Nutrient-Gene Interactions

Methylenetetrahydrofolate Reductase C677T Polymorphism, Folic Acid

and Riboflavin Are Important Determinants of Genome Stability
in Cultured Human Lymphocytes
Michiyo Kimura,* Keizo Umegaki,* Mitsuru Higuchi,* Philip Thomas**
and Michael Fenech**1
*The National Institute of Health and Nutrition, 123-1 Toyama, Shinjuku-ku, Tokyo 162-8636, Japan;

Takasaki University of Health and Welfare, 371 Nakaorui, Takasaki, Gunma 370-0033, Japan
and **CSIRO Health Sciences and Nutrition, Adelaide BC, Adelaide, SA, Australia 5000

Folate plays an important role in the maintenance of

genomic stability, mainly by providing methyl groups for the
synthesis of deoxythymidine triphosphate (dTTP)2 from deoxyuridine monophosphate (dUMP), and methionine from
homocysteine (Hcy) (1 4). Methionine is subsequently converted to S-adenosyl methionine (SAM), which provides
methyl groups for the maintenance methylation of cytosinephosphate-guanosine dinucleotide (CpG) islands and other
intervening sequences containing CpG. Figure 1 presents a
simplified diagram of the relevant metabolic pathways and the
key enzymes that control these pathways. Methylenetetrahydrofolate reductase (MTHFR) is a pivotal enzyme that con-

trols the bioavailability of folate for dTTP synthesis and maintenance methylation of CpG. The activity of MTHFR can be
reduced in three ways: 1) by polymorphisms in the gene
sequence that alter its affinity for a substrate or cofactor; 2) by
a high concentration of methionine or SAM, which inhibit
MTHFR activity; and 3) by a low concentration of its cofactor
flavin adenine dinucleotide (FAD) or of riboflavin, the precursor of FAD (5,6). Reducing MTHFR activity increases the
5,10-methylenetetrahydrofolate concentration, whereas it decreases the 5-methyltetrahydrofolate concentration. Such a
situation is expected to 1) favor synthesis of dTTP over methylation of CpG, 2) minimize uracil incorporation into DNA
and the chromosome breaks caused by uracil (7,8); and 3)
increase the Hcy concentration.
The C677T polymorphism reduces MTHFR activity by
50% and is associated with reduced risk for a variety of
cancers, such as leukemia (9), lymphoma (10) and colorectal
cancer (11), but also with increased risk for Down syndrome
(12), neural tube defects (13) and cervical cancer (14). The
polymorphism is also linked to reduced methylation of CpG
DNA in lymphocytes (15). The contrasting effects of the
C677T polymorphism may reflect the relative importance of
CpG methylation and dUMP methylation in the etiology of
the diseases listed above. Therefore, it is important to thor-

1
To whom correspondence should be addressed.
E-mail: michael.fenech@csiro.au.
2
Abbreviations used: APOP, apoptotic cell; BNC, binucleated cell; CBMN,
cytokinesis-block micronucleus assay; CC, wild-type methylenetetrahydrofolate
reductase C677T homozygote; CpG, cytosine-phosphate-guanosine dinucleotide; dTTP, deoxythymidine triphosphate; dUMP, deoxyuridine monophosphate;
FAD, flavin adenine dinucleotide; Hcy, homocysteine; HF, high folic acid (100
nmol/L); HR, high riboflavin (500 nmol/L); LF, low folic acid (20 nmol/L); LR, low
riboflavin (0 nmol/L); MNi, micronuclei; MTHFR, methylenetetrahydrofolate reductase; NEC, necrotic cell; NPB, nucleoplasmic bridge; NBUD, nuclear bud; SAM,
S-adenosyl methionine; TT, mutant methylenetetrahydrofolate reductase C677T
homozygote.

0022-3166/04 $8.00 2004 American Society for Nutritional Sciences.

Manuscript received 24 June 2003. Initial review completed 31 July 2003. Revision accepted 6 October 2003.
48

Downloaded from jn.nutrition.org at UMD of New Jersey G F Smith Library on April 16, 2009

ABSTRACT We tested the hypothesis that methylenetetrahydrofolate reductase (MTHFR) C677T polymorphism,
folic acid deficiency and riboflavin deficiency, independently or interactively, are important determinants of
genomic stability, cell death, cell proliferation and homocysteine (Hcy) concentration in 9-d human lymphocyte
cultures. Lymphocytes of seven wild-type (CC) and seven mutant (TT) homozygotes were cultured under the four
possible combinations of deficiency and sufficiency of riboflavin (0 and 500 nmol/L) and folic acid (20 and 100
nmol/L) at a constant L-methionine concentration of 50 mol/L. Viable cell growth was 25% greater in TT than in
CC cells (P 0.05) and 32% greater at 100 nmol/L folic acid than at 20 nmol/L folic acid (P 0.002). The
comprehensive cytokinesis-block micronucleus assay was used to measure micronuclei (MNi; a marker for
chromosome breakage and loss), nucleoplasmic bridges (NPB; a marker of chromosome rearrangement) and
nuclear buds (NBUD, a marker of gene amplification). The MNi levels were 21% higher in TT cells than in CC cells
(P 0.05) and 42% lower in the high folic acid medium than in the low folic acid medium (P 0.0001). The NBUD
levels were 27% lower in TT cells than in CC cells (P 0.05) and 45% lower in the high folic acid medium than in
the low folic acid medium (P 0.0001). High riboflavin concentration (500 nmol/L) increased NBUD levels by 25%
(compared with 0 nmol/L riboflavin) in folate-deficient conditions (20 nmol/L folic acid medium; P 0.05), and there
was an interaction between folic acid and riboflavin that affected NBUD levels (P 0.042). This preliminary
investigation suggests that MTHFR C677T polymorphism and riboflavin affect genome instability; however, the
effect is relatively small compared with that of folic acid. J. Nutr. 134: 48 56, 2004.

MTHFR, FOLIC ACID, RIBOFLAVIN AND GENOME STABILITY

49

oughly understand the effect of the MTHFR C677T polymorphism and other metabolic factors that affect the activity of
MTHFR on chromosomal instability, an important risk factor
in cancer.
To test these various factors we developed an in vitro
system of culturing lymphocytes for 9 d with concentrations of
micronutrients that are within or close to the physiological
range (16 19). We used this system in combination with the
cytokinesis-block micronucleus (CBMN) assay in its comprehensive mode (18,19) to measure various markers of genotoxicity and cytotoxicity that are important in assessing the effect
of micronutrients on genomic stability and cell death. These
markers include 1) micronuclei (MNi), a marker of chromosome breakage and/or loss; 2) nucleoplasmic bridges (NPB), a
marker of chromosome rearrangement; 3) nuclear buds
(NBUD), a marker of gene amplification; 4) necrosis (NEC);
and 5) apoptosis (APOP) (Fig. 2).
We previously used this system to show that MNi, NPB and
NBUD levels increase markedly with a decline in folic acid
concentration from 120 to 12 nmol/L, which coincides with
the physiological range in the serum of individuals consuming
an unsupplemented diet (8 to 35 nmol/L) (6,20,21). It is
important to note that these markers of genome instability all
positively correlate with each other, suggesting that folic acid
deficiency affects the generation of breakage-fusion-bridge cycles, which is a hallmark of genomic instability in several types
of cancer cells (22). However, we were unable to show that the
MTHFR C677T polymorphism affected genomic stability under the conditions of in vitro culture used in that experiment.
We reasoned that the supraphysiological concentrations of
methionine (100 mol/L) and riboflavin (530 nmol/L) in the
RPMI 1640 medium might have altered the activity of
MTHFR T677T and MTHFR C677C in such a way that
differing effects on genome instability between genotypes became indiscernible (17). The range of serum concentration of
methionine in healthy subjects is 20 to 30 mol/L (23) and
that of riboflavin varies from 5 to 50 nmol/L (6). An excess of
methionine may increase SAM concentration, which inhibits
MTHFR, whereas an excess of riboflavin may increase FAD
concentration, which may be expected to promote MTHFR

FIGURE 2 Genome damage and cell death biomarkers scored in

the comprehensive cytokinesis-block micronucleus (CBMN) assay. The
CBMN assay allows the measurement of all possible outcomes following a genome damage event. In this assay a cell with genome damage
either may undergo cell death via apoptosis or necrosis or may survive
and undergo further nuclear division. In the latter case, dividing cells are
recognized as binucleated cells (BNC) by blocking cytokinesis with
cytochalasin-B. The BNC are then scored for the following genome
damage events: 1) micronuclei (MNi), which originate from lagging
whole chromosomes or broken chromosome fragments and are therefore a marker of chromosome breakage and chromosome loss events,
the latter being due to defects in centromere or spindle structure; 2)
nucleoplasmic bridges (NPB), which originate from dicentric chromosomes caused by misrepair of chromosome breaks and are therefore a
marker of chromosome rearrangement; and 3) nuclear buds (NBUD),
which are the mechanism by which the nucleus eliminates amplified
DNA resulting from breakage-fusion-bridge cycles generated by NPB.
The NBUD level therefore provides a measure of gene amplification. An
increase in MNi, NPB and NBUD levels is indicative of an increase in
genome instability commonly seen in cancer. For futher details of the
CBMN assay, refer to Fenech (18,19).

Downloaded from jn.nutrition.org at UMD of New Jersey G F Smith Library on April 16, 2009

FIGURE 1
The main metabolic
pathways by which folate, cobalamin,
choline, methionine, pyridoxine and riboflavin affect DNA methylation, synthesis
and repair. Abbreviations: B6, pyridoxine; B12, cobalamin; BHMT, betaine:
homocysteine methyltransferase; DHF,
dihydrofolate; DMG, dimethylglycine; FAD,
flavin adenine dinucleotide; 5-MeTHF,
5-methyltetrahydrofolate; 5,10-MeTHF,
5,10-methylenetetrahydrofolate; MS,
methionine synthase; MTHFR, methylenetetrahydrofolate reductase; SAM,
S-adenosyl methionine; SHM, serine
hydroxymethyltransferase; THF, tetrahydrofolate; TS, thymidylate synthase.

KIMURA ET AL.

50

TABLE 1
Gender, age and genotype characteristics
of subjects in this study1
Variable
Gender
Male
Female
Age2 (range), y

CC, n 7

3
4
56.6 8.04
(4467)

3
4
55.14 8.34
(4266)

5
2
0

4
3
0

6
1
0

6
1
0

1 Abbreviations: CC, wild-type methylenetetrahydrofolate reductase C677T homozygote; MTHFR, methylenetetrahydrofolate reductase; TT, mutant methylenetetrahydrofolate reductase C677T homozygote. Genotypes: AA, wild-type MTHFR A1298C homozygote; AC,
MTHFR A1298C heterozygote; CC, mutant MTHFR A1298C homozygote; AA*, wild-type MS A2756G homozygote; AG, MS A2756G heterozygote; GG, mutant MS A2756G homozygote; MS, methionine synthase.
2 Values are means SD.

activity, particularly in the variant enzyme with decreased

coenzyme-binding activity (24).
Therefore we repeated these experiments using a physiological concentration of methionine in combination with deficient or adequate concentrations of riboflavin and folic acid.
In addition, we measured the homocysteine concentration of
the medium to validate the model. The results suggest that 1)
the hypothesis that folic acid, riboflavin and MTHFR C677T
polymorphism interact to affect chromosomal instability is, to
a certain extent, correct, and 2) the in vitro model can be used
to study the effect of gene-nutrient interaction on genome
stability.
MATERIALS AND METHODS
Recruitment of subjects and genotyping. The study was approved by the Human Experimentation Ethics Committee of CSIRO
Health Sciences and Nutrition, Adelaide, Australia. Subjects were
recruited from a cohort of volunteers whose genotype for MTHFR
C677T, MTHFR A1298C and methionine synthase A2756G was
previously determined. The volunteers selected for the study included
seven MTHFR 677 TT (mutant type)homozygotes and MTHFR 677
CC (wild type) homozygotes matched for age, gender and the
MTHFR A1298C and MS A2756G polymorphisms (Table 1). Genotyping was performed in duplicate using published methods (13,25).
Lymphocyte culture, viability and cell growth assays. Blood (90
mL) was collected from each subject after an overnight fast and
before breakfast. The RPMI 1640 culture medium was custom made
in the laboratory to achieve the required concentrations of folic acid,
riboflavin and L-methionine. All other constituents of the medium
were standard for RPMI 1640 as previously described (26), and were
all purchased from Sigma (St. Louis, MO). Dialyzed fetal calf serum
(FCS; 5%; Trace Biosciences, Victoria, Australia) and 10 kU/L
interleukin 2 (Roche Diagnostics, Basel, Switzerland) were added to
the medium. The dialyzed FCS contained 356 pmol/L cobalamin and
9 nmol/L folic acid, which equated to 17.8 pmol/L cobalamin and
0.45 pmol/L folic acid in the complete medium. The folic acid
concentration of the medium did not change after 3 d of culture.
Lymphocytes were cultured at a concentration of 0.5 109 cells/L in
10-mL volumes in eight 25-mL culture flasks (Sarstedt, Adelaide,

FIGURE 3 Experimental design. Abbreviations: CBMN, cytokinesis-block micronucleus assay, performed by adding cytochalasin-B on
d 8 and harvesting cytokinesis-blocked cells on d 9; Hcy, homocysteine
assay; PHA, phytohaemagglutinin (mitogen).

Downloaded from jn.nutrition.org at UMD of New Jersey G F Smith Library on April 16, 2009

MTHFR 1298 genotype

AA
AC
CC
MS 2756 genotype
AA*
AG
GG

TT, n 7

Australia) after stimulation with phytohaemagglutinin according to

an established protocol (16,17). Briefly (Fig. 3), on d 3 and 6 the
number of viable cells was determined using trypan blue exclusion
and electronic cell counting, then the lymphocytes were subcultured
in fresh medium at a concentration of 0.5 109 viable cells/L. The
viable cell counts on d 3, 6 and 9 were used to estimate the cell
proliferation rate. An increase in the viable cell count indicates that
the proportion of cells completing replicative DNA synthesis and cell
division is greater than the proportion of nondividing and/or dying
cells. Preliminary dose-response experiments were conducted to determine the range of concentrations of methionine, riboflavin and
folic acid that would sustain cell growth over a 9-d period. Each
micronutrient was studied individually while maintaining the other
micronutrients at their typical concentrations in RPMI 1640 medium.
Cytokinesis-block micronucleus (CBMN) assay. On d 8 a
0.75-mL aliquot of the cell culture was placed in a culture tube and
incubated for a further 24 h in the presence of cytochalasin-B (4.5mg/
L). At the end of this period cells were harvested by cytocentrifugation (Shandon; Southern Products, Cheshire, U.K.) and stained using
DiffQuik (LabAids, Brisbane, Australia). The slides were coded and
scored using established scoring methods and criteria (18). Binucleated cells (BNC; 1000 cells) were scored for the presence of MNi,
NBUD and NPB (Fig. 2). The levels of mononucleated, binucleated,
multinucleated, apoptotic and necrotic cells were determined by
scoring 500 cells. All experiments on each subject were conducted in
duplicate, and each experiment on each matched pair was commenced on the same day. The MNi, NBUD and NPB provided a
measure of chromosome breakage or loss, chromosome rearrangement
and gene amplification, respectively. The MNi also provided an
indirect measure of genome hypomethylation, because satellite 2 and
satellite 3 DNA hypomethylation causes the loss of chromosomes 1,
9 and 16 as MNi (27), and the MNi level is directly correlated with
genome hypomethylation (21). Concurrent positive control assays
with lymphocytes exposed to -radiation (1.5 Gy) were conducted
each week, yielding MNi levels of 150 to 220 MNi per 1000 BNC.
The slides were coded and scored by a single individual who was not
aware of the treatment group or genotype of the samples.
The Hcy concentration in d-9 medium was measured by fluorescence polarization immunoassay using the Abbott AXSYM system
(Abbott Laboratories, Axis-Shield, Oslo, Norway). The coefficient of
variation for duplicate measurements was 5%. The Hcy concentration was measured to determine whether the in vitro model replicated
the known in vivo effects of MTHFR genotype, folate and riboflavin
on homocysteine concentration (6,12,13,15).
Statistical analysis of data. All data were log-transformed before
statistical analysis. Results for the various culture media were compared using parametric repeated-measures one-way ANOVA and the
Bonferroni multiple-comparison test. Means of two groups were compared using the matched-pair Students t-test. Repeated-measures
two-way ANOVA was used to determine whether there were significant two-way interactions between folic acid and riboflavin, folic
acid and genotype, riboflavin and genotype, as well as to determine
the percentage of total variation attributable to these factors. These
analyses, as well as the test for significance between the slopes of the
regression lines for the viable cell count dose-response curves, were
conducted using Prism 4.0 software (GraphPad, San Diego, CA).
Effect size, the ratio of the difference between group means and the SD
(28), was also measured. CSS Statistica (Statsoft, Tulsa, OK)

MTHFR, FOLIC ACID, RIBOFLAVIN AND GENOME STABILITY

was used for three-way ANOVA to identify any significant three way
interactions, ANOVA post hoc to determine the means for each
independent variable category (e.g., CC or TT genotype) and Pearsons test to determine the extent and significance of correlation
among all measured variables. Values of P 0.05 were considered
significant.

RESULTS
Preliminary experiments showed a clear dose response in
cell growth at folic acid concentrations between 4 and 100

nmol/L; 20 nmol/L was the lowest concentration that produced an increase in the number of viable cells. There was also
a dose response in cell growth at methionine concentrations
between 4 and 100 mol/L; 16 mol/L was the lowest concentration that adequately sustained cell growth (Fig. 4A, B).
Riboflavin concentrations from 0 to 500 nmol/L did not affect
cell growth (Fig. 4C).
Based on the preliminary data, we fixed the methionine
concentration at 50 mol/L, a concentration that is closer to
the physiological range and yet not too low that it might
compromise cell growth at the lowest tolerable folic acid
concentration. The low and high concentrations of folic acid
selected were 20 nmol/L (LF) and 100 nmol/L (HF), respectively. These folic acid concentrations are within or close to
the physiological range in plasma, which ranges from 7
nmol/L in subjects with a negative folate balance to 50
nmol/L in subjects consuming 400 g/d of folate (7,8,21).
The low and high concentrations of riboflavin selected were 0
nmol/L (LR) and 500 nmol/L (HR), respectively, to provide a
deficient dose and a supraphysiological dose [plasma riboflavin
concentrations normally range from 5 to 50 nmol/L (6)] and
maximize the chances of eliciting an effect on homocysteine or
genome stability. We then tested the genome stability, viable
cell number and medium Hcy concentration of MTHFR 677
CC and TT lymphocytes cultured for 9 d in each of the four
combinations of low and high folic acid and riboflavin media
(i.e., LFLR, LFHR, HFLR and HFHR; Table 2).We also
calculated the percentage of total variation that could be
explained by genotype, folic acid or riboflavin concentration
for each variable measured and the effect size of genotype, folic
acid and riboflavin concentration on each variable measured
(Table 3).
Cell growth was significantly increased in lymphocytes cultured in the HF media and in those with the TT genotype
(Tables 2 and 3). Folic acid concentration and genotype
accounted for 16.5 and 11.3% of the increase, respectively;
however, these two factors did not interact with each other or
with riboflavin. To verify that the culture conditions and
genotype affected the metabolism and bioavailability of folate,
we measured Hcy concentration on d 9. The concentration
was significanly decreased by increasing concentrations of folic
acid and riboflavin (Table 2). Genotype, folic acid and riboflavin accounted for 8, 10 and 5% of the decrease in Hcy
concentration, respectively (Table 3); however, there was no
interaction among the three factors.
Levels of the chromosomal stability biomarkers MNi,
NBUD and NPB were minimized in the high folic acid media
(Table 2). Folic acid accounted for 30, 39 and 12% of the
variance in MNi, NBUD and NPB levels, respectively, and
genotype accounted for 4 and 11% of the variance in MNi and
NBUD levels, respectively (Table 3). The NBUD levels were
27% lower in TT compared with CC cells (P 0.002),
whereas MNi levels were 21% higher in TT compared with
CC cells (P 0.05). Riboflavin affected only the NBUD
levels, which were 25% higher in cells cultured in the LFHR
medium than in cells cultured in the LFLR medium (P 0.05)
when the data for the CC and TT genotypes were combined
(Table 2). Folic acid and riboflavin interacted to affect NBUD
levels significantly (P 0.042). Folic acid and riboflavin did
not affect apoptosis and necrosis (data not shown). However,
there was a marginal (14%) reduction in apoptosis in the TT
compared with the CC genotype (P 0.10), with genotype
accounting for 3% of the decrease in apoptosis rate (P
0.05; Table 3).
To analyze the effect of MTHFR C677T polymorphism
directly, we combined the data for cells cultured in each of the

Downloaded from jn.nutrition.org at UMD of New Jersey G F Smith Library on April 16, 2009

FIGURE 4 Growth of human lymphocytes in media containing

various concentrations of folic acid (A), L-methionine (B) and riboflavin
(C). Values are means SEM, n 2 subjects. The slopes of the
dose-response regression lines of 50 and 100 nmol/L folic acid differ
from those of the lower concentrations (P 0.001). The slopes of the
dose-response regression lines of 50 and 100 mol/L methionine differ
from those of the lower concentrations (P 0.001). The slopes of the
dose-response regression lines of the various riboflavin concentrations
do not differ.

51

KIMURA ET AL.

52

TABLE 2
Comparison of genome stability biomarkers, homocysteine and number of viable cells in human lymphocytes grown for 9 d in
culture media containing different combinations of folic acid and riboflavin concentrations1,2
Culture medium
MTHFR C677T genotype

LFLR

LFHR

HFHR

P-value3

9.8 2.3b
12.2 1.3c
11.0 1.3b

12.1 1.6b
14.2 2.4bc
13.2 1.4b

0.0024
0.0038
0.0001

19.6 2.7b
15.5 1.6b
17.5 1.6c

20.3 1.9b
12.4 1.0b
16.4 1.5c

0.0001
0.0001
0.0001

10.3 1.3
11.1 2.6b
10.7 1.4b

14.0 1.6
10.5 1.5b
12.3 1.2ab

0.2268
0.0439
0.0162

1.36 0.17ab
1.67 0.14a
1.51 0.12b

1.29 0.09b
1.39 0.13b
1.34 0.08b

0.0161
0.0007
0.0001

0.92 0.07a
1.13 0.16a
1.02 0.09a

0.86 0.06a
1.09 0.15a
0.98 0.08a

0.0001
0.0001
0.0001

HFLR

MNi, n/1000 BNC

CC
TT
CC & TT

16.9 1.6ab
23.5 3.4a
20.2 2.0a

20.8 3.8a
22.4 3.4ab
21.6 2.5a
NBUD, n/1000 BNC

32.1 2.4a
23.1 3.7a
27.6 2.5a

38.7 5.2a
30.0 3.7a
34.4 3.3b
NPB, n/1000 BNC

CC
TT
CC & TT

14.6 1.8
13.6 3.0b
14.1 1.7ab

15.3 1.9
20.3 3.6a
17.8 2.1a
Hcy, mol/L

CC
TT
CC & TT

1.67 0.15a
1.93 0.20a
1.80 0.12a

1.44 0.09ab
1.74 0.20a
1.59 0.11ab
Viable cells, n 109/L

CC
TT
CC & TT

0.69 0.05b
0.86 0.11b
0.77 0.06b

0.65 0.05b
0.83 0.11b
0.74 0.06b

1 Values are means SEM; n 7 for CC and TT; n 14 for CC & TT. Means in a row with superscripts without a common letter differ, P 0.05.
2 Abbreviations used: BNC, binucleated cell; CC, wild-type methylenetetrahydrofolate reductase C677T homozygote; Hcy, homocysteine; HFHR,

high folic acid (100 nmol/L) and high riboflavin (500 nmol/L) culture medium; HFLR, high folic acid (100 nmol/L) and low riboflavin (0 nmol/L) culture
medium; LFHR, low folic acid (20 nmol/L) and high riboflavin (500 nmol/L) culture medium; LFLR, low folic acid (20 nmol/L) and low riboflavin (0 nmol/L)
culture medium; MNi, micronuclei; MTHFR, methylenetetrahydrofolate reductase; NPB, nucleoplasmic bridge; NBUD, nuclear bud; TT, mutant
methylenetetrahydrofolate reductase C677T homozygote.
3 Repeated-measures one-way ANOVA of log-transformed data.

four media and compared cells with the CC and TT genotypes

cultured under LF (20 nmol/L folic acid) and HF (100 nmol/L
folic acid) conditions (Fig. 5AD). The NBUD level was
lower, and the rate of cell growth was greater, in TT cells than
in CC cells under both LF and HF conditions. The greater
MNi level in TT cells than in CC cells was not significant
when the LF and HF data were analyzed separately (Fig. 5A).
The Hcy concentration was greater in TT cells than in CC
cells, but only under LF conditions.
This study measured seven variables related to cell division,
genome stability, cell death and folate metabolism. To clarify
the interrelationship among these factors, we performed a
cross-correlation analysis, as summarized in Table 4. The Hcy
concentration correlated positively with MNi level and negatively with APOP cells. The NBUD level correlated positively with APOP cells, MNi level and NPB level. The number of viable cells on d 9 correlated negatively with APOP
cells and NBUD level. The NPB, NBUD and MNi levels
correlated positively with each other, and APOP and NEC
cells correlated positively with each other.
DISCUSSION
Chromosome breakage, loss and rearrangement are important initiating events in cancer; however, they also play an
important role during the evolution of cancer when a genome

instability phenotype is established (19,29 33). Gene mutations and gene silencing are known to play a critical role in the
inactivation of genes involved in DNA repair, in cell cycle
control, in the appropriate segregation of chromosomes during
mitosis and in apoptosis (34,35), but less is known regarding
the effect of dietary factors on the genome instability phenotype. We therefore focused our research on the development of
a tissue culture model that may help to predict the effects of
diet on the chromosomal stability of human lymphocytes
depending on genotype. This could have important applications in the following fields: 1) determining the optimal concentration of micronutrients for genome stability, as a guide to
establishing recommended dietary allowances for the prevention of genome damage (4,36), and 2) the development of
optimal culture media for the growth of cells required in the
biotechnology industry (e.g., cell lines for protein production,
stem cells for tissue repair and lymphocytes for cancer immunotherapy).
The folic acidmethionine pathway (Fig. 1) is particularly
relevant to the control of genome stability and involves a
number of critical enzymes for which several polymorphisms
have been identified. In addition, several of these enzymes
require vitamins as cofactors; e.g., MTHFR, methionine synthase and serine hydroxymethyltransferase require riboflavin
(as a precursor for FAD), cobalamin and pyridoxine, respec-

Downloaded from jn.nutrition.org at UMD of New Jersey G F Smith Library on April 16, 2009

CC
TT
CC & TT

MTHFR, FOLIC ACID, RIBOFLAVIN AND GENOME STABILITY

53

TABLE 3

TABLE 4

Effect of genotype, folic acid and riboflavin on variables

measured in human lymphocytes grown in various culture
media, expressed as percentage of total variation (V)
accounted for by genotype, folic acid concentration and
riboflavin concentration and their effect size (ES)13

Cross-correlation matrix of variables measured in human

lymphocytes grown in various culture media1,2

Genotype

Folic acid

Riboflavin

V, %

ES

V, %

ES

V, %

ES

MNi
NBUD
NPB
APOP
NEC
Hcy
Viable cells

3.8*
11.0
0.1
2.9*
0.5
7.9*
11.3*

0.33*
0.59
0.05
0.30
0.12
0.56*
0.67*

29.7**
39.0**
12.3*
0.0
0.0
9.9
16.5

0.93**
1.10**
0.57
0.00
0.02
0.62*
0.76

1.3
1.6
4.3
1.7
1.0
5.0
0.4

0.19
0.22
0.34
0.23
0.18
0.46
0.10

1 Values determined by two-way ANOVA; * P 0.05, P 0.005,

** P 0.0005.
2 Effect size is expressed as the ratio of the difference between the
mean values for the two categories for each independent variable [i.e.,
genotype, mutant (TT) and wild type (CC); folic acid, high and low
concentration; or riboflavin, high and low concentration] and the standard deviation for the dependent variable.
3 Abbreviations: APOP, apoptotic cell; Hcy, homocysteine; MNi,
micronuclei; NEC, necrotic cell; NPB, nucleoplasmic bridge; NBUD,
nuclear bud.

tively. Therefore, this pathway provides an ideal opportunity

to study the effects of nutrition on genome instability. The
present study is the first to define how folic acid, riboflavin and

FIGURE 5 Comparison of micronuclei (MNi), a biomarker of chromosome breakage or loss, per 1000 binucleated cells (BNC) (A); medium homocysteine (Hcy) concentration (B); nuclear buds (NBUD), a
biomarker for gene amplification, per 1000 BNC (C); and number of
viable cells (D) in human lymphocytes of the wild-type (CC) and mutant
(TT) homozygotes of the methylenetetrahydrofolate reductase (MTHFR)
C677T genotypes cultured in media with high and low concentrations
of folic acid (F). Values are means SEM, n 7. Bars marked with an
asterisk differ from CC, P 0.05.

APOP

MNi

NPB

NBUD

Hcy

0.09
0.06
0.12
0.27*

0.42
0.17
0.30*
0.36*
0.36*

0.24
0.13
0.60**
0.32*
0.10
0.13

r
Viable
cells
NEC
APOP
MNi
NPB
NBUD

0.21

0.33*
0.42

0.16
0.26
0.16

1 Values are estimated Spearman correlation coefficients; * P

0.05, P 0.005, ** P 0.0005.
2 Abbreviations: APOP, apoptotic cell; Hcy, homocysteine; MNi,
micronuclei; NEC, necrotic cell; NPB, nucleoplasmic bridge; NBUD,
nuclear bud.

MTHFR C677T polymorphism interact to determine the

chromosomal stability of lymphocytes at physiologically relevant concentrations of folic acid, riboflavin and methionine in
vitro. However, the results must be considered with some
caution, because the culture conditions may not predict precisely what happens in vivo; the RPMI 1640 medium may be
deficient in key micronutrients involved in DNA repair (e.g.,
zinc) or contain supraphysiological concentrations of other
micronutrients involved in the folatemethionine cycle such
as choline, the precursor of betaine. The concentration of
choline is unlikely to be an important contributor to methionine
synthesis in this culture system, because betaine-homocysteinemethyl-transferase is not expressed in lymphocytes (37). The
type of folate (i.e., 5-methyltetrahydrofolate or folic acid)
might also affect the results, although we previously showed
that chromosomal instability does not differ when these two
types of folate are compared in the same culture system over
the concentration range used in this experiment (38).
The results suggest that folic acid and methionine are two
of the key determinants of lymphocyte growth in culture
media and that riboflavin does not affect it. In addition, it is
evident that the TT genotype provided a significant growth
advantage over the CC genotype over the long-term culture
period of 9 d. The growth advantage of the TT genotype may
be related to reduced cell cycle delay, which may be caused by
the nuclear budding process that occurs during S-phase (39),
because TT cells express significantly fewer nuclear buds than
CC cells. An alternative explanation is the slight reduction in
apoptosis rate in TT cells compared with CC cells.
A novel aspect of the present study is the observation that
the NBUD level was markedly higher in cells cultured in
LFHR medium compared with LFLR medium, indicating that
excess riboflavin may be genotoxic at low folate concentrations. It is important to note that the NBUD level is also the
chromosomal instability marker that was most affected by the
MTHFR polymorphism (i.e., the NBUD level was lower in TT
cells than in CC cells). These data suggest that the mechanism
that causes NBUD (a biomarker for gene amplification) under
folate-deficient conditions may in fact be aggravated either
when the riboflavin concentration is increased or when the
CC genotype is present. A high riboflavin concentration may
increase the activity of MTHFR, which could cause folate to
provide methyl groups for methionine synthesis rather than for
thymidylate synthase. The net result could be increased uracil

Downloaded from jn.nutrition.org at UMD of New Jersey G F Smith Library on April 16, 2009

Variable

NEC

54

KIMURA ET AL.

in the DNA, which could lead to the generation of breakagefusion-bridge cycles, leading to gene amplification and the
removal of amplified DNA by nuclear budding, as explained in
our recent reviews (19,22).
One puzzling result of this study is that, although NBUD
were markedly modified by genotype (i.e., reduced levels in TT
cells), there was no trend toward a reduction of NPB and MNi
levels in TT cells compared with CC cells. In fact. MNi levels
increased marginally in TT cells. This appears to be counterintuitive, given that uracil in DNA correlated positively with
MNi in our previous studies using this system (16) and in other
studies (7,8). However, MNi may originate not only from
chromosome breakage caused by uracil in DNA but also from
chromosome loss events that may be caused by hypomethylation of DNA (3). The TT genotype increases DNA hypomethylation in lymphocytes in vivo (15), and DNA hypomethylation in lymphocytes in vitro or in vivo increases the
loss of chromosomes 1, 9 and 16, which are then included in
MNi (40,41). Therefore, the marginal increase in MNi levels
in TT cells may be due to increased chromosome loss events.
A possible weaknesses of our study is that we did not
measure DNA methylation directly. However, as indicated
above, MNi levels increase under demethylating conditions
[e.g., in 5-azacytidine treatment or defects in DNA methyl
transferase, as in immunodeficiency, centrometric region instability and facial anomalies (ICF) syndrome] (27,42,43) and
therefore provide a measure of genome hypomethylation. Furthermore, there is a direct relationship between CpG hypomethylation and MNi expression in vivo (21). In addition, MNi
levels under folate-deficient conditions correlate with uracil in
DNA, which is another marker of DNA hypomethylation
(16). These observations are supported by the tendency shown
in our study for MNi levels to increase in MTHFR TT cells
that exhibit CpG hypomethylation (15) and increased MNi
levels in lymphocytes in vivo (44).
The complexity of the interrelationships among MTHFR
genotype, folic acid and riboflavin is illustrated in Figure 6,
which provides a succinct mechanistic framework for the
results of this study. This mechanistic framework predicts that
1)genome instability from breakage-fusion-bridge cycles and
aneuploidy is minimized when folate concentration is increased, 2) high MTHFR activity minimizes genome hypomethylation and aneuploidy caused by chromosome loss or

gain at the expense of increased breakage-fusion-bridge cycles

and vice versa, 3) high riboflavin concentration in the presence of low folate concentration increases the risk of breakagefusion-bridge cycles, 4) low riboflavin concentration in the
presence of low folate concentration maximizes the risk of
genome hypomethylation and aneuploidy caused by chromosome loss or gain, and 5) MNi in MTHFR C677C and
MTHFR T677T may not appear to be very different because a
low MTHFR activity may decrease MNi caused by uracil and
chromosome breakage but also increase MNi originating from
chromosome loss or gain caused by CpG hypomethylation and
vice versa. These mechanistic interrelationships among
MTHFR genotype, folate, riboflavin and genome instability
may explain why the MTHFR C677T polymorphism reduces
risk for certain cancers, such as leukemia (9), lymphoma (10)
and colorectal cancer (11), but increases risk for Down syndrome (12), neural tube defects (13) and cervical cancer (14).
We suggest that prevention of chromosome breakage and
breakage-fusion-bridge cycles caused by uracil in DNA may be
more relevant to the prevention of cancers such as lymphoma
and leukemia, whereas prevention of CpG hypomethylation,
which may be associated with chromosome loss or gain, could
be more relevant to the minimization of risk for cancers caused
by integration and expression of parasitic DNA (e.g., human
papilloma virus in cervical cancer) and/or cancers caused by
aneuploidy (31) and developmental defects caused by aneuploidy, such as Down syndrome.
As noted above, we are uncertain as to how precisely the in
vitro culture conditions used in this study reflect conditions in
vivo. However, they may constitute a close approximation,
because the Hcy concentration data from this study are in
good agreement with in vivo data that show an increment in
Hcy concentration for the TT genotype, low folic acid and low
riboflavin (6,45). Both low folate and MTHFR C677T polymorphism increase plasma Hcy in vivo (46). Our in vitro
experiments replicated this effect, showing a clear increment
in medium Hcy concentration in the MTHFR TT cells compared with the MTHFR CC cells and an increase with the
reduced folic acid concentration. The Hcy concentration also
increased moderately with the reduced riboflavin concentration, which corresponds with in vivo data (45). The extent of
the total variation in Hcy concentration accounted for by the
MTHFR C677T genotype was 7.9%, which is comparable to

Downloaded from jn.nutrition.org at UMD of New Jersey G F Smith Library on April 16, 2009

FIGURE 6 Mechanistic framework

explaining the interrelationship between
methylenetetrahydrofolate
reductase
(MTHFR) genotype, riboflavin (R) and folic
acid (F) with respect to 1) CpG methylation and uracil in DNA; 2) aneuploidy
and micronuclei (MNi) originating from
chromosome loss events; 3) MNi originating from acentric chromosome fragments; nuclear buds (NBUD), nucleoplasmic bridges (NPB) and breakagefusion-bridge (BFB) cycles; 4) initiation of
cancer caused by cytosine-phosphateguanosine denucleotide (CpG) hypomethylation and aneuploidy; and 5) initiation of cancer caused by increased BFB
cycles, MNi originating from acentric chromosome fragments, NBUD and NPB. *For
brevity, other carcinogenic mechanisms induced by altered genome methylation,
such as silencing of tumor suppressor
genes and/or activation of oncogenes, are
not included in the diagram.

MTHFR, FOLIC ACID, RIBOFLAVIN AND GENOME STABILITY

ACKNOWLEDGMENTS
We are grateful to the volunteers who made this study possible by
kindly donating a blood sample. The assistance of Julie Turner,
Felicia Bulman and Carolyn Salisbury at various occasions throughout the project is gratefully acknowledged. Professor Bruce N. Ames
and Susan Mashiyama are kindly thanked for their comments and
suggestions regarding the interpretation of the results obtained. We
would also like to indicate our gratitude to Phil Leppard for his
valuable advice on statistical analysis of the data.

LITERATURE CITED
1. Selhub, J. & Rosenberg, I. H. (1996) Folic acid. In: Present Knowledge
in Nutrition (Ziegler, E. E. & Filer, L. J., eds.), 7th ed., pp. 206 219. ILSI Press,
Washington, D.C.
2. Choi, S. & Mason, J. B. (2002) Folate status: effects on pathways of
colorectal carcinogenesis. J. Nutr. 132: 2413S2418S.
3. Fenech, M. (2001) The role of folic acid and vitamin B12 in genomic
stability of human cells. Mutat. Res. 475: 56 67.
4. Ames, B. N. & Wakimoto, P. (2002) Are vitamin and mineral deficiencies a major cancer risk? Nature Rev. Cancer 2: 694 704.
5. Rivlin, R. S. (1996) Riboflavin. In: Present Knowledge in Nutrition
(Ziegler, E. E. & Filer, L. J., eds.), 7th ed., pp. 206 219. ILSI Press, Washington,
D.C.
6. Hustad S., Ueland, P. M., Vollset, S. E., Zhang, Y., Bjorke-Monsen, A. L.
& Schneede, J. (2000) Riboflavin as a determinant of plasma total homocysteine: effect of modification by the methylenetetrahydrofolate reductase C677T
polymorphism. Clin. Chem. 46(8): 10651071.
7. Blount B. C., & Ames, B. N. (1995) DNA damage in folate deficiency.
Bailleres Clin. Haematol. 8: 461 475.
8. Blount, B. C., Mack, M. M., Wehr, C. M., MacGregor, J. T., Hiatt, R. A,
Wang, G., Wickramasinghe, S. N, Everson, R. B. & Ames, B. N. (1997) Folate
deficiency causes uracil midincorporation into human DNA and chromosome
breakage: implications for cancer and neuronal damage. Proc Natl. Acad. Sci.
U.S.A. 94: 3290 3295.
9. Skibola, C. F., Smith, M. T., Kane, E., Roman, E., Rollinson, S., Cartwright
R. A. & Morgan, G. (1999) Polymorphisms in the methylenetetrahydrofolate
reductase gene are associated with susceptibility to acute leukaemia in adults.
Proc. Natl. Acad. Sci. U.S.A. 96: 12810 12815.
10. Matsuo, K., Suzuki, R., Hamajima, N., Ogura, M., Kagami, Y., Taji, H.,
Kondoh, E., Maeda, S., Asakura, S., Kaba, S., Nakamura, S., Seto, M., Morishima,
Y. & Tajima, K. (2001) Association between polymorphisms of folate- and
methionine-metabolising enzymes and susceptibility to malignant lymphoma.
Blood 97(10): 32053209.
11. Giovannucci, E. (2002) Epidemiological studies of folate and colorectal neoplasia: a review. J. Nutr. 132: 2350S2355S.
12. James, S. J., Pogribna, M., Pogribny, I. P., Melnyk, S., Hine, R. J., Gibson,
J. B., Yi, P., Tafoya, D. L., Swenson, D. H., Wilson, V. L. & Gaylor, D. W. (1999)
Abnormal folate metabolism and mutation in the methylenetetrahydrofolate reductase gene may be maternal risk factors for Down syndrome. Am. J. Clin. Nutr.
70: 495501.
13. Van der Put, N., Gabreels, F., Stevens, E. M., Smeitink, J. A., Trijbels,
F. J., Eskes, T. K., van den Heuvel, L. P. & Blom, H. J. (1998) A second
common mutation in the methylene tetrahydrofolate reductase gene: an additional risk factor for neural tube defects? Am. J. Hum. Genet. 62: 1044 1051.
14. Goodman, M. T., McDuffie, K., Hernandez, B., Wilkens, L. R., Bertram,
C. C., Killeen, J., Le Marchand, L., Selhub, J., Murphy, S. & Donlon, T. A. (2001)
Association of methylenetetrahydrofolate reductase polymorphism C677T and
dietary folate with the risk of cervical dysplasia. Cancer Epidemiol. Biomark. Prev.
10: 12751280.
15. Friso, S., Choi, S., Girelli, D., Mason, J. B., Dolnikowski, G. G., Bagley,
P. J., Olivieri, O., Jacques, P. F., Rosenberg, I. H., Corrocher, R. & Selhub, J.
(2002) A common mutation in the 5,10-methylenetetrahydrofolate reductase
gene affects genomic DNA methylation through an interaction with folate staus.
Proc. Natl. Acad. Sci. U.S.A. 99(8): 5606 5611.
16. Crott, J. W., Mashiyama, S. T., Ames, B. N. & Fenech, M. (2001)
MTHFR C677T polymorphism does not alter folic acid deficiency-induced uracil
incorporation into primary human lymphocyte DNA in vitro. Carcinogenesis 22:
1019 1025.
17. Crott, J. W., Mashiyama, S. T., Ames, B. N. & Fenech, M. (2001) The
effect of folic acid deficiency and MTHFR C677T polymorphism on chromosome
damage in human lymphocytesin vitro. Cancer Epidemiol. Biomark. Prev. 10:
1089 1096.
18. Fenech, M. (2000) The in vitro micronucleus technique. Mutat. Res.
455: 8195.
19. Fenech, M. (2002) Chromosomal biomarkers of genomic instability
relevant to cancer. Drug Discov. Today 7(22): 1128 1137.
20. Fenech, M., Dreosti, I. E. & Rinaldi, J. R. (1997) Folate, vitamin B12,
homocysteine status and chromosome damage rate in lymphocytes of old men.
Carcinogenesis 18: 1329 1336.
21. Fenech, M., Aitken, C. & Rinaldi, J. (1998) Folate, vitamin B12, homocysteine status and DNA damage in young Australian adults. Carcinogenesis 19:
11631171.
22. Fenech, M. & Crott, J. W. (2002) Micronuclei, nucleoplasmic bridges
and nuclear buds induced in folic acid deficient human lymphocytes evidence
for breakage-fusion-bridge cycles in the cytokinesis-block micronucleus assay.
Mutat. Res. 504(12): 131136.
23. Ubbink, J. B., van der Merwe, A., Delport, R., Allen, R. H., Stabler, S. P.,
Reizler, R. & Vermaak, W. J. (1996) The effect of a subnormal vitamin B6 status
on homocysteine metabolism. J. Clin. Invest. 98: 177184.
24. Ames, B. N., Elson-Schwab, I. & Silver, E. A. (2002) High-dose vitamin
therapy stimulates variant enzymes with decreased coenzyme binding affinity
(increased K(m)): relevance to genetic diseases and polymorphisms. Am. J. Clin.
Nutr. 75(4): 616 658.
25. Leclerc, D., Campeau, E., Goyette, P., Adjalla, C. E., Christensen, B.,
Ross, M., Eydoux, P., Rosenblatt, D. S., Rozen, R. & Gravel, R. A. (1996)

Downloaded from jn.nutrition.org at UMD of New Jersey G F Smith Library on April 16, 2009

results with in vivo data, i.e., 9.0% of variation accounted for

by the MTHFR C677T genotype (47). These data indicate
that the in vitro model approximates the in vivo situation
quite well. The positive correlation of Hcy concentration with
cell proliferation fits well with other studies (48 50), as does
the positive correlation with MNi, which is precisely the
relationship we have reported in our ex vivo studies in older
men (20) and young adults (21). These data and the relatively
strong negative correlation of Hcy concentration with apoptosis further support the emerging hypothesis that elevated Hcy
concentration is a risk factor for carcinogenic events
(20,21,51).
The significant positive correlation among NPB, MNi and
NBUD is consistent with our previous studies and provides
further evidence that folate deficiency induces breakage-fusion-bridge cycles. The evidence for breakage-fusion-bridge
cycles as a key mechanism in cancer may explain not only
gene amplification and rapid evolution of the karyotypic abnormalities of cancer but also centrosome abnormalities,
which may result when cytokinesis is inhibited by the presence
of anaphase bridges induced by chromosome breakage and
rearrangement or loss of telomeric DNA (19,30,33). The fact
that these events occur at high frequency with moderate folate
deficiency within the normal physiological range (22) underscores the relevance of folate deficiency as an important
risk factor for cancer.
Given that the MTHFR 677 TT homozygotes have a lower
risk of developing adult acute lymphocytic leukemia (9) and
lymphoma (10), it is important to ask whether any of the
biomarkers examined in this study may be potential risk factors
for these cancers. It appears from this preliminary data that
NBUD is the biomarker most strongly related to the TT
genotype, whereas MNi, NBUD and NPB are generally negatively correlated with folic acid, deficiency of which is a risk
factor for a variety of cancers (4,11,52). Which of these
biomarkers is the strongest predictor of cancer risk should be
determined by prospective epidemiological studies; such studies have already linked elevated chromosome aberration rate
with cancer risk (53).
In conclusion, this study shows that an in vitro culture
system with physiologically relevant concentrations of folic
acid, riboflavin and methionine is a valid model for studying
the effects of nutrition on genome instability. However, given
the relatively small number of subjects in this study, the
observed effects of the MTHFR genotype and riboflavin must
be considered preliminary. This model can be further improved for use in predicting in vivo events by developing a
culture medium that matches the physiological concentrations
of all other micronutrients and metabolites found in body
fluids. We anticipate that such models can be used to determine the optimal micronutrient concentrations required to
minimize genome instability in genetic subgroups and more
specifically in individuals.

55

56

KIMURA ET AL.
instability and immunodeficiency syndrome caused by mutations in a DNA methyltransferase gene. Nature 402: 187191.
42. Sawyer, J. R., Swanson, C. M., Wheeler, G. & Cunniff, C. (1995) Chromosome instability in ICF syndrome: formation of micronuclei from multibranched
chromosome 1 demonstrated by fluorescence in situ hybridisation. Am. J. Med.
Genet. 56(2): 203209.
43. Stopper, H., Korber, C., Gibis, P., Spencer, D. L. & Caspary W. J. (1995)
Micronuclei induced by modulators of methylation: analogs of 5-azacytidine.
Carcinogenesis 16(7): 16471650.
44. Botto, N., Andreassi, M. G., Manfredi, S., Masetti, S., Cocci, F., Colombo,
M. G., Storti, S., Rizza, A. & Biagini, A. (2003) Genetic polymorphisms in folate
and homocysteine metabolism as risk factors for DNA damage. Eur. J. Hum.
Genet. 11(9): 671 678.
45. Jacques, P. F., Kalmbach, R., Bagley, P. J., Russo, G. T., Rogers, G.,
Wilson, P.W.F., Rosenberg, I. H. & Selhub, J. (2002) The relationship between
riboflavin and plasma total homocysteine in the Framingham Offspring Cohort is
influenced by folate status and the C677T transition in the methylenetetrahydrofolate reductase gene. J. Nutr. 132: 282288.
46. DeBree, A., Verschuren, W. M., Kronhout, D., Kluijtmans, L. A. & Blom,
H. J. (2002) Homocysteine determinants and the evidence to what extent
homocysteine determines the risk of coronary heart disease. Pharmacol. Rev.
54(4): 599 618.
47. Kluijtmans, L. A., Young, I. S., Boreham, C. A., Murray, L., McMaster, D.,
McNulty, H., Strain, J. J., McPartlin, J., Scott, J. M. & Whitehead, A. S. (2003)
Genetic and nutritional factors contributing to hyperhomocysteinemia. Blood
101(7): 24832488.
48. Buemi, M., Marino, D., Di Pasquale, G., Floccari, F., Ruello, A., Aloisi, C.,
Corica, F., Senatore, M., Romeo, A. & Frisina, N. (2001) Effects of homocysteine on proliferation, necrosis, and apoptosis of vascular smooth muscle cells in
culture and influence of folic acid. Thromb. Res. 104(3): 207213.
49. Sun, C. F., Haven, T. R., Wu, T. L., Tsao, K. C. & Wu, J. T. (2002)
Serum total homocysteine increases with the rapid proliferation rate of tumor cells
and decline upon cell death: a potential new tumor marker. Clin. Chim. Acta
321(12): 55 62.
50. Zhang, Q., Zeng, X., Guo, J. & Wang, X. (2002) Oxidant stress mechanism of homocysteine potentiating Con A-induced proliferation in murine splenic
T lymphocytes. Cardiovasc. Res. 53(4): 10351042.
51. Wu, L. L. & Wu, J. T. (2002) Hyperhomocysteinemia is a risk factor for
cancer and a new potential tumor marker. Clin. Chim. Acta 322: 2128.
52. Thompson, J. R., Fitzgerald, P., Willoughby, M.L.N. & Armstrong, B. K.
(2001) Maternal folate supplementation in pregnancy and protection against
acute lymphoblastic leukaemia in childhood: a case-control study. Lancet 358:
19351940.
53. Bonassi, S., Hagmar, L., Stromberg, U., Montagud, A. H., Tinnerberg, H.,
Forni, A., Heikkila, P., Wanders, S., Wilhardt, P., Hansteen, I. L., Knudsen, L. E. &
Norrpa, H. (2002) Chromosomal aberrations in lymphocytes predict human
cancer independently of exposure to carcinogens. Cancer Res. 60(6): 1619 1625.

Downloaded from jn.nutrition.org at UMD of New Jersey G F Smith Library on April 16, 2009

Human methionine synthase: cDNA cloning and identification of mutations in

patients of the cblG complementation group of folic acid/cobalamin disorders.
Hum. Mol. Genet. 5: 18671874.
26. Moore, G. E. & Woods, L. K. (1976) Culture media for human cells
RPMI 1603, RPMI 1634, RPMI 1640 and GEM 1717. In: Tissue Culture Association Manual, vol. 3, pp. 503508. Tissue Culture Association, Rockville, MD.
27. Suzuki, T., Fuji, M. & Ayusawa, D. (2002) Demethylation of classical
satellite 2 and 3 DNA with chromosomal instability in senescent human fibroblasts. Exp. Gerontol. 37: 10051014.
28. Norman, G. R. & Streiner, D. L. (1994) Biostatistics. Mosby-Year Book,
St. Louis, MO.
29. Gisselson, D., Bjork, J., Hoglund, M., Mertens, F., Dal Cin, P., Akerman M.
& Mandahl, N. (2001) Abnormal nuclear shape in solid tumours reflects mitotic
instability. Am. J. Pathol. 158(1): 199 206.
30. Gisselson, D., Jonson, T., Martins, C., Mandahl, N., Jin, Y., Mertens, F. &
Jin, C. (2002) Centrosomal abnormalities, multipolar mitoses and chromosomal instability in head and neck tumours with dysfunctional telomeres. Br. J.
Cancer 87: 202207.
31. Li, R. H., Sonik, A., Rasnick, D. & Duesberg, P. (2000) Aneuploidy vs.
gene mutation hypothesis of cancer: recent study claims mutation but is found to
support aneuploidy. Proc. Natl. Acad. Sci. U.S.A. 97(7): 3236 3241.
32. Saunders, W. S., Shuster, M., Huang, X., Gharaibeh, B., Enyenihi, A. K.,
Petersen, I. & Gollin, S. M. Chromosomal instability and cytoskeletal defects in
oral cancer. Proc. Natl. Acad. Sci. U.S.A. 97(1): 303308, 2000.
33. Maser, R. S. & DePinho, R. A. (2002) Connecting chromosomes, crisis
and cancer. Science 297: 565569.
34. Zingg, J. M. & Jones, P. A. (1997) Genetic and epigenetic aspects of
DNA methylation on genome expression, evolution, mutation and carcinogenesis.
Carcinogenesis 18: 869 882.
35. Jones, P. A. & Baylin, S. B. (2002) The fundamental role of epigenetic
events in cancer. Nature Rev. Genet. 3: 415 428.
36. Fenech, M. (2001) Recommended dietary allowances for genomic
stability. Mutat. Res. 480 481: 5154.
37. Wang, J. A., Dudman, N. P., Lynch, J. & Wilcken, D. E. (1991) Betaine:
homocysteine methyl transferasea new assay for the liver enzyme and its
absence from human skin fibroblasts and peripheral blood lymphocytes. Clin.
Chim. Acta 204(13): 239 249.
38. Wang, X. & Fenech, M. (2003) A comparison of folic acid and 5-methyl
tetrahydrofolate for prevention of DNA damage and cell death in human lymphocytes in vitro. Mutagenesis 18(1): 81 86.
39. Shimizu, N., Itoh, N., Utiyama, H. & Wahl, G. M. (1998) Selective
entrapment of extrachromosomally amplified DNA by nuclear budding and micronucleation during S phase. J. Cell Biol. 140: 13071320.
40. Guttenbach, M. & Schmid, M. (1994) Exclusion of specific human
chromosomes into micronuclei by 5-azacytidine treatment in lymphocyte cultures. Exp. Cell Res. 211: 127132.
41. Xu, G. L., Bestor, T. H., Bourchis, D., Hsieh, C. L., Tommerup, N., Bugge,
M., Hulten, M., Qu, X., Russo, J. J. & Veigas-Peguinot, E. (1999) Chromosome