Dual Role of Dietary Curcumin Through Attenuating AFB1-Induced Oxidative Stress and Liver Injury via Modulating Liver Phase-I and Phase-II Enzymes Involved in AFB1 Bioactivation and Detoxification

It is well understood that liver cytochrome p450 enzymes are responsible for AFB1 bioactivation, while phase-II enzymes regulated by the transcription factor nuclear factor-erythroid-2-related factor 2 (Nrf2) are involved in detoxification of AFB1. In this study, we explored the potential of curcumin to prevent AFB1-induced liver injury by modulating liver phase-I and phase-II enzymes along with Nrf2 involved in AFB1 bioactivation and detoxification. Arbor Acres broiler were divided into four groups including control group (G1; fed only basal feed), curcumin alone-treated group (G2; 450 mg/kg feed), AFB1-fed group (G3; 5 mg/kg feed), and curcumin plus AFB1 group (G4; 5 mg AFB1+450 mg curcumin/kg feed). After 28 days, liver and blood samples were collected for different analyses. Histological and phenotypic results revealed that AFB1-induced liver injury was partially ameliorated by curcumin supplementation. Compared to AFB1 alone-treated group, serum biochemical parameters and liver antioxidant status showed that curcumin supplementation significantly prevented AFB1-induced liver injury. RT-PCR and western blot results revealed that curcumin inhibited CYP enzymes-mediated bioactivation of AFB1 at mRNA and protein level. Transcription factor Nrf2, its downstream genes such as GSTA3, and GSTM2 mRNA, and protein expression level significantly upregulated via dietary curcumin. In addition, GSTs enzyme activity was enhanced with dietary curcumin which plays a crucial role in AFB1-detoxification. Conclusively, the study provided a scientific basis for the use of curcumin in broiler’s diet and contributed to explore the multi-target preventive actions of curcumin against AFB1-induced liver injury through the modulation of phase-I and phase-II enzymes, and its potent anti-oxidative effects.


INTRODUCTION
Recently, several reports demonstrated that aflatoxin B 1 (AFB 1 ) is probably the most common researched mycotoxin in the world (Liu and Wu, 2010;Abrar et al., 2013). Many Aspergillus species such as Aspergillus flavus and Aspergillus parasiticus produce AFB 1 (Nakai et al., 2008). It is considered to be one of the most harmful and potent aflatoxins among all other known mycotoxins (Ferenčík and Ebringer, 2003). AFB 1 causes many adverse effects such as mutagenicity, carcinogenicity, hepatotoxicity, and immunotoxicity in humans, poultry and in many other animal species (Stettler and Sengstag, 2001;Kalpana et al., 2012). AFB 1 has the ability to affect more than one system at a time and cause serious alterations in the normal homeostasis of the affected organism (Ferenčík and Ebringer, 2003;Herzallah, 2013). It has been placed in Group-1 human carcinogen by the International Agency for Research on Cancer (IARC) (Zuckerman, 2006). AFB 1 leads to carcinogenicity in humans by p53 gene mutation at codon 249 and formation of DNA adducts (Aguilar et al., 1993(Aguilar et al., , 1994. Previous reports showed that the liver is the major target organ of AFB 1 , as the liver plays a crucial role in drug or toxin metabolism and detoxification in the body (Grant et al., 2001;Rawal et al., 2010). AFB 1 causes liver toxicity including infiltration of mononuclear and heterophilic cells and proliferation of bile duct along with changes in liver morphology, such as increase in liver weight, enlargement, congestion, necrosis, and pallor discoloration (Newberne and Butler, 1969;Hussain and Anwar, 2008), but the susceptibility of a species to AFB 1 depends on various factors such as type of breed, immunity and nutritional status, age, sex, dose, length of exposure to AFB 1 , and managemental conditions (Richard, 2007).
The use of naturally occurring phytochemicals to combat human diseases and prevent toxicity is of growing scientific and public interest due to their potential therapeutic uses (Sharma et al., 2005). Curcumin is a polyphenol commonly composed of three types of curcuminoids, bisdemethoxycurcumin, demethoxycurcumin, and diferuloylmethane, derived from the rhizome of Curcuma longa (Common name: turmeric) famous for its pharmacological and therapeutic properties (Anand et al., 2008). Several reports have documented the anti-inflammatory, anti-oxidative, anti-cancerous, and potent chemotherapeutic properties of curcumin (Aggarwal et al., 2003;Joe et al., 2004;Liew and Mohd-Redzwan, 2018;Limaye et al., 2018;Rauf et al., 2018). It has been proved earlier that curcumin is effective against AFB 1 -induced oxidative stress El-Bahr, 2015). Moreover, previous studies demonstrated that curcumin modulated phase-I enzymes activity and has the potential to control AFB 1 -induced carcinogenesis and toxic effects (Firozi et al., 1996;Appiah-Opong et al., 2007;Nayak and Sashidhar, 2010). Cytochrome p450 (phase-I) enzymes are responsible for the biotransformation of AFB 1 into its harmful metabolites (Yunus et al., 2011). Earlier reports showed that several CYP enzymes (CYP1A1, CYP1A2, CYP1A5, CYP2A6, CYP3A37, and CYP3A4) were involved in bioactivation of AFB 1 into its harmful electrophilic metabolite aflatoxin-8,9-epoxide (AFBO) and to AFM 1 , AFP 1 , AFQ 1 , etc. (Gregorio et al., 2015). AFBO, the most potent metabolite may quickly react with DNA to form DNA adducts or cause mutations (Aguilar et al., 1993), or conjugated with phase-II enzymes such as glutathione S-transferases (GST) (Johnson et al., 2014). Importantly, the sensitivity of a species to AFB 1 is not only related to AFB 1 bioactivation via CYP enzymes but also related to detoxification enzymes (phase-II enzymes) (Klein et al., 2000). It is well known that the transcription factor nuclear factor-erythroid-2-related factor 2 (Nrf2) plays a crucial role in regulating several pathways, including drug metabolism, antioxidant, and detoxification such as phase-II enzymes like superoxide dismutase, catalase, NAD(P)H:quinine oxidoreductase-1, uridine 5-diphospho (UDP) glucuronosyltransferase, heme oxygenase-1 (HO-1), glutathione S-transferase, glutathione peroxidase (GPx), thioredoxin, and uridine 5-diphospho (UDP)glucuronosyltransferase (Li et al., 2008). Besides, Slocum and Kensler's (2011) studies demonstrated that up-regulation of Nrf2-pathway and its downstream genes (GSTs) prevented chemical carcinogenesis . Moreover, Nrf2 plays an important cytoprotective role in disease prevention, especially at the initiation stage of cancer development (Krajka-Kuźniak et al., 2017). Thus, inhibition of phase-I enzymes (involved in AFB 1 -bioactivation) and activation of Nrf2-pathway and its downstream genes (related to AFB 1 -detoxification) could be a novel pharmacological approach for the prevention of AFB 1induced liver injury and oxidative stress in broilers. The present study was designed to shed light on the multi-target effects of curcumin through inhibition of phase-I enzymes, and activation of Nrf2 as well as phase-II enzymes against AFB 1 -induced liver injury and ameliorating oxidative stress in Arbor Acres broiler liver. The study contributed to provide a biochemical basis for the supplementation of curcumin in broiler's diet along with its chemopreventive action against AFB 1 -induced liver injury.

Ethics Statement
All the experiments were performed according to animal ethics guidelines and approved protocols of the Harbin Veterinary Research Institute of the Chinese Academy of Agricultural Sciences [SYXK (Hei) 2012-2067].

Experimental Chickens, Diets, and Sample Collection
One-day-old healthy Arbor Acres broiler chickens were bought from Yi Nong commercial hatchery (Registration number; 230108799294096, Heilongjiang, China). The broilers were divided in four groups (15 chicks/group) and assigned to different cages. Chickens were provided with 12 h light and 12 h dark cycle and fed ad libitum for 28 days. Four dietary groups were: (G1) control group; fed only basal diet, (G2) curcumin control group: basal diet incorporated with 450 mg curcumin/kg feed; (G3) AFB 1 group: basal diet contaminated with 5.0 mg AFB 1 /kg feed; and (G4) curcumin+AFB 1 group: basal diet incorporated with 5.0 mg AFB 1 and 450 mg curcumin/kg feed. After 28 days, chickens were first weighed, euthanized with sodium pentobarbital, and sacrificed in an antiseptic environment. Liver was collected, weighed and stored at −80 • C for further analysis.

Serum Biochemical Analysis
Serum biochemical markers were measured to evaluate liver toxicity. 5 ml blood sample was collected from the vein wing before slaughtering of chickens. Serum was separated after centrifugation at 3000 g for 1.5 h and biochemical analysis was carried out at the same time. Serum samples were analyzed for alanine aminotransferase (ALT), aspartate aminotransferase (AST), alkaline phosphatase (ALP), and gamma-glutamyl transferase (GGT) activities in an automatic biochemical analyzer (Model 7080, Japan Hitachi Limited).

Histological Observation
Liver morphology was assessed by histopathological examination. Fresh liver tissue pieces were stained with hematoxylin and eosin (HE) as previously described (Wang X.H. et al., 2016). In brief, liver sections (1 cm × 1 cm) were fixed in 10% formalin for 24 h. The fixed liver tissues were embedded in paraffin after processed through a series of graded ethanol and dimethyl benzene. 4-µm tissue sections were mounted on glass slides, and finally stained with HE dye for microscopic examination (Light microscope; Nikon E100, 40× magnification).

Measurement of Antioxidant Enzyme Activities
A piece of fresh liver tissue (0.4-0.5 g) was immediately taken out after slaughtering and chilled in 0.85% NaCl at 0 • C. It was then dried and homogenized as previously described (Soetikno et al., 2013). The homogenized solution was then centrifuged (12,000 × g for 15 min) at 4 • C and the supernatants were analyzed for antioxidant activity. All the kits were bought from Nanjing Jiancheng Bioengineering Factory (China). GSH-Px (Glutathione peroxidase, kit; A005), SOD (Superoxide dismutase, kit; A001-1) and MDA (Malonaldehyde, kit; A003-1) activities were measured in the supernatant according to the instructions of the kit.

RNA Extraction and Real-Time Quantitative Polymerase Chain Reaction (qPCR)
TRIzol (Invitrogen Inc., Carlsbad, CA) reagent was used to isolate total RNA . RNA quality was verified by measuring the 260/280 ratio (Wang J. et al., 2016). Toyobo first strand cDNA synthesis kit (Osaka, Japan) was used for the synthesis of cDNA. The target genes and β-actin loading control gene specific primers are given in Table 1. The kit bought from Toyobo Company Ltd. (Japan) was utilized for RT-PCR reaction in a Roche LightCycle instrument (Shanghai, China). The data were analyzed according to 2 − C t method (Livak and Schmittgen, 2001).

Western Blot Analysis
Western blotting was performed as described previously (Lu et al., 2017). In brief, fresh liver tissues were homogenized in RIPA and PMSF. 5 × SDS-PAGE loading buffer was added, followed by protein degeneration via boiling for 5 min. The loaded protein were separated by 10 or 12% SDS-PAGE, and transferred to nitrocellulose membrane (NC). The membranes were incubated for 1 h in non-fat dry milk in TBST. CYP1A1 Inc., Canada) protein was detected using specific primary antibodies incubated overnight at 4 • C. Secondary anti-mouse or anti-rabbit horseradish peroxidase conjugated IgG was used for 1 h, and immune-complexes were detected using enhanced chemiluminescence (ECL) detection. Signals were visualized on a computer using Image J software (National Institute of Health, United States).

Statistical Analysis
The data significance were analyzed by one way ANOVA followed by LSD test using the statistical package for social science (SPSS, version 21.0), and p < 0.05 was considered statistically significant. The experiments were performed at least three times unless otherwise mentioned. The data are represented as means ± standard errors. GraphPad Prism (Version 6.01) was used to draw all the graphs with standard error bars.

Curcumin Prevented AFB 1 -Induced Liver Injury
Liver sections from control group (Figure 1; G1) and curcumin control group (Figure 1; G2) showed normal morphology. Dietary AFB 1 induced severe liver injury (Figure 1; G3) compared to control and curcumin control group. Severe to moderate lesions such as hepatic granular degeneration, lymphocyte infiltration, and hepatic steatosis were observed in AFB 1 -fed group. The supplementation of curcumin in the diet partially ameliorated AFB 1 -induced hepatic lesions (Figure 1; G4). These results confirmed the preventive effects of curcumin against AFB 1 -induced liver injury.
The changes in serum biochemical parameters were inconsistent with the liver histological analysis (Figure 1). In addition, body weight and liver weight were recorded to study the changes in phenotypic effects (Table 2). Notably, compared to control ( Table 2; G1) and curcumin control group (Table 2; G2), body weight significantly (p < 0.05) decreased in AFB 1 -fed group ( Table 2; G3). Moreover, liver weight significantly (p < 0.05) decreased in AFB 1 -fed group. However, the supplementation of dietary curcumin ( Table 2; G4) reversed AFB 1 -induced changes in body and liver weight significantly (p < 0.05).

Inhibition of Phase-I Enzymes
The effect of curcumin, AFB 1, and curcumin+AFB 1 diet was investigated at the mRNA and protein level. Figure 3 shows four major CYP enzymes (CYP1A1, CYP1A2, CYP3A4, and CYP2A6) mRNA expression levels. Compared to control (Figure 3; G1) and curcumin control group (Figure 3; G2), AFB 1 (Figure 3; G3) significantly (p < 0.05) upregulated the mRNA expression level of the four CYP enzymes. However, curcumin supplementation (Figure 3; G4) in diet significantly (p < 0.05) down-regulated AFB 1 -induced increase in CYP enzymes. Protein expression levels (Figure 4) showed the same trends as mRNA expression levels. Compared to control (Figure 4; G1) and curcumin control group (Figure 4; G2), a significant (p < 0.05) increase has been noted in all CYP enzyme protein levels in AFB 1 -fed group (Figure 4; G3). While curcumin supplementation (Figure 4; G4) significantly (p < 0.05) decreased the protein expression levels of CYP1A1, CYP1A2 and CYP3A4, the decrease is not statistically significant for CYP2A6 (p > 0.05) protein expression. The data clearly showed that CYP1A1, CYP1A2, CYP3A4 and CYP2A6 actively involved in AFB 1 bioactivation. However, curcumin supplementation significantly down-regulated the expression level of these enzymes at mRNA and protein level, and hence inhibited CYP enzymes-mediated bioactivation of AFB 1 .

Hepatic GST Enzyme Activity
GST enzymes play an important role in the detoxification (conjugation) of AFB 1 metabolites. Figure 6B displays GST enzyme activity detected in liver cytosol. It has been noted that AFB 1 -supplemented diet (G3) significantly (p < 0.05) reduced GST enzyme activity compared to control group (Figure 6B; G1) and curcumin control group (Figure 6B; G2). However, GST activity was markedly (p < 0.05) elevated in curcumin+AFB 1 -fed group (Figure 6B; G4) relative to AFB 1fed group (Figure 6B; G3). These findings showed that curcumin actively increased GST enzyme activity to catalyze the harmful metabolites of AFB 1 .

DISCUSSION
Numerous studies reported that AFB 1 is a well-known hepatotoxic and hepatocarcinogen and affects many systems at a time (Rawal et al., 2010). The current study showed that AFB 1 significantly modulated serum biochemical parameters. Liver histological observations showed abnormal morphological signs in AFB 1 -treated group, while oxidative stress parameters clearly revealed that AFB 1 induced oxidative stress in hepatocytes. Liver weight markedly increased and body weight was significantly suppressed in the current study in AFB 1 -fed group. These abnormal findings related to AFB 1 toxicity were in general agreement with previous studies (Kumar and Balachandran, 2009;Ibrahim, 2013;El-Bahr, 2015;Muhammad et al., 2017). Notably, our data showed that dietary supplementation of curcumin ameliorated AFB 1 -induced oxidative stress and reversed serum biochemical changes. AFB 1 -induced abnormal morphological signs in the liver partially disappear with curcumin treatment. The above findings were in agreement with some earlier reports that curcumin significantly mitigated AFB 1 -induced toxicity in broiler liver (Gowda et al., 2008). It has been reported earlier that CYP1A1, CYP1A2, CYP2A6, and CYP3A4 are responsible for the bioactivation of AFB 1 in poultry species (ducks, quails, turkeys and chickens) (Diaz et al., 2010a,b). In this study, liver CYP1A1, CYP1A2, CYP3A4, and CYP2A6 were significantly upregulated in AFB 1fed group showing its involvement in AFB 1 biotransformation into harmful metabolites. Interestingly, the inhibition of these enzymes may be a promising treatment to prevent AFB 1induced liver toxicity. Importantly, the use of curcumin markedly reduced the mRNA and protein expression levels of the selected CYP enzymes in broiler liver in vivo. Our previous study demonstrated that curcumin inhibited CYP2A6 enzyme activity (in a dose dependent manner) involved in AFB 1 -bioactivation . Zhang et al. (2016) also reported that curcumin significantly inhibited CYP enzymes-mediated bioactivation of AFB 1 into AFBO. Similarly, it has been noted that AFB 1 -induced increase expression of  chicken CYP2H1 and CYP1A1 was alleviated with dietary curcumin (Yarru et al., 2009). Moreover, curcumin attenuated Benzo[a]Pyrene-induced oxidative stress and reduced DNA adducts in mice via inhibiting CYP1A1/1A2 activities (in vivo) (Garg et al., 2008), decreased CYP3A activity in rats (Kim et al., 2015), and modulated CYP1A2, CYP3A4, CYP2B6, CYP2C9, and CYP2D6 activity in humans (Appiah-Opong et al., 2007). Thus, curcumin could possibly prevent AFB 1 -induced hepatotoxicity though the synergistic effects of increased antioxidant capacities and inhibition of the cytochrome enzymes mediated activation to AFBO. However, the mechanism behind cytochrome P450 enzymes inhibition is still unknown and further molecular studies are needed to know the exact molecular mechanism of curcumin-mediated CYP450 inhibition. Transcription factor Nrf2 is an attractive target in toxin chemoprevention due to its cytoprotective response and combat oxidative stress leading to prevention of DNA damage, carcinogenesis, and removal of toxic metabolites from the body (Slocum and Kensler, 2011). Previous reports demonstrated that curcumin is a potent inducer of Nrf2-pathway and its downstream genes involved in many cytoprotective mechanisms (Sahin et al., 2012;Soetikno et al., 2013). Nrf2 plays a crucial role in chemoprevention and reduce liver injury via up-regulating GST enzymes responsible for the detoxification of AFB 1 harmful metabolites such as AFBO (Poapolathep et al., 2015), and meanwhile upregulate antioxidant defense system (Soetikno et al., 2013). The nuclear translocation of AFB 1 is associated with many cellular proteins carrying signals for its translocation and subsequent binding to DNA, which results in DNA damage or causes deterioration of cellular proteins (Ricordy et al., 2002). Intriguingly, AFB 1 -induced impairment of specific protein functions or DNA damage decreased cell survival in cultured cells (Besaratinia et al., 2009). Previous studies demonstrated that protein biosynthesis is weakened due to the formation of AFB 1 -DNA adducts. In addition, AFB 1 could change the activity of enzymes such as cyclic nucleotide phosphodiesterase, protein kinase, and Ca 2+ -ATPase, alter hepatic protein phosphorylation FIGURE 6 | Western blot analysis of Nrf2, GSTA3, and GSTM2 involved in AFB 1 -detoxification. (A) Nrf2, GSTA3, and GSTM2 protein expression level of the four treated groups shown. The intensity of the protein bands was quantified by densitometry analysis. The levels of GAPDH gene were used as internal standard control. The values represented as mean ± SE (n = 3). (B) Cytosolic GST enzyme activity (U/mg protein) of the four treated groups presented in the figure. Treatments represented as (G1) control group: fed only basal diet, (G2) curcumin control group: basal diet + 450 mg curcumin/kg feed, (G3) AFB 1 group: basal diet + 5.0 mg AFB 1 /kg feed and (G4) curcumin+AFB 1 group: basal diet + 5.0 mg AFB 1 + 450 mg curcumin/kg feed. The value of P < 0.05 was considered statistically significant. * p < 0.05 vs. control group (G1), * * p < 0.05 vs. curcumin control group (G2) and * * * p < 0.05 vs. AFB1 group (G3). (Mistry et al., 1996;Ricordy et al., 2002), and causes p53 mutation which may develop carcinogenesis in experimental animals (Aguilar et al., 1994). Similarly, our data showed that Nrf2 mRNA and protein expression level significantly downregulated in AFB 1 -fed group. Consequently, the downstream genes of Nrf2 such as GSTA3, and GSTM2 mRNA and protein expression levels markedly decreased in AFB 1 -fed group. Hence, it is speculated that the reduction in protein level could be due to AFB 1 -induced impairment in protein synthesis in broiler hepatocytes. Meanwhile, curcumin supplementation reversed AFB 1 -induced decrease in Nrf2, GSTA3, and GSTM2 mRNA and protein expression level. Furthermore, curcumin markedly increased GST enzyme activity involved in detoxification (conjugation) of AFB 1 . Similarly, previous studies reported that GST enzymes play a major role in prevention of AFB 1induced toxic effects and carcinogenesis (Larsson et al., 1994), and could prevent AFB 1 -induced hepatic carcinoma in fisher F344 rats via Nrf2-pathway (Eaton and Schaupp, 2014). We found that curcumin could inhibit cytochrome P450 enzymesmediated bioactivation of AFB 1 along with activation of Nrf2-pathway. In addition, our data showed that curcumin alleviated oxidative stress and caused alteration in AFB 1 -induced changes in serum biochemical parameters. These evidences suggested that curcumin could effectively inhibit AFB 1 -induced toxicity in broilers liver. Recently, many studies reported the chemopreventive effects of curcumin against AFB 1 -induced liver injury in several experimental animals (Larsson et al., 1994;Gowda et al., 2008;Yarru et al., 2009;Poapolathep et al., 2015;Zhang et al., 2016). However, these studies only investigated the preventive effects of curcumin through phase-I enzymes or phase-II enzymes against AFB 1 -induced liver injury. In contrast, in this study, we scrutinized the multi-target effects of curcumin regulating both phase-I enzymes and phase-II enzymes via Nrf2pathway involved in AFB 1 bioactivation and detoxification. It has been noted from our previous study that 450 mg/kg curcumin provided maximum amelioration and significantly inhibited CYP2A6-mediated AFB 1 bioactivation . Therefore, in this study, we selected the dose of curcumin (450 mg/kg feed) to explore its preventive effects against dietary AFB 1 at a dose of 5 mg/kg feed. However, previous studies reported that curcumin is regarded a safe compound at a dose of 12 g/day (Basak et al., 2015), acts as anti-inflammatory, antitumor, and anti-oxidant, etc. (Abdel-Wahhab, 2015), regulates pharmacological activities in a direct or indirect way and modulates various signaling molecules including cell survival proteins, apoptosis, drug resistance protein, etc. (Basak et al., 2015;Abdel-Wahhab et al., 2016), and possess the ability to scavenge free radicals (Trujillo et al., 2013). Our data also showed that curcumin prevented liver dysfunction due to its antioxidant activity as well as modulating drug metabolizing enzymes including liver phase-I enzymes, Nrf2, and phase-II enzymes. In addition, higher studies are needed to investigate the preventive effects of curcumin and AFB 1 -induced undesirable changes in other cellular pathways such as cell cycle arrest, autophagy, apoptosis and cell death.

CONCLUSION
In conclusion, we showed that dietary AFB 1 negatively altered serum biochemical parameters, reduced activities of liver antioxidative enzymes, suppressed Nrf2 and phase-II enzymes involved in AFB 1 detoxification, and markedly increased the expression level of CYP enzymes involved in AFB 1 biotransformation, while dietary curcumin significantly reversed AFB 1 -induced alterations in serum biochemical parameters and enhanced liver antioxidant enzyme activities. Moreover, curcumin inhibited CYP enzymes-mediated bioactivation of AFB 1 and upregulated Nrf2 defense system including GST enzyme activity. Therefore, the above novel findings suggested that dietary curcumin is recommended as a prophylactic measure to prevent AFB 1 -induced hepatocellular toxicity and oxidative stress.

AUTHOR CONTRIBUTIONS
XZ supervised all the experiments. IM contributed to paper writing. HW, XS, and MH performed the practical work and completed the experiments. XW, ZL, PC, and MAH provided help during experiments.

FUNDING
The National Natural Science Foundation of China (registration number: 31172369) and the Natural Science Foundation of Heilongjiang Province (registration number: ZD201405) provided funding for this work.