4-Phenylbutyric | Phosphorylase Signal

FoodandChemicalToXicology

Involvement of PERK-CHOP pathway in fumonisin B1- induced cytotoXicity in human gastric epithelial cells

Song Yu, Bingxuan Jia, Yunxia Yang, Na Liu, Aibo Wu∗
SIBS-UGENT-SJTU Joint Laboratory of Mycotoxin Research, CAS Key Laboratory of Nutrition, Metabolism and Food Safety, Shanghai Institute of Nutrition and Health, Shanghai Institutes for Biological Sciences, University of Chinese Academy of Sciences, Chinese Academy of Sciences, 200031, Shanghai, PR China

A R T I C L E I N F O

Keywords: Fumonisin B1 Apoptosis
ER stress
PERK-CHOP pathway GES-1 cells
Flow cytometry

A B S T R A C T

Fumonisin B1 (FB1) is a mycotoXin, produced by Fusarium verticillioides and Fusarium proliferatum, and a common fungal contaminant of maize worldwide. Its potential health hazard as a natural toXin is well documented in human and domestic animals. However, the molecular mechanism and the key factors responsible for FB1- induced cytotoXicity have not been elucidated. In this study, we ﬁrst examined the cytotoXicity induced by FB1 in human gastric epithelial cell line (GES-1). We found that FB1 notably decreased cell viability and induced apoptotic cell death. Furthermore, the levels of ER stress markers were signiﬁcantly increased after FB1 exposure and the ER stress inhibitor 4-phenylbutyric acid strongly suppressed FB1-induced cytotoXicity. Interestingly, the inhibition of PERK activity by GSK2606414 or shPERK3 blocked FB1-induced apoptotic cell death and cell proliferation suppression, which indicated that the cytotoXicity induced by FB1 was dependent on this signalling pathway. Moreover, myriocin could relieve FB1-induced ER stress and prevent cell death, which implied that the disruption of sphingolipid metabolism is an apical event for FB1-induced cytotoXicity. In the present study, we demonstrated that the ER stress-related PERK-CHOP signalling pathway is a novel mechanism for FB1-induced cytotoXicity and the gastrointestinal injury caused by FB1 should be concerned in the future.

1. Introduction

MycotoXins constitute a class of toXins produced as natural con- taminants under certain environmental conditions by fungus in food- stuﬀs. The Food and Agriculture Organization (FAO) of the United Nations estimated that more than 25% of cereal crops are lost annually due to mycotoXin contamination that aﬀects the whole food chain. MycotoXins in foods can adversely aﬀect the health of human and an- imal. They exhibit toXic, mutagenic, teratogenic and carcinogenic ef- fects (Acaroz, 2019). Measures to overcome mycotoXin contamination have been seriously challenged globally because of their thermal sta- bility and resistance to acid-base media (Shi et al., 2018). Fumonisins are a group of mycotoXins produced by Fusarium verticillioides and Fu- sarium proliferatum primarily that have been widely found as con- taminants all over the world. In 2011, the results of a survey of my- cotoXin contamination of 4327-grain samples around the world showed that the proportion testing positive for fumonisin B in North America,

Central Europe, Africa, South Asia and Southeast Asia was 27%, 51%, 58%, 56% and 55%, respectively. Speciﬁcally, the highest prevalence was observed in South America, for which the fumonisin B-positive rate was 76% (average level of 1.501 mg/kg) (Biomin, 2011). Fumonisin B mainly contaminates corn and corn products. In areas with high levels of pollution, people whose dietary staple is corn are at higher risk of exposure. For example, in East Africa and Linxian, China, the incidence of fumonisin B-related oesophageal cancer is much higher than it is in areas with lower levels of contamination (Kimanya, 2015; Wang et al., 2008).
Fumonisin B1 (FB1), the most prevalent member of the fumonisins, has been shown to be responsible for most of the toXicological eﬀects associated with these contaminants (Janse van Rensburg et al., 2017; Silva et al., 2017; Waśkiewicz et al., 2012). FB1 causes diverse toXic eﬀects in human and domestic animals, including neurotoXicity, he- patotoXicity and carcinogenesis, through oXidative stress, apoptosis, necrosis and alterations in cell proliferation and diﬀerentiation (Escriva

Abbreviations: 4-PBA, 4-phenylbutyric acid; ATF4, activating transcription factor 4; Bip, glucose-regulated protein 78; CCK-8, Cell Counting Kit-8; CHOP, C/EBP homologous protein; Ctrl, control; ER, endoplasmic reticulum; FB1, fumonisin B1; GES-1, human gastric epithelial cell line; GIT, gastrointestinal tract; GSK, GSK2606414; H-DMEM, dulbecco’s modiﬁed eagle medium with high glucose; IECs, intestinal epithelial cells; ISP-1, myriocin; LDH, lactic dehydrogenase; PCD, programmed cell death; PERK, protein kinase R-like ER kinase; Sa, sphinganine; So, sphingosine; SPT, serine palmitoyltransferase

∗ Corresponding author. Building 34, 294 Taiyuan Road, Shanghai, China.
E-mail address: [email protected] (A. Wu).et al., 2015). Several lines of evidence indicate that the toXicity of FB1 is caused by the accumulation of free sphingoid bases via inhibition of ceramide synthase, which catalyses the formation of ceramide from sphinganine to sphingosine. Grenier et al. found that FB1 exposure leads to increased sphinganine (Sa)/sphingosine (So) ratios in the liver and plasma of pigs (Grenier et al., 2012), and the same phenomenon was found in Sprague Dawley rats (Hahn et al., 2015). In animal models, the Sa/So ratio serves as a speciﬁc biomarker for the evaluation of fumonisin toXicity (Riley et al., 2011). Yin et al. reported that pharmacologically inhibiting the accumulation of free sphinganine by myriocin leads to the abrogation of FB1-induced autophagic cell death in MARC-145 cells (Yin et al., 2015). However, the deﬁnitive me- chanism of toXicity inducement remains unknown. We need to make a better understand on the mechanisms behind FB1 toXicity to develop more eﬀective methods to minimize the health risks associated with exposure to FB1.

Apoptosis is programmed cell death and plays an important role in
maintaining development processes (Varela-Nieto et al., 2019). In ad- dition, it is involved in many diseases (Jaeschke et al., 2018; Zhang et al., 2018a). EXposure to external stress factors can disrupt the homeostasis of the endoplasmic reticulum (ER), thereby triggering apoptosis and eliminating damaged cells. Studies have shown that ER stress-mediated apoptosis contributes to the toXiﬁcation initiated by mycotoXin in past years. Zearalenone mediated the cell death of RAW
264.7 macrophages by inducing ER stress, however, cell death was prevented by 4-PBA and shCHOP (Chen et al., 2015). OchratoXin-in- duced autophagy in the liver and kidneys of pigs was related to ER stress (Gan et al., 2017). Protein kinase R-like ER kinase (PERK), an ER- located type I transmembrane protein, was activated by ER stress in- duced by an increase in the pro-apoptotic transcription factor C/EBP homologous protein (CHOP), which regulated cell apoptosis (Chen et al., 2013; Marciniak et al., 2004; Masuda et al., 2013), thereby participating in the occurrence and development of various diseases, such as diabetes (Wang et al., 2010), tumours (Bobrovnikova-Marjon et al., 2010), neuron degeneration (Park et al., 2008) and myocardial ischemia-reperfusion injury (Yu et al., 2016). Based on these results, we hypothesized that the PERK-CHOP signalling pathway may also be a critical signal transduction mechanism in FB1-induced cytotoXicity.
Since the gastrointestinal tract is the ﬁrst physiological barrier
against food contaminants, it is the ﬁrst target for toXicants (Akbari et al., 2017). FB1 has been studied for its adverse eﬀects on the function of the gastrointestinal tract (GIT) in animals. Ingestion of FB1 induced an increase in heat shock proteins in the GIT (Lalles et al., 2010) and modulated intestinal microbial homeostasis (Antonissen et al., 2015). FB1 disrupts the barrier function (Bouhet et al., 2004; Lalles et al., 2009) and alters chemokine expression in intestinal epithelial cells (IECs) and gut tissues (Bouhet et al., 2006). Nevertheless, the eﬀect of FB1 on human GIT toXicity remains unclear. In this research, in order to obey the rules of 3R, human gastric epithelial cell line (GES-1) would be used as in vitro model to explore the molecular mechanism of FB1 toXicity. The results revealed that the ER stress-related PERK-CHOP signalling pathway is a novel mechanism for FB1-induced cytotoXicity.

2. Materials and methods

2.1. Chemicals

Fumonisin B1 (ab142433) was from Abcam (Cambridge, MA, USA). Annexin V-FITC Apoptosis Detection Kit I (556547) was obtained from BD Pharmingen (San Diego, CA, USA). A Cell Counting Kit-8 (CCK-8) and CytotoXicity Lactic Dehydrogenase (LDH) assay kit were purchased from DOJINDO Laboratories (Kumamoto, Japan). The primary anti- bodies used for immunoblotting analyses against Caspase-3 (9668S), Bcl2 (15071S), Bax (2772S), Bip (3177S), ATF4 (11815S), PERK (5683S), CHOP (2895S), Actin (3700S), Cytochrome C (4280S), and
GADD45α (4632S) were purchased from Cell Signalling Technology
(Danvers, MA, USA). NC membrane (FFN02) was from Beyotime Biotech (Nantong, China). Bicinchoninic acid assay kit (20201ES90) was purchased from YEASEN Biotechnology (Shanghai, China). PrimeScript™ RT Master MiX (RR036A) and SYBR® PremiX EX Taq™ II (RR820A) were obtained from Takara Biomedical Technology (Beijing, China). Trypsin EDTA Solution A (0.25%) (03-050-1B) and penicillin- streptomycin-amphotericin B solution (03-033-1B) were obtained from BioInd (Kibbutz Beit, Israel). Myriocin (ISP-1) (476300-5 MG) was purchased from Merck/Millipore (Billerica, MA, USA). Dulbecco’s modiﬁed eagle medium with high glucose (H-DMEM) (SH30243.01) was purchased from HyClone (South Logan, UT, USA). The fetal bovine serum and TRIzol reagent were obtained from Invitrogen (Waltham, MA USA). GSK2606414 (HY-18072-5mg) and 4-phenylbutyric acid (4- PBA) (HY-A0281-5 g) were acquired from MCE (Monmouth, NJ, USA).

2.2. Cell culture and treatments

Human gastric epithelial cell line (GES-1) was purchased from Beijing Beina Chuanglian Biotechnology Institute (Beijing, China). The GES-1 cells were cultured in H-DMEM (HyClone) supplemented with 10% fetal bovine serum (Invitrogen) and penicillin-streptomycin-am- photericin B (BioInd). After plating when cells were 60–70% conﬂuent,
cells were treated with FB1 and/or other agents.

2.3. Cell viability assay and membrane leakage assay

Cell viability assay and cell membrane leakage assay were carried out with a Cell Counting Kit-8 and a CytotoXicity LDH assay kit (DOJINDO) according to the manufacturers’ instructions, respectively. Each group had ﬁve duplicate wells on a microtiter plate, and a mi- crotiter plate was set as an individual experiment. The results are re-
presentative of four individual experiments. Absorbance was recorded with a Tecan GENios Pro ﬂat-panel reader.

2.4. Cell death analysis

Cell death was measured by ﬂow cytometry assay using an Annexin V-FITC Apoptosis Detection Kit I (BD Pharmingen). The GES-1 cells
were treated with FB1 (0–40 μM) for 48 h and then stained with PI and Annexin V-FITC according to the manufacturers’ instructions.

2.5. Immunoblotting

The cells at 60–70% conﬂuence were treated with FB1 for 48 h and then washed once with phosphate-buﬀered saline and harvested. The harvested cells were lysed for 30 min and then spun at 13,000 rpm for
20 min at 4 °C. The soluble fraction was collected, and the protein concentration was measured by a bicinchoninic acid assay (YEASEN Biotechnology). Then, the protein samples and pre-stained molecular weight markers were subjected to 10% or 12.5% sodium dodecyl sul- fate-polyacrylamide gel electrophoresis and transferred to NC mem- branes (Beyotime Biotech) by electroblotting at 4 °C. Subsequently, the membranes were blocked with 10% non-fat dry milk in Tris-buﬀered saline and Tween 20 for 2 h at room temperature and incubated with speciﬁc antibodies at a 1:1000 or 1:2000 dilution overnight at 4 °C. The second antibody, IgG labelled with horseradish peroXidase, was diluted to 1:2000 and incubated with the membrane at room temperature for
2 h, and the stained proteins were visualized by an electro- chemiluminescence detection system. Snygene-Image Systems were used to quantify the band density.

2.6. Real-time PCR

Total RNA from the GES-1 cells was prepared with TRIzol reagent (Invitrogen). RNA samples were used to synthesize cDNA using PrimeScript™ RT Master MiX (Takara). Each sample was quantiﬁed using SYBR® PremiX EX Taq™ II (Takara). The primer sequences and additional information are shown in Table S1 and Table S2. Gene ex- pression was analysed by the comparative CT method. The expression level of each gene was normalized to the expression level of actin by the standard curve method.

2.7. shRNA-mediated knockdowns

Four shRNAs speciﬁc to human PERK (see Table S3) and a negative control, shNC, were synthesized and cloned into pGPU6/GFP/Neo by the Jima Gene Company (Shanghai, China). GES-1 cells were then transfected with vectors carrying shPERK or shNC vectors by transfec- tion reagents (DNAfectin™ Plus transfection reagent, ABM, Canada) to knock down the gene.

2.8. Statistical analysis

Data shown throughout this study represent the mean ± SD of four individual experiments. Data were assessed using a one-way or two-way ANOVA with Tukey or post hoc Bonferroni testing to assess progressive changes between groups by Prism software (GraphPad Prism 5). All probabilities were two-sided, and a value of P < 0.05 was considered statistically signiﬁcant. Statistical analyses were performed with GraphPad Prism 5 software. All methods were carried out in accordance with the relevant guidelines and regulations.

3. Results

3.1. FB1 inhibited cell proliferation and increased LDH levels in GES-1 cells

Cell viability and cell membrane leakage are two independent in-

3.2. Eﬀects of FB1 on cell death in GES-1 cells

We further investigated the eﬀect of FB1 on the rate of cell death for the GES-1 cells. Flow cytometry analysis (Annexin V-FITC/PI staining) was used to quantify the extent of apoptosis and necrosis in the total cell population. PI stains cells in late apoptosis and necrotic cells, and Annexin V-FITC stains cells undergoing the early stages of apoptosis. After incubation with FB1 for 48 h, the percentage of cell death in the
10, 20 and 40 μM FB1 treatment groups remarkably increased, com- pared with control group (Fig. 2A and B). To determine the possible
mechanism of cell death induced by FB1, we analysed the markers of apoptosis and DNA damage by immunoblotting and real-time PCR. The protein expression levels of total caspase-3 were signiﬁcantly decreased and those of cleaved caspase-3 were signiﬁcantly increased compared with control group. The protein expression levels of Bax, GADD45α and
Cytochrome C were also dramatically enhanced; however, the expres-
sion of Bcl2 was obviously reduced (Fig. 2C and D). In addition, the mRNA expression levels showed the same trend (Fig. 2E).

3.3. FB1 induced ER stress in GES-1 cells

It is known that ER stress is activated and induces apoptosis when cells are threatened by the external environment. We veriﬁed whether FB1 caused ER stress in the GES-1 cells. Glucose-regulated protein 78 (Bip), activating transcription factor 4 (ATF4) and CHOP are bio- markers of ER stress. The immunoblotting results showed that the levels of Bip, ATF4 and CHOP were signiﬁcantly increased after the cells were
treated with 20 μM FB1 for 48 h (Fig. 3A and B). In addition, the mRNA expression levels were also signiﬁcantly upregulated (Fig. 3C). These
ﬁndings suggested that FB1 induced ER stress in GES-1 cells.

3.4. ER stress involved FB1-induced cytotoxicity
dicators often used to evaluate
cytotoXicity.
The CCK-8 kit and

CytotoXicity LDH assay kit were used to detect the cytotoXicity of FB1 in GES-1 cells. As shown in Fig. 1, cell viability was decreased and lactic dehydrogenase (LDH) was increased in a dose- and time-dependent manner in the presence of 2.5–40 μM FB1. After 24 h of exposure to
FB1, the GES-1 cells were treated with 10 μM FB1, which resulted in a
signiﬁcant decrease in cell viability, the cell viability was gradually decreased with increased dose by time. The GES-1 cells had the lowest
viability level under 40 μM FB1 after 48 h (Fig. 1A). Simultaneously, when GES-1 cells were treated with 5 μM FB1 for 24 h, the LDH leakage
rates were signiﬁcantly diﬀerent from the rate in the control group. The LDH leakage rate was reached at the highest level, compared with control group at a 40 μM dose after 48 h (Fig. 1B).

Our previous results showed that FB1 induced cytotoXicity and ER stress. To further conﬁrm that ER stress is involved in the cytotoXicity induced by FB1, we exposed the GES-1 cells to the ER stress inhibitor 4- phenylbutyric acid (4-PBA) before we applied the FB1 treatment. As shown in Fig. 4A–C, 4-PBA strongly reduced the expression levels of Bip
and CHOP in the FB1-treated cells. In addition, 4-PBA blocked the FB1-
induced cell death and cell proliferation suppression. The results from the ﬂow cytometry analysis showed that the FB1-induced cell death rate decreased after 4-PBA treatment with no signiﬁcant diﬀerence compared with control group (Fig. 4D and E). 4-PBA observably in- creased the cell viability in the FB1-treated cells (Fig. 4F). In summary, the results demonstrated that FB1-induced cytotoXicity depended on ER
Fig. 1. FB1-induced cytotoXicity in GES-1 cells as shown by reducing cell viability and increasing LDH leakage. GES-1 cells were treated with FB1 (0, 2.5, 5, 10, 20, and 40 μM) for 24 and 48 h and then evaluated. (A) Cell viability was assessed by a Cell Counting Kit-8 cell proliferation assay. (B) Membrane integrity was determined by measuring the LDH leakage in cell culture media using a CytotoXicity LDH assay kit. Each group had ﬁve duplicate wells on a microtiter plate, and a
microtiter plate was set as an individual experiment. Data represent the mean ± SD of four individual experiments based on ﬁve duplicate wells (*p < 0.05,
**p < 0.01 and ***p < 0.001).
Fig. 2. FB1-induced apoptosis in GES-1 cells. (A–B) Cells were treated with FB1 for 48 h. Results from the data analysis of the GES-1 cell death induced by FB1, as measured by Annexin V-FITC/PI. GES-1 cells were treated with 20 μM FB1 for 48 h and then harvested. The markers of apoptosis and DNA damage were determined by immunoblotting (C) and real-time PCR (E). (D) Results from the protein quantiﬁcation analysis by ImageJ. Data are depicted as the mean ± SD from four
individual experiments (*p < 0.05, **p < 0.01 and ***p < 0.001).stress.

3.5. FB1 induced cytotoxicity through the ER stress-related PERK-CHOP signalling pathway

To substantiate the hypothesis that FB1 mediates cytotoXicity through the PERK-CHOP signalling pathway, we constructed shPERKs to knock down PERK activity and then observed the changes to cyto- toXicity and the expression levels of downstream genes. According to the experimental data, PERK was knocked out by shPERK1, shPERK2
and shPERK3 (Fig. 5A and B). We selected shPERK3 to perform sub- sequent experiments. As shown in Fig. 5C and D, we transfected shPERK into cells before subjecting them to the FB1 treatment and found that the shPERK transfection successfully reduced the levels of PERK, ATF4, and CHOP (Fig. 5C and D). In addition, FB1-induced cell death was suppressed and the cell viability was increased (Fig. 5E and F). Fur- thermore, we also used the PERK inhibitor GSK2606414 before FB1 treatment and found that it had the same eﬀect as shPERK (Fig. 5G and H). Overall, our data indicated that FB1 induced cytotoXicity through the ER stress-related PERK-CHOP signalling pathway.

Fig. 3. FB1 induced ER stress in GES-1 cells. GES-1 cells were treated with 20 μM FB1 for 48 h and then harvested. The markers of ER
stress were determined by immunoblotting
(A) and real-time PCR (C). (B) Results from the protein quantiﬁcation analysis by ImageJ. Data are depicted as the mean ± SD from four individual experi- ments (*p < 0.05, p < 0.01 and p < 0.001).
Fig. 4. FB1-induced apoptosis was ER stress-dependent. GES-1 cells were exposed to the ER stress inhibitor 4-PBA before treatment with 20 μM FB1. (A) The expression of ER stress markers was detected by immunoblotting, and the band quantiﬁcation of Bip (B) and CHOP (C) was performed using ImageJ software. (D–E) Data analysis of the GES-1 cell death induced by FB1 as measured by Annexin V-FITC/PI. (F) The cell viability was evaluated after treated with 20 μM FB1 for 48 h
with ER stress inhibitor 4-PBA. Each group had ﬁve duplicate wells on a microtiter plate, and a microtiter plate was set as an individual experiment. Data are depicted as the mean ± SD from four individual experiments (*p < 0.05, **p < 0.01 and ***p < 0.001).

Fig. 5. Blocking the PERK-CHOP signalling pathway with shPERK and GSK2606414 mitigated the cytotoXicity that had been induced by FB1. (A) GES-1 cells were treated with a vehicle, control shRNA, or PERK shRNA. PERK expression was detected by immunoblotting, and the bands were quantiﬁed using ImageJ software (B).
(C) GES-1 cells were treated with shPERK and then cultured with FB1 for 48 h. The expression of PERK, ATF4 and CHOP in GES-1 cells was assayed by im- munoblotting, and the bands were quantiﬁed using ImageJ software (D). (E–H) shPERK and GSK2606414 suppressed FB1-induced cell death and cell proliferation. Each group had ﬁve duplicate wells on a microtiter plate, and a microtiter plate was set as an individual experiment. Data are depicted as the mean ± SD from four
individual experiments (*p < 0.05, **p < 0.01 and ***p < 0.001).
Fig. 6. Disruption of sphingolipid metabolism was an apical event in FB1-induced cytotoXicity in GES-1 cells. The cells were treated with 20 μM FB1 in the presence or absence of ISP-1 for 48 h. (A–B) Eﬀects of ISP-1 on FB1-induced cell death were analysed by Annexin V-FITC/PI staining. (C) Viability of the cells treated with 20 μM FB1 for 48 h with ISP-1. Each group had ﬁve duplicate wells on a microtiter plate, and a microtiter plate was set as an individual experiment. (D) The expression of ER stress markers was detected by immunoblotting, and the Bip band (E) and the CHOP band (F) were quantiﬁed using ImageJ software. Data are
depicted as the mean ± SD from four individual experiments (*p < 0.05, **p < 0.01 and p < 0.001).
3.6. Sphingolipid metabolism disorder contributed to the ER stress-mediated cytotoxicity induced by FB1

As FB1 has a structure similar to that of sphingosine, it can disrupt sphingolipid metabolism by inhibiting ceramide synthase, leading to the accumulation of intracellular free sphingoid bases. This accumula-

induces programmed cell death (PCD); and causes diseases in plants (Ormancey et al., 2019; Qin et al., 2017), however, there was a lack of studies on the toXicity of FB1 to other major functional tissues. The tissue speciﬁcity and key molecular mechanisms of FB1 toXicity have not been fully elucidated. Due to the increasingly strict regulations on using animal and ensuring their welfare, cell-based systems are moretion has been suggested to play a key role in FB1toXicity favourable and applicable for toXicity evaluations. We conducted a(Chandrasekhar and Sreelakshmi, 2012). To determine the relationship between FB1-induced cytotoXicity and the accumulation of free sphin- goid bases, we measured the rates of cell death and proliferation in the presence of myriocin (ISP-1) (Hojjati et al., 2005), a potent speciﬁc inhibitor of serine palmitoyltransferase (SPT), which is the ﬁrst enzyme in the sphingolipid biosynthesis pathway, to prevent the accumulation of free sphingoid bases. As shown in Fig. 6A and B, FB1-induced cell death was completely abolished in the presence of ISP-1. Moreover, ISP- 1 almost completely prevented the overall decrease in cell proliferation that had been induced by FB1 (Fig. 6C). The increase in the levels of ER stress markers (Bip and CHOP), which have been identiﬁed as playing

preliminary toXicity evaluation of FB1 on human renal epithelial cell line HK-2, liver cell line LX-2 and gastric epithelial cell line GES-1. Our data showed that the GES-1 cells were more sensitive to FB1 compared to the sensitivity of the LX-2 or HK-2 cells (data not shown), and the cell viability of the GES-1 cells was observably reduced and the rate of cell apoptosis was remarkably increased in a dose- and time-dependent manner after FB1 treatment (Figs. 1 and 2). These data suggested that FB1 is likely to cause GIT damage.
In previous studies, apoptosis was found to be a consequence of FB1- induced toXicity. Zhang et al. discovered that FB1 could induce apop- tosis in PK15 cells via mitochondria-dependent channels that are critical roles and being largely responsible for the cytotoXicity inducedMmediated by oXidative stress (Zhang et al., 2018b). FB1 induced by FB1, was also diminished in the presence of the inhibitor (Fig. 6D–F). These results indicated that disruption of sphingolipid metabolism had a key role in the FB1-induced PERK-CHOP-mediated cytotoXicity in GES-1 cells.

4. Discussion

Here, we used human gastric epithelial cell line (GES-1) as an in vitro study model and demonstrated that the ER stress-related PERK- CHOP signalling pathway is a novel mechanism for FB1-induced cyto- toXicity. Although numerous studies have demonstrated that FB1 in- duces a sphingolipid metabolism disorder; acts as a neurotoXicity (Domijan and Abramov, 2011), hepatotoXicity (Sharma et al., 2006), and nephrotoXicity (Muller et al., 2012) in both human and animals;
apoptosis in LLC-PK1 renal epithelial cells via a sphinganine- and cal- modulin-dependent pathway (Kim et al., 2001). FB1 also induced apoptosis or cell cycle arrest in CV-1 cells through the tumour necrosis factor pathway (Ciacci-Zanella and Jones, 1999). In vivo experiments have shown that, after acute exposure, FB1 induced alterations in the cytokine expression and the apoptosis signalling genes in mouse liver and kidney (Bhandari and Sharma, 2002). We found that FB1 treatment signiﬁcantly reduced GES-1 cell viability and induced cell death in a dose- and time-dependent manner. In addition, FB1 also enhanced the level of apoptotic marker expression. However, the observed cell apoptosis may have been mainly caused by FB1 treatment at low doses or for a short time, while both apoptosis and necrosis may be the results of FB1 treatment at high doses and over an extended time.
The endoplasmic reticulum (ER) is the main site of protein and lipid
Fig. 7. The signalling pathways involved in FB1-induced apoptotic cell death in the GES-1 cells.synthesis, membrane biogenesis, Xenobiotic detoXiﬁcation and cellular calcium storage, and perturbation of ER homeostasis leads to stress and the activation of the unfolded protein response. ER transmembrane proteins dissociate with the ER chaperone molecular Bip, which binds to the accumulating unfolded proteins and transduces the ER stress signalling pathway. This signalling pathway upregulates the expression of molecular chaperones in the ER, subsequently restoring homeostasis and promoting cell survival. However, with severe and prolonged ER stress, the pro-apoptotic factor CHOP is activated and induces apoptosis in some cell types (Forus et al., 1994; Kim et al., 2019; Luo et al., 2017; Oyadomari et al., 2002). Chen et al. veriﬁed that ER stress was involved in the zearalenone-induced cell death of RAW 264.7 macrophages in which the activation of CHOP plays a critical role (Chen et al., 2015). Singh et al. demonstrated that FB1 could induce the upregulation of the expressed levels of ER stress markers (PERK, Bip and CHOP) in HepG2 cells and mouse liver (Singh and Kang, 2017). However, whether ER stress plays an important role in FB1-induced toXicity remains un- known. In the present study, markers of ER stress (Bip and CHOP) were found to be upregulated in the FB1-treated cells. Furthermore, we also found that 4-PBA (an ER stress inhibitor) markedly reduced the FB1- induced upregulation of Bip and CHOP in GES-1 cells and subsequently increased the percentage of cell death. To further explore the down- stream signalling pathway, shPERK and GSK2606414 were used to in- hibit PERK activation. The results of these experiments suggested that suppressed PERK activation prevented the expression of CHOP and the apoptotic cell death induced by FB1. These ﬁndings indicated that the ER stress-related PERK-CHOP pathway plays a pivotal role in FB1-in- duced apoptotic cell death.

As discussed above, FB1 competitively inhibits ceramide synthase. Inhibition of ceramide synthase by FB1 resulted in elevated in- tracellular free sphinganine base (van der Westhuizen et al., 2004). The perturbation of sphingolipid metabolism has been considered a key event that leads FB1 to exert its toXic eﬀects (Soriano et al., 2005). ISP-1 eﬀectively prevented the accumulation of the free sphingoid bases that had been induced by FB1 in pig kidney epithelial cells (He et al., 2001). Therefore, we investigated the contribution of sphingolipid metabolism to FB1-induced apoptotic cell death in GES-1 cells. ISP-1 is a chemical that speciﬁcally inhibits the activity of SPT, a key enzyme responsible for the de novo synthesis of sphinganine. In our study, the eﬀect of FB1 on ER stress and apoptotic cell death was evaluated in the presence of ISP-1. The results demonstrated that FB1-induced apoptotic cell death had been almost completely prevented in the presence of ISP-1, a ﬁnding that was consistent with a decrease in the levels of the ER stress markers (Bip and CHOP). The data strongly indicate that disruption of sphingolipid metabolism is an essential event involved in the FB1 triggering of apoptotic cell death.
In summary, in this in vitro study, we found that FB1 signiﬁcantly
decreased cell viability and induced apoptotic cell death in GES-1 cells. The ER stress-related PERK-CHOP signalling pathway plays a pivotal role in FB1-induced apoptotic cell death. This study oﬀers new insights into the molecular mechanisms of FB1 cytotoXicity (Fig. 7). Gastro- intestinal injury caused by FB1 would be concerned in the future on the basis of these scientiﬁc ﬁndings. Next, we will build appropriate in vitro and in vivo models to study the adverse eﬀects of FB1 on oesophageal and gastrointestinal functions. The dose, exposure time and signalling pathway described in this study may provide reference for future stu- dies.

Author contributions section

Song Yu: Conceptualization, Methodology, Formal analysis, Validation, Writing – Original Draft, Writing – Review & Editing
Bingxuan Jia: Formal analysis, Investigation Yunxia Yang: Resources, Investigation
Na Liu：Software, Writing – Review & Editing
Aibo Wu: Conceptualization, Funding acquisition, Writing – Review & Editing

Declaration of competing interest

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

Acknowledgements

We would like to thank the National Key Research and Development Program of China (2017YFC1600304), MOST (2016YFE0112900 and
201513006-02-3), National Natural Science Foundation of China (31772087) and Shanghai Agriculture Applied Technology Development Program, China (2019-02-08-00-02-F01145) for the ﬁ- nancial support.

Appendix A. Supplementary data

References

Acaroz, U., 2019. Determination of the total aﬂatoXin level in red pepper marketed in afyonkarahisar, Turkey. Fresenius Environ. Bull. 28 (4A), 3276–3280.
Akbari, P., Braber, S., Varasteh, S., Alizadeh, A., Garssen, J., Fink-Gremmels, J., 2017. The intestinal barrier as an emerging target in the toXicological assessment of mycotoXins. Arch. ToXicol. 91, 1007–1029.
Antonissen, G., Croubels, S., Pasmans, F., Ducatelle, R., Eeckhaut, V., Devreese, M.,
Verlinden, M., Haesebrouck, F., Eeckhout, M., De Saeger, S., Antlinger, B., Novak, B., Martel, A., Van Immerseel, F., 2015. Fumonisins aﬀect the intestinal microbial homeostasis in broiler chickens, predisposing to necrotic enteritis. Vet. Res. 46, 98.
Bhandari, N., Sharma, R.P., 2002. Fumonisin B1-induced alterations in cytokine expres- sion and apoptosis signaling genes in mouse liver and kidney after an acute exposure. ToXicology 172, 81–92.
Biomin, 2011. Biomin’s MycotoXin Survey Report 2011. . http://www.biomin.net.
Bobrovnikova-Marjon, E., Grigoriadou, C., Pytel, D., Zhang, F., Ye, J., Koumenis, C., Cavener, D., Diehl, J.A., 2010. PERK promotes cancer cell proliferation and tumor growth by limiting oXidative DNA damage. Oncogene 29, 3881–3895.
Bouhet, S., Hourcade, E., Loiseau, N., Fikry, A., Martinez, S., Roselli, M., Galtier, P.,
Mengheri, E., Oswald, I.P., 2004. The mycotoXin fumonisin B1 alters the proliferation and the barrier function of porcine intestinal epithelial cells. ToXicol. Sci. 77, 165–171.
Bouhet, S., Le Dorze, E., Peres, S., Fairbrother, J.M., Oswald, I.P., 2006. MycotoXin fu-
monisin B1 selectively down-regulates the basal IL-8 expression in pig intestine: in vivo and in vitro studies. Food Chem. ToXicol. 44, 1768–1773.
Chandrasekhar, S., Sreelakshmi, L., 2012. Formal synthesis of fumonisin B1, a potent sphingolipid biosynthesis inhibitor. Tetrahedron Lett. 53, 3233–3236.
Chen, F., Li, Q., Zhang, Z., Lin, P., Lei, L., Wang, A., Jin, Y., 2015. Endoplasmic reticulum stress cooperates in zearalenone-induced cell death of RAW 264.7 macrophages. Int. J. Mol. Sci. 16, 19780–19795.
Chen, Y.J., Su, J.H., Tsao, C.Y., Hung, C.T., Chao, H.H., Lin, J.J., Liao, M.H., Yang, Z.Y.,
Huang, H.H., Tsai, F.J., Weng, S.H., Wu, Y.J., 2013. Sinulariolide induced hepato- cellular carcinoma apoptosis through activation of mitochondrial-related apoptotic and PERK/eIF2alpha/ATF4/CHOP pathway. Molecules 18, 10146–10161.
Ciacci-Zanella, J.R., Jones, C., 1999. Fumonisin B1, a mycotoXin contaminant of cereal

grains, and inducer of apoptosis via the tumour necrosis factor pathway and caspase activation. Food Chem. ToXicol. 37, 703–712.
Domijan, A.M., Abramov, A.Y., 2011. Fumonisin B1 inhibits mitochondrial respiration and deregulates calcium homeostasis–implication to mechanism of cell toXicity. Int.
J. Biochem. Cell Biol. 43, 897–904.
Escriva, L., Font, G., Manyes, L., 2015. In vivo toXicity studies of fusarium mycotoXins in the last decade: a review. Food Chem. ToXicol. 78, 185–206.
Forus, A., Florenes, V.A., Maelandsmo, G.M., Fodstad, O., Myklebost, O., 1994. The
protooncogene CHOP/GADD153, involved in growth arrest and DNA damage re- sponse, is ampliﬁed in a subset of human sarcomas. Cancer Genet. Cytogenet. 78, 165–171.
Gan, F., Hou, L., Zhou, Y., Liu, Y., Huang, D., Chen, X., Huang, K., 2017. Eﬀects of
ochratoXin A on ER stress, MAPK signalling pathway and autophagy of kidney and spleen in pigs. Environ. ToXicol. 32, 2277–2286.
Grenier, B., Bracarense, A.P., Schwartz, H.E., Trumel, C., Cossalter, A.M., Schatzmayr, G., Kolf-Clauw, M., Moll, W.D., Oswald, I.P., 2012. The low intestinal and hepatic
toXicity of hydrolyzed fumonisin B(1) correlates with its inability to alter the meta- bolism of sphingolipids. Biochem. Pharmacol. 83, 1465–1473.
Hahn, I., Nagl, V., Schwartz-Zimmermann, H.E., Varga, E., Schwarz, C., Slavik, V., Reisinger, N., Malachova, A., Cirlini, M., Generotti, S., Dall’Asta, C., Krska, R., Moll, W.D., Berthiller, F., 2015. Eﬀects of orally administered fumonisin B(1) (FB(1)),
partially hydrolysed FB(1), hydrolysed FB(1) and N-(1-deoXy-D-fructos-1-yl) FB(1) on the sphingolipid metabolism in rats. Food Chem. ToXicol. 76, 11–18.
He, Q., Riley, R.T., Sharma, R.P., 2001. Fumonisin-induced tumor necrosis factor-alpha expression in a porcine kidney cell line is independent of sphingoid base accumula- tion induced by ceramide synthase inhibition. ToXicol. Appl. Pharmacol. 174, 69–77.
Hojjati, M.R., Li, Z., Zhou, H., Tang, S., Huan, C., Ooi, E., Lu, S., Jiang, X.C., 2005. Eﬀect
of myriocin on plasma sphingolipid metabolism and atherosclerosis in apoE-deﬁcient mice. J. Biol. Chem. 280, 10284–10289.
Jaeschke, H., Duan, L., Akakpo, J.Y., Farhood, A., Ramachandran, A., 2018. The role of apoptosis in acetaminophen hepatotoXicity. Food Chem. ToXicol. 118, 709–718.
Janse van Rensburg, B., McLaren, N.W., Flett, B.C., 2017. Grain colonization by fumo-
nisin-producing Fusarium spp. and fumonisin synthesis in South African commercial maize in relation to prevailing weather conditions. Crop Protect. 102, 129–136.
Kim, I.Y., Shim, M.J., Lee, D.M., Lee, A.R., Kim, M.A., Yoon, M.J., Kwon, M.R., Lee, H.I.,
Seo, M.J., Choi, Y.W., Choi, K.S., 2019. Loperamide overcomes the resistance of colon cancer cells to bortezomib by inducing CHOP-mediated paraptosis-like cell death. Biochem. Pharmacol. 162, 41–54.
Kim, M.S., Lee, D.Y., Wang, T., Schroeder, J.J., 2001. Fumonisin B(1) induces apoptosis in
LLC-PK(1) renal epithelial cells via a sphinganine- and calmodulin-dependent pathway. ToXicol. Appl. Pharmacol. 176, 118–126.
Kimanya, M.E., 2015. The health impacts of mycotoXins in the eastern Africa region. Curr.
Opin. Food Sci. 6, 7–11.
Lalles, J.P., Lessard, M., Boudry, G., 2009. Intestinal barrier function is modulated by short-term exposure to fumonisin B(1) in Ussing chambers. Vet. Res. Commun. 33, 1039–1043.
Lalles, J.P., Lessard, M., Oswald, I.P., David, J.C., 2010. Consumption of fumonisin B1 for
9 days induces stress proteins along the gastrointestinal tract of pigs. ToXicon 55, 244–249.
Luo, J., Xia, Y., Luo, J., Li, J., Zhang, C., Zhang, H., Ma, T., Yang, L., Kong, L., 2017. GRP78 inhibition enhances ATF4-induced cell death by the deubiquitination and stabilization of CHOP in human osteosarcoma. Cancer Lett. 410, 112–123.
Marciniak, S.J., Yun, C.Y., Oyadomari, S., Novoa, I., Zhang, Y.H., Jungreis, R., Nagata, K.,
Harding, H.P., Ron, D., 2004. CHOP induces death by promoting protein synthesis and oXidation in the stressed endoplasmic reticulum. Genes Dev. 18, 3066–3077.
Masuda, M., Miyazaki‐Anzai, S., Levi, M., Ting, T.C., Miyazaki, M., 2013.
PERK‐eIF2α‐ATF4‐CHOP signaling contributes to TNFα‐induced vascular calciﬁca- tion. J. Am. Heart Assoc. 2.
Muller, S., Dekant, W., Mally, A., 2012. Fumonisin B1 and the kidney: modes of action for renal tumor formation by fumonisin B1 in rodents. Food Chem. ToXicol. 50, 3833–3846.
Ormancey, M., Thuleau, P., van der Hoorn, R.A.L., Grat, S., Testard, A., Kamal, K.Y.,
Boudsocq, M., Cotelle, V., Mazars, C., 2019. Sphingolipid-induced cell death in Arabidopsis is negatively regulated by the papain-like cysteine protease RD21. Plant Sci. 280, 12–17.
Oyadomari, S., Koizumi, A., Takeda, K., Gotoh, T., Akira, S., Araki, E., Mori, M., 2002.
Targeted disruption of the Chop gene delays endoplasmic reticulum stress-mediated diabetes. J. Clin. Investig. 109, 525–532.
Park, H.J., Kim, H.J., Park, H.J., Ra, J., Zheng, L.T., Yim, S.V., Chung, J.H., 2008.
Protective eﬀect of topiramate on kainic acid-induced cell death in mice hippo- campus. Epilepsia 49, 163–167.
Qin, X., Zhang, R.X., Ge, S., Zhou, T., Liang, Y.K., 2017. Sphingosine kinase AtSPHK1 functions in fumonisin B1-triggered cell death in Arabidopsis. Plant Physiol. Biochem. 119, 70–80.
Riley, R.T., Voss, K.A., Coulombe, R.A., Pestka, J.J., Williams, D.E., 2011. Developing
Mechanism-Based and EXposure Biomarkers for MycotoXins in Animals. pp. 245–275.
Sharma, N., He, Q., Sharma, R.P., 2006. Amelioration of fumonisin B1 hepatotoXicity in mice by depletion of T cells with anti-Thy-1.2. ToXicology 223, 191–201.
Shi, H., Li, S., Bai, Y., Prates, L.L., Lei, Y., Yu, P., 2018. MycotoXin contamination of food
and feed in China: occurrence, detection techniques, toXicological eﬀects and ad- vances in mitigation technologies. Food Control 91, 202–215.
Silva, J.J., Viaro, H.P., Ferranti, L.S., Oliveira, A.L.M., Ferreira, J.M., Ruas, C.F., Ono, E.Y.S., Fungaro, M.H.P., 2017. Genetic structure of Fusarium verticillioides popula- tions and occurrence of fumonisins in maize grown in Southern Brazil. Crop Protect.
99, 160–167.
Singh, M.P., Kang, S.C., 2017. Endoplasmic reticulum stress-mediated autophagyactivation attenuates fumonisin B1 induced hepatotoXicity in vitro and in vivo. Food Chem. ToXicol. 110, 371–382.

Soriano, J.M., Gonzalez, L., Catala, A.I., 2005. Mechanism of action of sphingolipids and their metabolites in the toXicity of fumonisin B1. Prog. Lipid Res. 44, 345–356.
van der Westhuizen, L., Gelderblom, W.C., Shephard, G.S., Swanevelder, S., 2004.
Disruption of sphingolipid biosynthesis in hepatocyte nodules: selective proliferative stimulus induced by fumonisin B1. ToXicology 200, 69–75.
Varela-Nieto, I., Palmero, I., Magarinos, M., 2019. Complementary and distinct roles of autophagy, apoptosis and senescence during early inner ear development. Hear. Res. 376, 86–96.
Wang, C., Guan, Y., Yang, J., 2010. Cytokines in the progression of pancreatic beta-cell
dysfunction. Internet J. Endocrinol. 2010, 515136.
Wang, J., Zhou, Y., Liu, W., Zhu, X., Du, L., Wang, Q., 2008. Fumonisin level in corn-based food and feed from Linxian County, a high-risk area for esophageal cancer in China. Food Chem. 106, 241–246.

Waśkiewicz, A., Beszterda, M., Goliński, P., 2012. Occurrence of fumonisins in food – an interdisciplinary approach to the problem. Food Control 26, 491–499.
Yin, S., Guo, X., Li, J., Fan, L., Hu, H., 2015. Fumonisin B1 induces autophagic cell death via activation of ERN1-MAPK8/9/10 pathway in monkey kidney MARC-145 cells. Arch. ToXicol. 90, 985–996.
Yu, L., Li, B., Zhang, M., Jin, Z., Duan, W., Zhao, G., Yang, Y., Liu, Z., Chen, W., Wang, S.,
Yang, J., Yi, D., Liu, J., Yu, S., 2016. Melatonin reduces PERK-eIF2alpha-ATF4- mediated endoplasmic reticulum stress during myocardial ischemia-reperfusion in- jury: role of RISK and SAFE pathways interaction. Apoptosis 21, 809–824.
Zhang, H., Diao, X., Li, N., Liu, C., 2018a. FB1-induced programmed cell death in he-
mocytes of Ostrinia furnacalis. ToXicon 146, 114–119.
Zhang, W., Zhang, S., Zhang, M., Yang, L., Cheng, B., Li, J., Shan, A., 2018b. Individual and combined eﬀects of Fusarium toXins on 4-Phenylbutyric apoptosis in PK15 cells and the protective role of N -acetylcysteine. Food Chem. ToXicol. 111, 27–43.