Two Faces of Arbutine in Hepatocelluler Carcinoma (HepG2) Cells: Anticarcinogenic Effect in LD50 Dose and Protective Effect Against Cisplatin Toxication Through its Antioxidant and Anti-inammatory Activity in LD0 Dose

Arbutine is one of the active substances used as a skin whitening agent in cosmetic products. Possible effects of arbutine on hepatocelluler carcinoma (HepG2) cells and cisplatin toxication in HepG2 cells were investigated in this study. Cytotoxicity, genotoxicity, oxidative stress, inammation, apoptosis and proliferation levels were determined in experimental groups established for the purpose above. It was determined that when low dose of α-arbutine (in LD 0 dose) was administered to HepG2 cells alone, it had no genotoxic and cytotoxic effects and no effects on inammation, apoptosis and proliferation. However, when low dose of arbutine was used with cisplatin, it was observed that oxidative stress, inammation, and genotoxicity levels increased as a result of cisplatin toxicity, but caspase 3 levels were not affected by this situation. As a result of high dose (in LD 50 dose) of α-arbutine administration to HepG2 cells, it was determined that it would have anticarcinogenic effects by increasing oxidative stress, genotoxicity, inammation and apoptosis and by suppressing proliferation. In the presented study it was determined that as well as α-arbutine had an anticarcinogenic effect on HepG2 cells in high doses, it might be protective to reduce the side effects caused by low dose (LD 0 dose) of cisplatin treatment. In addition, it was concluded that α/β-arbutine-including cosmetic products were safe for cancer patients because α/β-arbutine had no effect on proliferation in HepG2 cells. In order to present the activity of arbutine isoforms more clearly it is recommended that studies should be conducted using healthy and different cell lines.


Introduction
Cancer is among the global problems affecting public health and economy. According to GLOBOCAN database data prepared by International Agency for Research on Cancer (IARC) using the data from 185 countries, 18.1 million people in the world were diagnosed as cancer and 9.6 million died of cancer in 2018. If cancer types are listed in terms of incidence, lung (11,6%), breast (11,6%), colon (10.2%), prostate (7,1%), stomach (5.7%) cancers can be uttered in the top ve. Liver cancer is the 6th most common cancer type with its 4.7% of incidence rate [1].
From past to present Cisplatin [cis-diamminedichloroplatin (II)] has been one the most commonly used chemotherapeutics whose biological activity was rst reported in 1965. Cisplatin is among the effective chemotherapeutics that can be also used for the treatment of hepatic cancer. Cisplatin is also effective against various types of cancer including carcinomas, germ cell tumors, lymphomas and sarcomas.
However, side effects of cisplatin (nephropathy, allergic reactions, decreased immunity to infections, bleeding and hearing loss especially in young patients, etc.) and developing drug resistance may restrict the activity of cisplatin in the treatment [2][3][4].
Arbutine is used as a powerful skin whitening agent in cosmetic industry due to its strong inhibitory effect on human tyrosinase activity [5]. It can also cure cystitis and urinary tract infections as it expels kidney stones and has antibacterial properties [6]. Moreover, it is commonly used in pharmaceutic industry due to its antioxidant, antimicrobial and anti-in ammatory activities [7]. Many studies have shown that it can be particularly effective on skin cancer [8,9]. Arbutine in glyloside form has two different isomers depending on the binding way of the hydroquinone bound to the anomeric carbon atom in the glucose molecule. These isomers are α-arbutine and βarbutine. Natural arbutine isolated from plants is β isomer [10]. Alpha arbutine is usually produced through the transglucosylation of hydroquinone by microbial glycosyltransferases. Seven different microbial enzymes that can produce α-arbutine including alpha amylase, sucrose phosphorylase, cyclodextrin glycosyltransferase, α-glucosidase, dextransucrase, amylosucrose and sucrose isomerase were identi ed [5]. In addition, deoxycarbutin (4-[(tetrahydro-2H-pyran-2-yl) oxy] phenol) obtained by reducing arbutine (removing hydroxides from the pattern in glucose ring in the hydroquinone side chain) was also produced synthetically and used in the composition of cosmetic products [11]. The conducted studies indicate that cytotoxicity of arbutine types is lower than hydroquinone and therefore it can be regarded as a safer skin whitening agent [11][12][13].
Hydroquinones have been used for nearly half a century in the treatment of melasma, hyperpigmentation and especially facial freckles. The use of hydroquinone which is a derivative of benzene, a carcinogenic substance, has recently been forbidden especially in cosmetic products [14]. One of the most important reasons is that it has been shown by scienti c studies conducted in clinics, experimental animals and cell culture models that hydroquinones can induce formations such as oxidative stress and DNA damage in the organism [15][16][17][18]. Hydroquinones can easily be formed through arbutine hyrolysis (Fig. 1).
Hydroquinone in low concentrations was found in human urine and plasma [19]. These hydroquinone amounts are lower than the amounts of arbutine taken with food. This means that arbutine isoforms release the hydroquinone in glycoslated form through hydrolysis only when they reach the tissues or cells they will affect. Therefore, arbutine isoforms can only have effects on some tissues and cells (including enzymes that can hydrolise the arbutine hydroquinone) without having toxic effects for the whole organism as hydroquinones do.

Figure 1 Hydrolise of Arbutine
Concerning the reliability of α-arbutine and β-arbutine in cosmetics, Scienti c Committee of Consumer Safety (SCCS) states that α-arbutine rate in cosmetics should have to be 2% in face creams and 0.5% in body lotions [20] and this rate can be increased to 7% in face creams for β-arbutine [21]. This information indicates us that β-arbutine isolated from natural sources is used in cosmetic products in higher rates than α-arbutine produced sythetically. In other words, it states that synthetically produced α-arbutine can have effects even in low concentrations. This can be due to the fact that α-arbutine inhibites the tyrosinose enzyme in 15 times lower concentrations than β-arbutine. Thus, the conducted studies indicate that while LD 50 doses which are effective in inhibation of tyrosinose enzyme are 2 mM for α-arbutine, they are 30 mM for β-arbutine [22,23].
New treatment protocols that are non-toxic because of their side effects, multi-targeted and do not cause drug resistance are needed in the treatment of liver cancer and other types of cancer. In this context, natural active substances with limited toxic effects continue to be alternative. The studies conducted in this eld indicate that natural active substances may contribute to the development of new solutions for cancer treatment. The effects of α-arbutine and cisplatin in HepG2 cells were investigated separately and together in this study. Thus, the possible bene ts/harms of arbutine against liver cancer and cisplatin intoxication in liver cancer were tried to be determined on an in-vitro model.

Materials And Methods
Cell lines and culture conditions HepG2 cells were obtained from the American-Type Culture Collection (ATCC) (Manassas, VA, USA). The complete medium used for the reproduction of cells was prepared using low glucose (1000 mg/L) DMEM (Sigma) including 10% of Fetal bovine serum (Copricorn), 1% of penicillin streptomycin (Sigma), 1% of sodium pyruvate (Sigma), 1% of glutamin (Sigma). All incubations administered to the cells throughout the study were performed in a CO 2 incubator (Panasonic) at 37 C and in a medium containing 5% of CO 2 .

Thawing and Culturing Cells
HepG2 cells were brought to the laboratory in cryotubes in liquid nitrogen. The frozen cells in the cryotubes were rst taken from liquid nitrogen to -80°C. After waiting for 10 minutes, it was left to dissolve by incubating in a hot water bath at 37°C for 1-2 minutes. In the meantime, 5 mL of medium was placed on a 15 mL of falcon tube. The cells in the cryotube which were about to dissolve completely in the water bath were taken to the prepared falcon tube (that medium was put into it). It was then gently pipetted and centrifuged immediately. At the end of the centrifuge, the medium was drawn with a pipette without contacting the cells accumulated on the falcon base. A second wash was performed by adding 2 mL of medium onto the cell palette again and gently pipetting. The medium remaining on the cells after the centrifuge was drawn again. HepG2 cells, which were dissolved as explained and puri ed from dimethyl sulfoxide (DMSO) in 2 washes, were dissolved in 1 mL of medium and then cultured to a 25 mL of ask. When the 25 mL of ask became capacious, the cells continued to be replicated by passaging to the 75 mL of ask.

Counting the Cells by Trypan Blue Method
When the cells are cultured to wellplates or asks in order to use in analyses, it is essential to culture the same / close number of cells in each ask or wellplate. For that reason, before the cells were cultured in well plates, how many cells per mL were determined and the cells were counted. For this purpose, trypan blue cell counting method was used. Trypan blue is a paint that enables live cells to be seen bright under the light. While live cells painted with trypan blue applied to the slide to be counted are seen as bright under the microscope, dead cells are seen as blue and at. The cells in four counting zones (A-D) consisting of 16 squares each on the counting slide were counted and averaged and therefore the cell counting was completed. Trypan blue method used in live cell counting was tried to be outlined in Fig. 2. The number of cells per milliliter of medium-cell suspension is calculated using the formula following the cell count. The amount of dilution is important in terms of the reliability of counting. The average number of cells between 50 and 150 is an indicator that the cell suspension was processed to ideal dilution. If the average number of cells is very high, volume of the cells in trypan blue-cell mixture is reduced and more diluted cell suspension is obtained and the count is repeated. In this case dilution rate will also increase.
The counting can be performed using the mixture obtained after the volume is completed to 100 µL with trypan blue (dilution rate is 10 or 5) by taking 10 or 20 µL of cell suspension in cases where the number of cells is very high. The following formula is used in order to determine the number of cell per mL: Number of live cells per mL = (Average Number of Cells) x (10 4 ) x (Dilution Factor) Cells were counted in the presented study as the following. Cells removed from the ask base by trypsinization and puri ed from trypsin by detripinization were dissolved in 4 mL of medium. 50 µL of cell suspension was taken and mixed with 50 µL of trypan blue (dilution rate is 2). Slides and lamellas (Neubauer counting chamber) to be used in counting were prepared. 10 µL of trypan-blue suspension mixture was taken by pipetting and mounted to the neubauer slide properly. Taking the averages of the cells counted in counting zones, the amount of cells per mL of cell suspension was found with the help of the formula. Afterwards, how many mL of the cell suspension should be taken according to the number of cells to be used in the applications was calculated and manipulations with the cells were initiated.

Cytotoxicity Analyses (MTT analyses)
Cytotoxicity levels of the arbutine isoforms, α-arbutine and β-arbutine, in HepG2 cells were determined by MTT (3-4,5-dimethyl-thiazolyl-2,5-diphenyltetrazolium bromide) method. Cancer cells were cultured in 96 piece-well plates as 5000 cells per 200 µL of medium in each well. Cells were incubated for 24 hours so that the cultured cells could take roots to the ask base. Following the incubation mediums of cells were exchanged with the mediums prepared by dissolving α-arbutine and β-arbutine in appropriate concentrations. The mediums with α/β-arbutines used in MTT analyses were prepared in 7 different concentrations between 0,5 and 200 mM. Each dose application was studied as at least 3 replications.
Mediums without arbutines were added into the control wells in the same volume. Then, cells were again left to incubation for 24 hours.
The MTT solution was prepared as in 5 mg / mL of concentration by dissolving the MTT salt (Sigma) in phosphate buffer (PBS) with pH 7.4. At the end of the incubation, MTT solution in a volume of 10% of the well volume was added to each well. Cells were incubated in this way for 2-4 hours more. Incubation may be limited to 2 hours if there is excessive formation of formazan crystals (purple) in the wells. Because excessive increase in colour density in wells may lead to higher optical densities to be obtained from spectrophotometers. In this case it is possible to get no results from MTT analyses. For that reason, it is important to monitor the colouration in wells after the rst two hours as the cells are incubated with MTT solution. If the formation of formazan crystals in the wells in MTT analysis is low, duration of the incubation can be extended to four hours. However, duration of incubation should not be exceed to 4 hours. Because long incubation durations may lead to coloration in dead cells due to the coloration in mediums by formazan crystals occuring in wells. This may cause to get false results from MTT analysis.
The medium in the wells was pipetted without damaging the formazan crystals formed after incubation with the MTT solution. Following this operation, 200 µL of DMSO was added to each well and formazan crystals were dissolved. Finally, optical density of the samples were almost immediately determined at 540 nm using ELISA microtype reader (Biotek, ELx800). Cell viability of the control group without any arbutine was regarded as 100 % and the effect of each dose on cell viability was calculated using the given formula below [24,10].
Cell viability (%) = [ (100 * Absorbance sample )/ (Absorbance control ) ] Lethal doses (such as LD 0 , LD 50 ) belonging to arbutine isoforms in HepG2 cells were determined using % viability rates that each dose created in cells. Lethal doses were calculated using Epa Probit Analysis Program (Version 1.5). After the lethal doses were determined in the study, experimental groups were created and administrations were initiated. The experimental groups created within the scope of the study were presented in Table 1. Table 1 Experimental groups created in the study and the administrations to cells

Groups
Administrations to the cells Group 1 :

Control
The medium used as a solvent to dissolve α-arbutine and saline solution used as a solvent for cisplatin were administered in the same volume.

Genotoxicity (Comet Assay and Micronucleus Test) Analyses
In order to determine how high and low doses of β-arbutin affect genotoxicity when administered to HepG2 cells, DNA damage and micronucleus frequencies were determined in the samples. 24 hours after the necessary operations in experimental groups indicated in Table 1, the cells were collected from the ask base by trypsinisation and detrypsinisation. Cell suspensions belonging to the experimental groups created in 1 mL of PBS (phosphate buffer seline) were used in the analyses.
Comet assay was used in order to determine DNA damage. 20 µL was taken from the cell suspension prepared for this purpose and mixed with 100 µL of LMA (low melting agarose) in eppendorf. All of the cell-LMA mixture in eppendorf was taken and mounted to slides prepared by treating with NMA (normal melting agarose) the day before. Each prepared material was painted following the lysis and electrophoresis stages. Painted preparations were scored by counting as 100 cells under uorescence microscope [25].
Numbers (frequencies) of micronucleus as a result of arbutine administrations in samples were determined using micronucleus test. For this purpose, the cells belonging to the samples in each group were spread on clean slides. Dried slides were xed with pure ethanol for 20 minutes. Then, slides were air dried. Cell nuclei were painted by waiting the dried slides for 20 minutes in 5 % of Giemsa solution. Three slides were prepared from each sample in the analyses. 1000 cells from each preparation were counted at 100X [26]. Micronucleus index values were evaluated over an average of 1000 cells.

Biochemical Analyses
Biochemical analyzes were performed on cell lysates obtained from cells. Cell lysates were prepared as the following. 24 hours after the mentioned administrations to the experimental groups (Table 1) cells were collected from falcon base following trypsinisation and detrypsinisation. Cells were nally washed with 1 mL of PBS. After the wash, 500 µL of lysis buffer (PBS solution containing 1 % of triton-X-100 and 8 % of protease inhibitor cocktail) was added onto the cells. They were pipetted. Then, they were sonicated (Binder) for 20 seconds in ice. It was waited for 40 seconds. Sonication was repeated for 10 times in this way. As a result of sonication, intracellular / organelle uids were allowed to pass into the lysis buffer. Insoluble proteins in lysis buffer despite all administrations were precipitated by centrifugation at 8500 rpm at 4 0 C for 10 minutes. The supernatants obtained in this way were used as cell lyzates in chemical analyses. Lysis buffer was used as a blank in analyses. Total protein, oxidative stress and proin ammatory cytokine levels were determined through biochemical analyses in the obtained cell lyzates.
Total antioxidant status (TAS) and total oxidant status (TOS) analyzes, which were analyzed as oxidative stress parameters, were spectrophotometrically conducted using commercial kits (Rell Assays). In addition, OSI levels were calculated by considering TAS levels determined as a general expression of antioxidants in the samples and TOS levels determined as a general expression of oxidants in the samples. OSI levels were calculated according to the following formula in accordance with the kit protocols [10].

Oxidative Stress Index (OSI) = [(TOS/TAS) x 100]
Proin ammatory cytokine (TNF-α, IL-6 and TGF-β) levels were determined at 540 nm in a microplate reader (Biotek ELx800) using human speci c ELISA kits (Sun Red). The data obtained from the biochemical analysis were normalized by dividing each sample's own total protein levels. Protein levels of cell lyzates used in biochemical analyses were determined by spectrophotometrical method at 595 nm with the help of a commercial ELISA kit (Fluka) in which Coommassi Brilliant Blue reactive was used.
Determination of RNA Isolation, cDNA Synthesis and mRNA Expression Levels Bcl-2 and Caspase 3 mRNA expression levels were analyzed by Reverse Transcription-Polymerase Chain Reaction (RT-PCR) method using total RNA isolated from control and treatment groups in HepG2 cells and cDNAs synthesized afterwards. In the analyses total RNAs belonging to HepG2 cells were isolated by using commercial kit (Gene Matrix). Concentration and quality of the obtained RNAs were determined in nanodrop (BioTek, Epoch 2). Among the isolated RNAs the ones with optical densities (OD 260/280 ) between 1.7 and 2.2 obtained at 260 nm and 280 nm were used in the study.
In order to be used as a template in the PCR reaction, 2 µg of total RNA belonging to each sample was taken and complementary DNA (cDNA) synthesis was rst performed through reverse transcriptase (RT).
For this purpose, cDNA synthesis kit was used (Thermo). Sybr Green PCR Master Mix (12.5 µL) and primer pair (oligonucleotide) were added in accordance with the protocols using 1.5 µL of the cDNA belonging to each sample. Primers are speci c for each transcription analysis and were determined using studies in the literatüre [24,28]. The primers presented in Table 2 were used as 100 ng in each RT-PCR reaction. With reference to the cycling threshold (Ct) values of the ampli cation curves of each sample in the analysis performed on the RT-PCR device (Biorad CFX-96), the relative changes in the mRNA expression levels of the target genes were calculated by 2 − ΔΔCt method [29]. Calculations were performed using REST 2009 software (Qiagen, Germany). mRNA expression levels were determined as fungible decrease or increase. Beta actin gene was used as the endogenous control. Expression levels of other genes were emended (normalized) according to the beta actin gene level of each sample.

Immunocytochemical Analyses
The

Statistical Analyses
The obtained data were identi ed as average ± standard deviation (SH) and SPSS 18 package program was used in data evaluation. Whether the data were normally distributed or not was initially tested. Oneway analysis of variance (ANOVA), one of the parametrical tests, and Duncan test as post hoc test were applied to the data with normal distribution. Kruskal-Wallis test, one of the non-parametrical tests, was applied to the analysis results without normal distribution to determine whether there was a statistical difference or not. Which groups from the control groups indicated statistical difference in the parameters in which there was a statistical difference between each other was determined by the Mann-Whitney U test.

Results
Determination of cytotoxicity levels of arbutine isoforms and cisplatin in HepG2 cells Data of the cells applied arbutine in different doses were compared with control group and the obtained % cell viability rates were presented in Fig. 3.  Lethal doses of both α-arbutine and β-arbutine and cisplatin in HepG2 cells were calculated using the data obtained from MTT analyses (Tablo 3). When the calculated lethal doses of α-arbutine and βarbutine are compared, it is observed that LD 50 dose of α-arbutine is lower. This means that the same rate of cytotoxicity in HepG2 cells can be achieved with lower doses of α-arbutine in LD 50 dose. In other words, toxicity of α-arbutine is higher than toxicity of β-arbutine. Therefore, α-arbutine was used for the rest of the study. Table 3. Lethal doses of cisplatin, α-Arbutine and β-Arbutine (LDs) The Effect of α-Arbutine in HepG2 Cells on Genotoxicity, Oxidative Stress and In ammation Conducted genotoxicity tests indicate that α-arbutine in HepG2 cells in LD 0 dose has no genotoxic effect.
However, as a result of the administration of α-arbutine and cisplatin in LD 50 doses, it was determined that both DNA damage (Comet scores) and micronucleus formation were statistically higher than the control group (Table 4) and therefore genotoxicity occurred. With the administration of cisplatin, it was determined that α-arbutine in LD 0 dose reduced the DNA damage (comet scores) and decreased the genotoxicity induced by cisplatin.  When the effect of arbutine in HepG2 cells on oxidative stress was examined, it was determined that arbutine did not affect TAS, TOS, OSI, MDA, GSH, NO parameters in LD 0 dose as compared to control group. It was observed that oxidative stress increased as levels TOS, OSI, MDA, NO went up when αarbutine was administered in LD 50 doses. It was determined that oxidative stress increased as OSI, NO levels went up when cisplatin was administered. However, it was observed that oxidative stress index decreased in the cells (Group 5) administered α-arbutine in LD 0 dose with cisplatin (Table 4).
In this studies conducted in order to identify the effect of α-arbutine on in ammation it was determined that arbutine in LD 0 dose did not affect the in ammation as compared to control group. It can be stated that acute in ammation occurs as IL-6 and TGF-β levels increase when α-arbutine is administered in LD 50 dose and TNF-α, IL-6 and TGF-β levels increase when cisplatin is administered. It was identi ed that in the group (Group 5) in which α-arbutine was administered in LD 50 dose along with cisplatin only TNF-α levels decreased as compared to the group in which cisplatin was administered.
The effect of cisplatin and α-arbutine on expression levels of caspase 3 and Bcl-2 mRNA in HepG2 cells mRNA expression levels of Bcl-2, a proliferative gene and caspase 3, an apoptotic gene, were determined when α-arbutine and cisplatin were administered to HepG2 cells in different doses (Fig. 5). While Cas3 and Bcl-2 gene expression levels did not change compared to control group as α-arbutine was administered to the cells in LD 0 dose, Bcl-2 levels were suppressed as α-arbutine was administered to the cells in LD 50 dose ( Fig. 5A and Fig. 5B). However, it was found that when cisplatin in LD 50 dose was administered to the cells, it suppressed the proliferation (Bcl-2) as expected from a chemotherapeutic agent, but induced apoptosis (Cas 3) (Fig. 5C). When the data of the group in which α-arbutine was administered to the cells in LD 0 dose along with cisplatin, it was observed that Bcl-2 levels were suppressed (Fig. 5D).

Figure 5 The effect of α-arbutine on apoptasis (Cas3) and proliferation (Bcl-2) in HepG2 cells
The numbers of Cas3 and p53 positive cells belonging to experimental groups As a result of immunocytological analyses, the numbers/rates of Cas3 + and p53 + cells in experimental groups were calculated and presented in Table 5. The obtained data refer that number of Cas3 + cells statistically increased in the groups in which α-arbutine or cisplatin was administered in LD 50 dose. It was determined that α-arbutine in LD 0 dose did not affect the number of Cas3 + cells. When p53 + cell rates % were examined, it was found out that α-arbutine in LD 0 dose decreased the number of p53 + cells; however, α-arbutine in LD 50 dose could noticeably induce p53-mediated apoptosis. The microscope images obtained from the slides used in the statistical evaluation of the experimental groups mentioned in Table 5 were presented in Fig. 6. The data were presented as average ± standard deviation (n = 6). Statistical differences between experimental groups are expressed in letters as superscripts ( a, b ). a p 0.05; refers to the groups that are statistically different from the control group. b p 0.05; refers to the groups that are statistically different from the cisplatin group.  While several antioxidants such as arbutine have a protective role for organisms and cells in low doses, they can be cytotoxic by inducing apoptosis and suppressing proliferation in high doses. Bioactive substances having protective effect in low doses and cytotoxic effect in high doses have been the subject of researches for many years due to their anticarcinogenic effects in cancer cells. In order to determine by which mechanisms α-arbutine in high doses induces apoptosis in HepG2 cells, in ammation, oxidative stress and genotoxicity levels were analyzed in experimental groups as well as apoptosis level. In the light of the ndings, it was tried to be outlined that by which mechanisms α-arbutine applied to HepG2 cells especially in LD 50 dose might have anticarcinogenic effect.
As a result of MTT analyses, it was determined that high doses of arbutine isoforms in HepG2 cells might have a cytotoxic effect. It was observed that as the doses of both α-arbutine and β-arbutine increased in HepG2 cells, cell viability decreased, but their cytotoxicity increased. No similar study was found in literature on HepG2 cells and arbutine. However, it is stated in the literature that arbutine does not indicate cytotoxicity in low doses [1], but it may have cytotoxic effects in high doses, therefore we can mention about the anticarcinogenic activity [10].
As a result of MTT analyses, it was determined that the cytotoxicity of α-arbutine, one of the isoforms of arbutine, was higher than β-arbutine in HepG2 cells. Different cytotoxicities of arbutine isomers may be related to the different pharmacokinetics of both isomers. It is stated also in a study conducted with a different cell line in literature that the cytotoxicities of α-arbutine and β-arbutine in MCF-7 cells are different. It is stated in the related study that β-arbutine has more cytotoxic effects than α-arbutine in MCF-7 cells [10].
Since it was also aimed to determine the effects of arbutine on cisplatin toxication in this study, LD 50 dose of cisplatin that might have cytotoxic effect in HepG2 cells were determined through MTT analyses.
It was determined that cytotoxicity of cisplatin in HepG2 cells rapidly increased as of 2 µg/mL of concentration. LD 50 dose of cisplatin for HepG2 cells was found as 11,06 µg/mL in the calculations. As the studies on HepG2 cells in literature are examined, it is stated that cytotoxic doses of cisplatin are higher. For instance, in a study using cisplatin doses in the 5-80 µg/mL concentration range, it is stated that cisplatin can only reduce proliferation in HepG2 cells by 30-40 % in high doses (40-80 µg / mL) [3]. Such different cytotoxicities of cisplatin in HepG2 cells can be related to the used cisplatin solution.
Because additional buffering agents such as mannitol can be added to cisplatin solutions in injectable form used for cancer patients in order to reduce toxic effects and side effects. Not the injectable cisplatin solution prepared for the treatment of cancer in humans, but the active substance cisplatin itself was used in the presented study. For that reason, toxication rate of cisplatin might have been found higher than the literature.
After the lethal doses of α-arbutine, β-arbutine and cisplatin were determined in this study, the second part of the study was initiated to determine the effects of arbutine on genotoxicity, oxidative stress, in ammation, apoptosis and proliferation in HepG2 cells. As previously mentioned, only α-arbutine was administered to the cells since the cytotoxicity of α-arbutine was found higher in this stages of studies.
It can be said that cisplatin in LD 50 dose has a genotoxic effect in HepG2 cells by increasing both DNA damage and micronucleus development. In literature it was stated with different studies that cisplatin caused genotoxicity by increasing DNA damage and micronucleus development [31,32]. It was identi ed that when α-arbutine was administered to HepG2 cells in LD 50 dose, α-arbutine had statistically more signi cant and higher genotoxic effects not as much as cisplatin as compared to control group. No study has been found in the literature on arbutine and HepG2 cells. However, in an in-vitro study it is stated that orally administered arbutine did not cause DNA damage in rat hepatocytes [33]. However, in a study on MCF-7 cells, a cell line out of HepG2, it is stated that β-arbutine, another isomer of arbutine, might have genotoxic effects by increasing both DNA damage and micronucleus development in high doses [10].
It was determined in this study that when α-arbutine was administered in LD 0 dose (group 5) after cisplatin toxication was induced in HepG2 cells, DNA damage decreased as compared to cisplatin group (Group 4). These data bring to mind that α-arbutine in LD 0 dose may be bene cial in reducing possible side effects of cisplatin in healthy cells by reducing toxicity caused by cisplatin toxication. In addition, the reason why low doses of α-arbutine reduce DNA damage may be related to the antioxidant activity of arbutine administered in low doses. In fact, in a study conducted by using U937 macrophage cells [34], it is stated that arbutine has antioxidant activity by reducing the oxidative stress caused by ionizing radiation in U937 cells. It is stated that arbutine shows this activity by neutralizing (scavenging) hydroxyl radicals.
It is seen in the studies conducted using arbutine and different cell lines that the use of arbutine for therapeutic purposes is in low doses. Studies indicate that arbutine in low doses may create antioxidant and anti-in ammatory effects without having any cytotoxicity. The effects of arbutine on oxidative stress and in ammation increasing as a result of toxication created with lipopolisaccharide in rat BV2 microglia cells were investigated in a study. The ndings indicate that arbutine reduces the oxidative stress and in ammation caused by the lipopolisaccharide without any cellular toxicity [35]. Protective effects of arbutine in low doses (50-100 ve 150 µM) determined as non-cytotoxic against the toxication created with tert-butyl hydroperoxide (t-BHP) in HepG2 cells were investigated in another study [36]. It is stated that the administration of arbutine in low doses before toxication increases the antioxidant power in HepG2 cells and therefore might reduce the cell deaths as a result of toxication. In this study cisplatin toxication was created and it was determined that both oxidative stress and in ammation increased in HepG2 cells as a result of toxication. It was also determined that α-arbutine treatment in low doses following the toxication with cisplatin in HepG2 cells reduced the oxidative stress by decreasing OSİ levels and the in ammation by decreasing TNF-α levels. (Table 4). Probably, these effects of arbutine in low doses may be related to strong antioxidant activity of Arbutine [37,34].
When a high dose of α-arbutine (57,471 mM) was administered to HepG2 cells, it was determined that oxidative stress was induced by the increase in MDA, NO and OSI levels and acute in ammation was induced by the increase in IL-6 and TGF-1β levels ( Table 4). No source studies have been found in the literature regarding the administration of arbutine in high doses in HepG2 cells. However, it is stated in a study with MCF-7 cells, another cell line, that β-arbutine used in a high dose may increase the oxidative stress and in ammation in cells [10].
The effects of arbutine, cisplatin and arbutine following the cisplatin toxication on apoptosis and proliferation were also aimed to be determined in our study. As a result of the analyses through RT-PCR and immunocytochemical methods it was determined that cisplatin caused apoptosis by inducing Cas3 expression in HepG2 cells. These data are parallel with the information in literature. Because it is stated that when cisplatin is administered to HepG2 cells, Cas3 can be induced and it can lead to apoptosis [38]. According to the obtained results, it can be said that cisplatin inducing the apoptotic Cas3 is an effective anticarcinogenic drug in HepG2 cells considering that it also supresses Bcl-2 mRNA gene expression levels as well (Fig. 5).
In the immunostochemical analyses, it was indicated that the number of p53 positive cells in the cisplatin group did not change compared to the control group. P53 is a highly important gene which plays a role in regulation of apoptosis and cellular cycle. Because cellular cycle can be deactivated by p53 effect for DNA repair in case of DNA damage. If DNA repair is not possible, apoptosis can be induced by p53 effect. Therefore, excessive inhibition or loss in function of p53 is considered among the reasons of cancer diseases. For that reason, if there is an induction / suppression of the p53 gene in HepG2 cells as a result of cisplatin toxication, it is important to clearly reveal it. As previously mentioned, although it was determined in the presented study that cisplatin did not cause any change in the number of p53 + cells in HepG2 cells, it is stated in the literature that p53 can be induced via cisplatin and deactivate the cell cycle [39,40]. Information in literature partly contradicts with our ndings. To determine and clarify the possible effects of cisplatin on p53 in hepatocellular carcinoma is important for explaining cisplatin activity in treatment. For that reason, it would be bene cial to carry out further studies on this subject.
It was determined that the number of p53 + cells increased (Table 5, Fig. 6) when HepG2 cells were treated with α-arbutine in LD 0 dose after administration of cisplatin. The fact that an agent used in combination with cisplatin induces apoptotic p53 means that anticarcinogenic activity is increased. Therefore, it can be considered that a low dose of arbutine treatment along with cisplatin in liver cancer may contribute to the current treatments. Hence, the studies with arbutine and HepG2 indicate that the number of apoptotic cells may increase as a result of the increase in Bax/Bcl-2 rate and the induction of Cas3 as the used arbutine dose increases [41]. In this context, it was determined in this study that while the number of p53 positive cells signi canty increased when α-arbutine was administered to HepG2 cells in LD 50 dose, Bcl-2 mRNA expression levels were suppressed. These data indicate that while high doses of arbutine induce apoptosis through p53, it may have anticarcinogenic effect (Fig. 7)   In conclusion, it was determined that when a low dose of α-arbutine (in LD 0 dose) alone was administered to hepatocellular carcinoma (HepG2) cells, it had no genotoxic and cytotoxic activity and did not affect in ammation, apoptosis and proliferation. In addition, it was determined that when a low dose of arbutine was used for therapeutic purpose following the cisplatin toxication, decreases in oxidative stress, in ammation and genotoxicity levels increasing as a result of cisplatin oxidation were observed; however, caspase 3 levels were not affected by this situation. In other words, arbutine administered in LD 0 dose reduces the severity of symptoms accompanying toxication without impairing the anticarcinogenic activity of cisplatin created by apoptosis in HepG2 cells (Fig. 8). These data bring to mind that when low doses of arbutine is used combined with cisplatin, it may be bene cial to reduce the side effects that may occur in other healthy cells. In order to determine this precisely, it may be useful to conduct similar studies using in-vitro models with healthy cell lines or in-vivo models using experimental animals.
As previously mentioned and the mechanism of activity is indicated in Fig. 7, it was determined that αarbutine may have anticarcinogenic effects on hepatocellular carcinoma cells by increasing oxidative

Conclusion
It is important to determine the effects of active substances such as arbutine on different disease models primarily in cell culture medium. Afterwards, if there are possible bene cial effects of the active substance whose effects are investigated, it is necessary to carry out further studies using modelings to be created with experimental animals and then phase studies in clinics. When the presented study is regarded in this context, the possible effects of arbutine as an active substance in liver cancer (HepG2) cells were investigated. It was determined that low doses of arbutine might be bene cial to remove the side effects of chemotherapeutics (cisplatin) used in cancer treatment; however, in high doses it might be used as an active substance with anticarcinogenic effect in hepatocellular carcinoma (HepG2) cells. In the light of the obtained ndings, it is thought that it might be bene cial to carry out further studies in order to present the possible effects of arbutine on liver cancer and metabolism in detail.

Declarations
Ethics approval and consent to participate: The study does not involve human participants, human data, or human tissue. In adition this article does not contain any studies with animals performed by any of the authors. Knowledge and Acknowledgement: Thanks to Ahmet Büyükben for helping in editing immunocytochemical images in the study. In addition, some parameters (MDA, GSH, NO levels) related to oxidative stress in the study were presented in a congress called "lnternational Congress on Science and  The methodology used in counting the cells by trypan blue method Figure 3 Cytotoxicity levels of α-arbutine ve β-arbutine in different doses in HepG2 cells Cytotoxicity of cisplatin in HepG2 cells   Mechanism that mediates the anticarcinogenic effects of α-arbutine in HepG2 cells (obtained by comparing the data of control group and the experimental group administered α-arbutine in LD50 dose.) Figure 8 Mechanism that mediates the cisplatin toxication reducing effects of α-arbutine in HepG2 cells (obtained by comparing the data of cisplatin and the experimental group administered LD0 dose of α-arbutine following cisplatin toxication).