Content uploaded by Patricia Wakimoto
Author content
All content in this area was uploaded by Patricia Wakimoto
Content may be subject to copyright.
© 2002 Nature Publishing Group
Maximum health and lifespan require metabolic har-
mony.It is commonly thought that the problem of
how to ensure adequate intake of the more than 40
essential micronutrients (vitamins, minerals and
other biochemicals that are required in small
amounts) has been solved for most of the world’s
population.Classical nutrient acute
DEFICIENCY diseases
such as scurvy, beri-beri, pernicious anaemia and
rickets are no longer prevalent.Acute deficiencies of
micronutrients are rare in developed countries, but
suboptimal nutrient intake — less than the
RECOM-
MENDED DAILY ALLOWANCE
(RDA) — is a widespread prob-
lem
(TABLE 1).Research indicates that considerable
metabolic damage can still occur when nutrient
intake levels fall below the RDA — even though they
might not cause acute disease. For example,the opti-
mum amount of folic acid or zinc that is required to
minimize DNA damage and maximize a healthy lifes-
pan is likely to be greater than the amount that is
needed to prevent acute disease.Furthermore,the
nutritional requirements of the elderly
2–4
are likely to
differ from those of younger people, and have not
been carefully examined.Nutritional requirements are
also likely to depend on genotype
5
.
Deficiencies in one aspect of the metabolic
network can cause repercussions in many systems.
A single deficiency,for example, can increase DNA
damage (and cancer),promote neuronal decay (and
cognitive dysfunction) or lead to mitochondrial dis-
ruption (accelerating ageing). The relationship
between diet and cancer has, historically, been
thought of in terms of exposure to potential carcino-
gens,such as alcohol or heterocyclic amines.Dietary
deficiencies,however, might be a much more impor-
tant factor in cancer risk.Evidence for a link between
various micronutrient deficiencies and DNA damage
has been accumulating in recent years
6–9
,but this has
been difficult to study and,as a result, has not been
the main focus for epidemiology researchers
10
.
The importance of nutrition in cancer development
is actively studied and is controversial.Several studies
have examined the association between diet and
specific cancers,including breast
11,12
,prostate
13
and col-
orectal cancer
14
.The associations between cancer and
specific dietary factors, such as meat,fruits and vegeta-
bles,and specific nutrients, such as vitamin D and sele-
nium,have been established
15–18
.The most convincing
epidemiological evidence for the role of dietary factors
in cancer risk is the inverse relationship between the
consumption of fruits and vegetables and many types
of cancers.Comprehensive reviews have shown that
people who consume the fewest fruits and vegetables,
compared to those who consume the most, have a
higher cancer incidence
16,19–21
.Most CASE–CONTROL STUDIES
ARE VITAMIN AND MINERAL
DEFICIENCIES A MAJOR
CANCER RISK?
Bruce N. Ames and Patricia Wakimoto
Diet is estimated to contribute to about one-third of preventable cancers — about the same
amount as smoking. Inadequate intake of essential vitamins and minerals might explain the
epidemiological findings that people who eat only small amounts of fruits and vegetables have
an increased risk of developing cancer. Recent experimental evidence indicates that vitamin
and mineral deficiencies can lead to DNA damage. Optimizing vitamin and mineral intake by
encouraging dietary change, multivitamin and mineral supplements, and fortifying foods might
therefore prevent cancer and other chronic diseases.
DEFICIENCY
Dietary intake of a vitamin or
mineral at a level that is less than
50% of the recommended daily
allowance — as distinguished
from acute deficiency. For
example,acute vitamin C
deficiency causes scurvy.
RECOMMENDED DAILY
ALLOWANCE
(RDA). The dietary-intake level
that is sufficient to meet the daily
nutrient requirements of nearly
all healthy individuals in a
defined group.
Nutrition Genomics Center,
Children’s Hospital Oakland
Research Institute, 5700
Martin Luther King Jr Way,
Oakland,California
94609-1673,USA.
Correspondence to B.N.A.
e-mail: bames@chori.org
doi:10.1038/nrc886
REVIEWS
694 | SEPTEMBER 2002 | VOLUME 2 www.nature.com/reviews/cancer
© 2002 Nature Publishing Group
NATURE REVIEWS | C ANCER VOLUME 2 | SEPTEMBER 2002 | 695
REVIEWS
CASE–CONTROL STUDY
An epidemiological study design
in which individuals are selected
based on the presence (case) or
absence (control) of disease.
Well-designed case–control
studies require that the two
groups are derived from the
same population.
PROSPECTIVE COHORT STUDY
An epidemiological study design
in which individuals with
known characteristics (such as
occupational exposure,smoking
and exercise) are enrolled and
followed over time for specific
outcomes.The rate of cancer (or
other disease) in the exposed
population is compared to that
in the unexposed population.
ME TA- ANA LYS IS
A retrospective analysis of the
results from different studies,
making certain assumptions, to
reach a conclusion that is based
on the pooled data.
than fruits and vegetables.Increased consumption of
whole grains has also been associated with a
decreased risk of several cancers
26,27
.A M ETA-ANALYSIS of
40 case–control studies showed that whole grains
were associated with a 40% decreased risk of devel-
oping stomach cancer and a 20% decreased risk of
developing cancers of the rectum and colon
27
.
Although epidemiological studies indicate an overall
higher cancer risk in people who consume the fewest
fruits and vegetablesor the fewest whole grains
(TABLE 2),
they have not, however, established a causal link
between consumption of these foods and cancer.
Also, it is not known which of the many dietary
constituents in fruits and vegetables or whole grains
are responsible for, or the cellular or molecular
processes that confer, the protective effects.It is worth
noting that meat — the main food source of iron,
zinc and B12 — is also an important source of
micronutrients
(TABLE 3).
The mechanisms of action of dietary micronutri-
ents are complex and are not fully understood.
Micronutrients might function as antioxidants,anti-
mitogens, anti-mutagens or in other ways
28–31
.
Impending research by both bench scientists and epi-
demiologists on the many factors that are involved in
the diet–cancer relationship,including gene–environ-
ment interactions,should begin to clarify the complex
relationship between diet and cancer
32
.Though DNA
damage is a well-established risk factor for cancer cau-
sation,it should be emphasized that cell-division rates
and other factors also contribute
33–37
.
Establishing a link between micronutrient intake,
DNA damage and cancer is only one important area
of research on diet and cancer,but it lends itself to an
inexpensive and practical solution. ‘Tuning up’
metabolism could reduce the incidence of cancer,as
well as have other health benefits,at little cost — a
full year’s supply of daily multivitamin and mineral
pills for one person costs less than a few packs of cig-
arettes.In this article,we focus on cancer protection
that is conferred by vitamins such as folate,B12, B6
and C,as well as the minerals iron and zinc
(TABLE 3).
indicate that a reduced consumption of fruits and
vegetables can double the risk of developing most
types of cancer
16,19–21
(TABLE 2). In 1992, a compre-
hensive review by Block and colleagues
19
found that
75% of nutritional studies reported a significant asso-
ciation between cancer protection and the consump-
tion of fruits and vegetables.A more recent review
(which did not include cohort studies) reported a
similar result,stating that 77% of nutritional studies
associated fruit and vegetable consumption with can-
cer protection
(TABLE 2).It should be noted that, for
breast and colon cancer,some large-scale
PROSPECTIVE
COHORT STUDIES
and recent case–control studies have
failed to show a link with low levels of fruit and
vegetable consumption
22–25
.
Whole grains are a better source of some vitamins
or minerals (such as vitamin B6 and magnesium)
Summary
• Acute deficiencies of vitamins and minerals are rare in developed countries, but
suboptimal nutrient intake — less than the recommended daily allowance (RDA)
— is a widespread problem. Research indicates that considerable metabolic damage
can still occur when nutrient intake levels fall below the RDA — even though they
might not cause acute disease.
• Evidence indicates that deficiencies of iron and zinc, and the vitamins folate,B12, B6
and C,can cause DNA damage and lead to cancer.
• New animal bioassays of nutritional deficiencies are needed,particularly for
studying cancer.
• Reduced folate intake has been associated with cancer.Folate,B6 and B12
deficiencies cause the incorporation of deoxyuracil into DNA,leading to DNA
breakage,and could promote tumorigenesis.
• The relationship of vitamin and mineral deficiencies and cancer is extremely
complex.An integrated analysis of the findings from epidemiological,animal-
model, metabolic and intervention studies,as well as from genetic polymorphism
research,is required.
• Approaches to eliminating micronutrient deficiencies include improving diet,
fortifying foods and providing multivitamin and mineral supplements. Prevention
strategies such as these could have a significant impact on cancer and public health,
with minimal risk being involved.
Table 1 | Micronutrient deficiencies in US individuals
Nutrient Population group Current RDA % Consuming less than % Consuming less than
the RDA half the RDA
Minerals
Iron Women 20–30 years 18 mg 75% 25%
Women 50+ years 8 mg 25% 5%
Zinc Women/men 50+ years 8/11 mg 50% 10%
Vitamins
Folate* Women 20+ years 400 µg 75% 50%
Men 20+ years 400 µg 75% 25%
B6 Women/men 20+ years 1.5/1.7 mg 50% 10%
B12 Women 20+ years 2.4 µg 25% 10%
Men 20+ years 2.4 µg 10% 5%
C Women/men 20+ years 75/90 mg 50% 25%
*Folate intake before US fortification in 1998. RDA, recommended daily allowance. Data adapted from REF.143; dietary intakes include
food fortification, but not supplement use.
© 2002 Nature Publishing Group
696 | SEPTEMBER 2002 | VOLUME 2 www.nature.com/reviews/cancer
REVIEWS
example,phytic acid (inositol hexaphosphate),which
is found at high levels in cereal grains and legumes,
forms a tight complex with zinc or iron that decreases
absorption. Circulating levels of nutrients also
depend on transport proteins or other co-enzymes,
and marker analysis can be affected by personal
habits,such as smoking, drug use or dietary factors.
Aspects of epidemiological study design can also
affect results
(BOX 1).
Because it is so difficult to determine the exact levels of
micronutrient intake in a person’s diet,the most direct
approach to assaying nutrient effects would be to
perform supplementation studies. Smaller clinical or
metabolic studies ofsubpopulations of individuals with
identified genetic alterations might be more valuable than
large epidemiological studies that include several geno-
types.
INTERVENTION STUDIES that measure chromosome
breaks or other DNA damage in small numbers of people
with a low intake of a micronutrient before, during and
after supplementation with the micronutrient might be
the most successful research approach.
Patterson et al.
28
published a comprehensive review
of epidemiological studies from 1980 to 2000,summa-
rizing the results of randomized,controlled cancer trials
that assessed the association between intake of
micronutrient supplements and cancer.Although some
individual studies of micronutrient supplement intake
have associated nutrient supplements with lower cancer
risk, the authors concluded that there is not sufficient
high-quality data on which to base firm conclusions.
Experimental models. In model systems that attempt
to correlate diet with cancer risk, there is always
uncertainty as to whether results obtained from the
Challenges to nutrition research
Epidemiological analysis. Epidemiological studies
investigate the link between exposure to one or more
variables, and a defined outcome,such as the inci-
dence of particular cancers in a carefully described
population. The exposure can be a nutritional
variable such as dietary pattern, consumption of an
individual food, nutrients or non-nutrient compo-
nents of foods, or chemical alterations that occur
during cooking or preservation. Diet is therefore very
complex to measure.
These complexities are multiplied when researchers
attempt to measure components of foods such as
micronutrients.Micronutrient status can be mea-
sured by markers in blood,urine or tissue, although
the use of biological markers to measure micro-
nutrient status has its own limitations. Nutrients are
never isolated alone,and the presence of one is usually
associated with the presence of another.For example,
β-carotene and vitamin C are markers for fruit and
vegetable intake. Studies that report on the effects
of a specific nutrient should therefore always be
viewed cautiously.
Another complication in forming associations
between diet and cancer is that the investigator must
make an a priori decision about the pertinent expo-
sure time,such as whether to study the cumulative
exposure over time, average exposure over time or
peak exposure at a crucial time, such as the first
trimester of pregnancy.There are also
CONFOUNDING
FACTORS
that must be considered when making associa-
tions with diet, such as sociodemographic factors,as
well as absorption, bioavailability,transport distribution
and measurement of a particular nutrient
38
.For
CONFOUNDING FACTOR
These occur because behaviour-
related variables of interest tend
to cluster.An exposure (for
example,vegetable
consumption) might be of
interest in protecting against a
particular cancer. However, if
smokers eat fewer vegetables
than non-smokers,we might
falsely attribute a risk reduction
to vegetables that is really owing
to the fact that a higher
proportion of vegetable-eaters
are non-smokers.Confounding
factors can be controlled for by
separating the smokers and the
non-smokers and asking
whether the vegetable–cancer
association is seen in both
groups, or by more
sophisticated,but conceptually
similar,statistical techniques.
INTERVENTION STUDY
Often called a clinical trial or
experimental study,an
epidemiological analysis of a
hypothesized cause–effect
relationship that is performed by
modifying a supposed causal
factor,such as lack of vitamin C
consumption,in a population.
Table 2 | Epidemiological studies showing cancer protection from fruits and vegetables
Cancer site 1992 review: 1992 review: 1997 review:
fraction of studies showing relative risk (median) (low fraction of studies
significant cancer protection
19
versus high quartile of showing significant
consumption)
19
cancer protection
16
Epithelial
Lung 24/25 2.2 11/13
Oral 9/9 2.0 13/15
Larynx 4/4 2.3 6/8
Oesophagus 15/16 2.0 15/18
Stomach 17/19 2.5 28/31
Pancreas 9/11 2.8 9/11
Cervix 7/8 2.0 4/6
Bladder 3/5 2.1 6/8
Colorectal 20/35 1.9 3/6
Colon 15/21 – –
Miscellaneous 6/8 – –
Hormone dependent
Breast 8/14 1.3 8/12
Ovary/endometrium 3/4 1.8 7/9
Prostate 4/14 1.3 1/6
Total 129/172 (75%) 126/164 (77%)
© 2002 Nature Publishing Group
NATURE REVIEWS | C ANCER VOLUME 2 | SEPTEMBER 2002 | 697
REVIEWS
RECALL BIAS
Occurs in individuals that
describe events (such as
exposures,diseases and
pregnancy outcome) of the past
in a non-comparable manner.It
is primarily a problem in
case–control studies when that
presence of the disease in one
group might result in differential
recall (for example,of alcohol
consumption or dietary
behaviour) between the cases
and controls.
Folate
Our current understanding of folate deficiency and its
relationship to cancer illustrates the importance of
considering the findings of all types of research —
epidemiological, molecular, clinical and interven-
tional — when investigating the link between diet and
cancer. There is much epidemiological evidence indi-
cating that low folate intake increases the risk of many
types of cancer
(TABLE 4).As shown in TABLE 4,reduced
folate intake has been associated with a higher risk of
colon cancer
14,41
,and long-term use of a folate-con-
taining supplement has been shown to lower the risk
of colon cancer by 75%
42
.There is also evidence that
low folate intake increases the risk of breast cancer
43,44
,
pancreatic cancer in smokers
45,46
, and gastric and
oesophageal cancers
47
.
In vitro studies have shown that folic-acid defi-
ciency causes a dose-dependent increase in uracil
incorporation into human lymphocyte DNA
48
(FIG.1).
Folate administration reduces DNA uracil incorpora-
tion and the occurrence of chromosome breaks in
human cells
7
.Ex vivo experiments have shown that
all the markers of chromosome damage in human
experimental system accurately reflect processes that
occur in tissues or whole organisms.Some studies
have used animal models to address the role of
dietary factors in causing or preventing cancer, lead-
ing to hypotheses about physiological mechanisms
and predictions about dose–response relationships
between dietary compounds and cancer risk.
However, there are limitations to extrapolating the
findings of animal studies to humans,owing, in part,
to differences in metabolic pathways,rates and lifes-
pans. In some cases, there are known differences
between species in the metabolism of vitamins.
Vitamin C, for example, is synthesized by rats and
mice,but must be provided in the diets of guinea pigs
and humans.
Relatively few animal studies have examined the
link between vitamin or mineral deficiency and can-
cer. Better animal models are needed to elucidate
mechanisms that are related to cancer prevention.
This might be a more effective strategy than testing
the effect of exposure to high doses of synthetic
chemicals, when humans are typically exposed to
only low doses
39,40
.
Box 1 | Limitations to specific types of epidemiological studies
A number of factors make it especially complex to perform epidemiological studies to associate diet and disease. In
case–control studies,participants are susceptible to
RECALL BIAS. Prospective cohort studies might not cover the period of
time during which subjects experienced a micronutrient deficiency that was crucial for the development of cancer.For
example, the predisposition to cancer might have occurred during germ-cell or fetal development,or during childhood.
Study duration might not also have had sufficient follow-up time for cancers to be manifest.
Most epidemiological studies use questionnaires to measure dietary intake,rather than assaying micronutrient levels
directly. In carefully designed studies,nutrient levels in the blood are analysed in combination with dietary data. One of
the primary reasons that epidemiological studies are sometimes inconclusive is that conclusions are drawn from data
that might contain measurement error in estimating micronutrient intake.Dietary questionnaires must be carefully
designed to cover all appropriate foods, nutrients and supplements.Nutrient estimates are derived from databases or
food-composition data tables and must be updated to reflect the ever-changing food supply,and to provide an accurate
assessment of nutrient intake.Another important methodological issue is that trials should carefully monitor dose and
duration of use of supplements, measurement of long-term micronutrient intake, and other aspects of a healthy lifestyle.
A final important methodological problem in performing epidemiological analyses of micronutrient
intervention and disease is that data from most people, who usually consume saturation levels of a particular
micronutrient, are combined with those of the smaller percentage of people (generally 10–20%), who have an
inadequate intake,which makes it easier to miss an association. Careful design and appropriate analyses of data are
required to help overcome some of these challenges.
Table 3 | Micronutrient sources and main contributing foods in the Western diet
Nutrient Richest food sources Primary sources in US diet
Folate Fruits and vegetables, including dark greens and Fortified cold cereal*, orange or grapefruit juice,
dried beans green salad, fibre or bran cereals , white bread
‡
Vitamin B12 Meat, fish, milk products, fortified cereals Beef, fortified cold cereal*, shellfish, low-fat milk
Vitamin B6 Fortified cereals, whole grains, meat Fortified cold cereal*, white potatoes, bananas,
chicken or turkey, beef
Vitamin C Citrus fruits and vegetables Orange or grapefruit juice, drinks/juices with
vitamin C, other fruit
Iron Meat Cold cereal
‡
, fibre or bran, white bread
‡
, rolls
‡
,
buns
‡
, bagels
‡
, pizza
‡
, hamburgers, other beef
Zinc Meat, eggs, nuts Beef, fortified cold cereal*, cheese or cheese
spread, mixed dishes with beef/pork/veal/lamb
*Fortified with essential vitamins and minerals.
‡
Enriched/fortified with iron, riboflavin, thiamine and niacin. Data adapted from REF.144.
© 2002 Nature Publishing Group
698 | SEPTEMBER 2002 | VOLUME 2 www.nature.com/reviews/cancer
REVIEWS
Unfortunately,the MTHFR-TTgenotype is not all
good news.As the lower level of MTHFR decreases the
methyl-THF pool,it increases serum levels of homocys-
teine — a risk factor for endothelial-cell damage and
cardiovascular disease
32
.Individuals who are homozy-
gous for this allele have a twofold increase in plasma
homocysteine levels.
Low dietary intake of folate, B12 or B6 has been
associated with a higher risk for developing cervical
squamous epithelial lesions in a case–control study
60
.In
contrast to studies mentioned above,individuals with the
MTHFR-TT genotype also had a higher incidence of
cervical cancer.The authors suggest this is due to a lack of
inhibitory methylation — that results from a decreased
methyl-THF pool — of human papillomavirus (HPV),
which is a significant cause of cervical cancer
60
.
Why is this polymorphism so frequent in popula-
tions in the northern United States? Although it is
credible that these populations were, historically,
chronically folate deficient because of their diet, can-
cer and heart disease come too late in life to select for
individuals with a lower risk. The MTHFR-TT geno-
type might be selected for based on its ability to
reduce uracil incorporation into sperm DNA,which
consequently results in less DNA damage in
offspring
6,61
.Levels of non-methyl THFs are posi-
tively associated with sperm count and density
61
.
Folate levels are therefore also important for male
reproductive function,and further support the con-
cept that folate deficiency can cause DNA damage.
Germ-line damage to the sperm has also been linked
to childhood cancers
62
.Further studies are required
to determine if ALL is higher in children whose
fathers have a poor diet
59
.
lymphocytes are minimized at folic-acid concentra-
tions that are higher than the RDA
48–50
.Folate supple-
mentation above the RDA has also been shown to
reduce chromosome breakage in humans
49,50
.
Although this study did not examine cancer inci-
dence,chromosome breaks have been linked to cancer
in other studies
51
.
A clue to the folate–cancer connection was the
discovery of a polymorphism (C677T) in the gene
that encodes methylene-THF reductase (MTHFR) —
the enzyme that reduces methylene-tetrahydrafolate
(CH
2
=THF) to methyl THF
(FIG.1)
.This polymor-
phism decreases the activity of MTHFR, which
increases the methylene-THF pool at the expense of
the methyl-THF pool,resulting in decreased incor-
poration of uracil into DNA and an increased
number of chromosome breaks.It is a common poly-
morphism in populations of people who live in
northern regions of the world,with 5–25% of indi-
viduals being homozygous for this polymorphism
and up to 50% being heterozygous.
Several studies have shown a two- to fourfold lower
risk of colon cancer in individuals who are homozygous
for the 677T allele of methylene-THF reductase
(MTHFR-TT), compared with individuals who are
homozygous for the C677allele and have a high folate
intake
52–55
.At low folate levels,however,the MTHFR-TT
genotype does not seem to be protective and might even
be a risk factor.Other studies on adenomatous polyps
show an increased risk for developing the MTHFR-TT
genotype
53,56,57
. Adult acute lymphocytic leukaemia
(ALL)
58
and childhood ALL
59
have also been inversely
associated with the MTHFR-TTgenotype,indicating
that folate deficiency might promote ALL.
Table 4 | Evidence for folate, B6 and B12 deficiency and cancer risk
Type of cancer Comments References
Colorectal cancer, Lower intake of folate was associated with higher risk of colon cancer; 14,41,42
adenomas long-term use of folate supplement lowers the risk of colon cancer by 75%
No association was seen between folate, B6 and B12 and risk of colorectal 146
hyperplastic polyps
Breast cancer Strong inverse association was seen between folate intake and risk of breast 43,44
cancer among women who drink alcohol, which interferes with folate absorption
No association was seen between folate and B6 and breast cancer 67
Pancreatic cancer Risk of pancreatic cancer in smokers was inversely associated with dietary folate 45,46
Oesophageal and Folate and vitamin B6 were inversely associated and vitamin B12 was positively 47
gastric cancers associated with these cancers in a case–control study
Acute lymphoblastic ALL was associated with the MTHFR genotypes, indicating a role of folate in 59
leukaemia the development of ALL
(ALL; children)
Acute lymphocytic A significant reduction in risk of ALL was found in those with the MTHFR 677TT 58
leukaemia genotype, indicating that folate deficiency might be a risk factor
(ALL; adults)
Cervical cancer Risk of invasive cervical cancer was elevated for women with higher serum 70
homocysteine
Dietary intakes of folate, B6 and B12 were inversely related to the risk of developing 60
cervical dysplasia
No statistically significant association between folate, B12 and cervical cancer 66
was found
Prostate cancer Lower intake of B6 was associated with prostate cancer 68
Lung cancer Lower intake of B6 (higher B6 serum level) was associated with lung cancer 69
© 2002 Nature Publishing Group
NATURE REVIEWS | C ANCER VOLUME 2 | SEPTEMBER 2002 | 699
REVIEWS
Vitamin B6 deficiency would also be expected to
result in uracil misincorporation. B6 deficiency causes
a decrease in the enzyme activity of serine hydrox-
ymethyl transferase, which supplies the methylene
group for methylene-THF
65
(FIG.1).So the methylene-
THF pool is decreased in instances of B6 deficiency,
leading to uracil incorporation and subsequent chro-
mosome breaks (B.N.A.,T. Shulz and S.Mashiyama,
unpublished observations). Several studies have
shown that,in individuals with the MTHFR-TTgeno-
type, vitamin B6 intake is associated with protection
against colon cancer and/or adenomas
53,55,57
.
Recent epidemiological studies have indicated an
association between B6 or B12 deficiency and cancer
prevention, but results are mixed
47, 66, 67
.In one
case–control study of diet and cancer, vitamin B6
deficiency was associated with prostate cancer
68
.A
significantly lower risk of lung cancer was found in
men with higher serum B6 levels
69
. Compared to
men with the lowest vitamin B6 concentrations,men
in the highest quintile had about one-half the risk of
lung cancer. No associations were made between B12,
folate or homocysteine and lung cancer. Serum
homocysteine levels were found to predict the risk
of developing invasive cervical cancer in a large
case–control study
70
.
Oxidation, micronutrients and cancer risk
Oxidants,such as radiation, are known mutagens, so
antioxidant micronutrients, such as vitamins C and E,
might function as anti-mutagens and anti-carcino-
gens
71,72
.The evidence that supplementation with these
vitamins lowers cancer risk is inconclusive.
Vitamin C. Both experimental and epidemiological
data indicate that vitamin C protects against stomach
cancer
47,71,73
.This is a plausible conclusion,as oxidative
damage from inflammation caused by Helicobacter
pylori infection is a risk factor for stomach cancer
74
.
Fruit and vegetable intake — the main dietary source
of vitamin C — is also inversely associated with stom-
ach cancer
(TABLE 1).Mayne and colleagues have also
reported an inverse association between vitamin C and
oesophageal adenocarcinoma
47
.
Many other studies, however, have reported
no effect of vitamin C on cancer risk. A thorough
review of intervention studies showed both posi-
tive
14
and negative
12
studies,and so the evidence is
inconclusive
75
.There are several reasons why a posi-
tive effect might not have been observed.The blood-
cell saturation of vitamin C occurs at about 100
mg/day in humans
76
.Evidence indicates that this level
minimizes DNA damage
9,77–79
.Perhaps the differing
results from various studies were caused by differ-
ences in whether vitamin C reached tissue saturation
levels in the population that was studied.If only a
small proportion of the population had inadequate
tissue saturation by vitamin C intake, a real effect
would be missed
(TABLE 1).Other factors that might
explain the difference in results include failure to
adequately assess vitamin C intake,failure to assess
Vitamins B12 and B6
Vitamin B12 deficiency would be expected to cause
chromosome breaks by the same uracil-misincorpo-
ration mechanism that is found with folate defi-
ciency
32
.Both B12 and methyl-THF are required for
the methylation of homocysteine to methionine
(FIG.1).
If cells are deficient in either folate or B12,homocys-
teine accumulates.When B12 is deficient,tetrahydro-
folate is trapped as methyl-THF,reducing the methyl-
ene-THF pool,which is required for methylation of
deoxyuridine monophosphate (dUMP) to
deoxythymidine monophosphate (dTMP). B12 defi-
ciency, like folate deficiency, therefore causes uracil to
accumulate in DNA (B.N.A., unpublished observa-
tions).In a study of healthy elderly men
63
and young
adults,increased chromosome breakage was associ-
ated with low dietary intake of either B12 or folate,
or with elevated levels of homocysteine
49,63,64
. B12
supplementation above the RDA was necessary to
minimize chromosome breakage
49,50
.
CH
2
=THF
TS
CH
3
-THF
MS
B12
B6
dUMP dTMP
SHMT
MTHFR
Serine
Methionine
Homocysteine
THF
DNA
Figure 1 | Incorporation of uracil in DNA. Folate (THF),
B6 or B12 deficiencies have all been associated with
cancer. Cellular depletion of any of these vitamins can
induce DNA damage by causing deoxyuridine (dUTP),
instead of deoxythymidine (dTTP), to be incorporated into
DNA. Methylene-THF reductase (MTHFR) converts
methylene-THF (CH
2
=THF) to methyl-THF (CH
3
-THF). The
CH
2
=THF pool is derived from folate and is required for the
methylation of deoxyuridine monophosphate (dUMP) to
deoxythymidine monophosphate (dTMP) by thymidylate
synthetase (TS). Folate deficiency decreases the amount of
CH
2
=THF, causing dUTP to accumulate in DNA (dashed
line), rather than dTTP, which results in chromosome
breaks
7
. B6 deficiencies decrease the activity of the enzyme
serine hydroxymethyl transferase (SHMT), which is required
to produce CH
2
=THF. The resulting smaller CH
2
=THF pool
results in dUTP incorporation into DNA. Methionine
synthetase (MS), a B12 and CH
3
-THF-dependent enzyme,
converts homocysteine to methionine. When B12 is
deficient, THF is trapped as CH
3
-THF. This reduces the size
of the CH
2
=THF pool, leading to increased dUTP
incorporation into DNA. When dUTP is incorporated into
DNA, it is normally excised by a glycosylase repair enzyme,
which induces transient single-strand breaks in the DNA.
Two opposing single-strand breaks cause a double-strand
chromosome break. Double-strand breaks are difficult for
cells to repair, and increase cancer risk
51
.
© 2002 Nature Publishing Group
700 | SEPTEMBER 2002 | VOLUME 2 www.nature.com/reviews/cancer
REVIEWS
Smoking is another producer of oxidative stress.A
smoker needs to consume 40% more vitamin C than a
non-smoker in order to maintain a comparable blood-
plasma level of vitamin C
91
.Several studies have exam-
ined the associations between paternal smoking,
vitamin C intake,oxidative damage to sperm DNA and
childhood cancer in the offspring. Smoking depletes
vitamin C, which is required to protect DNA in sperm
against oxidative damage
9,92,93
.Smokers,or men with
low ascorbate intake, have lower seminal-plasma
ascorbic-acid levels and higher levels of oxidative DNA
damage in their sperm than either non-smokers or
men with adequate ascorbate intake
9,92
.
Unfortunately, there have been few studies that have
rigorously examined the effect of paternal vitamin C
level by itself on cancer in offspring.Nevertheless, there
is evidence that children with fathers who smoke have
an increased rate of childhood cancer
94–97
.An epidemio-
logical study from China makes a particularly strong
case that ALL,lymphoma and brain cancer are each
increased three- to fourfold in children of male
smokers
97
.The associations were strongest in men with
the highest number of
PACK YEARS of smoking. It seems
likely, given the available evidence,that the cancer risk to
offspring of male smokers would be higher when
dietary antioxidant intake is low.
Other micronutrients in addition to iron and zinc,
which are discussed below, have antioxidant potential.
These include carotenoids,vitamin E and selenium.
These have been the focus of experimental and epi-
demiological research to determine the association with
cancer risk
16
(BOX 2).
whether people were using vitamin supplements or
failure to take into account the effect of modifiers
such as body-mass index or smoking (which decreases
plasma vitamin C levels)
80
.
Many studies have investigated the effects of vita-
min C supplementation in humans using biomarkers
of oxidative damage to DNA,lipids (lipid oxidation
releases mutagenic aldehydes) and protein
81–85
.For
example, intervention studies with antioxidant supple-
ments (100 mg per day of vitamin C,28 mg per day of
vitamin E and 25 mg per day of β-carotene) were
found to decrease DNA strand breaks in lymphocytes,
as measured by the
COMET ASSAY
77
.Subsequent studies
showed β-carotene by itself was ineffective at reducing
the number of DNA breaks
86
,but that vitamin C was
effective alone
87
.
Studies in rats have shown that the spontaneous
oxidative damage occurs at a rate of about 66,000
DNA adducts per diploid cell
88,89
and,unlike uracil
misincorporation, is likely to occur with equal fre-
quency on both strands. Repair of oxidative adducts
by glycosylase results in transient single-strand breaks
in DNA. Increased oxidative damage is therefore
associated with low vitamin C intake
9
.Individuals
who are deficient in both folate and antioxidant
intake would have higher levels of both oxidative
damage and of uracil incorporation in their DNA,
and be expected to have a high level of double-strand
DNA breakage.Radiation (an oxidative mutagen)
and folate deficiency have been shown to act syner-
gistically in causing chromosome breakage in tissue-
culture cells
90
.
COMET ASSAY
A technique that uses
electrophoresis of immobilized
single cells to measure DNA
strand breaks.
Box 2 | Other nutritional factors linked to cancer
Evidence continues to accumulate to support the importance of several other micronutrients, such as vitamin D,
calcium, niacin and selenium,in cancer development. Epidemiological findings, although not entirely consistent,
indicate that there is a relationship between vitamin D deficiency and several types of cancer,primarily colorectal
cancer and colorectal adenomas
18,127
.Vitamin D is a hormone that is synthesized in skin that has been exposed to
ultraviolet light.Vitamin D deficiency increases cell proliferation, and so is a risk factor for cancer
128
. Populations in
northern regions of the world are chronically vitamin D deficient unless they drink fortified milk. People with darker
skin in northern regions of the world who don’t drink fortified milk (dark skin is associated with lactose intolerance)
are at the greatest risk for vitamin D deficiency and cancer.
Several studies have reported a weak association between increased calcium intake and decreased risk of colorectal
cancer, but high calcium intakes (above the recommended daily allowance (RDA)) have also been associated with an
increased risk of prostate cancer
129,130
. More research is needed to clarify the role of calcium, vitamin D and related factors
in colorectal and prostate cancer.
Epidemiological studies, including cross-sectional, case–control and prospective studies,support the inverse
association between serum selenium levels and lung,colorectal and prostate cancer in men
131–133
. Several selenocysteine-
containing proteins are part of the antioxidant defence network. The evidence for selenium has been felt to be sufficient
to justify a large clinical trial on prostate cancer
134
.In a large international case–control study, an increase in niacin —
one of the B vitamins — along with antioxidant nutrients were found to be associated with a decrease in oral cancer (the
mouth, pharynx and oesophagus)
135,136
. Niacin is a component of NAD, which is involved in the polyADPribose defence
against DNA strand breaks.
Other nutritional factors that are implicated in cancer include obesity/type II diabetes,alcohol,fat, lack of fibre
and phytochemicals
12,137
.Research on obesity and cancer is an active area; it includes research that examines not
only body-mass index, but also factors that are intimately related to obesity and being overweight, such as physical
activity and the hormone insulin
138–141
. The relationships between fibre, fat,phytochemicals and cancer have been
extensively reviewed
16,142
, but findings are not consistent.The effect of fibre is difficult to separate from other
highly correlated dietary variables in whole grains, fruits and vegetables, whereas the effect of fat is difficult to
separate from calorie intake.
© 2002 Nature Publishing Group
NATURE REVIEWS | C ANCER VOLUME 2 | SEPTEMBER 2002 | 701
REVIEWS
shown to cause this cancer
115
.Replenishment of zinc in
zinc-deficient rats can induce apoptosis in oesophageal
epithelial cells, which could reduce the risk of
oesophageal cancer
118
.
Zinc is essential for testicular development and
spermatogenesis
119
.Zinc concentration is hundreds of
times greater in seminal plasma than in blood plasma,
and it is believed to be important for spermatogenesis
and maintaining the stability of spermatozoa
106
.Zinc
concentrations in blood and seminal plasma are
correlated positively with sperm-cell density, and
lower zinc concentrations are found in infertile men
compared with fertile men
120
. Combined zinc and
folate supplementation increased the total normal
sperm count in subfertile men in a randomized inter-
vention study, although an independent effect of the
two nutrients was not seen
121
.Zinc deficiency leads to
increased oxidative damage to testicular-cell DNA (as
measured by oxo
8
dG); this might also contribute to
childhood cancer
113
.
Conclusions and future directions
More than 40 micronutrients are required in the
human diet. The effects of deficiencies of these
micronutrients,both individually and in combina-
tion,often leads to DNA damage,which can lead to
cancer. These effects can be studied in human cell cul-
tures,and followed-up through intervention studies
in people that have dietary deficiencies in specific
micronutrients. Another active area of nutrition
research has been the study of polymorphisms in
genes that encode metabolic enzymes, and how these
affect cell processing of particular micronutrients.
This research is useful in establishing causality
between micronutrient deficiencies and cancer
32,58–60
.
Micronutrient deficiencies are relatively common,
but can be easily remedied.Approaches to eliminat-
ing micronutrient deficiencies include improving
diet,fortifying foods and encouraging multivitamin
and mineral supplements that contain the RDA levels.
Cancer-prevention strategies such as these could have
a significant impact on public health, with minimal
risk being involved
122–124
.
Though consumption of the RDA levels of
micronutrients is of minimal risk, some people take
too many supplements. Mae West’s quote “too much
of a good thing is wonderful”does not apply to nutri-
tional supplements.Too much of a mineral or even a
vitamin can be toxic.The RDA committees are now
listing upper limits (UL) for minerals and vitamins.
Cancer-intervention trials of β-carotene, which
involved doses that were well above those that would
be obtained from a normal diet, had a deleterious
effect, reinforcing the importance of exceeding the
RDA of micronutrients
125,126
.
People who eat very few fruits and vegetables are
likely to have an inadequate intake of many micronu-
trients, such as folic acid and vitamin C, which
contributes to DNA damage, cancer and other
degenerative diseases.In addition, dietary deficiencies
of micronutrients that are not derived primarily from
Iron. The epidemiological data on iron and cancer are
mainly limited to studies of iron excess.Excessive iron
has long been known to catalyse oxidation in vitro.
Increased risk of human cancer is associated with excess
iron
98,99
. The increased risk of hepatic carcinoma in
individuals with cirrhosis caused by
HAEMOCHROMATOSIS
indicates a link between iron overload and cancer.
Several epidemiological studies have reported associa-
tions between increased iron status and colorectal
cancer.A recent review of 26 publications on iron and
colorectal cancer risk found that approximately three-
quarters of the larger studies supported the association
of excess iron with colorectal cancer risk
100
.Excess iron
also seems to lead to oxidative DNA damage in rats,
which is reversed by vitamin E
101
.
But iron deficiency,as well as iron excess,leads to
oxidative DNA damage
102
.Iron deficiency is one of the
most common micronutrient deficiencies — it affects
2 billion women and children worldwide,and about
25% of menstruating women in the United States
alone
(TABLE 1)
.How does iron deficiency cause oxida-
tive damage? One mechanism involves haem defi-
ciency (haem uncouples mitochondria and causes
oxidant release and mitochondrial DNA damage),as
well as the loss of some important iron-containing
defence enzymes, such as catalase and haem oxygenase
II
103
.Data from epidemiological studies that link iron
deficiency to cancer incidence, however, are lacking.
Zinc.Intake of the trace element zinc is below the
amount that is considered to be adequate or desirable
in many populations
104
(TABLE 1).Zinc is found in all
body tissues and is one of the most abundant intra-
cellular elements.Acute zinc deficiency is not easily
achieved in adult humans
105
,but when it does occur
it causes various health effects
104,106,107
.Zinc is a com-
ponent of more than a thousand DNA-binding pro-
teins that contain zinc fingers,as well as copper-zinc
superoxide dismutase,the oestrogen receptor and
synaptic transmission proteins
106
.The TP53gene,
which encodes p53 in humans,is mutated in half of
human tumours,and loss of zinc binding disrupts its
ability to mediate the DNA-damage response
108,109,110
(E.Ho and B. N.A.,unpublished observations). Zinc
deficiency also causes loss of function of zinc-con-
taining DNA-repair enzymes
110
,thereby compromis-
ing the ability of the cell to repair the damage and so
promoting tumorigenesis.
Rats placed on a zinc-deficient diet and the off-
spring of zinc-deficient rhesus monkeys both have a
higher incidence of chromosome breaks
111,112
.These
chromosome breaks seem to be caused by oxidative
damage
111,113
,which could result from a loss of activity
of copper-zinc superoxide dismutase or formamidopy-
rimidine glycosylase — a zinc-containing DNA-repair
enzyme that repairs oxidized guanine
114
.
Zinc deficiency might also contribute to oesophageal
cancer in humans.In conjunction with a single low
dose of a chemical carcinogen (nitrosamine),it has
been shown to cause oesophageal tumours in
rats
115–117
, and zinc deficiency alone has also been
PACK YEARS
The number of years of tobacco
use, multiplied by the number of
packs per day. For example,1
pack year is 20 cigarettes per day
for 1 year,40 cigarettes per day
equals 2 pack years.
HAEMOCHROMATOSIS
A genetic disorder and the most
common form of iron overload
disease, which is characterized
by iron deposition in the liver
and other tissues as a result of a
small increase in intestinal iron
absorption over many years. It
most often affects white
northern Europeans: 1 in 8–12 is
a carrier of the abnormal gene,
and men are five times more
likely to be diagnosed with
haemochromotosis than
women.
© 2002 Nature Publishing Group
702 | SEPTEMBER 2002 | VOLUME 2 www.nature.com/reviews/cancer
REVIEWS
regulate normal cell function,and how their deficien-
cies can alter normal metabolism.‘Tuning-up’human
metabolism,which varies with genetic constitution and
changes with age, could prove to be a simple and inex-
pensive way to minimize DNA damage,prevent cancer,
improve health and prolong a healthy lifespan
2–4,145
.
1. Doll, R. & Peto, R. The Causes of Cancer (University Press,
Oxford, 1981). (Reference deleted in proof.)
This classical paper discusses the risk of cancer and
the link to lifestyle factors.
2. Hagen, T. M. et al. Feeding acetyl-L-carnitine and lipoic acid
to old rats significantly improves metabolic function while
decreasing oxidative stress. Proc. Natl Acad. Sci. USA 99,
1870–1875 (2002).
3. Liu, J. et al. Memory loss in old rats is associated with
brain mitochondrial decay and RNA/DNA oxidation: partial
reversal by feeding acetyl-L-carnitine and/or R-alpha-lipoic
acid. Proc. Natl Acad. Sci. USA 99, 2356–2361 (2002).
4. Liu, J., Killilea, D. & Ames, B. N. Age-associated
mitochondrial oxidative decay: improvement of carnitine
acetyltransferase substrate-binding affinity and activity in
brain by feeding old rats acetyl-L-carnitine and/or R-alpha-
lipoic acid. Proc. Natl Acad. Sci. USA 99, 1876–1881
(2002).
5. Ames, B. N., Elson-Schwab, I. & Silver, E. A. High-dose
vitamin therapy stimulates variant enzymes with decreased
coenzyme binding affinity (increased Km): relevance to
genetic disease and polymorphisms. Am. J. Clin. Nutr. 75,
616–658 (2002).
6. Ames, B. N. DNA damage from micronutrient deficiencies is
likely to be a major cause of cancer. Mutat. Res. 475, 7–20
(2001).
Discusses how various micronutrient deficiencies can
lead to DNA damage and cancer.
7. Blount, B. C. et al. Folate deficiency causes uracil
misincorporation into human DNA and chromosome
breakage: Implications for cancer and neuronal damage.
Proc. Natl Acad. Sci. USA 94, 3290–3295 (1997).
The results of this study provide evidence that folate
status can be a key determinant of DNA strand
breaks.
8. MacGregor, J. T. Dietary factors affecting spontaneous
chromosomal damage in man. Prog. Clin. Biol. Res. 347,
139–153 (1990).
9. Fraga, C. G. et al. Ascorbic acid protects against
endogenous oxidative damage in human sperm. Proc. Natl
Acad. Sci. USA 88, 11003–11006 (1991).
10. Willett, W. C. Eat, Drink, and Be Healthy (Simon & Schuster
Source, New York, 2001).
11. Willett, W. C. Diet and breast cancer. J. Intern. Med. 249,
395–411 (2001).
12. Willett, W. C. Diet and cancer: one view at the start of the
millennium. Cancer Epidemiol. Biomarkers Prev. 10, 3–8
(2001).
Describes the process by which diet and cancer
relationships have been investigated
epidemiologically in the past few decades, using fat
and breast cancer as an example. Dietary
methodology issues, as well as other epidemiological
issues, are discussed.
13. Meyer, J. P. & Gillatt, D. A. Can diet affect prostate cancer?
Br. J. Urol. 89, 250–254 (2002).
14. Kim, Y. I. in Present Knowledge in Nutrition (eds Bowman,
B. A. & Russell, R. M.) 573–589 (ILSI International Life
Sciences Institute, Washington DC, 2001).
15. Truswell, A. S. Meat consumption and cancer of the large
bowel. Eur. J. Clin. Nutr. 56 (Suppl. 1), S19–S24 (2002).
16. World Cancer Research Fund. American Institute for Cancer
Research, Food Nutrition and the Prevention of Cancer: a
global perspective (World Cancer Research Fund,
Washington DC, 1997).
This is the most comprehensive review so far on the
many aspects of the relationship of diet to cancer.
17. Mehta, R. G. & Mehta, R. R. Vitamin D and cancer. J. Nutr.
Biochem. 13, 252–264 (2002).
18. Plantz, E. A. et al. Plasma 1,25 dihydroxy- and 25-
hydroxyvitamin D and adenomatous polyps of the distal
colorectum. Cancer Epidemiol. Biomarkers Prev. 9,
1059–1065 (2000).
19. Block, G., Patterson, B. & Subar, A. Fruit, vegetables and
cancer prevention: a review of the epidemiologic evidence.
Nutr. Cancer 18, 1–29 (1992).
This review found a statistically significant protective
effect of fruit and vegetable consumption in 128 out of
156 studies in which results were expressed in terms
of relative risk.
20. Steinmetz, K. A. & Potter, J. D. Vegetables, fruit, and cancer.
I. Epidemiology. Cancer Causes Control 2, 325–357 (1991).
This was the first comprehensive scientific review that
concluded that consumption of higher levels of
vegetables and fruit is associated consistently,
although not universally, with a reduced risk of cancer
at most sites.
21. Steinmetz, K. A. & Potter, J. D. Vegetables, fruit, and cancer
prevention: a review. J. Am. Diet Assoc. 96, 1027–1039
(1996).
22. Terry, P. et al. Fruit, vegetables, dietary fiber, and risk of
colorectal cancer. J. Natl Cancer Inst. 93, 525–533
(2001).
23. Smith-Warner, S. et al. Fruits, vegetables and adenomatous
polyps. Am. J. Epidemiol. 155, 1104–1113 (2002).
24. Smith-Warner, S. et al. Intake of fruits and vegetables and
risk of breast cancer: a pooled analysis of cohort studies.
JAMA 285, 769–776 (2001).
25. Voorrips, L. E. et al. Vegetable and fruit consumption and
risks of colon and rectal cancer in prospective cohort sutdy:
The Netherlands Cohort Study on Diet and Cancer. Am. J.
Epidemiol. 152, 1081–1092 (2000).
26. Chatenoud, L. et al. Whole grain food intake and cancer
risk. Int. J. Cancer 77, 24–28 (1998).
27. Jacobs, D. R., Marquart, L., Slavin, J. & Kushi, L. H. Whole-
grain intake and cancer: an expanded review and meta-
analysis. Nutr. Cancer 30, 85–96 (1998).
28. Patterson, R. E., Kristal, A. R. & Neuhouser, M. L. in Primary
and Secondary Preventive Nutrition (eds Bendich, A. &
Deckelbaum, R. J.) 21–43 (Humana Press, Totowa, New
Jersey, 2001).
29. Steinmetz, K. A. & Potter, J. D. Vegetables, fruit, and
cancer. II. Mechanisms. Cancer Causes Control 2,
427–442 (1991).
30. Murillo, G. & Mehta, R. G. Cruciferous vegetables and
cancer prevention. Nutr. Cancer 41, 17–28 (2001).
31. Milner, J. A. Strategies for cancer prevention: the role of diet.
Br. J. Nutr. 87, S265–S272 (2002).
32. Ames, B. N. Cancer prevention and diet: help from single
nucleotide polymorphisms. Proc. Natl Acad. Sci. USA 96,
12216–12218 (1999).
33. Ames, B. N., Shigenaga, M. K. & Gold, L. S. DNA lesions,
inducible DNA repair, and cell division: three key factors in
mutagenesis and carcinogenesis. Environ. Health Perspect.
101 (Suppl. 5), 35–44 (1993).
34. Ames, B. N. Carcinogenesis mechanisms: the debate
continues. Science252, 902 (1991).
35. Cairns, J. Mutation and cancer: the antecedents to our studies
of adaptive mutation. Genetics 148, 1433–1440 (1998).
36. Henderson, B. E. & Feigelson, H. S. Hormonal
carcinogenesis. Carcinogenesis 21, 427–433 (2000).
37. Tomlinson, I. & Bodmer, W. Selection, the mutation rate and
cancer: ensuring that the tail does not wag the dog. Nature
Med. 5, 11–12 (1999).
38. Wild, C. P., Andersson, C., O’Brien, N. M., Wilson, L. &
Woods, J. A. A critical evaluation of the application of
biomarkers in epidemiological studies on diet and health.
Br. J. Nutr. 86 (Suppl 1), S37–S53 (2001).
39. Ames, B. N. & Gold, L. S. Paracelsus to parascience: the
environmental cancer distraction. Mutat. Res. 447, 3–13
(2000).
40. Gold, L. S., Slone, T. H. & Ames, B. N. What do animal
cancer tests tell us about human cancer risk? Overview of
analyses of the carcinogenic potency database. Drug
Metab. Rev. 30, 359–404 (1998).
This paper describes the Carcinogenic Potency
Database (CPDB), which analyses and standardizes
the literature of chronic carcinogenicity tests in
laboratory animals. It discusses the usefulness of
CPDB in addressing questions about the use of
animal bioassays in the evaluation of potential cancer
risks to humans.
41. Giovannucci, E. et al. Alcohol, methyl-deficient diets and risk
of colon cancer in men. J. Natl Cancer Inst. 87, 265–273
(1995).
42. Giovannucci, E. et al. Multivitamin use, folate, and colon
cancer in women in the nurses’ health study. Ann. Intern.
Med. 129, 517–524 (1998).
43. Rohan, T. E., Jain, M., Howe, G. R. & Miller, A. B. Dietary
folate consumption and breast cancer risk. J. Natl Cancer
Inst. 92, 266–269 (2000).
44. Zhang, S. et al. A prospective study of folate intake and the
risk of breast cancer. JAMA 281, 632–637 (1999).
45. Stolzenberg-Solomon, R. et al. Dietary and other methyl-
group availability factors and pancreatic cancer risk in a
cohort of male smokers. Am. J. Epidemiol. 153, 680–687
(2001).
46. Stolzenberg-Solomon, R. Z. et al. Pancreatic cancer risk
and nutrition-related methyl-group availability indicators in
male smokers. J. Natl Cancer Inst. 91, 535–541 (1999).
47. Mayne, S. T. et al. Nutrient intake and risk of subtypes of
esophageal and gastric cancer. Cancer Epidemiol.
Biomarkers Prev. 10, 1055–1062 (2001).
48. Crott, J. W., Mashiyama, S. T., Ames, B. N. & Fenech, M. F.
Methylenetetrahydrofolate reductase C677T polymorphism
does not alter folic acid deficiency-induced uracil
incorporation into primary human lymphocyte DNA in vitro.
Carcinogenesis 22, 1019–1025 (2001).
49. Fenech, M. F., Aitken, C. & Rinaldi, J. R. Folate, vitamin B12,
homocysteine status and DNA damage in young Australian
adults. Carcinogenesis19, 1163–1171 (1998).
50. Fenech, M. Micronutrients and genomic stability: a new
paradigm for recommended dietary allowances (RDAs).
Food Chem. Toxicol. 40, 1113–1117 (2002).
51. Hagmar, L. et al. Chromosomal aberrations in lymphocytes
predict human cancer: a report from the European Study
Group on Cytogenetic Biomarkers and Health (ESCH).
Cancer Res. 58, 4117–4121 (1998).
52. Chen, J. et al. A methylenetetrahydrofolate reductase
polymorphism and the risk of colorectal cancer. Cancer Res.
56, 4862–4864 (1996).
53. Le Marchand, L. et al. B-vitamin intake, metabolic genes,
and colorectal cancer risk (United States). Cancer Causes
Control 13, 239–248 (2002).
54. Ma, J. et al. Methylenetetrahydrofolate reductase
polymorphism, dietary interactions, and risk of colorectal
cancer. Cancer Res. 57, 1098–1102 (1997).
55. Slattery, M. L., Potter, J. D., Samowitz, W., Schaffer, D. &
Leppert, M. Methylenetetrahydrofolate reductase, diet, and
risk of colon cancer. Cancer Epidemiol. Biomarkers Prev. 8,
513–518 (1999).
56. Levine, A. J. et al. The methylenetetrahydrofolate
reductase 677C→T polymorphism and distal colorectal
adenoma risk. Cancer Epidemiol. Biomarkers Prev. 9,
657–663 (2000).
57. Ulrich, C. M. et al. Colorectal adenomas and the C677T
MTHFR polymorphism: evidence for gene-environment
interaction? Cancer Epidemiol. Biomarkers Prev. 8, 659–668
(1999).
58. Skibola,C. F. et al. Polymorphisms in the
methylenetetrahydrofolate reductase gene are associated
with susceptibility to acute leukemia in adults. Proc. Natl
Acad. Sci. USA 96, 12810–12815 (1999).
Shows the power of single-nucleotide polymorphisms
in epidemiological studies to clarify mechanisms and
risk factors in the area of diet and disease prevention.
59. Wiemels, J. L. et al. Methylenetetrahydrofolate reductase
(MTHFR) polymorphisms and risk of molecularly defined
subtypes of childhood acute leukemia. Proc. Natl Acad. Sci.
USA 98, 4004–4009 (2001).
60. Goodman, M. T. et al. Association of
methylenetetrahydrofolate reductase polymorphism C677T
and dietary folate with the risk of cervical dysplasia. Cancer
Epidemiol. Biomarkers Prev. 12, 1275–1280 (2001).
61. Wallock, L. M. et al. Low seminal plasma folate
concentrations are associated with low sperm density and
count in male smokers and nonsmokers. Fertil. Steril. 75,
252–259 (2001).
fruits and vegetables,such as zinc,iron, and the vita-
mins E,niacin,B6 and B12,also seem to contribute to
DNA damage.We anticipate that other micronutri-
ents will be added to this list in the coming years.With
more research,we will develop a better understanding
of the mechanisms by which specific micronutrients
© 2002 Nature Publishing Group
NATURE REVIEWS | C ANCER VOLUME 2 | SEPTEMBER 2002 | 703
REVIEWS
Folate deficiency in men causes uracil incorporation in
sperm DNA and decreased sperm count and quality.
62. Mayr, C. A., Woodall, A. A. & Ames, B. N. in Preventive
Nutrition: The Comprehensive Guide for Health
Professionals (eds Bendich, A. & Deckelbaum, R. J.)
373–386 (Humana Press, Inc., Totowa, New Jersey, 2001).
63. Fenech, M. F., Dreosti, I. E. & Rinaldi, J. R. Folate, vitamin
B12, homocysteine status and chromosome damage rate in
lymphocytes of older men. Carcinogenesis 18, 1329–1336
(1997).
64. Fenech, M. Micronucleus frequency in human lymphocytes
is related to plasma vitamin B12 and homocysteine. Mutat.
Res. 428, 299–304 (1999).
65. Stabler, S. P., Sampson, D. A., Wang, L. P. & Allen, R. H.
Elevations of serum cystathionine and total homocysteine in
pyridoxine-, folate-, and cobalamin-deficient rats. J. Nutr.
Biochem. 8, 279–289 (1997).
66. Alberg, A. J. et al. The risk of cervical cancer in relation to
serum concentrations of folate, vitamin B12, and
homocysteine. Cancer Epidemiol. Biomarkers Prev. 9,
761–764 (2000).
67. Wu, K. et al. A prospective study on folate, B12, and
pyridoxal 5′-phosphate (B6) and breast cancer. Cancer
Epidemiol. Biomarkers Prev. 8, 209–217 (1999).
68. Key, T. J., Silcocks, P. B., Davey, G. K., Appleby, P. N. &
Bishop, D. T. A case–control study of diet and prostate
cancer. Br. J. Cancer 76, 678–687 (1997).
69. Hartman, J. T. et al. Association of B-vitamins pyrodoxal
5′-phosphate (B(6)), B(12) and folate with lung cancer risk in
older men. Am. J. Epidemiol. 153, 688–694 (2001).
70. Weinstein, S. J. et al. Elevated serum homocysteine levels
and increased risk of invasive cervical cancer in US women.
Cancer Causes Control 12, 317–324 (2001).
71. Jacob, R. A. in Nutrition in Health and Disease (eds Shils, M.,
Olson, J. A. & Shike, M.) 467–484 (Williams & Wilkins,
Baltimore, 1999).
72. Ames, B. N., Shigenaga, M. K. & Hagen, T. M. Oxidants,
antioxidants, and the degenerative diseases of aging. Proc.
Natl Acad. Sci. USA 90, 7915–7922 (1993).
73. Kono, S. & Hirohata, T. Nutrition and stomach cancer.
Cancer Causes Control 7, 41–55 (1996).
74. Zhang, Z. W., Abdullahi, M. & Farthing, M. J. Effect of
physiological concentrations of vitamin C on gastric cancer
cell and Helicobacter pylori. Gut50, 165–169 (2002).
75. Moller, P. & Loft, S. Oxidative DNA damage in human white
blood cells in dietary antioxidant intervention studies. Am. J.
Clin. Nutr. 76, 303–310 (2002).
76. Levine, M. et al. Vitamin C pharmacokinetics in healthy
volunteers: evidence for a recommended dietary allowance.
Proc. Natl Acad. Sci. USA 93, 3704–3709 (1996).
77. Duthie, S. J., Ma, A., Ross, M. A. & Collins, A. R. Antioxidant
supplementation decreases oxidative DNA damage in
human lymphocytes. Cancer Res. 56, 1291–1295 (1996).
78. Harats, D. et al. Citrus fruit supplementation reduces
lipoprotein oxidation in young men ingesting a diet high in
saturated fat: presumptive evidence for an interaction
between vitamins C and E in vivo. Am. J. Clin. Nutr. 67,
240–245 (1998).
79. McCall, M. R. & Frei, B. Can antioxidant vitamins materially
reduce oxidative damage in humans? Free Radic. Biol. Med.
26, 1034–1053 (1999).
This review examines the scientific evidence that
supplementation of humans with vitamin C, vitamin E
or β-carotene lowers in vivo oxidative damage to
lipids, proteins or DNA based on the measurement of
oxidative biomarkers.
80. Block, G. in Vitamin C: The State of the Art in Disease
Prevention Sixty Years after the Nobel Prize (eds Paoletti, R.,
Sies, H., Bug, J., Grossi, E. & Poli, A.) 51–58
(Springer–Verlag, Milano, 1998).
81. Byers, T. & Guerrero, N. Epidemiologic evidence for vitamin
C and vitamin E in cancer prevention. Am. J. Clin. Nutr. 62,
13855–13925 (1995).
82. Chan, S. W. Y. & Reade, P. C. The role of ascorbic acid in
oral cancer and carcinogenesis. Oral Dis. 4, 120–129
(1998).
83. Jacobs, E. J. et al. Vitamin C, vitamin E, and multivitamin
supplement use and stomach cancer mortality in the
Cancer Prevention Study II cohort. Cancer Epidemiol.
Biomarkers Prev. 11, 35–41 (2002).
84. Jacobs, E. J. et al. Vitamin C and Vitamin E supplement use
and colorectal cancer mortality in a large American Cancer
Society cohort. Cancer Epidemiol. Biomarkers Prev. 10,
17–23 (2001).
85. Dietrich, M. et al. Antioxidant supplementation decreases
lipid peroxidation biomarker F(2)-isoprostanes in plasma of
smokers. Cancer Epidemiol. Biomarkers Prev. 11, 7–13
(2002).
86. Collins, A. R., Olmedilla, B., Southon, S., Granado, F. &
Duthie, S. J. Serum carotenoids and oxidative DNA damage
in human lymphocytes. Carcinogenesis 19, 2159–2162
(1998).
87. Panayiotidis, M. & Collins, A. R. Ex vivo assessment of
lymphocyte antioxidant status using the Comet Assay.
Free Rad. Res. 27, 533–537 (1997).
This study reports results that the three main
antioxidants have a demonstrable, short-term
protective effect against oxidative DNA damage,
which is widely regarded as an important element in
the aetiology of cancer.
88. Beckman, K. B., Saljoughi, S., Mashiyama, S. & Ames, B. N.
A simpler, more robust method for the analysis of
8-oxoguanine in DNA. Free Radic. Biol. Med. 29, 357–367
(2000).
89. Helbock, H. J. et al. DNA oxidation matters: the HPLC-EC
assay of 8-oxo-deoxyguanosine and 8-oxo-guanine. Proc.
Natl Acad. Sci. USA 95, 288–293 (1998).
90. Branda, R. F. & Blickensderfer, D. B. Folate deficiency
increases genetic damage caused by alkylating agents and
gamma-irradiation in Chinese hamster ovary cells. Cancer
Res. 53, 5401–5408 (1993).
91. Institute of Medicine, Food and Nutrition Board. Dietary
Reference Intakes for Vitamin C, Vitamin E, Selenium and
Carotenoids (National Academy Press, Washington DC,
2001).
92. Fraga, C. G., Motchnik, P. A., Wyrobek, A. J., Rempel, D. M.
& Ames, B. N. Smoking and low antioxidant levels increase
oxidative damage to sperm DNA. Mutat. Res. 351, 199–203
(1996).
93. Lykkesfeldt, J. et al. Ascorbate is depleted by smoking and
repleted by moderate supplementation: a study in male
smokers and nonsmokers with matched dietary antioxidant
intakes. Am. J. Clin. Nutr. 71, 530–536 (2000).
94. Sorahan, T. et al. Childhood cancer and parental use of
tobacco: deaths from 1971 to 1976. Br. J. Cancer 76,
1525–1531 (1997).
95. Sorahan, T., Lancashire, R. J., Prior, P., Peck, I. & Stewart, A. M.
Childhood cancer and parental use of alcohol and tobacco.
Ann. Epidemiol. 5, 354–359 (1995).
96. Sorahan, T., Lancashire, R. J., Hulten, M. A., Peck, I. &
Stewart, A. M. Childhood cancer and parental use of
tobacco deaths from 1953 to 1955. Br. J. Cancer 75,
134–138 (1997).
97. Ji, B.-T. et al. Paternal cigarette smoking and the risk of
childhood cancer among offspring of nonsmoking mothers.
J. Natl Cancer Inst. 89, 238–244 (1997).
Lymphoma, acute lymphocytic leukaemia and brain
cancer are each increased three- to fourfold in the
offspring of men who smoke.
98. Toyokuni, S. Iron-induced carcinogenesis: the role of redox
regulation. Free Radic. Biol. Med. 20, 553–566 (1996).
99. Yip, R. in Present Knowledge in Nutrition (eds Bowman, B. A.
& Russell, R. M.) 311–328 (ILSI Press, Washington DC,
2001).
100. Nelson, R. L. Iron and colorectal cancer risk: human studies.
Nutr. Rev. 59, 140–148 (2001).
101. Zhang, D. et al. Vitamin E inhibits apoptosis, DNA
modification, and cancer incidence induced by iron-
mediated peroxidation in Wistar rat kidney. Cancer Res. 57,
2410–2414 (1997).
102. Walter, P. W. et al. Iron deficiency and iron excess damage
mitochondria and mitochondrial DNA in rats. Proc. Natl
Acad. Sci. USA 99, 2264–2269 (2002).
103. Atamna, H., Walter, P. W. & Ames, B. N. The role of heme
and iron-sulfur clusters in mitochondrial biogenesis,
maintenance, and decay with age. Arch. Biochem. Biophys.
397, 345–353 (2002).
104. King, J. C., Shames, D. M. & Woodhouse, L. R. Zinc
homeostasis in humans. J. Nutr. 130, 1360S–1366S
(2000).
Provides background information on zinc metabolism
and describes how adjustments in zinc homeostasis
occur in animals and humans.
105. Wood, R. J. Assessment of marginal zinc status in humans.
J. Nutr. 130, 1350S–1354S (2000).
106. Walsh, C. T., Sandstead, H. H., Prasad, A. S., Newberne, P. M.
& Fraker, P. J. Zinc: health effects and research priorities for
the 1990s. Environ. Health Perspect. 102 (Suppl 2), 5–46
(1994).
107. King, J. & Keen, C. L. in Nutrition in Health and Disease (eds
Shils, M., Olson, J. A. & Shike, M.) 223–239 (Williams &
Wilkins, Baltimore, 1999).
108. Sarkar, B. Metal replacement in DNA-binding zinc finger
proteins and its relevance to mutagenicity and
carcinogenicity through free radical generation. Nutrition 11,
646–649 (1995).
109. Pavletich, N. P., Chambers, K. A. & Pabo, C. O. The DNA-
binding domain of p53 contains the four conserved regions
and the major mutation hot spots. Genes Dev. 7,
2556–2564 (1993).
110. Ho, E. & Ames, B. N. Zinc deficiency induces APE/Ref–1
expression and induces DNA damage in C6 glioma cells.
Free Rad. Biol. Med. 31, 518 (2001).
111. Castro, C. E., Kaspin, L. C., Chen, S.-S. & Nolker, S. G. Zinc
deficiency increases the frequency of single-strand DNA
breaks in rat liver. Nutr. Res. 12, 721–736 (1992).
112. Olin, K. L. et al. Maternal dietary zinc influences DNA strand
break and 8-hydroxy-2′-deoxyguanosine levels in infant
rhesus monkey liver. Proc. Soc. Exp. Biol. Med. 203,
461–466 (1993).
113. Oteiza, P. L., Olin, K. L., Fraga, C. E. & Keen, C. L. Oxidant
defense systems in testes from zinc-deficient rats (44040).
Proc. Soc. Exp. Biol. Med. 213, 85–91 (1996).
114. O’Connor, T. R., Graves, R. J., de Murcia, G., Castaing,
B. & Laval, J. Fpg protein of Escherichia coli is a zinc
finger protein whose cysteine residues have a structural
and/or functional role. J. Biol. Chem. 268, 9063–9070
(1993).
115. Newberne, P. M., Broitman, S. & Schrager, T. F. Esophageal
carcinogenesis in the rat: zinc deficiency, DNA methylation
and alkyltransferase activity. Pathobiology 65, 253–263
(1997).
116. Fong, L. Y. Y., Lau, K.-M., Huebner, K. & Magee, P. N.
Induction of esophageal tumors in zinc-deficient rats by
single low doses of N-nitrosomethylbenzylamine (NMBA):
analysis of cell proliferation, and mutations in H-ras and p53
genes. Carcinogenesis18, 1477–1484 (1997).
117. Fong, L. Y. Y., Li, J., Farber, J. L. & Magee, P. N. Cell
proliferation and esophageal carcinogenesis in the zinc-
deficient rat. Carcinogenesis 17, 1841–1848 (1996).
118. Fong, L. Y., Nguyen, V. T. & Farber, J. L. Esophageal cancer
prevention in zinc-deficient rats: rapid induction of apoptosis
by replenishing zinc. J. Natl Cancer Inst. 93, 1525–1533
(2001).
119. Anderson, M. B., Lepak, K., Farinas, V. & Geroge, W. J.
Protective action of zinc against cobalt-induced testicular
damage in the mouse. Reprod. Toxicol. 7, 49–54 (1993).
120. Xu, B., Chia, S. E. & Ong, C. H. Concentrations of cadmium,
lead, selenium and zinc in human blood and seminal
plasma. Biol. Trace Element Res. 40, 49–57 (1994).
121. Wong, W. Y. et al. Effects of folic acid and zinc sulfate on
male factor subfertility: a double-blind, randomized,
placebo-controlled study. Fertil. Steril. 77, 491–498
(2002).
122. Block, G. in Antioxidant Food Supplements in Human
Health (eds Packer, L., Hiramatsu, M. & Yoshikawa, T.)
45–54 (Academic Press, San Diego, 1999).
123. Scholl, T. O. et al. Use of multivitamin/mineral prenatal
supplements: influence on the outcome of pregnancy.
Am. J. Epidemiol. 146, 134–141 (1997).
Examines the association of prenatal multivitamin and
mineral supplements during the first two trimesters of
pregnancy and birth outcomes. Risk for preterm
delivery and low birth weight was reduced with
supplement use.
124. Willett, W. C. & Stampfer, M. J. What vitamins should I be
taking, Doctor? N. Engl. J. Med. 345, 1819–1824
(2001).
125. Albanes, D. et al. Alpha-tocopherol and beta-carotene
supplements and lung cancer incidence in the alpha-
tocopherol, beta-carotene cancer prevention study: effects
of base-line characteristics and study compliance. J. Natl
Cancer Inst. 88, 1560–1570 (1996).
126. Omenn, G. S. et al. Risk factors for lung cancer and for
intervention effects in CARET, the Beta-Carotene and
Retinol Efficacy Trial. J. Natl Cancer Inst. 88, 1550–1559
(1996).
127. Levine, A. J. et al. Serum 25-hydroxyvitamin D, dietary
calcium intake, and distal colorectal adenoma risk. Nutr.
Cancer 39, 35–41 (2001).
128. Kallay, E. et al. Vitamin D receptor activity and prevention of
colonic hyperproliferation and oxidative stress. Food Chem.
Toxicol. 40, 1191–1196 (2002).
129. Chan, J. M., Stampfer, M. J., Gann, P. H., Gaziano, J. M. &
Giovannucci, E. L. Dairy products, calcium and prostate
cancer risk in Physicians’ Health Study. Am. J. Clin. Nutr. 74,
549–554 (2001).
130. Giovannucci, E. L. et al. Calcium and fructose intake in
relation to risk of prostate cancer. Cancer Res. 58, 442–447
(1998).
131. Clark, L. C. et al. Decreased incidence of prostate cancer
with selenium supplementation: results of a double-blind
cancer prevention trial. Br. J. Urol. 81, 730–734 (1998).
132. Nomura, A. M., Lee, J., Stemmermann, G. N. & Combs, G.
F. J. Serum selenium and subsequent risk of prostate
cancer. Cancer Epidemiol. Biomarkers Prev. 9, 883–887
(2000).
133. Yoshizawa, K. et al. Study of prediagnostic selenium level in
toenails and the risk of advanced prostate cancer. J. Natl
Cancer Inst. 90, 1184–1185 (1998).
© 2002 Nature Publishing Group
704 | SEPTEMBER 2002 | VOLUME 2 www.nature.com/reviews/cancer
REVIEWS
134. Klein, E. A. et al. SELECT: the next prostate cancer
prevention trial. Selenum and Vitamin E Cancer Prevention
Trial. J. Urol. 166, 1311–1315 (2001).
135. Negri, E. et al. Selected micronutrients and oral and
pharyngeal cancer. Int. J. Cancer 86, 122–127
(2000).
136. Franceschi, S. et al. Role of macronutrients, vitamins and
minerals in the aetiology of squamous-cell carcinoma of the
oesophagus. Int. J. Cancer 86, 626–631 (2000).
137. Greenwald, P., Clifford, C. K. & Milner, J. A. Diet and cancer
prevention. Europ. J. Cancer 37, 948–965 (2001).
This review covers the evidence for a diet and
cancer relationship, including the topics of
vegetables and fruits, dietary fibre, micronutrients,
phytochemicals, dietary fat, anthropometry/physical
activity and alcohol. It also discusses the emerging
evidence of gene–nutrient interactions and future
directions.
138. Gupta, K., Krishnaswamy, G., Karnad, A. & Peiris, A. N.
Insulin: a novel factor in carcinogenesis. Am. J. Med. Sci.
323, 140–145 (2002).
139. IARC International Agency for Research on Cancer, WHO.
Weight Control and Physical Activity (IARC Press, Lyon,
2002).
140. Key, T. J. Epidemiology of breast cancer. Lancet Oncol. 2,
133–140 (2001).
141. McCann, J. Obesity, cancer links prompt new
recommendations. J. Natl Cancer Inst. 93, 901–902
(2001).
142. Barnes, S. Role of phytochemicals in prevention and
treatment of prostate cancer. Epidemiol. Rev. 23, 102–105
(2001).
143. Wakimoto, P. & Block, G. Dietary intake, dietary patterns,
and changes with age: an epidemiolgical perspective.
J. Gerontol. 56, 65–80 (2001).
144. Block, G. Analysis of: National Health and Nutrition
Examination Survey III 1988–1994 Public-Use Data Files
(National Center for Health Statistics, Hyattsville, Maryland).
145. Ames, B. N. The metabolic tune up: metabolic harmony and
disease prevention. J. Nutr. (in the press).
146. Ulrich, C. M. et al. Lack of association between the
C677T MTHFR polymorphism and colorectal hyperplasic
polyps. Cancer Epidemiol. Biomarkers Prev. 9, 427–433
(2000).
Acknowledgements
This work was supported by grants from the National Foundation
for Cancer Research, the US Department of Energy, the Wheeler
Fund for the Biological Sciences at the University of California
Berkeley, the Ellison Medical Foundation and the National Institute
of Environmental Health Sciences Center. We thank L. Gold and
J. Nides for their many useful comments.
Online links
DATABASES
The following terms in this article are linked online to:
Cancer.gov: http://www.cancer.gov/cancer_information/
adult acute lymphoblastic leukaemia | brain cancer | breast cancer |
cervical cancer | childhood acute lymphoblastic leukaemia |
colorectal cancer | hepatic carcinoma | lung cancer | lymphoma |
oesophageal cancer | oral cancer | pancreatic cancer | prostate
cancer | stomach cancer
GenBank: http://www.ncbi.nih.gov/Genbank/
HPV
LocusLink: http://www.ncbi.nlm.nih.gov/LocusLink/
catalase | haem oxygenase II | MTHFR | TP53
FURTHER INFORMATION
The International Bibliographic Information on Dietary
Supplements (IBIDS): http://dietary-
supplements.info.nih.gov/databases/ibids.html
Linus Pauling Institute Micronutrient Information Center:
http://osu.orst.edu/dept/lpi/infocenter/index.html
National Institute of Health Office of Dietary Supplements:
http://www.cc.nih.gov/ccc/supplements/
Scientific evaluation of US dietary reference intakes:
http://www.iom.edu/IOM/IOMHome.nsf/Pages/FNB+DRI
Access to this interactive links box is free online.
