Full text of DOI:10.1093/carcin/22.4.661

Carcinogenesis vol.22 no.4 pp.661–664, 2001
Repair of sulfur mustard-induced DNA damage in mammalian
cells measured by a host cell reactivation assay
Z.Matijasevic1, M.L.Precopio, J.E.Snyder and
D.B.Ludlum
However, the toxicity and vesicating properties of monofunc-
tional derivatives of sulfur mustard, such as 2-chloroethyl
ethyl sulfide (CEES), which generate similar monoadducts but
do not form cross-links (2), suggest that monoadducts also
contribute to the biological effects of sulfur mustard.
Department of Biochemistry and Molecular Pharmacology, University of
Massachusetts Medical School, Worcester, MA 01655-0126, USA
1To whom correspondence should be addressed
Previously, the cellular repair of SM-damaged DNA has
been demonstrated either by measuring the disappearance of
alkyl groups from DNA (7,8) or by monitoring the occurrence
of non-semiconservative DNA synthesis (‘repair synthesis’) in
cells exposed to sulfur mustard (9). The removal of DNA
cross-links specifically has been demonstrated by several
investigators in both E.coli and mammalian cells (3,8,10–12).
However, the specific cellular repair pathways and enzymes
that act on SM-induced DNA adducts have not been completely
established. In vitro studies have shown that bacterial 3-
alkyl adenine DNA glycosylase II releases both of the SM
monoadducts, 7HETEG and 3HETEA, from SM-modified
DNA indicating that base excision repair may play a role in
repairing sulfur mustard lesions (13). The involvement of
another repair pathway in eukaryotic organisms, nucleotide
excision repair (NER), has been suggested by the study of
Kircher et al. (14) who have shown that yeast mutants deficient
in nucleotide excision repair are much more sensitive to sulfur
mustard than wild-type cells.
Although the biochemical studies mentioned above indicate
that DNA repair processes act on SM-modified DNA, they do
not demonstrate whether or not the damaged DNA has been
restored to a functional state. To address this issue, other
investigators have used a variety of host cell reactivation
assays to demonstrate functional repair of DNA after damage
by antitumor agents (15–18). In this manuscript, we describe
the use of a dual luciferase reporter assay to demonstrate
cellular repair of mustard-induced DNA damage.
We have first established that NER-competent Chinese
hamster ovary cells are more able to withstand the toxic effects
of SM than are NER-deficient cells. Then, using the dual
luciferase host cell reactivation assay, we have shown that
NER-competent cells are able to repair SM-damaged reporter
plasmid and bring luciferase expression from the plasmid to
higher levels than in NER-deficient cells, thus correlating
DNA repair with SM toxicity.
We have also used the host cell reactivation assay to examine
cellular repair of the damage caused by the single armed
mustard, CEES. As shown in Figure 1, CEES forms monoad-
ducts that are very similar to those formed by SM, thus
allowing the effects of monoadducts on survival and repair to
be examined independently from the effects of cross-links.
These studies have shown that the toxicity of CEES is also
decreased in NER-competent cells in comparison with NER-
deficient cells. Luciferase expression from plasmid damaged
by CEES is enhanced in NER-competent cells compared
with NER-deficient cells, indicating that the monofunctional
mustard adducts formed by CEES are cytotoxic and are also
substrates for NER. From these results, we conclude that this
DNA damage is thought to be the initial event that causes
sulfur mustard (SM) toxicity, while the ability of cells to
repair this damage is thought to provide a degree of natural
protection. To investigate the repair process, we have
damaged plasmids containing the firefly luciferase gene
with either SM or its monofunctional analog, 2-chloroethyl
ethyl sulfide (CEES). Damaged plasmids were transfected
into wild-type and nucleotide excision repair (NER) defi-
cient Chinese hamster ovary cells; these cells were also
transfected with a second reporter plasmid containing
Renilla luciferase as an internal control on the efficiency
of transfection. Transfected cells were incubated at 37°C
for 27 h and then both firefly and Renilla luciferase intensit-
ies were measured on the same samples with the dual
luciferase reporter assay. Bioluminescence in lysates from
cells transfected with damaged plasmid, expressed as a
percentage of the bioluminescence from cells transfected
with undamaged plasmid, is increased by host cell repair
activity. The results show that NER-competent cells have
a higher reactivation capacity than NER-deficient cells for
plasmids damaged by either SM or CEES. Significantly,
NER-competent cells are also more resistant to the toxic
effects of SM and CEES, indicating that NER is not only
proficient in repairing DNA damage caused by either agent
but also in decreasing their toxicity. This host cell repair
assay can now be used to determine what other cellular
mechanisms protect cells from mustard toxicity and under
what conditions these mechanisms are most effective.
Introduction
Sulfur mustard, bis-(2-chloroethyl) sulfide (SM), is a bifunc-
tional alkylating agent that has cytotoxic, mutagenic and
vesicant properties, and is considered carcinogenic by the
IARC (1). Sulfur mustard interacts with cellular DNA to
form the cross-link, di-(2-guanin-7-yl-ethyl)-sulfide, and two
monoadducts, 7-(2-hydroxyethylthioethyl) guanine (HETEG)
and 3-(2-hydroxyethylthioethyl) adenine (HETEA) (reviewed
in ref. 2). DNA modification by SM has been shown to
interfere with replication and transcription and is probably
responsible for its various toxicities (3–5).
Since the resistance of Escherichia coli cells to the lethal
effect of SM correlates with their ability to remove cross-
links, it has generally been assumed that the formation of
DNA cross-links is a major cause of SM toxicity (3,6).
Abbreviations: CEES, 2-chloroethyl ethyl sulfide; HETEA, 3-(2-hydroxy-
ethylthioethyl) adenine; HETEG, 7-(2-hydroxyethylthioethyl) guanine; NER,
nucleotide excision repair; SM, sulfur mustard, bis-(2-chloroethyl) sulfide.
© Oxford University Press
661

Z.Matijasevic et al.
Fig. 1. Structures of SM and CEES and the adducts they form with DNA.
Sulfur mustard forms the indicated cross-link and two monoadducts with
R ϭ OH; CEES forms only monoadducts with R ϭ H.
Fig 2. Survival of CHO cells after exposure to SM or CEES. Viable cell
numbers were determined by trypan blue exclusion and are expressed as a
percent of control on day 1 (D1), day 2 (D2) and day 3 (D3). Wild-type
cells (s); NER-deficient cells (d).
relatively simple host cell reactivation assay can be used to
determine what repair mechanisms restore mustard-damaged
DNA to a functional state and under what conditions they are
most effective.
densitometrically from a UV photograph of the ethidium bromide-stained
agarose gel.
Materials and methods
Materials
Transfection conditions
Bis-(2-chloroethyl)sulfide (SM) was supplied by the US Army Institute of
Chemical Defense (Aberdeen Proving Ground, MD) and 2-chloroethyl ethyl
sulfide (CEES) was purchased from Aldrich (Milwaukee, WI). The luciferase
reporter vectors pGL3-Control (containing the firefly luciferase gene) and
pRL-TK (containing the Renilla luciferase gene), the transfection reagent
TransFast and the Dual-Luciferase Reporter Assay System were purchased
from Promega (Madison, WI). Plasmid vectors were propagated in Escherichia
coli strain JM109 and plasmid DNAs were purified by using the EndoFree
Plasmid Mega kit (Qiagen, Chatsworth, CA). Chinese hamster ovary (CHO)
cell lines AA8 (wild type) and UV41 (NER group 4) were obtained from the
American Type Culture Collection. These cells were grown at 37°C in
monolayer culture in α-modified minimum essential medium supplemented
with 10% fetal bovine serum and antibiotics.
For transfection experiments, cells were plated in 24-well plates at a density
of 2ϫ104 cells/cm2 and incubated for 24 h at 37°C. Transfection was
performed using the liposome-based transfection reagent, TransFast. Optimal
conditions for transfection were established as 0.5 µg plasmid DNA per well,
at a charge ratio of transfection reagent to DNA of 1:1, and a ratio of pGL3
to pRL-TK of 10:1. Twenty-four hours after plating, the growth medium was
replaced with 0.2 ml serum-free medium containing the transfection mixture.
After the cells were incubated for 1.5 h at 37°C, the transfection reagent was
diluted by the addition of 1 ml complete medium and cells were further
incubated for luciferase expression.
Preliminary experiments with undamaged pGL3 and pRL-TK plasmids
have shown that the difference in genetic background between the AA8 and
UV41 cells does not affect the kinetics of expression of either firefly or
Renilla luciferase. The level of expression of both luciferases increases during
the first 25 h of incubation at 37°C and then reaches a plateau. Therefore in
repair experiments, cells were incubated for 27 h after transfection to allow a
maximal level of luciferase expression. At that time, cells were approximately
90% confluent.
Sulfur mustard exposure and cytotoxicity determinations
Cells were plated in 12-well plates at a density of 2ϫ104 cells/cm2. After
24 h, the medium was replaced with fresh medium containing the indicated
concentrations of SM or CEES; dilute solutions of these compounds in
absolute alcohol were prepared immediately before treatment. Cells were
exposed to SM or CEES for 1 h at room temperature in a SterilchemGard
hood and then incubated in fresh medium at 37°C. At the indicated times,
cell viability was determined by the trypan blue exclusion assay.
Cell lysis and assay for luciferase activity
Cells were lysed in multi-well plates with 150 µl/well passive lysis buffer
provided with the Dual Luciferase Reporter Assay kit; lysates were stored
frozen at –20°C until they were assayed. The assays for firefly luciferase
activity and Renilla luciferase activity were performed sequentially in one
reaction tube using 20 µl aliquots of cell lysates. The supplier’s standard
protocol for the dual luciferase assay was followed. The luminescent signal
from the luciferase reaction was monitored by a single-sample luminometer
(Monolight 2010; Analytical Luminescence Laboratory, San Diego, CA) with
spectral sensitivity over the range 360–620 nm. The values shown are means
Ϯ SD from three to five separate assays.
Alkylation of plasmid DNA
Purified pGL3 DNA was dissolved in TE buffer (10 mM Tris–HCl, 1 mM
EDTA, pH 7.4) at a concentration of 1 µg/µl and incubated with SM or CEES
in a SterilchemGard hood at room temperature for 1 h. DNA was precipitated
with ethanol, dissolved in TE buffer and stored at –20°C until used for
transfection. Aliquots were analyzed by gel electrophoresis on 1% agarose
gel. The percent conversion of supercoiled DNA (Form I) into the nicked
circular form (Form II) after exposure to SM or CEES was determined
662

Repair of sulfur mustard-induced DNA damage
Fig 4. DNA damage caused by SM and CEES. The percent conversion of
the firefly luciferase plasmid to Form II is plotted versus the concentration
of mustard to which it was exposed.
Fig 5. Relative luciferase expression in wild type (s) and NER-deficient
cells (d); values are means Ϯ SD from three to five separate assays. See
text for details.
two plasmids, one containing the damaged firefly luciferase
gene and the other (pRL) containing an undamaged Renilla
luciferase gene. The intensity of firefly luciferase enzyme
activity can then be compared with the level of Renilla
luciferase activity as a control. This protocol is shown in
Figure 3.
Fig 3. Scheme for monitoring DNA repair using damaged luciferase
reporter gene in CHO cells. The level of firefly luciferase activity relative to
Renilla luciferase activity is a measure of how efficiently the damaged
firefly luciferase gene has been repaired.
The firefly luciferase gene was damaged as described in the
Materials and methods by exposing plasmid pGL3 to either
SM or CEES in vitro. Damaged plasmid was then recovered
and used in the host cell reactivation assays as shown in
Figure 3.
Results
The data in Figure 2 show that CHO cells exposed to either
SM or CEES are protected from cytotoxicity by the NER
mechanism. Referring to data on day 3 (D3), Ͼ50% of NER-
competent cells survive exposure to 20 µM SM while fewer
than 2% of the NER-deficient cells survive. A similar difference
is noted after exposure to CEES except that concentrations of
CEES Ͼ10-fold greater than those of SM are required to
produce the same level of cytotoxicity. Again at day 3, survival
for wild-type cells is close to 100% after exposure to 300 µM
CEES, but only ~5% for NER-deficient cells.
Based on the assumption that unrepaired DNA damage is
responsible for SM cytotoxicity, we would expect that the
increased survival shown in Figure 2 would be accompanied
by an increase in cellular repair of DNA. This has been
confirmed by measuring cellular DNA repair directly with the
host cell repair assay as described below. In this assay, a
plasmid that contains a gene for firefly luciferase is damaged
by SM or CEES and then transfected into the cells that are to
be evaluated for repair. Cells that can repair the DNA damage
will express the luciferase gene at a higher level (i.e. the cells
will ‘reactivate’ the gene). This results in an increased intensity
of firefly luciferase bioluminescence that can be used to
determine the extent of repair.
Damage to plasmid DNA can be detected by measuring the
conversion of the supercoiled form of the plasmid (Form I) to
the nicked circular form (Form II). We assume that the extent
of DNA damage that causes this conversion parallels the DNA
damage that interferes with expression of the luciferase gene.
Accordingly, we have measured conversion of the plasmid to
Form II as an indication of the DNA damage caused to the
firefly luciferase gene by SM and CEES. The level of conver-
sion of Form I to Form II is shown in Figure 4. As expected,
both SM and CEES convert Form I to Form II, but it takes
an ~10-fold higher concentration of CEES to cause as much
damage as is caused by a given concentration of SM.
The host cell reactivation data in Figure 5 show the extent
to which this damage was repaired in the two different cell
lines. In these experiments, plasmid containing damaged firefly
luciferase gene was transfected separately into either wild-
type or NER-deficient CHO cells. As shown in Figure 5, wild-
type cells can return luciferase expression to normal if the
plasmids are damaged with low concentrations of SM or
CEES. In contrast, NER-deficient cells show much less repair,
and levels of firefly luciferase expression fall off as the
concentrations of SM or CEES are raised.
To compensate for variations in the efficiency of transfection
and other experimental variables, luciferase reporter gene
assays are usually run using dual transfection with DNA from
It is possible that some of the damage to the luciferase gene
663

Z.Matijasevic et al.
could be repaired by mechanisms other than NER in the NER-
deficient cells, but the lack of a shoulder on the curve for
NER-deficient cells at low concentrations of damaging agent
suggests that NER is needed to restore full expression. In any
case, the data in Figure 5 indicate that the host cell reactivation
assay can be used to examine repair of SM- or CEES-induced
DNA damage within the cell. It is also apparent from the right
hand panel of Figure 5 that NER removes the monofunctional
adducts caused by CEES and, by analogy, the monofunctional
adducts caused by SM as well.
References
1.IARC (1987) IARC Monographs on the evaluation of carcinogenic risks
to humans: Overall evaluations of carcinogenicity: An updating of IARC
Monographs Volumes 1 to 42 (suppl. 7). IARC, Lyon, pp. 259–260.
2.Papirmeister,B., Feister,A.J., Robinson,S.I. and Ford,R.D. (1991) Medical
Defense Against Mustard Gas: Toxic Mechanisms and Pharmacological
Implications. CRC Press, Boca Raton.
3.Lawley,P.D. and Brookes,P. (1965) Molecular mechanism of the cytotoxic
action of difunctional alkylating agents and of resistance to this action.
Nature, 206, 480–483.
4.Masta,A., Gray,P.J. and Phillips,D.R. (1996) Effect of sulphur mustard on
the initiation and elongation of transcription. Carcinogenesis, 17, 525–532.
5.Roberts,J.J., Brent,T.P. and Crathorn,A.R. (1968) The mechanism of the
cytotoxic action of alkylating agents on mammalian cells: inactivation of
the DNA template and its repair. In Campbell,P.N. (ed.) The Interaction
of Drugs and Subcellular Components in Animal Cells. J. & A. Churchill,
London, WI.
6.Venitt,S. (1968) Interstrand cross-links in the DNA of Escherichia coli B/
r and BS–1 and their removal by the resistant strain. Biochem. Biophys.
Res. Commun., 31, 355–360.
7.Ball,C.R. and Roberts,J.J. (1970) DNA repair after mustard gas alkylation
by sensitive and resistant Yoshida sarcoma cells in vitro. Chem. Biol.
Interact., 2, 321–329.
8.Crathorn,A.R. and Roberts,J.J. (1966) Mechanism of the cytotoxic action
of alkylating agents in mammalian cells and evidence for the removal of
alkylated groups from deoxyribonucleic acid. Nature, 211, 150–153.
9.Roberts,J.J., Crathorn,A.R. and Brent,T.P. (1968) Repair of alkylated DNA
in mammalian cells. Nature, 218, 970–972.
10.Papirmeister,B. and Davison,C.L. (1964) Elimination of sulfur mustard-
induced products from DNA of Escherichia coli. Biochem. Biophys. Res.
Commun., 17, 608–617.
11.Reid,B.D. and Walker,I.G. (1969) The response of mammalian cells to
alkylating agents II. On the mechanism of the removal of sulfur-mustard-
induced cross-links. Biochim. Biophys. Acta, 179, 179–188.
12.Mol,M.A.E., van der Schans,G.P. and Lohman,H.M. (1993) Quantification
of sulfur mustard-induced DNA interstrand cross-links and single-strand
breaks in cultured human epidermal keratinocytes. Mutat. Res., 294,
235–245.
Discussion
The survival curves in Figure 2 show that the difunctional
agent SM is ~10-fold more cytotoxic than CEES for both cell
lines, in agreement with previous observations (2). Since CEES
forms monofunctional adducts that are similar to those formed
by SM, this difference in cytotoxicity between SM and CEES
has been assumed to be the result of cross-link formation (2).
The data (Figure 2, left panel) show that NER competent cells
are more resistant to the cytotoxic action of SM than are NER
deficient cells suggesting that the cytotoxic cross-link is
removed by the NER repair mechanism. Again, this would
agree with previous observations that the NER pathway recog-
nizes bulky adducts like DNA cross-links.
Since the data (Figure 2, right panel) show that NER
competent cells are also more resistant to the cytotoxic action
of CEES, we can conclude that NER repairs the monoadducts
formed by CEES and, presumably, the similar monoadducts
formed by SM as well. These adducts, as well as the cross-
link, would probably be classified as bulky because of the size
of their adducted groups.
Host cell reactivation data provide direct information on
cellular repair of DNA damage, and the data in Figure 5
support the role of the NER mechanism in protecting cells
from mustard toxicity since firefly luciferase gene damaged
by either SM or CEES is restored to a higher level of expression
in the NER-competent CHO cells. This reinforces the belief
that DNA repair mechanisms protect against SM toxicity.
Thus the host cell reactivation assay described in this
manuscript demonstrates that there is a direct relationship
between resistance to SM and CEES toxicity and the cellular
repair of DNA damage caused by these agents. This not only
validates the hypothesis that DNA damage is the root cause
of mustard toxicity, but provides a method of testing environ-
mental conditions such as hypothermia that may increase the
extent of DNA repair.
13.Matijasevic,Z., Stering,A., Niu,T.-Q., Austin-Ritchie,P. and Ludlum,D.B.
(1996) Release of sulfur mustard-modified DNA bases by E.coli 3-
methyladenine DNA glycosylase II. Carcinogenesis, 17, 2249–2252.
14.Kircher,M., Fleer,R., Ruhland,A. and Brendel,M. (1979) Biological and
chemical effects of mustard gas in yeast. Mutat. Res., 63, 273–289.
15.Dabholkar,M., Eastman,A. and Reed,E. (1990) Host-cell reactivation of
cisplatin-damaged pRSVcat in a human lymphoid cell line. Carcinogenesis,
11, 1761–1764.
16.Ruan,S., Okcu,M.F., Ren,J.P., Chiao,P., Andreeff,M., Levin,V. and
Zhang,W. (1998) Overexpressed WAF1/Cip1 renders glioblastoma cells
resistant to chemotherapy agents 1,3-bis(2-chloroethyl)-1-nitrosourea and
cisplatin. Cancer Res., 58, 1538–1543.
17.Sheibani,N., Jennerwein,M.M. and Eastman,A. (1989) DNA repair in cells
sensitive and resistant to cis-diamminedichloroplatinum(II): Host cell
reactivation of damaged plasmid DNA. Biochemistry, 28, 3120–3124.
18.Yen,L., Woo,A., Christopoulopoulos,G., Bastist,G., Panasci,L., Roy,R.,
Mitra,S. and Alaoui-Jamali,M.A. (1995) Enhanced host cell reactivation
capacity and expression of DNA repair genes in human breast cancer cells
resistant to bi-functional alkylating agents. Mutat. Res., 337, 179–189.
Acknowledgements
This work was supported by the US Army Medical Research and Materiel
Command under Contracts DAMD17-96-C-6073 and DAMD17-00-C-0012.
Received October 25, 2000; revised December 21, 2000;
accepted December 28, 2000
664