2×10⁴ cells/mL. Next, pre-warmed 0.5% standard-gelling-temperature agarose, containing approximately 2000 cells, was placed on the microscope slides and spread sequentially (placed sequentially for more than 10 min) on the slides. The slides were immersed in an alkaline solution (pH >13) at 4°C for >1 h to lyse the samples. The slides were then submerged in a horizontal electrophoresis chamber containing a fresh cold alkaline solution (pH 12.3) for 20 min to unwind the DNA, and electrophoresis was performed at 300 mA for 20 min. Subsequently, the samples were neutralized with Tris-HCl (pH 7.5) for 15 min and dehydrated. Finally, the samples were stained with DAPI for 5 min, and images were captured using a fluorescent microscope (IX81-FV1000, Olympus). Using the Comet Assay Software Project, each sample was statistically analyzed using approximately 30 comet images.

RNA extraction and quantitative polymerase chain reaction (qPCR)

After drug treatment, the cells were collected with the Trizol reagent (15596–026, Invitrogen) for total RNA extraction. Equal amounts of RNA were reverse transcribed into complementary DNA using a cDNA reverse transcription kit (AT311; Transgene, Beijing, China). qPCR was performed using the TB Green Premix Ex Taq™ (Tli RNaseH Plus) (RR420A, Takara, Tokyo, Japan) on a QuantStudio™ 3 Real-Time PCR Instrument (A28132, Thermo Fisher Scientific, Waltham, MA, USA). Samples were amplified in two steps: the first step at 95°C (3 s), followed by 95°C (5 s) and 60°C (31 s) for 40 cycles. Relative quantification was determined by the ΔΔCt method.

The primer sequences were as follows:

NFE2L2 forward, 5’-AAACCAGTGGATCTGCCAAC-3’.

NFE2L2 reverse, 5’-GACCGGGAATATCAGGAACA-3’

ACTB forward, 5’-CACCATTGGCAATGAGCGGTTC-3’

ACTB reverse, 5’-AGGTCTTTGCGGATGTCCACGT-3’

HO-1 forward, 5’-TCAGGCAGAGGGTGATAGAAG-3’

HO-1 reverse, 5’-TTGGTGTCATGGGTCAGC-3’

SOD-1 forward, 5’-TGTGGCCGATGTGTCTATTG-3’

SOD-1 reverse, 5’-GCGTTTCCTGTCTTTGTACTTTC-3’

Measurement of Nrf2-antioxidant response element (ARE)-dependent transcriptional activity

The ARE-driven reporter gene construct pGL4.27-ARE-NRF2-SPE was purchased from MIAOLING BIOLOGY (P40937, Hubei, China). HL-7702 cells were transfected with pGL4.37 plasmid using jetPRIME (101000046, Polyplus-transfection, Illkirch, France). Briefly, HL-7702 cells were seeded and grown to approximately 70% confluence. The related plasmid was transfected into cells with the transfection reagent for 4-6 h and then replaced with a fresh culture medium. Cells stably transfected with the ARE reporter were exposed to 10 μM gefitinib and/or 1.25 mM APAP for 24 h. Luciferase activity in cell lysates was measured using the Duo-Lite Luciferase Assay System (DD1205-01; Vazyme, Jiangsu, China).

Immunofluorescence assay

After specific treatments, HL-7702 cells grown in 96-well plates were washed twice with PBS and fixed with fresh 4% paraformaldehyde (P6148, Sigma-Aldrich) for 20 min at 25°C. Then, the cells were permeabilized with 0.1% Triton X-100 in PBS for 10 min at 4°C and blocked with 4% bovine serum albumin (B2064, Sigma-Aldrich) in PBS for 30 min at 37°C. Thereafter, the cells were incubated with primary antibody against Nrf2 (ab62352; Abcam, Cambridge, UK) at 4°C overnight. After washing with PBS, the cells were incubated with Alexa Fluor 488-conjugated secondary antibodies (A11008; 1:100, Thermo Fisher Scientific, Waltham, MA, USA) at room temperature for 1 h, stained with DAPI for 5 min, and imaged using a fluorescence microscope (IX81-FV1000, Olympus).

Statistical analysis

Statistical comparisons of multiple groups were performed using one-way analysis of variance. All experiments were conducted in triplicate. Data were analyzed using GraphPad Prism 6.0 and expressed as mean ± standard error of mean (SEM). P value of <0.05 was considered statistically significant.

Combination of gefitinib and APAP could directly induce hepatocyte death via the apoptotic pathway

To observe hepatotoxicity, we designed the administration of gefitinib and APAP in ICR mice (Fig. 1A). Gefitinib and APAP doses were chosen based on clinical recommendations (Peacock et al., 2011; Miyauchi et al., 2022). A low dose of APAP was used to simulate the amount required for treating fever symptoms (Jaeschke, 2015). The mice were divided into four groups of six animals in each: vehicle control with 0.5% CMC-Na; gefitinib at a dosage of 75 mg/kg intragastric administration once a day for 5 weeks; APAP at a dosage of 100 mg/kg intragastric administration three days a week continuously for 5 weeks; gefitinib plus APAP. Mice treated with gefitinib, APAP, or gefitinib plus APAP tolerated the drugs well. Treatments with only gefitinib or APAP resulted in no significant liver injury compared with the vehicle control based on blood biochemical indices. In contrast, enhanced ALT and AST levels were observed in the gefitinib plus APAP group (Fig. 1B). Although there was no significant change in relative liver weight between the gefitinib group and the gefitinib plus APAP group (Fig. 1C), we further investigated the liver tissue sections of the mice using H&E staining. The livers of the mice in the gefitinib plus APAP group showed severe liver injury, as evidenced by structural disorders, cell swelling, and karyopyknosis with an eosinophilic cytoplasm, whereas those of the mice in gefitinib-only and APAP-only groups showed less damage (Fig. 1D). We used human hepatocytes HL-7702 and human hepatoma cells HepG2 to explore the potential hepatotoxic effect of gefitinib plus APAP in vitro. As illustrated in Fig. 1E, APAP reduced hepatocyte survival rate in a concentration-dependent manner. After 48 h of single administration of 1.25 mM APAP, the cell survival rate was 84.29 ± 2.96%. We chose this concentration for follow-up experiments because it had less effect on hepatocytes, and 5- or 10-mM APAP was used to mimic the overdose. We observed a further decrease in cell survival after combining gefitinib with a low concentration of APAP. This result was consistent with the number of hepatocytes observed under a microscope (Fig. 1F). Taken together, these data proved that the combination of gefitinib and APAP could potentially exacerbate hepatotoxicity in vivo and in vitro.

Figure 1. Combination of gefitinib and APAP induced hepatotoxicity in mice and cell death in hepatocytes. (A) Schematic diagram gefitinib and APAP co-administration. ICR mice were treated with 75 mg/kg gefitinib daily, with additional 100 mg/kg APAP given on three days a week by intragastric administration continuously for 5 weeks (n=6), and liver tissues were harvested after the mice were sacrificed. (B) Serum samples were collected to analyze the levels of ALT and AST (n=6). (C) Relative liver weights were calculated (n=6). (D) Liver sections were stained with H&E for histopathological analysis. Yellow arrowheads indicated the cellular level showing karyopyknosis with an eosinophilic cytoplasm that were suggestive of cell death in specific regions. For 400× magnification, scale bar: 50 μm. (E) Cell proliferation inhibition was examined by the SRB assays. HL-7702 cells were exposed to different concentrations of gefitinib and/or APAP for 48 h. Survival rates were calculated and shown as mean ± SEM from three independent experiments (n=3). (F) HL-7702 cells were exposed to indicated dose of gefitinib (10 μM) and/or APAP (1.25 mM) for 24 h. Cells were photographed by microscopy. The data are expressed as the mean ± SEM; *p<0.05; **p<0.01; ***p<0.001. ALT, alanine aminotransferase; APAP, acetaminophen; AST, aspartate aminotransferase; BW, body weight; GEFI, gefitinib; LW, liver weight; H&E, hematoxylin and eosin; SRB, sulforhodamine B.

Cell death involves a signaling cascade comprising several effector molecules and has unique biochemical characteristics, including various forms of apoptosis, necroptosis, pyroptosis, ferroptosis, and autophagy-dependent cell death. Given that the combination of gefitinib and APAP can directly inhibit hepatocyte survival, we investigated the specific mechanisms underlying hepatocyte death. We used a variety of cell death inhibitors, such as apoptosis inhibitor Z-VAD-FMK, necroptosis inhibitor necrostatin-1, pyroptosis inhibitor Z-YVAD-FMK, ferroptosis inhibitor ferrostatin-1 and autophagy inhibitor 3-MA to investigate the cell death. The appropriate drug concentration was determined based on relevant studies (Liao et al., 2019; Chen et al., 2020; Zhang et al., 2021a) and the drugs were used in combination with gefitinib plus APAP for 48 h in HL-7702 cells. The data (Fig. 2A) indicated that apoptosis induced by the combination of gefitinib and APAP was the main factor involved in regulating hepatocytes survival. We investigated the pattern of hepatocyte death induced by the combination of gefitinib and APAP. Annexin V/PI staining and flow cytometry were used to examine hepatocyte apoptosis. Annexin V-PI-, Annexin V-PI+, Annexin V+PI-, and Annexin V+PI+ stains were used to stain viable, necrotic, early apoptotic, and late apoptotic cells. The HL-7702 cells were incubated with gefitinib and/or APAP at the indicated concentrations for 24 h. As shown in Fig. 2B, most hepatocyte deaths were caused by apoptosis and rarely by necrosis. In particular, the rate of apoptosis in the gefitinib plus APAP group was higher than those in the gefitinib-only and APAP-only groups. Furthermore, western blot was used to analyze the levels of the key apoptotic protein, cleaved poly (ADP-ribose) polymerase (c-PARP). We found that the gefitinib-only and APAP-only groups have slightly increased c-PARP levels, which were remarkably enhanced in the gefitinib plus APAP group (Fig. 2C, Supplementary Fig. 1). Next, we performed the classic apoptotic TUNEL assay to determine the apoptotic cell population in liver tissue. After combined stimulation with both agents, a considerable amount of green fluorescence was observed, which is one of the signals of late apoptosis (Fig. 2D). The data also showed that apoptosis induced by gefitinib plus APAP was partially reversed in the presence of the apoptosis inhibitor, Z-VAD-FMK (Fig. 2E). In sum, the above results demonstrate that the combination of gefitinib and APAP causes increased hepatocyte death by exacerbating apoptosis.

Figure 2. APAP enhanced gefitinib-triggered apoptosis. (A) Cell proliferation inhibition was examined by the SRB assays. HL-7702 cells were exposed to gefitinib, APAP and/or different inhibitors: 20 μM Z-VAD-FMK, 10 μM Z-YVAD-FMK, 1μM ferrostatin-1, 20 μM necrostatin-1 and 1mM 3-MA for 48 h. (B-C) HL-7702 cells were treated with 10 μM gefitinib and/or 1.25 mM APAP for 24 h. (B) The apoptosis rates were analyzed by flow cytometry with an FITC Annexin V Apoptosis Detection Kit. (C) The expression levels of c-PARP and ACTB were determined by western blot. (D) Liver sections were stained with TdT-mediated dUTP nick end labeling (TUNEL) and 4’,6-diamidino-2-phenylindole (DAPI). Scale bar=50 μm. Quantitative analysis was performed to detect apoptotic cells (n=3). (E) HL-7702 cells treated with 10 μM gefitinib and 1.25 mM APAP were combined with or without 20 μM Z-VAD-FMK for 24 h. The expression levels of c-PARP and ACTB were determined by Western blot. All experiments were performed in triplicate. For western blots, one of three similar experiments is shown, and densitometric analysis was carried out. Data are presented as the mean ± SEM (n=3); n.s=no significance; *p<0.05; **p<0.01; ***p<0.001. APAP, acetaminophen; Fer-1, ferrostatin-1; GEFI, gefitinib; Nec-1, necrostatin-1; SRB, sulforhodamine B; 3-MA, 3-Methyladenine.

Combination of gefitinib and APAP induced DNA damage response via ROS generation

There is growing evidence that ROS accumulation is involved in hepatotoxicity and that ROS plays a considerable role in cell apoptosis (Yan et al., 2019). APAP overdose results in the overproduction of reactive free radicals and N-acetyl-p-benzoquinone-imine, which induces ROS production, DNA damage, and liver cell damage (Yan et al., 2018; Du et al., 2022). However, the effect of a lower dose of APAP on ROS production remains elusive. Therefore, we explored whether gefitinib plus APAP could activate ROS generation in the hepatocytes. We used a ROS assay kit to measure intracellular ROS changes in HL-7702 cells following different treatments. As elucidated in Fig. 3A, the combination of gefitinib and APAP resulted in increased ROS generation compared with single agents, implying that ROS generation may be associated with the gefitinib plus APAP-induced hepatotoxicity.

Figure 3. Combination of gefitinib and APAP induced accelerated ROS generation, resulting in a DNA damage response. (A-E) HL-7702 cells were treated with 10 μM gefitinib and/or 1.25 mM APAP for 24 h. (A) A ROS assay kit showed ROS generation in HL-7702 cells. (B) The expression levels of γH2AX and ACTB were determined by western blot. (C) Typical images of DNA damage after treatment detected by comet assay. Scale bar=50 μm. (D) Quantitative analysis was performed on the percentage of DNA tail, n=50 cells per group. (E) HL-7702 cells treated with 10 μM gefitinib and 1.25 mM APAP were combined with or without 1 mM NAC for 24 h. The expression levels of c-PARP and ACTB were determined by western blot. All experiments were performed in triplicate. For western blots, one of three similar experiments is shown, densitometric analysis was carried out. Data are presented as the mean ± SEM (n=3); n.s=no significance; **p<0.01; ***p<0.001. APAP, acetaminophen; GEFI, gefitinib; NAC, N-acetyl-L-cysteine.

Oxidative DNA damage caused by amplified ROS generation is one of the most common types of DNA damage in cells, in which the double strand of the cellular DNA breaks. Phosphorylation of histone H2AX (γH2AX) is considered a biomarker that can reflect the extent of DNA damage and repair (Mah et al., 2010). Moreover, accumulating evidence indicates that H2AX phosphorylation commonly occurs during apoptosis (Lu et al., 2006; Liu et al., 2019; Wang et al., 2022). Thus, we hypothesized that increased DNA damage may be responsible for increased apoptosis. We measured the γH2AX level and showed that the γH2AX level was greatly increased when gefitinib was combined with APAP (Fig. 3B). We conducted an alkaline single-cell gel electrophoresis (comet) assay to investigate DNA damage (Olive and Banáth, 2006). As expected, the amount of tail DNA after electrophoresis increased in the gefitinib plus APAP group (Fig. 3C, 3D). We demonstrated that the combination of gefitinib and APAP was associated with DNA damage. Previous studies have demonstrated that ROS accumulation mediates DNA damage and induces apoptosis (Lee et al., 2022; Park et al., 2022). As shown in Fig. 3E, the ROS scavenger NAC inhibited the increase in c-PARP levels induced by gefitinib plus APAP. In conclusion, these data indicate that ROS generation and DNA damage are involved in hepatocyte apoptosis under the combination of gefitinib and APAP.

Liver injury exacerbated by combination of gefitinib and APAP might be related to Nrf2 activation disorders

High levels of ROS are associated with oxidative stress and disease progression. Nrf2 is a key upstream transcription factor for cellular resistance to oxidative stress and plays an antioxidant and anti-mortality role by upregulating heme oxygenase-1 (HO-1), superoxide dismutase 1 (SOD1), and other antioxidant proteins to scavenge ROS under oxidative stress (Zhang, 2006; Done and Traustadóttir, 2016; Jayasuriya et al., 2021). Guo et al. (2021) found that Nrf2 protein levels in hepatocytes were significantly decreased by crizotinib, which in turn caused an abnormal accumulation of ROS in hepatocytes and ultimately induced cell death (Guo et al., 2021). Immunoblot analysis of mouse liver tissue lysates, as well as HL-7702 and HepG2 cells, showed that gefitinib plus APAP significantly reduced the upregulation of Nrf2 induced by gefitinib alone (Supplementary Fig. 2, 3, Fig. 4A), implying that APAP strengthened gefitinib-induced ROS generation by reducing Nrf2 levels. Furthermore, we explored the potential regulation of the reduced Nrf2 levels following APAP treatment. We found a non-significant decrease in Nrf2 transcriptional levels (Fig. 4B), which suggested Nrf2 protein downregulation was not associated with transcriptional suppression. Next, we focused on the regulation of Nrf2 stability. The stability and cellular distribution of Nrf2 are tightly controlled by the inhibitory binding protein Keap1. Keap1 binds to Nrf2 and activates the ubiquitin-proteasome pathway for degradation (Baird and Yamamoto, 2020). As illustrated in Fig. 4C and Supplementary Fig. 3, the combination of gefitinib and APAP significantly increased Keap1 protein levels in HL-7702 and HepG2 cells, whereas gefitinib greatly decreased the level of Keap1.

Figure 4. Liver injury exacerbated by gefitinib in combination with APAP might be related to Nrf2. (A-E, G) HL-7702 cells were treated with 10 μM gefitinib and/or 1.25 mM APAP for 24 h. (A) The expression levels of Nrf2 and ACTB were determined by western blot. (B) The transcription level of NFE2L2 was determined by qPCR. (C) The expression levels of Keap1 and ACTB were determined by western blot. All experiments were performed in triplicate. (D) The transcription levels of HO-1 and SOD-1 was determined by qPCR. (E) Representative images of HL-7702 stained with Nrf2 (green) and DAPI (blue) were shown by immunofluorescence assay. Scale bar=10 μm. (F) Induction of the ARE-dependent luciferase reporter activity by gefitinib and/or APAP. HL-7702 cells stably expressing an ARE-reporter gene were treated with 10 μM gefitinib and/or 1.25 mM APAP for 24 h and the luciferase activity in the cell lysates was measured. (G) The expression levels of p62 and ACTB were determined by western blot. All experiments were performed in triplicate. For western blots, one of three similar experiments is shown, and densitometric analysis was carried out. Data are presented as the mean ± SEM; n.s=no significance; *p<0.05; **p<0.01; ***p<0.001. APAP, acetaminophen; ARE, antioxidant response element; GEFI, gefitinib.

The antioxidant genes, HO-1 and SOD1, are induced by Nrf2 nuclear translocation. HO-1 and SOD-1 transcriptional levels (Fig. 4D) detected in cells treated with gefitinib plus APAP were significantly downregulated compared with those in the gefitinib-only group, which may be responsible for the weakened resistance to oxidative stress. ARE signaling pathway is a major defense mechanism against oxidative stress (Motahari et al., 2015). ARE luciferase reporter assays were used to analyze Nrf2-driven gene transcription activity. The luciferase activity of the cells was lower in the gefitinib plus APAP group than in the gefitinib-only group (Fig. 4E). Additionally, nuclear localization of Nrf2 following drug treatment was observed using immunofluorescence. As shown in Fig. 4F, the cytoplasmic and nuclear distribution ratios of Nrf2 were similar in the control and APAP-only groups, whereas Nrf2 accumulated in the nucleus following gefitinib treatment. However, increased translocation of Nrf2 from the nucleus to the cytosol was observed in the gefitinib plus APAP group. These results suggest that the combination of gefitinib and APAP activates oxidative stress through Nrf2.

Moreover, the p62/Keap1/Nrf2 antioxidative signaling pathway is involved in protection against ferroptosis in hepatocellular carcinoma cells cells. p62 prevents Nrf2 degradation and enhances Nrf2 nuclear accumulation via Keap1 (Sun et al., 2016). To examine the upstream signaling pathways involved in drug-induced Keap1 and Nrf2 expression, the cells were exposed to gefitinib and/or APAP, and p62 protein was analyzed by western blot. As shown in Fig. 4G and Supplementary Fig. 2, when the two drugs were combined, gefitinib-induced upregulated expression of p62 was inhibited by APAP, leading to the accumulation of Keap1. In summary, our results demonstrate that gefitinib plus APAP induces liver injury via the p62/Keap1/Nrf2 signaling pathway-mediated inhibition of apoptosis.

Overall, our preliminary study suggests that Nrf2 levels might be associated with the exacerbation of liver injury. This could be a way to reduce the risk of hepatotoxicity by precisely regulating the level of Nrf2 or using ROS scavengers.

In this study, we revealed the critical role of apoptosis in the exacerbation of gefitinib plus APAP-induced hepatotoxicity and provided mechanistic insight into ROS-promoting pathways. The liver injury occurred after the co-administration of clinically appropriate doses of gefitinib and APAP in ICR mice. Mechanistically, the combination of gefitinib and APAP downregulated gefitinib-induced increase in the p62/Keap1/Nrf2 signaling pathway in hepatocytes, thus increasing intracellular ROS production, leading to DNA damage and enhanced apoptosis.

Gefitinib at a dose of 250 mg once daily (Ramalingam et al., 2020) and APAP at ≤2 g per day are safe for an adult (García-Román and Francés, 2020). In the animal model, following dose conversion, the experimental dose of gefitinib (75 mg/kg) was twice the clinically safe dose, whereas that of APAP (100 mg/kg) was within the clinically safe dose range. The in vitro concentration of gefitinib used in this study was determined based on the literature reports. In some studies, the mean maximum plasma concentration of gefitinib was approximately 492-679 ng/mL (1.0-1.5 μM) (Zhao et al., 2005, 2013). Nigade et al. (2017) demonstrated that the ratio of liver to plasma concentrations of gefitinib in mice 2 h after administration was 20.83 ± 8.49. In accordance with the literature, 10 μM of gefitinib used in this study is suitable for the clinical environment and the plasma level of patients. In most studies on the mechanism of liver damage caused by APAP overdose, concentrations used include 5 mM, 10 mM, or 20 mM (Xie et al., 2014). However, we chose a concentration range that was less likely to cause severe liver toxicity: 0, 0.3125, 0.625, 1.25, 2.5 mM for screening. We chose 10 μM gefitinib and 1.25 mM APAP to conduct follow-up experiments due to their lesser effect on hepatocytes.

Previous studies have reported that APAP overdose induces predominantly necrotic or programmed hepatocyte necrosis and rarely involves apoptosis, but the effects of safe and regular low-dose APAP administration in the liver remain unclear. Our results show that low-dose APAP had little effect on hepatocyte survival and caused only a slight increase in apoptosis. When APAP was combined with gefitinib, we speculated that the gefitinib-induced liver injury was exacerbated by an increase in ROS produced by gefitinib and the promotion of hepatocyte apoptosis. In vivo, modeling studies have found the effect of APAP on the pharmacokinetic parameters of sorafenib, lapatinib, and erlotinib, which resulted in a significant increase in the area under the plasma concentration-time curve and the maximum concentration of the oncology drugs, which may increase the intensity of adverse effects due to enhanced inhibition of p-glycoprotein by APAP (Karbownik et al., 2017, 2018, 2020). Patients taking imatinib concurrently with APAP for pain relief may experience severe hepatotoxicity and even death (Polson and Lee, 2005; Thia et al., 2008). Further systemic studies are required to elucidate the in vivo effects of APAP on the pharmacokinetic profile of gefitinib. APAP may cause higher concentrations of and exposure to gefitinib and a higher risk of gefitinib hepatotoxicity during long-term combination therapy.

p53 protein is a central biomarker of cellular responses to various types of damage and regulates apoptosis and autophagy in response to oxidative stress (Shi and Dansen, 2020). We detected the expression level of the P53 protein and found no obvious increase after administration (Supplementary Fig. 4). Therefore, we focused on Nrf2 as a key regulatory factor of toxicity.

To explore the possible mechanisms underlying gefitinib plus APAP-induced oxidative stress, we evaluated Nrf2 protein expression, an antioxidant transcription factor involved in redox homeostasis, which plays a vital role in drug-induced liver injury (Chao et al., 2018). Activation of the Nrf2 pathway can effectively protect the liver from oxidative stress. However, Nrf2 deletion or activation disorders can exacerbate oxidative stress-induced cytotoxicity, contributing to cellular dysfunction, apoptosis, and even cell death (Ma, 2013). Our results suggest that the combination of gefitinib with APAP may promote oxidative stress through the negative regulation of the p62/Keap1/Nrf2 signaling pathway compared with gefitinib monotherapy, thus exacerbating gefitinib-induced ROS generation and amplifying gefitinib-induced hepatotoxicity. Notably, a low dose of APAP reduced the level of p62 protein, which was otherwise upregulated by gefitinib and needs further exploration.

Several studies have revealed that p62 is instrumental in regulating the Keap1/Nrf2 pathway. p62 autophagy-dependent degradation of Keap1 can result in the attenuation of Nrf2 ubiquitination and increased protein stability (Komatsu et al., 2010). p62 participates in autophagy by regulating the formation of protein aggregates, a catabolic process designed to degrade and recycle cellular components and damaged organelles in response to various stress conditions, such as oxidative stress. Gefitinib has been reported to induce ROS production and thus cause multiple stress-activated autophagy in cancer cells. Gefitinib induces maladaptive autophagy and thus promotes apoptosis in hepatocytes, indicating a negative effect of autophagy on gefitinib-induced hepatotoxicity (Chen et al., 2019; Luo et al., 2021). In contrast, autophagy protects against APAP-induced liver injury by removing APAP-protein adducts in mouse and human hepatocytes (Ni et al., 2016; Chao et al., 2018; Jin et al., 2021). To exclude the interference of autophagy, we selected the autophagy inhibitor, 3-MA, in combination with the two drugs and observed its effect on gefitinib plus APAP-induced hepatocyte death using the SRB method. The results (Fig. 2A) demonstrated that autophagy inhibitors did not contribute significantly to gefitinib plus APAP-induced hepatocyte death, indicating that autophagy was not involved in regulating gefitinib plus APAP-induced cell death.

Our study indicates that the p62/Keap1/Nrf2 antioxidative signaling pathway is involved in the combination with gefitinib and APAP-induced apoptosis. Preventing the expression of p62 promotes the degradation of Nrf2 and attenuates subsequent Nrf2 nuclear accumulation through the accumulation of Keap1. To the best of our knowledge, this is the first study to demonstrate that the combination of gefitinib and APAP enhances oxidative stress via the p62/Keap1/Nrf2 signaling pathway.

This work was supported by National Natural Science Foundation of China (No.82104315), Natural Science Foundation of Zhejiang Province of China (No. LQ22H310002), Special Pharmacy Project of Zhejiang Pharmaceutical Association (No.2023ZYY31), Yangtze River Delta Health Scientific Research Project of Zhejiang Province (No.2023CSJ-3-A002) and Youth Fund Project of Hangzhou Red Cross Hospital (No.HHQN2023007).

All authors declare that they have no conflict of interest.

Conceptualization and study design: P. L., H. Y. and J. J.; Experimental work: J. X. and X. H.; Animal models: X. H. and Y. Z.; Analysis and interpretation of data: J. X., X. H., Y. Z. and Z. X., X. C.; Writing the draft manuscript: J. X. and X. H.; Review and editing the manuscript: Z. X., X. C., B. Y., Q. H., P. L., H. Y. and J. J.; Funding acquisition: H. Y. All authors have read and agreed to the published version of the manuscript.