Bisphenol S Leads to Cytotoxicity-Induced Antioxidant Responses and Oxidative Stress in Isolated Rainbow Trout (Oncorhyncus Mykiss) Hepatocytes

Background: Bisphenol S (BPS) is a chemical compound that is utilized in the plastic industry as an alternative to bisphenol A (BPA). The toxic effects of BPS in sh is less known and limited. Therefore, in the present study, it was aimed to investigate the inuence of BPS on rainbow trout (Oncorhyncus mykiss) hepatocytes in vitro. Methods and Results: For this purpose, the hepatocytes of the sh were isolated, and then the cultured cells were treated with increasing concentrations of BPS (0, 15.63, 31.25, 62.50, 125, 250, and 500 µM) for 24 h. The cytotoxic impact of BPS was determined in the culture media using lactate dehydrogenase assay and then, the antioxidant defence indicators were assayed. The results showed that concentration-dependent increases were observed in the percentage of cytotoxicity. The superoxide dismutase activity was reduced, while the catalase and glutathione peroxidase activity was elevated with all of the BPS concentrations. The glutathione S-transferase (GST) activity was signicantly increased with a BPS concentration of 31.25 µM or higher, while GST theta 1-1 activity was decreased with the same concentrations of BPS. The reduced glutathione content was decreased signicantly with a BPS concentration of 31.25 µM or higher, and the malondialdehyde content increased with BPS concentrations of 125, 250, and 500 µM. Conclusions: The ndings determined herein suggested that BPS causes cytotoxicity in sh hepatocytes and could lead to oxidative stress, resulting hepatotoxicity in sh. Thus, the utilization of BPS instead of BPA as safe alternative in industry should be re-evaluated in the future for environmental health.


Introduction
It is well-known that bisphenol A (BPA) is a an endocrine-disrupting chemical, as well as a suspected carcinogen that is used in the production of plastic materials, food packaging, medical equipment, dental sealants, baby bottles, adhesives, ame retardants, toys, and thermal receipts [1,2]. Due to its potential hazardous impacts in humans and risks to wildlife, scienti c milieu and regulators have raised concerns regarding use of BPA throughout the world [3]. Consequently, efforts toward restriction and legislation for usage of BPA have been taken by the European Union and United States [4,5] and safer alternatives have been sought instead of BPA in industrial applications. Thus, manufacturers have been prompted to replace BPA with its structural derivatives, such as bisphenol S (BPS), so as to comply with those regulations [6,7]. BPS has been frequently utilized as an electroplating solvent, wash-fastening agent, and component of phenolic resin [8]. This analog is commonly used in daily products, such as epoxy glues, thermal receipts, canned food stuffs, paper currencies, luggage tags, food cartons, and baby bottles [9,10,11]. Products in which BPS has been used as a component, such as thermal papers and plastics, are labelled as 'BPA-free' in marketing [12]. Environmental monitoring studies have reported the presence of BPS in indoor dust, foodstuff, sediments, surface waters, and sewage sludges in different countries [6, 13,14]. Average concentrations of BPS in the rivers of Japan, Korea, China, and India were reported to be detected from undermined levels, up to 15, 42, 135, and 7200 ng/L, respectively [14]. The estimated daily dietary intake (ng/kg body weight/day) of BPS for infants, toddlers, children, teenagers, and adults, were averagely determined to be 1.72, 4.34, 2.49, 1.60, and 1.31 ng/kg body weight, respectively, in the United States [13]. In the human urine samples that were examined in various countries around the world, BPS concentrations of 1.18 ng/mL, 0.933 µg/g creatinine (Cre) were determined in Japan, as well as 0.299 ng/mL, 0.304 µg/g Cre in the United States, 0.226 ng/mL, 0.223 µg/g Cre in China, 0.172 ng/mL, 0.126 µg/g Cre in Kuwait, and 0.160 ng/mL, 0.148 µg/g Cre in Vietnam.
As inhabitants of aquatic environments, sh could be inevitably exposed to BPA and its analogs via their surrounding milieu. In a study that was recently, BPA was detected in the liver and muscle of Trachurus trachurus, Dicentrarchus labrax, and Scomber colias from the North East Atlantic Ocean, as well as bisphenol B and bisphenol E in the muscle [15]. Experimental studies over the last decade have shown that exposure to BPS causes developmental abnormalities, reproductive impairment, and hormonal imbalance in sh. The exposure of adult zebra sh pairs to concentrations of 0.5 µg/L to 50 µg/L of BPS has been reported to cause reduced egg production, increased 17β-estradiol levels in both sexes, decreased testosterone levels accompanied by up-regulated cyp19 and down-regulated cyp17 and 17βhsd transcripts, as a consequence of being in uenced negatively of hypothalamic-pituitary-gonad axis, and a delayed rate of hatching. Moreover, hatchability success and malformations in F1 generation due to the impairment of development have also been reported [16]. The treatment of zebra sh (Danio rerio) with BPS resulted in a skewed sex ratio towards the side of females, low testosterone levels in males, signi cant escalations in plasma vitellogenin, decreased levels of plasma thyroxine and triiodothyronine, decreases in egg production and sperm count, and lower hatching rate [17]. Apart from these studies in sh, it was also documented that BPS depresses hematological functions and promotes cardiovascular function in Sprague-Dawley rats. BPS has been also reported to possess estrogenic, androgenic, and anti-androgenic activity, and genotoxic impacts [10,18,19,20,21].
The characterization of oxidative stress includes an imbalance between anti-oxidants and pro-oxidants that can be stimulated by xeno-estrogens and antioxidant system indicators, which are useful tools for the determination of harmful effects of a speci c pollutant [22]. In a previous study, BPS was found to cause signi cant increases in intracellular reactive oxygen species (ROS) levels and glutathione Stransferase (GST) activity in the marine rotifer, Brachianus koreanus. Qiu et al. [23] reported that the exposure of sh (Cyprinus carpio) primary macrophages to low concentrations of BPS resulted in the generation of nitric oxide and ROS levels, increased total antioxidant capacity, and lipid peroxidation, as well as impaired phagocytic activity and the immune response involved in estrogen receptors. However, very scarce information regarding BPS cytotoxicity in sh is available in the literature data.
Fish cells are inexpensive and suitable tools that can be used for testing chemicals in toxicological studies [24]. Therefore, in this study, the isolated and primary cultured hepatocytes of Oncorhyncus mykiss were used, as a result of it being a test animal that was recommended for use in toxicological studies by the Organization for Economic Co-operation and Development [25], to shed light on the harmful effects of BPS at a cellular level.

Material And Method
Fish Five juvenile rainbow trout (Oncorhyncus mykiss) (fork length: 20− 21.5 cm; total weight: 106−136 g) were used in the study. The sh were obtained from a locally owned rainbow trout farm that was located in Van Province, Turkey.

Preparation Of Isolated Hepatocytes
The process of hepatocyte isolation from the sh was performed using the method given tin the study of Mortensen et al. (2006) [26], with some modi cations. First, the livers were aseptically dissected from the sh and put into Petri dishes that contained cold Ca 2+ -free solution-I (7.14 g/L of NaCl, 0.36 g/L of KCl, 0.15 g/L of MgSO 4 , 1.6 g/L of Na 2 HPO 4 , 0.4 g/L of NaH 2 HPO 4 , 0.31 g/L of NaHCO 3 , and 20 mg/L of EGTA). Next, the samples of liver tissue were minced while still in this solution, and then all of the blood was removed, as well as any remaining rude components of tissue, such as connective tissue and vessels. As a next step, the liver was dissociated mechanically into even smaller pieces via the use of surgical blades and ne forceps. Next, pieces of tissue, which were now whitened, were transferred into solution-II, which possessed of the same compounds that were included in solution-I, with the addition of 0.11 g/L of CaCl 2 rather than EGTA, as well as 0.025 mg/mL of type IV collagenase, which was used to break the Ca 2+ -dependent cell-cell connections and enzymatically dissociate the cells. After the pieces of tissue had been softened in solution-II for 10 min, they were pulverized using pipette tips of varying sizes.
Next, the crude particles were then removed via the use of a stainless steel sieve and the cell suspension that it yielded was very carefully passed through a sterile injector needle to further dislocate the cells. Then, the suspension was placed into Eppendorf tubes, which were then centrifuged at 60 × g for 3 min. The cell pellet was then suspended in Leibovitz (L-15) medium containing 0.38 g/L of NaHCO3 and 1% (v/v) antibiotic-antimycotic. After it had been washed for the second time with L-15, the cells were resuspended in medium. After this, the cells were counted by using a Thoma slide, and the cell viability was assessed using the Trypan blue exclusion test. In the examination of the cell culture, the cells were observed to determine if they possessed > 90% viability.

Hepatocyte Culture And Bps Treatment
The isolated hepatocytes were seeded into 24-well culture plates that were coated with collagen-I (Gibco,  15.63, 31.25, 62.50, 125, 250, and 500 µM concentrations of BPS, which were then applied to the cells for a period of 24 h. Care was taken to ensure that the nal ethanol concentration in the media was not greater than 0.5%. In the control group, the cells were treated with L-15 medium that only contained 0.5% absolute ethanol. Five replicate wells were constructed for the experimental groups. A Leica DMI 6100B inverted microscope (Leica Microsystems GmbH, Wetzlar, Germany) was used to check the cell cultures.

Lactate Dehydrogenase Cytotoxicity Test
The culture media was examined to determine the amount of lactate dehydrogenase (LDH) leakage that had occurred, and this measurement was then used in the determination of the cytotoxic effects that resulted from the BPF following treatment for 24 h. After the treatments had been completed, removal of the culture media from the wells was performed, which was then transferred into Eppendorf tubes. Following this, 10 µL of the samples was put into a 96-well plate and an assay kit to determine LDH cytotoxicity (Catalog no: ab65393, Abcam, Cambridge, UK) was then used to conduct an assay of the LDH activity, following the manufacturer's instructions. As a nal step, a DAS A3 ELISA plate reader (DAS Rome, Italy) was used in the measurement of the optic density (OD) values, which was performed at 450 nm. In each of the experiments, 5 replicates were conducted. All measurements conducted were in duplicate, and calculation of the percentage of cytotoxicity was done using the formula that was provided in the kit procedures for each sample, as given in Eq. (1): 1

Measurement Of Antioxidant Defense Indicators
After the exposures had been completed, the culture media very carefully pipetted out of the wells and the hepatocytes were then rinsed in 1 mL of ice-cold phosphate buffered saline that had a pH of 7.4. Then, the obtained 2 × 10 − 6 / mL cell suspensions were transferred into Eppendorf tubes, which were then lysed in a glass-porcelain ultrasonic homogenizer (Jencons Scienti c Co., Herts, UK). Next, the homogenate was then centrifuged at 15,000 × g for 15 min. All of these procedures were either conducted at 4°C or on ice. As the next step, the supernatant fractions were taken out and then used to determine the antioxidant defense indicators. In each of the experiments, 5 replicates were conducted.
Spectrophotometric measurement of the SOD activity was conducted with a commercial kit (Ransod, Randox Lab., Crumlin, County Antrim, UK) by following the manufacturer's instructions. The method listed in the kit, which followed that given by Suttle and McMurray [27], consisted of an xanthine-xanthine oxidase system that was utilized in the generation of superoxide radicals, which reacted with 2-(4- iodophenyl)-3-(4-nitrophenol)-5-phenyltetrazolium chloride (INT) and resulted in the formation of red formazan dye. Measurement of the SOD activity was then conducted using the degree of inhibition determined from this reaction. It was determined that 1 unit of SOD was the amount to result in 50% inhibition of the reduction rate of INT under the conditions that were given in the assay. All of the measurements were conducted at 37°C and 505 nm, and the activity was expressed as unit/mg protein.
Spectrophotometric measurement of the catalase (CAT) activity was performed using the method given by Aebi [28]. The CAT activity was determined via an assay of the decrease of the consumption of H 2 O 2 at 240 nm. CAT activity was expressed as nmole of H 2 O 2 consumed/min/gram of protein.
Measurement of the glutathione peroxidase (GPx) activity was performed using a commercial obtained Ransel kit (Randox Laboratories Ltd., Co. Antrim, UK) 37°C and 340 nm, following the manufacturer's instructions. This assay followed the method given by Paglia and Valentine [29]. The oxidation of glutathione is catalyzes by GPx via cumene hydroperoxide. When glutathione reductase and NADPH are present, the glutathione that has been oxidized (GSSG) is converted immediately into its reduced form, with the concomitant oxidation of NADPH to NADP+. Calculation of the GPx activity was performed from the decreases observed in the absorbance values. Expression of the GPx activity was given as units per grams of protein.
Determination of glutathione S-transferase (GST) activity was performed in line with the method given by Habig et al. [30]. Optimization was performed for this using the ELISA Microplate Reader System, which was used on a Multiskan FC microplate photometer (Thermo Fisher Scienti c Inc., Waltham, MA, USA) (Yilmaz and Işcan, 2014) [31]. The contents of the reaction medium for the total GST activity measurements were 0.1 M of phosphate buffer at a pH of 7.4, 1 mM of 1-chloro-2,4-dinitrobenzene, and 1 mM of GSH. The reaction was started through the addition of 25 µL of the sample, which comprised 33.28-73.4 µg per well, and the absorbance data were obtained at 25°C, for 10 min, at 340 nm The contents of the reaction medium for the GSTT1-1 isozyme activity measurements were 0.1 M of phosphate buffer at a pH of 6.5, 0.25 mM of 1,2 -epoxy-3-(p-nitrophenoxy)-propane, and 0.5 mM of GSH. The reaction was started through the addition of 100 µL of the sample, which comprised 133.2-293.6 µg per well, and the absorbance data were obtained at 360 nm, for 10 min, at 25°C [32]. Speci c activity was calculated by using Eq. (2): 2 Here, ΔA/Δt is the absorbance change per minute, ε is the extinction coe cient of the substrates (9.6 mM − 1 cm − 1 for CDNB and 0.5 mM − 1 cm − 1 for EPNP), and DF is the dilution factor.
Determination of the total thiol content was performed following the method that was described by Sedlak and Lindsay [33], which had been optimized for the ELISA Microplate Reader System. The standard curve was built with the measurement results of the GSH standards in the concentration range of 0.1-1 mM. Next, 50 µL of sample, comprising 66.6-146.8 µg/well, was added to 30 µL of 0.2 M Tris buffer at a pH of 8.2. Right after that, 20 µL of 2 mM 5,5′-Dithiobis-(2-Nitrobenzoic acid) (DTNB) and 100 µL of methyl alcohol were added into each of the wells. The plate was then incubated, under dark conditions, for 30 min at 25°C. The absorbance was measured at 405 nm. Calculation of the total amount of thiol within the samples was performed by using the slope value of the standard curve, the results of which were expressed as nmole/mg of protein.
The content of malondialdehyde (MDA), which is a product of lipid peroxidation, was spectrophotometrically measured at 532 nm, following the method of Beuge and Aust [34]. Brie y, 0.1 mL of 150 mM Tris-HCl buffer, at a pH of 7.1, was added on the supernatant (0.1 mL). Next, 0.1 mL of 1.5 mM ascorbic acid and 0.1 mL 1 mM FeSO 4 were added to this mixture. The total volume of this mixture was completed to 1 mL with addition of distilled water and then the tubes were incubated for 15 min at 37°C. As a next step, 1 mL of trichloroacetic acid (10%) and 2 mL of thiobarbituric acid (0.375%) were added to the mixture, and the tubes were incubated in boiling hot water for 15 min. After the tubes had been cooled, centrifugation was performed at 3000 rpm for 10 min. Spectrophotometric measurements were carried out at 505 nm. Calculation of the concentration of MDA in samples was performed using the standard curve that was derived from external 1,1,3,3-tetraethoxypropane standards. The results were expressed as nmole/mg of protein.
Spectrophotometric assay of the total protein content within the supernatant fractions was performed using the method that was given by Bradford [35], using bovine serum albumin as the standard.

Statistical Analyses
All of the statistical analyses were conducted using IBM SPSS Statistics for Windows 20.0 (IBM Corp., Armonk, NY, USA). One-way ANOVA and the Duncan multiple comparison post-hoc test were used to analyze differences among the groups. The results were expressed as the mean ± the standard error of the mean (SEM). P < 0.05 was considered statistically signi cant.

Results
After the experiments had been completed, it was observed in the microscopic examination of the cells that the hepatocytes in the control group exhibited a healthy appearance (Fig. 1a). The hepatocytes that has been treated exhibited degenerative changes, such as apoptotic gures or shrinkage, and they had did not have as healthy of an appearance as the cells in the control group (Fig. 1b).
As a result of the treatment that were applied in this study, it was observed that all of the BPF concentrations resulted in a signi cant increase LDH leakage into the media. The percentage of cellular cytotoxicity increased gradually with the increasing BPS concentrations (P < 0.05). It was observed that cytotoxicity was occurred with BPS concentrations of 1.  2). All of the increased levels of cytotoxicity were determined to be statistically signi cant, except for the 15.63 µM concentration of BPS The hepatocyte SOD activity exhibited a signi cant decrease (P < 0.05; Fig. 3a), whereas the CAT activity was observed to signi cantly increase (P < 0.05; Fig. 3b). The GPx activity was signi cantly increased with all of the BPS concentrations ( Fig. 4a; P < 0.05). The total GST speci c activity was also detected to be elevated (P < 0.05) by the increasing amount of BPS in the growth media. On the contrary, speci c activity of GSTT1-1 isozyme in the samples, except the one that had the lowest BPS concentration, of 15.63 µM, were all lower than (P < 0.05) those that were in the control group (Figs. 4b,c). An alteration similar to the change in the GSTT1-1 isozyme was observed in the amount of total thiol groups considered as an indicator of the GSH pool of the cell. It was determined that the GSH pool shrank (P < 0.05) in all treatment groups in all of the experimental groups, relative to the control group, except for the one that was treated with the lowest BPS concentration (Fig. 5a). The MDA content remained unchanged with 15.63, 31.25, and 62.50 µM concentrations of BPS, while it displayed statistically signi cantly increases with 125, 250, and 500 µM concentrations of BPS (Fig. 5b).

Discussion
The percentage of hepatocyte cytotoxicity was gradually increased after the BPS treatment according to the LDH cytotoxicity test. SOD and CAT constitute the rst barrier for the elimination of superoxide radicals, as well as hydrogen peroxide, in the antioxidant system. As a free radical scavenging enzyme, SOD is responsible for the dismutation of highly-reactive and toxic superoxide anions into molecular oxygen and hydrogen peroxide.
The current results showed that all of the concentration of BPS led to general decreases in the SOD activity. In accordance with the results determined herein, lower SOD activity were determined in the reproductive tissues of male rat offspring that had been exposed to BPA and its analogs including BPS. The levels of SOD activity were also reduced in human red blood cells treated with BPA, BPAF, and BPF for 4 and 24 h, whereas BPS did not create signi cant changes in the SOD activity [40]. Kose et al. [37] also reported diminished SOD activity in RWPE-1 cells that had been incubated with BPA and BPF for 24 h, except for BPS. The differences between the studies may have arisen from the susceptibility of the cell types to BPS or the severity of the oxidative stress by BPS. A probable reason for deceased SOD activity in this study might have arisen from the excessive production of superoxide anions as a result BPS exposure, as the substrate of the enzyme might behaved as an oxidant [41]. SOD is susceptible to oxidation and it was demonstrated that the enzyme was inactivated by hydrogen peroxide [42]. Dimitrova et al. [43] reported that superoxide radicals or the generated hydrogen peroxide via superoxide radical transformation might have caused oxidation of the cysteine in the enzyme that resulted in decreased SOD activity. Thus, increased levels of superoxide anions or excessive hydrogen peroxide levels in the hepatocytes after BPS exposure might have caused such an effect in the current study. CAT, which is an essential antioxidant enzyme for cells, degrades hydrogen peroxide into H 2 O and oxygen. The generating hydrogen peroxide induces the activity of the enzyme. It was reported that CAT activity was elevated in mouse liver and renal cells after incubation for 12 h with BPS depending on the ROS production in those cells. In addition, BPS is capable of interacting with the enzyme via binding to the Gly 117 residue on the substrate channel, thus effecting hydrogen bonding networks [43]. Supporting those ndings, the results determined herein showed that the exposure of rainbow trout hepatocytes to BPS in uenced the CAT activity. Elevated CAT activity in the cells suggested that the enzyme was capable of eliminating of hydrogen peroxide, even in the presence of the decreased SOD activity [45]. The removal of toxic hydrogen peroxide and a wide variety of organic peroxides, which are produced as a consequence of oxygen metabolism, is carried out by the selenium-dependent antioxidant enzyme GPx. The enzyme acts as ROS scavenger in cells and uses GSH as a substrate during those processes. Exposure of human red blood cells to a BPS concentration of 500 µg/L for 4 h caused a signi cant decrease in the GPx activity, while no change was observed at the end of 24 h with the same concentration [40]. In another study, a signi cant reduction in the GPx activity was reported in RWPE-1 cells that were exposed to a BPS concentration of 108 µM for 24 h [37]. These reports indicated that the sensitivity or compensation role of GPx against BPS cytotoxicity might have changed with treatment using different concentrations and time intervals according to the cell type. On the other hand, in this study, signi cant increases in the GPx activity were determined at all of exposure concentrations of BPS. The ndings herein were more likely to be related with the production of hydrogen peroxide or other organic peroxides, or with the protective role of the enzyme against cellular membrane damage. Supporting the current study results, similar ndings were also determined following 24-h treatment with BPF in rainbow trout hepatocytes [46].
GSH is a part of the second-line of antioxidant defense. This ubiquitous tripeptide (L-glutamine, cysteine, and glycine), which is usually the most prevalent intracellular thiol, can react directly or indirectly with ROS and during the metabolic free radical scavenging role of GSH, the sulfhydryl group of the molecule transforms into an oxidized GSH (GSSG) disul de compound [47]. Both increased and decreased levels of GSH in cells are considered to be indicators of oxidative stress, of which increases occur under mild oxidative stress, depending on the compensation function of the molecule against free radicals, as a result of adaptive mechanisms including its synthesis, while its level could be decreased under severe oxidative stress because of disturbed adaptation mechanisms [48]. In the research conducted herein, GSH content was decreased with a BPS concentration of 125 µM or higher, which suggested the presence of oxidative stress. In parallel with those ndings, signi cant reductions in the GSH content was reported in zebra sh (Danio rerio) embryos after a short time exposure to BPA. Maćczak et al. [40] reported that BPA and its analogs, including BPAF and BPF, except for BPS, induced the strongest decreases in human red blood cells. However, elevated GSH levels were found in RWPE-1 cells that were exposed to BPS [37]. GSH is used by GST and GPx as a conjugating molecule, which aids in the facilitation of the hazardous effects of xenobiotics [49]. Decreased GSH content may result from an increase in the use of GSH by both GST and GPx, as well as the insu cient synthesis of GSH by glutathione reductase [45,[50][51][52]. In parallel with the literature data, concomitant increases were observed in the GST and GPx activity in the hepatocytes with reduced GSH levels in this study. Thus, it can be concluded that GSH was decreased as a consequence of the disturbance of its synthesis due to its failure to have an adaptive mechanism or increase in the molecule by the enzymes GST and GPx.
The up-regulation of some genes, which have vital functions in drug/xenobiotic metabolism and antioxidant defense system, such as UDP-glucuronosyltransferase, CAT and GST, were reported in the liver of marine sh Paralichthys olivaceus after exposure to BPS [53]. Moreover, 12-24 h of incubation with BPA caused the upregulation of GSTM in another marine sh, Kryptolebias marmoratus [54]. Similar results were also reported as elevated GST speci c activity in the liver tissues of freshwater species like Pimephales promelas (Zare et al. 2018) [55] and Oryzias latipes [56]. In the current study, BPS treatment of Oncorhyncus mykiss hepatocyte culture unveiled the very same response of elevated total GST speci c activity, demonstrating the practicality of the usage of this parameter in the testing of damage of xenobiotic exposure to aquatic animals. There have been such dose-responsive in vivo studies conducted to reveal the effects of BPS on marine organisms; however, the number of studies based on the application of this chemical directly onto hepatocyte cultures is very limited. The study of the function and nature of GSTs in the liver tissues of aquatic organisms in a cause and effect relation could enable the determination of the dimension of the problem created by BPA and its derivatives, like BPS.
GST theta is known as the rst isozyme evolved among others, which consists of 2 different types: GSTT1 and GSTT2, sharing 55% protein sequence identity, and, to date, it has been subjected to characterization in several aquatic organisms, such as Macrobrachium rosenbergii, Ruditapes philippinarum, and Apostichopus japonicus [57]. Although its major sites of expression are the gills and epithelium in the olfactory and digestive systems of rainbow trout [58], weak but debatable speci c activity of GSTT1-1 was measured in the liver hepatocyte culture for the rst time in this current study. GST theta types have been considered to play a role in xenobiotic biodegradation, even though their substrate speci cities are different from each other [59], which might explain the decrease in the measured speci c activity with the increasing BPS concentration in the media, while the total GST activity was scaled up in the treatment groups.
Oxidants, including radicals and ROS, may attack carbon-carbon double-bond-containing lipids, such as polyunsaturated fatty acids in biological membranes. During the oxidation of unsaturated fatty acids, a range of changes, including hydrogen removal from a carbon and the insertion of oxygen instead of it, occurs in target lipids involving glycolipids, phospholipids, and cholesterol. Finally, lipid peroxyl radicals as well as hydroperoxides cause an impairment in membrane function and leads to apoptosis in cells [60]. Among lipid peroxidation products, MDA is the most mutagenic and it has been widely used as a biomarker, which re ects indirectly the lipid peroxidation levels following free radical attacks in cells [ Dawley rats that were exposed to BPA, and its analogs, with concomitant increases in the ROS levels, indicating that the accumulation of ROS could attack unsaturated fatty acids, and nally, lead to lipid peroxidation as the result of BPS. On the other hand, Maćczak et al. [40] stated that BPA and its analogs, BPAF, BPF, and BPS, induced ROS levels in red blood cells, whereas BPS did not cause increased lipid peroxidation, as was induced by BPA and the other analogs. Russo et al. [39] reported that the cytotoxicity of bisphenol analogs was slightly related with phospholipophilicity in different cell lines. Thus, it can be concluded that signi cant increases in the MDA content observed in this study might have arisen from the elevated levels of ROS, caused by BPS, which lead to accelerated lipid peroxidation.
In summary, the present research is the rst report to demonstrate toxic effects of BPS on isolated rainbow trout hepatocytes. BPS, used as a substitute instead of BPA as an alternative in the food packaging or other application areas of industry, changed the levels of antioxidant defense enzymes, namely SOD, CAT, GPx, GST, and GSTT1, decreased the GSH content, and caused lipid peroxidation in the cells. The results provided evidences that BPS is not a safer alternative or innocent agent that can be used as an alternative to BPA. Authorities should reconsider the utilization of BPS in the industry and take measures to prevent environmental contamination of this compound, which may affect the antioxidant system in organisms.

Declarations Author Contributions
This research was conceived by BK. The isolation and culturing of the hepatocytes, and the application of the treatments were performed by BK, HA, ED, MB and FY. The cytotoxicity testing was performed by BK, HA, ED, and FY. The antioxidant defense system indicators were measured by BK, CY, HA, ED, CF, FY, and MB. Analyzing of the data and interpreting the results were performed by BK and CY. The drafting and editing of the manuscript were performed by BK and CY. The manuscript was nalized by BK.

Funding
The authors declare that no nancial support was received for either the research or the writing of this article.

Compliance with Ethical Standards
Ethical approval The procedures that were conducted within this research were all performed in line with the procedures set forth by the National and Institutional Regulations for the Protection of Animal Con ict of Interest The authors declare that they have no con icts of interest with regards to this research.

Data availability
All of the data and the material that were used and analyzed in the course of the study herein can be obtained from the author for correspondence upon reasonable request.

Figure 1
Photomicrographs showing the rainbow trout cultured hepatocytes. a healthy control cells clustering themselves into groups. b Cells that were exposed to a BPF concentration of 500 µM for 24 h, which display degenerative changes (shown with arrows).    The different letters shown indicate the statistically signi cant differences that were observed between the treatments.