(CANCER RESEARCH 55. 4(129-411.15. September 15. I
Inhibition of Etoposide (VP-ló)-induced DNA Recombination
Frequency by Bcl-2 Protein Overexpression1
and Mutant
Hisako Hashimoto, Satadal Chatterjee, and Nathan A. Berger2
enls tff Medicine and Hiochemislry anil Cancer Center, Case Western Resen'e University, School of Medicine, Cleveland, Ohio 44106-4937
ABSTRACT
addition to its role in physiological apoptosis, Bcl-2 inhibits the
apoptotic process induced by many anticancer drugs, including DNA-
Bcl-2 has been shown to inhibit apoptosis induced by several anticancer
agents and to cause a dissociation between etoposide (VP-lot-induced
protein-cross-linked
DNA strand breaks and VP-16-induced cell death.
We suggested previously that VP-16-induced cytotoxicity is mediated by a
damaging agents and topoisomerase active agents (10). Recently.
Kamesaki et al. (11) showed that Bcl-2 inhibits apoptotic cell death
induced by etoposide (VP-16),3 a commonly used anticancer agent the
series of events leading from cleavable complex formation to aberrant
UNA recombination, as measured by sister chromatid exchange (SCK)
and Southern blot analysis of the hypoxanthine phosphoribosyl transferase (hprt) gene mutations. To further evaluate this hypothesis and to
determine whether Bcl-2 could affect any steps leading to the aberrant
DNA recombination process, we stably transfected an expression vector
containing human Bcl-2 cDNA into V79 Chinese hamster cells. This
transfection resulted in overexpression of the Bcl-2 gene product. We
subsequently quantitated the relationship between VP-16-induced cyto
toxicity, DNA strand breaks, SCK, and mutant frequency at the II/HIlocus
in these Bcl-2-overexpressing cells. Two independent Bcl-2-overexpressing
cell lines, IH '1.2/2 and ISC1.2/4. showed 3-5 times higher survival at 15 u\i
VP-16 compared with parental V79 cells or control NeoR cells that were
obtained by transfecting V79 cells with the expression vector containing
the G-418 resistance gene only. DNA single-strand breaks induced by
VP-16 were similar in parental V79, control NeoR, BCL2/2, and BCL2/4
cells. In contrast, VP-16 induced significantly less SCE in Bcl-2-overexpressing cell lines compared with parental V79 and control NeoR cells.
The SCE/chromosome induced by 15 UMVP-16 were 0.65, 0.42, 0.09, and
0.10. respectively, in V79, NeoR, BCL2/2, and BCL2/4. In addition, there
was an excellent correlation between VP-16-induced SCE and cytotoxicity
in all cell lines. Furthermore, VP-16-induced mutant frequencies at the
hprt locus were 5-10 times less in BCL-2/2 and ISCI.-2/4 cells than those
observed in the V79 or NeoR control cells. These results indicate that
overexpression of Bcl-2 is associated with reduction in VP-16-induced
genetic recombination, mutation, and cytotoxicity. Moreover, they suggest
that Bcl-2 modulates cytotoxicity of VP-16 between cleavable complex
formation and subsequent induction of DNA recombination events. Thus,
our results provide important support for the hypothesis that VP-16induced cytotoxicity is associated with aberrant recombination events,
including gene deletions and rearrangements.
mutant frequency at the hprt locus, and cytotoxicity of etoposide.
MATERIALS
Chemicals.
AND METHODS
Etoposide (VP-16)
was a gift from Bristol Laboratories
(Syracuse, NY) and was prepared in DMSO at the concentration of 1 mg/ml,
then diluted in a-MEM just before use. G-418 was purchased from GIBCOBRL (Gaithersburg, MD) and dissolved in distilled water at 100 mg/ml G-418,
filtered, and stored at -20°C.
INTRODUCTION
The Bcl-2 gene was originally identified at the juxtaposition of the
IgH locus in the 14; 18 translocation in human follicular B cell
lymphomu (1-3). This gene encodes a M, 26,000 membrane protein
that localizes in the inner mitochondria! membrane, endoplasmic
rcticulum, and nuclear envelope (4-6), and that has the unique
property of enhancing cell survival rather than promoting cell prolif
eration (7). Its capacity to promote cell survival is associated with its
ability to interfere with programmed cell death or apoptosis (7), which
plays an important role in a wide variety of physiological elimination
processes ranging from removal of immune cells in the thymus (8) to
the elimination of redundant neurons in the nervous system (9). In
Received 10/31W: accepted 7/13/95.
The costs of publication of [his 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 study was supported in part by National Cancer Institute Grants R29 CA65920,
Pill CA51I83. and P30 CA43703.
' To whom requests for reprints should be addressed,
mechanism of action of which is mediated by its effects on topoi
somerase II. They showed that Bcl-2 did not reduce the formation of
VP-16-induced protein-cross-linked
DNA strand breaks and sug
gested that Bcl-2 inhibited VP-16-induced cell death at some step
after cleavable complex formation (11). We suggested previously that
VP-16-induced DNA recombination, gene deletions, and rearrange
ments, as measured by the induction of SCE and mutation at the hprt
locus, are downstream events after cleavable complex formation. In
addition, we showed that VP-16-induced SCE frequency correlated
directly with drug-induced cytotoxicity (12-14). We proposed that
obstruction to the replication fork by drug-stabilized cleavable com
plex results in an increase in nonhomologous or aberrant recombina
tion, which contributes to the mechanism of VP-16-induced cell death
(13). The recent observations by Kamesaki et al. (11) provide a
unique opportunity to evaluate our proposed hypothesis of VP-16induced cell death. Thus, we transfected and overexpressed the Bcl-2
gene in V79 Chinese hamster cells and quantitated the relationship
between VP-16-induced cytotoxicity, strand break formation, SCE
formation, and mutant frequency at the hprt locus in these Bcl-2overexpressing cells. The results presented here show that Bcl-2
overexpression does not alter cleavable complex formation: however,
it is associated with a significant reduction in VP-16-induced SCE,
at Case Western Reserve
University, Biomedicai Research Building 3 West, 1091X1Euclid Avenue. Cleveland, OH
44106-4937.
Cell Culture. V79 Chinese hamster cells were maintained in a-MEM
supplemented with 7 mM glutamina, 10% heat-inactivated PCS, 100 units/ml
penicillin G, and 100 fig/ml streptomycin, buffered to pH 7.2 with 25 mm
HEPES.
Transfection of Bcl-2 Plasmid. The human Bcl-2 expression plasmid
(pR509-8-45), prepared by Dr. John C. Reed (La Jolla Cancer Research
Foundation, La Jolla. CA) (15), containing the human Bcl-2 cDNA subcloned
into the retroviral expression plasmid (pBC140), prepared by Drs. E. Gilboa
and B. Cullen (Duke University, Durham, NC), and the control retroviral
expression plasmid (pBC140), which contains the cytomegalovirus promoter/
enhancer and G-418 resistance gene, were kindly provided by Dr. Hiroshi
Kamesaki (Georgetown University, Washington, DC). These plasmids were
stably introduced into V79 cells by the calcium precipitation method (16, 17)
with Protection Mammalian Transfection Systems (Promega, Madison. Wl).
V79 cells were plated the day before transfection. The plasmids were purified
by phenol-chloroform extraction, dissolved in TE buffer [10 mM Tris-HCI (pH
8.0)-1 mM EDTA] and mixed with 124 mM CaCl2, 1 X HBS buffer [50 mM
HEPES (pH 7.1)-280 mM NaCl-1.5 mM Na,HPO4|. and then added to 50%
3 The abbreviations used are: VP-16, VP-16-213 designation for etoposide; SCE. sister
chromatid exchange; hprt, hypoxanthine phosphoribosyl transferase; G-418, gcneticin;
6-TG, 6-ihioguanine.
4029
Downloaded from cancerres.aacrjournals.org on June 15, 2017. © 1995 American Association for Cancer Research.
HcIO AND VIMh-INDl'Å’U
confluent colls. After 16-h incubation, the cells were harvested and replated in
T75 flasks at a concentration of 1 X Id'1 cells/flask. After 32-h incubation,
growth medium was exchanged for selection medium containing 400 /xg/ml
G-418. which was replaced with fresh G-418-containing medium at 3-4-day
intervals for 2 weeks, after which resistant colonies were isolated with cloning
rings, suhcullured. and expanded.
Immunoblot Assays. Expression of the Bcl-2 plasmids in transfected V79
cells was confirmed by immunoblotling using a mouse rnAb to human Bcl-2
protein (DAKO-Bcl-2-124, DAKO, CarpinterÃ-a, CA) as described previously
(18). Briefly, cells were washed and scraped with a rubber policeman into
ice-cold PBS (138 mw NaCI-8.1 mM Na,HPO4-2.7 mM KCI-1.2 mM KH2PO4,
pH 7.4). After centrifugation (400 X g for 5 min), cell pellets were resuspended in 0.15 M NaCI. 10 mM Tris-HCI (pH 7.4), 5 min EDTA, 1% Triton
X-100. I mM phenylmethyl sulfonyl fluoride. 0.23 units/ml aprotinin, IO /XM
leupeptin. and I mvi henzamidine. Alter incubation on ice for 10 min. samples
were centrifuged at 14.000 X i; for 20 min. and the concentration of protein in
the supernatant was measured by the Bio-Rad protein assay kit (Bio-Rad.
Hercules, CA). Twenty /xg of postlysate
SDS-polyacrylamidc gels, electrophoresed.
protein were applied to 12.5%
and transferred lo lmmobilon-P.
polyvinylidene difluoride membrane (Millipore. Bedford, MA; Ref. 18). The
mouse mAb to the human M, 25.000 Bcl-2 protein (DAKO-Bcl-2-124) was
used at a 1:200 dilution and visualized by the horseradish peroxidase-conjugated secondary antibody and enzyme chemiluminescence detection system
reagents (Amersham, Arlington Heights. IL). As an internal standard, we
reprohed Ihe same membrane with mouse mAb to actin, Actin (Ab-1).
(Oncogene Science. Uniondale, NY) used at a 1:1000 dilution and visualized
as above.
Cell Survival Studies. V7l) parental cells, Bcl-2-overexpressing V79 cells
(BCL-2/2 and BCL-2/4), and control transfected G-418-resistant V79 cells
(NeoR2 and NeoR4) in exponential growth were trypsinized and diluted to
seed 500-1000 cells in 100-mm Falcon dishes containing 10 ml culture
medium with or without 400 /xg/ml G-418. After 4-h incubation at 37°Cto
allow for cell attachment, cells were treated with the desired concentration of
etoposide for I h and then washed twice and incubated at 37°Cfor 7 days with
etoposide-free
medium with or without 400 /xg/ml G-418. Colonies were fixed
and stained for 30 min with a solution containing 3 g méthylène
blue and l g
NaOH in 750-ml 0.9% NaCI solution and 250 ml 37% (w/w) formaldehyde.
Survival w as corrected for the cloning efficiency (number of colonies/number
of cells plated) in the absence of VP-16-treatment.
Detection of Internueleosomal
DNA Fragmentation. VP-lfS-induced
DNA ladder was detected by the method of Kamesaki el ill. (11). Briefly,
logarithmic cells were treated with 20 /tM VP-16 for 1 h. At 48 h after VP-16
treatment, cells were trypsinized and washed. Cells (1 X 10'') were lysed on ice
with 10 mM Tris-HCI. 10 mM EDTA. and 0.2% Triton X-KK) (pH 7.5) and then
centrifuged at 14.000 rpm for 10 min at 4°Cto eliminate intact high molecular
weight DNA. Low molecular weight fragmented DNA was extracted from the
supernatant by phenol-chloroform extraction, followed by ethanol precipita
tion, and then electrophoresed in a 2% agarose gel for 2 h at 60 V. DNA was
visualized by staining the gel with ethidium bromide.
Alkaline' Klution. Protein cross-linked DNA single-strand breaks were
measured by the alkaline elution method of Kohn el ill. (19). After 18-h
labeling with 0.02 /xCi/ml [MC|thymidine (50 mCi/mmol. Amersham), cells
were treated with various concentrations
(¡INI.IK
AI TI-.RAÕÕONS
frequency of DNA single-strand
breaks was expressed in rad equivalents as
described previously (19).
SCE. SCE measurements were performed by the method of Goto et al.
(20). Briefly, early logarithmic cells were treated with various concentrations
of etoposide for 1 h at 37°C.Drug was removed by washing the cells twice
with cold PBS. The cells were then incubated with culture medium containing
5-bromo-2-deoxyuridine (10 /IM) in the dark at 37°Cfor 24 h in Ihe case of all
nontreated cells and BCL-2/2 and BCL-2/4 cells treated with 5 /xM VP-16, 36
h in the case of V79 and NeoR cells treated w ith 5 /XMVP-16 and BCL-2/2 and
BCL-2/4 cells treated with 15 /XMVP-16, and 48 h in the case of V79 and
NeoR cells treated with 15 /¿MVP-16. which were the times required for
approximately 2 cell divisions. At 23, 35, and 47 h, colcemid (0.15 /xg/ml) was
added to each flask for the final I h of incubation before harvest. Slides were
prepared and air-dried as described previously, treated with 0.075 M KCI for 10
min at 37°C, fixed with ice-cold methanol: acetic acid (3:1) solution, and
stained in Hoechst 33258 (50 ng/ml) for 15 min. The slides were rinsed with
distilled water, air dried, and mounted with several drops of Mcllvuinc solution
(0.196 M Na,HPO4-0.002 M citric acid, pH 8.25). The slides were then
simultaneously heated at 50°Con a slide warmer and exposed to Black Ray
long wave UV (Fisher. Pittsburgh, PA) for IO min. The slides were rinsed and
subsequently stained with 2.5% Gurr's Giemsa in phosphate buffer (0.01 M
KH,PO4-().()1 M K2HPO4, pH 6.8) for up to 15 min. All procedures described
above were performed in the dark. SCE and chromosome number were
determined on 25 intact metaphase cells for each preparation and expressed as
SCE per chromosome.
Measurement of VP-16-induced Mutant Frequency at the hprt Locus.
The method for selection of hprl~ mutants was described previously (14).
Briefly, all cell lines were maintained tor 2 weeks in «-MEM containing 100
/xM hypoxanthine. 0.4 /XMaminopterin. and 16 /XMthymidine («-MEM-HAT)
with or without G-418 to eliminate sponlaneous hpn~ mutants. At least
1 X Id7 logarithmic cells were treated with various concentrations of VP-16
for 24 h and washed, then allowed an expression time of 10 days, during which
they were subcultured in «-MEMwith or without G-418 every 2 days. Finally,
1 X IO7 cells were replated at the density of 1 X l()4/ml in fifty 75-cnr flasks
(2 X 1If/flask) and exposed to 2 /xg/ml 6-TG for 14 days. After 14 days
exposure to 6-TG. the number of 6-TG-resistant colonies was counted. Mutant
frequency (MO was calculated as follows: Mf = cloning efficiency in the
presence of 6-TG/cloning efficiency in the absence of 6-TG. Mf value at the
hpri locus was log transformed for statistical analysis.
RESULTS
Overexpression of Bcl-2 in V79 Cells. We selected two independ
ent G-418-resistant V79 clones, BCL-2/2 and BCL-2/4 (Fig. 1), which
were transfected with the Bcl-2 expression plasmid (pR 509-8-45) and
overexpressed Bcl-2. In addition, we selected as controls two G-418resistant V79 clones, NeoR2 and NeoR4, transfected with the same
plasmid (PB140) without the Bcl-2 cDNA. All of these clones have
been maintained and grown continuously in a-MEM media contain
ing 4(H) /xg/ml G-418. Bcl-2 overexpression in BCL-2/2 and BCL-2/4
V79 NeoR2 NeoR4
of VP-16 for I h and then detached
BCL2/2 BCL2/4
by gentle scraping with a rubber policeman, washed once, and resuspended in
ice-cold PBS. Cells (2.5 x Id5) were deposited by mild suction on 2-/im
polycarbonate filters (Nucleopore, Pleasanton. CA). An equal number of
L1210 mouse leukemia cells labeled with 0.02 ¿iCi/ml |'H]thymidine
(2
45kD
—Actin
25kD
- Bcl-2
Ci/mmol. Amersham) were irradiated with 600 rad and collected on each filter
to serve as internal standards. The cells were treated with 2.5 ml lysis buffer
(0.02 M disodium-EDTA-0.1 M glycine-2% SDS. pH 10.0) for 30 min. deprotcinized for another 30 min with 0.5 mg/ml proteinase K in the same lysis
buffer, and washed with 3-ml 0.02 M EDTA, pH 10.0. The DNA was eluted in
the dark at pH 12.5 with a 2% (w/v) tetrapropylammonium hydroxide solution
(Fisher, Pittsburgh, PA), 0.02 M EDTA, and 0.2% SDS at a rate of 0.04 ml/min
using a peristaltic pump (Gilson. Middletown, Wl). Ten 90-min fractions were
collected in scintillation vials and radioactivities of I4C and 3H were measured
by using a Liquid Scintillation C'ounler (Beckman, Arlington Hights. IL).
Processing of the radioactivity
was as described by Kohn el al. (19). The
Fig. I. ImmunoMot analysis of Bcl-2 protein in BCL-2/2. BCL-2/4. NeoR2. NeoK4.
¡indparental V7l) cell lines. Twenty /ig of prolein from eaeh cell line were analv/ed hy
SDS-PACii: and ¡mmunoblouing as described in "Materials and Methods." The M, 45.IHHI
actin prolein was immunohlotted
molecular weight in thousands.
as an internal standard on Ihe same membrane, kl),
4030
Downloaded from cancerres.aacrjournals.org on June 15, 2017. © 1995 American Association for Cancer Research.
Bcl-2 AND VP-lh-INWKTO (ll-NITir
ALTERATIONS
cells and its absence in the control-transfected NeoR2 and NeoR4
cells is shown in Fig. 1 by immunohlot assay. Overexpression of
Bcl-2 has been found to be stable for more than 7 months.
Cytotoxicity of Etoposide (VP-16). Fig. 2 shows the clonogenic
cell survival curves for V79, BCL-2/2, BCL-2/4, NeoR2, and NeoR4
cell lines after 1-h exposure with various concentrations of VP-16.
The VP-16 concentrations needed to reduce survival by 50% (IC5(P)
and 90% (1C,,,,) are compared in Table 1 for the different cell lines.
These studies clearly demonstrate that the Bcl-2-overexpressing cell
lines are significantly more resistant to VP-16 than are the V79
parental cell line or the O-41K-resistant. transfected control cell lines.
Because VP-16 toxicity is highly dependent on DNA synthesis, the
03 DOZ <
o o
100
Fig. 3. Internucleosomal DNA fragmentation induced by 20 ¡inVP-1 h for 1 h. Low
molecular weight, fragmented DNA was extracted from 1 x HI" cells and electrophoresed
in a 2r/r agarose gel as described in "Materials and Methods." MW. A DNA///V/U/MI
molecular si/e marker.
oc
10
20
30
40
50
60
VP-16
Fig. 2. Cytotoxicity of VP-16 in V79 (•).control-transfcclcd cell lines NcoR2 |O| and
NcoR4 (+ ]. and Bcl-2-ovcrexprcssing cell lines BCL-2/2 |LI| and BCL-2/4 [•].Cells
were trealed with indicated drug concentration l'or l h at 37°C,and colonies were assayed
as described in "Materials and Methods." Each experiment was performed in triplicale.
Filini*, means of al least 3 separate experiments: han, SE. Plating efficiencies of V7l).
NeoR2. NcoR4. BCL-2/2. and BCI.-2/4 were 95.0, 97.1, 97.2, 90.2, and 64.4%,
respectively.
Table 1 Ctnn[ninni\'i' t'fftrcl\ nf Rcl-2
tni VP-16-imiuct'il <Tfi
ICW"(flM)15.3
Cell
linesV79
+ 0.4
±0.4
NeoR2
20.5 ±0.9
6.8 ±0.2
6.4 + 0.3
NeoR4
17.8 ±1.9
17.7+1.5
BCL2/2
46.0 ±3.1
15.6 + 0.7VP-16
BCL2/4VP-16IC5(,"6.0
41.8 ±2.9
" 1C5(, and ICg,, values were determined for cells treated with VP-16 for l h at 37°C
as described in "Materials and Methods." Differences between V7°.NeoR2. and NeoR4
versus BCL-2/2 and BCL-2/4 are significant al /' < 0.01.
possibility was considered that variable S-phase fractions in the dif
ferent cell lines might have contributed to the different levels of
etoposide cytotoxicity. To assess this possibility, we analyzed the cell
cycle distribution of these cell lines by using the software ModFit Cell
Cycle Analysis on the Elite ESP flow cytometer (Coulter Corp.,
Kendall, FL). The fraction of cells in S-phase during exponential
growth was as follows: V79, 50.9%; NeoR2, 58.4%; NeoR4, 50.7%;
BCL-2/2, 50.8%; and BCL-2/4, 40.7%. Even though BCL-2/4 cells
have a 10% decrease in S-phase cells compared to V79 cells, this
decrease cannot account for the 2.7-fold increase in the observed
resistance to etoposide-induced toxicity in this cell line. Furthermore,
in the case of BCL-2/2 cells, the S-phase fraction was identical to the
V79 control cells, thus eliminating the possibility that variations in
cell cycle distribution due to Bcl-2 overexpression are contributing to
the differential susceptibility.
Etoposide-induced Internucleosomal DNA Fragmentation. As
shown in Fig. 3. 20 /LIMVP-16 clearly induced DNA fragmentation in
V79 and control NeoR2 cells. In contrast, Bcl-2-overexpressing cell
lines showed a much smaller extent of VP-1 (»-inducedDNA fragmen
tation. Thus, it appears that cytotoxicity of VP-16 in V79 and control
cells may be associated with induction of apoptosis and DNA frag
mentation, whereas the decrease in VP-16 cytotoxicity in Bcl-2overexpressing cells is associated with a reduction in VP-16-indiiced
DNA fragmentation.
Etoposide-induced DNA Single-Strand Breaks (SSB). After en
tering into the nuclei, etoposide or other topoisomerase Il-targeted
drugs induce formation of protein-cross-linked DNA strand breaks
termed cleavable complexes (21, 22). Upon treatment with proteindenaturing conditions, these complexes become frank DNA strand
breaks. Many cell lines that are resistant to these drugs have been
shown to have a decrease in drug-induced cleavable complex or DNA
strand break formation (23-25). However, Fig. 4 shows that there are
no differences between the frequency of etoposide-induced, dosedependent protein-cross-linked DNA strand breaks in the Bcl-2-overexpressing cell lines and the parental V79 or the control transfected
cell lines (P > 0.05). Thus, VP-16 induces the same level of cleavable
4031
Downloaded from cancerres.aacrjournals.org on June 15, 2017. © 1995 American Association for Cancer Research.
Bcl-2 AND VP-16-INDUCED
O
ta
ments that can lead to cell death (13). Thus, we analyzed the mutant
frequency at the hprt locus in Bcl-2-overexpressing cell lines to
evaluate whether the observed decrease in VP-16-induced cytotoxicity is associated with a decrease in VP-16-induced mutant frequency.
Fig. 7 shows that VP-16 induces fewer mutants in Bcl-2-overexpressing cell lines than in parental V79 and NeoR controls. Because most
of the VP-16-induced mutations at the hprt gene locus have been
shown to result from deletions and/or rearrangements (13), these
results suggest that Bcl-2 overexpression interferes with the VP-16induced genetic recombination process that may be responsible for
VP-16-induced cytotoxicity.
The differential effects of VP-16 on cytotoxicity and genetic events
in Bcl-2-overexpressing cell lines are summarized in Table 2. VP-16induced cytotoxicity is significantly different in the different cell
lines, with Bcl-2-overexpressing cell lines showing marked resistance
to VP-16, despite the fact that the drug induces similar levels of
1000
V)
-X
a
o
m
•o
ALTERATIONS
mutations can be readily measured at the hprt locus (14). Because
cells that have mutations in this gene cannot metabolize 6-TG to
6-thioinosine monophosphate, which is subsequently metabolized to
6-thiopurine triphosphate that kills the cells, the mutant cells can
survive in the presence of 6-TG. We have also shown that these
mutations induced by VP-16 are associated with genetic rearrange
1500
"5
oo
GENETIC
500
-
C
(O
I—
»-
co
o
C»
C
co
0
5
10
15
20
0.80
VP-16
(MM)
Fig. 4. VP-16-induced DNA single-strand breaks. V79 (•),NeoR2 (O). BCL-2/2 (D),
and BCL-2/4 (•)cells were treated with the indieated drug concentrations for l h at 37°C.
DNA single strand breaks were assayed by alkaline elution as described in "Materials and
Methods." Poinis. means of at least 3 independent experiments; bars, SE.
complex formation in the sensitive and resistant cell lines, suggesting
that variations in this early response to VP-16 are not likely to
contribute to differences in drug susceptibility.
Induction of Sister Chromatic! Exchange by VP-16. Because
previous studies have shown that VP-16 is a potent inducer of SCE
and that the frequency of SCE can be directly related to cytotoxicity
(12, 26, 27), we evaluated VP-16-induced SCE in these cell lines. Fig.
5 shows the relationship between VP-16 concentration and SCE
induced in V79, BCL-2, or NeoR cell lines. In all cell lines, VP-16
induced SCE in a dose-dependent fashion, but in Bcl-2-overexpressing cells, VP-16 induced significantly fewer SCE than in V79. In
BCL-2/2 and BCL-2/4, 15 /IM VP-16 induced four times less SCE
compared to the control-transfected cell line and six times less com
pared to the V79 parental cell line (P < 0.001). Whereas some
Bcl-2-transfected cells have been reported to remain in G,, (7, 28), this
possibility could not have contributed to decreased SCE frequency
because this assay system analyzes SCE only in cells that have
undergone two cycles of cell division.
Fig. 6 shows the results obtained when etoposide-induced alter
ations in clonogenic survival are plotted against drug-induced SCE for
the different cell lines. Clearly, there is an excellent correlation
(correlation coefficient = -0.98) in all cell lines between etoposideinduced SCE and cytotoxicity. These results extend those we reported
previously showing a similar correlation between etoposide-induced
SCE and cytotoxicity in a variety of V79-derived cell lines with
various non-Bcl-2-dependent etiologies for etoposide resistance (12).
Taken together, our results suggest that overexpression of Bcl-2
protein is associated with inhibition of the cytotoxicity of etoposide at
a stage before SCE formation.
VP-16-induced Mutant Frequency at hprt Locus. Using 6-TG as
a selective agent, we have shown previously that VP-16-induced
0.70
-
0.60
-
CD
E
o
W
o
E
o
i_
£2
O
LU
O
C/3
•o
<D
O
3
•o
0.20 -
C
0.10
-
0.00
10
VP-16
15
20
(|jM)
Fig. 5. Formation of SCE in different cell lines after treatment with the indicated
concentration of VP-16 for I h at 37°C.Background SCE were subtracted from the net
SCE formed after VP-16 treatment to provide the number of SCE induced by VP-Ih.
Results are expressed as SCE/chromosome. Points, mean of 25 metaphase spreads; burs,
SE. Induced SCE/chromosome for V79 (•),NeoR2 (O). BCL-2/2 (D), and BCL-2/4 (•)
background SCE/chromosome were: V79, 0.2ܱ0.07; NeoR2, Ü.21±0.09; BCL-2/2,
0.13 ±0.11; and BCL-2/4, 0.11 ±0.03.
4032
Downloaded from cancerres.aacrjournals.org on June 15, 2017. © 1995 American Association for Cancer Research.
Bcl-2 AND VP-lh-INDUCED
CiliNl-TIC AI.THRAÕÕONS
tion bypass mechanisms that includes both homologous and nonho-
100
mologous recombination events (13). Fig. 8 provides a modification
of the original scheme showing that the recombination frequency can
be quantified by measuring SCE and providing the basis for our
observation that a correlation exists between SCE and eytotoxicity
(12). Thus, with exposure to increasing drug concentrations, there will
be an increase in both homologous and nonhomologous recombina
tion. Because homologous recombination creates no genetic change,
we expect to see no resultant toxicity. In contrast, nonhomologous
recombination has the capacity to cause gene deletions and/or rear
rangements, the consequences of which will depend on the nature of
the gene affected. If an essential gene is affected, its deletion should
lead to a loss of the essential gene product and consequent cell death.
If a nonessential gene is deleted such as in the case of hprl, it should
have no effect on cell survival. However, the rate of such events can
be quantitated under selective conditions such as growth in 6-TG. If
the drug-induced recombination process results in rearrangement of
CE
D
CO
100
to
ó
10
0.00
0.20
0.40
0.60
0.80
o
C
Induced SCE/Chromosome
l-'ig. o. Relationship between cytotoxicity and induced SCE-.after exposure to different
coneenlrations of VP-I6 for l h at 37°C.Data are derived from Figs. 2 and 5. Puinis. line
of regression fit by the equation v = (1.933e""'"7411with a correlation coefficient of -0.98.
V79. • NcoR2. O: BCL-2/2. D; BCL-2/4, •
cleavahlc complex formation in all cell lines. In contrast, the Bcl-2overexpressing cell lines show a marked decrease in SCE formation
and mutant frequency, which is consistent with the Bcl-2 inhibition of
VP-16-induced eytotoxicity. Thus, Bcl-2 overexpression does not
affect VP-16-induced clcuvablc complex formation; however, it is
associated with a significant reduction in drug-induced SCE forma
tion, mutant frequency, and eytotoxicity.
o
3o0
<a
**
3
0.00
0.20
0.40
0.60
0.80
1.00
DISCUSSION
Whereas most topoisomerase II active agents lead to dose-depen
dent increases in cleavable complex formation (29), we and others
have proposed that cleavable complex formation is not the immediate
cause of drug-induced cell death (11, 12, 30, 31). This proposal is
based on a variety of observations that show cell cycle and/or kinetic
differences, as well as cell line-dependent discrepancies (30) in the
relationship between cleavable complex formation and eytotoxicity
(31). This dissociation has been emphasized further by a recent report
by Kamesaki et til. (11) showing that Bcl-2 overexpression alters the
relation between cleavable complex formation and VP-16-induced
eytotoxicity. In addition, these studies (11) provide further support for
the use of Bcl-2 as a molecular probe to dissect the sequence of events
leading to etoposide-induced eytotoxicity.
We suggested previously a pathway for topoisomerase II active
agent-induced cell death in which the drug first induces formation of
a cleavable complex; the cleavable complex presents an obstruction to
replication fork progress, which in turn leads to activation of replica
VP-16
Fig. 7. VP-16-indueed mutant frequencies at the l¡¡>n
livcus. Cells were treated with
indicated drug concentrations for 24 h. Mutant frequency was assayed as described in
"Materials and Melhods." V70-. •;NeoR2, O; BCL-2/2, D; BCL-2/4, •
Table 2 Comparativi1 effect* of Hcl-2 overexprexsion {in ctmxt'tiucnccx
ofVP-lfi irt'iiiim-ni"
Cell
Cleavahle
equivalent)V79
lines
(rad
NeoR2
BCL2/2
BCL2/41
complex
112 ±37
1126± 122
1035 ±52
1072 ±66Mutant
frequency
(XI(P')11.3
(%)89.4
±3.7
±0.10
0.42 ±0.05
81.3
10.8 ±3.2
1.95 ±0.75
0.09 ±0.05
44.3
0.65 ±0.45SCE/chromosome0.65
0.10 ±0.02Cvtotoxicity
47.7
±0.7
±1.7
±3.2
±2.2
" All values were determined on cells treated with 15 /¿M
VI1-16 for l h at 37°C,except
for mutant frequencies that were determined tin cells treated with 0.5 ¿IM
VI*-16 for 24 h
at 37°C.Values for eytotoxicity are given as the 'V clonogenic cells killed and arc
presented as the inverse of survival measurements in Table 2. Cleavable complex
formation in V7li and NeoR2 shows the same frequency as in BCL-2/2 and BCL-2/4
(P > 0.05). V7*i and NeoR2 are significantly different from BCL-2/2 and BCL-2/4 in
mutant frequency. (P < 0.005). SCE/chromosome (P < 0.(X)1), and cytotoxicily (P < 0.01 ).
4033
Downloaded from cancerres.aacrjournals.org on June 15, 2017. © 1995 American Association for Cancer Research.
Bcl-2 AND VP-16-INDUCED
GENETIC ALTERATIONS
VP-16
I
inhibits etoposide-induced cell death somewhere between the induc
tion of cleavable complex formation and SCE formation. These stud
ies show clearly that Bcl-2 overexpression prevents VP-16-induced
genetic recombination and mutations.
Topo II
VP-16/TopoII/DNAComplex1
Replication
Block
ACKNOWLEDGMENTS
*
Replication
BlockBypass
We thank Dr. Hiroshi Kamesaki for plasmids pBC140 and pR 509-8-45.
GeneticRecombination?-
REFERENCES
Horaologous
-Non-Homologous-
1. Tsujimoto. Y.. Finger L. R., Yunis. J.. Nowell. P. C.. and Croce. C. M. Cloning of the
chromosome breakpoint of neoplastic B cells with the 1(14:18) chromosome translo
No genetic
Mutationin
Mutation
In
cation. Science (Washington DC), 226: 1097-1099. 1984.
Mutation in
Non-essential
Alteration
Essential
Oncogene or
2. Bakhshi. A., Jensen.J. P., Goldman, P.. Wright. J. J.. McBride. O. W., Epstein, A. L.,
Genes
Genes
Tumor Supressor gene
and Korsmeyer. S. J. Cloning the chromosomal breakpoint of 1(14:18) human lymphomas: clustering around JH on chromosome 14 and near a transcriptional unii on
18. Cell, 41: 899-906, 1985.
Deletion
ofEssentialGeneinNon-essentialGene
inOncogene
3. Cleary, M. L., and Sklar. J. Nucleotide sequenceof a 1(14:18) chromosomal break
orTumor
Products1VCell Products1»Normal
point in follicular lymphoma and demonstration of a breakpoint cluster region near a
GeneProducts<rMalignant
Supressor
transcriptionally active locus on chromosome 18. Proc. Nati. Acad. Sci. USA, 82:
7439-7433, 1985.
Death*Alteration
Normal Cell
CellSurvival
Transformation 4. Hockenbery, D.. Nunez. G.. Milliman. C, Schreiber. R. D.. and Korsmeyer. S. J.
Bcl-2 is an inner mitochondrial membrane protein that blocks programmed cell death.
Survival
ExeptUnder
SelectiveConditions4Alteration
Nature (Lond.). 348: 334-336, 1990.
5. Chen-Levy, Z., Nourse. J., and Cleary. M. The Bcl-2 candidate proto-oncogene
product is a 24-kilodallon integral-membrane protein highly expressed in lymphoid
Fig. 8. Proposed pathway for VP-1ft induced cytotoxicity and genetic alterations.
cell lines and lymphomas carrying the 1(14:18). Mol. Cell. Biol., V: 701-710, 1989.
Numbers on figure indicate steps along pathway measured in this manuscript. 1, VP-166. Alnemri. E. S.. Robertson, N. M., Fernandes, T. F.. Croce, C. M., and Litwack, G.
Topo II-DNA complex formation measured by cleavable complex formation: 2. genetic
Overexpressed full-length human BCL2 extends the survival of baculovirus-infected
recombination frequency measured by sister chromatid exchange: 3. cell death measured
Sf9 insect cells. Proc. Nati. Acad. Sci. USA, XV: 7295-7299, 1992.
by clonogenic survival; and 4. nonesscntial gene mutant frequency measured by clono7. Vaux. D. L., Cory, S., and Adams, J. M. Bcl-2 promotes the survival of haemopoietic
genic survival of hprr mutants selected for growth in 6-TG.
cells and cooperates with c-ttiyc to immortalize prc-B cells. Nature (Lond.), 335:
440-442, 1988.
8. Veis, D. J., Seutman. C. L.. Bach. E. A., and Korsmcyer. S. J. Expression of the Bcl-2
protein in murine and human lymocytes and in peripheral T lymphocytes. J. Immuan oncogene and/or deletion of a tumor suppressor gene, the process
noi., 151: 2546-2554. 1993.
could lead to malignant transformation. This hypothesis provides a
9. Allsopp. T. E., Wyatt, S., Paterson.H. F., and Davies, S. M. The proto-oncogene bcl-2
can selectively rescue ncurotrophic factor-dependent meurons from apoptosis. Cell,
unifying mechanism to explain both the cytotoxicity and the increased
73: 295-307. 1993.
mutant frequency induced by topoisomerase II active agents, as well
10. Miyashita, T., and Reed, J. C. Bcl-2 oncoprolein blocks chemotherapy-induced
as the recently reported development of VP-16-induced secondary
apoplosis in a human leukemia cell line. Blood, XI: 151-157, 1993.
leukemias with characteristic chromosomal rearrangements (32-34).
11. Kamesaki, S., Kamesaki. H., Jorgenscn. T. J.. Tanizawa. A.. Pommier. Y., and
Cossman, J. Bcl-2 protein inhibits etoposide-induced apoptosis through its effects on
Moreover, the scheme shown in Fig. 8 provides the basis for direct
events subsequent to topoisomcrase Il-induced DNA strand breaks and their repair.
Cancer Res., 53: 4251-4256, 1993.
measurements of different components of this hypothesis. Thus, the 3
12. Chatterjee. S., Trivedi. D.. Petzold. S. J.. and Berger, N. A. Mechanism of epipodopotential consequences of nonhomologous recombination should be
phyllotoxin-induced cell death on poly(adenosine diphosphate-ribose) synthesis-de
proportional (although not equal) because they are all derivatives of
ficient V79 Chinese hamster cell lines. Cancer Res.. 50: 2713-2718. 1990.
the same process, and an increase in drug-induced cell death
13. Berger, N. A.. Chatterjee, S.. Schmotzer, J. A., and Helms, S. R. Etoposide (VP-162l3)-induced gene alterations: potential contribution to cell death. Proc. Nati. Acad.
should be reflected in a proportional increase in SCE and mutant
Sci. USA, XX: 8740-8743, 1991.
frequency. This scheme further allows for identification of steps
14. Hashimoto. H.. Chatterjee. S.. and Berger. N. A. Mutagenic activity of topoisomerase
I inhibitors. Clin. Cancer Res.. /: 369-376, 1995.
that could be affected by different inhibitors or processes such as
15. Reed, J. C., Cuddy, M.. Haldar, S., Croce, C.. Nowell, P.. Makover, D., and Bradley,
overexpression of Bcl-2. Thus, if an agent inhibits the process
K. BCL-2 mediated tumorigcnicity of a human T-lymphiod cell line: synergy with
MYC and inhibition by BCL-2 antisense. Proc. Nati. Acad. Sci. USA, 87:
between cleavable complex formation and genetic recombination,
3660-3664. 1990.
then it should not affect the measurement of cleavable complex
16. Graham, E. L.. Vander. E. A. J.. and Heijneker. H. L. Size and location of
formation but should result in a reduction in cell death, SCE
the transforming region in human adenovirus type 5 DNA. Nature (Lond.), 251:
687-691. 1974.
formation, and mutant frequency.
17. Wigler, M., Silverstein. S.. Lee, L. S., Pellicer, A., Cheng, Y. C., and Axel, R.
In the present study, we developed cell lines that overexpress Bcl-2
Transfer of purified virus thymidine kinase gene to culture mouse cells. Cell, //:
to investigate the relation between etoposide-induced cleavable com
222-232, 1977.
18. Reed,J. C., Meisler, L., Tanaka, S., Cuddy, M., Yum, S., Geyer, C., and Pleasure,D.
plex formation, SCE, mutant frequency at the hprt locus, and cyto
Differential expression of bcl-2 protooncogene in neuroblastoma and other human
toxicity. Our results confirm those of Kamesaki et al. (11) by showing
lumor cell lines of neural origin. Cancer Res.. 51: 6529-6538, 1991.
that overexpression of Bcl-2 decreases etoposide-induced cytotoxicity
19. Kohn, K. W.. Ewig, R. A. G.. Erickson, L. C., and Zwelling, L. A. Measurement of
strand breaks and crosslinks by alkaline elution. In: F. C. Friedberg, F. C. Hanawalt,
but docs not interfere with drug-induced cleavable complex forma
(eds.), DNA Repair: A Laboratory Manual of Research Procedures. Vol. IB. pp.
tion. Our results provide additional insight into this process by dem
379-401. New York: Marcel Dekker, Inc., 1981.
onstrating that Bcl-2 overexpression interferes with VP-16-induced
20. Goto. K.. Masada. S.. Kano. Y.. and Sugiyama. T. Factors involved in differential
giemsa staining of sister chromatids. Chromosoma, 66: 351-359, 1978.
cytotoxicity by acting somewhere between cleavable complex forma
21. Schneider. E.. Hsiang. Y. H.. and Liu. L. F. DNA topoisomerases as anticancer drug
tion and genetic recombination events that ultimately lead to cell
targets. Adv. Pharmacol., 21: 149-183, 1990.
22. Chen, G. L., Yang, L.. Rowe, T. C., Halliga, B. D., Tewey, K. M., and Liu. L. F.
death. These results provide important support for the proposed mech
Nonintercalaclive drugs interfere with the breakage-reunion reaction of mammalian
anism of cytotoxicity and mutagenicity induced by topoisomeraseDNA topoisomerase II. J. Biol. Chem., 25V: 13560-13566, 1984.
active agents and the notion that cleavable complex formation is an 23. Spiridonidis, C. A., Chatterjee, S., Petzold. S. J., and Berger. N. A. Topoisomerase
Il-dependent and -independent mechanisms of eloposide resistance in Chinese ham
early event, whereas SCE is a downstream event on the pathway of
ster cell lines. Cancer Res.. 4V: 644-650, 1989.
etoposide-induced cytotoxicity (13). Whereas the precise mechanism
24. Pommier. Y.. Schwanz. R. E.. Zwelling. L. A.. Kerrigan. D., Mattern, M. R.,
of Bcl-2 action remains to be elucidated, our results indicate that it
Charcosset, J. Y., Jacquemin-Sablon, A., and Kohn, K. W. Reduced formation of
4034
Downloaded from cancerres.aacrjournals.org on June 15, 2017. © 1995 American Association for Cancer Research.
nei-: ANDVMMNum-.n c;iMTICALTIKATIONS
25.
26.
27.
28.
29.
protein-associated DNA Strands breaks in Chinese hamster cells resistant to topoisomerase II inhibitors. Cancer Res., -/fi: 61 1-616, 1986.
Glisson. B., Gupla. R.. Smallwood-Kentro. S., and Ross. W. Characterization of
acquired cpipodophyllotoxin resistance in a Chinese hamster ovary cell line: loss of
drug-stimulated DNA cleavage activity. Cancer Res., 46: 1934-1938, 1986.
Singh. B.. and Gupta. R. S. Mutagenic responses of thirteen anticancer drugs on
mutation induction at multipul genetic loci and on sister chromatid exchanges in
Chinese hamster ovary cells. Cancer Res.. 43: 577-584, 1983.
Pommier, Y., Kerrigan. D., Covery, J. M.. Kao-shan, C.. and Whang-Peng, J. Sister
chromatid exchanges, chromosomal aberrations and cytotoxicity produced by antitumor topoisonierase II inhibitors in sensitive (DC3F) and resistant (DC3F/9-OHE)
Chinese hamster cells. Cancer Res., 48: 512-516, 1988.
Chiou, S. K., Rao. L., and White, E. Bcl-2 blocks p53-dependent apoplosis. Mol. Cell
Biol., 14: 2556-2563, 1994.
Long, B. H., Musial. S. T.. and Brattain. M. Ci. Comparison of cytotoxicity and DNA
breakage activity of congeners of podophyllotoxin including VP-16-213 and VM 26:
a quantitative structure activity relationship. Biochemistry. 23: 1183-1188. 1984.
30. Pommier. Y.. Leteurtre. F.. Fesen, M. R.. Fujimori. A., Kohlhagen. G., and Kohn,
K. W. Cellular determination of sensitivity and resistance to DNA topoisomcrase
inhibitors. Cancer Invest., 12: 530-542, 1994.
31. Chow, K. C., and Ross, W. E. Topoisomerase-specific drug sensitivity in relation to
cell cycle progression. Mol. Cell. Biol., 7: 3119-3123, 1987.
32. Penderson-Bjergaard. J., and Philip. P. Balanced translocations involving chromo
some bands Ilq23 and 2lq22 are highly characteristics for myelodysplasia and
leukemia following therapy with cytostatic agents at DNA-topoisomerase II. Blood.
78: 1147-1148, 1991.
33. Smith, M. A., Rubinstein, L., Cazenava. L., Ungcrleider, R. S., Maurer, H. M., Hcyu,
R.. Khan. F. M., and Gehan. E. Report of the cancer therapy evaluation program
monitoring plan for secondary acute myeloid leukemia following treatment with
epipodophyllotoxins. J. Nati. Cancer Insù «S:554-558. 1993.
34. Sugita. K.. Furukawa. T.. Tsuchida, M.. Okawa. Y.. Nakazawa. S., Akatsuka. J.,
Ohira, M.. and Nishikawa. K. High frequency of etoposide (VP-l6)-relatcd secondary
leukemia in children with non-Hodgkin's lymphoma. Am. J. Pediatr. Hemalol.
Oncol.. IS: 99-104.
1993.
4035
Downloaded from cancerres.aacrjournals.org on June 15, 2017. © 1995 American Association for Cancer Research.
Inhibition of Etoposide (VP-16)-induced DNA Recombination
and Mutant Frequency by Bcl-2 Protein Overexpression
Hisako Hashimoto, Satadal Chatterjee and Nathan A. Berger
Cancer Res 1995;55:4029-4035.
Updated version
E-mail alerts
Reprints and
Subscriptions
Permissions
Access the most recent version of this article at:
http://cancerres.aacrjournals.org/content/55/18/4029
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 15, 2017. © 1995 American Association for Cancer Research.