Introduction 

Humans are constantly exposed to factors causing oxidative stress including pollutants, radiation and oxidized food [1]. Oxidative stress, defined as a loss of balance between the cellular concentration of reactive oxygen species and the cell’s antioxidant capacity, is implicated in the onset of various diseases [2]. The human body has several endogenous systems [3] with which it can protect itself against oxidative stress, but antioxidant factors acquired from food also play a key role. Indeed, certain micronutrients obtained from food have potent antioxidant properties and may play an important role in maintaining the oxidative/antioxidative balance, especially if the diet is rich in these constituents [4]. In recent years, a variety of vegetables that contain antioxidants potentially capable of preventing oxidative stress reactions, such as those mediated by the formation of free radical species have been studied. Tomatoes, spinach, green peppers and cabbage are important sources of vitamin C. In vivo, this vitamin acts as scavenger of oxygen radicals and also as competitive inhibitor of nitrosamine synthesis from nitrite and amines in vivo [5]. Isothiocyanates are a family of molecules which are abundant in cruciferous vegetables such as broccoli, watercress and cauliflower. 

Sulforaphane, the best known isothiocyanate, induces drug metabolizing enzymes such as glutathione Stransferase A1/2 isoforms and NAD(P)H:quinone oxidoreductase (NQO1) in primary hepatocytes [6]. Whole grains are good source of B group vitamins, vitamin E, some minerals (zinc, magnesium and phosphorous), and they contain a variety of phytochemicals such as phytoestrogens, phytate, proteins, polysaccharides, phenols and lignans that are able to minimize oxydive damage [7]. All these components may act synergically [8]. By contrast, refined grains have a reduced nutrient content as the milling process results in the loss of dietary fibre, vitamins, minerals, lignans, phytoestrogens, phenolic compounds and phytic acid [9]. Many wheat proteins contain reduced sulfhydryl groups, which can have some free radical scavenging activity. Phytic acid can protect tissues against oxidative reactions by sequestering and inactivating pro-oxidative transition metals [3]. In epidemiological studies, whole grain consumption is associated with improvements in body mass index (BMI) [10] and insulin sensitivity [11] as well as with lower incidences of type 2 diabetes [12], cardiovascular diseases [13], and colorectal cancer [14].

 Little is known about how cereals effect cells and to our knowledge, no research has yet been done on the antioxidant properties of whole grain products in primary hepatocytes. Several studies have shown that some phytochemicals can modulate antioxidant and phase II enzymes through the activation of nuclear factor E2-related protein (Nrf2) [15]. Nrf2 is a basicleucine zipper transcription factor that under basal conditions, is present in an inactive form in the cytoplasm, bound to the Kelchlike ECH- associated protein 1 (Keap1) [16]. Various agents including Antioxidant Response Element (ROS) and weak electrophiles (e.g. isothiocyanates) can alter the Keap1-Nrf2 protein complex and free Nrf2 through phosphorilation or alkylation of one or more of the 27 cysteine residues in Keap1 [17]. When this occurs, Nrf2 translocates into the nucleus. Upon activation, Nrf2 dimerizes with a small Maf protein then binds to antioxidant responsive element (ARE) sites in the promoter regions of antioxidant and phase II genes, thereby inducing their transcription [18].

 In recent years, many authors have suggested the existence of cross-talk between Nrf2/ARE and the nuclear factor-kappa B (NF-kB) signaling pathways in response to inflammation [19–21]. The Nrf2 and NF-kB signaling pathways interface at several points to control the transcription or function of downstream target proteins [22]. In addition, ROS now appear to act as second messengers in numerous signaling pathways [23–24]. One signaling pathway that engages in cross-talk with ROS involves NF-kB family transcription factors [25–27]. It had already been shown twenty years ago by Schreck and coworkers [28] that oxidative stresses, such as addition of extracellular hydrogen peroxide, can induced NF-kB nuclear translocation in several cell lines. The NF-kB family is made up of NF-kB1 (p50), NFkB2 (p52), RelA (p65), c-Rel and RelB. In the absence of stimuli, NF-kB, is associated with the inhibitor protein, IkBa, and sequestered in the cytosol. Upon stimulation with a NF-kB inducers, IkBa is rapidly phosphorylated on two serine residues (S32 and S36), which targets the inhibitor for ubiquitination and degradation by proteosome. Lisosan G is a powder obtained from Triticum Sativum (wheat) and it is registered with the Italian Ministry of Health as a nutritional supplement. In the production process, the wholegrain is first ground to a rough powder. From this intermediary product, the bran and germ are separated and collected for further treatment which consists in the following: water is added to moisten the mix, then selected microbic starting cultures are inoculated to initiate fermentation.

 The starting cultures typically consist of a mix of lacto-bacillus and natural yeast strains. Once the product is sufficiently fermented, it is dried. The resulting dry powder is now Lisosan G, which is widely used in food production thanks to its rich nutritional content. It contains vitamins, minerals and polyunsatured fatty acids as well as having significant antioxidant activity [29]. In vivo, Lisosan G protects against cisplatin induced toxicity [30], and a recent paper showed that Lisosan G helps prevent microcirculatory dysfunction [31]. The authors of these works suggested that the protective effect of Lisosan G could be associated with the attenuation of oxidative stress and the preservation of antioxidant enzymes. To date, no studies have attempted to identify the molecular mechanism that determines antioxidant properties of Lisosan G. For this reason, in the present study, we investigated the effects of Lisosan G on the antioxidant and drug-metabolising enzymes at transcriptional, catalytic and protein levels using cultures of primary rat hepatocytes.

 Materials and Methods 2.1 Chemicals Lisosan G is registered as nutritional supplement by the Italian Minister of Health and was supplied by Agrisan Company, Larciano (PT), Italy. Collagenase; dexamethasone; insulin; glucagon; penicillin/streptomycin; ampicillin/kanamycin; fetal bovine serum; 4-(2-hydroxyethyl)-1-piperazineethanesulfonic acid (HEPES); Tween 20; phenylmethylsulfonyl fluoride (PMSF); leupeptin; apoprotein; pepstatin; tunicamycin; Williams E medium; bovine serum albumin (BSA); epidermal growth factor (EGF); b-nicotinamide adenine dinucleotide reduced (NADPH); ethylenediaminetetraacetic acid (EDTA); ethylene glycol tetraacetic acid (EGTA); glutathione (GSH); Glutathione disulfide (GSSG); and hydrogen peroxide (H2O2) were all supplied by Sigma-Aldrich (St. Louis, MO). Rabbit polyclonal anti-Nrf2 (sc-13032), anti-heme oxygenase-1 (sc-10789), NFkB (sc-7178), b-actin (sc-130657), PARP-1 (sc-25780) and goat anti-rabbit (1:2000 or 1:5000) were from Santa Cruz Biotechnology Inc. (Santa Cruz, CA). Collagen (type I) was prepared by the method of Beken et al. (1998). 2.2 Primary rat hepatocytes isolation, culture and treatments Hepatocytes were isolated from 200–300 g Wistar male rats with free access to drinking water and food and on a 12 h light/ dark cycle. The research with the use of animals was approved by the Italian Ministry of Health in compliance with European Community law n. 116/92. The approved protocol number is 10/ 09. The animals were anesthetized with an intraperitoneally injection of Zoletil H (40 mg/kg) and then subjected to midline laparotomy in order to exteriorize the liver and isolate the portal vein. A needle was inserted into the portal vein and then the liver was perfused as described previously [32]. 

After filtration and centrifugation, the cell viability was determined by trypan blue exclusion. The cells were dispersed in Williams E medium containing 39 ng/ml dexamethasone, 0.5 U/ ml insulin, 0.007 mg/ml glucagon, 5 mg/ml penicillin and streptomycin, 5 mg/ml ampicillin and kanamycin and 10% fetal bovine serum. The cells were plated at a density of 4.56106 cells/ 8 ml on 100 mm cell culture dish pre-coated with 3 ml of collagen (type I) solution (1 mg/ml). The cultures were maintained at 37uC in 5% CO2 in a humidifier incubator. After 5 h, the medium was replaced with serum-free Williams E medium supplemented with 2% BSA, 7.5 mg/ml hydrocortisone 21-hemisuccinate sodium salt and EGF 20 ng/ml. Cultures were maintained in this medium at 37uC and 5% CO2 for 24 h. After this period, the medium was removed and a second layer of type I collagen was added to create a collagen-gel sandwich culture [33] and after 45 minutes serumfree Williams E medium was added again. The cells were maintained for additional 24 h before treatments. Cells treatments were divided in four different groups: in the first group (control), the cells were treated with medium only; in the second one with Lisosan G 0.7 mg/ml (Lis); in the third group with H2O2 200 mM and in the last group, the H2O2 200 mM was added after 1 h pretreatment with Lisosan G 0.7 mg/ml (Lis+H2O2).

 We have used different concentrations of H2O2 (10–500 mM) and we chose 200 mM (good toxicity); we also used for Lisosan G different concentrations from 0.1 to 2.8 mg/ml and 0.7 mg/ml was the best concentration in terms of maximum protective effect and without toxicity. The cytotoxic effects compared with the vehicle-only controls, was also measured by lactate dehydrogenase assay (data not shown). We have also performed experiments in function of time of treatments of H2O2 and Lisosan G and we chose the time of treatment on the basis of the best results. 2.3 Enzymatic activities After 24 h the end of treatment, the medium was removed and a collagenase solution was added. After 30 minutes, recovered cells were centrifuged (4006g) for 3 minutes at 4uC. The cell pellet was sonicated and used for the microsomal preparation [34]. Total protein concentration was determined by the method of Lowry [35]. NAD(P)H:quinone oxidoreductase (NQO1) activity was measured by the method of Bensen et al. [36]. Glutathione-Stransferase (GST) activity was quantified as previously described by Habig et al. [37] using 1-chloro-2,4-dinitrobenzene as substrate. Catalase activity was monitored following the H2O2 decomposition at 240 nm, as described by Cao and Li [38]. Heme oxygenase-1 (HO-1) activity was determined by the method of Naughton et al. [39]. Lactate dehydrogenase activity was assayed as previously described [40]. Reduced GSH was measured using the method previously described by Hissin and Hilf [41]. Lipid peroxidation was monitored by determining the production of malondialdehyde (MDA)-like products, quantified by the reaction with thiobarbituric acid (TBA) as reported by Stacey et al. [42]. 

2.4 RNA Extraction and cDNA synthesis Total cellular RNA was extracted from primary rat hepatocytes 4 h after ITC treatment, using the RNeasy Mini Kit (Qiagen, Valencia, CA), following the supplied protocol. RNA was quantified using NanoDrop (Celbio, Mi, Italy); its purity and integrity were evaluated by checking the absorbance ratio at 260– 280 nm and assessing the sharpness of 18S and 28S ribosomal RNA bands on agarose gel stained with ethidium bromide. Genomic DNA elimination and reverse transcription of total RNA were performed using QuantiTect Reverse Transcription Kit (Qiagen). 2.5 RT-PCR Two microliters of cDNA were added to a PCR Master Mix (GoTaq Green Master Mix, Promega, Madison, WI) for the amplification reaction (various cycles) performed using for each transcript 400 nM of forward–reverse primers for heme oxygenase-1 (GenBank accession no. NM_012580.2), NQO1 (GenBank accession no. NM_017000.3), b-actin, as housekeeping gene, (GenBank accession no. NM_031144.2) and the annealing temperature indicated in Table 1. The DNA fragments were separated on ethidium bromide-stained 1% agarose gel and visualized by transillumination with ultraviolet light. Bands obtained from five independent rat experiments were quantified by an Image J software. The results have been normalized to bactin levels and are expressed as percentages of control. Results are reported as means 6 SD of cells from five independent experiments using five rats. 2.6 Preparation of nuclear fractions Nuclear and cytosolic extracts were prepared by previously established methods [43]. 

Briefly, hepatocytes were washed twice with 16phosphate buffer saline (PBS). Cells were then harvested in 1 ml of PBS and centrifuged at 800 g for 3 min at 4uC. The pellet was carefully resuspended in 200 ml of cold hypotonic buffer, consisting of 10 mM HEPES (pH 7.9), 10 mM KCl, 0.1 mM EDTA, 0.1 mM EGTA, 1 mM dithiothreitol and complete protease inhibitor cocktail (Sigma antiprotease cocktail P8340). After addition of NP40 to a final concentration of 0.3%, the cells were vortexed and centrifuged at 800 g for 3 min at 4uC. The resulting nuclear pellet was resuspended in 30 ml of cold nuclear extraction buffer (20 mM HEPES (pH 7.9), 0.4 M NaCl, 1 mM EDTA, 1 mM EGTA, 1 mM dithiothreitol, 25% glycerol and protease inhibitors) and incubated on ice for 30 min. The nuclear extract was finally centrifuged at 15000 g for 15 min at 4uC. The supernatant containing nuclei proteins was aliquoted and stored at 230uC. 2.7 Immunoblot analysis Nuclear and microsomal proteins from primary rat hepatocytes were separated according to Laemmli [44] on SDS-10% (v/v) 1.0 mm thick polyacrylamide gels and then electrophoretically transferred onto nitrocellulose membranes following the method of Towbin et al. [45].