Humans are continuously exposed to HAs, as demonstrated by detection of these compounds in cooked foods and human urine samples. The presence of 2-amino-3,8-dimethylimidazo[4,5-f]-quinoxaline (MeIQx)-DNA adducts found in human tissues, such as colon, rectum and kidney increased the concern about the health consequences of the intake of these compounds .

     Although methods for estimation of HA dietary exposure in the population are still imprecise , in some studies epidemiological evidence suggests that consumption of high amount of HAs-containing meat products may induce colorectal, pancreatic and urothelial cancer . On the other hand, other studies did not show such association . Today, more than 20 mutagenic/carcinogenic HAs have been isolated and identified in cooked foods . Their formation is the result of complex reactions that involve creatine, free amino acids and carbohydrates through the Maillard Reaction (MR). The development of MR also occurs through a free radical mechanism which has been shown to play an important role in the formation of imidazoquinoxalines and imidazoquinolines .

    The pathway of HA formation is shown in Fig.23. Creatine forms the amino-imidazo part of the molecule by cyclization and water elimination, whereas the remaining parts of the IQ compounds arise from Strecker degradation products, such as pyridine and pyrazine .

    This hypothesis was verified in model systems , and the results were later confirmed also in meat-based system.     

    The presence of HAs in foods depend on many factors such as cooking method, time and temperature, the presence of relative amounts of precursors, enhancers and inhibitors, lipids, antioxidants and the water content. .In particular, supplementation with antioxidants is considered to be an effective measure to reduce HA exposure because of the hypothetical free radical pathway leading to HA formation. Single antioxidant compound and complex mixtures of antioxidants have been demonstrated to inhibit HA-induced mutagenesis or carcinogenesis. It is likely, that this effect is the final result of different actions interfering at various steps of the HA formation and of HA-toxic activity. In fact, antioxidants can act as inhibitors along the different pathways of the reaction, preventing the mutagens formation,


fig.23.Scheme of HA formation and possible sites for interference with antioksidants.

                back to top


fig.24.Scheme of HA formation and possible sites for interference with antioksidants.UDP-glucurosonosyl transferase (UDPGT);Sulfotransferases(SULT);Glutathione-S-transferase(GST);Prostaglandin H (PGH)-sinthase.


                                                                                                                                         back to top


through radical quenchers and free radical scavengers activity (as shown in Fig.23); as blocking agents, preventing the biotransformation of premutagens into reactive metabolites by inhibiting metabolic activation, by stimulating detoxification enzymes, or by scavenging reactive molecules; as suppressing agents modulating intracellular processes, which are involved in DNA repair mechanisms, tumour promotion and tumour progression (see Fig.24).

    The aim of this review is to present the current knowledge on the ability of antioxidants to modulate HA-induced mutagenicity/carcinogenicity.

    To better underline the differences of action and the potential efficacy of antioxidants to minimize the human health risk associated to cooked meat/fish consume the review was organised in two main paragraphs dealing on the effect exerted by synthetic and natural antioxidants, respectively.


 Natural antioxidants

    When tested as single substances, various natural compounds show marked antimutagenic activity in the Ames test and other mutation assays, but little is known about such activity when these compounds are present in food interacting with other food components, during cooking/storage and in the lumen of gastrointestinal tract. The studies about the activity of naturally occurring antioxidants on HAs have been divided into two groups, the first reviewing papers based on use of pure compounds and the second using food extracts and whole foods.

                                                                                                                                         back to top

 Activity of pure compounds

Antioxidant vitamins

    In a real food system, such as fried fish fibre, prepared by boiling, deboning, eviscerating, separating and pressing of snake fish meat, the addition of Vitamin C and α-tocopherol did not show any consistent effect on the formation of HAs. On the contrary, previous studies showed that these antioxidants were able to inhibit mutagen formation during the frying of beef . The addition of Vitamin E to the surface of the beef patties before frying, at the concentration of 1–10% of the fat content, produced a reduction of PhIP formation after cooking in the range of 45–75% . These results can be due to the ability of tocopherols to inhibit free radical formation and/or producing compounds which may react with HA precursors and prevent formation of 4,8- DiMeIQx .

    The capability to decrease active metabolite formation is presumably responsible of the reduction of IQ-DNA adductformation exerted by leaves of Kidachi aloe, which are rich in α-tocopherol, and ß-carotene . Anyway antimutagenic effects and possible mechanisms of action of several vitamins against genotoxic HAs have been reported by Edenharder and co-workers. Results showed a lack of efficacy of ascorbic acid and α-tocopherol against HA-induced mutagenicity in the Salmonella/reversion assay.

    Opposite results were obtained by Snyderwine and coworkers  who hypothesised that ascorbic acid could prevent the oxidative break-down of N-hydroxy-IQ, enabling the mutagen to remain active for a longer time in vitro with a consequent enhancement of mutagenic activity. In fact, 0.5 and 1mM ascorbic acid, added to the plates, produced 11,000 and 11,500 revertants per plate compared to 1000 revertants per plate produced when N-hydroxy-IQ was co-workers  was probably due to some metabolites of ascorbic acid or of N-hydroxy-IQ generated by cytosolic enzymes .

Phenolic antioxidants

    Lee and co-workers suggested that flavones inhibited formation of IQ-type mutagens in a glycine/glucose/creatine model system reducing the formation of Maillard reaction products . Edenharder and co-workers  some years later reported that, among the 30 flavonoids tested, all of them but two inhibited the mutagenic activity in the Salmonella assay; in presence of diosmetin and isorhamnetin a slight increase was observed. Among flavones and flavonols, the inhibitory effects increased in dependence on number and position of hydroxyl functions. This figure suggests that in poly-hydroxylated compounds the number and the position of OH groups are involved in the modulation of the activity of the enzymes responsible for IQ n-hydroxylation inhibition .

    These in vitro evidences were not confirmed by an in vivo study performed to verify the effect of several flavonoids on hepatocarcinogenesis. Hirose and co-workers reported that quercetin, rutin, curcumin, daidzin, ferulic acid and genistein were co-carcinogens for MeIQx-induced hepatocarcinogenesis, because they were able to enhance GST-P positive focus development.

    Also studies on ellagic acid and nordihydroguaiaretic acid (NDGA) gave apparently contradictory results. Ellagic acid reduced the formation of MeIQx but enhanced that of PhIP while NDGA had the opposite effect . The enhancement of MeIQx formation by NDGA could be explained according to the authors by the same alkylation mechanism described by Pearson and co-workers  for BHT, while in the case of PhIP formation some reactive groups derived from ellagic acid were supposed to be involved .

    Consistent evidences have been found using other phenolic antioxidants such as (-)-epigallocatechin gallate (EGCG) which has been reported to suppress the formation of imidazoquinoxaline-type HAs in heated mixtures of creatinine, amino acids and sugars. Kato and co-workers demonstrated that generation of the mutagens in a heated chemical model system was effectively prevented by phenolic antioxidants, sesamol, esculetin and EGCG in a dose-dependent manner (being MeIQx content reduced to 35% of control in the presence of 100mM EGCG) and in heated-and-dried bonito meat by pre-treatment with 0.5% EGCG or 5% green tea extract. The mutagenicity of heated-and-dried bonito meat, tested on S. typhimurium TA98 strain with metabolic activation (by liver microsomal S9 system) after blue rayon extraction and determined as number of histidine revertant colonies, was dramatically reduced by pre-treatment with 0.5% EGCG and 5% green-tea extract resulting less than 50 and 30% respectively. The authors postulated these antioxidants acted preventing the formation, through Maillard reaction , of the unstable pyrazine cation radical. In a subsequent study, phenolic antioxidants such as green tea catechins and their major component EGCG, two flavonoids (luteolin and quercetin), and caffeic acid were found to decrease up to 75% both MeIQx and PhIP, and to reduce the total mutagenicity of heated mixtures . Also in animal experiments green-tea catechins have been shown to exert inhibitory effect against GST-P positive focus development.

    As far as the mechanisms of action of phenolic antioxidants, beside the direct scavenging and reducing action and the modulation of enzymes involved in the detoxification system also a mechanism involving signal transduction has been described. The alteration of pathways that control apoptosis or cell-proliferation could explain the protective effect observed during the initiation phase of HA-induced carcinogenesis by several phenolic antioxidants.

                                                                                                                                         back to top

Activity of food-extracts and whole food

    In the last years, many studies have been performed to test the effective capability and the mechanism of action of rich-antioxidant food in preventing HA-induced mutagenicity/ carcinogenicity. The attention for the use of whole food or food extracts is due to the need of increase our knowledge on the behaviour of antioxidants in conditions which can resemble as much as possible the reality.

Fruits and vegetables

    Almost all studies present in the literature consistently showed a strong antimutagenic activity, detected in liverenzyme mediated bacterial assays, exerted by fruits and vegetables against HA action. This effect was strictly correlated to the presence in these foods of vitamins and related compounds (ascorbic acid, ß-carotene, retinal, retinoic acid,α-tocopherol etc.) having antioxidant properties .

     Carotenoid-rich foods

     Vitaglione and co-workers  have reported that carotenoids from tomatoes inhibit the formation of imidazoquinolines both in chemical model system, containing as precursors creatine, glucose and glycine, and in meat juice model system, based on freezedried bovine meat juice. In particular using carotenoid extract at a concentration of 1000 ppm, inhibitions of 36 and 11% of IQx and MeIQx formation respectively in the chemical system and of 13% of MeIQx and of 5% of 4,8- DiMeIQx in the meat juice model system was observed. Quercetin, the main tomato flavonoid, tested at 10 ppm in the meat juice model system, exerted an inhibition of HA formation of 67% .

    N-hexane extracts of some carotenoid-rich fruits and vegetables such as apricots, oranges, Brussels sprouts, carrots, yellow-red peppers, and tomatoes were directly tested on histidine-deficient strains of Salmonella typhimurium . This study demonstrated a reduced mutagenic activity of IQ probably due to carotenoids (α-, ß-carotene, lycopene),xanthophylls (ß-cryptoxanthin, lutein), and also carotenoid esters (oranges) contained in the tested extracts. In particular, 100µg of orange extract reduced the bacterial mutagenicity of IQ by 27%. Previously solvent extracts from 13 fruit and 12 vegetable residues had been demonstrated to exhibit antimutagenic activity against IQ and 2-amino-3,4- dimethylimidazo[4,5-f]quinoline (MeIQ) in S. typhimurium TA 98 assay .

                                                                                                                                         back to top

    Anthocyanin-rich foods

    Tart cherry tissue has been also demonstrated to exert inhibitory effect on HA formation in ground beef patties .

     The potential beneficial effect of fruits has been recently confirmed by Miyata and co-workers  who showed that grapefruit juice suppressed PhIP-induced colon DNA damage in a concentration-dependent manner. The migration of DNA and frequency of tailed colon nuclei as indicators of DNA damage were measured. In particular, a 40% reduction of DNA damage was observed in F344 rats given 60 mg/kg of PhIP by gavage after pre-treatment with grapefruit juice for 5 days. This result was independent of PhIP absorption in the intestine because serum level of PhIP was comparable between grapefruit juice—pretreated and non pretreated rats.

    Furthermore, it has been reported that genotoxic activity of PhIP was strongly reduced in a dose-related manner by blueberries, blackberries, red grapes, kiwi, watermelon, parsley, and spinach and that protection by beverages, fruits, and vegetables against genotoxicity of HAs might take place by enzyme inhibition .

    The administration of 5% purple corn color, an anthocyanin extracted from the seeds of corn, in rats pre-treated with 1,2-dimethylhydrazine (DMH) and given PhIP in the diet, has been evaluated to reduce development of colorectal carcinogenesis . Furthermore, cacao liquor proanthocyanidins showed to inhibit in vitro mutagenicity of PhIP, as well as rat pancreatic carcinogenesis in the initiation stage, but not mammary carcinogenesis induced by PhIP.


    The addition of soy protein concentrates (SPC) to beef patties also have shown to inhibit the mutagen formation in pan-fried beef patties during cooking. This effect was not related to protein alone but mainly to the phenolic antioxidant component in SPC such as chlorogenic acid. More recent studies have not confirmed this finding. In fact Lan and Chen found that soy sauce-marinated pork, eggs, and bean cakes show an increase of HAs. Moreover, adding soybean-oil during frying of fish fibre the same effect of enhancing HA formation was exerted .


    Polyphenolic compounds from tea are effective inhibitors of HA formation in model systems. Several studies report that tea polyphenols attenuate mutagenic activities of HAs by various mechanisms such as: inhibition of NADPH cytochrome P450 reductase , inhibition of mutagenic activity of N-hydroxylated HAs in the absence of S9 in vitro  and electrophilescavenging. The first mechanism has been recently recapitulated using specific isoforms of human cytochromes P450, including inhibition by epigallocatechin-3-gallate (EGCG) of human CYP1A2-mediated activation of PhIP in genetically engineered Salmonella strains . Data showed that the protection towards HA action depended not only from quality and concentration of antioxidants but also from their relative levels in food and from the influence of other minor food constituents on their activity.

    This correlation has been demonstrated in green- and black-tea extracts whose antimutagenic activity reduction corresponded with reduction in antioxidants and in particular with a decrease in concentration of three catechins (catechin, epigallocatechin gallate and epigallocatechin) . Furthermore, an artificial tea produced by mixing nine of the major constituents found in green tea (including high level of EGCG and several other polyphenols) exhibited a smaller antimutagenic potency in Salmonella assays compared with the complete tea .

    The bioavailability of potential antimutagenic compounds is another key-factor that has to be taken into account. Krul and co-workers evaluated the influence of the food matrix on the bioavailability of tea antioxidants using an in vitro gastrointestinal model, simulating the conditions in the human digestive tract. In particular, they investigated whether black tea and green tea preserved their antimutagenic properties against HAs, alone or together with milk at different fat content, after passage through the in vitro model. The antimutagenic activity of the filtrates obtained introducing in the gastrointestinal model tea extracts alone, together with MeIQx and with whole or semi-skimmed or skimmed milk, was determined in the Ames test. The maximum inhibition of mutagenicity in Salmonella assay was observed with black tea filtrates, while adding together with the tea extract whole, semi-skimmed or skimmed milk, a reduction of the antimutagenic activity of 22, 42 and 78%, respectively, was observed. The authors hypothesized that these findings were due to the formation of milk proteintea polyphenol complexes, resistant to gastric hydrolysis, with consequent reduction of absorbed antimutagenic compounds. Furthermore, antimutagens of green-tea, such as catechins (90% of dry weight) and quercetin, bind stronger to milk proteins than the antimutagens of blacktea, thus explaining why the effect of green-tea was more affected by milk than that of black tea . When tea and MeIQx were added together into a digestion model system, MeIQx mutagenicity was efficiently inhibited, green tea had a slightly stronger antimutagenic activity than black tea. Moreover, purified polyphenols from black tea extracts are more potent inhibitors of mutagenicity caused by PhIP in the S. typhymurium TA98 assay than the polyphenols from green tea extract . Application of polyphenols extracted from both tea varieties to the surfaces of ground beef before cooking is able to inhibit the formation of mutagens in a dose-related fashion . Furthermore, both theaflavins and theafulvins from black tea brews are able to prevent, at a dose-range 0.1–0.5 mg/ml, in a concentration-dependent manner, the DNA damage caused by heterocyclic amine 3-amino-1-methyl-5H-pyrido[4,3-b]indole (Trp-P-2) .

    The chemical mechanisms implicated in the inhibition by tea extracts of both HA-formation and HA-induced mutagenicity/ carcinogenicity, have been amply studied.

    Stavric and co-workers  reported that extracts of various teas (green, oolong, orange pekoe, decaffeinated orange pekoe) and herbal teas (chamomile, orange spice, linden flowers), were able to inhibit the Salmonella assay mutagenic activity of eight different HAs. The effect was strongly dependent from the concentration of tea extracts and it decreased at high concentrations. In particular, studies on the antimutagenic actions towards IQ of tea extracts (green, pouching, oolong and black tea) using a Salmonella/microsome assay showed that they were due to a combination of two distinctive mechanisms: inhibition of the cytochrome P450-mediated metabolism of IQ to its mutagenic metabolite form and interaction with IQ promutagens and metabolites in such a way to reduce its mutagenic potential .

    Tea polyphenols could have also additional mechanisms of protection against HAs, such as the induction of UDP-glucuronosyl transferase (UDPGT), another enzyme involved in phase II metabolism of HAs. This is the prevalent mechanism associated to the inhibitory activities of white, green, and black teas against IQ and PhIP-induced colonic aberrant crypt foci (ACF) and DNA adducts in the rat . In fact rats fed with white or green teas extracts (1.5 or 2% w/v), in place of drinking water, before treatment with PhIP or IQ had reduced parent compounds and sulfamate in the urine. Accordingly, the concentration of ring glucuronides increased .

    White tea is also able to inhibit PhIP-induced colonic ACF inducing glutathione-S-transferase (GST)  thus assessing glutathione-dependent reduction of N-acetoxyheterocyclic amines back to the parent compound .

    On the contrary, liver S9 or microsomes from rats given aqueous extracts of green tea enhanced the mutagenic activity of 2-amino-6-methyl-pyrido[1,2-a:3,2-d]imidazole (Glu-P-1) and IQ in the Salmonella assay  through the induction of CYP1A and consequent enhanced formation of N-hydroxylated HAs, which were substrates for further activation by bacterial O-acetyltransferase and by enzyme S9 . 

                                                                                                                                        back to top


    Olive oil

     Monti and co-workers  reported that olive oil inhibited the formation of different IQ-type HA, in a chemical model system. The fresh made oil was more effective than the same oil after 1 year of storage thanks to the presence in the first of largest amount of dihydroxyphenylethanol derivatives. Their data supported the hypothesis that HA formation partially involved free-radical reactions, thus a reaction between phenolic compounds and key intermediates of HA formation could also contribute to reduce the levels of HAs.

    Previously Johansson and co-workers  reported that using frying fats with an initial high antioxidant level lower amounts of MeIQx and DiMeIQx in the beef burgers and in the pan residues after frying were obtained.


    Thyme, marjoram, rosemary and Monascus red extracts added, each at three different concentrations (4, 10, and 50 mg that is about 0.0005, 0.0025 and 0.00625%), in chemical model system, caused an increase of PhIP formation stronger for thyme, marjoram and Monascus red in comparison to rosemary flavour . No correlation could be found between the antioxidative properties of the tested food additives, measured by the Rancimat system, and the formation of PhIP. In fact PhIP content increased independently to pro- or anti-oxidative properties of flavours. The extracts of thyme and Monascus red showed an antioxidative potential measured toward rapeseed oil oxidation while marjoram and rosemary showed pro-oxidative properties. On the contrary, some years before Murkovic and co-workers  found that the addition of spices including rosemary, thyme and sage to the ground beef, reduced the concentrations of HAs in fried beef. Balogh and co-workers also showed that rosemary oleoresin added directly to the ground beef patties or to the surface of the patties before frying, at the concentration of 1–10% of the fat content, produced a reduction of PhIP formation after cooking of 44%. 

                                                                                                                                            back to top