Effects of the combination of brefeldin A and tunicamycin on endoplasmic reticulum stress and apoptosis in human normal hepatocytes

Graphical abstract

Brefeldin A (BFA) and tunicamycin (TM) induce a mild, non-canonical endoplasmic reticulum (ER) stress response in normal human hepatocytes. Treatment with BFA + TM upregulates BiP and ATF4 without activating the PERK-eIF2α-CHOP axis, leading to weak, caspase-12-dependent but caspase-3-independent apoptosis. These findings highlight the need to evaluate the potential hepatotoxicity of BFA-based therapies.

Highlights

• •BFA and TM combination mildly induces apoptosis in normal human hepatocytes.
• •Apoptosis is caspase-12-dependent but independent of CHOP and caspase-3 pathways.
• •BFA triggers non-canonical ER stress without PERK phosphorylation or CHOP induction.
• •BFA+TM enhances BiP/ATF4 expression, suggesting potential hepatotoxicity concerns.

Abstract

Background

Due to its ability to inhibit the growth of hepatoma cells, brefeldin A (BFA) has been considered a promising drug candidate for liver cancer. However, there is limited research on its safety profile and potential impacts when administered alone or in combination with other anticancer drugs.

Objective

To evaluate the safety of BFA in combination with tunicamycin (TM, a candidate anticancer drug) in human normal liver cells (HL-7702) in terms of its ability to induce endoplasmic reticulum (ER) stress and apoptosis.

Methods

HL-7702 cells were exposed to BFA (0–2.5 mg/L) and TM (0–5 mg/L), either alone or in combination, for 24 h. Cell viability was measured using the CCK-8 assay, and apoptotic rates were determined using flow cytometry. The mRNA and protein levels of key factors related to cell proliferation, ER stress, and apoptosis were determined using quantitative RT-PCR and Western blot, respectively.

Results

BFA and TM, either alone or in combination, significantly reduced the viability of HL-7702 cells. BFA alone and BFA + TM combination could weakly induce apoptosis, increase the expression of caspase 12, and reduce the protein level of proliferating cell nuclear antigen (PCNA). BFA alone and BFA + TM combination could significantly increase the mRNA and protein levels of binding immunoglobulin protein (BiP) and activating transcription factor 4 (ATF4), but did not affect the mRNA and protein levels of C/EBP homologous protein (CHOP) and poly (ADP-ribose) polymerase-1 (PARP-1).

Conclusion

This study demonstrates that BFA, alone and in combination with TM, exerts mild pro-apoptotic effects on HL-7702 cells, independent of the CHOP and caspase-3 pathways. These findings underscore the necessity of evaluating the potential hepatotoxicity of BFA-based therapies, particularly in combination treatments, to ensure their safe clinical application.

1. Introduction

Brefeldin A (BFA) is a protein transport inhibitor that exerts its inhibitory effect by blocking protein transport between the endoplasmic reticulum (ER) and Golgi apparatus, as well as by interfering with protein glycosylation. (Alonso et al., 2006, Zeng et al., 2019, Massarweh et al., 2016, Uedono et al., 2021) BFA can induce ER stress in cancer cells, leading to apoptosis, (Burdett, 2002, Bouchut et al., 2014) which is similar to the common anti-tumor mechanisms of many chemotherapy drugs. (Aslan et al., 2021, Shi et al., 2022) Therefore, BFA has been considered a very promising anti-tumor drug. On the other hand, tunicamycin (TM) is a commonly used ER stress inducer. It can specifically inhibit the dolichol phosphate cycle in the process of N-linked glycosylation, impairing proper protein folding and leading to the accumulation of unfolded proteins in the ER. (Liu et al., 2020, Han et al., 2013) TM has also been considered a potential anti-tumor drug. In addition, the ability of BFA to induce apoptosis requires the activation of the caspase family, which can be effectively suppressed by specific caspase inhibitors. (Guo et al., 1998) Therefore, the combined use of BFA and TM have great potential to treat liver cancer. (Li et al., 2025, Markouli et al., 2020) ER stress is closely linked to apoptosis through multiple molecular pathways, and failure of the unfolded protein response (UPR) to restore ER homeostasis can initiate apoptosis. There are many signaling molecules involved in ER stress-induced apoptosis, such as C/EBP homologous protein (CHOP), c-Jun N-terminal kinase (JNK), and caspase-12. Under normal conditions, the expression level of CHOP is relatively low; however, during ER stress, its expression can be upregulated through the PERK-eIF2α-ATF4 pathway, leading to the initiation of apoptosis. (Huang et al., 2025) Caspase-12 can specifically senses internal perturbations in the ER, converting them into apoptotic signals and playing a crucial role in ER stress-induced apoptosis. (Nakagawa et al., 2000, Lakshmanan et al., 2011).

Hepatocellular carcinoma (HCC) is one of the common liver cancers that seriously threaten human health. (Vogel et al., 2022) The combination of targeted therapy (such as sorafenib, lenvatinib, and donafinib) and immunotherapy (such as pembrolizumab) has been used to treatment HCC patients. However, due to rapid development of drug resistance in cancer cells, the therapeutic efficacy of these treatments is still not satisfactory. Therefore, discovering other alternative drugs is of great clinical value. The therapeutic potential of BFA and TM in HCC treatment, either alone or in combination, is undoubtedly anticipated. Nevertheless, studies also show that BFA exhibits ER stress-inducing effects in both cancer cells and normal liver cells. (Tian et al., 2017, Wang et al., 2023) Therefore, the safety of BFA as an anti-cancer drug still needs to be further confirmed. Herein, we investigated the inhibitory effects of BFA alone and in combination with TM on the viability of human normal hepatocytes (HL-7702). Moreover, we observed whether the inhibitory effects were associated with ER stress-induced apoptosis. The findings are expected to provide evidence for the clinical application of BFA and TM in HCC treatment.

2. Materials and methods

2.1. Cells

HL-7702 cells are a human normal hepatocyte cell line originally derived from a normal young male. Cells were generously provided by Professor Xujun Qin from Shaanxi University of Traditional Chinese Medicine, who originally purchased them from the Stem Cell Bank of Shanghai Institute of Cell Biology, Chinese Academy of Sciences, China. cells were cultured in high-glucose Dulbecco’s Modified Eagle medium (DMEM, Cytiva, Hangzhou, China) containing 10% fetal bovine serum (SORFA, Huzhou, China) and 1% penicillin–streptomycin (Solarbio, Beijing, China) at 37℃ and 5% CO₂ in a Galaxy 170S incubator (Eppendorf, Germany). At the logarithmic growth phase, cells were digested with 0.25% trypsin-EDTA (Solarbio, Beijing, China), suspended in DMEM, and treated with BFA (purity > 98%, Beyotime Biotechnology, Shanghai, China) and TM (purity > 98%, Shanghai Macklin Biochemical, Shanghai, China), either alone or in combination. DMSO (0.1%, Sigma-Aldrich, St. Louis, MO, USA) was used as vehicle control.

2.2. Cell viability

Cell viability was measured using CCK-8 (Wuhan Boster Biological Technology, Wuhan, China). Briefly, cells were seeded in 96-well plates (6 wells per group, 1.5 × 10⁵ cells per well). In dose–response experiments, cells were treated with different concentrations of BFA (0.1, 0.25, 0.5, 1.0, and 2.5 mg/L) and TM (1, 2, 3, 4, and 5 mg/L), either individually or in combination, for 24 h. After treatment, 10 µL of CCK-8 solution was directly added to each well (a final volume of 100 µL), followed by incubation for 1–2 h with gentle shaking. The absorbance (optical density, OD) at 450 nm was measured using a 680 microplate reader (BioTek, VT, USA). Cell viability was calculated as follows: Cell viability (%) =(OD_treatment − OD_blank)/(OD_vehicle − OD_blank) × 100%. Blank controls contain an equal volume of culture medium and CCK-8 reagent without cells. All experiments were performed in at least three independent biological replicates, each with six technical replicates.

2.3. Flow cytometry

Based on viability assays, 0.25 mg/L BFA and 1 mg/L TM were selected for subsequent mechanistic studies to ensure a clear cellular response while maintaining sufficient viability. (Huang et al., 2017, Pommepuy et al., 2003, Guha et al., 2017) Cells were treated with 0.25 mg/L BFA and 1 mg/L TM, either individually or in combination, for 24 h. Then, cells were digested with 0.25% trypsin, collected by centrifugation (800 g, 5 min), and re-suspended in 100 µL of 1 × binding buffer (1 × 10⁶ cells/mL). After adding 5 µL of Annexin V-FITC and 8 µL of propidium iodide (PI) (Dalian Meilun Bio, Dalian, China), cells were incubated at room temperature in the dark for 15 min. Apoptosis was detected using an AccuriTM C6 flow cytometer (BD Bioscience, CA, USA), and the results were analyzed using Cell Quest software. Data were collected from three independent biological replicates.

2.4. Western blot

Cells were incubated in 6-well plates (3 × 10⁶ cells/well) at 37°C overnight. Then, cells were treated with BFA (0.25 mg/L) and TM (1 mg/L) for 24 h. After treatment, cells were washed with PBS and lysed in 200 µL of cell lysis buffer containing RIPA lysis buffer, phosphatase inhibitor cocktail, and protease inhibitor cocktail (98:1:1, Boster Biological Technology, Wuhan, China) on ice for 10 min. Cell lysate was collected, and protein levels were determined using the bicinchoninic acid (BCA) assay (Boster Biological Technology, Wuhan, China). Proteins were electrophoretically separated using SDS-polyacrylamide gels at 80 V for 150 min, followed by transferring to a polyvinylidene difluoride (PVDF) membrane (Immobilon-P transfer membrane, Merck KGaA, Darmstadt, Germany) using the wet transfer method at 400 mA for 60 min. The membrane was blocked with 5% skim milk in PBST (PBS containing 1% Tween 20) at room temperature for 2 h. After incubation with primary and secondary antibodies (Table 1), an enhanced chemiluminescence (ECL) detection solution (ThermoFisher, NJ, USA) was added, and the images were captured using a BG-gdsAUTO imaging analysis system (Baygene Biotechnologies, Shanghai, China). Protein levels were quantitatively analyzed using Image-Pro Plus 6.0 (Media Cybernetics, MD, USA). All experiments were performed in at least three independent biological replicates, each with three technical replicates.

Table 1.

Antibodies used for Western blots.

Antibodies	Manufacturer	Dilution ratio
Mouse anti-human β-actin	Protein Tech, Wuhan, China	1:2000
Rabbit anti-human PCNA	Boster Bio Tech, Wuhan, China	1:1000
Mouse anti-human PARP-1	Protein Tech, Wuhan, China	1:1000
Rabbit anti-human Caspase-12	Protein Tech, Wuhan, China	1:500
Rabbit anti-human Caspase-3	Protein Tech, Wuhan, China	1:500
Rabbit anti-human BiP/GRP78	CST, Boston, America	1:1000
Rabbit anti-human PERK	Protein Tech, Wuhan, China	1:1000
Rabbit anti-human phospho-PERK	Beyotine, Shanghai, China	1:1000
Rabbit anti-human ATF4	Protein Tech, Wuhan, China	1:1000
Rabbit anti-human eIF2α	Protein Tech, Wuhan, China	1:1000
Rabbit anti-human phospho-eIF2α	CST, Boston, America	1:1000
Rabbit anti-human CHOP	Protein Tech, Wuhan, China	1:500
Goat anti-mouse IgG	Protein Tech, Wuhan, China	1:5000
Goat anti-Rabbit IgG	Protein Tech, Wuhan, China	1:5000

Abbreviations: PCNA, proliferating cell nuclear antigen; PARP-1, poly (ADP-ribose) polymerase-1; BiP, binding immunoglobulin protein; PERK, protein kinase R-like endoplasmic reticulum kinase; ATF4, activating transcription factor 4; eIF2α, eukaryotic translation initiation factor 2α; CHOP, C/EBP homologous protein.

2.5. Quantitative real-time polymerase chain reaction (qRT-PCR)

Cells were incubated in 6-well plates (1.2 × 10⁷ cells/well) at 37°C overnight, followed by treatment with BFA (0.25 mg/L) and TM (1 mg/L) for 24 h. Then, cells were harvested, and total RNA was extracted using RNAiso Plus reagent (TaKaRa, Japan). The concentration and purity of total RNA were determined using a NanoDrop TM 8000 spectrophotometer (Thermo Fisher Scientific, Wilmington, USA). cDNA was synthesized using a PrimeScript RT Master Mix kit (TaKaRa, Japan), and qRT-PCR was performed using a SYBR Premix Ex Taq™ Ⅱ kit (TaKaRa, Japan) on a LineGene 9600 Plus real-time PCR detection system (Hangzhou Bioer Technology, China) according to the manufacturer’s protocol. The primers used for qRT-PCR are listed in Table 2. Relative mRNA levels were normalized to the level of the internal reference gene (β-actin) using the 2^-ΔΔCt method. β-actin was chosen as a reference gene based on its widespread use and stability in HL-7702 cells under various conditions. (Cui et al., 2015, Zheng et al., 2018, Gao et al., 2016) All experiments were performed with three independent biological replicates, each with three technical replicates.

Table 2.

Primers of the genes used for qRT-PCR.

Genes	Primer sequence (5′-3′)	Amplicon (bp)
β-actin	Forward: TGGCACCCAGCACAATGAA	186
	Reverse: CTAAGTCATAGTCCGCCTAGAAGCA
BiP	Forward: TTCCCAGCCCCTCAGATAC	162
	Reverse: CAAAGAGAAGCACCAAGGAGAC
ATF4	Forward: TTCCGAGATTCCATCCTACG	103
	Reverse: AGGCTCACAAACGAATGGAC
CHOP	Forward: GCGATATGCTGTGGTGCTTA	121
	Reverse: AGCTCCAAGTGAAACCAGGA
PERK	Forward: ACGATGAGACAGAGAGTTGCGAC	126
	Reverse: ATCCAAGGCAGCAATTCTCCC
Caspase-3	Forward: CTCGGTCTGGTACAGATG	126
	Reverse: GGTTAACCCGGGTAAGAATGTGCA
Caspase-12	Forward: CACCAGTCCTCAGACAGCACATTC	144
	Reverse: AGACTCTGGCAGTTACGGTTGTTG
PARP1	Forward: AGGCTTGAAAAGCCCTAAAGG	132
	Reverse: CTGCTTGTTGAAGATGAGTAGC
eIF2α	Forward: CCTGGCAAGAATGCAGTCAC	186
	Reverse: GGCGAAACCAATGTATTTCTGGA

Abbreviations: BiP, binding immunoglobulin protein; ATF4, activating transcription factor 4; CHOP, C/EBP homologous protein; PERK, protein kinase R-like endoplasmic reticulum kinase; PARP-1, poly (ADP-ribose) polymerase-1; eIF2α, eukaryotic translation initiation factor 2α.

2.6. Statistical analysis

Data are expressed as mean ± standard deviation (SD) and were statistically analyzed using GraphPad Prism 9.0 (GraphPad Software, CA, USA). Differences between groups were determined using one-way ANOVA, followed by Dunnett’s test. Differences were considered statistically significant at P < 0.05.

3. Results

3.1. BFA and TM inhibit the viability of HL-7702 cells

Compared to the blank control and vehicle control, both BFA (0.1–2.5 mg/L) and TM (1–5 mg/L) significantly reduced cell viability in a dose-dependent manner (Fig. 1A, 1B). Moreover, the combination of BFA (0.25 mg/L) and TM (1 mg/L) significantly reduced cell viability compared to BFA alone (Fig. 1C).

Fig. 1.

Effects of BFA and TM on the viability of HL-7702 cells at 24 h. (A) BFA, (B) TM, and (C) BFA + TM. Cells were treated with BFA and TM, either alone or in combination, for 24 h. * p < 0.05 compared with the blank control, # p < 0.05, compared with DMSO, and a p < 0.05, compared with 0.1 mg/L BFA.

3.2. Combination of BFA and TM inhibits the expression of proliferating cell nuclear antigen (PCNA)

To investigate whether BFA and TM inhibited cell proliferation, we measured the protein levels of PCNA using Western blot. PCNA is an essential protein for DNA replication and has been widely used as a biomarker for cell proliferation. (Cayrol et al., 1998, Waseem and Lane, 1990 May) The result showed that BFA alone significantly reduced PCNA levels after 24-h treatment, but TM alone did not (Fig. 2). Notably, BFA + TM also significantly decreased PCNA levels compared to the blank control (P < 0.05). No significant difference in PCNA levels was observed between BFA alone and BFA + TM combination.

Fig. 2.

Effects of BFA and TM on PCNA levels. HL-7702 cells were treated with BFA (0.25 mg/L) and TM (1 mg/L), either individually or in combination, for 24 h. The experiment was performed in triplicates. * P < 0.05 compared with the blank control.

3.3. TM enhances the ability of BFA to induce apoptosis

As shown in Table 3, BFA + TM significantly increased the early apoptotic rates (P < 0.05), but neither BFA nor TM alone did (P > 0.05). On the other hand, both BFA and BFA + TM significantly increased the late apoptotic rates (P < 0.05), with BFA + TM showing a higher induction ability than BFA alone (P < 0.05). TM alone had no effects on both the early and late apoptotic rates (P > 0.05). While the observed increase in total apoptosis (from 6.5% in control to 11% with BFA + TM) is statistically significant, its biological impact is relatively modest compared to strong apoptosis inducers.

Table 3.

Effects of BFA and TM on the induction of apoptosis in HL-7702 cells.

	Early apoptotic rate (%)	Late apoptotic rate (%)	Total apoptotic rate (%)
Blank control	2.07 ± 0.38	4.70 ± 1.56	6.77 ± 1.94
DMSO	2.27 ± 0.55	4.20 ± 1.44	6.47 ± 1.37
0.25 mg/L BFA	3.27 ± 0.25	6.53 ± 2.21*	9.53 ± 1.26*
1 mg/L TM	3.00 ± 1.06	3.20 ± 0.00	6.47 ± 0.25
0.25 mg/L BFA + 1 mg/L TM	3.77 ± 1.60*	7.47 ± 2.63*#	11.23 ± 1.21*#

HL-7702 cells were treated with BFA (0.25 mg/L) and TM (1 mg/L), either individually or in combination, for 24 h. Data are presented as means ± SD (n = 3). * P < 0.05 compared with blank control; # P < 0.05 compared with 0.25 mg/L BFA.

3.4. Combination of BFA and TM increases the levels of caspase-12 but not PARP1 and cleaved caspase-3

To elucidate the mechanisms underlying the ability of BFA + TM to induce apoptosis in HL-7702 cells, we measured the expression levels of several key proteins associated with apoptosis, including PARP-1, caspase-12, and caspase-3 (Fig. 3). The results showed that BFA and BFA + TM significantly increased the protein levels of caspase-12 (P < 0.05), with cells treated with BFA + TM showing a higher increase than those treated with BFA alone (P < 0.05). However, BFA and TM, either alone or in combination, did not affect the protein levels of PARP-1, full-length and cleaved caspase-3. These results suggest that the ability of BFA to induce ER stress could be enhanced by TM, but BFA did not cause PARP-1 and caspase-3-mediated apoptosis. These results suggest that BFA led to ER stress through caspase-12 activation, but the subsequent apoptotic process appeared to be independent of PARP-1 and caspase-3.

Fig. 3.

Effects of BFA and TM on the levels of apoptosis-related proteins (mean ± SD, n = 3). HL-7702 cells were treated with BFA (0.25 mg/L) and TM (1 mg/L), either individually or in combination, for 24 h. The experiment was performed in triplicates. (A) PARP-1, (B) caspase-12, and (C) full-length and cleaved caspase-3. * P < 0.05, compared with the blank control; # P < 0.05, compared with 0.25 mg/L BFA.

Moreover, we measured the mRNA levels of PARP-1, caspase-12, and caspase-3 (Fig. 5). BFA and BFA + TM significantly increased the mRNA levels of PARP-1 and caspase-12 compared to the controls (P < 0.05), with BFA + TM upregulating a higher mRNA level than BFA alone (P < 0.05). In addition, BFA and TM, either alone or in combination, did not influence the mRNA levels of caspase-3, and TM alone did not affect the mRNA levels of these three genes.

Fig. 5.

Effects of BFA and TM on the mRNA levels of apoptotic genes (PARP-1, caspase-12, and caspase-3) in HL-7702 cells. HL-7702 cells were treated with BFA (0.25 mg/L) and TM (1 mg/L), either individually or in combination, for 24 h. Data are presented as means ± SD (n = 3) * P < 0.05, compared with blank control; ^#P < 0.05, compared with 0.25 mg/L BFA.

3.5. Combination of BFA and TM upregulates the expression of ER stress-related proteins (BiP and ATF4)

To observe the effect of BFA and TM combination on the induction of ER stress, we measured the expression levels of several key proteins involved in ER stress, including BiP, PERK, p-PERK (phosphorylated PERK), eIF2α, p-eIF2α (phosphorylated eIF2α), ATF4 and CHOP using Western blot (Fig. 4). BFA alone significantly increased the protein levels of BiP and ATF4 (P < 0.05), but had no effects on the levels of PERK, eIF2α, p-eIF2α, and CHOP. TM alone significantly increased the protein levels of BiP, PERK, and ATF4 (P < 0.05), but did not affect the levels of p-PERK, eIF2α, p-eIF2α, and CHOP. Notably, BFA + TM significantly increased the level of PERK (P < 0.05). The absence of increased p-PERK, p-eIF2α, and CHOP expression in BFA-treated cells suggests that BFA induces ER stress in normal hepatocytes through a non-canonical pathway that bypasses PERK phosphorylation and CHOP induction.

Fig. 4.

Effects of BFA and TM on the levels of ER stress-related proteins (mean ± SD, n = 3). HL-7702 cells were treated with BFA (0.25 mg/L) and TM (1 mg/L), either individually or in combination, for 24 h. (A) BiP, (B) PERK, (C) p-PERK, (D) eIF2α, (E) p-IFf2α, (F) ATF4, and (G) CHOP. * p < 0.05, compared with the blank control.

Furthermore, we measured the mRNA levels of BiP, PERK, eIF2α, ATF4, and CHOP using qRT-PCR (Fig. 6). BFA alone significantly increased the mRNA levels of BiP, eIF2α, and ATF4 (P < 0.05), but had no effects on the expression of PERK and CHOP. TM alone significantly increased the mRNA level of PERK, but had no impact on the levels of BiP, eIF2α, ATF4, and CHOP. BFA + TM significantly increased the mRNA levels of BiP, PERK, eIF2α, and ATF4 (P < 0.05), but did not affect the expression of CHOP. Notably, BiP and ATF4 levels were significantly higher in cells treated with BFA + TM than those treated with BFA alone (P < 0.05). Taken together, these results suggest that BFA could remarkably increase the expression of BiP and ATF4 in HL-7702 cells at both transcriptional and translational levels, which could be enhanced by TM.

Fig. 6.

Effects of BFA and TM on the levels of ER stress-related genes (BiP, PERK, eIF2α, ATF4, and CHOP) in HL-7702 cells. HL-7702 cells were treated with BFA (0.25 mg/L) and TM (1 mg/L), either individually or in combination, for 24 h. Data are presented as means ± SD (n = 3). * P < 0.05, compared with the blank control; # P < 0.05, compared with 0.25 mg/L BFA.

4. Discussion

BFA has been regarded as a promising candidate for anticancer drugs due to its high anti-tumor activity. (Xue et al., 2010, Gao et al., 2019) Studies have found that BFA can induce apoptosis in various cancer cells; such as melanoma, prostate cancer, breast cancer, liver cancer, and colon cancer. (Li et al., 2025, Huang et al., 2017, Luo et al., 2020, Gao et al., 2023) However, its potential toxic side effects on normal cells should be considered. Since the liver is the primary organ for drug metabolism, drug-induced hepatotoxicity has become one of the important factors restricting drug development. Here, we used a normal human liver cell line (HL-7702) as a model to address this challenge. The results show that BFA only induced relatively weak apoptosis in HL-7702 cells. However, when administered in combination with TM (a common ER stress-inducer), the harmful effect of BFA increases to a certain degree, which should be considered in clinical practice.

Generally, BFA exerts the anti-tumor effect through its apoptosis-inducing ability in tumor cells. (Markouli et al., 2020, Gruber et al., 2023) However, it remains unclear whether BFA could induce apoptosis in normal hepatocytes, which is very important for its clinical application. To address this question, we observed the apoptosis-inducing ability of BFA in normal hepatocytes, along with the effects on the protein and mRNA levels of several key apoptosis-related factors. The results showed that BFA alone or combination of BFA and TM caused relatively low apoptotic rates of about 9.53% or 11.23% in HL-7702 cells, which are markedly lower than 32.23% in hepatocytes caused by acetaminophen (a widely used pain reliever). (Bai et al., 2018) Caspase-12 is an ER stress-specific protease and also participates in the apoptotic process. (Zhang et al., 2023) Here; we found that BFA alone and its combination with TM significantly elevated the protein levels of caspase-12 in HL-7702 cells, indicating that BFA could cause ER stress to some degree in normal hepatocytes and this ability could be enhanced by TM. This enhancement might be due to the synergistic inhibition of protein glycosylation by BFA and TM. (Kopp et al., 2024, Yoon et al., 2023) In addition; we found that BFA had no influence on the protein levels of PARP-1 and caspase-3, as well as the cleavage of caspase-3. PARP-1 is mainly involved in cellular poly(ADP-ribose) metabolism. (Liu et al., 2022, Chen et al., 2024, Lu et al., 2024) After activation, PARP-1 catalyzes the degradation of nicotinamide adenine dinucleotide (NAD) into nicotinic acid and ADP. ADP binds to receptor proteins to synthesize PAR, which promotes DNA repair by recruiting a large number of DNA repair proteins. (Virág and Szabó, 2002, Wang et al., 2019, Pascal, 2018) Caspase-3 is a key effector enzyme in apoptosis, and its cleavage is a necessary step for apoptosis. Therefore, we inferred that BFA might not cause DNA damage and trigger caspase-3-mediated apoptosis in HL-7702 cells. Although BFA could increase the apoptotic rate of HL-7702 cells to some extent, its ability to induce apoptosis in normal hepatocytes was relatively weak, and it might be caspase-12-dependent and caspase-3-independent. In fact, there are many cases in which caspase-3 is not activated during apoptosis. (Zulliger et al., 2011, Karmakar et al., 2006, Mo et al., 2014).

Many factors can induce apoptosis, such as ER stress, DNA damage, and mitochondrial dysfunction. (Faccioli et al., 2025, Yan et al., 2021, Kuang et al., 2025) Here, we observed that BFA and its combination with TM significantly increased the protein and mRNA levels of BiP and ATF4 in HL-7702 cells. BiP is a marker protein for ER stress, and its increased expression suggests the occurrence of unfolded protein response. (Chen et al., 2015) In general, ER stress can induce increased expression of ATF4, (Tang et al., 2024) which prevents compensatory cell proliferation and tumorigenesis in hepatocytes by maintaining glutathione synthesis and inhibiting ferroptosis-dependent inflammatory cell death. (He et al., 2023) Interestingly, our study found that BFA did not affect the expression of other proteins involved in the typical PERK pathway, such as p-PERK, p-eIF2α, and CHOP. This differed from our previous findings in HepG2 hepatoma cells, where BFA and TM combination robustly activated the PERK-eIF2α-ATF4-CHOP signaling axis. (Li et al., 2025) This discrepancy suggests a fundamental difference in how normal hepatocytes and hepatoma cells sense and respond to BFA-induced ER stress. A regulatory network composed of 4EHP (eIF4E-homologous protein), NELF-E, the 40S ribosomal subunit, and eIF3 subunits has been found to regulate ATF4 expression. Knockdown of these factors inhibits the translation of ATF4, indicating a regulatory mechanism that is not directly dependent on eIF2α phosphorylation. (Walsh et al., 2025 Dec 23) CHOP is an important regulatory factor in ER stress-induced apoptosis, and it induces apoptosis by upregulating the expression of a pro-apoptotic protein Bax. (Dag et al., 2025) Strong ER stress can induce oxidative stress and impair mitochondrial function, leading to CHOP-dependent apoptosis. (Qureshi et al., 2023) However, mild ER stress is an adaptive response of cells and does not lead to cell death. In this study, we found that BFA and TM did not affect the mRNA and protein levels of CHOP in HL-7702 cells. As mentioned above, we also found that BFA did not affect PARP-1 and cleavage of caspase-3, and these results suggest that BFA might have limited apoptosis-inducing effects in normal hepatocytes.

This study has some limitations. First, the assessment of combined effects was performed at a single concentration combination (0.25 mg/L BFA + 1 mg/L TM). A full dose-dose matrix analysis would be necessary to formally classify the interaction (e.g., synergy, additivity) between BFA and TM. Second, the stability of β-actin as a single reference gene for qRT-PCR normalization was not verified with other housekeeping genes under our specific treatment conditions. While β-actin is commonly used for HL-7702 cells, future studies should employ multiple reference genes to ensure robust normalization.

Taken together, our results demonstrated that BFA and TM, either alone or in combination, exerted a mild pro-apoptotic effect in normal hepatocytes. Notably, the induction of apoptosis was CHOP- and caspase-3-independent but caspase-12-dependent. BFA could also trigger a non-canonical ER stress response in normal hepatocytes, which differs from the pathway observed in liver cancer cells and might be protective. Given the hepatotoxicity, albeit mild, the adverse effects of BFA warrant particular attention and strict control in clinical applications. Further studies directly comparing the relative potency and efficacy of BFA in tumor versus normal liver cells are essential to define its therapeutic window.

Declaration of competing interest

The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper.