Betacyanins, major components in Opuntia red-purple fruits, protect against acetaminophen-induced acute liver failure

Acetaminophen (APAP) misuse or overdose is the most important cause of drug-induced acute liver failure. Overdoses of acetaminophen induce oxidative stress and liver injury by the electrophilic metabolite N-acetyl-pbenzoquinone imine (NAPQI). Plant-based medicine has been used for centuries against diseases or intoxications due to their biological activities. The aim of this study was to evaluate the therapeutic value of Opuntia robusta and Opuntia streptacantha fruit extracts against acetaminophen-induced liver damage and to identify the major biocomponents on them. Opuntia fruit extracts were obtained by peeling and squeezing each specie, followed by lyophilization. HPLC was used to characterize the extracts. The effect of the extracts against acetaminopheninduced acute liver injury was evaluated both in vivo and in vitro using biochemical, molecular and histological determinations. The results showed that betacyanins are the main components in the analyzed Opuntia fruit extracts, with betanin as the highest concentration. Therapeutic treatments with Opuntia extracts reduced biochemical, molecular and histological markers of liver (in vivo) and hepatocyte (in vitro) injury. Opuntia extracts reduced the APAP-increased expression of the stress-related gene Gadd45b. Furthermore, Opuntia extracts exerted diverse effects on the antioxidant related genes Sod2, Gclc and Hmox1, independent of their ROSscavenging ability. Therefore, betacyanins as betanin from Opuntia robusta and Opuntia streptacantha fruits are promising nutraceutical compounds against oxidative liver damage.


Introduction
Acute liver failure (ALF) is a rare and unpredictable clinical syndrome, characterized by sudden, severe liver dysfunction associated with coagulopathy and hepatic encephalopathy (Khandelwal et al., 2011). An important cause of ALF is unintentional misuse of over-thecounter (OTC) pain medication, in particular acetaminophen, the most commonly used OTC product in the United States (Wolf et al., 2012). Acetaminophen, or paracetamol, 4-hydroxy-acetanilide, N-acetyl-p-aminophenol (APAP) is a safe and effective analgesic and antipyretic OTC drug when used as recommended (Wang et al., 2017). However, APAP misuse or overdose can lead to ALF and APAP overdose is currently the leading cause of ALF in adults in Western countries (Fontana, 2008;Kim et al., 2015;Larson et al., 2005). At therapeutic doses, APAP is conjugated by glucuronidation or sulphation in the liver and excreted into the urine (> 90%). A small amount is excreted unchanged and < 10% is biotransformed by cytochrome P450 enzymes into the reactive intermediate N-acetyl-p-benzoquinone-imine (NAPQI), which under normal conditions is inactivated by reduced glutathione (GSH) (Eugenio-Pérez, Montes de Oca-Solano, & Pedraza-Chaverri, 2016;Lancaster, Hiatt, & Zarrinpar, 2015). At high doses of APAP, the glucuronidation and sulphation pathways are saturated resulting in excessive production of NAPQI causing depletion of liver GSH. NAPQI then forms covalent bonds (adducts) with proteins and non-protein thiols, initiating alkylation of proteins, lipid peroxidation of membranes, imbalance of intracellular calcium homeostasis, production of reactive oxygen species (ROS) and reactive nitrogen species (RNS), ATP depletion and eventually cell death (McGill & Jaeschke, 2013;Seki, Brenner, & Karin, 2012). The only approved treatment for APAP overdose is N-acetylcysteine (NAC), a precursor of GSH. This reduces oxidative stress and liver injury (Ferner, Dear, & Bateman, 2011). However, NAC is not always effective and liver transplantation is the last therapeutic option. Therefore, there is an urgent need for novel and effective interventions to improve the prognosis of APAP-induced ALF.
The aim of this study was to investigate the therapeutic effect of fruit extracts of two Opuntia species, Opuntia robusta and Opuntia streptacantha on APAP-induced hepatotoxicity both in vivo and in vitro, and to identify the main component(s) possibly related to their protective properties.

Plant materials and preparation of extracts
Ripe fruits of Opuntia robusta and Opuntia streptacantha were collected from randomly selected plants in a semi-arid region of Aguascalientes, México (21°46′55.86″ N, 102°6′16.08″ O, and 1994 m above sea level). The juice extraction of each Opuntia fruit species was carried out by using a Braun J500 juice extractor (Braun, GmbH, Taunus, Germany) and juice was collected into 50 ml dark tubes to remove non-soluble fibers by centrifugation at 5000 rpm for 15 min at 4°C. After that, the juice extracts were filtered through an 8-μm pore size Whatman filter paper, frozen at −80°C and lyophilized as described previously (González-Ponce et al., 2016).

Betacyanins content
The betacyanins content was performed as described by (Sumaya-Martínez et al., 2011). Juice extracts were reconstituted in 50 ml of deionized water and clarified by centrifugation at 12,000g for 15 min at 15°C. Determination was carried out spectrophotometrically at 535 nm and the concentration was calculated using the follow equation: where: A = absorbance 535 nm, DF = dilution factor, MW = molecular weight (550 g/mol), ∈= extinction coefficient (60,000 L/mol cm), and 1 = width of the spectrophotometer cell (1 cm). The quantification was performed in triplicate on a Biotek PowerWave XS microplate reader and the results were expressed as mg of betacyanins equivalents/L.
A known concentration of betanin (10 mg/ml), gallic acid (0.5 mg/ ml) and quercetin (0.5 mg/ml) standards from Sigma-Aldrich (St. Louis, MO, USA) were used to identify the main biocomponents in the Opuntia extracts by comparing retention time and spectra at 535, 280 and 360 nm, respectively.

Animals
Adult male Wistar rats (200-250 g) were used for the in vivo and in vitro studies. The animals were obtained from the animal facility of the Universidad Autónoma de Aguascalientes (for the in vivo experiments) and University Medical Center Groningen (for the in vitro experiments) and kept in polypropylene cages at room temperature (25 ± 2°C) with food and water ad libitum. Experiments were approved by and performed according to the guidelines of the local committee for care and use of laboratory animals (permission No. 6415A of the committee for care and use of laboratory animals of the University of Groningen and Mexican governmental guideline NOM-033-ZOO-1995).
2.6. Experimental design 2.6.1. In vivo experiment Albino male Wistar rats (200 -250 gr) were randomly divided into seven groups (n = 10): Group 1 -Control; Group 2 -APAP; Group 3 -Opuntia robusta (Or) treated; Group 4 -Opuntia streptacantha (Os) treated; Group 5 -APAP + Or treated; Group 6 -APAP + Os treated; and Group 7 -APAP + NAC. Rats (groups 2, 5, 6 and 7) were intoxicated with a single dose of APAP (500 mg/kg, intraperitoneally, Sigma-Aldrich). After 0.5 h, rats in the appropriate groups were therapeutically treated with a single dose of Opuntia extract (800 mg/kg, orally) (González-Ponce et al., 2016) or NAC (300 mg/kg, intraperitoneally, Sigma-Aldrich) (Geng et al., 2015). After 6 h of APAP intoxication samples of blood and liver tissue were collected from six animals of each group for the assessment of biochemical markers of hepatic damage and for RNA isolation. Liver tissue from the other animals was collected 24 h after APAP intoxication for histological evaluation.
Biochemical markers of liver damage, alanine aminotransferase (ALT), aspartate aminotransferase (AST), lactate dehydrogenase (LDH), and alkaline phosphatase (ALP) were measured spectrophotometrically (Varian UV visible spectrophotometer, model DMS80, Varian, Inc., CA, USA) in plasma using commercial kits (SPINREACT, Girona, Spain). The values represent the mean of six samples ± standard error of the mean (SEM) and are expressed as IU/L. Hepatic GSH content in tissue homogenates from experimental animals was determined according to (Hissin & Hilf, 1976), using o-phtaldehyde (OPT) as the fluorescent reagent. The fluorescence intensity was measured at 420 nm using 350 nm as the excitation wavelength using a luminescence spectrophotometer (Model LS-50B, PerkinElmer Inc., Waltham, MA, USA). The values represent the mean of six samples ± SEM and are expressed as µg/g. Determination of malondialdehyde (MDA), a product of lipid peroxidation, was performed using the thiobarbituric acid reactive substance (TBARS) method according to (Uchiyama & Mihara, 1978) with some modifications. Samples were measured spectrophotometrically (Varian UV visible spectrophotometer, model DMS80) at 530 nm. The values represent the mean of six samples ± SEM and are expressed as nmol/100 mg. Histological analysis was performed by collecting liver tissue from the experimental animals 24 h after APAP intoxication. Animals were anesthetized with sodium pentobarbital and systemically perfused with saline solution (sodium chloride 0.9%), containing 0.5% heparin and 0.1% procaine and fixed in situ with neutral formalin (10%). The hepatic tissue was embedded in paraffin blocks and sections of 5 µm were prepared with a microtome RM2125RT (Leica Biosystems, USA). The sections were stained with hematoxylin/eosin (H&E). Liver tissue images were obtained using a slide scanner NanoZoomer 2.0 HT (Hamamatsu Photonics, Japan) and Aperio ImageScope Pathology slide viewer software (Leica Biosystems).

In vitro experiments
Stock solutions of acetaminophen (APAP, 2 mol/L in DMSO) and Nacetylcysteine (NAC, 1 mol/L in PBS) were prepared for all the in vitro experiments. Opuntia cactus fruit extracts were sterilized through filtration (0.2 µm pore size) before use.
Hepatocyte cultures were divided into seven groups following the same set up as in the in vivo experiments. Cells from Groups 2, 5, 6 and 7 were treated with 10 mmol/L APAP for biochemical and molecular assays and 20 mmol/L for cell death assays (González-Ponce et al., 2016). After 0.5 h, cells were therapeutically treated with a single dose of each Opuntia extract (16.5 mg of lyophilized extract ≈ 10 mg/mL) or NAC (5 mmol/L) (Odewumi et al., 2011). Cells were harvested at 24 h after APAP intoxication for biochemical assays and RNA isolation. LDH assay was used to determine necrotic cell death and performed 24 h after APAP intoxication as described by (Verhaag et al., 2016). Percentage of LDH released was calculated by measuring the LDH activity in both the medium and cell lysates. Determination of LDH in each group was performed in triplo per experiment and values represent the mean of three different experiments ± SEM.
After 24 h of APAP intoxication, SYTOX® Green nuclei acid stain (Invitrogen) was added to the cells for 15 min at 37°C (1:40,000, diluted in William's E medium) to determine necrotic cell death by fluorescence microscopy (DMI6000B, Leica Microsystems, Germany) at 450-490 nm as reported by (Conde de la Rosa et al., 2006).

RNA isolation, reverse transcription and quantitative real-time PCR
Total RNA from in vivo and in vitro samples was isolated using Trireagent (Sigma-Aldrich), following manufacturer's protocol. RNA quantity and quality were determined using the Nanodrop spectrophotometer (Thermo Scientific, Wilmington, DE, USA). Reverse transcription PCR (RT-PCR) was performed with 2.5 μg of RNA using the Moloney murine leukemia virus (M−MLV) reverse transcriptase system and random nanomers from Life Technologies (Breda, The Netherlands). RT-PCR was performed in 3 steps: 10 min at 25°C, 1 h at 37°C and 5 min at 95°C with the GeneAmp PCR system (Applied Biosystems, Nieuwekerk a/d IJssel, the Netherlands). Quantitative realtime PCR (qPCR) was performed using 4 μl 20-fold diluted cDNA in combination with 2x master mix (Eurogentec, Maastricht, The Netherlands) in a total volume of 20 μl. 18S mRNA levels were used as housekeeping gene. Fluorescence was measured using the 7900HT Fast Real-Time System, and SDS 2.3 software (Applied Biosystems) (Verhaag et al., 2016). Results are expressed as fold induction and each value represents the mean of four samples (in vivo) and three different experiments (in vitro) ± SEM. Primers and probes are listed in Supplemental Table 1.

Statistical analysis
Data acquired from the experiments were statistically analyzed using GraphPad Prism 5 software (La Jolla, CA, USA). Considering a normal distribution of the values, a one-way analysis of variance (ANOVA) and a post-hoc Dunnett's multiple comparison test were used to compare the experimental groups and to determine significant differences with a confidence interval of 95%. For the betacyanins determination and time-response curves of the in vitro studies a two-tail unpaired T-test was performed to compare the control and treated groups at each time point with a confidence interval of 99%.

Betacyanins content
The amount of betacyanins present in the Opuntia robusta and Opuntia streptacantha fruit extracts are shown in Table 1. Opuntia robusta fruit extract had a significantly higher concentration of betacyanins (2.21 fold; P < 0.01) compared to Opuntia streptacantha fruit extract suggesting a more potent biological activity of Opuntia robusta Values represent the mean of three different measurements ± SD. ¶ P < 0.01 vs Opuntia streptacantha.  fruit extract treatment against APAP-induced hepatotoxicity in comparison to Opuntia streptacantha fruit extract.

Characterization of Opuntia extracts by HPLC
Betacyanins, specifically betanin and its isomer (isobetanin) are the most abundant antioxidant-related components in the Opuntia cactus fruit extracts. No other phenolic acids (280 nm) or flavonoids (360 nm) with comparable intensity were identified (Supplemental Figs. 1 and 2). The chromatograms from Opuntia robusta (Fig. 1B) and Opuntia streptacantha (Fig. 1C) at 535 nm were compared to the chromatogram of the betanin standard (Fig. 1A). The standard showed a retention time of 18 and 20 min for betanin and isobetanin, respectively. Both Opuntia extracts showed two peaks at the same retention time as the betanin standard confirming the presence of betanin and isobetanin. In both extracts there were no additional peaks in the whole chromatogram (55 min) ensuring that betanin and isobetanin are the major components in these extracts. The amount of both betanin and isobetanin appeared to be higher in the Opuntia robusta extract as compared to the Opuntia streptacantha extract.

Biochemical markers of liver damage
The levels of the biochemical markers of hepatic injury in plasma and tissue homogenates are shown in Fig. 2 There was a significant decrease of 83.3% of GSH content in liver tissue of the APAP group (131.72 ± 6.25 µg/g) compared to the control group (788.59 ± 28.75 µg/g) (P < 0.05) (Fig. 2-E). Treatment with Opuntia robusta and streptacantha fruit extracts preserved the GSH content in liver tissue of APAP-intoxicated rats with a non-significant reduction of 34.1% and 15.4% compared to the control group ( Fig. 2-E). Treatment with Opuntia extracts alone did not induce alterations in the total GSH content. NAC was also effective in maintaining the levels of hepatic GSH in the APAP-treated group with a nonsignificant reduction of 8.9% compared to the control group (Fig. 2-E).
APAP intoxication induced a significant increase of MDA levels of 119.4% (130.35 ± 10.34 nmol/100 mg) in liver tissue as compared to the control group (59.40 ± 2.75 nmol/100 mg) (P < 0.05) (Fig. 2-F). Treatment with Opuntia fruit extracts and NAC after APAP intoxication significantly reduced (P < 0.05) levels of MDA (53.6% for APAP + Or, 47.3% for APAP + Os, and 44.9% for APAP + NAC groups) to control levels ( Fig. 2-F). Opuntia extracts alone did not change the levels of MDA compared to the control group.

Relative mRNA expression of oxidative stress-related genes
After 6 h of APAP intoxication, liver tissue was collected to quantify the relative mRNA expression of the main antioxidant enzymes (Sod2, Hmox1, Gclc) and the cell survival promotor Gadd45b (Fig. 3).

Histopathology
APAP intoxication induced significant hydropic degeneration (cellular edema) and focal necrosis in the hepatocytes near the central vein (centrilobular) (Fig. 4-B). In addition, the normal structure of hepatic parenchyma (polygonal form of the cells and hepatic sinusoids) was disrupted in the APAP group (Fig. 4-B) compared to the control group which showed a normal architecture of liver (Fig. 4-A). Treatment with Opuntia extracts (Fig. 4-E,F) or NAC (Fig. 4-G) after APAP intoxication reduced focal necrosis and ballooning degeneration of the central hepatocytes (centrilobular) of the hepatic acinus (zone III). Opuntia extracts alone did not induce alterations in the morphology of the hepatic lobule (central area) (Fig. 4-C,D). Opuntia robusta treatment appeared to be more protective than Opuntia streptacantha and NAC with respect to the histopathological changes induced by APAP.

In vitro experiments
3.4.1. LDH leakage Primary hepatocytes exposed to a single dose of APAP showed significant LDH release (3.8 fold increase) into the medium after 24 h compared to the control group (Table 2). Therapeutic treatment with Opuntia robusta or Opuntia streptacantha after APAP intoxication significantly reduced LDH release to control levels indicating improved survival compared to the APAP group (P < 0.05) ( Table 2). Opuntia extracts alone did not induce liver cell death after 24 h of exposure (P > 0.05 vs control). NAC treatment was also effective in protecting the hepatocytes against APAP-induced cell death and significantly reduced LDH release compared to the APAP group (P < 0.05) ( Table 2).

Sytox green stain
Cell membrane disruption and necrotic cell death induced by APAP was confirmed using the cell-impermeable fluorescent dye SYTOX® Green. As shown in Fig. 5, necrotic cell death was dramatically increased 24 h after APAP intoxication compared to the control group (Fig. 5-A). Therapeutic treatment with Opuntia extracts (Fig. 5-E,F) or NAC (Fig. 5-G) considerably reduced necrotic cell death in primary hepatocytes exposed to APAP compared to the APAP group (Fig. 5-B). Treatment with Opuntia extracts alone did not alter membrane permeability of the primary hepatocytes ( Fig. 5-C,D).

Relative mRNA expression of oxidative stress-related genes
The mRNA level of Sod2 did not change up to 24 h after APAP exposure but was significantly reduced (62.2%) after 24 h of intoxication compared to the control (P < 0.05) (Fig. 6-A). The mRNA levels of antioxidant enzymes Hmox1 and Gclc were significantly increased (766 and 328%, respectively) after 24 h of APAP intoxication (P < 0.05) ( Fig. 6-B,C). mRNA level of the cell stress sensor Gadd45b gradually increased after exposure to APAP and peaked (197%) at 24 h after APAP exposure (Fig. 6-D).
Opuntia extracts and NAC displayed diverse effects on the APAPinduced changes in oxidative stress-related genes: therapeutic treatment with Opuntia robusta and Opuntia streptacantha fruit extracts of APAP-intoxicated hepatocytes restored Sod2 expression (42.2 and 43.6% vs APAP group), whereas therapeutic treatment with NAC did not restore Sod2 expression. Interestingly, Opuntia robusta and Opuntia streptacantha fruit extracts alone induced Sod2 expression compared to controls (74.2 and 130%, respectively). With regard to Hmox1, Opuntia extracts, in contrast to NAC (91.1% vs control group), did not attenuate the APAP-induced increase of Hmox1 (762% for APAP + Or; and 613% for APAP + Os vs control group). Opuntia extracts alone moderately, but not significantly, increased Hmox1 expression compared to control. Yet another effect was observed for Gclc: Opuntia robusta and Opuntia streptacantha fruit extracts further increased the APAP-induced increase of Gclc (789 and 939% vs control group, respectively; or, 107.7 and 119.6% vs APAP group, respectively), whereas NAC attenuated 196% the APAP-induced increase of Gclc expression. Opuntia extracts alone did not significantly change (P > 0.05) Gclc expression compared to the control group. Finally, both Opuntia robusta and Opuntia streptacantha fruit extracts, and NAC tended to significantly attenuate (94, 151 and 116%, respectively) the APAP-induced increase of Gadd45b expression (Fig. 7).

Discussion
Opuntia spp. fruits contain many bioactive components with potential health benefits but the exact composition is dependent on physical, chemical, geographical and environmental factors. Thus, it is important to identify the main bioactive compounds that are responsible for the potential protective mechanisms.
In this study we quantified spectrophotometrically the betacyanin content and determined by HPLC analysis that betalains, specifically betacyanins, are the most important components in extracts of Opuntia robusta and Opuntia streptacantha fruits. In our previous study, we quantified betalains, flavonoids, ascorbic acid and total phenolics in Opuntia robusta and Opuntia streptacantha fruit extracts by spectrophotometry and reported that betalains are the second major component after total phenolics (González-Ponce et al., 2016). In support, (Stintzing et al., 2005), reported that betacyanins are the second major group of components after total phenolics in the fruits of Opuntia ficusindica clones, although it is important to remark that betacyanins such as betanin and its isomer might be detected as phenolic compounds due to the presence of a phenolic ring in their structure. They identified betanin and isobetanin as the most abundant betacyanins in these clones, although they also identified additional betacyanins such as gomphrenin I, betanidin and neobetanin. (Serra et al., 2013), showed that betacyanins are the major components in hydroalcoholic extracts obtained from Opuntia ficus-indica and Opuntia robusta.
Our results demonstrate the hepatoprotective effect of therapeutic treatment with betacyanin-rich Opuntia purple fruit extracts against APAP-intoxication both in vivo and in vitro. The protective effect is mainly due to the reduction of oxidative stress induced by the free radical NAPQI. In vivo, Opuntia extracts reduced the biochemical markers of liver damage; diminished the hepatic levels of malondialdehyde and restored the levels of glutathione, indicating diminished oxidative stress; and improved the hepatic architecture, specifically at the centrilobular region (zone III of the hepatic acinus) where the expression of the CYP2E1 isoform is highest and APAP is biotransformed into the electrophilic metabolite NAPQI causing most damage in this region (Abdelmegeed, Moon, Chen, Gonzalez, & Song, 2010). In vitro, treatment with Opuntia extracts reduced LDH leakage into the medium and Sytox green nuclear staining, indicating reduced necrotic cell death. Of note, our results indicate that the treatment with Opuntia extracts may have therapeutic value, since the protective effect of the extracts was observed when administered after APAP intoxication, both in vivo and in vitro. We have previously demonstrated the protective effect of the prophylactic consumption of both extracts (González-Ponce et al., 2016). In addition, the protective effect appeared to be at least as effective as observed with NAC, the currently used treatment for APAPinduced acute liver failure, with Opuntia robusta being slightly more protective than Opuntia streptacantha.
The antioxidant status of cells is dependent on many factors, including several oxidative stress-related enzymes like mitochondrial superoxide dismutase 2, heme oxygenase 1 and the rate-limiting enzyme in glutathione synthesis, glutamate-cystein ligase.
Superoxide dismutases (SOD) play a key role in the protection against reactive oxygen species (ROS). They catalyze the conversion of superoxide anions (O 2 .-) into hydrogen peroxide (H 2 O 2 ) and oxygen (O 2 ). Two types of SOD enzymes (Sod1 and Sod2) are distinguished: cytoplasmic Sod1 and mitochondrial Sod2 (Wang, Branicky, Noë, & Hekimi, 2018). (Chen et al., 2015), described that increasing the activity of Sod2 reduced glycochenodeoxycholic acid (GCDCA)-induced mitochondrial oxidative stress in rat hepatocytes. On the other hand, Sod2 has also been related to tumorigenicity, both as a tumor suppressor and as tumor promotor (Hempel et al., 2011). Both Hmox1 and Gclc are inducible target genes of the oxidative stress-responsive transcription factor Nfe2l2. Gclc plays an important role in the synthesis of GSH. (Botta et al., 2006), demonstrated that overexpression of Gclc in transgenic animals protects the liver against APAP-induced liver injury. (Kay et al., 2010), reported that the treatment with ajoene, a component of garlic, increased GSH content through Nfe2l2 activation and induction of Gclc, protecting HepG2 cells and hepatocytes against oxidative stress. Hmox1 is another Nfe2l2-regulated antioxidant enzyme. It is an ubiquitous stress-responsive enzyme with several functions in tissue homeostasis (Kim et al., 2011). We have previously shown that overexpression of the oxidative stress-responsive enzyme Hmox1 protects hepatocytes against apoptosis via inhibition of superoxide anion- induced JNK activity (Conde de la Rosa et al., 2008). (Chiu, Brittingham, & Laskin, 2002), described that Hmox1 is an important antioxidant enzyme in the protection against APAP-induced hepatotoxicity. Although these oxidative stress-related genes are important, little is known about their role and regulation during APAP intoxication and their regulation by natural products.
In the present study, APAP alone increased the gene expression of Hmox1, Gclc and Gadd45b in vivo and in vitro. Expression of Sod2 was increased in vivo and decreased in vitro by APAP. The induction of Hmox1 and Gclc is in accordance with exposure to oxidative stress and their regulation by the oxidative stress responsive transcription factor Nfe2l2. In contrast, Sod2 is not an exclusive target gene of Nfe2l2 since it has been described that its expression is also modulated by the transcription factors NFkb and Sp1-dependent p53. p53 is a tumor suppressor protein and is known to modulate cell survival and apoptotic pathways. p53 target genes are involved in cell proliferation (e.g. Gadd45) and apoptotic cell death (e.g. Fas, Bax) (Vogelstein, Lane, & Levine, 2000). (Dhar et al., 2010) observed that gene expression of Sod2 is regulated in a dose-dependent manner by p53 via the transcription factors NFkb and Sp1. They propose that p53 has bi-directional effects leading to either cell survival or cell death by suppressing or activating target genes like Sod2. At present, the explanation for the opposite regulation in this study of Sod2 in vivo and in vitro is not clear, although it is very likely that the presence of other liver cell types in the in vivo situation, including inflammatory cells with activated NFkb (cytokine release) and abundant ROS production, lead to a different response in the regulation of Sod2. It should also be noted that for the mRNA expression studies, RNA was isolated under non-lethal conditions, both in vivo and in vitro.
Opuntia extracts alone enhanced the cytoprotective defenses by significantly increasing the expression of Sod2 in vivo and in vitro.
These results indicate that the Opuntia extracts not only contain compounds that scavenge reactive oxygen species, but also contain factors that actually increase the expression of antioxidant genes.
Therapeutic treatment with Opuntia extracts prevented the APAPinduced increase of Sod2, Hmox1 and Gclc mRNA expression in vivo. However, Opuntia extracts exerted divergent effects in vitro: although they normalized Sod2 expression, they did not attenuate the APAP-induced increase in Hmox1 expression and even further increased the APAP-increased expression of Gclc. The reason for these divergent effects may be that oxidative stress induces the expression of oxidative stress-related genes and therefore, antioxidants attenuate these changes, but that in this case components in the Opuntia extracts modulate the expression of these genes independent of their ROS scavenging effects.
Finally, Gadd45 is a family of genes which are induced in response to (patho)physiological stresses. Gadd45 proteins have important functions as regulators of the cell cycle, cell survival or apoptosis, DNA repair and genomic stability (Ueda, Kohama, Kuge, Kido, & Sakurai, 2017). Gadd45b is an early predictor of liver dysfunction and stress  (Tian et al., 2011). (Papa et al., 2004), demonstrated the cytoprotective effect of Gadd45b via the activation of NFκB and the capacity to bind and block MKK7 an essential activator of pro-apoptotic JNK signaling. In addition, Gadd45b knock-out mice show decreased hepatocyte proliferation and increased programmed cell death after partial hepatectomy compared to wildtype mice (Papa et al., 2008). A recent study showed that APAP toxicity induced Gadd45b expression, which was further increased by the protective agent metformin and reduced JNK phosphorylation. Finally, increased cell death and sustained JNK phosphorylation was detected in primary hepatocytes with Gadd45b deficiency after sub-toxic doses of APAP (Y.-H. Kim et al., 2015). Together, these data indicate that Gadd45b is not only a sensor for cellular stress but also protects against cellular stress. In our study, we observed that APAP induced Gadd45b expression both in vivo and in vitro and that Opuntia extracts alone did not modulate Gadd45b expression. These results are in line with Gadd45b being a sensor of cellular stress. In addition, both in vivo and in vitro, Opuntia extracts reduced APAP-induced Gadd45b expression, again in line with Gadd45b being a sensor of cellular stress and Opuntia extracts relieving APAP-induced stress.

Conclusion
In conclusion, we observed a therapeutic effect of Opuntia robusta and Opuntia streptacantha against APAP-induced hepatoxicity. Opuntia robusta appeared to be slightly more protective, probably due to the higher amount of betacyanin compounds than Opuntia streptacantha. In addition, the Opuntia extracts were at least as potent as NAC in the protection against APAP-induced hepatotoxicity. Furthermore, in addition to scavenging reactive oxygen species, we show that Opuntia extracts modulate the expression of important oxidative stress-related genes at the transcriptional level. In the current study, the therapeutic action of Opuntia extracts was investigated 30 min after APAP intoxication. Further studies are required to investigate whether more delayed administration of the extracts is effective as well. In addition, it will be interesting to investigate whether Opuntia extracts protect against other hepatotoxic drugs (e.g. diclofenac), non-drug hepatoxicity like bile acids (cholestatic liver diseases) or fatty acid-induced lipotoxicity (non-alcoholic steatohepatitis). Finally, studies are required using purified components of the extracts to confirm the identity of the protective agents in order to facilitate clinical application.

Declaration of Competing Interest
The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper.
Aguascalientes and the University of Groningen for supporting this study.

Funding
This work was supported by the Graduate School of Medical Sciences (GSMS) of University of Groningen; and National Council of Science and Technology Mexico (CONACYT) [grant number 336940].