Hepatotoxicity of bispHenol a under conditions of differential supplementation witH retinoids

classical xenoestrogenic in vivo effects of bisphenol a (2,2-bis(4-hydroxyphenyl)propane, BPa) are well-described in the literature, however the molecular mechanisms of BPa-induced hepatotoxicity are not fully characterized. the work is aimed to assess biochemical markers of BPa induced hepatotoxicity under conditions of differential supplementation with retinoids. We demonstrate that the absence of hepatic retinyl esters as the main form of vitamin a storage provides for a resistance to BPa induced liver damage. retinoid supplementation increases the hepatotoxic effects of bisphenol a, evidenced in higher indexes of oxidative damage of lipids, proteins and non-protein thiol groups as well as increase of serum alanine aminotransferase activity and myeloperoxidase activity in liver parenchyma. the absence of hepatotoxicity signs when hepatic retinoid stores are depleted and their presence during normal or excessive retinoid supplementation suggest that hepatic retinoid availability is one of the factors determining the hepatotoxicity of bisphenol a.

classical xenoestrogenic in vivo effects of bisphenol a (2,2-bis (4-hydroxyphenyl)propane, BPa) are well-described in the literature, however the molecular mechanisms of BPa-induced hepatotoxicity are not fully characterized.the work is aimed to assess biochemical markers of BPa induced hepatotoxicity under conditions of differential supplementation with retinoids.We demonstrate that the absence of hepatic retinyl esters as the main form of vitamin a storage provides for a resistance to BPa induced liver damage.retinoid supplementation increases the hepatotoxic effects of bisphenol a, evidenced in higher indexes of oxidative damage of lipids, proteins and non-protein thiol groups as well as increase of serum alanine aminotransferase activity and myeloperoxidase activity in liver parenchyma.the absence of hepatotoxicity signs when hepatic retinoid stores are depleted and their presence during normal or excessive retinoid supplementation suggest that hepatic retinoid availability is one of the factors determining the hepatotoxicity of bisphenol a. k e y w o r d s: retinoids, bisphenol a, hepatotoxicity, xenobiotics.
H epatotoxicity is currently one of the pri mary targets for biomedical research due to ever increasing toxic pressure on liver as an organ of homeostasis and detoxification [1].The adequate experimental models are indispensable for better understanding of biochemical mechanisms of hepatotoxicity progression and for development and testing of novel approaches to counteract this patholo gy [2].Bisphenol A (2,2-bis(4-hydroxyphe nyl)propane, BPA) is one of the potential in vivo hepatotoxicity inducer and is widely used as a mono mer for polycarbonate productions [3].There is a worrying constant danger of chronic human expo sure to low-level doses of BPA due to its leaching from BPA-containing household products (plastic containers, compact disks, DVDs, protective and corrective eyewear etc.) [4].The classical toxic ef fects of BPA are primarily determined by its xenoes trogenic qualities due to similarities between chemi cal structure of BPA and 17-β estradiol resulting in high affinity binding with estrogen receptors [5,6].Although toxic xenoestrogenic effects of BPA are described in detail, the molecular mechanisms be hind BPA-induced hepatotoxicity are still poorly un derstood and require further study.The detoxifica tion of this xenobiotic, primarily in liver, generates highly reactive bisphenol metabolites and activates free-radical processes [7].
Retinoids, including vitamin A and its metabolites (retinyl esters, retinol, retinoic acid) [8] may modulate toxicity of xenobiotics.Numerous studies [9,10] prove that retinoids control liver detoxifica tion system through regulation of expression of cyto chrome p450 isoforms by binding to nuclear retinoic acid receptor (RAR α, β, γ) and retinoid X receptors (RXR α, β, γ) [11].Since CYP-mediated metabolism of xenobiotics may produce more toxic highly reac tive intermediates, the substances that can affect the activity of detoxification system may define toxici ty of xenobiotics.It is thus sensible to investigate BPA toxicity under differential supplementation with retinoids.
The aim of this work was to evaluate the bio chemical parameters of BPA-induced hepatotoxicity under differential supplementation with retinoids.

materials and methods
The experimental animals were C57BL/6J mice (wild type, WT) of 2.5-3 month age, with body mass of 20-25 g.The animals, which are from a congenic subline of C57 mice, were kindly provided by prof.W. Blaner from the Institute of Human Nutrition, BPA was diluted in corn oil (used as vehicle) and given orally daily for 3 days in doses of 50 mg/ kg of body mass, corresponding to the lowest ob servable adverse effect level (LOAEL) dosage [13].
We used the following methodology to inves tigate BPA hepatotoxicity under differential supple mentation with retinoids: • Depletion of hepatic vitamin A stores (as retinyl esters) was modeled by usage of C57BL/6J transgenic mice that are unable to store retinyl esters due to knockout of lecithin:retinol acyltransferase (EC 2.3.1.135)gene (Lrat -/-), their phenotype has been characterized thoroughly [14]; • Vitamin A overconsumption was modeled by gavage of retinyl acetate (Rac) in a very high dose of 3000 IU at 12 h intervals for 3 days (the physiological dose is 30 IU) [15].
The animals were divided into the following groups (5 to 6 animals per group): • Group I (control group I) -wild type animals receiving the vehicle alone; • Group II (experimental group I) -wild type animals receiving 50 mg/kg of BPA per os; • Group III (experimental group II) -wild type animals receiving 50 mg/kg of BPA and 3000 IU of retinyl acetate per os; • Group IV (control group II) -Lrat -/-animals receiving the vehicle alone; • Group V (experimental group III) -Lrat -/- ani mals receiving 50 mg/kg of BPA per os; • Group VI (experimental group IV) -Lrat -/- animals receiving 50 mg/kg of BPA and 3000 IU of retinyl acetate per os.
Toxic lesions of liver were evaluated biochemi cally 72 h after start of the experiment.The animals were euthanized under light ether anesthesia.The liver was excised and blood collected from inferior vena cava.Blood serum was obtained by centrifuga tion at 1500 g for 15 min.
The extent of liver lesions was assayed through catalytic activity of alanine aminotransferase (ALT, EC 2.6.1.2) in blood serum by Reitman and Frankel method kit (Felicit Diagnostika, Dnipro, Ukraine) and expressed as IU/L.Myeloperoxidase activity (MPO, EC 1.11.1.7)as an indicator of inflammatory infiltration of hepatic tissue was assayed as has been described [16] and expressed as ΔD×min 1 ×mg 1 of protein.
Oxidative degradation of liver biomolecules was assayed as content of thiobarbituric acidreac tive substances (TBARS), protein carbonyl groups, protein and non-protein thiol groups.TBARS level was assayed after method [17] that is based on re action between these substances and thiobarbitu ric acid, the resulting absorbance was measured at λ = 532 nm (ε = 1.56×10 5 cm 1 ×M 1 ).TBARS con tent was expressed as nmol/mg of protein.Protein thiol groups content was assayed after method [18], which is based on reaction of protein carbonyls with DNPH, the resulting absorbance was measured at λ = 370 nm (ε = 21×10 3 cm 1 ×M 1 ).Carbonyl groups level was expressed as nmol/mg of protein.Protein and nonprotein thiol groups in liver were assayed by reaction with Ellman's reagent that produces yellow hued 2-nitro-5-thiobenzoate anion, the intensity of color is proportional to thiol content [19,20].Protein and nonprotein thiols were separated by sedimen tation in 20% trichloroacetic acid.The content was expressed as nmol/mg of protein.
Protein concentration was determined by Low ry method [21].
Statistical data processing was done with Mi crosoft Excel software using one-way analysis of variance followed by Tukey's honest significant dif ference test.The differences were considered sig nificant if P ≤ 0.05.In graphic representation, the values are indicated by Latin letters (a, b, c) if the differences between them are significant, e.g.values indicated by a differ significantly from those indi cated by b and c.Values indicated by the same letter do not differ significantly.

results and discussion
The results of our studies demonstrate that BPA administration to wild type animals ( sufficient with regards to their hepatic retinoid stores) caused symptoms of hepatotoxicity to manifest on 72 nd h of experiment.This is evidenced by 1.6-fold increase in blood serum ALT activity and twofold increase in hepatic MPO activity as a consequence of freeradical processes elevation, due to primarily BPA metabolism [22].
We observed no significant differences in blood serum ALT activity and hepatic MPO activity of animals with no hepatic retinoid stores (Lrat -/-animals) in comparison with those of control animals (Fig. 1), indicating that BPA administration to Lrat -/-ani mals did not cause liver lesions.On the other hand, the administration of 3000 IU of vitamin A to such animals (deprived with endogenous retinoid stores) caused 61% increase in ALT and 6-fold increase in MPO in comparison to animals administered with BPA.This evidences onset of hepatotoxicity that is dependent on availability of retinoids.Comparative ly, administration of very high doses of vitamin A to wild-type animals caused the BPA-induced liver lesions to become more severe, expressed as twofold rise in blood serum ALT activity and 5-fold rise he patic MPO activity, correspondingly (Fig. 1).
These primary biochemical indicators of hepa totoxicity in wildtype animals were corroborated by severe oxidative damage to liver lipids and proteins, which may result either from effects of toxic BPA radicals or from BPA-induced activation of free-radical processes [23].The most pronounced changes were those of TBARS and protein carbonyl groups, the levels of which rose by twofold and 1.3times un der effect of BPA, correspondingly (Fig. 2, a, B).We also found decreased protein and nonprotein thiol levels in comparison to control by 41 and 66%, cor respondingly (Fig. 2, c, D).This oxidative damage to liver biomolecules was the basis of hepatotoxicity onset as evidenced by increased blood serum ALT and hepatic parenchyma MPO activity (Fig. 1).
Conversely, Lrat -/-animals did not have statisti cally significant changes in levels of TBARS, protein carbonyl groups and protein and nonprotein thiol groups in comparison to control group of wildtype animals that were not subjected to BPA administra tion (Fig. 2).It is obvious that lack of hepatic retinoid stores (as retinyl esters) provides for defense against oxidative damage to liver biomolecules.Administra tion of 3000 IU of vitamin A to such animals caused decrease in protein and non-protein thiols, increase in TBARS and protein carbonyl groups, which indi  cates freeradical damage to liver biomolecules as sociated with availability of retinoids (Fig. 2).More over, the simultaneous administration of BPA and very high doses of retinyl acetate resulted in even more pronounced damage to liver protein and lipid components in wild-type animals (Fig. 2).It needs to be mentioned that administration of 3000 IU of vitamin A to wild-type and knockout animals with out administration of BPA was not associated with significant changes in the investigated parameters, which remained at control levels (data not shown).

Fig. 2. content of products of oxidative degradation of cellular biomolecules in liver. a -content of thiobarbituric acid-reactive substance; B -protein carbonyl groups; c -protein thiol groups; D -non-protein thiols
The effects indicating relation BPA toxicity to retinoid availability have to be mediated by their ability to affect components of liver detoxification system [9,10], and especially CYPs -the key en zymes of phase I detoxification.The sped up me tabolism of BPA aimed to detoxify it paradoxically leads to production of higher amounts of its toxic metabolites -2-or 3-OH-BPA and quinone forms [25], which in turn potentiates BPA hepatotoxicity.We conceivably observed this under administration of very high doses of vitamin A. Lack of retinoids provides for organism's tolerance to BPA-induced hepatotoxicity, while administration of very high vi tamin A doses aggravates the hepatotoxic processes.
A similar effect was detected in our previ ous studies on the model of thioacetamideinduced hepatotoxicity.In that model, similarly, the lack of endogenous stores of retinoids in animals provided for endurance against acute thioacetamideinduced liver lesions, and additional supplementation with alimentary retinoids potentiated the hepatotoxicity of the xenobiotic [24].The modulating effects of retinoids were demonstrated on models of CCl 4 and acetaminophen-induced hepatotoxicity.In works [25][26][27] it has been demonstrated that additional sup plementation with retinoids may aggravate processes associated with toxic lesions of the liver induced by administration of xenobiotics.
The results of this work and literature analy sis [8] allow to uncover common patterns and suggesting that retinoids are actively involved in pro cesses of in vivo biotransformation of xenobio tics.This interaction has to rely on realization of genomic effects of retinoids via their nuclear receptors, pri marily through retinoid participation in regu lation of metabolic detoxification pathways [8].On the other hand, the results of this and other studies [24] indicate that retinoiddependent potentiation of bio transformation of xenobiotics is not always a posi tive result in itself, as it may generate more toxic intermediates, which eventually will aggravate the hepatotoxic effects.Our results demonstrate that this problems needs to be investigated in detail, in cluding perspectives for analysis of expression and activity of specific elements of cellular detoxifica tion system, features of free-radical damage to cel lular compartments and activity of components of antioxidant system.The involvement of retinoids in processes of detoxification, additionally, will lead do depletion of these essential compounds and will result in marginal vitamin A deficiency associated with unbalancing of other retinoiddependent meta bolic pathways.
We have therefore established that lack of hepatic retinoid stores provides for organism's enduran ce against BPA-induced liver lesion.Ad ditional supplementation with retinoids potentiates the hepatotoxic effects of BPA resulting from oxi dative dama ge to hepatic biomolecules, evidenced by increased blood serum ALT activity and hepatic parenchyma MPO activity.Since these indicators of hepatotoxicity are absent under lack of hepatic reti nyl esters and start to appear under normal or higher vitamin A doses, we conclude that retinoid content is one of the factors determining BPA hepatotoxicity.К л ю ч е в ы е с л о в а: ретиноиды, бисфе нол А, гепатотоксичность, ксенобиотики.

Fig. 1 .
Fig. 1.Blood serum alanine aminotransferase (a) and hepatic parenchyma myeloperoxidase (B) activity in experimental animals.Vehicle -corn oil; BPa -bisphenol a; rac -retinyl acetate.here and for Fig. 2 the data indicated by indexes (a, b, c) differ statistically significant, P < 0.05