Mycotoxins are an extremely diverse group of environmentally persistent compounds produced by fungi, which when ingested, inhaled or absorbed from environmental sources, can cause adverse health effects or even death in humans and animals [1] . They can contaminate various agricultural commodities, especially maize and wheat, either before harvest or under post-harvest conditions [2] [3] and considered as the most important chronic dietary risk factor, higher than synthetic contaminants, food additives or pesticide residues [4] . Aflatoxin B₁ (AFB₁) and other naturally occurring aflatoxins (AFs) have been classified as group 1 human carcinogen because of their role in aetiology of liver cancer [5] , notably among subjects who are carriers of hepatitis B virus surface antigen (HBsAg) [6] . AFs are considered to be genotoxic carcinogens [7] . The FAO/WHO Joint Expert Committee on Food Additives (JECFA) concluded that the exposure of even <1 ng/kg body weight could contribute to the risk of liver cancer [8] . Moreover, ochratoxin A (OA), another mycotoxin, is classified in group 2B as possible human carcinogens [5] . Consequently, mycotoxin contamination is becoming one of the most insidious challenges to food safety.

Previous studies have demonstrated that multiple mycotoxins can co-contaminate crops and foods intended for both animal and human consumption [9] . Individual mycotoxicosis occurs seasonally on certain areas that hinder an implementation of an effective prophylactic measure [10] . However, interactions between given mycotoxins are still unclear [11] . The presence of a mixture of these toxins may present a problem in terms of determining clinical symptoms of an individual mycotoxicosis. Concomitant exposure to mycotoxins, such as aflatoxin and fumonisin, has been associated with various teratogenic, mutagenic, estrogenic, neurogenic and immunotoxic effects, as well as growth faltering, cancer, and even death in acute incidences [5] .

A new practical and effective strategy for reducing food-borne exposure to mycotoxins is the inclusion of various binding agents or sorbents in the diet to adsorb mycotoxins in the gastrointestinal tract of animals and reduce bioavailability, and toxicity. Montmorillonite, bentonite, and hydrated sodium calcium aluminosilicate (HSCAS), as anticaking agents for animal feed, have been reported to prevent disease associated with aflatoxicosis in farm animals, including chicks, turkey poults, and pigs [12] and laboratory animals [11] . Montmorillonite belong to the structural family called 2:1 phyllosilicates, which present a structure composed by two tetrahedral layers formed by Si and O atoms, fused with an octahedral layer with aluminum and magnesium atoms bonded to oxygen and hydroxyl groups [13] . Organically modified nanoclay (OMNC) is the clay modified with organic surfactants. These hydrophobic materials have attracted much interest because they have found wide applications. Organoclays have also been tested for treating ground and surface water and for other toxic organic chemicals from pharmaceuticals and pesticides industries. Organoclays can offer dramatic performance improvements in many other adsorption applications, including removing oil; grease; heavy metals; and polychlorinated biphenyl; organic matter; such as humic and fulvic acids; poly-nuclear and polycyclic aromatics; and sparingly soluble hydrophobic; chlorinated organics. Removing radio-nuclides, including pertechnetate, from water is another application with tremendous potential [14] . The aim of the current study was to develop organo-modified montmorillonite for the protection against the combined toxicity of AFB₁ and OA in rats.

XRD of OMNM is presented in Figure 1. The XRD pattern of OMNM was recorded and basal spacings of 19.80 Å was observed. The expansion in the basal spacing of the montmorillonite due to the intercalation of CTAB was calculated as Δ d = d ? 12.60 Å (where d is the basal spacing of the CTAB-treated clay and 12.60 Å is the thickness of a clay layer) and it was found to be 7.20 Å.

The morphology of OMNM was carried out using SEM (Figure 2). SEM photomicrographs of nano mont-

morillonite modified with CTAB indicated that the physical appearance of the clay particles changed significantly. Gathered agglomerations with severely curled or crumpled structures were formed much more easer in OMNM. The grain boundaries were steadily disappeared and the flakes of clay minerals are dispersed in the matrices. The laser particle size distribution of OMNM was carried out using SFM-1 Desk Top Planetary Ball Miller (MTI) for 3 hours. The results revealed that the average size of OMNM was 120 nm (Figure 3).

The biological evaluation of OMNM to protect against mycotoxins toxicity revealed that ALT, AST, ALP and G-GTP were significantly increased in the groups treated with AFB₁ and/or OA and this increase was more pronounced in the group received the combined treatment (Table 1). Treatment with OMNM alone has no significant effect on ALT and AST however; it caused a significant decrease in ALP and a significant increase in G-GTP compared to the control levels. Treatment with OMNM plus AFB₁ and/or OA resulted in significant improvements in all the tested parameters toward the control levels and it succeeded to normalize AST and G-GTP in the group received OA.

The effects of different treatments on kidney function parameters (Table 2) revealed that both the mycotoxins increased uric acid, urea and createnine although OA has severe toxicological effect on kidney function compared to AFB₁. The combined treatment with AFB₁ and OA showed synergistic toxicological effect than the single treatment. Treatment with OMNM did not affect uric acid or createnine however; it induced a significant decrease in urea. Treatment with OMNM succeeded to normalize uric acid and createnine in the groups treated with AFB₁ or OA and createnine in the group treated with AFB₁ plus OA and significantly decreased the level of uric acid and urea in the group treated with the combined mycotoxins (Table 2).

The data presented in Table 3 indicated that treatment with AFB₁ and/or OA resulted in a significant decrease in serum TAC and hepatic GSH and CAT. This decrease was pronounced in the AFB₁ alone-treated group and was more pronounced in the group treated with AFB₁ plus OA compared to OA-treated group. Treatment with OMNM alone did not induce any significant effect on serum TAC and hepatic CAT however; it induced a significant increase in hepatic GSH. Animals treated with AFB₁ and/or OA plus OMNM showed a significant im-

Within each column, means superscript with different letters are significantly different (P < 0.05).

provement in the antioxidant enzymes activities in serum and hepatic tissue although these parameters were still significantly lower than the control group.

The effect of different treatments on oxidative stress markers (NO and MDA) in liver tissue and tumor marker (AFP) in serum are presented in Table 4. These results showed that both the mycotoxins induced a significant

Within each column, means superscript with different letters are significantly different (P < 0.05).

Within each column, means superscript with different letters are significantly different (P ≤ 0.05).

Within each column, means superscript with different letters are significantly different (P ≤ 0.05).

increase in MDA, NO and AFP levels compared to the control group. The increase in these parameters was pronounced in the AFB₁-treated group and more pronounced in the group received the combined treatment of AFB₁ and OA. Treatment with OMNM did not induce any significant changes in these parameters. However, OMNM succeeded to normalize AFP in the groups treated with AFB₁ and OA and the groups treated with AFB₁ or OA accompanied with a significant improvement in NO in the group treated with AFB₁ plus OA and hepatic MDA in the groups treated with the single or combined mycotoxins (Table 4).

The aforementioned biochemical results were confirmed by the histological examination for the liver and kidney tissues. The microscopic examination of the liver of the control animals revealed normal architecture of hepatic lobule (Figure 4(a)). The liver of AFB₁-treated rats showed thickening in the wall of the portal tract with necrosis, fibrosis and bile duct proliferation with fatty droplets in the hepatocytes around the central vein (Figure 4(b)). Animals treated with OA showed pronounced hepatic histological changes around the enlarged portal tracts than the central vein in the form of necrosis obliteration or proliferation in the bile ducts surrounded by aggregation of fibrous tissues (Figure 4(c)). However, animals treated with AFB₁ plus OA showed disorganization of hepatic cords and vascuolar destruction, bile duct proliferation and aggregation of inflammatory cells (Figure 4(d)).

The microscopic examination of liver sections from rats treated with OMNM showed normal radiating hepatocytes and central veins (Figure 5(a)). However those treated with OMNM with AFB₁ showed maintenance of cellular integrity in the central area around the central vein without any inflammation and foci of regenerating nodule (Figure 5(b)). The liver of animals treated with OMNM plus OA showed regeneration in hepatocytes with normal acidophilic cytoplasm and central vein although local foci of the inflammatory cells are seen (Figure 5(c)). However, the liver of animals treated with AFB₁ and OA plus OMNM showed same picture of normal hepatocytes and prominent decrease in the different abnormal archetictuer (Figure 5(d)).

The histological examination of the kidney cortex of the control animals showed normal renal corpuscle with parietal layer of Bowman’s capsule, glomerulus, preserved renal space and normal proximal and distal convoluted tubules were also seen (Figure 6(a)). The examination of the sections in the kidney cortex of AFB₁-treated rat showed shrunken or atresia in some renal corpuscle and tubular necrosis with pyknosis of their epithelial cells were formed. The same group showed tubular necrosis and obliteration, the epithelial cells in some tubules were disintegrated or pyknotic. The glomerular capillaries were expanded, less cellularity and vacuolated (Figure 6(b)). The kidney cortex of the rats treated with OA showed an increase in number of shrinking or damaged renal corpuscle and foci of tubular necrosis and dilatation with epithelial cells pyknosis (Figure 6(c)). However, the kidney section of the rats treated with AFB₁ plus OA showed marked increase in tubular necrosis, vacuolation and obliteration with more cleared pyknotic epithelial cells. The same sections showed that the necrotic tubules were scattered others had vacuolar degeneration with pyknotic nuclei, interstitial haemorrhage and fibrous tissues were also seen with hyperplastic changes glomeruli (Figure 6(d)).

The kidney cortex of the rats treated with OMNM showed less affected tubules, nearly normal (Figure 7(a)). However, animals treated with AFB₁ plus OMNM showed disappearance of the intracytoplasmic vacuolation in renal tubules and decrease in foci of necrosis. The glomeruli showed hyperplastic changes with narrowing of their capsular spaces (Figure 7(b)). Whereas, kidney cortex of animals treated with OA plus OMNM showed improvement in renal tubules in the form of disappearance of the intracytoplasmic vacuolation in renal tubules or necrosis and increase in size of renal corpuscle with dilated capillaries and narrowing of their capsular spaces (Figure 7(c)). Moreover, animals treated with AFB₁ and OA plus OMNM showed improvement in most of renal tubules and renal corpuscle where the capillary tufts were surrounded by capsular spaces (Figure 7(d)).

The expansion in the basal spacing of the montmorillonite due to the intercalation of CTAB was 7.20 Å. This

morphological observation of SEM suggested that CTAB ion molecules intercalate in to the interlayers of montmorillonite with a monolayer arrangement. The shifting of these peaks to lower 2θ and increase in d-spac- ing confirms the increase in gap between clay platelets and diffusion of Cetyl-trimethylammoniumbromide (CTAB) into the layer of silicates [15] . These changes in the morphologies and particle sizes indicate that the intercalation was accompanied by adsorption. In general, the loading of surfactant onto clay is a “self-assembly” process [18] .

Regarding to the biological evaluation of OMNM, it is documented that the combined toxicity of mycotoxins is very hard to predict because it is influenced by several factors, including chemistry and mechanism of action, toxicodinamics and toxicokinetics, experimental design and endpoints of the study as well as statistical aspects [19] . This means that multiple mycotoxins can affect certain targets and initiate more than one event in the cell machinery leading to extremely complicated cell response [20] . The typical clinical picture of a disease is a result of mixed intoxication and interactions between mycotoxins [11] .

The novel technology applied for protecting against mycotoxins toxicity is the utilization of adsorbents mixed with foods which are supposed to bind efficiently mycotoxins in the gastrointestinal tract. Clay eating has been recorded from traditional human societies and is considered ‘culturally acceptable’ in many African countries and China [21] . The adsorption and ion exchange using natural, synthetic and modified inorganic and organic solids have been explored [22] . In the current study, we evaluated the protective effects of the organic modified nano montmorillonites (OMNM) against the toxicity resulted from the exposure to AFB₁ and/or OA in rat model. The selected doses of AFB₁ and OA were based on our previous work [23] [24] however; the selected dose of OMNM was based on our in vitro study (unpublished, data). Animals received the two mycotoxins singly or in combination showed a significant increase in ALT, AST ALP and G-GTP activities. This increase in transaminases in mycotoxins-treated animals is indicative for changes in the hepatic tissues and biliary system [11] . However, the increased activity of ALP may indicate degenerative changes and hypofunction of the liver and the elevated activity of G-GTP in serum is consistent with severe injury of both liver lysosomes and mitochondria [23] . Moreover, there is some evidence that OA induces bone changes which may also be a contributory factor to the increased ALP activity [24] [25] . On the other hand, the significant increase in uric acid, urea and createnine observed in the animals treated with AFB₁ and/or OA may indicate protein catabolism and/or renal dysfunction [23] [24] . These results clearly indicated that both AFB₁ and OA have stressful effects on the hepatic and renal tissues, consistent with those reported in the literature of mycotoxicosis [26] .

AFP is considered specific biomarker for liver cancer and it is synthesized mainly in the fetal stage; practically no production of this marker occurs in the normal adult. However, when some adult cells are transformed to cancer cells, the synthesis of AFP commences again. In the current study, the elevated serum level of AFP in the animals treated with AFB₁ and/or OA indicated that both agents are potent hepatocarcinogen, enhance reactive oxygen species (ROS) formation and causes oxidative DNA damage, which may play a role in their carcinogenicity [27] [28] . Therefore, the current study affirmed that AFB₁ and/or OA can induce hepatotoxicity and regeneration in liver cells in rats as indicated by the elevation of AFP level in serum. Similar to the current observations, AFB₁ administration resulted in the elevation of serum AFP level in both ducks [29] and rats [27] and in OA-treated rats [23] [24] .

In the current study, NO was found to be increased significantly in the animals treated with AFB₁ and/or OA. Although the role of NO in cell death is complex, Moon and Pyo [30] stated that NO is produced by macrophages and it plays an important role in tumor conditions. The generation of NO by the inducible nitric oxide synthase (iNOS) plays a key role in the cytokine-mediated cell destruction [31] . Consequently, the increased in NO level reported herein in the animals treated with the mycotoxins suggested that these mycotoxins preferentially affect macrophage functions [11] .

The current results also revealed a decrease in TAC and GSH in the liver and catalase in serum of AFB₁ and/ or OA-treated rats which are might indirectly lead to an increase in oxidative DNA damage [32] . Moreover, the reduced level of TAC may be explained by the association of glutathione peroxidase (GPX) with AFB₁ or its metabolites [32] . Several studies on the mechanisms of mycotoxins-induced liver injury have demonstrated that glutathione and TAC play an important role in the detoxification of the reactive and toxic metabolites of these mycotoxins, and the liver necrosis begins when the glutathione stores are almost exhausted [33] . Similar observations have been reported in weaned piglets received low doses of AFB₁ and OA [10] . Moreover, Gautier et al. [34] stated that OA does evoke oxidative stress, which may contribute at least in part to OA renal toxicity and carcinogenicity in rats during long-term exposure. Moreover, OA was found to decrease GSH and TAC which may be explained by the conjugation of GSH with OA or its metabolites [25] .

It is well documented that LP is one of the main manifestations of oxidative damage and it has been found to play an important role in the toxicity and carcinogenicity. However, the antioxidant enzymes represent the major defense system against liver injury and carcinogenesis. Several reports indicated that exposure to either AFB₁ or OA increased LP in liver. In the current study, AFB₁ and/or OA administration enhanced LP as indicated by the significant increase in MDA level which directly results of free radical-mediated toxicity [35] . In a previous works, Abdel-Wahhab et al. [23] [25] reported that free radicals are known to attack the highly unsaturated fatty acids of the cell membrane to induce LP which considered a key process in many pathological events and is one of the reactions induced by oxidative stress [36] . Another mechanism for OA-induced injury was suggested by Pfohl-Leszkowicz et al., [37] who reported that the ability of OA to generate free radicals and to enhance lipid peroxidation has been linked to the genotoxicity expressed by DNA adduct formation and to the disturbance of calcium homeostasis due to an impairment of the endoplasmic reticulum membrane. The cellular damage induced by ROS was estimated by monitoring the lipid peroxidation (LP), which is a well-known indicator of cellular damage by oxidative stress. Furthermore, it is well documented that AFB₁ is metabolized by the mixed- function oxidase system to a number of hydroxylated metabolites and to AFB₁ 8,9-epoxide, which binds to DNA, forming covalent adducts [38] . Also AFB₁ is known to produce membrane damage through increased lipid peroxidation [33] . However, OA appears to produce many of the effects in the cell such as the increase of the permeability of the cell to Ca²⁺ [39] . Both the enhanced cellular concentration of Ca²⁺ and the presence of the prooxidant OA uncouple oxidative phosphorylation resulting in an increased leakage of electrons from the respiratory chain. This generates O²⁻ and hence H₂O₂ lack of an adequate supply of NAD (P) H and GSH to permit H₂O₂ consumption by the GSH dependent glutathion peroxidase and NAD (P) H dependent glutathion reductase. Furthermore, an increased concentration of free iron within the cell stimulates the production of OH via the Fenton like reaction due to mobilization of Fe²⁺ by Ca²⁺. This results in further cell damage and may be one of the mechanisms that OA exerts its toxic effects [24] . The biochemical results were further confirmed by the histological examination of the liver tissue. The microscopic examination of the liver tissues revealed severe histological changes typical to those reported in the literature of mycotoxicoses [23] .

In the current study, treatment with OMNM plus AFB₁ and/or OA resulted in a significant improvement in all the biochemical parameters tested and the histological picture of the liver. Several reports suggested that the novel technology applied for protecting animals against mycotoxins toxicity is the utilization of adsorbents mixed with foods which are supposed to bind efficiently mycotoxins in the gastrointestinal tract. The adsorption and ion exchange using natural, synthetic and modified inorganic and organic solids have been explored [22] . Clay minerals, which are currently used as an anti-caking agent, may remove mycotoxins from aqueous solutions [40] . The absorbability of different clay minerals differs according to their specific surface area [41] . These clay minerals act as potential ionic exchangers for mycotoxins due to their low cost, high abundance, easy manipulation, and harmlessness to the environment. On the other hand, the addition of different clays to diet did not show any health risk to human or in laboratory and farm animals [42] . In previous works, the dietary inclusion of clay minerals at a level of 0.5% (w/w) did not show any toxic effect regarding biochemical, haematological and immunological parameters in laboratory animals [43] . Furthermore, the addition of calcium montmorillonite did not show any toxic effects in the equilibrium of vitamins and gain minerals in the human blood [44] .

The current results clearly indicated that rats treated with montmorillonite nanoclay plus AFB₁ and/or OA showed a significant improvement in all the parameters tested and support the hypothesis that montmorillonite nanoclay bind the AFB₁ and/or OA in the gastrointestinal tract of the animals and the montmorillonite-myco- toxins complex is stable and does not affected by the different metabolizing enzymes. In this concern, Hassan et al. [45] suggested that montmorillonite may posse three types of active binding sites: 1) those located at basal planes within interlayer channels, 2) those located on the surface and 3) those located at the edges of clay particles. Moreover, previous reports indicated that montmorillonite has the property of adsorbing organic substances either on its external surfaces or within its inter laminar spaces by the interaction with or substitution for the exchange cations present in their spaces [40] .

Although there is a lot of evidence for the good technological performance of nanocomposites, safety issues are also of importance. Available data on clay’s toxicity is still scarce, but different authors have already described toxic effects induced by montmorillonite and organoclays [46] . However, these toxic effects were suggested to be due to the modifier used to synthesize the organoclay [46] . Therefore, in the present study, the safety use of montmorillonite-based nanoclays modified with CATB as a commercial surfactant to protect against AFB₁ and/or OA in vivo was studied. The results showed no toxic effects for OMNM itself when used at 0.5% (w/w) in rat diet. In this concern, Houtman et al. [47] reported that montmorillonite-based nanoclays had no cytotoxicity on HepG2 cell lines although these authors suggested that the type of modifiers is very importance to improve the compatibility with the polymer matrix. Others studies have evaluated the toxicity of the commercial non modified montmorillonite in the same cell lines [46] . The same authors observed that HepG2 and Caco-2 exposed to organo modified clay did not present higher significant reductions of viability with respect to the controls in the range of concentrations assayed. Moreover, Sharma et al. [48] did not obtain any cytotoxic effects in Caco-2 exposed to organo modified clay. In this case, the concentration used of the modified clays showed the same behavior compared to that of the unmodified montmorillonite reported in our previous work [45] , indicating that the modifiers employed could not involve changes in the safety profile of the modified clay.

In the case of organoclays, the oral pathway is the most important entrance route for the consumers, and they should know the possible effects of the ingestion of these nanosubstances to the gastrointestinal tract [49] . There are a limited number of toxicological studies in the literature about commercial modified clays. Baek et al. [50] evaluated the toxicity effects in human normal intestinal cells (INT-407) in a short and long term exposure, 24, 48, 72 h, and, 10 days to montmorillonite. Thereby, a decrease in cell proliferation showed at all times assayed. On the one hand, significant differences in the short term assays were found above 100 μg/mL concentration levels, on the other hand, a significant inhibition of normal colony formation in the long term was observed at all concentrations tested. Even though, alterations in LDH release were only observed at the highest concentrations at 48 and 72 h. Also, the oligo (styrene-co-acrylonitrile)-modified clay montmorillonite showed an increased LDH release activity and cell viability reduction at a concentration of 1 g/L in mouse embryonic fibroblast (NIH 3T3) cells and human embryonic kidney 293 (HEK 293) cells [51] . The implication of oxidative stress, inflammation or DNA damage, among others, could be related to micro and nanoparticle exposure [52] .

For this reason, levels of NO and GSH were assayed in the current study, obtaining insignificant differences with respect to the control group on the GSH levels in the liver. In this concern, Sharma et al. [48] reported that modified clays did not induce ROS production in Caco-2. Furthermore, Baek et al. [50] reported the evaluation of ROS production in INT-407 cells exposed to montmorillonite at the highest concentration (1000 μg/mL) at all three time points assayed (24, 48 and 72 h). Moreover, it was reported that organo modified montmorillonite did not induce leakage of IL-6, biomarker of an inflammatory response, in any of the cell lines [47] . Montmorillonite, on the other hand, has been reported to rapidly lyse neutrophils and erythrocytes in vitro. Furthermore, it can stimulate chemiluminescence, the neutrophil oxidative metabolic burst [53] .

From the results of the current study, it can be concluded that both AFB₁ and/or OA at the tested doses induce severe hepatic toxicity. Treatment with AFB₁ reveals that this toxin is hepatonephrotoxic however; it is mainly targeted to the liver whereas, treatment with OA revealed that it is also hepatonephrotoxic but it is mainly targeted to the kidney. The co-occurrence of the two mycotoxins suggests the synergistic and adding-up interactions of AFB₁ and OA. OMNM is safe itself and it succeeds to eliminate and even prevent the toxicity of these mycotoxins. Moreover, OMNM is suggested a promise candidate for the protection against multi-mycotoxins in the highly incidence area.

This work was supported by the National Research Centre (Dokki, Cairo, Egypt) project # 10070112 and the Faculty of Pharmacy, Cairo University.

The authors declare that there are no conflicts of interest.