Carcinogenesis vol.29 no.4 pp.722–728, 2008
doi:10.1093/carcin/bgn033
Advance Access publication February 6, 2008
Cells deficient in oxidative DNA damage repair genes Myh and Ogg1 are sensitive to
oxidants with increased G2/M arrest and multinucleation
Yali Xie1,2,, Hanjing Yang3, Jeffrey H.Miller3, Diana
M.Shih4, Geoffrey G.Hicks1,2, Jiuyong Xie1 and Robert
P.Shiu1,2,
1
Department of Physiology and 2Manitoba Institute of Cell Biology,
University of Manitoba, 730 William Avenue, Winnipeg, MB R3E 0W9,
Canada, 3Department of Microbiology, Immunology, and Molecular Genetics
and the Molecular Biology Institute and 4Division of Cardiology,
Department of Medicine, University of California at Los Angeles, Los
Angeles, CA 90095, USA
To whom correspondence should be addressed. Y. Xie, Tel: þ1 204 480 1328;
Fax: þ1 204 789 3934; Email: [email protected]; or R. Shiu,
Tel: þ1 204 789 3327; Fax: þ1 204 789 3934;
Email: [email protected]
Oxidative stress generated from endogenous and exogenous sources causes oxidative DNA damage. The most frequent mutagenic
base lesion 7,8-dihydro-8-oxoguanine and the resulting mismatched adenine are removed by OGG1 and MYH in mammals.
Deficiencies in human MYH or mouse MYH and OGG1 result in
tumor predisposition but the underlying molecular mechanism is
not fully understood. To facilitate the study of the roles of MYH
and OGG1 in the protection against oxidative stress, we generated
mouse embryonic fibroblast cell lines deficient in these genes. Myh
and Ogg1 double knockout cells were more sensitive than wild
type to oxidants (hydrogen peroxide and t-butyl hydroperoxide),
but not to cis-platinum or g-irradiations. The low dosage oxidative stress resulted in more reduction of S phase and increase of
G2/M phase in Myh2/2Ogg12/2 cells than in wild-type cells, but
a similar level of cell death in both cells. The oxidants also induced
more multinucleated cells in Myh2/2Ogg12/2 cells than in wildtype, accompanied by centrosome amplification and multipolar
spindle formation. Thus, under oxidative stress, Myh and Ogg1
are likely required for normal cell-cycle progression and nuclear
division, suggesting multiple roles of Myh and Ogg1 in the maintenance of genome stability and tumor prevention.
Introduction
Oxidative stress generated from oxidant by-products of normal cellular metabolism or exogenous sources such as smoking plays a key role
in tumorigenesis, causing oxidative DNA damages (1–3). The most
frequent mutagenic base lesion, 7,8-dihydro-8-oxoguanine, can result
in adenine mismatches during DNA replication (4,5). OGG1 and
MYH, two major enzymes involved in mammalian oxidative DNA
damage repair, remove 7,8-dihydro-8-oxoguanine and mismatched
adenine, respectively, preventing G to T mutations (6–9). Inherited
defects in human MYH are associated with G to T mutations in a tumor
suppressor gene APC in colorectal tumors (10). Defective MYH has
been observed in 7.5–22.5% cases of inherited colorectal adenomatous polyposis, which eventually develops into cancer, in patients
without inherited APC mutations (11,12). Knockout of either Myh
or Ogg1 in mice does not cause tumors; however, knockout of both
genes predispose mice to various tumors including lymphoma, lung
and ovarian tumors, suggesting synergistic roles of Myh and Ogg1 in
the tumor prevention (13–15). Codon 12 in K-ras oncogene appears to
be a downstream target of oxidative DNA damage in lung tumorigenesis in these mice (13). It is unknown whether deficiencies in Myh
and Ogg1 also cause other changes in addition to the tumor suppressor gene APC and the K-ras oncogene, especially under oxidative
Abbreviations: H2O2, hydrogen peroxide; MEF, mouse embryo fibroblast;
TBH, t-butyl hydroperoxide.
stress. Previously, there were no cell models deficient in Myh or
Myh and Ogg1 available to conveniently address this question. Only
Ogg1/ cell line exists (14). Therefore, we have developed Myh/
and Myh/Ogg1/ embryonic fibroblast cell lines from corresponding knockout mice and examined their phenotypes associated with
oxidative stress.
Materials and methods
Generation of wild-type, Myh/, Ogg1/ and Myh/Ogg1/ cell lines
Heterozygotes for Myh (13) and Ogg1 (14) mice in a mixed C57BL/6J and 129
background were backcrossed with C57BL/6J to N4 generation. Littermates of
Myhþ/ or Ogg1þ/ mice were crossed to generate wild-type, Myh/ and
Ogg1/ embryo fibroblasts. The probability (1/16) was too low to obtain
wild-type and Myh/Ogg1/ embryos in littermates by crossing Myhþ/
Ogg1þ/ mice because the average litter size of C57BL/6J is only seven
(16). Therefore, offspring of the littermates of the mice that generated wildtype, Myh/ and Ogg1/ embryos were crossed to generate Myh/Ogg1/
embryos. Embryo fibroblasts were obtained from 13.5-day-old embryo using
a standard procedure and genotyped by polymerase chain reaction as described
previously (13,17). The primary fibroblasts were repeatedly passaged to establish cell lines as described before (17). When a confluent culture could be
continually passaged at a 1:3 ratio more than four times after crisis, it was
considered immortal and frozen in liquid nitrogen. All cells were cultured in
Dulbecco’s modified Eagle’s medium supplemented with 10% fetal bovine
serum, 2 mM glutamine, 100 lM non-essential amino acids and 100 U/ml
penicillin–streptomycin (Invitrogen, Burlington, Ontario) at 37°C in a humidified atmosphere with 5% CO2.
Assays of sensitivity to DNA-damaging agents
All experiments in this study were carried out within 10 passages after establishment of cell lines. Also, prior to treatment, all the cells were allowed 16–18 h
attachment following plating and cells were subconfluent at the end of the
experiments. Confluent cultures were trypsinized and cells were plated at
a density of 8 103 cells per well in 24-well dishes. Cells were exposed to
various concentrations of hydrogen peroxide (H2O2), t-butyl hydroperoxide
(TBH) and cis-platinum for 72 h. For c-irradiation, 4 104 cells were seeded
in 60 mm dishes and irradiated using a cesium137 source at a rate of 2.4 Gy/min.
Control cells were mock treated. The number of surviving cells was determined
72 h later using crystal violet staining as described previously (18).
Trypan blue exclusion, apoptosis and colony formation assays
For trypan blue exclusion assay, both suspended and attached cells were collected after exposures to H2O2 for 24 or 72 h and stained with 0.2% trypan
blue. Dead and live cells were counted under a microscope. To investigate if
apoptosis occurs in these cells after treatment, Annexin V–FITC Apoptosis
Detection Kit (BioVision, Mountain View, California) was used according to
the manufacturer’s instruction. At least 10 000 cells in each condition were
analyzed with Fluorescence activated cell sorting (FACS) using a FACScan
flow cytometer (Becton-Dickinson, Mississauga, Ontario). For the colony formation assay, 2 102 cells were plated in 60 mm dishes for the controls. To
yield 20–70 colonies per dish as control plates, 4 102 cells were plated and
treated with H2O2. Media with or without H2O2 were changed twice a week.
After 12–14 days, colonies were stained with crystal violet and only colonies
with .50 cells were counted. Three independent experiments were performed.
Cell-cycle analysis
After exposure to H2O2 or TBH for 24 or 72 h, cells were incubated with 10
lM 5-Bromo-2#-deoxyuridine (BrdU, Sigma, Oakville, Ontario) for 1 h. Suspended and attached cells were collected and then fixed in 70% cold ethanol.
Incorporated BrdU was identified by FITC-conjugated anti-BrdU antibodies
(BD Biosciences, Mississauga, Ontario) and DNA was labeled with 7-aminoactinomycin D (7AAD). Cells were then analyzed with a FACScan flow cytometer (Becton-Dickinson) using CellQuest software (Becton-Dickinson). At
least 10 000 cells in each condition were analyzed. Cells with DNA content ,2N were not included.
Immunofluorescent microscopy
After exposures to H2O2 and TBH for 72 h, cells were fixed with 10% formalin
and cold methanol and permeabilized with 0.2% Triton X-100. Fixed cells
Ó The Author 2008. Published by Oxford University Press. All rights reserved. For Permissions, please email: [email protected]
722
Characterization of Myh/Ogg/ cell line
were incubated with mouse monoclonal antibodies for c-tubulin (Sigma) at
room temperature for 1 h. FITC-conjugated anti-mouse IgG secondary antibodies (FI-2000, Vector Laboratories, Burlington, Ontario) were used at a dilution of 1:1000 at room temperature for 1 h. Cy3-conjugated anti-b-tubulin
antibodies were then used at a dilution of 1:1000 for microtubule detection. In
some experiments, phalloidin–Tetramethylrhodamine isothiocyanate (TRITC,
1:2000) was used to stain actin for visualizing the whole cell. Cells were
mounted and then examined under a fluorescent microscope with the appropriate filters. Multiple nuclei and centrosomes in single cells were confirmed
by actin or b-tubulin staining of the cells (Figure 5).
Statistical analysis
t-test was used to assess statistical significance.
Results
Fig. 1. Genotype analysis of cell lines. Genotypes of wild-type (lanes 3 and 4),
Myh/Ogg1/ (lanes 5 and 6), Myh/ (lanes 7 and 8) and Ogg1/ (lanes 9
and 10) cell lines are as indicated. Control DNA is from a Myhþ/Ogg1þ/
mouse (lanes 1 and 2). Lanes 1, 3, 5, 7 and 9 show Myh genotyping.
Polymerase chain reaction amplification of the wild-type Myh allele (labeled
as Myhþ) and the mutant Myh allele (Myh) produces 0.26 and 0.38 kb DNA
fragments, respectively. Lanes 2, 4, 6, 8 and 10 show Ogg1 genotyping.
Polymerase chain reaction amplification of the wild-type Ogg1 allele
(labeled as Ogg1þ) and the mutant Ogg1 allele (Ogg1) produces 0.14 and
1.1 kb DNA fragments, respectively. Lane M displays DNA marker and the
sizes of each fragment are indicated on the left.
Generation of embryonic fibroblast lines
To facilitate the study of mechanisms in oxidative DNA damage and
the roles of Myh and Ogg1, we generated wild-type, Myh/, Ogg1/
and Myh/Ogg1/ mouse embryo fibroblasts (MEFs) by crossing
littermates of the respective gene knockout mice. We developed these
primary MEFs to cell lines by spontaneous immortalization (17). The
MEFs, which could be continually passaged at 1:3 ratio more than
four times after crisis, were considered immortal and frozen for further study. The wild-type cell lines were established in 3 months,
whereas the three knockout cell lines took 5 months. The genotypes
of these cell lines were confirmed by polymerase chain reaction as
shown in Figure 1. After establishment, the doubling time for the
Myh/Ogg1/ cell line was 22.2 h under the normal growth condition, which was similar to that of the wild-type, 22 h. There was no
apparent difference in the morphology of the two cell lines. The
establishment of these cell lines provides possibilities for the experiments that could not be done previously due to the limited life span of
primary MEFs (approximately five passages).
Sensitivity to DNA-damaging agents
To determine whether cells deficient in Myh and Ogg1 display an
altered sensitivity to DNA-damaging agents, the cells were exposed
to H2O2, TBH, cis-platinum and c-irradiation. To mimic
Fig. 2. Sensitivity of wild-type and Myh/Ogg1/ cells to DNA-damaging agents. (A) H2O2; (B) TBH; (C) cis-platinum and (D) c-irradiation. In (A) and (B),
the values represent average percentages of surviving treated/untreated cells relative to wild-type cells after exposures to oxidants for 72 h. In (C) and (D), the
values represent average percentages of surviving treated/untreated cells determined after 72 h exposures in (C) and 72 h after exposures in (D), respectively. The
average values were based on three to six independent experiments, each of which was conducted in duplicates or triplicates.
723
Y.Xie et al.
Fig. 3. Cell death, apoptosis and colony formation of wild-type and Myh/Ogg1/ cells after exposures to H2O2. (A) The number of dead wild-type and
Myh/Ogg1/ cells after 24 or 72 h exposures to H2O2. Both suspended and attached cells were collected and examined by trypan blue exclusion assay.
Values represent average ± standard deviation (n 5 4). (B) Quantification of apoptotic cells, after 24 or 72 h exposures to 65 lM H2O2, with Annexin V/7AAD–
FACS analysis. The value represents the average percentage ± standard deviation (n 5 3). The dead cells are 7AA D positive [upper left (UL) þ upper right (UR)]
and the apoptotic cells are Annexin V positive [upper right (UR) þ lower right (LR)]. (C) and (D) Colony formation assay of untreated and treated wild-type and
Myh/Ogg1/ cells. The percentages in (D) represent the averages of three independent experiments.
physiological conditions, prolonged low dosages of oxidants were
used. After 72 h exposures to various concentration of H2O2, percentages of surviving treated versus untreated cells deficient in Myh
or Ogg1 did not differ significantly from that of wild-type, but
Myh/Ogg1/ cells showed significantly lower survival than the
wild-type (P , 0.01, Figure 2A). To exclude the effects of variable
catalase activity in these cell lines, another oxidant, TBH, which is
poorly hydrolyzed by catalase (19), was also tested. The survival rates
of treated versus untreated Myh/Ogg1/ cells were also significantly lower than that of the wild-type (P , 0.03, Figure 2B). To test
whether Myh/Ogg1/ cells were also sensitive to other DNAdamaging agents, cells were exposed to cis-platinum or c-irradiation
whose major damages are repaired by nucleotide excision repair
and double strand break repair, respectively (20,21). There was
no significant difference in survival between Myh/Ogg1/ and
wild-type cells exposed to these agents (Figure 2C and D). Thus,
Myh/Ogg1/ cells appear to be specifically sensitive to oxidants,
H2O2 and TBH.
724
Cell death, apoptosis and cell-cycle progression in wild-type and
Myh/Ogg1/ cells under oxidative stress
Sensitivity of Myh/Ogg1/ cells to oxidants can result from either
an increase in cell death due to necrosis or apoptosis or a decrease in
growth. To investigate these possibilities, we first monitored cell death
that occurred under prolonged treatment of low dosage H2O2 using the
trypan blue exclusion method. The number of dead Myh/Ogg1/
cells was not significantly different from that of wild-type cells after
exposures to H2O2 for 24 or 72 h. The number of dead cells in both
cell lines did not increase as the H2O2 dosage increased under the
experimental conditions (up to 65 lM H2O2, Figure 3A). We then
examined these cells with Annexin V–FACS analysis that reveals
apoptotic and dead cells (Figure 3B). In this assay, after 24 h exposures to 65 lM H2O2, both dead and apoptotic cells did not increase
significantly as compared with untreated cells in both wild-type and
Myh/Ogg1/ cell lines. After 72 h exposures to 65 lM H2O2, the
number of dead cells increased as compared with 24 h exposures in
both wild-type and Myh/Ogg1/ cell lines, but there was no
Characterization of Myh/Ogg/ cell line
significant difference between the two cell lines. These findings are
consistent with the results in trypan blue exclusion assay (Figure 3A).
Apoptotic cells also increased after 72 h exposures as compared with
24 h in both cell lines, but there was no significant difference between
the two cell lines (Figure 3B). We further carried out colony formation
assay. Myh/Ogg1/ cells formed fewer colonies than wild-type
under treatments with 20 lM H2O2 in 14 days (Figure 3C and D).
This suggested that Myh/Ogg1/ cells might proliferate slower
than wild-type since they had no difference in cell death and apoptosis
under H2O2 treatments.
To investigate whether deficiencies in Myh and Ogg1 alter the cellcycle profiles under oxidative stress, wild-type and Myh/Ogg1/cells were exposed to oxidants for 24 or 72 h following 1 h BrdU
labeling and subjected to FACS analysis. The percentage of S phase
cells was reduced and that of G2/M phase cells was increased in both
Myh/Ogg1/ and wild-type cells under oxidative stress (Table I
and Figure 4). However, fewer S phase and more G2/M phase cells
were observed in Myh/Ogg1/ than in wild-type cells after exposures to H2O2 and TBH. After 72 h exposures to 65 lM H2O2, reduction of S phase and increase of G2/M phase in Myh/Ogg1/
cells were significantly different from wild-type (P , 0.01). Also, the
percentage of cells containing DNA .4N was significantly increased
in Myh/Ogg1/ cells (27.2 ± 1.7%) than in wild-type (12.2 ±
1.1%) after 72 h exposures to 65 lM H2O2 (P , 0.01, Figure 4)
but was not different in untreated cell lines (6.1 ± 0.8% and 5.7 ±
0.3%, respectively). There was also no significant difference in the
reduction of G0/G1 phase cells between wild-type and Myh/Ogg1/
cells in the presence of H2O2 and TBH (P 5 0.1).
Multinucleation in wild-type and Myh/Ogg1/ cells induced by
oxidants
Exposures to oxidants for 72 h also induced multinucleated cells in
both wild-type and Myh/Ogg1/ backgrounds (Figure 5). The
percentages of multinucleated cells increased as H2O2 dosage
Table I. Flow cytometric analysis of Myh and Ogg1 proficient and deficient cells after exposures to TBH and H2O2
Relative percentagea of S phase
Treatment
None
TBH
10 lM 24
20 lM 24
H2O2
65 lM 72
65 lM 24
Relative percentagea of G2/M phase
Wild-type
Myh/Ogg1/
100 (±3.0)b
100 (±2.1)
100 (±2.2)
100 (±1.3)
h
h
98.3 (±5.3)
82.6 (±9.7)
80.0 (±5.2)
36.7 (±7.1)c
115 (±6.0)
179.9 (±30.6)
118.9 (±16.1)
230.3 (±26.5)c
h
h
72.3 (±2.3)
73.0 (±2.4)
39.0 (±20.3)
16.0 (±1.7)c
216.4 (±2.5)
174.2 (±4.2)
266.9 (±65.1)
357.5 (±20.1)c
Wild-type
Myh/Ogg1/
a
Relative percentage, percentage of treated cells in S or G2/M phase relative to that of untreated cells. The actual average values of untreated cells in S and G2 phase
were 54.2 and 17.3% in wild-type and 39.3 and 26.4% in Myh/Ogg1/ cells.
b
±, standard deviation (n 5 3). Cells were exposed to TBH or H2O2 for 24 or 72 h, following 1 h BrdU labeling before FACS analysis. Percentages of S and G2
phase cells were calculated with dot plots using CellQuest software (Figure 4 and Materials and Methods).
c
Significantly different from wild-type.
Fig. 4. Flow cytometric analysis of wild-type and Myh/Ogg1/ cells in response to H2O2 exposures. Cells were exposed to H2O2 for 24 or 72 h following 1 h
incubation with BrdU, which was identified by anti-BrdU–FITC antibodies. Dot plots show the incorporation of BrdU into DNA as an indication of DNA synthesis
and 7AAD fluorescence as an indication of DNA content. At least 10 000 cells per condition were analyzed. The outlined regions represent areas from which data
were taken for analysis (see Table I).
725
Y.Xie et al.
increased in both cell lines; however, there was a higher percentage of
multinucleated cells in Myh/Ogg1/ cells (26 ± 2.2%) than in
wild-type (14 ± 4%, Figure 5B). This is consistent with what we have
seen in FACS analysis (Figure 4); that there are higher percentages of
cells with DNA content .4N in Myh/Ogg1/ background than in
wild-type after 72 h exposures to H2O2. These observations suggest
a role of Myh and Ogg1 in preventing the formation of multiple nuclei
under oxidative stress.
In investigating the possible cause of multinucleation in these cells,
we observed some multinucleated cells in telophase (Figure 5C),
suggesting that multinucleated cells can still undergo cytokinesis.
We also examined centrosomes, an organelle that functions as a microtubule-organizing center, controlling the polarity of microtubules
and generating mitotic spindles that pull chromosomes apart to form
two nuclei in mitosis (22,23). Some evidence suggests centrosome as
a part of the network that integrates repair signals in response to some
genotoxic stress (23). To investigate if centrosomes were involved
in these cells under oxidative stress, centrosomes were visualized
by an immunofluorescent staining with antibodies against c-tubulin,
a component of centrosomes. About 94% of untreated wild-type and
Myh/Ogg1/ cells contained one to two centrosomes per cell
(Figure 6A and B). After exposures to oxidants for 72 h, the centrosome number increased significantly in both cell lines and was not
correlated with the number of nuclei (Figure 6A). There was a higher
percentage of cells with multiple centrosomes in Myh/Ogg1/
(54.8 ± 5.1% in H2O2 and 36.5 ± 9.3% in TBH) than in wild-type
background (40.5 ± 5.8% in H2O2 and 20.2 ± 10.1% in TBH, both
P , 0.01), suggesting a role of Myh and Ogg1 in preventing multiple
centrosome formation under oxidative stress (Figure 6B). We found
that some cells with multiple centrosomes had only one or two nuclei
(Figure 6A). Also, the percentages of cells with multiple centrosomes
in both Myh/Ogg1/ and wild-type backgrounds were higher than
the percentage of cells with multiple nuclei under the same oxidative
stress condition (Figures 5B and 6B and data not shown). These observations suggest that the formation of multiple centrosomes may be
an earlier event than the formation of multiple nuclei under oxidative
stress.
To investigate whether the multiple centrosomes were functional,
which led to abnormal mitotic spindle organization and chromosome
segregation, a double immunostaining with c- and b-tubulin antibodies was performed. b-Tubulin is a major component of mitotic spindles. In untreated cells, two centrosomes were colocalized with bipolar
spindle poles and associated with aligned chromosomes (Figure 6C).
In treated cells, multipolar spindles radiated from multiple centrosomes, which were colocalized with the poles and associated with
misaligned chromosomes (Figure 6C). The multiple centrosomes
Fig. 5. Multinucleated cells in wild-type and Myh/Ogg1/ cells under oxidative stress. (A) Whole cells were visualized with fluorescent staining with
phalloidin–TRITC for actin. Nuclei were stained with 4#,6-diamidino-2-phenylindole (DAPI). (B) Percentages of multinucleated cells in wild-type and Myh/Ogg1/
cells after exposures to H2O2 for 72 h. Values were the average of four independent experiments. For each experiment, 250–2000 cells per condition were counted.
(C) Multinucleated Myh/Ogg1/ cells undergoing cytokinesis after an exposure to H2O2. Cells were stained with DAPI and Cy3-conjugated anti b-tubulin antibodies.
726
Characterization of Myh/Ogg/ cell line
Fig. 6. Centrosome amplification in wild-type and Myh/Ogg1/ cells under oxidative stress. (A) Multiple centrosomes in Myh/Ogg1/ cells after
exposures to 65 lM H2O2 or 15 lM TBH for 72 h. It was visualized by immunofluorescent staining using antibodies against c-tubulin. Untreated Myh/Ogg1/
cells displayed one to two centrosomes and treated cells showed multiple centrosomes and nuclei in single cells. (B) Percentages of cells with multiple
centrosomes in wild-type and Myh/Ogg1/ cells after exposures to H2O2 and TBH for 72 h. Values were the average of four independent experiments. For each
experiment, 100–300 cells per condition were counted. (C) Colocalization of multiple centrosomes and multipolar spindle poles. Cells were double stained with
antibodies against c- and b-tubulin to visualize the association of centrosomes and mitotic spindles. Chromosomes were stained with DAPI to visualize their
alignment and association with centrosomes and spindles in metaphase.
were functional as microtubule-organizing centers and generated multipolar spindles that help chromosomal segregation in multiple directions, a process known to increase the risk of formation of multiple
nuclei and chromosomal aneuploidy (22,23).
Discussion
Oxidative stress is unavoidable and continuously causes oxidative
DNA damage (1). Although the importance of the damage repair
genes, Myh and Ogg1, has been highlighted in tumor predisposition
in humans defective in MYH and mice deficient in Myh and Ogg1
(5–8,10–13), there were no Myh/ and Myh/Ogg1/ cell lines
available to conveniently study the functions and mechanisms of these
genes at a cellular level. In this study, we established wild-type, Myh/,
Ogg1/ and Myh/Ogg1/ MEF cell lines derived from knockout
mice. Deficiencies in both Myh and Ogg1 increased the sensitivity of
cells to oxidants with G2/M phase accumulation, accompanied by
increased centrosome amplification and formation of multiple nuclei.
727
Y.Xie et al.
Therefore, Myh and Ogg1 are likely required for normal cell-cycle
progression and cell division under oxidative stress, suggesting multiple roles of Myh and Ogg1, in addition to mutation avoidance in
APC and K-ras genes reported previously (10–13).
Although deficiency in Myh or Ogg1 alone had no effects on cells’
sensitivity to H2O2, cells deficient in both genes did have significantly
increased sensitivity. The synergistic effect of these two genes in
response to oxidants was similar to that in tumor prevention in mice
and mutation avoidance in Escherichia coli (5,13). Low dosage
oxidants induced more G2/M accumulation and less S phase in
Myh/Ogg1/ cells than in wild-type. It was likely that under
oxidative stress, cells deficient in Myh and Ogg1 possessed higher
7,8-dihydro-8-oxoguanine content and adenine mismatches than
wild-type cells, thus affecting cell-cycle progression. G2/M accumulation induced by H2O2 has been seen in other cell lines (24,25), but
the involvement of Myh and Ogg1 in this process is revealed for the
first time in the present study. The increased percentage of multinucleated cells in Myh/Ogg1/ background may also contribute to
the increased sensitivity to oxidants due to inappropriate nuclear division and therefore less surviving cells than in wild-type.
Oxidative stress contributes to senescence and formation of tumors
where multinucleated cells have been found (1,22,26–29). Centrosome amplification is frequently seen in cancer cells, but it has not
been related to senescence. In this study, we have shown that prolonged treatment with low dosage oxidants induced the formation of
multiple nuclei accompanied by centrosome amplification and multipolar spindle formation. The centrosome duplication cycle appeared
to progress faster than the cell cycle under oxidative stress (Figure 6A).
It appeared that centrosomes had already been duplicated even before
cytoplasmic division commenced. Amplified centrosomes could generate multipolar mitotic spindles, pulling chromosomes in multiple
directions which could be responsible for the formation of multinucleated cells (Figure 6C) (22,23,30). Although centrosome amplification caused by H2O2 exposures was observed in other types of cells in
a recent study, the mechanism remains unclear (31). In this study, we
have shown that deficiency in both Myh and Ogg1 contributes to
centrosome amplification and multinuclear formation, suggesting
a role of the two genes in the control of the centrosome cycle under
oxidative stress. Additional studies are necessary in order to uncover
the mechanism. Centrosome amplification has been implicated in
causing chromosomal instability in tumorigenesis (30). Therefore,
deficiency in Myh and Ogg1 may also contribute to genome instability
in tumorigenesis via centrosome amplification. Our findings suggest
multiple roles of Myh and Ogg1 in the maintenance of genome stability and in tumor prevention. The cell lines we have developed can
be useful tools for studying the mechanism of oxidative DNA damage
at the cellular and molecular levels.
Funding
USA National Cancer Institute (CA 85952) to J.H.M.; The CancerCare Manitoba Foundation to G.G.H.; National Cancer Institute of
Canada (#016355) to J.X.; The Thorlakson Foundation to J.X. and
R.S.
Acknowledgements
We thank Dr Aldons J.Lusis (University of California at Los Angeles) and
Dr Etienne Leygue (University of Manitoba) for allowing us to use their
equipment. We are grateful to Drs Ludger Klewes, Charlton Cooper, Shihua
He and Spencer Gibson for technical assistance and helpful discussions. We
thank Irene Y. Xie for editing the manuscript.
Conflict of Interest Statement: None declared.
References
1. Ames,B.N. et al. (1993) Oxidants, antioxidants, and the degenerative diseases of aging. Proc. Natl Acad. Sci. USA, 90, 7915–7922.
728
2. Asami,S. et al. (1996) Increase of a type of oxidative DNA damage, 8hydroxyguanine, and its repair activity in human leukocytes by cigarette
smoking. Cancer Res., 56, 2546–2549.
3. Asami,S. et al. (1997) Cigarette smoking induces an increase in oxidative
DNA damage, 8-hydroxydeoxyguanosine, in a central site of the human
lung. Carcinogenesis, 18, 1763–1766.
4. Nishimura,S. (2001) Mammalian Ogg1/Mmh gene plays a major role in
repair of the 8-hydroxyguanine lesion in DNA. Prog. Nucleic Acid Res.
Mol. Biol., 68, 107–123.
5. Michaels,M.L. et al. (1992) Evidence that MutY and MutM combine to
prevent mutations by an oxidatively damaged form of guanine in DNA.
Proc. Natl Acad. Sci. USA, 89, 7022–7025.
6. Boiteux,S. et al. (2000) The human OGG1 gene: structure, functions, and
its implication in the process of carcinogenesis. Arch. Biochem. Biophys.,
377, 1–8.
7. Slupska,M.M. et al. (1999) Functional expression of hMYH, a human homolog of the Escherichia coli MutY protein. J. Bacteriol., 181, 6210–6213.
8. Yang,H. et al. (2001) Enhanced activity of adenine-DNA glycosylase
(Myh) by apurinic/apyrimidinic endonuclease (Ape1) in mammalian base
excision repair of an A/GO mismatch. Nucleic Acids Res., 29, 743–752.
9. Isogawa,A. (2004) Functional cooperation of Ogg1 and Mutyh in preventing G: c–.T: a transversions in mice. Fukuoka Igaku Zasshi, 95, 17–30.
10. Al-Tassan,N. et al. (2002) Inherited variants of MYH associated with somatic G:C–.T: a mutations in colorectal tumors. Nat. Genet., 30, 227–232.
11. Sampson,J.R. et al. (2003) Autosomal recessive colorectal adenomatous
polyposis due to inherited mutations of MYH. Lancet, 362, 39–41.
12. Sieber,O.M. et al. (2003) Multiple colorectal adenomas, classic adenomatous
polyposis, and germ-line mutations in MYH. N. Engl. J. Med., 348, 791–799.
13. Xie,Y. et al. (2004) Deficiencies in mouse Myh and Ogg1 result in tumor
predisposition and G to T mutations in codon 12 of the K-ras oncogene in
lung tumors. Cancer Res., 64, 3096–3102.
14. Klungland,A. et al. (1999) Accumulation of premutagenic DNA lesions in
mice defective in removal of oxidative base damage. Proc. Natl Acad. Sci.
USA, 96, 13300–13305.
15. Minowa,O. et al. (2000) Mmh/Ogg1 gene inactivation results in accumulation of 8-hydroxyguanine in mice. Proc. Natl Acad. Sci. USA, 97, 4156–4161.
16. Silver,L.M. (1995) Mouse Genetics-Concepts and Applications. Oxford
University Press, New York.
17. Todaro,G.J. et al. (1963) Quantitative studies of the growth of mouse embryo cells in culture and their development into established lines. J. Cell
Biol., 17, 299–313.
18. Carnero,A. et al. (2000) p16INK4A and p19ARF act in overlapping pathways in cellular immortalization. Nat. Cell Biol., 2, 148–155.
19. Winston,G.W. et al. (1983) The generation of hydroxyl and alkoxyl radicals
from the interaction of ferrous bipyridyl with peroxides. Biochem. J., 216,
415–421.
20. Hoeijmakers,J.H. (2001) Genome maintenance mechanisms for preventing
cancer. Nature, 411, 366–374.
21. Husain,I. et al. (1985) Repair of cis-platinum-DNA adducts by ABC excinuclease in vivo and in vitro. J. Bacteriol., 163, 817–823.
22. Fukasawa,K. (2005) Centrosome amplification, chromosome instability
and cancer development. Cancer Lett., 230, 6–19.
23. Loffler,H. et al. (2006) Structure meets function–centrosomes, genome maintenance and the DNA damage response. Exp. Cell Res., 312, 2633–2640.
24. Chang,D.K. et al. (2003) Effect of H(2)O(2) on cell cycle and survival in
DNA mismatch repair-deficient and -proficient cell lines. Cancer Lett., 195,
243–251.
25. Hardman,R.A. et al. (2001) Involvement of mammalian MLH1 in the apoptotic response to peroxide-induced oxidative stress. Cancer Res., 61,
1392–1397.
26. Ames,B.N. (1989) Mutagenesis and carcinogenesis: endogenous and exogenous factors. Environ. Mol. Mutagen., 14 (suppl. 16), 66–77.
27. Folpe,A.L. et al. (2005) Perivascular epithelioid cell neoplasms of soft
tissue and gynecologic origin: a clinicopathologic study of 26 cases and
review of the literature. Am. J. Surg. Pathol., 29, 1558–1575.
28. Evans,H.L. (2002) Multinucleated giant cells in plantar fibromatosis. Am. J.
Surg. Pathol., 26, 244–248.
29. Ren,R. et al. (2005) Fine-needle aspiration cytology of a liver metastasis of
follicular dendritic cell sarcoma. Diagn. Cytopathol., 32, 38–43.
30. D’Assoro,A.B. et al. (2002) Centrosome amplification and the development of cancer. Oncogene, 21, 6146–6153.
31. Chae,S. et al. (2005) Centrosome amplification and multinuclear phenotypes are induced by hydrogen peroxide. Exp. Mol. Med., 37, 482–487.
Received September 26, 2007; revised January 17, 2008;
accepted January 26, 2008