Dietary modulation of mitochondrial DNA deletions and copy number after chemotherapy in rats.

Abstract

Mitochondrial DNA (mtDNA) is particularly susceptible to mutation by alkylating agents, and mitochondrial damage may contribute to the efficacy and toxicity of these agents. We found that folate supplementation decreased the frequency of the "common deletion" (4.8kb, bases 8103-12,936) in liver from untreated rats and from animals treated with cyclophosphamide but not 5-fluorouracil (5-FU). The relative abundance of mitochondrial DNA was greater after chemotherapy but there was no effect of diet. Rats fed with a purified diet had fewer mitochondrial deletions than those maintained on a cereal-based diet after chemotherapy. These results indicate that diet can modulate the extent of mitochondrial damage after cancer chemotherapy, and that folic acid supplementation may be protective against mitochondrial DNA deletions.

Full text

Page 1

Mutation Research 501 (2002) 29–36
Dietary modulation of mitochondrial DNA deletions
and copy number after chemotherapy in rats
Richard F. Brandaa,b,∗, Elice M. Brooksb, Zhuan Chenb,
Shelly J. Naudc, Janice A. Nicklasa,b
aDepartment of Medicine, University of Vermont, 32 N. Prospect Street, Burlington, VT 05405, USA
bThe Vermont Cancer Center, University of Vermont, 32 N. Prospect Street, Burlington, VT 05405, USA
cMedical Biostatistics, University of Vermont, Burlington, VT 05405, USA
Received 14 October 2001; received in revised form 15 November 2001; accepted 20 December 2001
Abstract
Mitochondrial DNA (mtDNA) is particularly susceptible to mutation by alkylating agents, and mitochondrial damage may
contribute to the efficacy and toxicity of these agents. We found that folate supplementation decreased the frequency of the
“common deletion” (4.8kb, bases 8103–12,936) in liver from untreated rats and from animals treated with cyclophosphamide
but not 5-fluorouracil (5-FU). The relative abundance of mitochondrial DNA was greater after chemotherapy but there was no
effectofdiet.Ratsfedwithapurifieddiethadfewermitochondrialdeletionsthanthosemaintainedonacereal-baseddietafter
chemotherapy. These results indicate that diet can modulate the extent of mitochondrial damage after cancer chemotherapy,
and that folic acid supplementation may be protective against mitochondrial DNA deletions. © 2002 Elsevier Science B.V.
All rights reserved.
Keywords: Mitochondrial DNA; Folate; Cyclophosphamide; 5-Fluorouracil
1. Introduction
The late complications of cancer chemotherapy
are of increasing concern. For example more patients
with Hodgkin’s Disease now die of treatment-related
causes than of Hodgkin’s Disease itself [1]. Of these
late complications, deaths from second malignancies
are the most common [1]. The risk of developing
myelodysplasia or acute leukemia increases with the
age of the patient and is linearly related to the total
dose of alkylating agents [1]. Karyotypic analyses of
∗Corresponding author. Present address: Genetics Laboratory,
University of Vermont, 32 N. Prospect Street, Burlington, VT
05401, USA. Tel.: +1-802-656-8355; fax: +1-802-656-8333.
E-mail address: rbranda@zoo.uvm.edu (R.F. Branda).
therapy-related myelodysplasia or acute myelogenous
leukemia demonstrate frequent deletions of chromo-
somes5and7,andlessoftenofchromosomesegments
17p, 12p and 20q [2]. Our laboratory has reported that
folic acid deficiency increases the percentage of intra-
genic deletions caused by alkylating agents in Chinese
hamster ovary cells and in human lymphoblastoid
cells [3,4]. These observations support the idea that
folate supplementation of deficient patients may de-
crease the frequency of intragenic deletions caused
by alkylating agents and thereby reduce the risk of
developing secondary malignancies. Mitochondrial
DNA (mtDNA) deletions also have been associated
with several types of malignancy, including breast,
gastric, colon and renal carcinomas [5]. Studies with
alkylating agents indicate that mtDNA is especially
0027-5107/02/$ – see front matter © 2002 Elsevier Science B.V. All rights reserved.
PII: S0027-5107(02)00014-3

Page 2

30
R.F. Branda et al./Mutation Research 501 (2002) 29–36
susceptible to mutation and that mitochondrial dam-
age may contribute to the cytotoxic activity of these
agents (reviewed in [6,7]). Because of the reports of
the possible involvement of mitochondrial alterations
in carcinogenesis [5,8], we investigated the relation-
ship of diet, and particularly nutritional folate status,
to mtDNA deletions after cancer chemotherapy in rats.
2. Materials and methods
2.1. Animals and diets
The research protocol was approved by the Insti-
tutional Animal Care and Use Committee of the Uni-
versity of Vermont. Female Fischer 344 rats weighing
approximately 60g were obtained from Charles River
Canada (St.-Constant, Que.). The rats were main-
tained in groups of three or four for 10 days and fed a
cereal-based rat chow (Harlan Teklad LM-485; Harlan
Teklad, Madison, WI). This diet consists of 19.92%
protein, 5.67% fat, 4.37% fiber and 4.05Kcal/g gross
energy. Its principal ingredients include ground corn,
soybean meal, ground oats, wheat middlings and al-
falfameal.Itcontainsnoanimalprotein.Thefolicacid
content is 8.21mg/kg, and the Vitamin B12content is
30.00?g/kg. Then the rats were housed individually
in stainless steel wire-bottomed cages. One group
continued on the cereal-based diet, while the others
were maintained on the AIN-93G purified rodent diet
(Dyets Inc., Bethlehem, PA) of varying folate content
(deficient, replete or supplemented) for 6 weeks. This
purified diet is based upon vitamin-free casein and
cornstarch. Folate-replete diets consisted of AIN-93G
with a vitamin supplement that provided 2mg folic
acid/kg and 25?g of Vitamin B12/kg of diet. The fo-
late deficient diet consisted of AIN-93G with vitamin
mix lacking folic acid. The rats receiving high fo-
late were fed the AIN-93G with vitamin supplement
and were injected i.p. daily with folic acid, 50mg/kg
dissolved in 8.4% sodium bicarbonate solution. Then
they were treated with a single intraperitoneal injec-
tion of increasing doses of either cyclophosphamide
or 5-fluorouracil (5-FU). Lyophilized CYTOXAN®
(Mead Johnson/Bristol Myers Squibb Co., Princeton,
NJ) was reconstituted with sterile water for injection,
USP (Abbott Laboratories, North Chicago, IL). A
20mg/ml solution of cyclophosphamide was used for
i.p. injections. 5-FU was obtained from American
Pharmaceutical Partners Inc. (Los Angeles, CA). A
50mg/ml solution of 5-FU was used for i.p. injections.
2.2. Measurement of mtDNA deletions
and copy number
Two weeks after chemotherapy administration,
liver was collected for subsequent analyses after car-
diac puncture and exsanguination of the rats. The
liver tissue was stored at −80◦C until processing.
Total hepatic DNA was isolated using the Qiagen
DNeasy Tissue kit (Valencia, CA). The amount of the
“common deletion” (4.8kb, bases 8103–12,936) in
liver was measured by quantitative co-amplification
of the mitochondrial D-loop and the mitochondrial
deletion using a real-time quantitative polymerase
chain reaction assay. The relative abundance of mi-
tochondrial DNA was determined by co-amplifying
mitochondrial D-loop versus rat ?-actin gene.
2.3. Oligonucleotide primers and TaqMan
probe design
Primers and probes for the rat D-loop, rat mito-
chondrial deletion and rat ?-actin were designed
using Primer Express software (PE, Foster City, CA)
with rat mitochondrial genome from GenBank (acces-
sion X14848). Primers and probes were synthesized
and HPLC purified by the PE Oligo Factory (Foster
City, CA) and primer limiting experiments were per-
formed to determine the proper primer concentrations.
Sequence for the primers and probes can be found in
Table 1.
2.4. Real-time PCR for mitochondrial deletion
Mitochondrial deletion expression was quantified
with a 5?VIC reporter and a 3?TAMRA quencher dye
and D-loop expression with a 5?6FAM reporter and a
3?TAMRA labeled quencher dye. PCR amplification
was carried out in a 50?l reaction consisting of 1×
TaqMan Universal Master Mix (Perkin-Elmer Applied
Biosystems, Foster City, CA), 200nM each mitochon-
drial deletion forward and reverse primers, 100nM
each D-loop forward and reverse primers, and 100nM
each mitochondrial deletion and D-loop probe. The
cycling condition included an initial phase of 2min at

Page 3

R.F. Branda et al./Mutation Research 501 (2002) 29–36
31
Table 1
Sequence of PCR primers and probes
Primers
Rat D-loop FWD
Rat D-loop REV
Rat Actin FWD
Rat Actin REV
Rat Mito Del FWD
Rat Mito Del REV
GGTTCTTACTTCAGGGCCATCA
GATTAGACCCGTTACCATCGAGAT
GGGATGTTTGCTCCAACCAA
GCGCTTTTGACTCAAGGATTTAA
AAGGACGAACCTGAGCCCTAATA
CGAAGTAGATGATCCGTATGCTGTA
Probes
Rat Mito DelVIC-TCACTTTAATCGCCACATCCA-
TAACTGCTGT-TAMRA
VIC-CGGTCGCCTTCACCGTTCCA-
GTT-TAMRA
6FAM-TTGGTTCATCGTCCATACGT-
TCCCCTTA-TAMRA
Rat Actin
Rat D-loop
50◦C, followed by 10min at 95◦C, 40 cycles of 15s at
95◦C and 1min at 60◦C. Each sample was assayed in
duplicate and fluorescence spectra were continuously
monitored by the 7700 Sequence Detection System
(Perkin-Elmer Applied Biosystems, Foster City, CA)
with sequence detection software version 1.6.3.
2.5. Real-time PCR for mitochondrial D-loop
D-loop expression was quantified with the probe
detailed and ?-actin expression was quantified with a
5?VIC reporter and a 3?TAMRA labeled quencher dye.
PCR amplification was carried out in a 50?l reac-
tion consisting of 1× TaqMan Universal Master Mix
(Perkin-Elmer Applied Biosystems, Foster City, CA),
200nM each ?-actin forward and reverse primers,
50nM each D-loop forward and reverse primers, and
100nM each ?-actin and D-loop probe. The cycling
condition included an initial phase of 2min at 50◦C,
followed by 10min at 95◦C, 40 cycles of 15s at 95◦C
and 1min at 60◦C.
2.6. Statistical analyses
Analysis of variance was used to test the signif-
icance of differences in mtDNA deletions and copy
number.
3. Results and discussion
These studies utilized livers from rats that were
maintained on a cereal-based diet or on a purified
diet of varying folate content. The amount of the
“common deletion” (4.8kb, bases 8103–12,936) was
measured by quantitative co-amplification of the mi-
tochondrial D-loop and the mitochondrial deletion
using a real-time reverse transcriptase polymerase
chain reaction assay with the specific primers de-
scribed in Table 1. The amount of mitochondrial
DNA (abundance) relative to genomic DNA was
determined by co-amplifying mitochondrial D-loop
versus rat ?-actin gene. This methodology allows not
only the determination of changes in the amount of
mitochondrial deletion relative to total mitochondrial
DNA after chemotherapy drug treatment in rats on
different diets but also the measurement of changes in
total mitochondrial DNA relative to genomic DNA.
In this system, a smaller ?CTdeletion (mitochondrial
deletion [CTdel]-mitochondrial D-loop [CTD-loop])
indicates more deletions, while a smaller ?CTcopy
number (mitochondrial D-loop [CTD-loop]-?-actin
[CT?-actin]) indicates less total mitochondrial DNA.
3.1. Mitochondrial deletions after
cyclophosphamide treatment
The ?CT values and relative expression (which
indicates the fold difference in deletions or copy
number compared to the folate-replete purified diet)
for mitochondrial deletions and copy number in livers
from rats treated with cyclophosphamide are shown in
Table 2. Overall, there was a significant effect of dose
(P ≤ 0.0001) and of diet (P ≤ 0.0001) on mitochon-
drial deletions in rats treated with cyclophosphamide
(Table 2). For any particular dose, the diet effects
were similar on mitochondrial deletions (P = 0.44).
Within the group that received no cyclophos-
phamide, the high folate group had significantly fewer
deletions than either the low folate (P ≤ 0.0001) or
replete (P = 0.006) groups. There was no significant
difference between the low folate and replete groups
(P = 0.22). These findings indicate that the folate
content of the diet influenced the number of spon-
taneous or background “common deletions” in rat
liver cells. Mitochondrial defects, particularly dele-
tions, accumulate in senescent tissues and have been
implicated in chronic progressive external ophthal-
moplegia, Kearns-Sayre Syndrome and its associated
condition, Pearson Syndrome, in ischemic heart dis-
ease, and in cirrhotic liver (reviewed in [9,10]). A

Page 4

32
R.F. Branda et al./Mutation Research 501 (2002) 29–36
Table 2
Mitochondrial DNA deletions and copy number in rat liver were measured 2 weeks after injection of cyclophosphamide (CTX) in the
indicated doses
Cereal-based dietPurified diet
Low folateFolate-repleteHigh folate
CTX (mg/kg)
?CT(na)Rel Exp
?CT(na)Rel Exp
?CT(na) Rel Exp
?CT(na) Rel Exp
Deletions
0
50
65
110
144
190
4.71 ± 0.53 (6)
4.11 ± 0.27 (6)
4.37 ± 0.35 (6)
4.09 ± 0.12 (6)
3.78 ± 0.21 (6)
4.73 ± 0.51 (5)
1.17
1.78
1.48
1.80
2.23
1.16
4.67 ± 0.23 (6)
4.47 ± 0.27 (6)
4.56 ± 0.73 (6)
4.09 ± 0.30 (3)
4.07 ± 0.18 (4)
4.37 ± 1.00 (2)
1.21
1.39
1.30
1.80
1.83
1.48
4.94 ± 0.43 (6)
4.82 ± 0.14 (6)
4.68 ± 0.28 (6)
4.38 ± 0.17 (4)
4.42 ± 0.22 (4)
4.32 ± 0.78 (2)
1.00
1.09
1.20
1.47
1.43
1.54
5.54 ± 0.64 (6)
5.04 ± 0.18 (5)
4.76 ± 0.54 (7)
4.85 ± 0.15 (3)
4.28 ± 0.17 (3)
3.62 (1)
0.66
0.93
1.13
1.06
1.58
2.50
Copy number
0
50
65
110
144
190
−6.73 ± 0.27 (6)
−8.01 ± 0.19 (6)
−8.22 ± 0.45 (6)
−7.32 ± 0.32 (6)
−7.60 ± 0.35 (6)
−7.41 ± 0.11 (5)
0.89
2.16
2.50
1.34
1.62
1.42
−6.78 ± 0.13 (6)
−7.86 ± 0.34 (6)
−7.78 ± 0.35 (6)
−6.86 ± 0.43 (3)
−7.44 ± 0.18 (4)
−7.18 ± 0.35 (2)
0.92
1.95
1.84
0.97
1.45
1.21
−6.90 ± 0.29 (6)
−7.82 ± 0.11 (6)
−7.95 ± 0.25 (6)
−7.13 ± 0.57 (4)
−7.37 ± 0.20 (4)
−7.11 ± 0.33 (2)
1.00
1.89
2.07
1.17
1.39
1.16
−6.83 ± 0.25 (6)
−7.64 ± 0.11 (5)
−7.85 ± 0.34 (7)
−7.38 ± 0.26 (3)
−6.94 ± 0.09 (3)
−7.45 (1)
0.95
1.67
1.93
1.39
1.03
1.46
The rats were maintained on either a cereal-based diet or a purified diet of varying folate content. The number of PCR cycles required
to exceed a threshold (CT) just above background was calculated for test and reference reactions. CTvalues were determined in duplicate
and averaged for the mitochondrial deletion and D-loop and then subtracted to obtain the ?CTdeletion, expressed as the mean ± S.E.M.
Relative expression was calculated using the equation 2−??C
T
where ??CTequals ?CTdel− ?CTcalwhere the calibrator is the ?CTof
0mg/kg CTX in the folate-replete purified diet. CT values were determined in duplicate and averaged for the mitochondrial D-loop and
?-actin and subtracted to obtain the ?CTcopy number, expressed as the mean±S.E.M. Relative expression (Rel Exp) was calculated using
the equation 2−??C
T
where ??CTequals ?CTcopynumber−?CTcalwhere the calibrator is the ?CTof 0mg/kg CTX in the folate-replete
purified diet. Relative expression indicates the fold difference in deletions or copy number compared to the folate-replete purified diet.
an: number of animals.
“common deletion” that is the result of recombination
between repeats occurs in both humans and rats. In
humans, a 4977bp deletion occurs by recombination
between copies of a 13bp repeat in the ATP8 (repeat
at 8470–8482) and NAD5 (repeat at 13,447–13,459)
genes while in rat, a 4834bp deletion occurs by
recombination between copies of a 16bp repeat in
the ATP6 (repeat at 8103–8118) and ND5 (repeat
at 12,937–12,952) genes [11,12]. The phenomenon
could be caused by a homologous recombinational
event in the region bordered by the repeats, via
slipped mis-pairing. In humans, this common deletion
is found with increasing frequency in aging tissues
and in approximately 30% of patients with chronic
progressive external ophthalmoplegia [9]. The “com-
mon deletion” is often used as a specific indicator
of general damage [12]. Experiments in adult and
senescent rats confirm that the “common deletion”
accumulates during aging [12,13]. At the same time
senescent rats have a higher mtDNA content in the
liver, heart and brain than the adult counterparts [13].
The data presented here suggest that folate deficiency
may enhance, and supplemental folate reduce the
frequency of these deletions.
Compared to untreated controls, rats that received
110 or 144mg/kg cyclophosphamide had significantly
more deletions in all dietary groups. For rats receiv-
ing 50 or 110mg/kg cyclophosphamide, there were
significantly fewer deletions in the high folate group
compared to low folate (P = 0.01). For rats that re-
ceived 50 or 144mg/kg cyclophosphamide, there were
significantly more deletions in rats on the cereal-based
diet than on the folate-replete diet (P < 0.01).
These data confirm previous studies indicating that
mitochondrial damage is increased after treatment
with alkylating agents [6,14–16]. Our findings also
provide evidence that diet can modulate the delete-
rious effects of alkylating agents on mtDNA. After
treatment with cyclophosphamide, rats maintained on
a purified diet and supplemented with folic acid had

Page 5

R.F. Branda et al./Mutation Research 501 (2002) 29–36
33
significantly fewer mtDNA deletions than rats fed
with the same diet but lacking folic acid. The com-
position of the diet also appeared to be important, in
that rats fed with the purified diet had fewer mtDNA
deletions than rats fed with a cereal-based diet. Since
the folate-replete purified diet and the cereal-based
diet contained approximately the same amounts of
folic acid, there are other, as yet unidentified, di-
etary constituents that can protect against alkylating
agent-induced mtDNA damage.
3.2. Effect of cyclophosphamide treatment
on mitochondrial copy number
Overall, there was a significant effect of dose
(P ≤ 0.0001) and of diet (P = 0.014) on mito-
chondrial DNA relative abundance in rats treated
with cyclophosphamide (Table 2). For any particular
dose, the diet effects were similar on mitochon-
drial relative copy number (P = 0.16) for inter-
action. Mitochondrial relative abundance increased
in cyclophosphamide-treated rats. Compared to un-
treated animals, differences were significant (P <
0.02) in all dietary groups except for low folate and
replete diets at the 110mg/kg dose level and high
folate at the 144mg/kg dose level. Rats on the low
folate diet had significantly fewer copies of mtDNA
than animals on the high folate diet at the 110mg/kg
level (P = 0.04) and significantly higher relative
copies than rats maintained on the high folate diet
at the 144mg/kg level (P = 0.04). Otherwise there
was no significant effect of diet on mtDNA relative
abundance in untreated or cyclophosphamide-treated
animals.Therewasnodifferenceinrelativecopynum-
ber in rats on the cereal-based versus folate-replete
diets in untreated or cyclophosphamide-treated rats.
Therefore, mtDNA relative abundance increased in
cyclophosphamide-treated animals, but there was no
consistent effect of diet on mitochondrial relative copy
numberinuntreatedorcyclophosphamide-treatedrats.
Treatment of patients with alkylating agents such
as cyclophosphamide has been associated with an in-
creased risk of secondary malignancies, particularly
acute myelogenous leukemia and its precursor condi-
tion, myelodysplasia [17]. Gattermann has described
mtDNA mutations in sideroblastic anemia and pos-
tulated that age-dependent accumulation of mtDNA
mutations in stem cells contributes to the development
of myelodysplasia [18]. Boultwood et al. reported that
mtDNA is invariably amplified in acute myelogenous
leukemia patients, with levels 2- to 50-fold higher than
levels found in blood or bone marrow samples from
normal individuals [19]). Since cyclophosphamide
treatment increased both mtDNA damage and relative
copy number, these mtDNA changes may contribute
to the role of alkylating agents in carcinogenesis [8].
3.3. Mitochondrial deletions and copy number
after 5-FU treatment
The ?CT values for mitochondrial deletions and
mtDNA relative abundance in livers from rats treated
with 5-FU are shown in Table 3. In the rats treated
with 5-FU, a significant dose by diet interaction was
found for both deletions and relative copy number
(P < 0.01) in an analysis of rats fed with the purified
diet of varying folate content. Significant differences
were found with regard to deletion number at three of
the five dosages in rats on the purified diet (Table 3).
At110mg/kg,thehighfolategroupwaslowerthanthe
low folate group; at the 325mg/kg dose level, the low
folate group was lower than the high folate or replete
groups; at the 546mg/kg dose level, the folate-replete
group was lower than the low folate group. With
regard to mtDNA relative abundance, significant dif-
ferences were found at only two dose levels: at the
110 and 144mg/kg dose levels, the high folate group
was higher than the low folate or folate-replete groups
(Table 3).
In a comparison of the mtDNA analyses of
rats maintained on the cereal-based diet and the
folate-replete purified diet and treated with 5-FU,
there was a significant dose by diet interaction for
both deletions and relative copy number (P < 0.01).
The folate-replete group had significantly fewer dele-
tions than the cereal group at nearly all dosage levels
(P ≤ 0.02) except at the 325mg/kg dose level, where
the cereal group had fewer deletions (P = 0.001).
In the rats that received no drug, these two dietary
groups were not significantly different. With regard
to mtDNA relative abundance, the two dietary groups
were significantly different at two dose levels: 325
and 420mg/kg. In both cases, the cereal group had
the higher relative copy number (P ≤ 0.001).
Thus, in contrast to the findings with cyclophos-
phamide, 5-FU treatment and diet had a more variable

Page 6

34
R.F. Branda et al./Mutation Research 501 (2002) 29–36
Table 3
Mitochondrial DNA deletions and copy number in rat liver measured 2 weeks after injection of 5-FU in the indicated doses
Cereal-based dietPurified diet
Low folate Folate-repleteHigh folate
5-FU
(mg/kg)
?CT(na) Rel Exp
?CT(na) Rel Exp
?CT(na)Rel Exp
?CT(na) Rel Exp
Deletions
0
110
144
325
420
546
4.71 ± 0.53 (6)
4.99 ± 0.26 (6)
3.50 ± 0.33 (6)
6.46 ± 0.87 (10)
4.07 ± 0.15 (6)
4.82 ± 0.40 (6)
1.17
0.97
2.71
0.35
1.83
1.09
4.67 ± 0.23 (6)
5.55 ± 0.33 (6)
4.22 ± 0.12 (5)
6.60 ± 1.09 (11)
4.61 ± 0.66 (6)
4.97 ± 0.18 (5)
1.21
0.66
1.65
0.32
1.26
0.98
4.94 ± 0.43 (6)
5.97 ± 0.65 (5)
4.71 ± 0.43 (5)
5.18 ± 0.86 (10)
4.55 ± 0.32 (5)
5.80 ± 0.61 (3)
1.00
0.49
1.17
0.85
1.31
0.55
5.54 ± 0.54 (6)
7.02 ± 1.13 (3)
4.55 ± 0.29 (4)
5.04 ± 0.88 (11)
4.49 (1)
5.93 (1)
0.66
0.24
1.31
0.93
1.37
0.50
Copy number
0
110
144
325
420
546
−6.73 ± 0.27 (6)
−7.11 ± 0.27 (6)
−6.03 ± 0.36 (6)
−8.92 ± 0.51 (10)
−6.95 ± 0.22 (6)
−9.28 ± 0.60 (6)
0.89
1.16
0.55
4.06
1.04
5.21
−6.78 ± 0.13 (6)
−7.02 ± 0.31 (6)
−6.08 ± 0.26 (5)
−8.02 ± 0.56 (11)
−6.51 ± 0.49 (6)
−9.18 ± 0.60 (5)
0.92
1.09
0.57
2.17
0.76
4.86
−6.90 ± 0.29 (6)
−7.00 ± 0.21 (5)
−5.95 ± 0.31 (5)
−8.02 ± 0.57 (10)
−5.92 ± 0.44 (5)
−8.93 ± 0.33 (3)
1.00
1.07
0.52
2.17
0.51
4.08
−6.83 ± 0.25 (6)
−7.87 ± 0.70 (3)
−6.96 ± 0.23 (4)
−7.96 ± 0.56 (11)
−5.62 (1)
−8.72 (1)
0.95
1.96
1.04
2.08
0.43
3.53
The rats were maintained on either a cereal-based diet or a purified diet of varying folate content. The number of PCR cycles required
to exceed a threshold (CT) just above background was calculated for test and reference reactions. CTvalues were determined in duplicate
and averaged for the mitochondrial deletion and D-loop and then subtracted to obtain the ?CTdeletion, expressed as the mean ± S.E.M.
Relative expression (Rel Exp) was calculated using the equation 2−??C
T
?CTof 0mg/kg 5-FU in the folate-replete purified diet. CTvalues were determined in duplicate and averaged for the mitochondrial D-loop
and ?-actin and subtracted to obtain the ?CT copy number, expressed as the mean ± S.E.M. Relative expression was calculated using
the equation 2−??C
T
where ??CTequals ?CTcopynumber−?CTcalwhere the calibrator is the ?CTof 0mg/kg 5-FU in the folate-replete
purified diet. Relative expression indicates the fold difference in deletions or copy number compared to the folate-replete purified diet.
an: number of animals.
where ??CTequals ?CTdel−?CTcalwhere the calibrator is the
effect on mtDNA. A significant dose by diet in-
teraction was found for both deletions and relative
abundance. The rats maintained on the folate-replete
purified diet had fewer deletions than rats fed with the
cereal-based diet, but there was no consistent effect
of nutritional folate status on deletions at the various
5-FU dosage levels. Relative mtDNA copy number
tended to be higher in the cereal-based and high folate
groups.
Currently there is a limited amount of information
available regarding the effect of diet on mtDNA dam-
age. Kang et al. showed that dietary restriction to 60%
of ad libitum intake in rats prevented the age-related
increase in mtDNA deletions in the liver but not the
brain [20]. More recently, Reddy et al. confirmed this
result using a 50% dietary restriction and a real-time
quantitative PCR, “TaqMan”, assay similar to the
one used in present study [21]. Dietary restriction is
believed to protect both nuclear and mtDNA from
oxidative damage [22]. Our studies suggest that a spe-
cificdietarycomponent,folicacid,alsomaybeprotec-
tive against accumulation of the “common deletion”.
The mechanism(s) by which folate might protect
against mtDNA damage is unclear at present. Mito-
chondria contain a substantial proportion of the
cell’s folate content [23–26]. Mitochondrial folate
metabolism is a primary source of one-carbon units
for cytoplasmic anabolic processes such as purine
and pyrimidine synthesis [24,27]. Therefore, one
could speculate that mitochondrial folate deficiency
may impair the synthesis and repair of both nuclear
and mitochondrial DNA. Another possible mecha-
nism is that folic acid deficiency, by causing elevated
homocysteine levels, decreases cellular antioxidant
potential (glutathione peroxidase, NKEF-? PAG, su-
peroxide dismutase, clusterin) and promotes oxidative
damage of mtDNA [28]. Regardless of mechanism,
these studies suggest that dietary supplementation

Page 7

R.F. Branda et al./Mutation Research 501 (2002) 29–36
35
with a non-toxic nutrient, folic acid, may decrease
the frequency of background mtDNA deletions and
of the mtDNA alterations associated with cancer
chemotherapy.
Acknowledgements
We thank Timothy Hunter for his assistance in
designing the primers and probes used in this
study. This work was supported by grants from the
Department of Defense (DAMD17-98-1-8345) and
the National Cancer Institute, Bethesda, MD (CA
41843). The content of the information does not
necessarily reflect the position or policy of the US
government, and no official endorsement should be
inferred. Automated sequencing was performed in the
Vermont Cancer Center DNA Analysis Facility and
was supported in part by NCI P30 CA 22435.
References
[1] A.C. Aisenberg, Problems in Hodgkin’s Disease management,
Blood 93 (1999) 761–779.
[2] N.J. Zeleznik-Le, G. Nucifora, J.D. Rowley, The molecular
biology of myeloproliferative disorders as revealed by
chromosomal abnormalities, Semin. Hematol. 32 (1995) 201–
219.
[3] R.F. Branda, A.R. Lafayette, J.P. O’Neill, J.A. Nicklas, Effect
of folate deficiency on mutations at the HPRT locus in Chinese
hamster ovary cells exposed to monofunctional alkylating
agents, Cancer Res. 57 (1997) 2586–2588.
[4] R.F. Branda, J.P. O’Neill, E.M. Brooks, L.M. Trombley, J.A.
Nicklas, The effect of folate deficiency on the cytotoxic
and mutagenic responses to ethyl methanesulfonate in human
lymphoblastoid cell lines that differ in p53 status, Mutat. Res.
473 (2001) 51–71.
[5] J.S. Penta, F.M. Johnson, J.T. Wachsman, W.C. Copeland,
Mitochondrial DNA in human malignancy, Mutat. Res. 488
(2001) 119–133.
[6] D. Neubert, W. Hopfenmuller, G. Fuchs, Manifestation of
carcinogenesis as a stochastic process on the basis of an
altered mitochondrial genome, Arch. Toxicol. 48 (1981) 89–
125.
[7] D. Decaudin, I. Marzo, C. Brenner, G. Kroemer, Mitochondria
in chemotherapy-induced apoptosis: a prospective novel target
of cancer therapy, Int. J. Oncol. 12 (1998) 141–152.
[8] L.G. Baggetto, Role of mitochondria in carcinogenesis, Eur.
J. Cancer 29A (1993) 156–159.
[9] E.J. Sherratt, A.W. Thomas, J.C. Alcolado, Mitochondrial
DNA defects: a widening clinical spectrum of disorders, Clin.
Sci. 92 (1997) 225–235.
[10] H. Yamamoto, M. Tanaka, M. Katayama, T. Obayashi,
Y. Nimura, T. Ozawa, Significant existence of deleted
mitochondrial DNA in a cirrhotic liver surrounding hepatic
tumor, Biochem. Biophys. Res. Commun. 182 (1992) 913–
920.
[11] M. Tanaka, W. Sato, K. Ohno, T. Yamamoto, T. Ozawa, Direct
sequencing of deleted mitochondrial DNA in myopathic
patients, Biochem. Biophys. Res. Commun. 164 (1989) 156–
163.
[12] W. Edris, B. Burgett, O.C. Stine, C.R. Filburn, Detection and
quantitation by competitive PCR of an age-associated increase
in a 4.8-kb deletion in rat mitochondrial DNA, Mutat. Res.
316 (1994) 69–78.
[13] M.N. Gadaleta, G. Rainaldi, A.M.S. Lezza, F. Milella, F.
Fracasso, P. Cantatore, Mitochondrial DNA copy number
and mitochondrial DNA deletion in adult and senescent rats,
Mutat. Res. 275 (1992) 181–193.
[14] V.Wunderlich,M.Schutt,
Preferential alkylation of mitochondrial deoxyribonucleic
acid by N-methyl-N-nitrosurea, Biochem. J. 118 (1970) 99–
109.
[15] R. Wilkinson, A. Hawks, A.E. Pegg, Methylation of rat liver
mitochondrial deoxyribonucleic acid by chemical carcinogens
and associated alterations in physical properties, Chem.-Biol.
Interact. 9 (1975) 157–167.
[16] V. Wunderlich, I. Tetzlaff, A. Graffi, Studies on nitroso-
dimethylamine: preferential methylation of mitochondrial
DNAinrats andhamsters,
(1971–1972) 81–89.
[17] D.J. Park, H.P. Koeffler, Therapy-related myelodysplastic
syndromes, Semin. Hematol. 33 (1996) 256–273.
[18] N. Gattermann, From sideroblastic anemia to the role of
mitochondrial DNA mutations in myelodysplastic syndromes,
Leuk. Res. 24 (1999) 141–151.
[19] J. Boultwood, C. Fidlers, K.I. Mills, P.M. Frodsham, R. Kusec,
A. Gaiger, R.E. Gale, D.C. Linch, T.J. Littlewood, P.A.H.
Moss, J.S. Wainscoat, Amplification of mitochondrial DNA
in acute myeloid leukemia, Br. J. Haematol. 95 (1996) 426–
431.
[20] C.-M. Kang, B.S. Kristal, B.-P. Yu, Age-related mitochondrial
DNA deletions: effect of dietary restriction, Free Rad. Biol.
Med. 24 (1998) 148–154.
[21] M.V. Reddy, H. Kurpad, D.R. Umbenhauer, T. Pippert, T.R.
Skopek, R. Dixit, K.P. Keenan, R.D. Store, Effect of dietary
restriction on age-related mitochondrial DNA deletions, Proc.
AACR 41 (2000) 235 (abstract).
[22] B.-P. Yu, Aging and oxidative stress: modulation by dietary
restriction, Free Rad. Biol. Med. 21 (1996) 651–668.
[23] C. Wagner, Symposium on the subcellular compartmen-
tation of folate metabolism, J. Nutr. 126 (1996) 1228S–
1234S.
[24] D.R. Appling, Compartmentation of folate-mediated one-
carbon metabolism in eukaryotes, FASEB J. 5 (1991) 2645–
2651.
[25] T.M. Trippett, S. Garcia, K. Manova, R. Mody, L. Cohen-
Gould, W. Flintoff, J.R. Bertino, Localization of a human
reduced folate carrier protein in the mitochondrial as well as
M. Bottger,A. Graffi,
Chem.-Biol.Interact.4

Page 8

36
R.F. Branda et al./Mutation Research 501 (2002) 29–36
the cell membrane of leukemia cells, Cancer Res. 61 (2001)
1941–1947.
[26] D.W.Horne,R.S.Holloway,
5-formyltetrahydrofolate in isolated rat liver mitochondria is
carrier-mediated, J. Nutr. 122 (1992) 2204–2209.
[27] S. Girgis, J.R. Suh, J. Jolivet, P.J. Stover, 5-Formy-
ltetrahydrofolate regulates homocysteine remethylation in
H.M.Said, Uptake of
human neuroblastoma, J. Biol. Chem. 272 (1997) 4729–
4734.
[28] P.A. Outinen, S.K. Sood, S.I. Pfeifer, S. Pamidi, T.J.
Podor, J. Li, J.I. Weitz, R.C. Austin, Homocysteine-induced
endoplasmic reticulum stress and growth arrest leads to
specific changes in gene expression in human vascular
endothelial cells, Blood 94 (1999) 959–967.