β-carotene is a proven anti-oxidant. Working in the lipid phase of the cell, carotenoids act as free radical traps, serving to quench reactive oxygen species. Based on this knowledge, a potential mechanism of action of carotenoids as anti-carcinogens may be via their anti-oxidant activity.
Vitamin C
Vitamin C or ascorbic acid has received much interest as a potential anti-carcinogenic agent and has been assessed for its anti-carcinogenic effects under a variety of experimental conditions. It is contained within a variety of fruits and vegetables, in particular citrus fruits such as oranges.
Humans and other primates lack the gene product responsible for gulonolactone oxidase formation, an enzyme involved in the synthesis of ascorbic acid [6], hence we must take in the vitamin in sufficient quantities in the diet, the current RDA being 60mg. A number of mechanisms of action have been proposed for the anti-carcinogenic effects of the vitamin.
Ascorbic acid works as an electron donor or reducing agent. It is feasible therefore to assume that, like β-carotene, the substance may be anti-carcinogenic by virtue of its anti-oxidant activity. The vitamin works as an anti-oxidant in the aqueous phase of the cell.
Another proposed anti-carcinogenic mechanism of action for vitamin C is based on its ability to compete for nitrites in their reaction with amines to yield carcinogenic nitroso compounds [6]. This reaction has been suggested to be significant in the prevention of stomach cancer by vitamin C [6].
There is obviously a significant amount of evidence suggesting an anti-carcinogenic role for vitamin C, however, an intake sufficient for the vitamin to exert an optimum effect has yet to be established.
The RDA of 60mg was chosen primarily as a level considered sufficient to prevent the occurrence of vitamin C deficiency, scurvy [6]. The key issue regarding intake however is whether preventing a deficiency of vitamin C is equivalent to its optimum intake for cancer prevention.
It is important to assess vitamin C requirements with respect to its functions within the body and thus to account for catabolic losses. In situ kinetics have been employed in many animal tissue experiments and have helped to determine precisely in which reactions ascorbic acid participates and how these reactions are regulated by the actual concentration of ascorbate. This in situ evidence then forms the basis of clinical assessments as to how much vitamin C is actually required.
There are no current recommendations as to the optimum intake of vitamin C for cancer prevention. For the maintenance of a healthy immune system, an intake of 2 grams per day or more is generally accepted as safe. This dose is normally taken in the form of vitamin supplements and to achieve this level of intake in the diet one would have to consume a considerable amount of vitamin-rich foods.
Vitamin E.
Vitamin E or α–tocopherol is an important anti-oxidant in plants, serving to protect the chloroplast from oxidative damage. The role of vitamin E in the prevention of cancer is unclear. It was originally proposed as an anti-cancer agent owing to its inverse relationship with colon cancer in certain studies. α–tocopherol is the most biologically active tocopherol and hence is the major form considered in dietary calculations of vitamin A intake [7]. Other tocopherols such as β– and γ–tocopherol have lower biological activities but are often present in far greater amounts in the diet [7]. The mechanism by which vitamin E is potentially anti-carcinogenic has yet to be elucidated but may be associated with its anti-oxidant activity. In contrast to vitamin C, vitamin E works in the lipid phase of the cell and is therefore an important protective molecule in biological membranes, for example in the protection against lipid peroxidation, (fig.2)
Figure 2. Vitamin E reacts with lipid hydroperoxide radicals yielding the hydroperoxide and α tocopherol (Timbrell).
LO2∙ + α-TH → LOOH + α-T
LO2∙ + α-T∙→ LOOH + α-TQ
In conclusion, diets generally high in fruits and vegetables have been demonstrated to have inverse correlations with a variety of cancers (Figure 3). Overall, the protective effects of these foods appear to be general and it is difficult to assign a preventative role to any one constituent. Evidence is mounting however concerning certain associations between specific nutrients in fruit and vegetables and a reduction in cancers at specific sites for example the mouth, oesophagus and stomach with vitamin C and carotenoids.
It has been reported that for many cancer sites, individuals with a “low” intake of fruit and vegetables experience up to twice the incidence of cancer compared to those with a high intake [8].
Figure 3. Relative risk of cancers at various sites among high vs. low consumers of fruit and vegetables [2].
0.8
0.6
0.4
0.2
0
Cancer site
As a result of this and other pieces of evidence, certain recommendations have been made as to the optimum intake of fruit and vegetables for potential cancer prevention. An intake of between 5 and 9 portions per day may be beneficial in preventing cancer [9] and in promoting a general sense of wellbeing. Further investigations regarding the potential anti-cancer activity of specific constituents of fruit and vegetables is required to optimise this method of chemoprevention, but for the moment we can conclude that a diet abundant in fruit and vegetables may give rise to a decreased risk of cancer overall.
Fibre.
“Fibre” is a term describing plant material that is abundant in cell walls. It shows high resistance to mammalian digestive enzymes, but is selectively sensitive to degradation by intestinal bacteria. “Fibre” encompasses many components, each with a specific physiological function [10].
There has been considerable interest in the role of dietary fibre in the aetiology of a number of diseases common to the western population. In 1972, Burkett et al proposed that dietary fibre plays an important role in the normal functioning of the gastro-intestinal tract and the potential for fibre to posses anti-carcinogenic effects is now recognised. Despite this, evidence has been confused by an uncertainty about which specific components of fibre exert the beneficial effect [10].
The role of fibre as a potential anti-carcinogen has been investigated by way of various epidemiological and experimental studies. Evidence for its beneficial effects in the prevention of colon cancer is particularly compelling, however such studies are not unanimous. Some studies investigating the role of the nutrient in the prevention of colorectal cancer have identified that fibre derived from cereals exerts no or little protective effects, whilst that obtained from fruit and vegetables is beneficial in being anti-carcinogenic. It is therefore extremely difficult to say with certainty whether it is actually the fibre, or other constituents of fruit and vegetables which has the beneficial effects.
Nevertheless the evidence for the role of fibre-rich foods in limiting the risk of colorectal cancer especially is mounting. Despite a genetic component of this cancer being well established [11], the disease is undoubtedly influenced by environmental factors, as illustrated by its occurrence in migrants from low to high risk areas [11]. Doll and Peto [12] have suggested that variations in dietary status may account for as much as 90 percent of the variation in rates of colorectal cancer among different populations. Numerous studies, for example Burkitt [13], have reported a lower incidence of colon cancer in certain regions of Africa, where the typical diet contains a high level of fibre.
What is the proposed mechanism of action of fibre as an anti-carcinogenic agent? The substance is known to shorten the intestinal transit time of food, thus potentially limiting the length of time that carcinogenic substances are held in the gut. In addition, fibre brings about an increase in the bulk and water content of gut material, which could potentially “dilute” any carcinogens present (table 2.)
In summary, the evidence regarding fibre in the prevention of colorectal cancer is inconclusive. Most studies assess the effects of fibre-rich foods. Such foods are typically rich in other micronutrients such as vitamins, hence one cannot be sure which specific components of the foods may be involved in cancer prevention. In addition, high fibre diets are typically low in calories and fat, two dietary constituents believed to play a causative role in cancer. The evidence regarding fibre intake and colon cancer however is compelling enough to warrant the recommendation of a diet high in fibre, which is undoubtedly also beneficial in promoting a general state of good health.
Food restriction.
A decreased incidence of cancer has been demonstrated not only by the administration of specific anti-mutagens and anti-carcinogens, but also by the restriction of food, which has also been illustrated to elongate the lifespan of experimental animals.
Although the potential anti-carcinogenic effects of calorific restriction have not been demonstrated directly in humans, there is compelling evidence that energy restriction and the resultant retarded growth rates give rise to a reduced incidence of breast cancer [2]. This evidence is supported by the use of height as an indicator of energy intake during the development of migrants switching from low to high calorie intake.
How does food restriction give rise to a lower incidence of tumours? The exact mechanism is not known although it is possibly via an inhibition of the generation of reactive oxygen species or due to a decrease in the rate of cell turnover.
It is extremely difficult to assess the relevance of this phenomenon to humans. The effects of calorific restriction would be extremely difficult to measure, particularly as low calorie diets are generally high in fruits, vegetables and fibre which are associated with a decreased incidence of neoplasms anyway. Maintaining weight within a desirable range and participating in a moderate level of exercise are current recommendations for a generally healthier lifestyle and may be beneficial in minimising the risk of cancer.
Dietary mutagens/carcinogens.
Dietary carcinogens can be roughly divided into two groups; direct carcinogens, which themselves have the potential to give rise to cancer and indirect carcinogens, which may also act as co-carcinogens.
Laboratory studies have identified a variety of dietary mutagens and carcinogens, both naturally occurring and synthetic (Table 4.). Such compounds include polycyclic aromatic hydrocarbons, heterocyclic amines, nitrosamines and mycotoxins. Despite this, epidemiological studies have identified consistently very few dietary constituents with the potential to cause cancer.
Table 3. Some proposed dietary carcinogens.
Natural carcinogens and mutagens in food.
A number of potential natural carcinogens have been identified in food, many of which are plant derived. Plants synthesise toxic chemicals in significant amounts, presumably as a primary defence against various predators. Such toxins have been studied for over 100 years and are still being discovered today [4]. Examples include;
-Estragole, safrole and methyleugenol, found in many edible plants, herbs and spices [15]. These substances are proven carcinogens in rodents and several of their metabolites are mutagenic. Black pepper contains small amounts of safrole and pepper has been shown to be carcinogenic in mice at doses far exceeding human exposure.
-Hydrazines, such as p-hydrazinebenzoate, which are contained in mushrooms. Most hydrazines are carcinogenic and mutagenic. One such compound, N-methyl-N-formylhydrazine has been shown to cause lung tumours in mice at relatively low levels [15]. Hydrazines are thought to be carcinogenic in part at least via the production of reactive oxygen species.
Certain microbial metabolites are also proven carcinogens [16], probably the best example being aflatoxins. Aflatoxins, some of the most potent mutagens and carcinogens known [15] are produced by certain strains of the fungi aspergillus flavus parasiticus and anomius [17]. The toxicological interest in aflatoxin has focused on aflatoxin B1, owing to its acute toxicity and carcinogenicity in humans and animals. Aflatoxin B1 is particularly abundant in peanuts and herbal spices and has been shown to induce liver tumours in experimental animals [16]. Epidemiological studies have also illustrated aflatoxin to be a potent human carcinogen, particularly in China and sub-Saharan Africa. DNA adducts have been found in humans living in these areas and p53 mutations found in hepatic tumours of those inhabiting high risk areas are the same as those observed experimentally [16].
In conclusion, although many of these plant and microbial derived substances are potent carcinogens, our exposure to them in the typical western diet is fairly insignificant, hence they are not a major factor when considering the influence of the Western diet on the occurrence cancer.
In addition to these naturally occurring carcinogens, other carcinogens are produced during the storage and preparation of food. Nitrosamines are formed during the pickling of food and its heating and result from the reaction between nitrite and secondary amines. Nitrosamines are found in significant amounts in pork and its derivatives, cheese and beer [18]. They are potent DNA methylating agents, particularly on guanine residues. Such mutations can result in mis-pairing such as GC-AT transitions. A cell’s ability to remove this error before cell division occurs is a critical determinant in the development of tumours. The dietary status of an organism may influence sensitivity to this carcinogen. Protein deficient diets reduce the toxicity of nitrosamines, but can give rise to an increased susceptibility to nitrosamine-induced kidney tumours as their metabolism is reduced and the kidney is exposed to a higher concentration of the carcinogen
Red meat and colon cancer.
Epidemiological studies have consistently identified an association between the intake of meat, particularly well-cooked red meat and cancer of the colon and rectum [2]. This association has lead to significant alarm among the general public over the last few years. Originally, fat was the suspected causative agent in meat. More recently however, the components were identified to be meat-related carcinogens such as heterocyclic amines [19]. These heterocyclic aromatic amines, particularly abundant in well cooked meat, are produced during the heating of free amino acids and have been shown to be potent mutagens in the Ames test [20]. In addition they have been shown to give rise to an increased incidence of cancer of the liver, large and small intestine, skin, breast and other sites in rodents.
The metabolism of these compounds to their hydroxylated derivatives is catalysed by cytochrome P4501A2. They are then further metabolised via N- and O-acetylation to their ultimately reactive forms. It has been demonstrated that the relative ability to metabolise such compounds is one determinant of the incidence of colo-rectal cancer.
The heterocyclic amines Me1Qx and Ph1P for example are efficiently absorbed into the systemic circulation following the ingestion of cooked food [21] and these two substances have been shown to be substrates for cytochrome P450 1A2, which catalyses their metabolism to genotoxic hydroxylamines. Me1Qx and Ph1P are hence classified as promutagens.
One recent study [15] examined the intake of meat, fat and fibre amongst health professionals and found that men who consumed beef, pork or lamb as a main dish, 5 or more times per week had a greater risk of developing cancer at various sites in comparison to those who ate such meat less than once per month.
Another study [20] investigated the long term feeding of a diet rich in well cooked beef in relation to the incidence of colon cancer in 1,2-Dimethylhydrazine treated rats. Lean beef was cooked by a variety of methods at different temperatures and levels of the major heterocyclic aromatic amines (HCAs) such as 2-amino-3,4,8-trimethylimidazo[4,5-F]quinoxaline were assayed by high performance liquid chromatography. Feeding of beef high in HCAs resulted in a higher incidence of DMH-induced colon tumours, hence the results provided further support for the hypothesis that the feeding of a diet high in well-cooked beef increases the incidence of colon cancer.
Environmental contaminants and food additives.
It is probable that a certain amount of environmental contaminants enter the food chain, some of which may be carcinogenic. Gold et al [22], estimated the total daily burden of synthetic pesticides to be around 0.09mg, but considering the average daily intake of burnt material (~200mg) and that of natural pesticides (~150mg), this is not thought to be significant.
No synthetic chemical can be deliberately added to food without rigorous toxicity testing nowadays. Doll and Peto [12] estimated that food additives have an etiological role in up to 2 percent of all deaths from cancer in the US. In fact it has been suggested that food additives may protect against cancer in some cases by reducing the microbial contamination of food.
Fat.
Fat currently accounts for around 40 percent of the total energy in a typical UK diet. There is a considerable amount of evidence to suggest that diets with a high fat content are correlated positively with a high risk of a number of cancers including cancer of the breast, colon, pancreas and prostate [14]. Epidemiological studies in humans however are merely suggestive of a link between a high intake of fat and cancer.
Animal studies have consistently shown that a high level of dietary fat serves as a promoter in the carcinogenic process. Various plausible mechanisms have been proposed to account for the relationship between a high fat intake and cancer.
It is feasible to assume that an appreciable amount of the fat in a typical diet may be oxidised. Unsaturated fatty acids and cholesterol can become oxidised fairly easily, for example during cooking [15]. The process of lipid peroxidation typically yields a variety of mutagens, promotors and other carcinogenic species such as cholesterol hyperoxide and various epoxides. The digestive tract, including the rectum and colon could therefore potentially be exposed to an array of fat-derived carcinogens.
Despite the fact that rodent feeding studies with oxidised fat have failed to present definitive results, the idea that fat may be carcinogenic by virtue of its oxidative derivatives is certainly a plausible hypothesis.
Another proposed mechanism by which fat may be carcinogenic is associated with fat breakdown by peroxisomes, intracellular organelles which have evolved to oxidise fatty acids. The removal of each two-carbon unit from a fatty acid generates one molecule of hydrogen peroxide which is a proven mutagen and carcinogen [15]. The majority of hydrogen peroxide produced in this way is broken down, however some inevitably escapes this catalase-dependent reaction, hence in the presence of iron-containing compounds, has the potential to generate hydroxyl radicals. Hydroxyl radicals are extremely reactive and in generated in proximity to DNA, have the potential to induce DNA damage, which is mutagenic and potentially carcinogenic.
Another proposed mechanism of fat toxicity could be via mitochondrial or peroxisomal membrane damage, initiated by abnormal fatty acids. Such membrane damage could potentially liberate harmful and potentially mutagenic oxygen derived radicals such as the superoxide radical.
A high fat diet is one of the factors implemented in peroxisome proliferation. This process, which can also be induced by certain hepato-carcinogenic compounds, has been recognised as a precursor to carcinogenesis, via the production of reactive oxygen species and consequential DNA damage. Despite the fact that a high fat diet induces a comparatively low level of peroxisome proliferation, this mechanism may in part at least, explain the carcinogenic consequences of a such a diet.
Dietary fat and breast cancer.
The epidemiology of breast cancer has been studied more than any other disease and its incidence has been attributed to a number of factors. These factors include reproductive patterns, physical activity, adiposity, alcohol intake and exposure to endogenous oestrogens [23].
The association between dietary fat intake and breast cancer however remains unresolved. Although there seems to be little doubt that nutritional status plays a key role in the aetiology of the disease, some believe that it is total energy intake rather than fat per se that is associated with the risk of developing breast cancer. There is some disagreement however as to which stage of life nutritional factors, such as fat intake become significant [24]. Some have suggested that dietary status during adult life, including dietary fat content may be a major determinant of breast cancer. Others believe that the risk of breast cancer is dependent upon a young woman’s diet before or during adolescence. One study [25] proposed that various components of the Western diet including fat can provoke hyperinsulinaemic insulin resistance at puberty which has been associated with abnormal ovarian steroidogenesis and anovulation which may initially decrease the risk of breast cancer. If obesity continues beyond puberty the high concentration of insulin like growth factor1 can interact with oestrogen receptors in mammary epithelium and lead to increased proliferation.
The relationship between a high intake of fat and the incidence of breast cancer has also been investigated by way of animal experiments. One such study [26] investigated the proliferation, development and DNA adduct levels in the mammary gland of rats administered with the heterocyclic amine 2-amino-1-methyl-6-phenylimidazo[4,5-b}pyridine (Ph1P) and given a high fat diet. The 43-day old rats were given 75mg/kg doses per day for 10 days and were then placed on a high fat diet of 23.5 percent corn oil or a low fat diet of 5 percent corn oil for up to 6 weeks. At various times the percentage of proliferating cells in the epithelium of the mammary gland was calculated. After 6 weeks on the diet the percentage of proliferating cells was significantly higher in the rats receiving the high fat diet, as were the levels of Ph1P-DNA adducts and these lead in some cases to cancer.
A number of scientists agree that adult height, which is partly determined by early energy intake, is positively correlated with the incidence of breast cancer [23]. Early ovulation, which is associated with a greater childhood weight gain is another phenonenom with a positive association with the disease. In summary the evidence implicating a high fat diet with the incidence of breast cancer is inconclusive. Epidemiological studies are not without their faults and it is difficult to assess the animal experiments in relation to human exposure scenarios therefore further studies in this area are required.
Conclusion.
In conclusion the human diet consists of a complex array of substances many of which are potentially carcinogenic. It is hence extremely difficult to assign a preventative or causative role to any one dietary constituent. It is extremely difficult also to distinguish causal relationships from indirect associations and there is the added complication of possible synergistic relationships between dietary components. Both epidemiological and experimental studies have their drawbacks. Human epidemiological studies are notoriously difficult to plan and interpret and laboratory studies on animals often involve the induction of tumours by various carcinogens, also the anti-carcinogen in question is administered in high doses therefore the relevance to human exposure scenarios is questionable. Nevertheless this is one area where extensive research has been conducted and this research can be used in the proposal of various dietary intervention strategies as part of a set of general lifestyle recommendations for healthy living including.
-An increase in fruit and vegetable consumption to 5 or more portions a day.
-An increase in cereal bran fibre intake.
-An adjustment of total fat intake to around 20% of total calories.
-An avoidance of salted and pickled foods.
-The consumption of alcoholic beverages in moderation.
-A moderation in the consumption of well cooked red meat.
Future research
Animal experiments and human epidemiological studies have provided numerous and consistent pieces of evidence implementing dietary factors in the occurrence of carcinogenesis, however the potential effects are diverse and hence are very difficult to quantify. It is for this reason that further research is required and such research may include:
-Assays of the mutagenic and carcinogenic potentials of specific dietary constituents, for example via the Ames test.
-Assays of DNA and protein adducts in humans.
-The sequence analysis of mutations within oncogenes and tumour suppresser genes in tumours.
-Investigations into the occurrence of mutations in xenobiotic metabolising enzymes and their potential contributions to diet-induced cancer.
-Further investigations into the mechanisms of mutagenesis and carcinogenesis.
References
-
Sugimura, T. Mutation Research, 402, 7-14, (1998).
-
DeMarini, D. M. Mutation Research, 400, 457-465, (1998).
-
Kusamran, W.R. Tepsuwan, A. & Piengchai, K. Mutation Research, 402, 247-258, (1998).
-
Dragsted, L. O., Strube, M. & Larsen, J. C. Pharmacology & Toxicology, 72 Suppl 1: 116-135, (1993).
-
Ziegler, R. G., Subar, A. F., Craft, N. E., Ursin, G., Patterson, B. H. & Graubard, B. I. Cancer Research, 52, 2060s-2066s, (1992).
-
Helzlsouer, K. J., Block, G., Blumberg, J., Diplock, A. T., Levine, M., Marnett, L. J., Schulplein, R. J., Spence, J. T. & Simic, M. G. Cancer Research, 54, 2044s-2051s, (1994).
-
Slattery, M. L., Edwards, S. L., Anderson, A. & Cann, B. Nutrition and cancer, 30(3), 201-206, (1995).
-
Block, G., Patterson, B., Subar, A. F. Nutrition & Cancer, 18, 2-20, (1992).
-
Weisburger, J. H. Mutation Research, 402, 331-337, (1998).
-
Austoker, J. BMJ, 308, 1310-1314, (1994).
-
Giovannucci, E., Rimm, E. B., Stampfer, M. J., Colditz, G. A., Ascherio, A. & Willett, W. C. Cancer Research, 54, 2390-2397, (1994).
-
Doll, R. & Peto, R., J. Natl. Cancer Inst., 66, 1191-1308, (1992).
-
Burkitt, D. P., Cancer(Phila.), 28, 3-13, (1971).
-
Kritchevsky, D. Cancer Suppl. 72, 3, 1011-1014, (1993).
-
Shibata, A., Pagamin-Hill, A., Ross, R. K. & Henderson, B. E. Br. J. Cancer, 66, 673-679, (1992).
-
Hirono, I. Jpn. J. Cancer Res., 84, 997-1006, (1993).
-
Sugimura, T., Kawachi, T., Nagao, M., Yahagi, T., Seino, Y., Okamoto, T., Shudo, K., Kosuge, T. & Itai, A. Proc. Jap. Acad., 53, 58-61, (1997).
-
Eisenbrand, G. & Tang, W. Toxicology, 84, 1-82, (1993).
-
Ames, B. N., Gold, L. S. & Willett, W. C., Proc. Natl. Acad. Sci. USA, 92, 5258-5265, (1995).
-
Pence, B. C., Landers, M., Dunn, D. M., Chwan, L. I. & Miller, F. Nutrition & Cancer, 30(3), 220-226, (1998).
-
Gooderhan, N. J., Murray, S., Lynch, A. M., Edwards, R. J., Yadollaki-Farsani, N., Blatt, C. & Rich, K. J., British Jou. of Clin. Pharmacol., 42(1),91-98, (1996).
- Glatt, H.R. /intern/ernkrebs.htmlErnahrung und Krebs aus der Sicht des toxikologen.
-
Walter, D. R. & Willer, C. Science, 264, 530-538, (1994).
-
Katsouyanni, K., Trichopoulou, A., Struver, S., Garas, Y., Kritselis, A., Kyriakou, G., Boyle, P. & Trichopoulos, D., British Jou. of Cancer, 70(3):537-541, (1994).
-
Stoll, B. A. International Jou. of Obesity, 22(11), 1035-1040, (1998).
-
Snyderwine, E.G., Davis, C.D., Schut, H.A.& Roberts-Thompson,S.J. Carcinogenesis, 19(7):1209-1215, (1998).