Download Cell Cycle-dependent Cytotoxicity of Alkylating

[CANCER RESEARCH 46, 2324-2329, May 1986]
Cell Cycle-dependent Cytotoxicity of Alkylating Agents: Determination of Nitrogen
Mustard-induced DNA Cross-Links and Their Repair in Chinese Hamster
Ovary Cells Synchronized by Centrifugal Elutriation1
David Murray2 and Raymond E. Meyn
Department of Experimental Radiotherapy, The University of Texas, M. D. Anderson Hospital and Tumor Institute at Houston, Houston, Texas 77030
ABSTRACT
Chinese hamster ovary cells were synchronized into the different
phases of the cell cycle by centrifugal vlut ria t¡onand treated with nitrogen
mustard (HN2) in order to investigate the role of DNA damage and
repair processes in the cell cycle-dependent cytotoxicity of this alkylating
antituntor agent. In agreement with previous studies, cell populations
enriched in <;, were the most sensitive to HN2, and those enriched in
late S phase-C-2 were more resistant, as determined by clonogenic assay.
Although the variation in surviving fraction through the cell cycle in
response to a single dose (3 UK/HII; 1.0 h) of HN2 was as great as a factor
of 10, complete dose-response curves generated for the most sensitive
and most resistant elutriator fractions indicated that such changes could
be accounted for by a ratio of D»values of only 1.4. Cells synchronized
by this same method were also analyzed for their relative levels of 11N 2induced DNA cross-linking using the sensitive technique of alkaline
elution. There was no significant difference in the levels of either DNA
interstrand or DNA-protein cross-links induced in the two elutriator
fractions described above immediately after the HN2 treatment. When
the amount of DNA cross-linking in the two fractions was measured 6 h
after treatment, considerable repair had occurred; however, there was no
measurable difference in the rate of repair of either type of cross-link
(i.e., DNA interstrand and DNA-protein) in the different phases. Differ
ences in DNA damage and repair processes could not therefore be
resolved within the confidence limits of available assays and probably
cannot account for the differential cytotoxicity of HN2 towards cells in
the different cell cycle phases.
INTRODUCTION
Since the initial observation that the cytotoxicity of nitrogen
and sulfur mustards depended on the position of the cell in the
cell cycle at the time of treatment (1), the phase specificities of
many bifunctional alkylating antitumor agents have been char
acterized using a variety of cell lines and synchronization tech
niques (for review, see Ref. 2). Despite this diversity of experi
mental approaches, a consistent pattern has emerged: cells in
M or early d are usually the most sensitive to such drugs,
followed by those at the d-S-phase transition, whereas cells in
mid to late S phase are the most resistant. The ultimate poten
tial for clinically exploiting such characteristics depends on a
more complete understanding of the relationships between
these variations in cytotoxicity and biochemical effects of the
drug in the different phases of the cell cycle. DNA is generally
thought to be the principal target for these drugs, based pri
marily on the results of studies with mutant bacterial cells that
show extreme sensitivity to alkylating agents and an inability
to remove alkylated bases from their genome (3, 4). Such
observations have been extended to mammalian systems using
genetically defined DNA repair-deficient mutants of CHO3 cells
isolated by Thompson et al. (S) on the basis of their sensitivity
to UV. These mutants are cross-sensitive to the antitumor drugs
c/s-platinum and mitomycin C (6) and to the bifunctional
alkylating agents melphalan and HN2 (2, 7), and they show an
impaired ability to remove DNA cross-links produced by these
drugs.
The objective of the present study was to investigate the
possible involvement of DNA damage and repair processes and
in particular the formation and removal of DNA cross-links in
the cell cycle-dependent response to a particular alkylating
agent, HN2; specifically, the following questions have been
addressed: (a) does the amount of HN2-induced DNA damage,
or the capacity of cells to repair this damage, vary in different
phases of the cell cycle? (b) Do such changes, if they occur,
correlate with the measured cell cycle phase-dependency of the
cytotoxicity of HN2? In order to address these questions, CHO
cells were synchronized by centrifugal elutriation, a method
that permits the rapid physical synchronization of cells in large
enough numbers to allow the quantitation of DNA damage.
DNA damage was measured using the sensitive alkaline elution
technique developed by Kohn (8) which has the advantage that
it allowed these investigations to be performed using sufficiently
low concentrations of HN2 such that the molecular measure
ments could be compared directly to the cytotoxicity. The
results suggest that there is no significant difference in initial
DNA damage induction in the different cell cycle phases and
furthermore that any possible alterations in the repair of DNA
damage in different phases of the cell cycle are too subtle to be
resolved with statistical significance and are therefore unlikely
to be a major determinant of the observed age-response pattern
for HN2 cytotoxicity.
MATERIALS
AND METHODS
Cell Culture Methods. CHO cells were maintained in exponential
monolayer culture at 37*C in a humidified 5% Å’>2-95%air atmosphere
in McCoy's 5A medium (Hsu's modification, Formula No. 78-5192;
GIBCO, Grand Island, NY) supplemented with 15% FBS (Irvine Sci
entific, Santa Ana, CA). Under these conditions, the cells had a 12- to
14 h doubling time and an 11.5-h generation time; the approximate
durations of the individual phases were 2.5 (d), 6.75 (S), 1.75 (G2),
and 0.5 h (M). Prior to elutriation, cells were harvested at 37°Cusing
0.025% trypsin containing DNase, 20 Mfi/nil. Trypsinization was
quenched after 7 min by adding an equal volume of complete medium,
and the cells were vigorously pipetted and syringed through a 22-gauge
needle to minimize clumping. Light microscopy was used to inspect
the quality of the single-cell suspensions.
Centrifugal Elutriation. Cells were synchronized by the centrifugal
Received 4/15/85; revised 1/15/86; accepted 2/3/86.
elutriation method (9, 10), which exploits the fact that cells continu
1This investigation was supported by USPHS grant CA 23270 awarded by the
ously increase in size as they progress through the cell cycle. This
National Cancer Institute.
2 To whom requests for reprints should be addressed, at Department of
method has several advantages (9) including the fact that it yields
Experimental Radiotherapy, Box 66, M. D. Anderson Hospital and Tumor
adequate synchrony for this application when compared with other
Institute, 6723 Bertner Avenue, Houston TX 77030.
techniques (10, 11). One disadvantage is that it does not allow charac
3The abbreviations used are: CHO, Chinese hamster ovary; HN2, nitrogen
terization of M cells independently of 62 cells. The centrifuge (Model
mustard; PSA, Puck's solution A: CLF, cross-link factor, GSH, reduced glutaJ-21C fitted with a Model .IF 6 elutriator rotor, Beckman Instruments,
thione; FBS, fetal bovine serum; MEM, minimum essential medium; SDS,
Palo Alto, CA) was operated at a constant rotor speed of 1700 rpm.
sodium dodecyl sulfate.
2324
Downloaded from cancerres.aacrjournals.org on June 14, 2017. © 1986 American Association for Cancer Research.
CELL CYCLE DEPENDENCE
Between 1 x 10* and 2.5 x 10" cells suspended in 20 ml of medium
were introduced into the elutnation chamber at a flow rate of 12 ml/
min and elutriated at room temperature by increasing the flow rate of
medium (a-MEM containing 5% FBS and DNase, 20 Mg/ml) through
the chamber in a stepwise manner from about 12 to about 30 ml/min.
In all, twelve SO-mlfractions were collected and placed on ice. Aliquots
were withdrawn for cell count and Coulter volume determinations, and
another aliquot was fixed (in 1 ml saline:3 ml 70% ethanol) for analysis
by flow cytometry to determine the purity of the individual fractions
using an ICP-21 flow cytome ter (Ortho Instruments, Westwood, MA).
The proportion of cells of each phase in a particular fraction was
determined as described by Johnston et al. (12).
Drug Treatments. Either asynchronous cells or cell fractions syn
chronized by centrifugal elutriation were sedimented by centrifugation,
resuspended in medium (containing 20% FBS) at a density of about 1.3
x 11)"/ml.and plated onto 60-mm tissue culture dishes to give about 2
x 10*cells/dish. After incubation for 30 min at 37*C almost complete
attachment of the cells was observed. The cells were then washed twice
with PSA and exposed to various concentrations of HN2 in serum-free
a-MEM for 0.5 or l h at 37'C. Stock solutions of HN2 (mechlorethamine hydrochloride; Merck, Sharp, and Dohme, West Point, PA) were
stored frozen in physiological saline at a concentration of 100 pg/ml.
Aliquots were thawed for 0.5 h at room temperature just before use
and then diluted to the required concentration with serum-free a-MEM.
After treatment, the cells were washed twice with PSA and either
trypsinized immediately (0.025% trypsin) for 7 min at 37*C (cells for
cytotoxicity or initial DNA damage determination) or were incubated
at 37*C in complete McCoy's medium for periods up to 6 h before
trypsinization to allow for repair of the H N2-induced DNA cross-links.
Cytotoxicity was assessed by colony-forming ability, surviving cells
being identified by their ability to form colonies of 50 cells or more.
Alkaline Elution. DNA cross-linking was estimated using the alkaline
elution methodology originally developed by Kohn (8), with slight
modification (6). Briefly, cells were labeled overnight with [MC]thymi-
OF DNA CROSS-LINKING
linear regression analysis. The error limits associated with the slope
values represent the 95% confidence interval. Any DNA strand breaks
induced by the drug treatment would be detected as an increased rate
of elution in the alkaline elution profiles obtained from 11N2 treated
unirradiated control samples that were run simultaneously with the 4 Gy-irradiated samples.
Glutathione. GSH was determined as follows. Various numbers (be
tween 0 and 10 x IO6) of cells were pelleted by centrifugation (2000
rpm for 10 min at 25*Q, washed by resedimenting from PSA, and then
resuspended in 4 ml of distilled water. After vortexing the suspension
for 5 min, metaphosphoric acid (1 ml, 25%) was added, and after a
further 5 min of vortexing, cell debris was removed by centrifugation
(3000 rpm for 20 min) followed by passage of the supernatant through
a 0.45-/»mMillex filter (Millipore Corp., Bedford, MA). GSH in the
supernatant was estimated using the fluorescent dye 0-phthalicdicarboxaldehyde (15). The relative fluorescence emission was a linear
function of the number of cells pelleted within this range but flattened
out somewhat at higher cell numbers (data not shown). The fluorescence
of the GSH standard solutions was also linear over the range 0-0.05
HIM.GSH concentrations were determined from the relationship
GSH = -- x 10-" mol/cell
¿nr
where n was the number (in millions) of cells assayed, /was the sample
fluorescence, and F was the fluorescence from a standard 0.01 HIM
GSH solution.
RESULTS
Cytotoxicity
The cell cycle phase dependence of HN2 cytotoxicity deter
mined after exposure of cells from each elutriator fraction to a
single dose (3 ng/m\ for 1 h) of HN2 is shown in Fig. 1. Cells
in G] were the most sensitive, followed by cells at the GrSphase transition; cells in mid to late S phase tended to be the
most resistant. Cells varied in sensitivity by a factor of about
dine (0.01 /iCi/ml; 50 mCi/mmol; Schwarz/Mann, Orangeburg, NY)
and then chased for 8 h with label-free medium. After drug treatment
and trypsinization, an appropriate volume to give 8 x 10s cells was
impinged onto a 25-mm polycarbonate membrane (2-pm pore; Nudepore Corp., Pleasanton, CA). The cells were then rinsed twice with icecold PSA containing 5 HIMEDTA, lysed with 5 ml of SDS lysis
solution (0.025 M EDTA-2% SDS, final pH 9.7) and then rinsed with
0.1
5 ml of 0.02 M EDTA (pH 10.3). The DNA was eluted overnight in
the dark with 0.1 M tetrapropyl ammonium hydroxide-)1.02 M EDTA
(free acid)-0.1% SDS (final pH 12.15) at a constant flow rate of 0.04
ml/min to give 10 equal fractions of about 3.5 ml. The DNA in the
resulting fractions and that remaining both on the membrane and on
the interior of the membrane holder at the completion of the elution
experiment was measured by liquid scintillation counting (6). The rate
of elution of DNA from HN2-treated cells from the membrane may be
influenced by both DNA-interstrand and DNA-protein cross-links (13).
Although the use of SDS lysis and elution solutions combined with
polycarbonate membranes served to minimize the effect of DNAprotein cross-links on the elution kinetics (14), additional experiments
0.001were performed in which proteinase K (0.5 mg/ml) was added to the
lysis solution and retained in contact with the membrane for 0.5 h prior
to the elution of the DNA. This treatment further reduces the influence
of DNA-linked protein molecules on the elution rate. Those lesions
resistant to digestion are presumed to be predominantly DNA-inter
strand cross-links, whereas those removed by the proteinase K are
0.0001
presumed to represent DNA-protein cross-links (13, 14).
14
12
10
8
Determination of DNA Cross-Linking. To determine DNA crossFraction Number
linking, the cell suspensions were irradiated on ice with 4 Gy of X-rays
Fig. 1. Age-response pattern for Chinese hamster ovary cells exposed to a
immediately prior to the alkaline elution to introduce a constant num
ber of DNA strand breaks (6). CLFs were calculated from the ratio of single concentration (3.0 Mg/ml) of HN2. Cells were separated into fractions
enriched in different phases of the cell cycle by centrifugal elutriation and then
the log of the fraction of uneluted DNA for the 4-Gy-irradiated control
treated with HN2 (l h at 37*C) in medium supplemented with 15% fetal bovine
to that for the 4-Gy-irradiated UN2-treated sample, both values being serum. Following treatment, the cells were plated out for colony formation. The
determined at the same eluted volume (21 ml in this case, although the early fractions (4-5) represent cells enriched in d (e.g., fraction 4 was 96% d,
3% S phase, 1%G2 + M); middle fractions (6-10) were enriched in S phase (e.g.,
relative CLFs did not depend greatly on the actual volume selected).
fraction 8 was 30% G,, 68% S phase, 2% <;, + M); and later fractions (11-13)
The slopes of the resulting cross-linking dose-response curves (ex
are enriched in G2 + M (e.g., fraction 12 was 5% G,, 30% S phase, 65% G2 +
pressed as change in CLF per 0.1 Mg/ml HN2) were determined by M). Points, average of three separate measurements; hurs. SE.
2325
Downloaded from cancerres.aacrjournals.org on June 14, 2017. © 1986 American Association for Cancer Research.
CELL CYCLE DEPENDENCE
OF DNA CROSS-LINKING
treatment protocol for the subsequent synchronized cell studies.
A treatment time of 0.5 h was used in all of these experiments
in an attempt to minimize the repair of lesions during treat
ment. Typical alkaline elution profiles from an experiment in
which cells were trypsinized immediately after treatment with
various concentrations of HN2 are shown in Fig. 4. The elution
rate for DNA from HN2-treated cells relative to untreated
control cells decreased progressively as the concentration of
HN2 was increased between 0-0.6 /¿g/ml,indicating extensive
DNA cross-linking. The alkaline elution technique was suffi
ciently sensitive that relatively low concentrations of HN2 were
required to produce detectable levels of cross-linking. The doseresponse curve relating the initial CLF (zero repair time) to the
HN2 concentration calculated from profiles such as those in
Fig. 4 is shown in Fig. 5A. The inclusion of proteinase K
digestion during the lysis stage decreased the measured cross-
0.03
0.01-
0.003
HN2 Concentration, pg/ml
Fig. 2. Dose-response survival curves for CHO cells from elutriator fractions
5 (•)and 10 (O) after exposure to varying concentrations of HN2. Drug treat
ments were performed at 37*C for 0.5 h in serum-free medium. Points, average
of three separate measurements. Ban, SE.
OS
"»TA
0.1-
0
40
»1200
40
801200
40
80
120
0.03-
CHAME. NUMBt ffWUh«ON* Contint)
Fig. 3. Typical histograms from flow cytometric analysis of the DNA content
nl (.-I) elutriator fraction 5, (B) elutriator fraction 10, and (Q the unseparated
asynchronous population.
36
9
Hours of Button
10 across the cell cycle in response to this particular treatment.
Fig. 2 shows complete dose-response survival curves determined
for two of the fractions, fractions 5 and 10, which were selected
on the basis that they were, respectively, the most sensitive and
most resistant fractions when exposed to the single concentra
tion of HN2 (Fig. 1). It should be noted that for the experiments
in Fig. 2 and in all subsequent data the treatment time was
shortened to 0.5 h, which reduced the overall HN2 cytotoxicity,
and serum-free a-MEM was used for the drug treatments since
preliminary experiments showed that both the cytotoxicity and
DNA cross-linking activity of HN2 were modified in a variable
manner by the presence of FBS. The shapes of the two survival
curves were identical within experimental error. The D0 values
(i.e., the dose required to reduce survival by a factor of e~\
calculated from the slope of the exponential part of the survival
curve) were 0.64 ng/m\ (fraction 5) and 0.90 Mg/ml (fraction
10), respectively, giving a D0 ratio of 1.4. Fraction 5 was
routinely composed of 85-90% GI cells; fraction 10 was usually
composed of about 45% S-phase and 55% G2 cells. Typical
DNA histograms from flow cytometric analysis of the unsepar
ated asynchronous population as well as elutriator fractions 5
and 10 are shown in Fig. 3. The degree of separation achieved
for the other elutriator fractions was similar to that reported
previously (11).
DNA Damage
Asynchronous Cells. Preliminary experiments were per
formed with asynchronous cells in order to establish a suitable
Fig. 4. Typical alkaline elution profiles for DNA from asynchronous Chinese
hamster ovary cells. The profiles are representative of the following treatment
schedules: (O) no drug treatment, no irradiation; (•)no drug treatment, 4-Gy
irradiated; ( ) HN2 (0.2 /ig/ml for O.S h) then 4-Gy irradiated; (T) HN2 (0.4
»ig/mlfor 0.5 h) then 4-Gy irradiated; (•)HN2 (0.6 /ig/ml for 0.5 h) then 4-Gy
irradiated. Irradiations were performed in vitro on ice just prior to alkaline elution
analysis.
HN2 Concentration, ug/ml
Time after treatment, h
Fig. 5. A, degree of relative DNA cross-linking as a function of HN2 concen
tration in asynchronous Chinese hamster ovary cells measured immediately after
treatment with HN2 (O.S h at 37*C); B. repair of DNA cross-links in asynchronous
Chinese hamster ovary cells treated with HN2 (0.6 ug/ml for O.S h at 37*C).
DNA cross-links were measured immediately after and at various times up to 9
h after treatment with HN2. Ban, SE.
2326
Downloaded from cancerres.aacrjournals.org on June 14, 2017. © 1986 American Association for Cancer Research.
CELL CYCLE DEPENDENCE
linking by 26 ±4% (SE). When HN2-treated cells (0.6 /tg/ml
for 0.5 h) were subsequently incubated in complete medium at
37°Cfor various periods after removal of the drug, the number
of cross-links decreased exponentially, with a repair half-time
of 4.8 h (Fig. SB). These repair kinetics are similar to those
reported for LI210 cells in vitro (16) and for mouse fibrosar
coma tumor cells in vivo (17). Proteinase K digestion reduced
the cross-linking measured at 6 h after treatment by 28 ±8%,
indicating that the two types of lesions (DNA interstrand and
DNA-protein cross-links) were repaired with similar kinetics
during this time (0-6 h), an observation that we have also
previously reported for mouse fibrosarcoma tumor cells In vivo
(17).
Synchronized Cells. Rather than obtaining limited data for
each elutriator fraction, we further characterized the response
of the two fractions, 5 and 10, that were the most sensitive and
resistant respectively to HN2 (Fig. 1). No DNA strand breaks
were detected in cells from either fraction at any time up to 6
h after treatment with HN2 at concentrations up to 0.6 /¿g/ml.
It is possible that some low level of strand breakage occurred,
but it may have been masked by DNA cross-linking. Extensive
DNA cross-linking was observed immediately after the treat
ment for both fractions 5 and 10 cells. Dose-response curves
relating total initial DNA cross-linking (i.e., DNA interstrand
plus DNA-protein) to the HN2 concentration for these fractions
are shown in Fig. 6A. Both dose responses were linear over this
concentration range, with slopes of 0.65 ±0.11 (fraction 5) and
0.64 ±0.11 (fraction 10), respectively, indicating that the
frequency of total DNA cross-link induction was identical in
these elutriator fractions at the 95% confidence level. Table 1
shows the results of separate experiments in which cross-linking
induced immediately after the treatment was measured either
with or without proteinase K digestion. Proteinase K reduced
the measured cross-linking by 21 and 32% in fractions 5 and
10, respectively; although these data suggest that there may be
a slightly greater proportion of DNA interstrand to DNAprotein cross-links in fraction 5 relative to fraction 10 cells, the
actual difference in the frequency of DNA interstrand crosslinking in the two fractions was barely significant at the 95%
/
5-
-B
4-
OF DNA CROSS-LINKING
Table 1 Proportion of proteinase K resistant to total DNA cross-linking in
elutriator fractions 5 and 10
Cross-linking was measured either immediately after or 6 h after treatment of
CHO cells with HN2 (0, 0.2, 0.4, or 0.6 pg/ml; 0.5 h treatment at 37'C). Dose
responses were constructed from these data, and the slopes were determined by
linear regression analysis.
Time
(h)0
no.5
total)0.93
Slope"
cross-linking
(% of
±0.21*
With
0.74 ±0.14
0
510
Without
0.91 ±0.25
0
With
0.62 ±0.14
10
0
Without
0.17 ±0.04
6
5
With
0.11 ±0.02
6
510
Without
0.19 ±0.04
6
WithProteinase-resislant
0.13 ±0.0279
6Fraction 10Proteinase-KWithout
68
65
69
'Change in CLF per HN2,0.1 fig/ml. Each slope was determined from at least
nine expérimental
points using three different drug doses.
* Mean ±95% confidence interval.
confidence level (Table 1).
The capabilities of cells from fractions 5 and 10 to repair
cross-links was also compared. Based on the expectation that
any differences in repair are probably relatively small, the
following procedure was used. Cells were exposed to various
doses of HN2 (0-0.6 ng/ml for 0.5 h) and then allowed to
repair for 6 h, after which time the cells were harvested and
assayed for DNA cross-links using alkaline elution as described
above. Complete dose-response curves for DNA cross-linking
were then constructed for each of the fractions (Fig. 6.0), and
by comparing the slopes of these 6-h dose responses with those
determined immediately after treatment (Fig. 6A) we were able
to obtain a statistically more reliable estimate of differences in
repair rate than could have been achieved from single-dose
experiments. The slopes of the 6-h dose-response curves for
total DNA cross-linking (i.e., DNA interstrand plus DNAprotein) were 0.19 ±0.03 (fraction 5) and 0.18 ±0.02 (fraction
10), corresponding to the removal of 71 and 72% of the initial
DNA cross-links for fractions 5 and 10, respectively; again,
these 6-h data showed no significant difference between the two
fractions at the 95% level of confidence. As indicated in Table
1, in separate experiments proteinase K digestion reduced the
measured cross-linking at 6 h by 35 (fraction 5) and 31%
(fraction 10), respectively.
Glutathione. Based on reports (18-20) that cellular thiols
such as GSH are potentially important moderators of cell
sensitivity to alkylating agents such as melphalan and also that
nonprotein sulfhydryl levels may be cell phase dependent (21,
22), GSH levels were determined for the synchronized cell
populations used in this study. The concentration of GSH in
the asynchronous cells was 2.42 ±0.47 nmol/106 cells. GSH
levels for the individual elutriator fractions are shown in Table
2 and generally increased through the cell cycle in a manner
that could be predicted on the basis of the increase in cell
volume; for example, the measured GSH levels in fractions 5
and 10 were 1.54 ±0.41 and 2.66 ±0.58 nmol/106 cells,
I
I3
§
0.2
0.4
0.6
O
0.2
0.4
respectively. When these values were corrected for the increase
in Coulter volume, the GSH levels were identical within exper
imental error, the relative normalized values being 1.0 (fraction
5) and 0.94 (fraction 10), confirming a previous report (23) that
the intraccllular GSH concentration in CHO cells is relatively
constant throughout the cell cycle.
0.6
HN2 Concentration,pg/ml
Fig. 6. Degree of relative DNA cross-linking as a function of HN2 concentra
tion in Chinese hamster ovary cells synchronized by centrifugal elutriation,
measured (A) immediately after or (B) 6 h after treatment with HN2 (0.5 h at
37*Q for cells from elutriator fractions 5 (•)and 10 (O). Points, average of either
10 (0-h data) or 6 (6-h data) separate measurements. Kars, SE.
DISCUSSION
Establishing the relationships between DNA damage, the
repair of this damage, and the cytotoxic effects of antitumor
2327
Downloaded from cancerres.aacrjournals.org on June 14, 2017. © 1986 American Association for Cancer Research.
CELL CYCLE DEPENDENCE
Table 2 Variation in GSH levels through the CHO cell cycle
Cells were synchronized by centrifugal elutriation, and the GSH in these
synchronized populations was determined by a modification of the procedure
described by Cohn and Lyle (15).
Elutriator
fraction
cells)34567.95
GSH concentration
(nmol/ 10*
0.46°.66
±
0.428
0.409
0.3810
0.5811
±0.5112
0.35.54
±
0.41.62
±
0.35.93
±
±
2.03 ±
2.12 +
2.66 ±
2.76
3.49 ±0.60
* Mean ±SE of four separate measurements.
agents is an important objective in understanding the mecha
nism of action of such drugs. The cell cycle specificity exhibited
by these agents provides an interesting situation in which to
address some of these basic questions. The data shown in Fig.
1 indicate that cells in (•,are the most sensitive to HN2, in
agreement with the results of previous studies relating to alkylating agents in general (2) and also with data specifically
relating to HN2 (24-26). Although the alkylation of DNA by
HN2 results in a variety of different lesions, diadducts such as
DNA inter-strand cross-links are presumed to be particularly
cytotoxic (8), based on the fact that this bifunctional drug is
considerably more cytotoxic than its monofunctional counter
part (27). Perhaps the best evidence relating DNA damage to
cytotoxicity is that the cytotoxic effects can be modified by
DNA repair (e.g., see Ref. 28), also suggesting that repair
processes themselves represent an important mechanism of
moderation of drug action; thus, any differences in the initial
level of damage and/or in the subsequent response of the cell
to that damage (i.e., in repair capacity) through the cell cycle
are probably important factors in determining the ultimate
pattern of cytotoxicity.
In previous studies of this type, survival has usually been
determined after exposure of synchronized cells to a single drug
concentration, although considerably more information may be
obtained when complete dose-response curves are generated for
each cell cycle phase (29). In initial experiments, therefore,
surviving fraction was determined over a range of drug concen
trations (Fig. 2) for the two elutriator fractions (5 and 10) that
showed the greatest differential sensitivity to killing by HN2.
This analysis revealed an important point in that although the
variation in surviving fraction through the cell cycle in response
to a single dose of HN2 was as great as a factor of 10 when the
treatment resulted in low survival levels (Fig. 1), this change
was in fact accounted for by a ratio of A> values of only 1.4.
Although this situation is simply a consequence of the divergent
survival curves, it is the change in survival curve parameters
reflected in the ratio of terminal slopes (i.e., of A> values) that
indicates differences in intrinsic cell sensitivity.
The first question that we addressed was whether there were
measurable variations in the initial yield of DNA damage in
cells treated with HN2 in different phases of the cell cycle. It is
clear from the data shown in Fig. 6A that there was no signifi
cant difference in total (DNA interstrand plus DNA-protein)
DNA cross-linking between elutriator fractions 5 and 10 im
mediately after the drug treatment; furthermore, the use of
proteinase K digestion to resolve any possible difference in the
contribution of DNA interstrand and DNA-protein cross-links
to the total cross-linking effect (Table 1) revealed that the
relative frequency of DNA interstrand cross-link induction in
OF DNA CROSS-LINKING
both fractions S and 10 cells was also indistinguishable at the
95% level of confidence. The observed 1.4-fold difference in
cytotoxicity cannot therefore be readily explained in terms of
alterations in initial cross-linking levels.
We therefore determined whether cells from the two fractions
differed in their capacity to remove these DNA cross-links;
however, cells from both fractions 5 and 10 appeared to be
equally efficient in the removal of cross-links (Fig. 6) in the
first 6 h after treatment. Proteinase K digestion again revealed
that the proportion of DNA interstrand to DNA-protein cross
links was identical within experimental error in each fraction
(Table 1), indicating that there was no differential repair of one
or the other type of lesion in the different phases; furthermore,
the fact that the proportion of DNA interstrand to DNAprotein cross-links measured after 6 h was identical to that
measured immediately after the drug treatment indicates that
both types of lesions (DNA interstrand and DNA-protein cross
links) were repaired with similar kinetics during this time;
however, the possibility of subtle changes that cannot be re
solved within the limitations of available DNA damage assays
cannot be completely ruled out. These observations agree with
a previous study (26) which also suggested that neither the
frequency nor repair of DNA interstrand cross-links varied
among CHO cells that were synchronized into the different cell
cycle phases and treated with HN2, even though the earlier
study (26) used a less sensitive assay for DNA cross-linking,
which required a much higher concentration of HN2 (approxi
mately, 6 ng/ml), and would have given a surviving fraction of
less than Id4. Because of the increased sensitivity of the
alkaline elution assay, DNA damage in the present study could
be determined at concentrations for which the degree of cell
killing was much lower (never greater than 50%). It should also
be noted that in a study of this type where the loss of detection
of the DNA cross-link is the measured end point, the possibility
remains that differences in the efficiency of subsequent repair
steps could still account for the observed differences in cytotox
icity.
In conclusion, we confirm the general finding that cells in ( ¡,
are more sensitive to HN2 than those in late S phase or G2.
This sensitivity is apparently a result neither of variations in
initial DNA cross-linking (either DNA interstrand or DNAprotein) nor of differences in DNA cross-link repair capacity in
the different cell cycle phases; on the other hand, considering
that the repair half-times measured here are on the order of 4
h, it seems reasonable to hypothesize that the sensitivity of cells
in the different cell cycle phases could be related to the proba
bility of repairing these lesions prior to the onset of DNA
synthesis, as suggested by Ludlum (30). In such a model, cells
treated in G2 would obviously have a survival advantage over
those treated at the d-S-phase transition.
ACKNOWLEDGMENTS
We thank Raul Laurel for excellent technical assistance.
REFERENCES
1. Walker, I. G., and Helleiner, C. W. The sensitivity of cultured mammalian
cells in different stages of the division cycle to nitrogen and sulfur mustards.
Cancer Res., 23: 734-738, 1963.
2. Meyn, R. I... and Murray, D. Cell cycle effects of alkylating agents. Pharmacol. Ther., 24:147-163, 1984.
3. Lawley, P. D., and Brookes, P. Molecular mechanism of the cytotoxic action
of difunctional alkylating agents and of resistance to this action. Nature
(Lond.), 206:480-483, 1965.
4. Kohn. K. W., Steigbigel, N. H., and Spears, C. L. Cross-linking and repair
of DNA in sensitive and resistant strains of /;'. coli treated with nitrogen
2328
Downloaded from cancerres.aacrjournals.org on June 14, 2017. © 1986 American Association for Cancer Research.
CELL CYCLE DEPENDENCE
mustard. Proc. Nati. Acad. Sci. USA, S3:1154-1161, 1965.
5. Thompson, L. H., Rubin, J. S., Cleaver, J. E., Whitmore, G. F., and
Brookman, K. A screening method for isolating DNA repair-deficient mu
tants of CHO cells. Somatic Cell Genet., 6:391-405,1980.
6. Meyn, R. E., Jenkins, S. F., and Thompson, L. H. Defective removal of
DNA cross-links in a repair deficient mutant of Chinese hamster cells. Cancer
Res., «.-3106-3110,1982.
7. Hoy, C. A., Thompson, L. H., Mooney, C. L., and Salazar, E. P. Defective
DNA cross-link removal in Chinese hamster cell mutants hypersensitive to
bifunctional alkylating agents. Cancer Res., 45:1737-1743, 1985.
8. Knhn. K. W. DNA as a target in cancer chemotherapy: measurement of
macromolecular DNA damage produced in mammalian cells by anticancer
agents and carcinogens. Methods Cancer Res., 16:291-345,1979.
9. Meistrich, M. L., Meyn, R. E., and Barlogie, B. Synchronization of mouse
L-P59 cells by centrifugal elutriation separation. Exp. Cell Res., 705: 169177, 1977.
10. Ordina, D. J., Meistrich, M. L., Meyn, R. E., Johnson, T. S., and White, R.
A. Cell synchrony techniques. I. A comparison of methods. Cell Tissue
Kinet., 77:223-236,1984.
11. Meyn, R. E., Meistrich, M. L., and White, R. A. Cycle-dependent anticancer
drug cytoioxiciiy in mammalian cells synchronized by centrifugal elutriation.
J. Nati. Cancer Inst., 64:1215-1219,1980.
12. Johnston, D. A., White, R. A., and Barlogie, B. Automatic processing and
interpretation of DNA distributions: comparison of several techniques. Cornput. Biomed. Res., //: 393-404, 1978.
13. Ewig, R. A. ( r., and Kohn, K. W. DNA damage and repair in mouse leukemia
1.1210 cells treated with nitrogen mustard, l,3-bis(2-chloroethyl)-l-nitrosourea. and other nitrosoureas. Cancer Res., 37:2114-2122, 1977.
14. Kohn, K. W., Ewig, R. A. G., Erickson, L. C, and /welling. L. A. Measure
ment of strand breaks and cross-links by alkaline elution. In: E. C. Friedberg
and P. C. Hanawalt (eds.), DNA Repair; A Laboratory Manual of Research
Procedures, Vol. IB, pp. 379-401. New York: Marcel Dekker Inc., 1981.
15. Cohn, V. H., and Lyle, J. A fluorometric assay for glutathione. Anal.
B¡oc
hem.. 74:434-440, 1966.
16. Ross, W. E., Ewig, R. A. G., and Kohn, K. W. Differences between melphalan
and nitrogen mustard in the formation and removal of DNA cross-links.
Cancer Res., J«:1502-1506, 1978.
17. Murray, D., and Meyn, R. E. Enhancement of the DNA cross-linking activity
OF DNA CROSS-LINKING
18.
19.
20.
21.
22.
23.
24.
25.
26.
27.
28.
29.
30.
of nitrogen mustard by misonidazole and diethyl maléate
in a mouse fibro
sarcoma tumor in vivo. Cancer Res., 44:91-96,1984.
Taylor, Y. C., Evans, J. W., and Brown, J. M. Mechanism of sensitization
of Chinese hamster ovary cells to melphalan by hypoxic treatment with
misonidazole. Cancer Res., 43:3175-3181,1983.
Vistica, D. T. Cellular pharmacokinetics of the phenylalanine mustards.
Pharmacol. Ther., 22:379-405,1983.
Somfai-Relle, S., Suzukake, K., Vistica, B. P., and Vistica, D. T. Reduction
in cellular glutathione by buthionine sulfoximine and sensitization of murine
tumor cells resistant to L-phenylalanine mustard. Biochem. Pharmacol., 33:
485^t90, 1984.
Ohara, H., and Terasima, T. Variations of cellular sulfhydryl content during
cell cycle of HeLa cells and its correlation to cyclic change of X-ray sensitivity.
Exp. Cell Res., 5«:182-185,1969.
Watanabe, M., and Horikawa, M. Analyses of differential sensitivities of
synchronized HeLa S3 cells to radiations and chemical carcinogens during
the cell cycle. Mutât.Res., 44:413-426, 1977.
Harris, J. W., and Teng, S. S. Sulfhydryl groups during the S-phase: com
parison of cells from (,,. plateau-phase <>,.and Go. J. Cell. Physiol., «7:9195, 1973.
Mauro, F., and Madoc-Jones, H. Age responses of cultured mammalian cells
to cytotoxk drugs. Cancer Res., 30:1397-1408,1970.
Bhuyan, B. K., and Fraser, T. J. Cytotoxicity of antitumor agents in a
synchronous mammalian cell system. Cancer Chemother. Rep., 58:149-155,
1974.
Clarkson, J. M., and Mitchell, D. L. The importance of DNA damage and
repair in the cell cycle sensitivity of CHO cells to nitrogen mustard. Radiât.
Res., 88:587-596,1981.
Wheeler, G. P., Bowdon, B. J., Adamson, D. J., and Vail, M. H. Effects of
certain nitrogen mustards upon the progression of cultured H.Ep. No. 2 cells
through the cell cycle. Cancer Res., 30: 100-111, 1970.
Roberts, J. J. The repair of DNA modified by cytotoxic, mutagenic and
carcinogenic chemicals. Adv. Radiât.Bini.. 7:211-436, 1978.
Drewinko, B., and Barlogie, B. Age-dependent survival and cell-cycle pro
gression of cultured cells exposed to chemotherapeutic drugs. Cancer Treat.
Rep., 60:1707-1717,1976.
Ludlum, D. B. Alkylating agents and the nitrosoureas. In: f. F. Becker (ed.),
Cancer, A Comprehensive Treatise, Vol. 5, pp. 285-307. New York: Plenum
Press, 1977.
2329
Downloaded from cancerres.aacrjournals.org on June 14, 2017. © 1986 American Association for Cancer Research.
Cell Cycle-dependent Cytotoxicity of Alkylating Agents:
Determination of Nitrogen Mustard-induced DNA Cross-Links
and Their Repair in Chinese Hamster Ovary Cells Synchronized
by Centrifugal Elutriation
David Murray and Raymond E. Meyn
Cancer Res 1986;46:2324-2329.
Updated version
E-mail alerts
Reprints and
Subscriptions
Permissions
Access the most recent version of this article at:
http://cancerres.aacrjournals.org/content/46/5/2324
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 14, 2017. © 1986 American Association for Cancer Research.