Abstract. The responses of cells with mutated DNA repair
pathways were compared for cisplatin, radiation and
combination treatments. The knockout of the nonhomologous
endjoining (NHEJ) pathway resulted in increased radiation
sensitivity, but no change in cisplatin response in the mouse
cells and increased radiosensitivity but decreased cisplatin
sensitivity in chicken cells. The mutation of the homologous
recombination repair (HR) pathway through XRCC3 in CHO
cells resulted in increased radiation and cisplatin sensitivity
and to a lesser extent for the Rad54 knockout in the DT40
chicken cells. The combination treatments of cisplatin and
radiation showed that inhibition of the HR repair pathway
resulted in super additive effects while the inhibition of the
NHEJ pathway in DT40 had no effect. In mouse cells the
knockout of the NHEJ pathway resulted in reduced super
additivity compared to the parental cell lines. These data show
that radiation, cisplatin and combination treatment damage is
affected differently by the various DNA repair pathways, which
could have a range of effects on combination treatments in
tumour cells expressing different levels of DNA repair in the
various repair pathways.
Cisplatin was discovered to be an effective chemotherapy
agent in the treatment of cancer (1-10) and has been
extensively investigated in cultured mammalian cells. It was
shown that cisplatin could form mono and bifunctional
adducts on the DNA and that cells had the ability to remove
such adducts using DNA repair systems (10-13). One such
repair system was shown to be nucleotide excision repair,
which when mutated in xeroderma pigmentosa cells could
cause enhanced sensitivity to cisplatin and, recently, the
homologous recombinational repair pathway has also been
implicated (14-18). Now, through gene mutation and gene
knockout techniques, it is possible to block cellular DNA
repair systems which process DNA single and double strand
breaks such as excision repair and nonhomologous,
endjoining and homologous recombination repair (HR) (19-
26). A number of studies have shown that the inactivation
of DNA repair genes can result in radiosensitization
(20,22,27-30). It has also been shown that such gene
inactivation and knockouts can effect cellular responses to
cisplatin (16,17,18,30-32). We and others have shown that
combined treatment of cisplatin and radiation can result in
additive or superadditive effects, supporting the concept of
radiosensitization may well be influenced by whether similar
or different DNA repair pathways are involved in processing
cisplatin and radiation damage in DNA. In order to explore
this question further, we set out to compare the responses
to radiation, cisplatin and combined treatments in cell lines
with the following natural and knockout mutations;
nonhomologous endjoining using Ku70 and Ku80 gene
knockout cells and homologous recombination using
XRCC3 gene mutated and Rad54 knockout cells.
Materials and Methods
The cell lines used in this study are described as follows: The
mouse embryo fibroblast line (MEF) and its derivative
nonhomologous endjoining knockout (NHEJ) cell line Ku80 were
kindly donated by Dr. G.C. Li (21) and their culture details have
been previously described (22). The CHO cell line AA8 and its
derivative irsISF is a mutant of the XRCC3 gene (XRCC3-/-) and
deficient in homologous recombination (HR) DNA repair. They
were kindly donated by Dr. L. Thompson and have been previously
described in detail (19-20). The DT40 chicken cell lines were
developed by Dr. Takeda (30) and kindly donated for these
experiments. The DT40 is the parental cell line, the DT40Rad54 is
the Rad54 gene knockout cell line deficient in HR repair. The
DT40Ku70 is a knockout in the NHEJ pathway and the
DT40Rad54Ku70 is a knockout line for both HR and NHEJ
Correspondence to: G.P. Raaphorst, Medical Physics Department,
Integrated Cancer Program, The Ottawa Hospital, 501 Smyth
Road, Ottawa, Ontario, Canada K1H 8L6. Tel: (613) 737-7700 Ext.
6727, Fax: (613) 247-3507, e-mail: firstname.lastname@example.org
Cisplatin, radiation, DNA repair, mutants,
ANTICANCER RESEARCH 25: 53-58 (2005)
A Comparison of Response to Cisplatin,
Radiation and Combined Treatment for Cells
Deficient in Recombination Repair Pathways
G.P. RAAPHORST, J- MAUDE LEBLANC and L.F. LI
Integrated Cancer Program/The Ottawa Hospital, 501 Smyth Rd., Ottawa, Ontario K1H 8L6, Canada
pathways. The details of the cells and their culturing have been
previously described (30).
The CHO and mouse cell lines were grown in a mixture of 1:1
DMEM and F12 medium supplemented with 10% fetal calf serum
and 0.1 mM MEM nonessential amino acids. The cells were grown
to plateau phase, then refed and, 48 hours after that, the
experiments were performed. The plating efficiencies were 60-90%
for the AA8 and irsISF cell lines and 20-25% for the MEF and Ku80
The DT40 cell lines were cultured in DMEM/F12 (Wisent,
Montreal, Canada) containing 1% penicillin streptomycin (Wisent),
0.88% tryptose phosphate broth solution (29.5g phosphate broth/L,
Sigma, Oakville, ON, Canada), 1% chicken serum (Wisent), 10%
fetal bovine serum (Wisent) and 50 ÌM ‚-mercaptoethanol (14.3
moles/L Sigma). The cells were grown in suspension in 100-mm
petri dishes. Cells were subcultured every two days and the
experiments were done in the exponential phase of the cell growth.
The day of the experiment, the cells were counted using the
hemocytometer. After treatment, the cells were plated in 15-mm
petri dishes. The plating efficiencies of the DT40, DT40Rad 54,
DT40Ku70 and DTRad54Ku70 cells were 70-90, 50-70, 30-50 and
For cisplatin treatment, cisplatin obtained from David Bull
Canada Inc. in isotonic saline (1 mg/ml) was diluted into the
culture medium at the required concentration. For the
concentration of 1 mg/ml, the dilution factor is 1000 and has no
significant effect on the medium. These solutions are pH buffered
at 7.2 and added directly to the cells. At the end of treatment, the
solutions were removed, cultures rinsed with warm isotonic buffer
and then medium was added. Fresh solutions were used for each
experiment. For the DT40 cells grown in suspension, cells were
centrifuged and the supernatant replaced with isotonic buffer. The
cells were then centrifuged one more time and the buffer was
replaced with fresh medium.
For irradiation, cells were irradiated in 25-cm2tissue culture
flasks at room temperature using a Pantak Bipolar Series Model
HF320 X-ray, operating at 250 kVp with 1.87 mm base aluminum
filtration giving a dose rate of 168 cGy/min.
After treatment, the cells were prepared for the colony survival
assay. The mammalian cells were trypsinized, counted and plated at
numbers to give about 50 to 100 colonies per 6-cm tissue culture
plate. For the DT40 chicken lines, the cells were plated into 6-cm
petri dishes at numbers for which they formed 50-100 colonies.
Survival was assayed using the suspension colony forming assay in
DMEM/F12 media preparation containing 1% methylcellulose,
4000 centipoises (Shin-Etsu Chemical Co. Ltd. Tokyo, Japan). All
dishes were placed in a 37ÆC incubator until colonies of 50 cells or
more were visible. At this point, all colonies were stained and
counted. Each experiment was repeated three times and the error
bars represent the standard error of the mean.
The radiation response of two cell line pairs is shown in
Figure 1. For the inhibition of recombination DNA repair
pathways, both the homologous recombination repair
mutant (irsISF) and the nonhomologous endjoining
knockout (Ku80) showed greater radiosensitivity than the
parental wild-type cell lines AA8 and MEF, respectively,
indicating both recombination repair pathways as being
important in radiation damage repair.
Figure 2 shows the radiation response of four chicken cell
lines. The DT40 wild-type cell line was much more
radioresistant than the derivative mutants Rad54 (HR gene
knockout) and Rad54Ku70 (both HR and NHEJ gene
knockout). The results for the Ku70 knockout were more
ANTICANCER RESEARCH 25: 53-58 (2005)
Figure 1. The radiation response is shown for two cell line pairs consisting
of the mouse embryo fibroblast parental line (MEF) and a Ku80 knockout
line (Ku80) and a Chinese hamster parental line (AA8) and an XRCC3
mutant line (irsISF).
Figure 2. The radiation response is shown for chicken lymphocyte cells
for the parental line DT40 and Rad54 knockout line (DT40Rad54), a
Ku70 knockout line (DT40Ku70) and a double knockout of Rad54 and
complex, showing low-dose high sensitivity followed by a
radiation-resistant plateau for which the cells become more
resistant than the parental line. This response was observed
earlier (30) and may be related to the up-regulation of the
HR system when the NHEJ system is inactivated and a
small subpopulation of cells in G1 may remain sensitive if
HR is less effective in G1cells before chromosomes are
The recombination repair pathways have profoundly
different effects on cisplatin responses. Figure 3 shows that
the knockout of NHEJ for the Ku80 cell line had no effect
on cisplatin response compared to its normal parental cell
line MEF. On the other hand, the mutation in XRCC3 of
the HR repair pathway in theirsISF cell line resulted in a
large increase in cisplatin sensitivity compared to the
parental cell line AA8. These data confirm that the HR
pathway is important in repair of cisplatin damage (30,32).
Figure 4 shows the response of DT40 cells and three
knockout mutants to cisplatin. The data show that the HR
knockout (Rad54) resulted in increased cisplatin sensitivity
comparing Rad54 to the wild-type. The knockout of NHEJ for
the Ku70 cell line resulted in increased resistance to cisplatin
and the knockout of both NHEJ and HR (DT40Rad54 Ku70)
reduced this resistance compared to Ku70, but it still
remained higher than the parental cell line, DT40.
The combined treatment with cisplatin and radiation was
also evaluated in the mutant cell lines in order to determine
whether any of these mutations affected the interaction
effects of these combined treatments. The results are shown
in Table π for cisplatin given for 1h and terminated 5 min
before irradiation. While experiments were done for a wide
range of cisplatin concentrations, the data are summarized
for two concentrations representative of the data and at the
level of maximum clinical achievability. For the chicken
DT40 cell lines, the parental line DT40 and the mutants
DT40K70 and DT40Rad54Ku70 showed additive effects of
the two treatments by assessing the interaction ratios
(survival of cisplatin alone times survival from X-rays alone
divided by survival of the combined treatments). A result of
greater than 1.0 indicates super additivity and the data were
not significantly different from 1.0. For the DT40Rad54
mutant, the results were significantly greater than 1.0
showing super additivity. For the rodent cell lines, the
results for hamster CHO AA8 showed additivity for the low
cisplatin concentration and super additivity for the higher
concentration, while the irsISF mutant for HR repair
showed super additivity for both concentrations. The mouse
MEF cell line showed super additivity for both
concentrations, while the NHEJ mutant showed super
additivity for the low concentration, but just additivity for
the higher concentration. It is also of interest to note that,
in general, the rodent cell lines showed a greater interaction
effect than the chicken cell lines.
The data presented in this study clearly show that DNA
repair pathways are different for repairing damage
stemming from radiation and cisplatin treatments, as
reflected in the interesting differences observed for
Raaphorst et al: Cisplatin and Radiation Responses in Repair Mutants
Figure 4. The response of the four chicken cell lines DT40 wild-type,
DT40 Rad54 knockout, DT40Ku70 knockout and the double knockout
DT40Rad54Ku70 is shown. Cisplatin exposure was for 1 hour after which
it was removed and cells were rinsed and given fresh medium.
Figure 3. The cisplatin response is shown for mouse and Chinese hamster
parental cells MEF and AA8 and their respective mutants Ku80 and
irsISF. Cisplatin exposure was given for 1 hour after which it was removed
and cells were rinsed and given fresh medium.
radiation and cisplatin responses for the recombination
repair pathway mutant cell lines. For the NHEJ repair
pathway knockout in Ku80, there was a dramatic increase
in radiation sensitivity but no change in the cisplatin
response, clearly indicating that NHEJ is active in
processing radiation damage but not cisplatin damage. On
the other hand, the inhibition of the HR pathway through
mutation of the XRCC3 gene in theirsISF cell line showed
increased sensitivity to both radiation and cisplatin. The
increased cisplatin sensitivity was about 10-fold in
concentration and confirms earlier results that HR is
important in processing cisplatin damage (18). This would
indicate that perhaps in S-phase and in G2, where HR is
more active (30), the impact of cisplatin would be greater
and this is supported in our earlier results showing
enhanced numbers of dead S-phase cells through cell cycle
The results for the DT40 cells are very complex. The HR
knockout through Rad54 supports the results for the irsISF
cell line in that it shows increased cisplatin sensitivity.
However, the knockout of the NHEJ pathway shows
complex behavior; cells became resistant to cisplatin and
resistant to high-dose radiation. We speculate that Ku70
knockout may result in up-regulation of the HR pathway
resulting in increased resistance to cisplatin and possibly to
high-dose radiation. The low-dose radiation sensitivity may
be related to the component of cells in the radiosensitive
G1/G0cell cycle phase, where also HR may be less effective.
The double knockout DT40Rad54Ku70 showed increased
cisplatin sensitivity compared to the Ku70 knockout, but it
was still more resistant than the parental cell line. The data
in Figure 4 show that the Rad54 knockout caused much less
cisplatin sensitization than the XRCC3 mutation shown in
Figure 3. Thus, Rad54 knockout may result in only a partial
HR inhibition and possibly the additional Ku70 knockout
may still allow for some up-regulation in the HR pathway.
We and others have shown that cisplatin radiosensitization
can act through the inhibition of cellular recovery of radiation
damage (10,35,37). In addition, cisplatin responses and
radiosensitization have been shown to affect DNA repair
Our results, shown in Table π, support the observation
that cisplatin radiosensitization can be influenced by DNA
repair systems. In the DT40Rad54 HR knockout line, the
interaction ratio for cisplatin and radiation was greater than
1.0 and greater than the parental line. Correspondingly, in
the CHO cell line theirsISF mutant also showed a greater
interaction ratio than the parental line AA8. On the other
hand, in the DT40 system the Ku70 NHEJ mutant was not
significantly different than the parental line, while in the
rodent MEF system Ku80 mutant was less than the parental
line. Thus, both the rodent and the chicken lines show that
perhaps the HR repair system influences the cellular
response to combined cisplatin and radiation treatment. The
results for the NHEJ system are less clear since in the
chicken system there is no effect, while in the rodent system
ANTICANCER RESEARCH 25: 53-58 (2005)
Table I. Combined treatment of cisplatin and radiation.
Cell line Cis conc
Cis alone Survival responseInteraction
X-rays Cis + X-ray3
X = 6 Gy
X = 3 Gy
X = 3 Gy
X = 2 Gy
X = 8 Gy
X = 6 Gy
X = 8 Gy
X = 2 Gy
1Interaction ratio = (survival from cisplatin x survival from X-rays/survival from the combined treatments)
2Interaction ratio marked with * show significant difference from 1.0
3The standard deviation on the results was less than ± 10%
4Cisplatin treatment was for 1 h and was terminated 5 min before irradiation was started.
the interaction ratio is less than for the parental line. It
should be noted that, while direct comparison between all
the different cell lines is not possible because of genetic
differences, the relative results of mutant to parental in each
group is possible and these relative comparisons do show a
consistent result for the involvement of the HR repair
In summary, these data show that altered expression of
recombination repair pathways can affect responses to
cisplatin, radiation and combined treatments and this could
have clinical implications in cancer treatment where tumour
cells may have different activity in the various DNA repair
1 Sledge GW: Cisplatin and platinum analogues in breast cancer.
Semin Oncol 19: 78-82,1992.
Piver MS: Ovarian carcinoma, a decade of progress. Cancer 54:
Taylor SG, Murphy AK, Vannetzel JM et al: Randomized
comparison of neoadjuvant cisplatin and fluorouracil infusion
followed by radiation versus concomitant treatment in advanced
head and neck cancer. J Clin Oncol 12: 385-395, 1994.
Palazzi M, Calaldo I, Gramaglia A, de Toma D, Milani F and
Ravasi G: Preoperative concomitant cisplatin/VP16 and
radiotherapy in stage III non-small cell lung cancer. Int J Rad
Oncol 27: 621-625, 1993.
Zietman AL, Shipley WU and Kaufman DS: The combination
of cisplatin based chemotherapy and radiation in the treatment
of muscle-invading transitional cell cancer of the bladder. Int J
Rad Oncol 27: 161-170, 1993.
Kelland LR: Cisplatin-based anticancer agents. In: Uses of
Inorganic Chemistry in Medicine. Ed. Farrell NP. pp.109-134,
Wen H, Zu JY, Lin L and Li T: Nasopharyngeal carcinoma
(NPC) treated by hyperfractionated irradiation plus alternating
chemotherapy. Radiother Oncol 43: 587, 1997.
Teicher BA: The role of platinum complexes in combined
modality therapy. In: Chemoradiation in Cancer Therapy. Ed.
Choy H. Humana Press, Totowa New Jersey, pp. 47-63, 2003.
Canal P: Platinum compounds: pharmacokinetics and
pharmacodynamics. In: A Clinician’s Guide to Chemotherapy,
Pharmacokinetics and Pharmacodynamics Eds. Grochow LB
and Ames MM, Williams and Wilkins, Baltimore, pp. 345-373,
10 Dewit L: Combined treatment of radiation and cis-
disminedichloroplatinum (II): A review of experimental and
clinical data. Int J Rad Oncol 13: 403-426, 1987.
11 Calsou P and Salles B: Role of DNA repair in the mechanisms
of cell resistance to alkylating agents and cisplatin. Cancer
Chemother Pharmacol 32: 85-89, 1993.
12 Sancar A and Sancar GB: DNA repair enzymes. Ann Rev
Biochem 57: 29-67, 1988.
13 Sibghat-Ullah, Intinsar H, Carleton W and Sancar A: Human
nucleotide excision repair in vitro: repair of pyrimidine dimers,
psoralen and cisplatin adducts by HeLa cell-free extract.
Nuclear Acids Res 17: 4471-4484, 1989.
14 Korberle B, Masters JR, Hartley JA and Wood RD: Defective repair
of cisplatin-induced DNA damage caused by reduced XPA protein
in testicular germ cell tumours. Current Biol 9: 273-276, 1999.
15 Zhen W, Evans MK, Haggerty CM and Bohr VA: Deficient
gene specific repair of cisplatin lesions in XP and FA cell lines.
Carcinogenesis 14: 919-924,1993.
16 Tanaka K, Nakatsu Y, Sayo M, Kuraoka I, Matsuda T,
Kobayashi T, Murai H and Nakane H: Nucleotide excision
repair defect and carcinogenesis in XPA-knocked out mice. J
Cell Mol Biochem, Supplement 21A: 272, 1995.
17 Li Q, Yu JJ, Mu C, Yunmbam MK, Slavsky D, Cross CL,
Bruton FB and Reed E: Association between the level of ERCC
expression and the repair of cisplatin-induced DNA damage in
human ovarian cancer cells. Anticancer Res 20: 645-652, 2000.
18 Zdraveski ZZ, Mello JA, Marinus MG and Essigmann JM:
Multiple pathways of recombination define cellular responses
to cisplatin. Chem Biol 7: 39-50, 2000.
19 Liu N, Lamerdin JE, Tebbs RS, Schild D, Tucker JD, Shen MR,
Brookman KW, Siciliano MJ, Walter CA, Fan W, Narayana LS,
Zhou ZQ, Adamson AW, Sorensen KJ, Chen DJ, Jones NJ and
Thompson LH: XRCC2 and XRCC3 new human Rad 51-family
members promote chromosome stability and protect against
DNA cross links and other damages. Mol Cell 1: 783-793, 1998.
20 Pierce AJ, Johnson RD, Thompson LH and Jasin M: XRCC3
promotes homology-directed repair of DNA damage in
mammalian cells. Genes Develop 13: 2633-2638, 1999.
21 Burgman P, Ouyang H, Peterson S, Chen DJ and Li GC: Heat
inactivation of Ku autoantigen: possible role in hyperthermic
radiosensitization. Cancer Res 57: 2847-2850, 1997.
22 Myint WK, Ng C and Raaphorst GP: Examining the non-
homologous repair process in cisplatin and radiation
treatments. Int J Rad Biol 78: 417-424, 2002.
23 Sobol RW, Norton JK, Kuhn R et al: Requirement of
mammalian DNA polymerase ‚ in base excision repair. Nature
379: 183-186, 1996.
24 Frosina G, Fortini P, Rossi O, Carrozzino F, Raspaglio G, Cox
LS, Lane DP, Abbondandolo A and Dogliotti E: Two pathways
for base excision repair in mammalian cells. J Biol Chem 271:
25 de Vries A and van Sleeg H: XPA knockout mice. Cancer Biol
26 Hoeijmakers JHJ: Human nucleotide excision repair
syndromes: molecular clues to unexpected intricacies. Eur J
Cancer 30A: 1912-1921, 1994.
27 Liang F, Han M, Romanienko PJ and Jasin M: Homology
directed repair a major repair pathway in mammalian cells.
Proc Nat Acad Science 95: 5172-5177, 1998.
28 Wachsberge PR, Li WH, Guo M, Chen D, Cheong N, Ling CC, Li
G and Iliakis G: Rejoining of DNA double strand breaks in Ku80-
deficient mouse fibroblasts. Radiation Res 151: 398-407, 1999.
29 Utsumi H, Tano K, Takata M, Takeda S and Elkind MM:
Requirement for repair of DNA double strand breaks by
homologous recombination in split-dose recovery. Radiation Res
155: 68-686, 2001.
30 Takata M, Sasaki MS, Sonoda E, Monison C, Hasimoto M,
Utsumi H, Yamaguchi Y, Wai I, Shinohara A and Takeda S:
Homologous recombination and nonhomologous end-joining
pathways of DNA double-strand break repair have overlapping
roles in the maintenance of chromosomal integrity in vertebrate
cells. EMBO J 17: 5497-5508,1998.
Raaphorst et al: Cisplatin and Radiation Responses in Repair Mutants