Interactions between aflatoxin B1 and dietary iron overload in hepatic mutagenesis

Toxicology 234 (2007) 157–166

Interactions between aflatoxin B1 and dietary iron overload in hepatic mutagenesis George A. Asare a , Michelle Bronz a , Vivash Naidoo b , Michael C. Kew a,∗ a

MRC/University Molecular Hepatology Research Unit, Department of Medicine, University of the Witwatersrand, Medical School, 7 York Road, Parktown 2193, Johannesburg, South Africa b Department of Anatomical Pathology, National Health Services Laboratories, Chris Hani Baragwana Hospital, Johannesburg, South Africa Received 15 December 2006; received in revised form 8 February 2007; accepted 16 February 2007 Available online 23 February 2007

Abstract Background/aim: Dietary aflatoxin B1 (AFB1 ) exposure and iron overload are important causes of hepatocellular carcinoma in sub-Saharan Africa. The aim of this study was to investigate if the two risk factors have an interactive effect. Methods: Four groups of Wistar albino rats were studied for 12 months. Group 1 (control) was fed the normal chow diet; group 2 (Fe) was supplemented with 0.75% ferrocene iron; group 3 (Fe + AFB1 ) was fed 0.75% ferrocene throughout and gavaged 25 ␮g AFB1 for 10 days; group 4 (AFB1 ) was gavaged 25 ␮g AFB1 for 10 days. Iron profile, lipid peroxidation (LPO), 8-hydroxydeoxyguanosine (8OHdG), oxidative lipid/DNA damage immunohistochemistry, superoxide/nitrite free radicals, cytokines IL6, IL-10, transaminases (ALT/AST) and Ames mutagenesis tests were performed. Results: LPO and ALT showed a significant (p < 0.05)/additive effect and 8OHdG a significant (p < 0.05)/multiplicative effect in the Fe + AFB1 group. IL-6 produced a negative synergy as against an additive antagonistic effect with IL-10. Massive deposits of 4hydroxynonenal (4-HNE) and 8OHdG were observed in liver sections of the Fe + AFB1 group, suggestive of multiplicative synergy. Significant levels of mutagenesis (p < 0.001) were observed in the Fe + AFB1 group. This multiplicative synergy was five-fold. Conclusion: Dietary iron overload and AFB1 have a multiplicative effect on mutagenesis. © 2007 Elsevier Ireland Ltd. All rights reserved. Keywords: Iron overload; Aflatoxin B1 ; Mutagenesis; Synergy; p53 mutation

1. Introduction The principal causes of the hepatocellular carcinoma (HCC) that occurs with such high frequency in eastern and south-eastern Asia and sub-Saharan Africa are chronic hepatitis B virus (HBV) infection and the high levels of dietary exposure to the mycotoxin, aflatoxin B1

∗ Corresponding author. Tel.: +27 11 488 3628; fax: +27 112 643 4318. E-mail address: [email protected] (M.C. Kew).

(AFB1 ). A multiplicative synergistic interaction occurs between the hepatocarcinogenic effects of these two risk factors (Kew, 2003). During recent years an additional environmental hepatocarcinogen has been identified in sub-Saharan Africa. Dietary iron overload occurs in some rural areas of the sub-continent, with a prevalence that may be as high as 10% (Gordeuk et al., 1986). It results from the consumption of large volumes over time of iron-rich traditional alcoholic beverages that are home-brewed in cast-iron pots or drums. As in the betterknown iron storage disease hereditary hemochromatosis (HH), the metal accumulates in its main storage site,

0300-483X/$ – see front matter © 2007 Elsevier Ireland Ltd. All rights reserved. doi:10.1016/j.tox.2007.02.009

158

G.A. Asare et al. / Toxicology 234 (2007) 157–166

the liver (Bothwell and Bradlow, 1960). Excess hepatic iron as a consequence of dietary iron overload has been shown to carry a relative risk of developing HCC of 10.6 (95% confidence intervals 1.5; 76.8) with a population attributable risk of 29 in one study (Mandishona et al., 1998) and relative risks of 23.5 (2.1; 225) (Gordeuk et al., 1996) and 3.1 (1.05; 9.4) (Moyo et al., 1998) in two others. The hepatocarcinogenicity of hepatic iron overload has been confirmed in an animal model (Asare et al., 2006a,b). Because contamination of certain foodstuffs by Aspergillus species and dietary iron overload may occur together in rural regions of sub-Saharan Africa, and because chronic HBV infection and HCC are more common in rural than in urban regions in most countries in the sub-continent (Kew, 1992), carcinogenic interactions both between AFB1 exposure and hepatic iron overload and between HBV infection and hepatic iron overload are possible. The purpose of this study was to ascertain in an animal model if exposure to AFB1 and dietary iron overload had an interactive effect on hepatic mutagenesis. 2. Materials and methods 2.1. Subjects, diet and experimental design Four groups of 20 Wistar albino rats were studied: group 1 was fed the standard chow diet (AIN-93G Formulation, manufactured by EPOL, South Africa and certified to be carcinogen-free); group 2 the standard chow diet supplemented with 0.75% ferrocene iron (dicyclopentadienyl iron), from Sigma (Germany); group 3 0.75% ferrocene iron throughout and gavaged 25 ␮g AFB1 for 10 days according to the protocol of Roebuck (2004); group 4 the standard diet and gavaged 25 ␮g AFB1 for 10 days. The rats were fed ad libitum. The Wistar albino rats received humane care in accordance with the guidelines of the Animal Ethics Committee of the University of the Witwatersrand. They were housed in plastic cages with stainless steel tops in the animal unit of the University’s Medical School, where room temperature, humidity and ventilation were controlled according to international standards. The rats were maintained at a 12-h light-cycle and were studied for 12 months. The rats were first anaesthetized with 0.1 ml/100 g wt. Anaket/Chinazin (4:1) for blood sampling and later sacrificed by euthanasia. Livers were harvested under aseptic conditions. One portion was snap frozen in liquid nitrogen (and later stored at −70 ◦ C) and the other, kept in 10% buffered formalin. 2.2. Assays 2.2.1. Determination of serum iron and non-transferrin-bound iron (NTBI) concentration Iron kit (Randox Laboratories, UK) was used to determine serum iron concentrations. For NTBI, a sensitive high liquid

chromatography (HPLC)-based method was used (McNamara et al., 1999). In brief, NTBI interacted with nitrilo-triacetic acid to form an iron-nitrilo-triacetic acid complex (iron-NTA), which was quantified by HPLC as a measure of NTBI. NTA does not compete with transferrin-bound iron. 2.2.2. Determination of blood superoxide (O2 •− ) concentration O2 •− levels were determined by the method of Kuldip et al. (1998), using heparinized whole blood. The blood was diluted (1:9) in phenol-free Hank’s balanced salt solution (HBSS). The 10 mM luminal stock was prepared in 0.1 M borate buffer and N-formylmethionyl-leucinylphenyl-alanine (fMLP). Stock 10 mM in DMSO was used as an agonist for the stimulation of free radicals from neutrophils. Basically, luminal reacts with an oxidizing species to produce larger and more measurable amounts of light at a peak wavelength of approximately 425 nm. The amount of O2 •− produced was measured by a four-channel luminometer interfaced with a computer. 2.2.3. Determination of nitrite (NO2 − ) concentration Nitrite assay was performed according to the Griess reaction and modified by Fiddler, 1977. In brief, nitrite reacts with sulphanilamide in the presence of N-1-napthylenediamine, under acidic conditions to form an azo compound that was measured spectrophotometrically at 550 nm. 2.2.4. Determination of total lipid hydroperoxide (LOOH) concentration Total lipid hydroperoxides (LOOH) were measured using the FOX II assay according to the protocol of Nourooz-Zadeh et al. (1994). Ferrous oxidation of xylenol orange (FOX) is based on the principle of the rapid peroxide-mediated oxidation of Fe2+ to Fe3+ under acidic conditions. The latter in the presence of xylenol orange, formed the Fe-xylenol orange complex that was measured spectrophotometrically at 560 nm. 2.2.5. Determination of 8OHdG concentration Hepatic levels of 8-hydroxy-2 -deoxy-guanosine (8OHdG) were determined using a commercial kit from the Japan Institute for the Control of Aging (Fukuroi, Japan). The 8OHdG test is a competitive in vitro enzyme-linked immunosorbent assay for the quantitative measurement of 8OHdG in tissue, serum and plasma. The test was performed according to the manufacturer’s instructions and the absorbance read at 450 nm on a microplate plate reader. 2.2.6. Determination of cytokine concentration A Bio-Plex cytokine commercial kit (BioRad, UK) was used. The BioPlex suspension array system uses multiplexing technology of up to 100 colour-coded beads conjugated to different specific reactants. Each reactant is specific for a different target molecule. Three different sets of colour-coated beads were used, each designated to IL-1␤, IL-6 and IL-10 detection. The cytokine assays were performed according to the manufacturer’s instructions.

G.A. Asare et al. / Toxicology 234 (2007) 157–166

2.2.7. Determination of AST/ALT/GGT levels AST and ALT concentrations were measured using an auto-analyser (Cobas Integra 400, Holliston, MA, USA) and kits from Roche Diagnostics (Hague Rd, Indianapolis, USA). GGT levels were determined colorimetrically using a commercial kit manufactured by Kat Medical, South Africa. 2.2.8. Ames mutagenicity test Ames mutagenicity test was performed according to the method of Maron and Ames (1983). The Salmonella typhimurium TA 100 and TA102 tester strains were used in the Ames reversion test. Assays were carried out on samples of fractionated liver homogenate, namely, whole homogenate, S9 fraction, nucleosomes, microsomes and the cytosolic fraction. All samples were assayed according to the standard plate incorporation test and challenged with S9-Mix. Dimethyl sulfoxide was used as a negative control whilst 10 ␮g/plate daunomycin (daunorubicin hydrochloride) and 1.5 ␮g/plate of sodium azide were used as positive controls for TA 102 and TA 100, respectively. 2.2.9. Immunohistochemistry Standard published protocols of Toyokuni et al. (1997, 1999), for the immunohistochemical detection of 8-hydroxydeoxyguanosine (8OHdG) were used. Trans 4-hydroxy-2 nonenol (4-HNE) was measured using the protocol of Ma et al. (1997). Paraffin embedded sections (3-␮m thick) were used. Briefly, slides were incubated at 56–60 ◦ C for 15 min for antigen retrieval. Sections were deparaffinized in two changes of xylene (5 min each) followed by two changes of absolute ethanol (3 min each), and briefly in water. Sections were incubated with 3% H2 O2 for 30 min to block endogenous peroxidases. This was followed by a similar wash in phosphate buffered saline (PBS). After incubation in PBS for 15 min, sections were blocked for monospecific binding of the antibody by incubation in 5% bovine serum albumin (BSA) for 15 min at room temperature. Sections were then covered in a humidified chamber with 300–500 ␮l of the rabbit primary antibody (5 ␮g/ml anti-8OHdG) in PBS, overnight at room temperature. PBS without primary antibody was used as negative control. Slides were rinsed with PBS and left in the buffer for 5 min. About 150 ␮l of secondary labelled polymer antibody [goat anti-mouse IgG present on a polymer (DakoCytomation, Denmark)] was applied onto the sections for at least 30 min at room temperature. Slides were washed with PBS (2×) and covered with 150 ␮l of DAB-substrate-chromogen system (3,3 -diaminobenzidine in chromogen solution from DakoCytomation, Denmark) for 5–30 min. Slides were gently rinsed and counter-stained with Meyer’s Haematoxylin solution for 10 min, followed by brief washing with running water. The assay procedure for 4-HNE detection was the same as 8OHdG. However, 2 ␮g/ml antiHNE was used as primary antibody [primary antibodies were obtained from the Japan Institute for the Control of Aging (Fukuroi, Japan)].

159

2.3. Statistical analysis The SAS statistical package was used. Statistical analysis was performed by independent Student’s t-test. A probability value of p < 0.05 was considered statistically significant. Analysis of variance (ANOVA) was performed by Bonferroni (Dunn) multiple comparison t-test. Pearson’s correlation coefficient was determined within groups. Multiple regression analysis was performed to determine predictive indicators of oxidative stress. The definition of additivity (Berenbaum, 1985; Loewe, 1953; Loewe and Muischnek, 1926) through the interaction index in a combination of c chemicals (here, c = 2), is represented by Eq. (1). Ei represents the concentration of the ith component alone and xi represent the concentration of the ith component in combination with the c agents that yields the same response: c  xi i=1

Ei

=1

(1)

If the left side of Eq. (1) is <1, then a greater-than-additive interaction occurs (e.g., synergism). If the left side of equation 1 is >1, then a less-than-additive interaction occurs (e.g., antagonism) (Crofton et al., 2005). The above was applied in a ration as follows after baseline subtraction of the control group’s value: [Fe] group + [AFB1 ] group : [Fe + AFB1 ] group

3. Results For clarity of results of bar charts: a = control group; b = Fe group; c = Fe + AFB1 group; d = AFB1 group. Additionally, in Figs. 1, 5 and 7: a = control group; b = Fe group; c = Fe + AFB1 group; d = AFB1 group. 3.1. Iron parameters Fig. 1 shows the serum iron/NTBI levels of rats at 12 months. b and c had the highest serum iron levels, significantly higher than a (p < 0.05 and p < 0.05, respectively) and d (p < 0.01 and p < 0.01, respectively). Similarly, b and c had significantly higher NTBI levels than a (p < 0.001 and p < 0.001, respectively) and d (p < 0.001 and p < 0.001, respectively). 3.2. Oxidative and nitrosative stress Fig. 2 shows the serum NO2 − levels. The level of NO2 − in c was significantly higher than the others. The p-values were as follows: a versus c (p < 0.001); b versus c (p < 0.01); d versus c (p < 0.01). With the superoxide (O2 •− ) anion, levels in the AFB1

160

G.A. Asare et al. / Toxicology 234 (2007) 157–166

Fig. 1. Serum iron and NTBI concentrations of the various groups at 12 months. A significant two-fold (p < 0.05) and seven-fold or more (p < 0.001) increase in serum iron and NTBI, respectively, was observed in b/c and b /c , respectively, compared to a and d.

group were significantly higher than the other groups (p < 0.05). The respective levels of O• 2 − production were as follows: control group (0.038 ± 0.038); Fe group (0.028 ± 0.018); Fe + AFB1 group (0.061 ± 0.064); AFB1 group (0.168 ± 0.134). All units were expressed as mV/Neutr × 109 (graph not shown). 3.3. Oxidative lipid damage From Fig. 3, b and c showed the greatest increase in LOOH in comparison to a and d. The interaction of the two toxins was additive (Table 1). c was significantly greater than b and d (p < 0.05 and p < 0.001,

Fig. 2. Serum nitrite levels of the various groups at 12 months. Nitrite level of c was significantly higher (p < 0.01) compared to the other groups.

Fig. 3. Serum lipid hydroperoxide (LOOH) levels in the various groups at 12 months. c had the highest LOOH concentration and was significantly greater than the other groups. Statistical differences between c vs. b and c vs. d were as follows: p < 0.05 and p < 0.001, respectively.

respectively). On the other hand, there were no significant differences in 8-IP levels of the Fe and Fe + AFB1 groups or Fe and AFB1 groups (data not shown). 3.4. Oxidative DNA damage The co-administration of the two toxins showed mark oxidative DNA damage (Fig. 4) with a multiplicative effect (Table 1). Significant differences (p < 0.001) were observed between a and c. Furthermore, statistical differences between c and b (p < 0.05) and, c and d were significant (p < 0.05).

Fig. 4. Oxidative DNA damage measured by 8OHdG levels in the various groups at 12 months. Significant differences (two-fold increase) were observed between a and c (p < 0.001). Furthermore, statistical differences between c and b and c and d were significant (p < 0.01 and p < 0.001, respectively).

G.A. Asare et al. / Toxicology 234 (2007) 157–166

161

Table 1 Summary of the interaction ratio between iron (Fe) and aflatoxin B1 (AFB1 ) with regards to the various test parameters S. No. 1 2 3 4 5 6 7 8 9 10 11 12

Test

Interaction ratio, (Fe + AFB1):(Fe) + (AFB1)

Comments

Serum nitrite

1.0:1.3

LOOH 8OHdG IL-6 IL-10 GGT AST ALT Ames test TA 100 Ames test TA 102 8OHdG immunohistochemistry 4-HNE immunohistochemistry

1.6:1 3.1:1 1.0:3.5 1.5:1 1.4:1 1.0:1 1.5:1 4.9:1 1.0:1.3 Microscopy—qualitative Microscopy—qualitative

Non-additive (due to the formation of peroxynitrite radical) Additive Multiplicative Negative interaction (see Fig. 5) Additive Additive Additive Additive Multiplicative Additive Multiplicative Multiplicative

3.5. Cytokines

3.6. Liver injury analysis

From Fig. 5, b and c had significantly lower IL-6 concentrations in comparison to a (p < 0.05 and p < 0.001, respectively). Differences between b and c were not significant. Furthermore, the interaction of the two toxins was negatively synergistic (Table 1). An inverse relationship was observed between IL-6 and IL-10 (Fig. 5). c had significantly raised levels in comparison to a (p < 0.01) and the effect of the two toxins was additive (Table 1). No significant differences were found between the c and d . Similarly, differences between the b and d were not significant. IL-1␤ results were not impressive (data not included).

A super additive effect of increased GGT was observed (Fig. 6, Table 1). b was significantly higher than a (p < 0.001) and d (p < 0.001). Similarly, c was significantly higher than a (p < 0.001) and d (p < 0.001). No significant differences were found between a and d, or between the b and c. Hepatocellular injury caused by iron and aflatoxin B1 is seen by markedly raised AST/ALT levels of b/b and c/c (Fig. 7). AST differences between b/c and a were statistically significant (p < 0.001 and p < 0.001, respectively). Co-administration of Fe and AFB1 had a fairly additive effect with respect to AST levels (Table 1). ALT level of c was also significantly raised in comparison to b (p < 0.001), and d (p < 0.001), reflecting significant hepatotoxicity of the two toxins.

Fig. 5. Cytokine levels (IL-6 and IL-10) in the various groups at 12 months. b and c had significantly lowered IL-6 levels compared to a (p < 0.05 and p < 0.001, respectively). An inverse relationship was observed with IL-10. c had significantly raised levels compared a and b (p < 0.01 and p < 0.01, respectively).

Fig. 6. Serum GGT levels of the various groups at 12 months. c compared to a and d was significant (p < 0.001 and p < 0.001, respectively) (b vs. c is not significant).

162

G.A. Asare et al. / Toxicology 234 (2007) 157–166

Fig. 7. Serum AST/ALT levels of the various groups at 12 months. b and c AST levels were significantly raised compared to a (p < 0.001 and p < 0.001, respectively). Similarly, d/b vs. c were statistically significant (p < 0.001 and p < 0.001, respectively). Furthermore, ALT levels of b /c vs. a were significant (p < 0.001 and p < 0.001, respectively) and d /b vs. c were also significant (p < 0.01 and p < 0.001, respectively).

ALT level of b was significantly raised compared to d (p < 0.01). 3.7. Mutagenesis A synergistic effect was observed in the mutagenesis test using S. typhimurium TA 100 bacteria strains (Fig. 8, Table 1). b did not show significantly higher mutagenesis levels compared to a and d. However c showed significant differences when compared to a, b and d (p < 0.001, p < 0.001 and p < 0.001, respectively). No significant differences were found between a and d,

or b and d. The negative and positive controls were 160 and 1230 colonies, respectively. The co-administration of the two toxins resulted in multiplicative synergism in the ratio of ≈5:1 for TA 100 (whole homogenate). However, an additive effect was observed using S. typhimurium TA 102 bacteria strains with respect to the co-administration of the two toxins (Table 1). The ironfed groups and the AFB1 group showed significantly higher mutagenesis levels in comparison to the control group (p < 0.05). However, no significant difference was observed between the Fe and AFB1 groups. There were significant differences between the Fe + AFB1 and Fe (p < 0.05) as well as Fe + AFB1 and AFB1 (p < 0.05) (no graph shown). The relative colony counts from the whole homogenate after background subtraction of 290 colonies (negative control) were as follows: control group (296 ± 17); Fe group (412 ± 6); Fe + AFB1 group (563 ± 50); AFB1 group (305 ± 5). The positive control read 670 colonies. 3.8. 8OHdG immunohistochemistry Photomicrographs 9a, 10a, 11a, and 12a represent 8OHdG immunohistochemistry of the control, AFB1 , Fe, and Fe + AFB1 groups, respectively, at 12 months. There was hardly any brown staining visible in Fig. 9a sections under microscopic observation, suggesting that in these tissues there was no significant detectable 8OHdG adducts. Brown stains indicating positive labelling of 8OHdG (arrowed yellow), were observed in the AFB1 and Fe liver sections (Figs. 10a and 11a). Few brown labelling was observed in these sections as well as granular labelling within the cells. However, intense 8OHdG labelling in aggregates was observed in the Fe + AFB1 sections especially around blood vessels. 3.9. 4-HNE immunohistochemistry In the control group liver sections (Fig. 9b) there was hardly any brown labelling observed after exposure to the anti-HNE antibodies. Brown cytoplasmic granular staining (arrowed yellow) was observed in the hepatocytes in the AFB1 , Fe, Fe + AFB1 groups of the rat liver at 12 months. Less intense and few deposits of 4-HNE staining were observed in the AFB1 and Fe sections (Figs. 10b and 11b, respectively). However, intense 4HNE staining of brown aggregates was observed in the Fe + AFB1 sections especially around vessels (Fig. 12b).

Fig. 8. Levels of mutagens produced and assessed by colony counts of spontaneous revertant colonies of S. typhimurium strain TA 100 in liver homogenates of the various groups at 12 months. c showed significant differences compared to all other groups (p < 0.001). Synergy was fivefold.

3.10. Correlations For the control group, AST was found to be positively correlated to serum NO2 − . For the Fe group, NO2 −

G.A. Asare et al. / Toxicology 234 (2007) 157–166

163

Figs. 9–12. Immunohistochemical detection of oxidative DNA and lipid damage in the livers of Wistar albino rats. Photomicrographs 9a and 9b for the control group are negative for 8OHdG and 4-HNE adducts, respectively. Photomicrographs 10a and 10b are mildly positive for 8OHdG and 4-HNE adducts, respectively, in the aflatoxin B1 group (group AFB1 ). Photomicrographs 11a and 11b are moderately positive for 8OHdG and 4-HNE adducts, respectively, in the iron-overloaded group (group Fe). Massive deposits of 8OHdG and 4-HNE adducts are seen in photomicrograghs 12a and 12b, respectively, in the group the received iron and aflatoxin B1 (group Fe + AFB1 ) suggesting multiplicative synergy.

164

G.A. Asare et al. / Toxicology 234 (2007) 157–166

levels were positively correlated with serum 8OHdG levels. Correlations for the Fe + AFB1 group showed that ALT positively correlated with 8OHdG levels and serum NO2 − levels were shown to positively correlate serum LOOH. The correlation matrix for the AFB1 group showed a positive correlation between serum AST and LOOH. However, serum NO2 − level was negatively correlated to O2 •− production. 4. Discussion The liver is the primary site for biotransformation of ingested AFB1 (Wild and Turner, 2002; Kensler et al., 2003; Sudakin, 2003). The parent molecule is harmless, but in the liver it is converted by members of the cytochrome P450 super-family to electrophilic intermediates. These metabolites are detoxified mainly by glutathione-S-transferase-mediated conjugation (Wild and Turner, 2002; Kensler et al., 2003; Sudakin, 2003). If, however, the quantity of AFB1 ingested in the diet exceeds the capacity of the phase II pathways to detoxify the mutagenic AFB1 -8,9-exo-epoxide formed or if, for any reason, the activity of these pathways is decreased (for example, by polymorphisms of the glutathione-Stransferase gene), the electrophilic metabolite accumulates and binds with high affinity to guanine bases in cellular DNA to form the 8,9-dihydro-8-(N7 -guanyl)-9hydroxy-AFB1 DNA (AFB1 -N7 -Guanine) adduct (Wild and Turner, 2002; Kensler et al., 2003; Sudakin, 2003). This adduct may give rise to guanine to thymine transversions in cellular DNA. The mechanisms responsible for a direct hepatocarcinogenic effect of iron have yet to be fully defined, although oxidative stress appears to be one mechanism (Asare et al., 2006b). Intracellular free iron is a catalyst for the formation of reactive oxygen and nitrogen species and consequently may cause oxidative damage to hepatocytes, DNA, protein, and lipids (Hagen et al., 1994; Jungst et al., 2004; Britton, 1996). Increased LPO is believed to be an important contributor to hepatocarcinogenesis in iron overload. 4-HNE, a major electrophilic by-product of lipid peroxidation caused by oxidative stress interacts with DNA to form exocyclic guanine products, which have been shown to be increased in a rat model of hepatocarcinogenesis (Marrogi et al., 2001). Deoxyguanosine residues in DNA are hydroxylated at the C8 position by hydroxyl radical or singlet oxygen to form 8OHdG, misreading of which results in guanine to thymine transversions (Ichiba et al., 2003; Cheng et al., 1992). Serum 8OHdG levels correlate well with serum iron levels. Both MDA and 4-HNE are genotoxic as well as cytotoxic (Cheng et al., 1992; Dabbagh et al., 1994).

The 4-HNE-DNA adduct is a potent mutagen in human cells and is preferentially formed at codon 249 of the p53 gene (Feng et al., 2004). 4-HNE also interacts with cellular repair proteins and may compromise DNA repair mechanisms (Feng et al., 2004). 8OHdG also correlates with the rate of DNA unwinding and strand breaks in liver tissue (Asare et al., 2006b). An association between DNA unwinding and the risk of HCC formation in HH has been described (Cheeseman, 1993). The bioavailability of iron in the experiment was not affected by co-administration of AFB1 , and the presence of iron overload in the rats receiving both agents was confirmed by the high levels of serum iron and NTBI. Similarly high iron values, accompanied by histochemical confirmation, have previously been reported with the same dietary iron regime (Asare et al., 2006a,b). The central regulator of iron homeostasis, hepcidin, is induced by the cytokines IL-6, IL-1␣ and IL-1␤ (Lee et al., 2004). Iron overload increases hepcidin production which serves to inhibit iron absorption and its release into circulation from macrophages. Hepcidin production is also influenced by inflammation, perhaps by depriving invading microorganisms of iron and inhibiting further iron absorption. In this study IL-6 production was lowered by the co-administration of iron and aflatoxin B1 (Fig. 5). This may account for the higher levels of serum iron in the same group compared to the iron group (Fig. 1). However, these differences were not statistically significant. Synergy of IL-10 was simply additive. The co-administration of iron and AFB1 had a super additive effect on the generation of reactive oxygen and species and oxidative stress in the liver, as evidenced by the increased total lipid hydroperoxide concentration (LOOH) and the generation of 8OHdG. Further evidence for a synergistic interaction between the two agents was provided by the immunohistochemical demonstration of greatly increased amounts 4-HNE-adduct formation and 8OHdG aggregates and the strongly positive Ames mutagenicity test using the S. typhimurium TA 100 bacteria strains. The synergism between the deleterious effects of excess tissue iron and the AFB1 appears to be taking place at the level of DNA damage, which would be in keeping with an increased hepatocarcinogenic potential. One possible mechanism of this interaction would be via the induction of the codon 249 ser p53 mutation by the dietary exposure to AFB1 (Hsu et al., 1991; Bressac et al., 1991) and the increased storage iron in the liver (Marrogi et al., 2001). A high frequency of p53 mutations has been reported in patients with hereditary hemochromatosis (Hussain et al., 2000; Vautier et al., 1999). This mutation may abrogate the functions of p53, including those involved in DNA repair and apoptosis, that

G.A. Asare et al. / Toxicology 234 (2007) 157–166

might be relevant to a synergistic interaction between AFB 1 and DNA damage induced by excess hepatic iron storage. In addition, a high level of non-heme iron in hepatocytes can inhibit nitric oxide-induced apoptosis by converting nitric oxide from a pro-apoptotic molecule to an anti-apoptotic molecule (Kim et al., 2000). References Asare, G.A., Mossanda, K.S., Kew, M.C., Paterson, A.C., KahlerVenter, C.P., Siziba, K., 2006a. Hepatocellular carcinoma caused by iron overload: a possible mechanism of direct hepatocarcinogenicity. Toxicology 219, 41–52. Asare, G.A., Paterson, A.C., Kew, M.C., Khan, S., Mossanda, K.S.A., 2006b. Iron-free neoplastic nodules and hepatocellular carcinoma without cirrhosis in Wistar rats fed a diet high in iron. J. Pathol. 208, 82–90. Berenbaum, M.C., 1985. The expected effect of a combination of agents: the general solution. J. Theor. Biol. 114 (3), 413–431. Bothwell, T.H., Bradlow, B.A., 1960. Siderosis in the Bantu: a combined histopathological and chemical study. Arch. Pathol. 70, 279–292. Bressac, B., Kew, M.C., Wands, J., Ozturk, M., 1991. Selective G to T mutations of p53 gene in hepatocellular carcinoma from southern Africa. Nature 350, 429–431. Britton, R.S., 1996. Metal-induced hepatoxicity. Semin. Liver Dis. 16, 3–12. Cheeseman, K.H., 1993. Lipid peroxidation in cancer. In: Halliwell, B., Arouma, A.I. (Eds.), DNA and Free Radicals. Ellis Horwood, London, pp. 109–144. Cheng, K.C., Cahill, D.S., Kasai, H., Nishimura, S., Loeb, L.A., 1992. 8-Hydroxyl guanine, an abundant form of oxidative DNA damage, causes G–T and A–C substitutions. J. Biol. Chem. 267, 166–172. Crofton, M.C., Craft, E.S., Hedge, J.M., Gennings, C., Simmons, J.E., Carchman, R.A., Carter Jr., W.H., DeVito, M.J., 2005. Thyroid-hormone-disrupting chemicals: evidence of dosedependent additivity or synergism. Environ. Health Perspec. 113, 11. Dabbagh, A.J., Mannion, T., Lynch, S.M., Frei, B., 1994. The effect of iron overload on rat plasma and liver oxidant status in vivo. Biochem. J. 300, 799–803. Feng, Z., Hu, W., Tang, M.S., 2004. Trans-4-hydroxy-2-noneal inhibits nucleotide excision repair in human cells: a possible mechanism for lipid peroxidation-induced cacinogenesis. Proc. Natl. Acad. Sci. USA 101, 8598–9602. Fiddler, R.N., 1977. Collaborative study of modified AOAC method of analysis for nitrite in meat and meat products. J. Assoc. Anal. Chem. 60 (3), 594–599. Gordeuk, V.R., Boyd, R.D., Brittenham, G.H., 1986. Dietary iron overload persists in sub-Saharan Africa. Lancet 1, 1310–1313. Gordeuk, V.R., McLaren, C.E., MacPhail, A.P., Deichsel, G., Bothwell, T.H., 1996. Associations of iron overload in Africa and hepatocellular carcinoma and tuberculosis. Blood 87, 3476–3479. Hagen, T.M., Huang, S., Curnutte, J., Fowler, P., Martinez, V., Wehr, C.M., Ames, B.N., Chisari, F.V., 1994. Extensive oxidative DNA damage in hepatocytes of transgenic mice with chronic active hepatitis destined to develop hepatocellular carcinoma. Proc. Natl. Acad. Sci. USA 91 (26), 12808–12812. Hsu, I.C., Metcalf, R.A., Sun, T., Welsh, J.A., Wang, N.J., Harris, C.C., 1991. Mutational hotspot in the p53 gene in human hepatocellular carcinomas. Nature 350 (6317), 427–428.

165

Hussain, S.P., Raja, K., Amstad, P.A., Sawyer, M., Trudel, L.J., Wogan, G.N., Hofseth, L.J., Shields, P.G., Billiar, T.R., Trautwein, C., Hohler, T., Galle, P.R., Phillips, D.H., Markin, R., Marrogi, A.J., Harris, C.C., 2000. Increased p53 mutation load in nontumorous human liver of wilson disease and hemochromatosis: oxyradical overload diseases. Proc. Natl. Acad. Sci. USA 97 (23), 12770–12775. Ichiba, M., Maeta, Y., Mukoyama, T., Saeki, T., Yasui, S., Kanbe, T., Okano, J., Tanabe, Y., Hirooka, Y., Yamada, S., Kurimasa, A., Murawaki, Y., Shiota, 2003. Expression of 8-hydroxy2 -deoxyguanosine in chronic liver disease and hepatocellular carcinoma. Liver Int. 23, 338–345. Jungst, C., Cheng, B., Gehrke, R., Schmitz, V., Nischalke, H.D., Ramakers, J., Schramel, P., Schirmacher, T., Caselmann, W.H., 2004. Oxidative damage is increased in human liver tissue adjacent to hepatocellular carcinoma. Hepatology 39, 1663– 1672. Kensler, T.W., Qian, G.S., Chen, J.G., Groopman, J.D., 2003. Translational strategies for cancer prevention in liver. Nat. Rev. 3, 321– 329. Kew, M.C., 2003. Synergistic interaction between aflatoxin B1 and hepatitis B virus in hepatocarcinogenesis. Liver Int. 23, 1–5. Kew, M.C., 1992. Chronic hepatitis-B virus infection and hepatocellular carcinoma in Africa. S. Afric. J. Sci. 88, 524–528. Kim, Y.M., Chung, H.T., Simmons, R.L., Billiar, T.R., 2000. Cellular non-heme iron content is a determinant of nitric oxide-mediated apoptosis, necrosis, and caspase inhibition. J. Biol. Chem. 275, 10954–10961. Kuldip, T., Abdel-Rahman, E., Paresh, D., 1998. Measurement of reactive oxygen species in whole blood and mononuclear cells using chemiluminescence. Meth. Mol. Biol. 108, 57–62. Lee, P., Peng, H., Terri, G., Wang, L., Beutler, E., 2004. Regulation of hepcidin transcription by interleukin-1 and interleukin-6. Proc. Natl. Acad. Sci. USA 102 (6), 1096–1910. Loewe, S., Muischnek, H., 1926. Effect of combinations: mathematical basis of a problem. Naunyn Schmiedebergs Arch. Exp. Pathol. Pharmakol. 114, 313–326. Loewe, S., 1953. The problem of synergism and antagonism of combined drugs. Arzeimittelforschung 3 (6), 285–290. Ma, Y., Zhang, D., Kawabata, T., Kiriu, T., Toyokuni, S., Uchida, K., et al., 1997. Copper and iron-induced oxidative damage in non-tumor bearing LEC rats. Pathol. Int. 47, 203–208. Mandishona, E., McPhail, A.P., Gordeuk, V.R., Kedda, M.A., Paterson, C., Rouault, T.A., Kew, M.C., et al., 1998. Dietary iron overload as a risk factor for hepatocellular carcinoma in black Africans. Hepatology 27, 1563. Maron, D.M., Ames, B.M., 1983. Revised methods for the Salmonella mutagenicity test. Mutat. Res. 113, 173–215. Marrogi, A.J., Khan, M.A., van Gijssel, H.E., Welsh, J.A., Rahim, H., Demetris, A.J., Kowdley, K.V., Hussain, S.P., Nair, J., Bartsch, H., Okby, N., Poirier, M.C., Ishak, K.G., Harris, C.C., 2001. Oxidative stress and p53 mutations in the carcinogenesis of iron overloadassociated hepatocellular carcinoma. J. Natl. Cancer Inst. 93 (21), 1652–1655. McNamara, L., MacPhail, A.P., Mandishona, E., Bloom, P., Paterson, A.C., Rouault, T.A., Gordeuk, V.R., 1999. Non-transferrin-bound iron and hepatic dysfunction in African dietary iron overload. J. Gastroenterol. Hepatol. 14 (2), 126–132. Moyo, V.M., Makunike, R., Gangaidzo, I.T., Gordeuk, V.R., McLaren, C.E., Khumalo, H., Saungweme, T., Rouault, T., Kiire, C.F., 1998. African iron overload and hepatocellular carcinoma. Eur. J. Hematol. 60, 28–34.

166

G.A. Asare et al. / Toxicology 234 (2007) 157–166

Nourooz-Zadeh, J., Tajaddini-Sarma, D.I., Wolff, S.P., 1994. Measurement of plasma hydroperoxide concentration by the ferrous oxidation-xylenol assay in conjunction with triphenylphosphine. Anal. Biochem. 220, 403–409. Roebuck, B.D., 2004. Hyperplasia, partial hepatectomy and the carcinogenicity of aflatoxin B1 . J. Cell Biochem. 91 (2), 243–249. Sudakin, D.L., 2003. Dietary aflatoxin exposure and chemoprevention of cancer: a clinical review. J. Toxicol. Clin. Toxicol. 41, 195–204. Toyokuni, S., Iwasa, Y., Kondo, S., Tanaka, T., Ochi, H., Hiai, H., 1999. Intracellular distribution of 8-hydroxy-deoxyguanosine. An immunocytochemical study. J. Histochem. Cytochem. 47 (6), 833–836.

Toyokuni, S., Tanaka, T., Hattori, Y., Nisjiyama, Y., Yoshida, A., Uchida, K., Hiai, H., Ochi, H., Osawa, T., 1997. Quantitative immunohistochemical determination of 8-hydroxy2 -deoxyguanosine by monoclonal N45.1: its application to ferric nitrilotriacetate-induced renal carcinogenesis model. Lab. Invest. 76, 365–374. Vautier, G., Bomford, A.B., Portmann, B.C., Metivier, E., Williams, R., Ryder, S.D., 1999. p53 mutations in British patients with hepatocellular carcinoma: clustering in genetic hemochromatosis. Gastroenterology 117 (1), 154–160. Wild, C.P., Turner, P.C., 2002. The toxicology of aflatoxins as a basis for public health decisions. Mutagenesis 17, 471–481.