[CANCER RESEARCH 58. 3712-3718,
Augusl 15. 1998]
Apoptosis, Reproductive Failure, and Oxidative Stress in Chinese Hamster Ovary
Cells with Compromised Genomic Integrity1
Charles L. Limoli,2 Andreas Hartmann, Lee Shephard, Chin-rang Yang, David A. Boothman, Jim Bartholomew,
and
William F. Morgan
Departments of Radiation Oncology and Radiology, University of California, San Francisco, California 94143-0750 [C. L. L, A. H„L. S., W. F. M.}: Life Sciences Division
1C. L L, A. H., L.S.: W.F.M./
and Structural Biology Division [J.B.I, Lawrence Berkeley National Laboratory, Berkeley, California 94720: and Department of Human
Oncology. University of Wisconsin Medical School and Comprehensive Cancer Center, Madison, Wisconsin 53792 ¡C-r.Y., D. A. B.j
and indirectly induced DNA lesions, including DNA base alterations,
DNA-DNA and DNA-protein cross-links, and single- and doublestrand breaks (4). Cells react rapidly to irradiation with a plethora of
biological responses, including cell cycle-specific growth arrest (re
viewed in Ref. 5), gene transcription (6), initiation of signal transduction pathways (7), and repair of damaged DNA (8). These early events
are likely to determine the later fate of the irradiated cells, i.e.,
whether a cell will necrose, senesce, apoptose, or, ultimately, survive
and proliferate. If a cell does survive, the initial biological response to
the radiation-induced injury may influence whether the cell differen
tiates normally, has a limited life span, or proliferates and begins to
acquire, via genomic instability, those characteristics associated with
the transformation of a normal cell to a cancerous one. Effects of
X-ray exposure in the progeny of surviving cells include persistent
reproductive cell death or lethal mutations (which are revealed by
reduced plating efficiency), giant cell formation, delayed mutation,
clonal heterogeneity, transformation, and persistent chromosomal in
stability [reviewed by Morgan et al. (3)]. The persistence of these
effects reinforces the notion that exposure to radiation has long-lasting
consequences that cannot be explained by the direct effects of irradi
ation. Rather, the emerging concept is that several tightly regulated
cellular processes are disrupted and normal baseline responses are
reset. The result seems to be a chaotic state from which cells can never
completely recover.
One pathway that is believed to prevent cells from acquiring these
so-called chaotic changes is apoptosis. Apoptosis involves a complex
series of biochemical reactions that result in the eventual elimination
of a cell from the proliferating population. This mode of cell death can
be induced by various cellular, chemical, and physical agents and is
regulated by intracellular enzymes acting on membrane, protein, and
nucleic acid substrates (9). The role of the tumor suppressor gene p53
is important in promoting apoptosis in compromised cells (10). Many
cell types that contain mutant versions of p53 are less likely to
undergo apoptosis: thus, they are more likely to survive a given insult
and acquire further deleterious change (11, 12). Although the precise
molecular mechanisms by which p53 mediates apoptosis are not
known, there are suggested links between p53, oxidative stress, and
the production of reactive oxygen species (13, 14). However, the
dependence of apoptosis on p53 is not absolute because apoptosis has
been reported in p53 null and mutant cells (15). In the study reported
here, we investigated apoptosis as a factor that not only contributes to
the persistence of reproductive cell death but also influences the state
of oxidative stress and the amount of free radical damage in irradiated
cells mutant in p53.
ABSTRACT
Chromosomal instability and persistent reproductive cell death show a
significant correlation after cells are exposed to ionizing radiation. To
examine the possible role of apoptosis in persistent reproductive cell
death, we analyzed subsets of chromosomally stable and unstable clones
for relationships between chromosome stability, reproductive integrity,
and apoptosis. All clones were generated from the GM10115 cell line and
derived from single progenitor cells surviving 10 Gy of X-rays, and all
measurements were made —¿60-80generations after irradiation. The
incidence of apoptosis, as measured by both annexin V binding of phosphatidylserine residues and terminal deoxynucleotidyl transferase label
ing of DNA strand breaks, was significantly higher in chromosomally
unstable clones than it was in chromosomally stable clones (P < 0.05;
ANOVA and Student's t test). Furthermore, statistical analyses of the
biological end points of persistent reproductive cell death and apoptosis
were consistent, showing R2 values of 0.78 and 0.76, respectively. These
results suggest that persistent reproductive cell death can, in part, be
explained by the predisposition of a fraction of the clonal population to
undergo apoptosis or necrosis. Immunological blot analyses of protein
levels and DNA bandshift assays confirmed the mutant status of p53 in the
host cell line, implying an apoptotic pathway that is independent of p53.
Induction of apoptosis by agents such as actinomycin D, etoposide, and
staurosporine and induction of necrosis by sodium azide were accompa
nied by an increase in the level of intracellular peroxy radicals and lipid
peroxidation products, two independent end points that are typically
associated with oxidative stress. Similar findings were observed in several
subclones showing persistent apoptosis. These results suggest that the
elevated levels of free radical damage that we detected were derived from
the fraction of cells dying by apoptotic or necrotic processes. Possible
mechanisms whereby oxidative stress may contribute indirectly to the
perpetuation of chromosomal instability are discussed.
INTRODUCTION
The progression of cancer is believed to result, in part, from the
accumulation of subtle genetic alterations that predispose an individ
ual to develop the abnormalities commonly associated with cancer.
Ionizing radiation is a tool with which to study carcinogenesis by
inducing the onset of many of these abnormalities within a reasonable
time frame. A long history links radiation exposure and the elevated
incidence of cancer ( 1), and the potential molecular mechanisms of
radiation oncogenesis have recently been described (2). Not surpris
ingly, substantial interest has focused on the role of genomic insta
bility in radiation carcinogenesis (3).
Exposure of cells to ionizing radiation results in a variety of directly
Received 3/20/98; accepted 6/17/98.
The costs of publication of this article were defrayed in part by the payment of page
charges. This article must therefore be hereby marked advertisement in accordance with
18 U.S.C. Section 1734 solely to indicate this fact.
1This work was supported by the Office of Health and Environmental Research.
United States Department of Energy, Contract DE-4459-011542; NIH Grant GM 54189
(to W. F. M.); National Cancer Institute. NIH. Grant CA67995-02 (to D. A. B.); Institu
tional National Research Service Award 5T32 ES07106 from NIH (to C. L. L.); and a
Deutsche Forschungsgemeinschaft (Germany) fellowship (to A. H.).
2 To whom requests for reprints should be addressed, at Department of Radiology,
LREH 314. University of California. San Francisco. CA 94143-0750. Phone: (415)4762793; Fax: (415)476-0721: E-mail: [email protected].
MATERIALS
AND METHODS
Cell Culture. All experiments used the human/hamster hybrid cell line
GM 10115, which contains one copy of human chromosome 4 in a background
of 20-24 hamster chromosomes. The growth of these cells has been described
(16). For X-irradiation. cells were exposed at a dose rate of 2.5 Gy/min at
ambient temperature using a Philips RT250 X-ray machine (250 kVp, 15 rnA;
half-value layer 1.0 mm Cu).
3712
Downloaded from cancerres.aacrjournals.org on June 17, 2017. © 1998 American Association for Cancer Research.
COMPROMISED
GENOMIC
INTEGRITY
Cell Populations. Three groups of GM10115 cells were studied: unirradiated control cell clones, clones of cells surviving 10 Gy of X-rays that showed
no chromosomal instability, and clones surviving 10 Gy of X-rays that
exhibited significant and persistent chromosomal instability. All of the chromosomally unstable clones used in this study had at least five distinct metaphase subpopulations of cells with rearrangements
Clones with fewer than three such subpopulations
Table 1 Induction of apopiosis in GMIOl 15 cells b\ X-rti\s and chemicals
Dose/
concentrationNoneActinomycin
Treatment
/1M15
somally stable. Details concerning the isolation, expansion, determination of
plating efficiency, and cytogenetic characterization of all subclones have been
described in detail (17).
Measurement of Apoptosis. To determine whether GM10115 cells could
be induced to undergo apoptosis in response to ionizing radiation, we exposed
exponentially growing cells to 5 or 10 Gy of X-rays and then cultured the cells
azideX-raysX-raysX-raysX-raysX-raysX-rays50
HIM5Gy5Gy5Gy10
Gy10
Gy10
GyAssay
" A total of 1000 cells were scored.
for 24, 48, or 72 h. To determine whether GM10115 cells could be chemically
induced to undergo apoptosis, we exposed exponentially growing cells to
staurosporine (8 /¿M),actinomycin D (50 /J.M).or etoposide (100 /J.M)for 24 h.
Sodium azide (15 mM) was used over the same time period as a control for
inducing necrosis, a nonapoptotic mode of cell death (18). Populations of cells
treated under each condition were then examined for apoptosis by means of
two assays: (a) annexin V-FITC binding of phosphatidylserine residues exter
were trypsinized. counted, and resuspended
Annexin V Binding and TdT Labeling Assays. Logarithmic-phase cul
tures of all clones used in the annexin V binding and TdT labeling assays were
harvested by incubating cells for 10 min at ambient temperature in a solution
of versene [I min EDTA, 40 mM Tris (pH 7.5), and 150 mM NaCl]. This
procedure constitutes a gentle harvest treatment that does not compromise cell
membrane integrity. Annexin V binding and TdT labeling assays were per
formed using the ApoAlert AnnexinV Apoptosis Kit (Clontech Laboratories,
Palo Alto, CA) and the ApopTag in situ Apoptosis Detection Kit (Oncor,
Gaithersburg, MD). following the instructions provided by the manufacturers.
After each assay, cells were fixed for 15 min in 50 ¡j.\of 4% neutral buffered
formalin, dropped onto glass slides, allowed to dry, covered with 20 /xl of
Antifade (Oncor), and fitted with glass coverslips. Slides were stored at 4°C
cellular content of MDA and 4-HNE was derived from standard
Assessment of p53 Status. Nuclear extracts were prepared from cells
exposed to various doses of ionizing radiation at selected times, after which
immunological blot and DNA bandshift assays were performed as described
previously ( 19, 20). Briefly, immunoblots of nuclear extracts were analyzed for
alterations in p53 protein levels by using an antibody to p53 (DO-1, horserad
ish peroxidase-conjugated)
purchased from Santa Cruz Biochemicals (Santa
Cruz, CA). Nuclear protein (10 jxg) from unirradiated or X-irradiated was
separated on denaturing 8% SDS-polyacrylamide gels, transferred to Immobilon-P membranes (Millipore, Bedford, MA), probed with the antibody
against p53, and detected by enhanced chemiluminescence
(Amersham.
Arlington Heights, IL).
Electromobility supershift assays were performed by using an oligonucleotide constructed to contain a consensus binding site for wild-type p53:
5'-GATCCGGACATGCCCGGGCATGTCCG-3
'. The 20 bases underlined
represent the consensus binding site for wild-type p53. These DNA oligomers
were then end-labeled with 12P and incubated with nuclear extracts prepared
from unirradiated and irradiated cells. p53 antibody was used to detect the
supershifted p53-DNA complexes. Where applicable. DNA bandshift and
immunological blots were quantified by Phosphorlmager (Molecular Dynam
ics) or Bio-Rad Gel DOC analysis, respectively. All experiments were per
formed three times, and human melanoma cells known to contain wild-type
p53 (Ul-Mel) were used as positive controls in these assays (19, 20). Equal
amounts of protein were loaded on the gels based on protein concentrations
determined by Bradford assays of each nuclear extract sample.
Determination of Peroxy Radicals. Levels of intracellular peroxy radicals
were assessed in clones by FACS analysis with the use of the dye H2DCFDA
(Molecular Probes, Eugene. OR). Exponentially growing cultures of clones
rescein.
in complete medium at 2 X IO6
cells/ml. Cells were incubated with 10 /J.MH-.DCFDA for 30 min at ambient
temperature; after dye loading, cells were quenched on ice to minimize efflux
of the dye until FACS analysis.
As positive controls for peroxy radical production, cells were preloaded
with H2DCFDA and placed on ice, and then 100-1000 /UMhydrogen peroxide
was added 10 min before FACS analysis. As positive controls for apoptosis
and necrosis, exponentially growing cells were treated with actinomycin D (50
IJLM,16 or 24 h) or etoposide (100 PLM.24 h) to induce apoptosis or sodium
azide (15 mM. 16 or 24 h) to induce necrosis. After these chemical treatments,
cells were loaded with H2DCFDA dye and analyzed by FACS.
FACS analysis was performed with an instrument constructed basically
according to the design of Steinkamp et al. (21). Signals from the photomultiplier were collected in an Oxford multichannel analyzer after analog-todigital conversion and transferred into Microsoft Excel format for display and
calculations.
Lipid Peroxidation Assays. Products of lipid peroxidation were measured
colorimetrically by using a Lipid Peroxidation Assay Kit (Calbiochem, San
Diego. CA), after which the extent of lipid peroxidation was determined by
MDA and 4-HNE assays according to the manufacturer's instructions. Total
nalized to the outer leaflet of the plasma membrane and (b) addition of
dNTP-FITC moieties to the 3'-hydroxyl termini of DNA strand breaks by
TdT.3
3 The abbreviations used are: TdT. terminal deoxynucleotidyl transferase: FACS.
fluorescence-activated
cell sorting; H,DCFDA. 2'.7'-dichlorodihydrofluorescein
diac
ciate; MDA, malonaldehyde; 4-HNE, 4-hydroxy-2(E)-nonenal;
DCF. 2'.7'-dichlorofluo-
time
cells"
positive cells"
after treatment
(h)24242424244872244872AnnexinV-positive
(%)4.941565617142022222623TdT(%)0.3266.5501.82.88.3256.21
DEtoposideStaurosporineSodium
¡MIOO/XM8
of human chromosome 4.
were considered chromo-
until analysis by fluorescence microscopy using a Zeiss Axioskop microscope
equipped with a dual-band pass FITC/Texas Red filter set.
OF CHO CELLS
curves
generated on the day of assay, and all determinations were done in duplicate
and corrected for the amount of protein in each sample.
Treatment of Cells with Xanthine/Xanthine
Oxidase. Logarithmicphase cells were resuspended in PBS containing 1 mM xanthine and then
treated for 15 min with 0-5 x 10~2 units of xanthine oxidase (Sigma
Chemical Co.. St. Louis. MO) at 37°C.After treatment, cell survival assays,
colony isolation, and potential chromosomal instability
formed in selected clones as described previously (16).
analysis were per
RESULTS
Cytogenetic and Clonogenic Characterization. After exposure
of cells to 10 Gy of X-rays, the status of chromosomal stability in the
GMIOl 15 subclones was assessed by fluorescence in situ hybridiza
tion of a labeled probe to the human chromosome. Details concerning
the analysis of these clones has been published (17). As a group,
unstable clones had significantly lower plating efficiencies than did
both the irradiated but chromosomally stable clones and the unirradi
ated control cells (P < 0.05; ANOVA and Student's t test).
Chemical and X-Ray Induction of Apoptosis. To investigate
whether some fraction of the persistent reproductive cell death phenotype could be a consequence of the persistent manifestation of
apoptosis, we conducted an extensive series of control experiments.
GMIOl 15 cells were indeed capable of undergoing apoptosis after
exposure to X-rays or to a variety of chemicals (Table 1). Apoptosis
was determined by assays for annexin V binding of phosphatidyl
serine residues and TdT labeling of DNA strand breaks. Treatment
with sodium azide was accompanied by much lower levels of apop
tosis, consistent with the observation that this agent induces necrotic
cell death (18).
3713
Downloaded from cancerres.aacrjournals.org on June 17, 2017. © 1998 American Association for Cancer Research.
COMPROMISED
GENOMIC
INTEGRITY
OF CHO CELLS
ated controls and 14 Chromosomally stable subclones had signifi
cantly fewer apoptotic cells than did the 12 Chromosomally unstable
subclones (P < 0.05; ANOVA and Student's t test). The morpholog
Table 2 Delayed apoptosis in chromosomally stable and unstable clones
cells'"(%)Chromosomally
cells*1
V-positive
Cloneno.PE'(%:Annexin1
(%)TdT-positive
ical appearance of apoptotic cells is shown in Fig. 1.
clonesControl'102103110114us1261291301321331411451461528772656474695260576984514759518.4146.513146.52412121815101216220.182.901.802.501.101.406.708.104.301.503.5912.814.08.108.00Chromoso
stable
Apoptosis and Persistent Reproductive Cell Death. The corre
lation between plating efficiency and the two assays used to measure
apoptosis is shown in Fig. 2. Correlation coefficients obtained be
tween plating efficiency and annexin V binding (0.78) or TdT labeling
(0.76) indicate an interdependence of these assays and, thus, an
association between apoptosis and persistent reproductive cell death.
p53 Protein Levels. To investigate whether the observed apoptosis
involved the p53 gene product, we used immunological blot analysis
with an antibody probe directed against the p53 protein. GM10115
cells showed little change in p53 protein levels 2 and 12 h after
cellular exposure to 0, 3, 6, or 12 Gy of X-rays (Fig. 3). This same
antibody, however, when used to probe nuclear extracts prepared from
human melanoma cells (Ul-Mel) known to contain wild-type p53 (19,
clones2710112324115138147CS9LSILS12314650674528362851467262331926162732323915318.02613.431.36.92.14.313.821.423.411.016.61.62.0
unstable
20), detected higher basal levels of p53 protein and an increase in p53
protein levels 2 h after exposure to 9 Gy of X-rays (Fig. 3). Sampleto-sample variations in nuclear protein levels were accounted for by
comparison with a-tubulin loading controls (data not shown).
" PE. plating efficiency.
* A total of 1,000 cells were scored.
' A total of 6.000 cells from unirradiated GM10115 subclones were scored.
Analysis of Apoptosis in Chromosomally Stable and Unstable
Subclones. Having established that apoptosis could be induced in
GM10115 cells, we sought to determine whether apoptosis was per
sisting over multiple generations in clones surviving radiation expo
sure. Subclones of GM10115 cells were maintained in exponential
growth and examined for apoptosis measured by the annexin V and
TdT assays ~60 and —¿80
generations after irradiation (Table 2). For
individual clones, the amounts of apoptotic cells found at each of
these time points were not significantly different. However, unirradi
p53 Functionality. To determine whether the absence of elevated
p53 protein levels after irradiation was due to mutant p53, the func
tionality of p53 in GM10115 cells was investigated by means of a
DNA bandshift assay. Nuclear extracts isolated from cells 2 and 12 h
after exposure to 0, 3, 6, or 12 Gy of X-rays contained little, if any,
p53 protein capable of binding the DNA and retarding its mobility
through the gel (Fig. 4). In contrast, nuclear extracts from Ul-Mel
cells contained p53 protein proficient at binding DNA, as evidenced
by the increase in apparent molecular weight of the DNA-p53 protein
complex (Fig. 4).
Intracellular Peroxy Radicals. To examine the production of
intracellular peroxy radicals, a measure of oxidative stress, we mon
itored DCF fluorescence (derived from the intracellular oxidation of
H,DCF, which, in turn, was derived from the enzymatic removal of
acetate moieties from H2DCFDA by cellular esterases). Using an
agent known to lead to intracellular free radical damage, we observed
a dose-dependent increase in the level of DCF fluorescence (as
evidenced by a right shift in peak fluorescence) after GM10115 cells
Fig. 1. Morphological appearance of apoptotic cells
detected by annexin V binding and TdT labeling, a and
b, annexin V binding assay; c and d, TdT labeling
assay; a and c. untreated control cells; b and d, cells
assayed 72 h after exposure to 5 Gy of X-rays.
3714
Downloaded from cancerres.aacrjournals.org on June 17, 2017. © 1998 American Association for Cancer Research.
COMPROMISED
GENOMIC
anDO
DDD
0°
•¿Dmc'S<uC
-20.0015.0010.005.00
rtiM resulted in a concomitant increase in the levels of MDA and
4-HNE, indicating lipid peroxidation (Fig. 6a). Having established the
A^A
response of the assay to a known oxidizing agent, we sought to
determine whether those chemicals that induced apoptosis or necrosis
would also elicit an increase in lipid peroxidation. As observed with
intracellular peroxy radicals, cells treated with actinomycin D, etopo
side, staurosporine, or sodium azide all showed a marked increase in
levels of MDA and 4-HNE (Fig. 6a). Lipid peroxidation was also
analyzed in several chromosomally stable and unstable subclones. As
with intracellular peroxy radicals, several subclones exhibited ele
vated levels of MDA and 4-HNE (Fig. 66).
Persistent Chromosomal Instability after Xanthine/Xanthine
Oxidase Treatment. As a further investigation of whether perturba
tions to the oxidative state of cells would result in chromosomal
instability, we treated cells with xanthine and xanthine oxidase, a
system known to generate Superoxide anions (22). Thirty clones
isolated from treatments resulting in three logarithmic orders of cell
kill and 20 clones isolated from treatments resulting in four logarith
mic orders of cell kill failed to show persistent chromosomal insta
bility (data not shown).
^A
•¿*A
AA
ADA A
C40.0035.00
-n
nn .D
20
40
60
80
100
Plating Efficiency (%)
j30.00
-25.0020.0015.00
9\O_c'S^^3T3
-10.005.00
AApd
LU^^T!35.00
*G
-n
A1
nti -bDDDDÖD
20
40
Ol CHO CFI.I.S
Lipid Peroxidation. To determine whether the observed apoptosis
was associated with lipid peroxidation (another end point commonly
linked to oxidative stress), we assayed cellular levels of MDA and
4-HNE. To ascertain the cellular response and sensitivity of the assay,
we first treated control cells with various concentrations of H2O2 for
30 min at 34°C.Increasing the H2O2 concentration from 0.25 to 10
-30.00
^^^0p^.oC
-25.00
INTEGRITY
DISCUSSION
A
A
1
60
AA—¿
A
80
1
100
Plating Efficiency (%
Fig. 2. Correlations between annexin V hinding. TdT labeling, and plating efficiency.
Annexin V binding Ui) and TdT labeling (b) were quantified by fluorescence microscopy
for chromosomally stable (A) and unstable (D) clones and plotted against their corre
sponding plating efficiencies. The correlation coefficients (R2i were 0.78 U/) and 0.76 ih}.
were treated with 100-1000 JU.MH2O2 (data not shown). Treatment of
cells with actinomycin D, etoposide, or sodium azide resulted in
elevated levels of intracellular peroxy radicals and concomitant in
creases in DCF fluorescence (Fig. 5a).
FACS analysis of chromosomally stable and unstable clones also
revealed increased levels of intracellular peroxy radicals, as demon
strated by enhanced DCF fluorescence compared to unirradiated con
trol cells (Fig. 5b). Although some of the stable clones exhibited lower
fluorescent signals than did the control cells (data not shown), all of
the unstable clones showed higher fluorescent signals than did the
controls.
Chromosomally stable and unstable subclones of GM10115 cells
with significant differences in plating efficiency (17) were used to
address the hypothesis that a reduction in plating efficiency in the
unstable clones was due to the persistence of apoptosis. Apoptosis was
measured by the binding of annexin V to externalized phosphatidylserine residues and the labeling of DNA strand breaks by TdT. The
reduced signal emanating from cells induced to undergo necrosis by
sodium azide, compared to the signal observed after treatment with
radiation or chemicals inducing apoptosis, suggests that these assays
detect specific end points associated with apoptosis. We did not use
the appearance of annexin V- and TdT-positive cells to distinguish
qualitatively between treatments or between chromosomally stable
and unstable cells: rather, we used these assays as quantitative meas
ures of variations in the frequency of apoptosis induced by various
agents or between distinct classes of subclones.
The induction of apoptosis through the use of X-rays and chemical
treatments confirmed that GM10115 cells can apoptose at early times
after insult (24-72 h). This mode of cell death is, thus, a normal
pathway for eliminating stressed or damaged cells. Furthermore, the
cell clones analyzed were found to be undergoing apoptosis ~60 and
—¿
80 generations after irradiation. Because apoptosis can occur days to
Cells
GM10115
U1-mel
Time after
Fig. 3. Immunological blot analysis of p53 pro
tein levels after X-irradiation. Nuclear extracts
from GM10115 and human melanoma Ul-mel
melanoma cells known to contain wild-type p53
were prepared at the indicated times after exposure
to X-rays and probed with an antibody specific to
p53.
p53
3715
Downloaded from cancerres.aacrjournals.org on June 17, 2017. © 1998 American Association for Cancer Research.
COMPROMISED
Cells
Time after
irradiation
OENOMIC
INTEGRITY
DNA
Oligo
taining
cubated
ated or
various
DNA - p53
ProteinComplex
U1-mel
GM10115
None
Gy
Fig. 4. Assessment of p53 functionality by DNA
bandshift assay. 32P-labeled DNA oligomers con
OF CHO CELLS
2h
02h3
6
12
012h3
12
612
•¿
I 4 4 *
a consensus binding site for p53 were in
with nuclear extracts isolated from irradi
unirradiated GMI0115 or Ul-mel cells at
times after irradiation.
H
«
Free DNA
Oligo —¿
ÉBá
months after insult suggests a basis for the persistence of reproductive
cell death after exposure to X-rays. Given the significant correlation
obtained between chromosomal instability and the persistence of
reduced plating efficiency in cells surviving irradiation (17, 23), our
results suggest that persistent reproductive cell death can be explained
in part by the predisposition of some fraction of the population to
undergo apoptosis. It is unlikely that apoptosis can account entirely
for the persistence of reproductive cell death, however. The de novo
formation of lethal mutations (24, 25), chromosome aberrations in
compatible with cell survival (17, 26, 27), and epigenetic factors such
as the bystander effect (28, 29) are also likely to contribute to this
persistent failure in reproductive integrity.
Analysis of cells induced to undergo apoptosis by treatment with
chemicals and X-rays indicated that the externalization of phosphatidylserine residues is a step that tends to precede the formation of DNA
strand breaks. Although most clones showing relatively high levels of
annexin V binding also showed relatively high levels of TdT labeling,
some did not (i.e., LSI2). The implications of this are uncertain but
suggest that the cascade of events transpiring during the process of
apoptosis may be disrupted in chromosomally unstable cells. These
disruptions might further enhance interclonal variability in the expres
sion of a given apoptotic end point and/or lead to differences in the
manifestation of such end points over time.
The ability to induce apoptosis in GM10115 cells by treatment with
several agents and the capacity of these cells to apoptose several
generations after the initial insult prompted investigation of possible
molecular mechanisms underlying this phenomenon. We were unable
to detect increased levels of p53 protein after X-irradiation, suggesting
that GM10115 cells contain a defect in p53 or a dysfunctional path
way regulating the p53 response to radiation. If GM10115 cells
contain a mutant p53, the mutation is likely to be within the DNAbinding region of the protein, a region that is known to be hypermutable (10, 30). The relationship between oxidative stress and apoptosis
was determined in a p53 mutant cell line suggesting, in parallel with
recent reports (13, 14), that there are p53-independent pathways for
modulating the redox state of a cell. It should be stressed that chro
mosomal instability induced by exposure of cells to ionizing radiation
appears to be independent of the cells' p53 status (31 ).
0.04
Control
. Actinomycin D (16h)
Sodium Azide (16h)
Etoposide (24h)
Actinomycin D (24h)
Sodium Azide (24h)
so
100
150
DGF Fluorescence
0.04 y
b
Control
20
40
60
100
DCF Fluorescence
Fig. 5. Detection of intracellular peroxy radicals, a, chemical-mediated induction of
DCF fluorescence. Cells were treated with actinomycin D (50 /¿M),etoposide (100 U,M).
or sodium azide ( 15 mM) for 16 or 24 h. loaded with H2DCFDA dye ( 10 U.M;30 min), and
analyzed by FACS. b, DCF fluorescence in chromosomally unstable subclones. Expo
nentially growing clones were harvested, loaded with dye (10 /XM;30 min), and quenched
on ice until FACS analysis. All results shown for FACS analysis represent the average of
at least two independent experiments.
3716
Downloaded from cancerres.aacrjournals.org on June 17, 2017. © 1998 American Association for Cancer Research.
COMPROMISED
GENOMIC
INTEGRITY
OF CHO CELLS
3.5
*
O)
3
2.5
T
I1'5
1
(0
<
Q
T
1
0.5
Fig. 6. a. lipid peroxidation products in chemi
cally treated cells. Cells were treated at 34°Cwith
various chemicals at the indicated concentrations
for 24 h unless otherwise indicated. All treatments
with H:O2 were for 30 min. Cells were then har
vested, lysed, and assayed for lipid peroxides. Col
umns, MDA and 4-HNE levels, expressed as
jimol/mg of total cellular protein: bars. SD. b. lipid
peroxidation products in chromosomally stable and
unstable clones. Chromosomally stable and unsta
ble clones were harvested, lysed, and assayed for
lipid peroxides. Columns. MDA and 4-HNE levels,
expressed as jamol/mg of total cellular protein:
bars. SD.
Control 0.25 mM
HA
1mM
HA
3mM
H2O2
10mM
H2O2
118
133
50 uM
Actinornycin 0
100 nM 8 |iM
15 mM
Etopo- Stauro- Sodium
side
sperine azide
(24 hr)
15 mM
Sodium
azide
(48 hr)
4 C
3.5
I •¿
2.5
5
t—j
LU 2
Z
X
1.5
•¿o
ra
1 -
T
0.5
OControl
103
114
chromosomally
To understand other ramifications of apoptosis in our stable and
unstable clones, we investigated whether oxidative stress could be
construed as a process capable of driving persistent chromosomal
instability and reproductive cell death. Oxidative stress has been
proposed to be a determinant for apoptosis in a number of cellular
systems (13, 14, 32, 33). When we treated cells with a series of
nonoxidizing chemical agents known to induce either apoptosis or
necrosis, we observed a time-dependent increase in the levels of
intracellular peroxy radicals and lipid peroxidation products. Because
the onset of induced cell death coincided with elevations in oxidative
stress, we propose that the increased intracellular free radical damage
we detected was derived from the fraction of cells destined to die.
Further support for this idea comes from studies showing that the
accumulation of peroxidation by-products is implicated in the loss of
membrane function and integrity (13, 18, 32, 33), that increased levels
of peroxides were found in cell populations containing apoptotic cells
(13, 32, 34), and that increased peroxide production did not precede
the onset of apoptosis but rather coincided with it ( 13). It is important
stable
CS9
LS12
chromosomally
115
138
unstable
to emphasize that, in these studies, the persistent increase in oxidative
stress was measured 60-80 generations after radiation exposure.
Furthermore, oxidative stress in chromosomally unstable clones was
also related to lower plating efficiencies and higher fractions of
annexin V- and TdT-positive cells.
We have demonstrated that H-,0-, (16) or Superoxide anions do not
induce chromosomal instability, suggesting these reactive oxygen
species may not be a major factor in initiating this phenotype. How
ever, Glutton and coworkers (35) have implicated oxidative stress as
a mechanism involved in maintaining genomic instability after cellu
lar exposure to radiation. The results presented here indicate that,
within some clones of cells surviving radiation exposure, subpopulations of cells are continuously being eliminated by apoptosis or
necrosis. This can contribute to the well-characterized phenotype of
lethal mutations or persistent reproductive cell death (24, 27). These
dying cells can also contribute to the increased levels of intracellular
peroxy radicals and lipid peroxidation end products associated with
oxidative stress. We point out that oxidative stress is not always lethal
3717
Downloaded from cancerres.aacrjournals.org on June 17, 2017. © 1998 American Association for Cancer Research.
COMPROMISED
GENOMIC
INTEGRITY
nor is it only detected in chromosomally unstable clones. We suggest
that any population of cells containing a subpopulation undergoing
apoptotic or necrotic elimination has the potential to show increased
levels of damage indicative of oxidative stress. Thus, dying cells
represent a potential source of oxidative stress, in which one ramifi
cation of dying cells within a viable population may include further
compromise to the fraction of cells destined to proliferate.
Reconciling a mechanism whereby oxidative stress serves to di
rectly perpetuate chromosomal instability is difficult, especially in
light of the limited diffusion distances imposed on the reactive oxygen
species mediating the biological effects of oxidative stress by their
inherently short half-lives (<1 s; Ref. 4). However, oxidative stress
may serve to indirectly perpetuate chromosomal instability by the
production of long-lived clastogenic factors (36, 37). Moreover,
chronic exposure to reactive oxygen species and their byproducts may
trigger changes in signal transduction, gene expression, or gap junc
tion communications within a cell or alter the extracellular environ
ment in ways that could influence the phenotype of chromosomal
instability (3, 37). Because of the variable nature of genomic insta
bility, it is possible that many of the biological end points associated
with this phenomenon are governed by a state of chaos; this chaotic
state, which a cell may enter after incurring genotoxic or cytotoxic
stress, could perturb the basal regulatory and signaling pathways
within a cell and disrupt cellular homeostasis.
ACKNOWLEDGMENTS
We thank Jeffrey L. Schwartz for his assistance in the analysis of apoptosis
and Mary McKenney for editing this manuscript.
REFERENCES
1. Sankaranarayanan. K.. and Chakraborty. R. Cancer predisposition, radiosensitivity
and the risk of radiation-induced cancers. I. Background. Radial. Res., 143: 121-143,
1995.
2. Cox, R. Molecular mechanisms of radiation oncogenesis. Int. J. Radiât.Biol., 65:
57-64, 1994.
3. Morgan, W. F., Day, J. P.. Kaplan. M. I.. McGhee. E. M., and Limoli. C. L. Genomic
instability induced by ionizing radiation. Radiât.Res., ¡46:247-258, 1996.
4. Ward. J. F. DNA damage produced by ionizing radiation in mammalian cells:
identities, mechanisms of formation, and reparability. Prog. Nucleic Acids Res. Mol.
Biol., 35: 95-125, 1988.
5. Maity, A.. McKenna. W. G., and Muschel, R. J. The molecular basis for cell cycle
delays following ionizing radiation: a review. Radiother. Oncol., 31: 1-13, 1994.
6. Fornace, A. J., Jr. Mammalian genes induced by radiation: activation of genes
associated with growth control. Annu. Rev. Genet., 26: 507-526, 1992.
7. Uckun. F. M., Schieven, G. L., Tuel-Ahlgren, L. M., Dibirdik. I., Myers, D. E.,
Ledbetter, J. A., and Song, C. W. Tyrosine phosphorylation is a mandatory proximal
step in radiation-induced activation of the protein kinase C signaling pathway in
human B-lymphocyte precursors (Published erratum in Proc. Nati. Acad. Sci. USA.
90: 3775. 1993). Proc. Nati. Acad. Sci. USA, 90: 252-256, 1993.
8. Ward. J. F. Biochemistry of DNA lesions. Radial. Res., 104 (Suppl.): S103-S111,
1985.
9. Kastan, M. B., Canman, C. E., and Leonard, C. J. P53, cell cycle control and
apoptosis: implications for cancer. Cancer Metastasis Rev., 14: 3-15, 1995.
10. Levine, A. J. p53, the cellular gatekeeper for growth and division. Cell, 88: 323-331,
1997.
11. Donehower, L. A.. Harvey. M., Slagle. B. L., McArthur, M. J., Montgomery, C. A. J.,
Butel, J. S.. and Bradley. A. Mice deficient for p53 are developmentally normal but
susceptible to spontaneous tumours. Nature (Lond.), 356: 345-347. 1992.
12. Symonds, H., Krall, L., Remington, L., Saenz-Robles, M., Lowe, S., Jacks, T.. and
Van Dyke, T. p53-dependent apoptosis suppresses tumor growth and progression in
vivo. Cell, 78: 703-711, 1994.
OF CHO CELLS
13. Lotem, J., Peled-Kamar. M., Groner, Y., and Sachs, L. Cellular oxidative stress and
the control of apoptosis by wild-type p53. cytotoxic compounds, and cytokines. Proc.
Nati. Acad. Sci. USA, 93: 9166-9171, 1996.
14. Polyak, K., Xia, Y.. Zweier. J. L.. Kinzler. K. W., and Vogelstein, B. A model for
p53-induced apoptosis. Nature (Lond.), 389: 300-305, 1997.
15. Clarke. A. R., Purdie. C. A., Harrison. D. J.. Morris. R. G.. Bird. C. C., Hooper, M. L.,
and Wyllie, A. H. Thymocyte apoptosis induced by p53-dependent and -independent
pathways. Nature (Lond.), 362: 849-852. 1993.
16. Limoli, C. L.. Kaplan, M. I., Phillips. J. W., Adair, G. M.. and Morgan. W. F.
Differential induction of chromosomal instability by DNA strand-breaking agents.
Cancer Res., 57: 4048-4056. 1997.
17. Limoli, C. L., Kaplan, M. I., Corcoran. J.. Meyers, M., Boothman, D. A., and Morgan,
W. F. Chromosomal instability and its relationship to other endpoints of genomic
instability. Cancer Res., 57: 5557-5563, 1997.
18. Lizard, G., Foumel, S., Genestier, L.. Dhedin, N., Chapul. C.. Flacher. M.. Mutin. M..
Panaye, G., and Revillard, J-P. Kinetics of plasma membrane and mitochondrial
alterations in cells undergoing apoptosis. Cytometry, 2l: 275-283, 1995.
19. Boothman, D. A., Fukunaga. N.. and Wang, M. Down-regulation of topoisomerase I
in mammalian cells following ionizing radiation. Cancer Res., 54: 4618-4626. 1994.
20. Boothman, D. A.. Davis, T. W., and Sahijdak, W. M. Enhanced expression of
thymidine kinase in human cells following ionizing radiation. Int. J. Radiât.Oncol.
Biol. Phys., 30: 391-398, 1994.
21. Steinkamp, J. A., Fulwyler, M. J., Coulter, J. R.. Hiebert, R. D., Homey, J. L., and
Mullaney. P. F. A new separator for microscopic particles and biological cells. Rev.
Sei. Instrum.. 44: 1301-1310. 1973.
22. Ward, A. J., Olive, P. L.. Burr. A. H.. and Rosin. M. P. Response of fibroblast cultures
from ataxia-telangiectasia patients to reactive oxygen species generated during in
flammatory reactions. Environ. Mol. Mutagen., 24: 103-111. 1994.
23. Marder, B. A., and Morgan. W. F. Delayed chromosomal instability induced by DNA
damage. Mol. Cell. Biol., 13: 6667-6677. 1993.
24. Seymour. C. B., Mothersill, C.. and Alper. T. High yields of lethal mutations in
somatic mammalian cells that survive ionizing radiation. Int. J. Radiât.Biol.. 50:
167-179, 1986.
25. Little. J. B., Nagasawa, H., Pfenning. T.. and Vetrovs. H. Radiation-induced genomic
instability: delayed mutagenic and cytogenetic effects of X rays and a particles.
Radiât.Res.. 148: 299-307, 1997.
26. Carrano. A. V. Chromosome aberrations and radiation-induced cell death. I. Trans
mission and survival parameters of aberrations. Mutât.Res., 17: 341-353, 1973.
27. Chang, W. P.. and Little, J. B. Delayed reproductive death in X-irradiated Chinese
hamster ovary cells. Int. J. Radial. Biol.. 60: 483-496, 1991.
28. Mothersill, C., and Seymour, C. Medium from irradiated human epithelial cells but
not human fibroblasts reduces the clonogenic survival of unirradiated cells. Int. J.
Radiât.Biol.. 71: 421-427, 1997.
29. Seymour, C. B., and Mothersill. C. Delayed expression of lethal mutations and
genomic instability in the progeny of human epithelial cells that survived in a
bystander-killing environment. Radial. Oncol. Invest., 5: 106-110. 1997.
30. Zambetti, G. P., and Levine, A. J. A comparison of the biological activities of
wild-type and mutant p53. FASEB J., 7: 855-865, 1993.
31. Kadhim, M. A., Walker. C. A., Plumb. M. A., and Wright. E. G. No association
between p53 status and a-particle-induced chromosomal instability in human lymphoblastoid cells. Int. J. Radiât.Biol., 69: 167-174, 1996.
32. Mancini, M., Anderson, B. O.. Caldwell. E., Sedghinasab, M., Paly. P. B., and
Hockenbery, D. M. Mitochondrial proliferation and paradoxical membrane depolar
ization during terminal differentiation and apoptosis in a human colon carcinoma cell
line. J. Cell Biol., 138: 449-469. 1997.
33. Kruman, I.. Bruce-Keller. A. J.. Bredesen. D., Waeg, G., and Mattson. M. P. Evidence
that 4-hydroxynonenal mediates oxidative stress-induced neuronal apoptosis. J. Neurosci., 17: 5089-5100. 1997.
34. Fernandez, A.. Kiefer, J., Fosdick, L., and McConkey, D. J. Oxygen radical produc
tion and thiol depletion are required for Ca(2 + )-mediated endogenous endonuclease
activation in apoptotic thymocytes. J. Immunol., /55: 5133-5139, 1995.
35. Glutton, S. M.. Townsend, K. M. S.. Walker. C.. Ansell, J. D., and Wright, E. G.
Radiation-induced genomic instability and persisting oxidative stress in primary bone
marrow cultures. Carcinogenesis (Lond.). 17: 1633-1639, 1996.
36. Emerit, 1. Reactive oxygen species, chromosome mutation, and cancer: possible
role of clastogenic factors in Carcinogenesis. Free Radical Biol. Med.. 16: 99-109.
1994.
37. Wiseman. H.. and Halliwell. B. Damage to DNA by reactive oxygen and nitrogen
species: role in inflammatory disease and progression to cancer. Biochem. J., 313:
17-29, 1996.
3718
Downloaded from cancerres.aacrjournals.org on June 17, 2017. © 1998 American Association for Cancer Research.
Apoptosis, Reproductive Failure, and Oxidative Stress in
Chinese Hamster Ovary Cells with Compromised Genomic
Integrity
Charles L. Limoli, Andreas Hartmann, Lee Shephard, et al.
Cancer Res 1998;58:3712-3718.
Updated version
E-mail alerts
Reprints and
Subscriptions
Permissions
Access the most recent version of this article at:
http://cancerres.aacrjournals.org/content/58/16/3712
Sign up to receive free email-alerts related to this article or journal.
To order reprints of this article or to subscribe to the journal, contact the AACR Publications
Department at [email protected].
To request permission to re-use all or part of this article, contact the AACR Publications
Department at [email protected].
Downloaded from cancerres.aacrjournals.org on June 17, 2017. © 1998 American Association for Cancer Research.