Mutagcncsis vol.13 no.5 pp.499-505, 1998
Caffeine effect on the mitotic delay induced by G2 treatment with
UVC or mitomycin C
A. Franchitto*, P. Pichierri+, P. Mosesso and F. Palitti*
University degh Studi della Tuscia, Dipartimento di Agrobiologia ed
Agrochimica, Via San Camillo de Lellis, 1-01100 Viterbo, Italy
*The first two authors contributed equally to this work.
It is well established that DNA lesions trigger cell cycle
check-points causing a mitotic delay that is required for
their repair before cells enter the mitotic phase. Caffeine,
in some cases, can remove this delay and consequently
potentiates the yield of induced chromosome aberrations.
The objective of this study was to test the effect of a
G2 treatment with S-dependent agents (UV light and
mitomycin C) on the cell kinetics of a G2 cell population
and evaluate whether post-treatments with caffeine could
modulate removal of the expected cell cycle delay. Cell
kinetics were monitored by analysing the mitotic index (MI)
values in combination with the 5-bromo-2'-deoxyuridine
(BrdUrd) labelling technique. Chinese hamster fibroblast
cultures (AA8) were treated in G2 phase of the cell cycle
with 8 and 15 J/m2 UV light or 0.1 and 0.6 Hg/ml mitomycin
C for 13 h. Post-treatments with caffeine were performed
at dose levels and recovery times where the mitotic indices
were substantially reduced. The results obtained showed
that both UV light and mitomycin C induced a G2 arrest,
as indicated by MI values and the absence of BrdUrdlabelled metaphases. For UV light the G2 block was
observed at lower and higher dose levels after 1.5 h, while
for mitomycin C it was observed only at the higher dose
level after 1 h. However, in both cases the block lasted
~1 h, after which, even though slowed down, the cell
population entered mitosis, as indicated by increased MI
values. This block was not removed by caffeine posttreatment In contrast, caffeine G2 post-treatment was
able to remove G2 arrest induced by Gl-S treatments.
Accordingly, our results suggest that both UV light- and
mitomycin C-induced damage must be processed during S
phase to allow caffeine to remove induced G2 blocks.
Introduction
It is known that treatments with mutagens induce DNA damage
in living cells leading to the activation of cell cycle checkpoint control mechanisms in different phases of the cell cycle
(Paulovich et al, 1997). Both prokaryotic and eukaryotic cells
delay cell division following damage to DNA. The threat
represented by damaged DNA is particularly serious during
the two major cell cycle events: DNA replication and mitosis.
Delays in progression from both Gl to S and from G2 to M
occur in most organisms (Tolmach et al, 1977; Painter and
Young, 1980; Lau and Pardee, 1982; Kaufmann et al., 1991;
O'Connor et al, 1992).
It is thought that these cell cycle check-points permit repair
of damaged DNA before replication or segregation of the
affected chromosomes (Hartwell and Weinert, 1989) and should
thus enhance cell survival and limit propagation of heritable
genetic errors (Weinart and Hartwell, 1988). DNA damage
may also induce apoptosis as an alternative strategy to prevent
propagation of damaged cells. The tumour suppressor protein
and transcription factor p53 is involved in both Gl arrest and
nucleotide excision repair (Livingstone et al., 1992; Dulic
et al., 1994; Abrahams et al, 1995). Through sequence-specific
DNA binding p53 can modulate transcription of a variety of
target genes following DNA damage (Cox and Lane, 1995);
such genes are likely to be involved in Gl arrest or apoptotic
cell death (Kastan et al, 1995).
In addition to this check-point, eukaryotic cells are able to
arrest in the G2 phase of the cell cycle after DNA damage
(Bernhard et al, 1994; Kaufmann, 1995; Murnane, 1995).
Arrest in the G2 phase is accompanied by an increase in
phosphorylation of Thrl4 and Tyrl5 of the catalytic p34
subunit of the cyclin-CdK complex and by a decrease in CdK
1 kinase activity in hamster cells as well as in human cells
(Hain et al, 1993, 1994; Kharbanda et al, 1994). For Chinese
hamster cells inhibition of phosphorylation has also been
shown for p40, a Cdk 2-like protein (Hain et al, 1994).
The G2 phase arrest is temporary and can be released
prematurely by caffeine, reducing the mitotic delay (LUckeHunle et al, 1983; Zampetti-Bossler and Scott, 1985; Darroudi
and Natarajan, 1987; Hain et al, 1994; Russell et al, 1995).
During the mitotic delay some of the induced damage that
could lead to chromosomal aberrations is repaired. Cancellation
of delay reduces the time available for such repair, so that
cells enter division with a higher proportion of DNA damage
than would normally be the case (Painter, 1980).
The ability of caffeine to potentiate the induction of chromosome aberrations is due to inhibition of repair of DNA damage
(Natarajan and Obe, 1983). However, it is also suggested that
mitigated delay and enhanced aberration frequency are not
necessarily causally related (Harvey and Savage, 1994). It is
usual to distinguish between two main types of mechanisms
for the production of chromatid-type aberrations; when induced
by UV, alkylating agents and many other chemicals, aberrations
arise by an S phase-dependent mechanism. This means that
the DNA lesions caused by these agents do not give rise to
chromatid-type aberrations before they have been replicated.
Thus, DNA lesions induced by S phase-dependent agents in
G2 do not result in chromatid aberrations until the second
cell cycle after their induction. Chromatid aberrations may,
however, be observed in the first mitosis after treatment when
the lesions have been induced during Gl or S in unreplicated
chromosomes or chromosome sites through 'misreplication'
(Evans and Scott, 1969). A caffeine post-treatment potentiates
the frequency of induced chromosome aberrations by S-phase
dependent agents, provided that the treatment has been done
in Gl or S phase of the cell cycle.
The aim of this work has been to test if a caffeine post-
T o whom correspondence should be addressed. Tel: +39 761 357 206; Fax : +39 761 357 242; Email: [email protected]
© UK Environmental Mutagen Society/Oxford University Press 1998
499
A.Franchltto et al
MMC
I 1,5hrsBrdUrd
1.5 h
+
Fix
Fix
Fix
Fix
colch icine
1 hr colch.
2 hrs colcti. 2 hrs colcti. 2 hre colch.
MMC
Washes
5 mM caffeine
I 1,5
1.5 hrs
h BrdUrd +
•
nolrd
colchicine
Fix
uvc
'
1,5 hrs BrdUrd+
colchicine
Fix
Fix
Fix
Fix
Fix
1 hr cdcrt. 1 hr colch. 2 hrs colch. 2 hrs colch.
Fix
UVC
J7
5 mM caffeine
i BrdUrd
colchicine
Fix
1 hr
r
1 hr colch.
Fig. 1. Experimental scheme of the G2 treatments.
treatment is also able to remove the G2 arrest caused by a G2
treatment with S phase-dependent agents.
Materials and methods
Cell cultures
Chinese hamster fibroblasts (AA8), kindly provided by Prof. F.Cortes (University of Sevilla, Sevilla, Spain) were cultured in McCoy's 5 medium, supplemented with 15% newborn calf serum and 2% L-glutamine and incubated at
37°C in a 5% CO} atmosphere (100% humidity nominal). Cells were
subcultured every 3-4 days at 3X lCr' cells/25 cm2 flask.
Approximately 18-20 h before treatment an appropriate number of 60 mm
Petri dishes were prepared from a smgle pool of cells. Each Petri dish was
seeded with 3X1O5 cells.
UV and MMC G2 treatment
A scheme of the experimental procedure is shown in Figure 1.
Immediately before treatment the cultures were shaken to remove mitotic
cells to avoid any contamination of G2 cells with mitotic cells. Cultures were
then exposed to 8 or 15 J/m2 UV without any medium, after one wash with
phosphate-buffered saline (PBS), or for 1.5 h to 0.1 or 0.6 ug/ml mitomycin
C (MMC) in fresh complete medium with 30 (ig/ml 5-bromo-2'-deoxyuridine
(BrdUrd). The UV light was generated by a Spectroline UV lamp (254 run).
At the end of treatment UV-treated cultures were resuspended in fresh complete
medium containing 30 Ug/ml BrdUrd, while MMC-treated cultures were
washed twice and resuspended in fresh complete medium. In both treatments
BrdUrd pulse labelling lasted 1.5 h.
Post-treatments with caffeine (1 h at a final concentration of 5 mM) were
performed only at dose levels where the mitotic indices (Mis) were substantially
reduced to check the sensitivity of the G2 block to caffeine. Colchicine
(10 (iM) was added to the cell cultures at the appropriate time before fixing
as described in Figure 1. Cultures were interrupted at intervals of 1-2 h over
a maximum period of 8 h and cytological microscope slides prepared according
to standard procedures.
500
Control cultures were handled under identical conditions.The experiments
were repeated three times and the results are representative of these repeats.
Staining of slides
At least three slides per test point were prepared. One set was stained with
3% Giemsa for 5 min to be scored for MI. The remaining slides were
processed using immunocytogenetic techniques with anti-BrdUrd antibodies
conjugated with the fluorochrome FTTC. The slides were denatured for 1 min
in 10 mM NaOH, 70% ethanol, dehydrated in a 70, 90 and 100% ethanol
series and air dried. The slides were then incubated in a moist chamber for
30 min with 100 (il/slide mouse anti-BrdUrd antibody (BoheringerMannheaim) diluted 1:100 in immunological buffer (PBS, 0.5% bovine serum
albumin, 0.5% Tween 20) under a 24X50 covershp. After incubation the
slides were washed three times with PBS and subsequently incubated with
100 (il/slide goat anti-mouse IgG-FITC antibody (Boheringer-Mannheaim)
diluted 5:100 in immunological buffer for 30 min. After three washes in PBS
and dehydration in ethanol, the slides were embedded in Vectashield mounting
medium (Vector Laboratories) containing 0.3 ug/ml propidium iodide (PI).
Scoring of slides
Mitotic indices were determined from 1000 cells analysed with a 100X oil
immersion objective.
BrdUrd labelling of mitotic cells was evaluated using a Zeiss (Axiophot)
fluorescence microscope equipped with single and dual bandpass filters for
FTTC and PI and a CCD camera (Photometrix) operated by IPLab spectrum
software. Images were captured through the above-mentioned system and
analysed on a 21 inch high resolution monitor.
For each test point, from at least 100 metaphases, metaphases were classified
as unlabelled (G2 cells) or early, mid or late replicating S phase cells according
to the BrdUrd labelling pattern, as also described by Savage in human
lymphocytes (Savage and Prasad, 1984). In our specific case we used
chromosome X as a marker to allocate each metaphase to the three different
S phase stages.
In Chinese hamster, chromosome X bears a short euchromatic arm which
is early replicating and a long heterochromatic arm which is late replicating.
According to these features, we allocated single metaphases to the different
Caffeine, UVC and mitomycm v.
Fig. 2. The different banding patterns of AA8 chromosomes in (a) late S phase, (b) mid S phase and (c) early S phase. The arrows indicate X chromosome
labelling in the three different S phase stages. It is clearly visible that in late and early S phase the two arms of the X chromosome are labelled by turns,
while in mid S phase the labelling is poorly distributed along trie two arms.
compartments as follows: (i) late S phase, the whole long arm of chromosome
X is labelled (Figure 2a); (ii) mid S phase, chromosome X is uniformly
labelled in both arms (Figure 2b); (iii) early S phase, the short arm of
chromosome X is labelled (Figure 2c).
Results
Analysis of mitotic indices
The MI data have been plotted and reported in Figure 3a and
b for UV- and MMC-treated cultures respectively. Analysis of
the results shows that at the higher dose level for UV the MI
was depressed to ~40% of the relevant control value, while at
the lower dose level the decrease in MI started 1.5 h after
treatment. The maximum inhibition of MI represented by a
decrease of ~80% of the relevant control value was observed
at both dose levels 2.5 h after treatment. The block lasted
~1 h, after which MI started to increase slowly. At the lower
dose level, the MI values reached the relevant control value
5.5 h later, while at the higher dose level it reached only 50%
of that value.
For MMC-treated cultures a marked depression in MI was
only seen at the higher dose level, with a drop to 80% of the
relevant control value 1 h later. This value increased slightly
1 h later, to 50% of the control value, after which it remained
constant until the end of observation 8 h later, while control
values progressively increased. At the lower dose level the MI
values were in line with the control values up to 2 h, after
which it dropped to ~50% of the control value in the next 2 h.
This represented the maximum inhibition of MI at this dose
level. From this point and until the end of observation (8 h)
the MI values, even though slowed, followed the same trend
as the control values. The fluctuations in MI in the UVC and
MMC controls almost certainly reflect perturbations introduced
by the mitotic shake-off, the medium change and, in the case
of the MMC experiment, the additional washes after a period
corresponding to the treatment and the differences in the
control frequencies between the UV and MMC experiments
is due to different handling of the cultures in the two experiments. The increase in the MI with time in the MMC controls
is, probably, due to a partial synchronization induced by the
mitotic shake-off and by the vigorous washes used to remove
MMC in the treated cultures. This trend in the MI values of
MMC controls has been observed in all three repeats of the
experiment and so the mitotic depression and the recovery
seen following MMC and UVC treatments should not be
considered an artefact.
Analysis of BrdUrd labelling
The percentages of labelled metaphases (early, mid and late
replicating S phases) at different recovery times have been
plotted and are presented in Figure 4a and b for UV- and
MMC-treated cultures respectively. The results show that the
appearance of labelled mitotic cells in the control was observed
2.5 h after pulse labelling, while for the UV-treated cells it
was delayed to 5.5 h. At this time the percentage of labelled
metaphases was slightly higher at 8 J/m2 compared with
15 J/m2, showing that this delay is dose dependent. After 5.5 h
this 'labelling index' increases with time but does not reach
the control values, at least in the time range analysed. Figure
501
A.FranchJtto et aL
14
12 -
- O - Ctrl
- O - 1 £ hrs 0.1 p&r* MMC
- A - 1 Jt hrs0J w)rtrt MMC
10 -
s
10
recovery (hrs)
recovery (hrs)
Fig. 3. Time course of the MI after G2 exposure to (a) UVC radiation and (b) MMC. After mitotic shake-off cells were treated with UVC or MMC, then
challenged in fresh complete medium and fixed at various intervals to determine the percent of mitotic cells. For each test point 1000 cells were analysed.
100
ca.
I
1
|
I
1
4
6
8
10
recovery (hrs)
recovery (hrs)
Fig. 4. Percentage of labelled metaphases after different recovery times of (a) UVC- and (b) MMC-treated cultures. Cultures were pulse labelled with
30 ng/ml BrdUrd for 1.5 h after UV radiation or during the MMC treatmenL Slides, prepared for each experimental point as described in Materials and
methods, were scored and 100 metaphascs were randomly analysed and subdivided into G2 (unlabelled metaphases) or S phase (labelled metaphases).
5a-c shows the distribution of labelled metaphases in the three
classes according to the BrdUrd labelling pattern (early, mid
and late S phase) following treatment with 8 and 15 J/m2. The
results show that early S phase is strongly delayed at both
dose levels compared with the control culture, while the
appearance of late S phase is slowed in a dose-dependent
manner.
For the MMC treatment, when the pooled data are analysed
(Figure 4b) a delay in the appearance of labelled metaphases
was observed at the higher dose level only. This delay appears
to be caused mainly by early S phase, which was markedly
slowed (Figure 6b).
Caffeine effect on UV- and MMC-induced mitotic delay
In order to study the effect of caffeine on the prevention of
the delay in G2 phase exit, we post-treated AA8 cells for 1 h
502
with 5 mM caffeine after treatment with UV or MMC. The
results obtained, reported in Table I, show that caffeine is
unable to remove the G2 block caused by UV irradiation and
MMC treatment when performed in the G2 phase, as indicated
by Ml values.
Discussion
In the present study we observed that both UVC and MMC
G2 treatments induced a delay in the G2 phase of Chinese
hamster fibroblasts (AA8). This delay has been also described
in other cell lines after UVB treatment (Hain et aL, 1996) and,
to a lesser extent, after UVA treatment (Banrud et aL, 1995).
Furthermore, the length of the G2 block depends on the dose
levels used, as previously described by Orren et al. (1995).
For both UV and MMC treatments, the arrest of cells is
Caffeine, UVC and mitomycln C
uw 80 -
- O - LR
- D - MR
- A - ER
t
/ ^^ /
/
\ /
6040 20 0 -
A
I
/
O
1
—a—oi
i
i
i
i
i
recovery (hn)
100
recovery (hre)
recovery (hre)
Fig. 5. Distribution of early (ER), mid (MR) and late S phase (LR) of (a) control, (b) 8 and (c) 15 J/m2 UVC-irradiated cells. Metaphases scored as S phase,
as described in Figure 4, were classified into ER, MR or LR according to the criteria described in Materials and methods.
immediate at higher dose levels, while at lower doses only
UV is able to induce G2 block, which is shifted by 1.5 h
compared with the higher dose level. This different behaviour
may be explained not only by different cytotoxic effects, but
also by different modulation of the block of S phase cells. In
fact, while for MMC the block at the higher dose level causes
a shift of 2 h in the appearance of early S phase cells, followed
by a rapid recovery, in the UV treatment at both dose levels,
early S phase cells and, to a lesser extent, mid S phase, even
though delayed as in MMC-treated cells, do not recover from
the block. In this case late S phase cells were slowed in a
dose-dependent manner. Nevertheless, the block was partially
reversible in both treatments and our results show that cells
were able to leave G2 phase gradually during the analysed
recovery periods. This suggests that early S cells are most
damaged by the treatment. Finally, our experimental data show
that a 5 mM caffeine post-treatment is unable to remove the
G2 delay caused by G2 treatment with MMC or UVC. That
this cell line is insensitive to caffeine can be excluded, as we
have treated AA8 cells in Gl-S phase with MMC or UV and
post-treated with caffeine in the G2 phase. In this experiment
(data not shown) it was confirmed that G2 arrest was removed
by a G2 caffeine post-treatment. It is not still clear how
caffeine removes the G2 arrest. Some authors believe that
caffeine acts predominantly in two ways: increasing the activity
of cdc2 kinase (Hain et al., 1994) or inhibiting DNA repair
and consequently causing premature entry into mitosis (Link
et al., 1995). Moreover, caffeine shows a high affinity for
single-stranded DNA and might mask damaged sites by
upstream blocking of the signalling pathway that links DNA
repair to the G2 check-point (Weibezahn and Coquerelle,
1986). In fact, caffeine has been shown in vitro to alter the
503
A.Franchltto et aL
recovery (hrs)
100
100
recovery (hrs)
recovery (hrs)
Fig. 6. Distribution of early (ER), mid (MR) and late S phase (LR) of (a) control, (b) 0.1 and (c) 0.6 jig/ml MMC-treated cells. Metaphases were classified
into ER, MR or LR as described in Figure 5.
ability of certain damage recognition proteins (Uvr A and
photolyase) to find and bind sites of UV damage in DNA in
bacterial extracts (Selby and Sancar, 1990).
Also, caffeine might interfere with DNA repair and the G2
check-point by inhibiting topoisomerases I and II (Shin et al.,
1990). Indirect molecular involvement of topoisomerases in
DNA repair has not yet been demonstrated (Stevsner and Bohr,
1993), but it has recently been proved that topoisomerase II
is involved in G2 check-point control (Downes et al., 1994).
Our data show that even if the G2 block is induced by G2
treatment with an S phase-dependent agent, its removal by
caffeine is dependent on 'processing' of the lesions through S
phase, in order that caffeine recognizes it. Actually, we cannot
exclude the possibility that G2 arrest, induced by UVC or
MMC, is not only dependent on DNA damage, but also on
injury to other cellular structures (Schneidermann et al., 1996).
504
I. caffeine effect on the MI of UVC- or MMC-treated G2 cells
G2 treatment
MI (%)
Control
0.1 (ig/mlMMC
0.6 (ig/ml MMC
Control + 5 mM caffeine
0.1 |ig/ml MMC + 5 mM caffeine
0.6 ng/ml MMC + 5 mM caffeine
Control
8 J/m2
15 J/m2
Control + 5 mM caffeine
8 J/m2 + 5 mM caffeine
15 J/m2 + 5 mM caffeine
4.9
4.9
2.7
1.5
2
1.3
3.3
0.5
0.7
2
0.1
0.5
Caffeine, UVC and mltomydn C
However, the UVC radiation doses used in this work essentially
cause DNA damage.
In conclusion, these data show that G2 treatment with UVC
or MMC is able to cause a G2 arrest and that caffeine does
not remove this G2 block, whereas it is able to remove a G2
arrest caused by Gl or S treatment with the same agent.
Therefore, it can be concluded that, in the case of the tested
S phase-dependent agents, in order to show a caffeine effect
the induced DNA lesions have to pass through S phase
Acknowledgements
This paper is dedicated to Professor Angelo Carere for his 60th birthday. This
work was supported by MURST (40%) and EU contract no. F14P-CT95-OO01.
References
AbrahamsJ'J., Schouten,R., Van Laar.T., Houweling,A., Terleth.C. and Van
der Eb.AJ. (1995) Different regulation of p53 stability in UV-irradiated
normal and DNA repair deficient human cells. Mutat. Res., 336, 169-180.
Banrud,H., Stokke.T., MoanJ. and Berg.K. (1995) S phase arrest and induction
of multjnucleated cells after exposure to ultraviolet radiation.
Carcinogenesis, 16, 1087-1094.
Bemhar^EJ., McKenna,W.G. and Muschel,RJ. (1994) Cyclin expression and
G2-phase delay after irradiation. Radiat. Res., 138, S64-S67.
CoxX-S., and Lane.D.P. (1995) Tumor suppressor kinases and clamps: how
p53 regulates the cell cycle in response to DNA damage. BioEssays, 17,
501-508.
Darroudi J\ and Natarajan^A.T. (1987) Cytological characterization of Chinese
hamster ovary X-ray sensitive mutant cells xrs 5 and xrs 6. I. Induction of
chromosomal aberrations by X-irradiation and its modulation with 3aminobenzamide and caffeine. Mutat. Res., 177, 133-148.
Downes.C.S., Clarke,DJ., MullingerAM., Gime'nez-AbianJ.F., Creighton,
A.M. and Johnson,R.T. (1994) A topoisomerase II-dependent G2 cycle
checkpoint in mammalian cells. Nature, 372, 467-470.
Dulic.V, Kaufmann.W.K., Wilson.SJ., Tlsty.T.D., LeesX"., HarperJ.W.,
EUedge.SJ. and Reed,S.I. (1994) p53-dependent inhibition of the cyclindependent kmase activities in human fibroblast during radiation-induced
Gl arrest. Cell, 76, 1013-1023.
Evans.H.J. and ScottJ). (1969) The induction of chromosome aberrations by
nitrogen mustard and its dependence on DNA synthesis. Proc. R. Soc.
Lond. B Bwl. Sci., 173, 491-512.
HainJ., Crompton,N.E.A., Burkart,W. and Jaussi.R. (1993) Caffeine release
of radiation induced S and G2 phase arrest in V79 hamster cells: increase of
historic mRNA levels and p34cdc2 activation. Cancer Res., 53, 1507-1510.
HainJ., Jaussi,R. and W(lrgler,F.E. (1994) Effects of ionizing radiation and
caffeine treatment on cyclin dependent kinase complexes in V79 hamster
cells. Cell Signalling, 6, 539-550.
HainJ., Weller,E.M., Jung,T. and Burkart,W. (1996) Effects of ionizing- and
UVB-radiation on proteins controlling cell cycle progression in human
cells: comparison of the MCF-7 adenocarcinoma and the SCL-2 squamous
cell carcinoma cell line. Int. J. Radiat. Biol., 70, 261-271.
Hartwell.L.H. and Weinert,T.A. (1989) Checkpoints: controls that ensure the
order of cell cycle events. Science, 246, 629-634.
Harvey.A.N. and SavageJ.R.K. (1994) A case of caffeine-mediated
cancellation of mitotic delay without enhanced breakage in V79 cells.
Mutat. Res., 304, 203-209.
Kastan.M.B., Canman.C.E. and Leonard.C.J. (1995) p53, cell cycle control
and apoptosis: implication for cancer. Cancer Metastasis Rev., 14, 3—15.
Kharbanda,S., SaleemA, Datta,R., Yuan,Z.M., Weichselbaum,R. and KufeJ).
(1994) Ionizing radiation induces rapid tyrosine phosphorylation of p34cdc2.
Cancer Res., 54, 1412-1414.
Kaufmann.W.K. (1995) Cell cycle checkpoints and DNA repair preserve the
stability of human genome. Cancer Metastasis Rev., 14, 31-41.
Kaufmann.W.K., BoyerJ.C, Estabrooks.L.L. and Wilson.SJ. (1991) Inhibition
of replicon initiation in human cells following stabilization of
topoisomerase-DNA cleavable complexes. Mol. Cell Biol., 11, 3711-3718.
Lau,C.C. and Pardee.A.B. (1982) Mechanism by which caffeine potentiates
lethality of nitrogen mustard. Proc. Natl Acad. Sci. USA, 79, 2942-2946.
Link,CJ., Evans,M.K-, CookJ.A., Muldoonji., Stevsner.T. and Bohr.V.A.
(1995) Caffeine inhibits gene-specific repair of UV-induced DNA damage
in hamster cells and in human xeroderma pigmentosum group C cells.
Carcinogenesis, 16, 1149-1155.
LivingstoneX-R-. White.A., SprouseJ., Livanos,E., Jacks.T. and Tlsty.T.D.
(1992) Altered cell cycle arrest and gene amplification potential accompany
loss of wild-type p53. Cell, 70, 923-935.
LUcke-Huhle.C, HicberX. and Wegner,R.D. (1983) Caffeine-mediated release
of alpha-irradiation-induced G2 arrest increases the yield of chromosome
aberrations. Int. J. Radiat. Biol., 43, 123-132.
MurnaneJ.P. (1995) Cell cycle regulation in response to DNA damage: a
historical perpective. Cancer Metastasis Rev., 14, 17-29.
NatarajanAT. and Obe.G. (1983) Influence of DNA repair on radiationinduced chromosomal aberrations in human peripheral lymphocytes. In
Ishihara,N. and Sasaki,M.S. (eds), Radiation Induced Chromosome Damage
in Man. Alan R.Liss, New York, NY, pp. 127-140.
O'ConnorJ'.M., Ferris.D.K., White.G.A., PinesJ., Hunter.T., Longo.D.L. and
Kohn.K.W. (1992) Relationships between cdc2 kinase, DNA cross-linking,
and cell cycle perturbations induced by nitrogen mustard. Cell Growth
Differential., 3, 43-52.
Orren.D.K., Petersen.L.N. and Bohr.V.A. (1995) A UV-responsive G2
checkpoint in rodent cells. Mol. Cell. Biol.. 15. 3722-3730.
Painter.R B. (1980) Effect of caffeine of DNA synthesis in irradiated and
unirradiated mammalian cells. J. Mol. Biol., 143, 289-301.
Painter.R.B. and Young.B.R. (1980) Radiosensitivity in ataxia-telangiectasia:
a new explanation. Proc. Natl Acad. Sci. USA, 77, 7315-7317.
Paulovich.A.G., JoczyskiJD.P. and Hartwell.L.H. (1997) When checkpoints
fail. Cell, 88, 315-321.
Russell.KJ., Weins.L.W., Demers.G.W., Galloway,D.A., Plon.S.E. and
Groudine,M. (1995) Abrogation of the G2 checkpoint results in differential
radiosensitization of Gl checkpoint-deficient and Gl checkpoint-competent
cells. Cancer Res., 55, 1639-1642.
SavageJ.R.K. and Prasad.R. (1984) Cytological sub-division of the S-phase
of human cells in asynchronous culture J. Med. Genet., 21, 204-212
SchneidermarvM.H., Schneiderman.G.S., Muhlmann-Diai^M.C. and Bedford,
J.S. (1996) The presence of DNA breaks and the formation of chromatid
aberrations after incorporation of 125IdUrd may be necessary but are not
sufficient to block cell cycle progression in G2 phase. Radiat. Res., 145,
17-23.
Selby.C.P. and SancarA (1990) Molecular mechanism of DNA repair
inhibition by caffeine. Proc. Natl Acad. Sci. USA, 87, 3522-3525.
Shin,G.G., StrayerJ.M., Waini^M.A. and Snapkajt.M. (1990) Rapid evaluation
of topoisomerase inhibitors: caffeine inhibition of topoisomerases in vivo.
Teratogen. Carcinogen. Mutagen., 10, 41-52.
Stevsner.T. and Bohr.V.A. (1993) Studies on the role of topoisomerases
in general, gene- and strand-specific DNA repair. Carcinogenesis, 14,
1841-1850.
Tolmach,LJ., Jones.R.W. and BusseJ>.M. (1977) The action of caffeine on
X-irradiated HeLa cells. I. Delayed inhibition of DNA synthesis. Radiat.
Res., 71, 653-665.
Weibezahn.K.F. and Coqucrelle.T.M. (1986) Relationship between double
strand break rejoining and G2 block formation in V79 cells. Radiat. Environ.
Biophys., 25, 13-21.
Weinert,TA and HartweUX-H. (1988) The RAD9 gene controls the cell cycle
response to DNA damage in Saccharomyces cerevisiae. Science, 241,
317-322.
Zampetti-Bosseler.F. and Scott,D. (1985) The effect of caffeine on X-rayinduced mitotic delay in normal human and ataxia-telangiectasia fibroblasts.
Mutat. Res., 143, 251-256.
Received on 15 September 1997; accepted on 30 March 1998
505