Thiol-activated DNA damage by α-bromo-2-cyclopentenone

Article abstract of DOI:10.1021/tx100282b

Some biologically active chemicals are relatively stable in the extracellular environment but, upon entering the cell, undergo biotransformation into reactive intermediates that covalently modify DNA. The diverse chemical reactions involved in the bioactivation of DNA-damaging agents are both fundamentally interesting and of practical importance in medicinal chemistry and toxicology. The work described here examines the bioactivation of α-haloacrolyl-containing molecules. The α-haloacrolyl moiety is found in a variety of cytotoxic natural products including clionastatin B, bromovulone III, discorahabdins A, B, and C, and trichodenone C, in mutagens such as 2-bromoacrolein and 3-chloro-4-(dichloromethyl)-5-hydroxy-2(5H)-furanone (MX), and in the anticancer drug candidates brostallicin and PNU-151807. Using α-bromo-2-cyclopentenone (1) as a model compound, the activation of α-haloacrolyl-containing molecules by biological thiols was explored. The results indicate that both low molecular weight and peptide thiols readily undergo conjugate addition to 1. The resulting products are consistent with a mechanism in which initial addition of thiols to 1 is followed by intramolecular displacement of bromide to yield a DNA-alkylating episulfonium ion intermediate. The reaction of thiol-activated 1 with DNA produces labile lesions at deoxyguanosine residues. The sequence specificity and salt dependence of this process is consistent with involvement of an episulfonium ion intermediate. The alkylated guanine residue resulting from the thiol-triggered reaction of 1 with duplex DNA was characterized using mass spectrometry. The results provide new insight regarding the mechanisms by which thiols can bioactivate small molecules and offer a more complete understanding of the molecular mechanisms underlying the biological activity of cytotoxic, mutagenic, and medicinal compounds containing the α-haloacrolyl group.

Full text of DOI:10.1021/tx100282b

ARTICLE
pubs.acs.org/crt
Thiol-Activated DNA Damage by R-Bromo-2-cyclopentenone
Mostafa I. Fekry,†,‡ Nathan E. Price,† Hong Zang,† Chaofeng Huang,† Michael Harmata,† Paul Brown,
J. Scott Daniels,*,^, and Kent S. Gates*,†,§
†Department of Chemistry, University of Missouri, 125 Chemistry Building Columbia, Missouri 65211, United States
‡Pharmacognosy Department, Faculty of Pharmacy, Cairo University, Kasr El-Aini, Cairo, Egypt 11562
§Department of Biochemistry, University of Missouri, 125 Chemistry Building Columbia, Missouri 65211, United States
Pfizer, Inc., 700 Chesterfield Parkway, Chesterfield, Missouri 63017, United States
S Supporting Information
b
ABSTRACT: Some biologically active chemicals are relatively
stable in the extracellular environment but, upon entering the
cell, undergo biotransformation into reactive intermediates
that covalently modify DNA. The diverse chemical reactions
involved in the bioactivation of DNA-damaging agents are both
fundamentally interesting and of practical importance in me-
dicinal chemistry and toxicology. The work described here
examines the bioactivation of R-haloacrolyl-containing molecules. The R-haloacrolyl moiety is found in a variety of cytotoxic
natural products including clionastatin B, bromovulone III, discorahabdins A, B, and C, and trichodenone C, in mutagens such as
2-bromoacrolein and 3-chloro-4-(dichloromethyl)-5-hydroxy-2(5H)-furanone (MX), and in the anticancer drug candidates
brostallicin and PNU-151807. Using R-bromo-2-cyclopentenone (1) as a model compound, the activation of R-haloacrolyl-
containing molecules by biological thiols was explored. The results indicate that both low molecular weight and peptide thiols
readily undergo conjugate addition to 1. The resulting products are consistent with a mechanism in which initial addition of
thiols to 1 is followed by intramolecular displacement of bromide to yield a DNA-alkylating episulfonium ion intermediate. The
reaction of thiol-activated 1 with DNA produces labile lesions at deoxyguanosine residues. The sequence specificity and salt
dependence of this process is consistent with involvement of an episulfonium ion intermediate. The alkylated guanine residue
resulting from the thiol-triggered reaction of 1 with duplex DNA was characterized using mass spectrometry. The results provide
new insight regarding the mechanisms by which thiols can bioactivate small molecules and offer a more complete under-
standing of the molecular mechanisms underlying the biological activity of cytotoxic, mutagenic, and medicinal compounds
containing the R-haloacrolyl group.
’ INTRODUCTION
with thiols initiates both DNA alkylation and generation of
reactive oxygen species (ROS).19-27 The natural products
varacin, lissoclinotoxin A, and thiarubrin C also generate ROS
following reactions with thiols.28-31 In addition, thiol-triggered
analogues of mitomycin C have been characterized.32,33
The diverse chemical reactions involved in the bioactivation of
DNA-damaging agents are both fundamentally interesting and of
practical importance in medicinal chemistry and toxicology.1,2,34
For example, elucidation of bioactivation processes can reveal
new chemical mechanisms by which small organic molecules can
deliver reactive intermediates to the interior of cells. The work
described here examines thiol-mediated bioactivation of R-
haloacrolyl-containing molecules. The R-haloacrolyl moiety is
found in a variety of cytotoxic natural products including
clionastatin B, bromovulone III, discorahabdins A, B, and C,
and trichodenone C, in mutagens such as 2-bromoacrolein and
3-chloro-4-(dichloromethyl)-5-hydroxy-2(5H)-furanone (MX),
Many DNA-damaging natural products are relatively stable in
the extracellular environment but, upon entering cells, undergo
biotransformation to reactive intermediates that covalently mod-
ify the genetic material.1,2 For example, DNA damage by mito-
mycins,3,4 saframycins,5 and dynemicin6 is activated by enzy-
matic reduction of their quinone moieties. Azomycin and
2-carboxyquinoline 1,4-dioxide are activated by enzymatic re-
duction of nitro and heterocyclic N-oxide groups, respectively.7,8
Bioactivation of acylfulvene natural products is carried out by
NADPH-dependent alkenal/one reductase,9 while DNA alkyl-
ation by pyrrolizidine alkaloids and alfatoxins is triggered by
cytochrome P450-mediated monooxygenation reactions.10-12
Intracellular thiols13,14 also have the potential to activate
DNA-damaging natural products. For example, DNA strand
cleavage by calicheamicin, dynemicin, and neocarzinostatin are
triggered by reactions with thiols.6,15-17 DNA alkylation by
acylfulvenes may be triggered by 1,4-addition of thiols.18 Re-
action of the Streptomyces-derived natural product leinamycin
Received: August 17, 2010
Published: January 20, 2011
r
2011 American Chemical Society
217
dx.doi.org/10.1021/tx100282b Chem. Res. Toxicol. 2011, 24, 217–228
|

Chemical Research in Toxicology
ARTICLE
mass spectrometer (Thermoelectron Corp., San Jose, CA). For direct
infusion ESI-MS, the solid phase extracted reaction was infused into
the mass spectrometer at a flow rate of 10 μL/min using a stainless
steel emitter on a Thermo-Fisher MaxIon source. The heated capillary
was maintained at 275 ꢀC. For MS/MS analysis, ions of interest were
selected with an isolation window of 2.0 Da. Fragmentation was
catalyzed via collision-induced dissociation (CID) and high-energy
C-trap dissociation (HCD), with normalized collision energies varying
from 10 to 40%. All spectra were acquired in the full profile mode with
a resolution of 100,000 at m/z 400. Each spectrum was generated by
signal averaging for approximately 20 s with a maximum injection time
of 1000 ms in MS mode and 2000 ms in MS/MS mode. All the data
were acquired using external calibration with a mixture of caffeine,
MRFA peptide, and Ultramark 1600, dissolved in a acetonitrile/water
(50:50, v/v). The multiply charged, pseudomolecular ions of the
alkylated peptides were subjected to CID in the LTQ and HCD in
the multipole collision cell adjacent to the Orbitrap mass analyzer.
For LC-MS/MS, the reaction mixtures were separated employing an
Agilent 1200 nanoflow LC system (Agilent, Wilmington, DE) utiliz-
ing a gradient of 0.1% formic acid and acetonitrile (3-45% acetoni-
trile over 45 min) and coupled to a C-18 reverse phase column (5 μm,
200 Å, Michrom Bioresources, Inc., USA) and a 15-cm fused silica
emitter (75 μm i.d., PicoTip EmitterTM/PicoFrit Self/P, New
Objective, USA). The LC eluent was introduced into the LTQ
Orbitrap mass spectrometer equipped with a nanoelectrospray ion
source (Thermoelectron Corp, Bremen, Germany). Full scan MS
spectra (m/z 300-2000 Da) were acquired with a resolution of
60,000 at m/z 400.
Reaction of 1 with 2-Mercaptoethanol. A mixture of 1 (80.5
mg, 0.5 mmol), 2-mercaptoethanol (35 μL, 0.5 mmol), and triethyl-
amine (70.3 μL, 0.5 mmol) were mixed in methanol (10 mL) and incuba-
ted at 37 ꢀC for 24 h. Following the removal of solvent, the major prod-
uct was purified by column chromatography on silica gel eluted with
40% ethyl acetate in dichloromethane. The compound 2-(2-hydroxyethyl)-
cyclopent-2-enone (6c) was obtained in approximately 40% yield
(contaminated with traces of 2-mercaptoethanol and 2-mercaptoetha-
nol disulfide). 1H NMR (300 MHz, CDCl3): δ 7.41 (t, J = 3 Hz, 1H),
3.78 (t, J = 6 Hz, 2H), 3.04 (t, J = 6 Hz, 2H), 2.72-2.68 (m, 2H),
2.54-2.51 (m, 2H).
In Vitro Metabolism of 1 in Human Hepatic S9 Fractions
and Hepatocytes . Human Hepatic S9 Fractions. The in vitro
metabolism of 1 was investigated using human hepatic S9 fractions.47 A
potassium phosphate-buffered reaction (0.1 M, pH 7.4) of 1 (100 μM),
hepatic S9 fractions (5 mg/mL), glutathione (2 mM), and MgCl2
(3 mM) was initiated by the addition of NADPH (2 mM) and incubated
at 37 ꢀC in borosilicate glass test tubes under ambient oxygenation for
50 min. Protein was precipitated by the addition of two volumes of
acetonitrile, and the resulting mixture was chilled at 4 ꢀC for 20 min
followedbycentrifugation. The residueswerereconstitutedin85:15(v/v)
water/acetonitrile, a solution reflective of the ensuing LC/MS analysis.
Hepatocytes. Cryopreserved human hepatocytes (1 ꢀ 106 cells/
mL) were incubated with 1 (100 μM) at 37 ꢀC for 4 h under controlled
atmospheric conditions (95% air/5% CO2). Cells were suspended in
Krebs-Hensleit buffer that was supplemented with glucose (5 mM) and
sodium pyruvate (1 mM). Cells were precipitated and reactions stopped
by the addition of two volumes of acetonitrile, followed by storage
at -20 ꢀC. Samples were prepared for LC-MS analysis by decanting the
supernatants and drying under a stream of nitrogen. The residues were
reconstituted in water/acetonitrile (85:15, v/v) in preparation for LC/
MS/MS analysis. LC/MS/MS characterization was completed employ-
ing an Agilent 1100 HPLC system coupled to a Symmetry C18 column
(5 μm, 3.8 ꢀ 150 mm; Waters Corporation, Milford, MA). Solvent A
was formic acid (0.1%), and solvent B was acetonitrile (fortified with
Figure 1. Bioactive compounds containing an R-haloacrolyl group.
and in the anticancer drug candidates brostallicin and PNU-
151807 (Figure 1).35-46 We employed R-bromo-2-cyclopentenone
as a model compound to examine the thiol-mediated bio-
activation of R-haloacrolyl-containing molecules. The evidence
presented here suggests that 1,4-addition of thiols to the
R-haloacrolyl moiety leads to the formation of an episulfo-
nium ion that can alkylate guanine residues in duplex DNA
(Scheme 1).
’ EXPERIMENTAL PROCEDURES
Materials. Materials were purchased from the following suppliers:
3-[N-morpholino]propanesulfonic acid (MOPS), 4-(2-hydroxyethyl)-
piperazine-1-ethanesulfonic acid (HEPES), 1,4-piperazinediethanesulfonic
acid (PIPES), ethylenediaminetetraacetic acid (EDTA), N,N,N0,N0-
tetraethylenediamine (TEMED), sigmacote, L-lysine, glutathione-S-
transferase from equine liver, and cysteamine were purchased from Sigma
Chemical Company; dimethylsulfate, sodium acetate, 2-cyclopenten-
1-one, 3-bromo-2-butanone, and N-methylaniline were purchased from
Aldrich Chemical; G-25 Sephadex, glutathione, and 2-mercaptoetha-
nol were purchased from Sigma-Aldrich; diethylenetriaminepentaace-
tic acid (DETAPAC) was purchased from Fluka Chemical Company;
piperidine was purchased from Alfa Aesar; xylene cyanol, bromophe-
nol blue, formamide, urea, and ammonium persulfate were purchased
from United States Biochemical; herring sperm DNA and TRIS-
borate-EDTA (TBE) were purchased from Roche Molecular Biochem-
icals; acrylamide was purchased through Fischer Scientific Inc., T4
polynucleotide kinase (T4-PNK) and bovine serum albumin (BSA)
were purchased from New England Biolabs; [γ-32P]-dATP was pur-
chased from Perkin-Elmer Life Sciences; and oligonucleotides were
purchased from Integrated DNA Technologies.
Reaction of 1 with the Peptide Probe γ-ECGHDRKAHYK.
The peptide probe, γ-ECGHDRKAHYK (200 μM), was incubated with
1 (200 μM) in potassium phosphate (0.1 M, pH 7.4) at 25 ꢀC for 0.5 h.
Following incubation, the modified peptide(s) was extracted from the
solution employing a solid phase (C8 HLB cartridges, Waters) chro-
matographic workup consisting of a series of acetonitrile-0.1% formic
acid mixtures to wash and subsequently elute the peptide. The peptide-
containing eluant resulting from an 80% acetonitrile (0.1% formic acid)
wash was subjected to electrospray ionization-tandem mass spectrom-
etry (ESI-MS/MS) and liquid chromatography-tandem mass spect-
rometry (LC/MS/MS) employing an LTQ-Orbitrap XL high-resolution
218
dx.doi.org/10.1021/tx100282b |Chem. Res. Toxicol. 2011, 24, 217–228

Chemical Research in Toxicology
ARTICLE
Scheme 1
0.1% formic acid). The initial mobile phase was 98:2 A/B (v/v) and by
linear gradient transitioned to 10:90 A/B over 22 min.
5 min in buffer (25 mM MOPS, pH 7, and 100 mM NaCl), followed by
slow-cooling to room temperature overnight (∼20 h). In a typical
alkylation reaction, the 50-32P labeled oligonucleotide duplex (3 μL,
approximately 50,000 cpm), HEPES buffer (2 μL of a 500 mM aqueous
solution, pH 7), DETAPAC (0.5 μL of a 40 mM aqueous solution),
2-mercaptoethanol (0.5 μL, 2.5 μL, or 5 μL of a 40 mM aqueous
solution), and 1 (0.5 μL, 2.5 μL, or 5 μL of a 40 mM solution in
acetonitrile) were mixed with an appropriate amount of distilled,
deionized water to give a final volume of 20 μL and the solution agitated
using a vortex mixer (final concentrations: 50 mM HEPES, pH 7; 1 mM
DETAPAC; 1 mM, 5 mM, or 10 mM of 2-mercaptoethanol; 1 mM,
5 mM, or 10 mM of 1).
The reaction was quenched by precipitation of the DNA. To the
mixture was added herring sperm carrier DNA (3 μL of a 1 mg/mL
solution), sodium acetate (2 μL of a 3 M, pH 5 solution), and absolute
ethanol (175 μL); (final concentrations: 15 μg/mL of carrier DNA,
30 mM of sodium acetate, 87.5% ethanol, final volume 200 μL). The
DNA was precipitated by cooling this mixture on dry ice for 45 min,
followed by centrifuging for 45 min at 14,000 rpm in a benchtop
microcentrifuge. The supernatant was removed, and the pellet was
washed twice with 70% ethanol-water (v/v). The DNA pellet was
redissolved in aqueous piperidine (100 μL of a 0.5 M solution) and
incubated at 90 ꢀC for 25 min.49 The solution was frozen on dry ice and
lyophilized for 1.5 h in a SpeedVac Concentrator at 37 ꢀC. The oligo-
nucleotides were then taken through two cycles of redissolution in 90 μL
of water, freezing, and lyophilization in a SpeedVac Concentrator at
37 ꢀC. The dried DNA fragments were dissolved in formamide loading
buffer and denatured at 90 ꢀC for 4 min, then immersed in ice water. An
equal number of counts were loaded in each lane of a 20% denaturing
polyacrylamide gel (19:1 cross-linking, containing 5 M urea). The gel
was electrophoresed at 800 V for approximately 9 h. The DNA frag-
ments in the gel were visualized by phosphorimager analysis (Molecular
Imager FX, Imaging Screen-K, cat 170-7841, Bio-Rad, using Quantity
One Version 4.2.3, Bio-Rad). Experiments involving other thiol or amine
Reaction of 1 with 2-Mercaptoethanol or Glutathione:
Qualitative Rate Measurements. A solution containing 1 (0.05
mM), 2-mercaptoethanol (0.5 mM, 1 mM, or 2 mM), HEPES buffer
(50 mM, pH 7), DETAPAC (1 mM as a chelator of adventitious trace
metals to suppress autoxidation of thiols and the production of radicals
that accompanies this process), and acetonitrile (0.06% v/v) was pre-
pared in HPLC grade water, vortex-mixed, and placed in a quartz cuvette
(1 cm path length). Absorption spectra were collected at 5 s intervals at
room temperature (Figure 3). Alternatively, the reaction of 1 with glu-
tathione was measured using HPLC. A solution containing sodium
phosphate buffer (60 μL of a 500 mM aqueous solution, pH 7), 1 (15 μL
of a 40 mM solution in acetonitile), and glutathione (150 μL of a 40 mM
aqueous solution) were mixed with an appropriate amount of distilled,
deionized water to give a final volume of 600 μL, and the solution
agitated using a vortex mixer and incubated at 37 ꢀC for 2 min (final
concentrations: 50 mM sodium phosphate buffer, pH 7; 1 mM of 1;
10 mM of glutathione; 2.5% (v/v) acetonitrile; 600 μL final reaction
volume). After incubation, 5 μL of the reaction mixture was analyzed by
HPLC using a reverse phase Supilcosil LC-18-S column (particle size,
5 μm;15 cm L ꢀ 2.1 mmID) ata flow rate of0.2 mL/minusing a gradient
of increasing methanol in water. The HPLC elution protocol was a
continuous gradient of 80-95% methanol over 30 min. The column was
then washed with 95% methanol for 10 min, followed by a continuous
gradient of 95-80% methanol over 5 min, then 80% methanol for
15 min.
Thiol-Triggered DNA Alkylation by 1. The single-stranded 20-
deoxyoligonucleotide 50-GTT CGT ATA TGG GAG GTC GCA TGT
G-30 was labeled using [γ-32P]dATP and T4 polynucleotide kinase
according to standard protocols48 and were purified on a 20% denatur-
ing polyacrylamide gel.48 The purified single-stranded radiolabeled
oligonucleotide was annealed with 1.5 equiv of the complementary
strand 50-CAC ATG CGA CCT CCC ATA-30 by heating to 95 ꢀC for
219
dx.doi.org/10.1021/tx100282b |Chem. Res. Toxicol. 2011, 24, 217–228

Chemical Research in Toxicology
ARTICLE
Figure 2. Mass spectra of the products obtained from the reaction of 1 with the peptide γ-ECGHDRKAHYK.
activating agents were carried out in an identical manner. Control
reactions examining alkylation by 3-bromo-2-butanone and 2-cyclopen-
tenone were carried out as described above except that 1 was replaced
with these molecules. Experiments in which the pH was varied (pH 6.1
and 8) employed a buffer consisting of a HEPES (50 mM) and PIPES
(50 mM) mixture.
LC/MS/MS Characterization of the Guanine Adducts Gen-
erated by the Reaction of 1 and Thiol with Duplex DNA.
Herring sperm DNA (double-stranded, 5 mM bp) was treated with 1
(5 mM) and 2-mercaptoethanol (5 mM) in HEPES buffer (pH 7,
50 mM) containing DETAPAC (1 mM). The reaction mixture was
incubated at 37 ꢀC for 72 h. The resulting alkylated DNA was heated at
220
dx.doi.org/10.1021/tx100282b |Chem. Res. Toxicol. 2011, 24, 217–228

Chemical Research in Toxicology
ARTICLE
Table 1. Calculated Exact Mass (calcd) and Measured Accurate Mass (obsvd) for the Native and Adducted Peptides
compound
[M þ 2H]2þ calcd
[M þ 2H]2þ obsvd
[M þ 3H]3þ calcd
[M þ 3H]3þ obsvd
ECGHDRKAHYK native peptide
672.3173
721.3356
761.3188
712.3304
672.3170
721.3359
761.3193
712.3308
448.5473
481.2262
507.8817
475.2227
448.5468
481.2255
507.8815
475.2226
4a
5a
6a
90 ꢀC for 45 min to release alkylated bases. Large DNA fragments were
then removed by filtering the solution through a microconcentrator filter
(YM-3, Microcon) by centrifugation at 10,000 rpm for 1.5 h. The result-
ing solution was lyophilized in preparation for LC/MS/MS analysis. The
resulting residue was dissolved in aqueous methanol (25% v/v) and a
portion (5 μL) subsequently diluted in 15% aqueous acetonitrile (45 μL).
LC/MS/MS characterization was completed employing an Agilent
1100 HPLC system coupled to a Symmetry C18 column (5 μm, 3.8 ꢀ
150 mm; Waters Corporation, Milford, MA). Solvent A was formic acid
(0.1%), and solvent B was acetonitrile (fortified with 0.1% formic acid).
The initial mobile phase was 95:5 A/B (v/v) and by linear gradient
transitioned to 20:80 A/B over 20 min. The flow rate was 0.4 mL/min.
The HPLC eluent was introduced via electrospray ionization directly
into a Finnigan LCQ Deca XPPLUS ion trap mass spectrometer (Thermo-
electron Corporation, San Jose, CA) operated in the positive ionization
mode. Ionization was assisted with sheath and auxiliary gases (ultra pure
nitrogen) set at 60 and 40 psi, respectively. The electrospray voltage was
set at 5 kV with the heated ion transfer capillary set at 300 ꢀC and 30 V.
Relative collision energies of 32% were used when the ion trap mass
spectrometer was operated in the MS/MS mode.
product eluting at 25.7 min, producing a [M þ 2H]2þ and a
[M þ 3H]3þ at m/z of 721.3359 and 481.2255, respectively
(Figure 2B), consistent with a structure such as 4a (Scheme 1).
The presence of the y9 fragment ion of 4a (m/z 1111.5743 Da)
and the corresponding y102þ fragment ion (m/z 656.8138 Da)
were again indicative of attachment of the small molecule to the
cysteine residue of the probe peptide. Interestingly, another
product eluting at 23.3 min was observed, producing a [M þ
2H]2þ and a [M þ 3H]3þ at m/z 761.3193 and 507.8815 Da
(Figure 2C), respectively, consistent with a phosphate adduct
such as 5a. Again, both the y9 ion (m/z 1111.5731 Da) and the
y102þ ion (m/z 696.7955 Da) were consistent with the afore-
mentioned attachment of the peptide to 1 via the cysteine thiol
group.
The stereochemistry and regiochemistry of the products
identified here (4a-6a) are not defined by the mass spectrom-
etry experiments. In the case of the sulfide, 6a, the results of a
model reaction led us to favor a structure in which the sulfur of
the attacking thiol is ultimately attached at the R-position
adjacent to the carbonyl residue. Specifically, 6c is observed as
a major product of the reaction between 1 and 2-mercaptoetha-
nol in methanol containing 1 equiv of triethylamine at 37 ꢀC.
NMR analysis of this product suggests that the sulfide side chain
resides at the R-position adjacent to the carbonyl residue. This
assignment is supported by the observation that the vinylic
proton in the product appears as a triplet (J = 3 Hz) at 7.41
ppm. This is in line with NMR data reported for structurally
analogous compounds.53
The reaction of thiols with 1 in neutral solution is quite facile.
We monitored the reaction of 1 with 2-mercaptoethanol in
HEPES buffer (50 mM, pH 7) containing acetonitrile (0.06%
v/v) using UV-vis spectrometry (Figure 3). A rapid decrease in
the absorbance associated with the starting material was observed
over the course of several minutes. In a separate HPLC assay, we
find that more than 90% of 1 is consumed after incubation for
2 min with glutathione (10 equiv, 10 mM) in sodium phosphate
buffer (50 mM, pH 7) containing 2.5% acetonitrile (v/v) at 37 ꢀC
(Supporting Information). In the absence of thiol, no degrada-
tion of 1 is observed within this time frame under the same
conditions. These findings are consistent with literature reports
showing that conjugate addition of thiols to R,β-unsaturated
ketones occurs readily.54-56
Although the data presented above indicates that compound 1
reacts readily with thiol groups, there are a variety of nucleophiles
in the intracellular environment that could also, in principle, react
with this compound. Accordingly, we set out to investigate
whether the principle intracellular thiol, glutathione, reacts with
1 in human hepatic subcellular fractions (S9) and in human
hepatocyte cells. We incubated 1 with human hepatic subcellular
fractions (S9) that were fortified with glutathione to probe the
formation of the corresponding thiol adduct of 1. Indeed, LC/
MS analysis of the reaction supernatant revealed the presence of
two isobaric metabolites producing a [M þ H]þ at m/z 388 Da
(Figure 4A), consistent with the addition of glutathione to 1 to
’ RESULTS
Reaction of r-Bromo-2-cyclopentenone (1) with the
Cysteine-Containing Probe Peptide γ-ECGHDRKAHYK, with
2-Mercaptoethanol, or with Glutathione in Human Hepatic
S9 Fractions. Compound 1 was prepared by the reaction of
2-cyclopentenone with Br2 in the presence of triethylamine as
described previously.50,51 To examine the reactivity of 1 with a
range of nucleophiles, we incubated 1 (200 μM) in a phosphate-
buffered (100 mM, pH 7.4) solution containing a peptide (γ-
ECGHDRKAHYK, 200 μM) bearing a variety of nucleophilic
side chains including carboxylate, thiol, alcohol, imidazole, and
amine groups. The peptide is an analogue of glutathione (γ-Glu-
Cys-Gly) bearing a H2N-His-Asp-Arg-Lys-Ala-His-Tyr-Lys-CO2H
peptide linked to the C-terminal glycine residue.52 Following a
30 min incubation (25 ꢀC), the reaction was subjected to solid
phase extraction (C18 Sep-Pak) followed by reversed-phase
LC-MS/MS analysis employing accurate mass spectroscopy
(Supporting Information). Several products arising from the
reaction of 1 with this nucleophilic probe peptide were ob-
served (Scheme 1 and Figure 2). Most notably, a product eluting at
26.8 min produced a [M þ 2H]2þ and a [M þ 3H]3þ at the
mass-to-charge ratios (m/z) of 712.3308 and 475.2226, respec-
tively (Table 1), consistent with adduct 6a arising from the
addition of the peptide to 1 (net addition of 80 Da to the pep-
tide, Scheme 1). Importantly, the presence of the y9 and y102þ
fragment ions at m/z 1111.5727 and 647.8070 (Figure 2D)
suggest that the peptide was linked to 1 via the cysteine residue.
This is indicated by a net 40 Da increase in the y102þ fragment
ion compared to the same ion seen for the native peptide
(607.7940 Da, Figure 2A), while the y9 ions in both spectra
appear at an m/z of 1111.5730 (isobaric). Upon closer exam-
ination of the reaction mixture, we observed two additional
peptide adducts. Specifically, LC/MS/MS analysis revealed a
221
dx.doi.org/10.1021/tx100282b |Chem. Res. Toxicol. 2011, 24, 217–228

Chemical Research in Toxicology
ARTICLE
Figure 3. Time-dependent change in the UV-vis spectra of the
reaction between 1 (50 μM) and 2-mercaptoethanol (1 mM) in 50 mM
HEPES buffer, pH 7, at 24 ꢀC.
Figure 5. Thiol-activated DNA damage by 1. The 50-32P-labeled
oligonucleotide 50-GTTCGTATATGGGAGGTCGCATGTG-30 was
treated with 1 and 2-mercaptoethanol. Reactions were conducted in
HEPES buffer (50 mM), pH 7.0, containing DETAPAC (1 mM) at
37 ꢀC for 23 h and 20 min, followed by Maxam-Gilbert workup and
analysis using 20% denaturing polyacrylamide gel. Lane G, Maxam-
Gilbert G-sequencing reaction. Lanes 1-8 employed duplex DNA,
where the underlined region of the oligonucleotide shown above is
double stranded: lane 1, duplex with Maxam-Gilbert workup (no 1 or
2-mercaptoethanol); lane 2, duplex treated with 1 (10 mM); lane 3,
duplex treated with 2-mercaptoethanol (10 mM); lane 4, duplex
treated with 1 (10 mM) and 2-mercaptoethanol (1 mM); lane 5,
duplex treated with 1 (10 mM) and 2-mercaptoethanol (5 mM); lane
6, duplex treated with 1 (10 mM) and 2-mercaptoethanol (10 mM);
lane 7, duplex treated with 1 (5 mM) and 2-mercaptoethanol (10 mM);
lane 8, duplex treated with 1 (1 mM) and 2-mercaptoethanol (10 mM).
Figure 4. Mass spectrometric analysis of products obtained from the
reaction of 1 with glutathione in human hepatocytes.
Therefore, we examined the potential for 1 to act as a thiol-
triggered DNA alkylating agent. Toward this end, the 50-32P-
labeled oligonucleotide 50-GTTCG8TATATG7G6G5AG4G3-
TCG2CATG1TG-30 (where the underlined region is double
stranded, and the subscripts represent the numbering of guanine
residues in Figures 5-7) was treated with various concentrations
of 1 and 2-mercaptoethanol in 50 mM HEPES buffer, pH 7.0, at
37 ꢀC for 24 h. Maxam-Gilbert workup (500 mM piperidine, 90
ꢀC, 25 min)49 and sequencing gel analysis revealed strand
cleavage selectively at guanine residues in the reactions that
contained both 1 and thiol (Figure 5). Control reactions invol-
ving the incubation of DNA with 1 alone or thiol alone showed
little or no strand cleavage upon Maxam-Gilbert workup. A
control reaction in which 1 was preincubated with 2-mercap-
toethanol (37 ꢀC, 26 h) prior to the addition of the DNA also
generated little or no DNA damage above background. This
yield compounds with the general structure 6b. The fragmenta-
tion spectra of the two products were nearly identical, producing
m/z 259 as a major fragment ion, corresponding to the loss of
glutamate (-129 Da) from the parent ion (Figure 4B). The loss
of glycine (-75 Da) produced an ion at m/z 313 and represented
an additional fingerprint fragment consistent with glutathione
conjugation. We observed the same two glutathione conjugates
when cryopreserved human hepatocyte suspensions (1 ꢀ 106
cells/mL) were incubated with 1 (Supporting Information).
Thiol-Triggered DNA Alkylation by r-Bromo-2-cyclopen-
tenone (1). A number of compounds containing the R-halo-
acrolyl fragment (Figure 1) are cytotoxic or mutagenic and
may derive biological activity through reactions with DNA.41,42
222
dx.doi.org/10.1021/tx100282b |Chem. Res. Toxicol. 2011, 24, 217–228

Chemical Research in Toxicology
ARTICLE
3-bromo-2-butanone produces no strand cleavage above back-
ground levels (Supporting Information). A separate control
experiment shows that 2-cyclopentenone also does not generate
any lesions that are labile to Maxam-Gilbert workup. Finally,
we observed that 2-bromo-2-cyclohexenone generates thiol-
dependent, labile lesions comparable to 1 (data not shown).
Overall the structure-activity data is consistent with the involve-
ment of an episulfonium ion in thiol-triggered DNA alkyla-
tion by 1, rather than direct alkylation by 1, 2, or 6 (Supporting
Information). DNA alkylation by charged alkylators such as
episulfonium ions can be inhibited by added salt.24,60-63 We
find that the addition of salt decreases the yield of DNA alkyl-
ation by 1 and 2-mercaptoethanol. Specifically, the addition of
1 M NaClO4 caused a 27 ( 3% decrease in overall alkylation yield.
We also examined the ability of glutathione and cysteamine to
trigger DNA alkylation by 1. We find that the use of these
different thiols leads to different DNA alkylation yields and
different sequence specificities (Figure 6). This observation is
consistent with the generation of different intermediates in which
the side chain of the attacking thiol has been incorporated into
the ultimate DNA-alkylating species. The relative overall yields
for thiol-triggered alkylation of duplex DNA by glutathione,
cysteamine, and 2-mercaptoethanol are 1:5.7:8 in double-
stranded DNA. It is noteworthy that glutathione, the major
thiol present in mammalian cells, is a rather weak activator of
the DNA alkylation process measured in this particular assay.
The yields for the alkylation of single-stranded DNA by 1 using
glutathione, cysteamine, and 2-mercaptoethanol were 1:1.9:1.7.
Thiol-triggered alkylation of single-stranded DNA by 1 is less
efficient than the alkylation of duplex DNA. Previous studies
have shown that small, positively charged alkylating agents such
as aziridinium ions preferentially alkylate duplex DNA over
single-stranded DNA.63
Figure 6. Alkylation of DNA by 1 triggered by different thiols at pH 7.
DNA alkylation reactions were carried out and analyzed using poly-
acrylamide gel electrophoresis as described in the legend of Figure 5
and in the Experimental Procedures. The sequence and numbering of
the guanine residues in the oligonucleotide are as follows: 50-GTT-
CG8TATATG7G6G5AG4G3TCG2CATG1TG-30. The bars represent
residues G1 to G8 moving left to right. Bar heights represent the relative
alkylation yields at each G-residue. Abbreviations: GSH = glutathione;
2-ME = 2-mercaptoethanol; CA = cysteamine; SS = single strand DNA;
DS = double strand DNA.
Addition of amines to R-haloacrolyl systems can generate
aziridines in organic solvents.64 Aziridines have the potential to
alkylate DNA.65,66 Accordingly, we examined whether amines
activate DNA alkylation by 1. We found that the addition of
N-methylaniline or L-lysine did not activate DNA alkylation by 1
under the conditions identical to those employed for the thiol-
triggered DNA alkylation by 1 (<1% alkylation yields; see Sup-
porting Information).
Figure 7. Alkylation of duplex DNA by 1 triggered by different thiols at
different pH values. DNA alkylation reactions were carried out as
described in the legend of Figure 5 with the exception that the buffers
used in the pH 6.1 and 8.0 assays consisted of a HEPES (50 mM) and
PIPES (50 mM) mixture. Bar heights represent relative alkylation yields
at various G-residues in the oligonucleotide (the sequence and number-
ing of the guanine residues in the oligonucleotide are as follows: 50-
GTTCG8TATATG7G6G5AG4G3TCG2CATG1TG-30). The bars re-
present residues G1 to G8 moving left to right. Abbreviations: GSH =
glutathione; 2-ME = 2-mercaptoethanol; CA = cysteamine.
The enzyme glutathione S-transferase catalyzes the addition of
glutathione to various electrophiles.67 We examined whether
glutathione S-transferase had any effect on the yields of
glutathione-triggered DNA alkylation by 1. We found that addi-
tion of equine liver glutathione S-transferase had no significant
effect on the yield of DNA strand cleavage caused by glutathione
and 1 upon Maxam-Gilbert workup (Supporting Information).
We examined the effects of pH on thiol-triggered DNA al-
kylation by 1 (Figure 7). In the case of 2-mercaptoethanol and
cysteamine, alkylation yields were highest at pH 7.0, with lower
yields observed at pH values of 6.1 and 8.0. For 2-mercaptoetha-
nol, the relative DNA alkylation yields obtained at pH 6.1,
7.0, and 8.0 are 1.5:2.9:1. For cysteamine, the relative DNA
alkylation yields at pH 6.1, 7.0, and 8.0 are 2.6:5.1:1. In the case of
glutathione, total alkylation yields are comparable at pH values of
6.1 and 7.0, while the yield of alkylation at pH 8.0 is decreased by
a factor of 0.7 (relative to pH 7.0). The basis of the observed pH
dependence remains uncertain. There are a number of reaction
steps that could be pH sensitive or subject to acid-base catalysis,
including the addition of thiol to the haloketone and the attack of
water, thiol, and DNA on the resulting reactive intermediate.
indicates that DNA modification is due to a transient intermedi-
ate generated in the reaction of 1 with thiol rather than the final
reaction products.
DNA alkylation by 1 and 2-mercaptoethanol displays sub-
stantial sequence specificity, with approximately 15 times more
alkylation seen at the most favored site versus the least favored
site in the 20-deoxyoligonucleotide duplex used in these experi-
ments (Figures 5 and 6).
The initial 1,4-addition of thiols to 1 (Scheme 1) is expected to
yield an R-bromoketone intermediate 2.54 Previous work has
shown that R-bromoketones have the potential to alkylate DNA.57-59
Therefore, we examined the ability of a simple R-bromoketone,
3-bromo-2-butanone, to generate labile lesions at guanine
residues in the 32P-labeled DNA duplex. Under conditions
identical to those used for DNA alkylation by 1, we find that
223
dx.doi.org/10.1021/tx100282b |Chem. Res. Toxicol. 2011, 24, 217–228

Chemical Research in Toxicology
ARTICLE
Figure 8. Mass spectrometric analysis of guanine adducts resulting from thiol-triggered DNA alkylation by 1. (A) LC/MS of the thermally labile adducts
released from duplex DNA treated with 1 and 2-mercaptoethanol. (B) MS/MS analysis of the m/z 310 ion of the major (above) and minor (below)
adducts eluting at 2.47 and 3.22 min, in panel A. (C) MS3 of the m/z 232 ion for the major adduct eluting at 2.47 min in panel A.
We employed LC/MS/MS to characterize DNA alkylation
products generated by 1 in the presence of 2-mercaptoethanol.
Mixed sequence, double-stranded herring sperm DNA was treat-
ed with 1 in the presence of 2-mercaptoethanol, followed by
thermal workup (90 ꢀC, 45 min) to induce depurination of the
anticipated guanine adduct.21,68 LC/MS/MS analysis revealed
two adducts (eluting at 2.47 and 3.22 min) that produced an
[M þ H]þ at m/z 310 Da (Figure 8A). Formation of these
adducts was only observed when 1, thiol, and DNA were all
present in the reaction mixture.
Scheme 2
spectroscopy experiments (Figures 2 and 4) are envisioned to
arise via an elimination/ring-opening reaction of the episulfo-
nium ion 3 (Scheme 2). Compound 6 is not likely to arise via
direct displacement of the bromide because nucleophilic dis-
placement of vinyl halides by thiols typically does not occur
under mild conditions unless there is an electron-accepting
substituent in the in 2-position,73 although a mechanism involv-
ing 1,4-addition of thiol, followed by the displacement of bro-
mine by a second equivalent of thiol, and subsequent elimination
of thiol cannot be ruled out. Mass spectrometry does not rig-
orously define the regiochemistry of thiol attachment in products
6, but two lines of evidence lead us to suggest that this product is
formed predominantly with the sulfide residue in the R-position
relative to the carbonyl (Scheme 2). First, this is the regiochem-
istry observed for the analogous elimination/ring-opening reac-
tion of a presumed bromonium ion intermediate generated in the
reaction of Br2 with cyclopentenone.50 Second, we find that this
regioisomer is the major isolable product arising from a model
reaction of 2-mercaptoethanol with 1 (Scheme 2). It remains
Furthermore, MS3 analysis of the major adduct provided informa-
tion regarding the site at which thiol-activated 1 becomes attached to
the guanine residue. A mass transition from 232 f 215, correspond-
ing to the loss of ammonia (-17 Da) is observed for the major
adduct (Figure 8C). MS3 mass transitions of 232 f 204 and 189,
corresponding to a loss of CO and HNCO are also observed. Collect-
ively, these findings argue against covalent attachment at the N2, N1,
or O6 positions of the guanine residue in this particular adduct.
’ DISCUSSION
We find that R-bromoacrolyl-containing compounds are thiol-
triggered DNA-alkylating agents. Overall, the results of our
experiments are explained by a mechanism involving the initial
1,4-addition 54 of the thiol group to the R,β-unsaturated ketone
followed by intramolecular displacement of the halide by the
sulfide group in 2 to produce an episulfonium ion intermediate 3
(Scheme 1).69-72 Products such as 6 observed in our mass
224
dx.doi.org/10.1021/tx100282b |Chem. Res. Toxicol. 2011, 24, 217–228

Chemical Research in Toxicology
ARTICLE
possible that lower yields of other regioisomers also may be
formed in the model reaction. Indeed, mass spectrometric
analysis of the products generated in the incubation of 1 with
glutathione in human hepatocytes reveals two isobaric metab-
olites having a mass-to-charge ratio consistent with the structure
6b (Scheme 1). We suggest that these products may represent
two regioisomers of 6b. Elimination reactions that yield 6
presumably will limit the efficiency with which 1 alkylates
nucleic acids and other macromolecules in cells. Products with
mass-to-charge ratios corresponding to the structures 4a and 5a
(Scheme 1) produced in the reaction of 1 with the peptide
probe can be envisioned to arise via an attack of water and
inorganic phosphate buffer ion on the episulfonium ion 3,
though it remains possible that these products are produced by
an attack on 2.
When activated by thiols, 1 generates guanine lesions residues
in DNA that lead to strand cleavage upon Maxam-Gilbert
workup (Figures 5 and 6). It is well known that N7-alkylguanine
residues are labile to Maxam-Gilbert workup.34,49 Accordingly,
this result is consistent with the generation of a DNA-alkylating
episulfonium ion in the reactions of 1 with thiol because
alkylation of guanine residues can be considered a characteristic
reaction of episulfonium ions.24,74 For example, the chemistry
seen here is reminiscent of 1,2-dibromoethane. Addition of gluta-
thione or alkylguanine transferase (AGT) to 1,2-dibromoethane
yields a 2-bromoethylsulfide intermediate that generates a DNA-
alkylating episulfonium ion.55,56,62 In fact, the strand cleavage
efficiency of 1 is comparable to that of other small, bioactive
episulfonium ions.34,49,68,75,76 We find that the addition of
NaClO4 (1 M) caused a 27 ( 3% decrease in the yield of
DNA alkylation by 1 and 2-mercaptoethanol. Again, this result is
consistent with the involvement of an episulfonium ion inter-
mediate, as it has been shown that the reactions of positively
charged alkylating agents, including episulfonium ions, diazo-
nium ions, and aziridinium ions with DNA are inhibited by the
addition of salt.24,60-63 For example, previous work showed that
addition of NaClO4 (0.1 M) caused 18% and 27% decreases in
the yields of DNA alkylation by leinamycin and chloroethyl
ethylsulfide-derived episulfonium ions, respectively.24 The
most favored alkylation site 50-GGG (the underlined G is
alkylated) for thiol-activated 1 (Figure 5) is broadly consistent
with the sequence specificity expected for episulfonium ion
intermediates.24,75 For example, previous experiments have
shown that 50-GGG in duplex DNA was a preferred site for
alkylation by the episulfonium ion derived from chloroethylethyl
sulfide.24
the two products observed in our experiment (Figure 8) repre-
sent the regioisomeric guanine adducts 7 and 8 (Scheme 1). The
MS/MS spectrum of the major adduct (Figure 8B, upper
trace) reveals a strong signal for a fragment ion at m/z 152 Da
(≈100% relative abundance), corresponding to the libera-
tion of guanine and relatively weak signals corresponding to
the loss 2-mercaptoethanol (m/z 232 Da). This suggests that
the major adduct possesses the regiochemistry shown in
structure 7 for which the more facile loss of proton in the
position R to the carbonyl group leads to the elimination of
guanine. In the minor adduct, the relative abundance of the
fragments at m/z 232 (≈100%) and 152 (g40%) are reversed
(Figure 8B, lower trace). This is consistent with the proposed
structure 8, in which the elimination reaction involving the
acidic R-proton leads to the m/z 232 fragment via the loss of
the 2-mercaptoethanol unit. Episulfonium ions typically
react to give trans ring-opening products;77 however, it
remains possible that the two observed products represent
cis and trans isomers.
The MS3 mass transitions observed for the major adduct argue
against covalent attachment of the lesion at the N2, N1, or O6
positions of this particular guanine adduct that is released upon
thermal workup. This mass spectroscopic data along with the
gel electrophoresis results and literature precedents are consis-
tent with the N7-attachment proposed in structures 7 and 8
(Scheme 1). First, N7-guanine adducts are known to generate
strand cleavage with Maxam-Gilbert workup.68 Second, the N7
assignment is consistent with the literature showing that the N7-
atom of guanine residues is the predominant alkylation site for all
episulfonium ions that have been characterized,21,24,68,74,75,78-82
while alkylation at N3-G by an episulfonium ion has not been
observed.74 In fact, there are no examples of small alkylating
agents that selectively alkylate guanine residues at the N3-
position in duplex DNA. Small episulfonium ions typically alkylate
DNA at several sites in addition to guanine residues, and it is likely
that nonlabile adducts are formed in our experiments that are
invisible to the thermal workup employed in our mass spec-
trometric analyses and the Maxam-Gilbert workup employed in
our gel experiments.74
An interesting aspect of the proposed mechanism is that the
side chain of the activating thiol is incorporated into the putative
episulfonium ion reactive intermediate. The data in Figure 6
clearly shows that the properties of the DNA-alkylating inter-
mediate generated by the reaction of 1 with thiol vary depending
upon the nature of the thiol employed. This observation is
consistent with the involvement of different intermediates in
which the side chain of the attacking thiol becomes incorporated
into the ultimate DNA-alkylating species. The major thiol pres-
ent in mammalian cells is glutathione (Scheme 1). However,
some pathogens contain cationic or glycosylated thiols such as
N1-monoglutathionylspermidine (E. coli), mycothiol (Actino-
bacteria), and trypanothione (Trypanosoma and Leishmania).83,84
Thus, R-haloacrolyl derivatives will yield structurally distinct
episulfonium ions in different organisms. As a result, R-haloa-
crolyl-containing agents have the potential to display distinct
biological activities that are dependent upon the thiol content
of different organisms. In this regard, it may be significant that
glutathione, the major thiol present in mammalian cells, is a
rather weak activator of guanine alkylation by 1 (Figure 6),
while the cationic thiol cysteamine is a relatively efficient
activator.
It is important to note that the observed alkylation at guanine
residues could, in principle, be mediated by the R-haloketone
intermediate 2. Indeed, previous work has shown that R-bromo-
ketones have the potential to alkylate DNA.57-59 However,
control experiments showed that the R-bromoketone, 3-bromo-
2-butanone produces no strand cleavage above background
levels under the conditions employed for the assays with 1. A
separate control experiment showed that 2-cyclopentenone also
does not generate Maxam-Gilbert labile lesions in DNA. Over-
all, these structure-activity results are consistent with the notion
that thiol activation of 1 generates a DNA-alkylating episulfo-
nium ion (Scheme 1).
Mass spectrometric analysis of the nucleobase lesions released
upon thermolysis of duplex DNA treated with 1 and thiol
revealed the products with mass-to-charge ratios consistent with
the attachment of 3 to guanine (Scheme 1). We speculate that
225
dx.doi.org/10.1021/tx100282b |Chem. Res. Toxicol. 2011, 24, 217–228

Chemical Research in Toxicology
ARTICLE
A number of studies have characterized adducts formed by the
direct, in vitro reaction between R-haloacrolyl derivatives and
DNA (in the absence of thiols).85-88 The products resulting from
these direct reactions are distinct from those described here. For
example, the reaction of 2-bromoacrolein with 20-deoxycytidine
yields a cyclic, bromine-containing adduct resulting from the
addition of the exocyclic N4-nitrogen to the carbonyl carbon and
1,4-addition of the endocyclic N3 nitrogen to the olefin.85-88
Rate constants for these thiol-independent DNA alkylation
reactions have not been reported. To the best of our knowledge,
there are no reports indicating that such thiol-independent
adducts have been observed in cell culture or in vivo. The inte-
’ AUTHOR INFORMATION
Corresponding Author
*( J.S.D.) Phone: (615) 322-0673. Fax: (615) 875-5520. E-mail:
scott.daniels@vanderbilt.edu. (K.S.G.) Phone: (573) 882-6763.
Fax: (573) 882-2754. E-mail: gatesk@missouri.edu.
Present Addresses
^Vanderbilt University Medical Center, Department of Pharma-
cology, Vanderbilt Program in Drug Discovery, Nashville, TN
37232, U.S.A.
’ REFERENCES
rior of cells contains high concentrations (about 1-10 mM) of
(1) Gates, K. S. (1999) Covalent Modification of DNA by Natural
Products, in Comprehensive Natural Products Chemistry (Kool, E. T., Ed.)
pp 491-552, Pergamon, New York.
(2) Wolkenberg, S. E., and Boger, D. L. (2002) Mechanisms of in
situ activation for DNA-targeting antitumor agents. Chem. Rev. 102,
2477–2495.
(3) Tomasz, M. (1995) Mitomycin C: small, fast and deadly (but
very selective). Chem. Biol. 2, 575–579.
(4) Rockwell, S., Sartorelli, A. C., Tomasz, M., and Kennedy, K. A.
(1993) Cellular pharmacology of quinone bioreductive alkylating
agents. Cancer Metastasis Rev. 12, 165–176.
(5) Lown, W. J., Joshua, A. V., and Lee, J. S. (1982) Molecular
mechanisms of binding and single strand scission of DNA by the
antitumor antibiotics saframycin A and C. Biochemistry 21, 419–428.
(6) Myers, A. G., Cohen, S. B., Tom, N. J., Madar, D. J., and Fraley,
M. E. (1995) Insights into the mechanism of DNA cleavage by
dynemicin A as revealed by DNA-binding and -cleavage studies of
synthetic analogs. J. Am. Chem. Soc. 117, 7574–7575.
13,14
thiols,
and we find that the addition of thiols to the
R-haloacrolyl moiety is rapid, reaching completion in minutes
under physiologically relevant conditions (Figure 3). The results
of our experiments with human hepatic subcellular fractions (S9)
provide evidence that human cells metabolize 1 via thiol con-
jugation pathways, even when other metabolic machinery such as
cytochrome P450 enzymes are present. Thus, thiol-dependent
adduct formation may compete effectively with thiol-indepen-
dent adduct formation. Our data indicate that the covalent
adducts generated following the thiol activation of R-haloacrolyl
compounds are quite different from those generated by the direct
reaction of these agents with DNA.85-88 Therefore, the potential
influence of thiols on the spectrum of DNA adducts should be
considered if biomarkers are sought for assessing human expo-
sure to this class of molecules.
Elements of the proposed mechanism (Scheme 1) involving
thiol-mediated generation of an episulfonium ion may shed ad-
ditional light on the molecular mechanisms underlying the
bioactivity of medicinally interesting R-haloacrolyl-containing
compounds (Figure 1). For example, the anticancer drug candi-
date brostallicin causes cytotoxic DNA alkylation.41,42 Further-
more, the cell-killing activity of brostallicin is potentiated by
glutathione and glutathione S-transferase enzymes.41,42 It has
been proposed that the R-bromoketone unit generated by the
addition of glutathione to the R-bromoacrolyl fragment is the
active DNA alkylating intermediate (a species analogous to com-
pound 2 in Scheme 1).41,42 While this is a reasonable suggestion
because bromoacetyl moieties tethered to noncovalent DNA
binding units are able to alkylate DNA,57-59 our data indicate
that an episulfonium ion intermediate also should be considered
as a possible alkylating intermediate in thiol-triggered DNA
alkylation by the anticancer agent brostallicin.
Understanding the mechanisms by which xenobiotics are con-
verted to reactive species inside cells is important in the fields of
medicinal chemistry and chemical toxicology.1,2,34,68,89-92 The
results reported here provide new insight regarding the bioacti-
vation of R-haloacrolyl-containing compounds. These molecules
present the cell with a sort of molecular snare, where attempts to
detoxify the electrophilic R,β-unsaturated ketone via thiol con-
jugation actually generate an intermediate of increased rather
than decreased reactivity.
(7) McClelland, R. A., Panicucci, R., and Rauth, A. M. (1987)
Products of the reductions of 2-nitroimidazoles. J. Am. Chem. Soc. 109,
4308–4314.
(8) Ganley, B., Chowdhury, G., Bhansali, J., Daniels, J. S., and Gates,
K. S. (2001) Redox-activated, hypoxia-selective DNA cleavage by
quinoxaline 1,4-di-N-oxide. Bioorg. Med. Chem. 9, 2395–2401.
(9) Gong, J., Neels, J. F., Yu, X., Kensler, T. W., Peterson, L. A., and
Sturla, S. J. (2006) Investigating the role of stereochemistry in the
activity of anticancer acylfulvenes: synthesis, reductase-mediated bioac-
tivation, and cellular toxicity. J. Med. Chem. 49, 2593–2599.
(10) Woo, J., Sigurdsson, S. T., and Hopkins, P. B. (1993) DNA
interstrand crosslinking by reductively activated FR900482 and
FR66979. J. Am. Chem. Soc. 115, 1199–1200.
(11) Iyer, R. S., and Harris, T. M. (1993) Preparation of aflatoxin B1
8,9-epoxide using m-chloroperbenzoic acid. Chem. Res. Toxicol. 6, 313–
316.
(12) Woo, J., Sigurdsson, S. T., and Hopkins, P. B. (1993) DNA
interstrand cross-linking reactions of pyrrole-derived, bifunctional elec-
trophiles: evidence for a common target site in DNA. J. Am. Chem. Soc.
115, 3407–3415.
(13) Meister, A., and Anderson, M. E. (1983) Glutathione. Annu.
Rev. Biochem. 52, 711–760.
(14) Bellomo, G., Vairetti, M., Stivala, L., Mirabelli, F., Richelmi, P.,
and Orrenius, S. (1992) Demonstration of nuclear compartmentaliza-
tion of glutathione in hepatocytes. Proc. Natl. Acad. Sci. U.S.A. 89, 4412–
4416.
(15) Chatterjee, M., Cramer, K. D., and Townsend, C. A. (1993)
Kinetic nature of thiol activation in DNA cleavage by calicheamicin.
J. Am. Chem. Soc. 115, 3374–3375.
(16) Myers, A. G., Cohen, S. B., and Kwon, B. M. (1994) A study of
the reaction of calicheamicin g1 with glutathione in the presence of
double-stranded DNA. J. Am. Chem. Soc. 116, 1255–1271.
(17) Myers, A. G. (1987) Proposed structure of the neocarzinostatin
chromophore-methyl thioglycolate adduct; a mechanism for the nu-
cleophilic activation of neocarzinostatin. Tetrahedron Lett. 28, 4493–
4496.
’ ASSOCIATED CONTENT
S
Supporting Information. Mass spectral data, gel electro-
b
phesis data, and HPLC data for the reaction of glutathione with 1.
This material is available free of charge via the Internet at http://
pubs.acs.org.
226
dx.doi.org/10.1021/tx100282b |Chem. Res. Toxicol. 2011, 24, 217–228

Chemical Research in Toxicology
ARTICLE
(18) McMorris, T. C., Kelner, M. J., Wang, W., Moon, S., and Taetle,
R. (1990) On the mechanism of toxicity of illudins: the role of
glutathione. Chem. Res. Toxicol. 3, 574–579.
(19) Asai, A., Hara, M., Kakita, S., Kanda, Y., Yoshida, M., Saito, H.,
and Saitoh, Y. (1996) Thiol-mediated DNA alkylation by the novel
antitumor antibiotic leinamycin. J. Am. Chem. Soc. 118, 6802–6803.
(20) Breydo, L., Zang, H., Mitra, K., and Gates, K. S. (2001) Thiol-
independent DNA alkylation by leinamycin. J. Am. Chem. Soc. 123,
2060–2061.
(21) Nooner, T., Dutta, S., and Gates, K. S. (2004) Chemical
properties of the leinamycin 20-deoxyguanosine adduct. Chem. Res.
Toxicol. 17, 942–949.
(22) Shipova, K., and Gates, K. S. (2005) A fluorimetric assay for the
spontaneous release of an N7-alkylguanine residue from duplex DNA.
Bioorg. Med. Chem. Lett. 15, 2111–2113.
(39) Eder, E., Deininger, C., Deininger, D., and Weinfurtner, E.
(1994) Genotoxicity of 2-halosubstituted enals and 2-chloroacrylonitrile
in the Ames test and the SOS-chromotest. Mutat. Res. 322, 321–328.
(40) Shen, Y.-C., Cheng, Y.-B., Lin, Y.-C., Guh, J.-H., Teng, C.-M.,
and C.-L., K. (2004) New prostanoids with cytotoxic activity from
Taiwanese octocoral Clavularia viridis. J. Nat. Prod. 67, 542–546.
(41) Beria, I., Baraldi, P. G., Cozzi, P., Calderelli, M., Geroni, C.,
Marchini, S., Mongelli, N., and Romagnoli, R. (2004) Cytotoxic alpha-
haloacrylic derivatives of distamycin A and congeners. J. Med. Chem. 47,
2611–2623.
(42) Geroni, C., Marchini, S., Cozzi, P., Galliera, E., Ragg, E.,
Colombo, T., Battaglia, M. H., D'Incalci, M., and Broggini, M. (2002)
Brostallicin, a novel anticancer agent whose activity is enhanced upon
binding to glutathione. Cancer Res. 62, 2332–2336.
(43) Fattorusso, E., Taglialatela-Scafati, O., Petrucci, F., Bavestrello,
G., Calcinai, B., Cerrano, C., Di Meglio, P., and Ianaro, A. (2004)
Polychlorinated androstanes from the burrowing sponge Cliona nigri-
cans. Org. Lett. 6, 1633–1635.
(23) Zang, H., Breydo, L., Mitra, K., Dannaldson, J., and Gates, K. S.
(2001) DNA alkylation by leinamycin can be triggered by cyanide and
phosphines. Bioorg. Med. Chem. Lett. 11, 1511–1515.
(24) Zang, H., and Gates, K. S. (2003) Sequence specificity of DNA
alkylation by the antitumor natural product leinamycin. Chem. Res.
Toxicol. 16, 1539–1546.
(44) LaLonde, R. T., and Ramdayal, F. (1997) Mucochloric acid
action on pX174 DNA: A comparison to other chlorine-substituted
2(5H)-furanones. Chem. Res. Toxicol. 10, 205–210.
(25) Behroozi, S. J., Kim, W., Dannaldson, J., and Gates, K. S. (1996)
DNA cleavage by 1,2-dithiolan-3-one 1-oxides: a class of thiol-activated
DNA cleaving agents structurally related to the antitumor antibiotic
leinamycin. Biochemistry 35, 1768–1774.
(26) Mitra, K., Kim, W., Daniels, J. S., and Gates, K. S. (1997)
Oxidative DNA cleavage by the antitumor antibiotic leinamycin and
simple 1,2-dithiolan-3-one 1-oxides: Evidence for thiol-dependent con-
version of molecular oxygen to DNA-cleaving oxygen radicals mediated
by polysulfides. J. Am. Chem. Soc. 119, 11691–11692.
(45) Perry, N. B., Blunt, J. W., and Munro, M. H. G. (1988)
Cytotoxic pigments from New Zealand sponges of the genus Latrunculia:
discorhabdins A, B, and C. Tetrahedron 44, 1727–1734.
(46) Teeyapant, R., Woerdenbag, H. J., Kreis, P., Hacker, J., Wray, V.,
Witte, L., and Proksch, P. (1993) Antibiotic and cytotoxic activity of
brominated compounds from the marine sponge Verongia aerophoba. Z.
Naturforsch., C: J. Biosci. 48, 939–945.
(47) Woolf, T., Ed. (1999) Handbook of Drug Metabolism, Marcel
Dekker, Inc., New York.
(27) Viswesh, V., Gates, K. S., and Sun, D. (2010) Characterization
of DNA damage induced by a natural product antitumor antibiotic
leinamycin in human cancer cells. Chem. Res. Toxicol. 23, 99–107.
(28) Chatterji, T., and Gates, K. S. (1998) DNA cleavage by
7-methylbenzopentathiepin: a simple analog of the antitumor agent
varacin. Bioorg. Med. Chem. Lett. 8, 535–538.
(29) Chatterji, T., and Gates, K. S. (2003) Reaction of thiols with
7-methylbenzopentathiepin. Bioorg. Med. Chem. Lett. 13, 1349–1352.
(30) Chatterji, T., Keerthi, K., and Gates, K. S. (2005) Generation of
reactive oxygen species by a persulfide (BnSSH). Bioorg. Med. Chem.
Lett. 15, 3921–3924.
(31) Wang, Y., Koreeda, M., Chatterji, T., and Gates, K. S. (1998)
Total synthesis and DNA-cleaving properties of thiarubrine C. J. Org.
Chem. 63, 8644–8645.
(32) Paz, M., and Tomasz, M. (2001) Reductive activation of
mitomycin A by thiols. Org. Lett. 3, 2789–2792.
(48) Sambrook, J., Fritsch, E. F., and Maniatis, T. (1989) Molecular
Cloning: A Lab Manual, Cold Spring Harbor Press, Cold Spring Harbor,
NY.
(49) Maxam, A. M., and Gilbert, W. (1980) Sequencing end-labeled
DNA with base-specific chemical cleavages. Methods Enzymol. 65, 499.
(50) Harmata, M., Barnes, C. L., Brackley, J., Bohnert, G., Kirchhoefer,
P., K€urti, L., and Rashatasakhon, P. (2001) Generation of cyclo-
pentadienones from 2-bromocyclopentenones. J. Org. Chem. 66, 5232–
5236.
(51) Smith, A. B. I., Branca, M. A., Guaciaro, M. A., and Wovkulich,
P. M. (1983) 2-Hydroxymethyl-2-cyclopentenone. Org. Syn. 61, 65–70.
(52) Mitchell, M. D., Elrick, M. M., Walgren, J. L., Mueller, R. A.,
Morris, D. L., and Thompson, D. C. (2008) Peptide-based in vitro assay
for the detection of reactive metabolites. Chem. Res. Toxicol. 21, 859–
868.
(53) Itami, K., Mitsudo, K., Fujita, K., Ohashi, Y., and Yoshida, J.
(2004) Catalytic intermolecular Pauson-Khand-type reaction: strong
directing effect of pyridylsilyl and pyrimidylsilyl groups and isolation of
Ru complexes relevant to catalytic reaction. J. Am. Chem. Soc. 126,
11058–11066.
(33) Wang, S., and Kohn, H. (1999) Studies on the mode of action of
mitomycin C (7) aminoethylene disulfides (BMS-181174 and KW-
22149): reactivity of 7-N-mercaptoethyl)mitomycin C. J. Org. Chem. 42,
788–790.
(34) Gates, K. S. (2009) An overview of chemical processes that
damage cellular DNA: spontaneous hydrolysis, alkylation, and reactions
with radicals. Chem. Res. Toxicol. 22, 1747–1760.
(54) Khatik, G. L., and Chakraborti, A. K. (2006) Catalyst-free
conjugated addition of thiols to R,β-unsaturated carbonyl compounds
in water. Org. Lett. 8, 2433–2436.
(35) Antunes, E. M., Beukes, D. R., Kelly, M., Samaai, T., Barrows,
L. R., Marshall, K. M., Sincich, C., and Davies-Coleman, M. T. (2004)
Cytotoxic pyrroloiminoquinones from four new species of South African
Latrunculid sponges. J. Nat. Prod. 67, 1268–1276.
(36) Radisky, D. C., Radisky, E. S., Barrows, L. R., Copp, B. R.,
Kramer, R. A., and Ireland, C. M. (1993) Novel cytotoxic topoisomerase
II inhibiting pyrroloiminoquinones from Fijian sponges of the genus
Zyzzya. J. Am. Chem. Soc. 115, 1632–1638.
(37) Amagata, T., Usami, Y., Minoura, K., Ito, T., and Numata, A.
(1998) Cytotoxic substances produced by a fungal strain fromo a
sponge: physico-chemical properties and structures. J. Antibiot. 51,
33–40.
(55) Esterbauer, H., Ertl, A., and Scholz, N. (1976) The reaction of
cysteine with R,β-unsaturated aldehydes. Tetrahedron 32, 285–289.
(56) Tacka, K. A., Dabrowiak, J. C., Goodisman, J., and Souid, A.-K.
(2002) Kinetic analysis of the reactions of 4-hydroperoxycyclopho-
sphamide and acrolein with glutathione, mesna, and Wr-1065. Drug
Metab. Dispos. 30, 875–882.
(57) Povsic, T. J., and Dervan, P. B. (1990) Sequence-specific
alkylation of double-helical DNA by oligonucleotide-directed triple-
helix formation. J. Am. Chem. Soc. 112, 9428–9430.
(58) Baker, B. F., and Dervan, P. B. (1989) Sequence-specific
cleavage of DNA by N-bromoacetyldistamycin. Product and kinetic
analysis. J. Am. Chem. Soc. 111, 2700–2712.
(38) Higa, T., Okuda, R. K., Severns, R. M., Scheuer, P. J., He, C. H.,
Xu, C., and Clardy, J. (1987) Unprecedented constituents of a new
species of acorn worm. Tetrahedron 43, 1063–1070.
(59) Taylor, M. J., and Dervan, P. B. (1997) Kinetic analysis of
sequence-specific alkylation of DNA by pyrimidine oligodeoxyribonu-
cleotide-directed triple-helix formation. Bioconjugate Chem. 8, 354–364.
227
dx.doi.org/10.1021/tx100282b |Chem. Res. Toxicol. 2011, 24, 217–228

Chemical Research in Toxicology
ARTICLE
(60) Hartley, J. A., Forrow, S. M., and Souhami, R. L. (1990) Effect of
ionic strength and cationic DNA affinity binders on the DNA sequence-
selective alkylation of guanine N7-positions by nitrogen mustards.
Biochemistry 29, 2985–2991.
(61) Liang, G., Encell, L., Nelson, M. G., Switzer, C., Shuker,
D. E. G., and Gold, B. (1995) Role of electrostatics in the sequence-
selective reaction of charged alkylating agents with DNA. J. Am. Chem.
Soc. 117, 10135–10136.
(62) Wurdeman, R. L., and Gold, B. (1988) The effect of DNA
sequence, ionic strength, and cationic DNA affinity binders on the
methylation of DNA by N-methyl-N-nitrosourea. Chem. Res. Toxicol. 1,
146–147.
(80) Lawley, P. D., and Phillips, D. H. (1996) DNA adducts from
chemotherapeutic agents. Mutat. Res. 355, 13–40.
(81) Povirk, L. F., and Shuker, D. E. (1994) DNA damage and
mutagenesis induced by nitrogen mustards. Mutat. Res. 318, 205–226.
(82) Mattes, W. B., Hartley, J. A., and Kohn, K. W. (1986) DNA
sequence selectivity of guanine-N7 alkylation by nitrogen mustards.
Nucleic Acids Res. 14, 2971–2987.
(83) Fahey, R. C. (2001) Novel thiols of prokaryotes. Annu. Rev.
Microbiol. 55, 333–356.
(84) Hand, C. E., and Honek, J. F. (2005) Biological chemistry of
naturally occurring thiols of microbial and marine origin. J. Nat. Prod. 68,
293–308.
(63) Gold, B., Marky, L. M., Stone, M. P., and Williams, L. D. (2006)
A review of the role of the sequence-dependent electrostatic landscape in
DNA alkylation patterns. Chem. Res. Toxicol. 19, 1402–1414.
(64) Mekonnen, A., and Carlson, R. (2006) Phase transfer catalyzed
aziridination of R-bromo-2-cyclopenten-1-one. Tetrahedron 62, 852–
856.
(65) Masta, A., Gray, P. J., and Phillips, D. R. (1994) Molecular basis
of nitrogen mustard effects on transcription processes: role of depurina-
tion. Nucleic Acids Res. 22, 3880–3886.
(66) Cussac, C., and Laval, F. (1996) Reduction of the toxicity and
mutagenicity of aziridine in mammalian cells harboring the Escherichia
coli fpg gene. Nucleic Acids Res. 24, 1742–1746.
(67) Armstrong, R. N. (1997) Structure, catalytic mechanism, and
evolution of the glutathione transferases. Chem. Res. Toxicol. 10, 2–18.
(68) Gates, K. S., Nooner, T., and Dutta, S. (2004) Biologically
relevant chemical reactions of N7-alkyl-20-deoxyguanosine adducts in
DNA. Chem. Res. Toxicol. 17, 839–856.
(69) Ogston, A. G., Holiday, E. R., Philpot, J. S. L., and Stocken, L. A.
(1948) The replacement reactions of beta, beta0-dichlorodiethyl sulf-
phide and of some analogues in aqueous solution: the isolation of beta-
chloro-beta0-hydroxy diethyl sulfphide. Trans. Faraday Soc. 44, 45–52.
(70) Yang, Y.-C., Szafraniec, L. L., Beaudry, W. T., and Ward, J. R.
(1988) Kinetics and mechanism of the hydrolysis of 2-chloroethyl
sulfides. J. Org. Chem. 53, 3293–3297.
(71) Thier, R., Muller, M., Taylor, J. B., Pemble, S. E., Ketterer, B.,
and Guengerich, F. P. (1995) Enhancement of bacterial mutagenicity of
bifunctional alkylating agents by expression of mammalian glutathione
S-transferases. Chem. Res. Toxicol. 8, 465–472.
(85) Meerman, J. H. N., Smith, T. R., Pearson, P. G., Meier, G. P.,
and Nelson, S. D. (1989) Formation of cyclic 1-N2-propanodeoxygua-
nosine and thymidine adducts in the reaction of the mutagen 2-bro-
moacrolein with calf thymus DNA. Cancer Res. 49, 6174–6179.
(86) Chenna, A., and Iden, C. R. (1993) Characterization of 20-
deoxycytidine and 20-deoxyuridine adducts formed in reactions with
acrolein and 2-bromoacrolein. Chem. Res. Toxicol. 6, 261–268.
(87) Le Curieux, F., Munter, T., and Kronberg, L. (1997) Identifica-
tion of adenine adducts formed in the reaction of calf thymus DNA with
mutagenic chlorohydroxyfuranones found in drinking water. Chem. Res.
Toxicol. 10, 1180–1185.
(88) Asplund, D., Kronberg, L., Sj€oholm, R., and Munter, T. (1995)
Reaction of mucochloric acid with adenosine: formation of 8-(N6-
adenosinyl)ethenoadenosine derivatives. Chem. Res. Toxicol. 8, 841–846.
(89) Gates, K. S. (2000) Mechanisms of DNA damage by leinamy-
cin. Chem. Res. Toxicol. 13, 953–956.
(90) Gates, K. S. (2007) Chemical Reactions of DNA Damage and
Degradation, in Reviews of Reactive Intermediates (Platz, M. S., Moss,
R. A. Jones, and M. J., Eds.) pp 333-378, John Wiley and Sons, Inc.,
Hoboken, NJ.
(91) Guengerich, F. P. (2000) Metabolism of chemical carcinogens.
Carcinogenesis 21, 345–351.
(92) Dahlmann, H. A., Vaidyanathan, V. G., and Sturla, S. J. (2009)
Investigating the biochemical impact of DNA damage with structure-
based probes: abasic sites, photodimers, alkylation adducts, and oxida-
tive lesions. Biochemistry 48, 9347–9359.
(72) Valadez, J. G., Liu, L., Loktionova, N. A., Pegg, A. E., and
Guengerich, F. P. (2004) Activation of bis-electrophiles to mutagenic
conjugates by human O6-alkylguanine-DNA transferase. Chem. Res.
Toxicol. 17, 972–982.
(73) Bernasconi, C. F., Killion, R. B. J., Fassberg, J., and Rappoport,
Z. (1989) Kinetics of reaction of β-methoxy-R-nitrostilbene with
thiolate ions. first direct observation of the intermediate in a nucleophilic
vinylic substitution. J. Am. Chem. Soc. 111, 6862–6864.
(74) Fidder, A., Moes, G. W. H., Scheffer, A. G., ver der Schans, G. P.,
Baan, R. A., de Jong, L. P. A., and Benschop, H. P. (1994) Synthesis,
characterization, and quantitation of the major adducts formed between
sulfur mustard and DNA of calf thymus and human blood. Chem. Res.
Toxicol. 7, 199–204.
(75) Humphreys, W. G., Kim, D.-H., Cmarik, J. L., Shimada, T., and
Guengerich, F. P. (1990) Comparison of the DNA-alkylating properties
and mutagenic responses of a series of S-(haloethyl)-substituted cysteine
and glutathione derivatives. Biochemistry 29, 10342–10350.
(76) Mattes, W. B., Hartley, J. A., and Kohn, K. W. (1986) Mechan-
ism of DNA strand breakage by piperidine at sites of N7 alkylation.
Biochim. Biophys. Acta 868, 71–76.
(77) Mueller, W. H., and Butler, P. E. (1968) Factors influencing the
nature of the episulfonium ion in sulfenyl chloride addition to terminal
olefins. J. Am. Chem. Soc. 90, 2075–2081.
(78) Pullman, A., and Pullman, B. (1981) Molecular electrostatic
potential of the nucleic acids. Q. Rev. Biophys. 14, 289–380.
(79) Beranek, D. T. (1990) Distribution of methyl and ethyl adducts
following alkylation with monofunctional alkylating agents. Mutat. Res.
231, 11–30.
228
dx.doi.org/10.1021/tx100282b |Chem. Res. Toxicol. 2011, 24, 217–228