Detection and quantification of 5-chlorocytosine in DNA by stable isotope dilution and gas chromatography/negative ion chemical ionization/mass spectrometry

Article abstract of DOI:10.1021/tx015578g

Hypochlorous acid (HOCl) is generated from activated phagocytes during infections and inflammation. One of the major products of HOCl reaction with DNA was 5-chlorocytosine (5Cl-Cyt). In this report, a gas chromatography/negative ion chemical ionization/m

Full text of DOI:10.1021/tx015578g

262
Chem. Res. Toxicol. 2002, 15, 262-268
Detection a n d Qu a n tifica tion of 5-Ch lor ocytosin e in DNA
by Sta ble Isotop e Dilu tion a n d Ga s Ch r om a togr a p h y/
Nega tive Ion Ch em ica l Ion iza tion /Ma ss Sp ectr om etr y
Hauh-J yun Candy Chen,* Shin-Wei Row, and Chia-Liang Hong
Department of Chemistry, National Chung Cheng University, 160 San-Hsing, Ming-Hsiung,
Chia-Yi 62142, Taiwan
Received October 25, 2001
Hypochlorous acid (HOCl) is generated from activated phagocytes during infections and
inflammation. One of the major products of HOCl reaction with DNA was 5-chlorocytosine
(5Cl-Cyt). In this report, a gas chromatography/negative ion chemical ionization/mass
spectrometry (GC/NICI/MS) assay with stable isotope dilution was developed for detection and
quantification of 5Cl-Cyt in DNA. During hydrolysis of DNA, 5Cl-Cyt undergoes spontaneous
deamination quantitatively forming 5-chlorouracil (5Cl-Ura). The stable isotope of 5Cl-Ura
with six mass units higher than the normal 5Cl-Ura was synthesized and used as internal
standard of the assay. The adduct-enriched fraction of DNA hydrolysate was derivatized with
pentafluorobenzyl bromide before GC/NICI/MS analysis with selected ion monitoring at [M -
181]- fragments of bispentafluorobenzylated 5Cl-Ura and its isotope analogue. The limit of
detection was 20 amol (S/N ) 8) of bispentafluorobenzylated 5Cl-Ura injected on column with
selective ion monitoring mode and the limit of quantification for the entire assay was 14 fmol
of 5Cl-Cyt. Analysis of hypochlorous acid-treated calf thymus DNA by both GC/NICI/MS and
HPLC/UV detection provided similar adduct levels and thus verified this new GC/NICI/MS
assay. Using this highly specific and ultrasensitive GC/NICI/MS method, the levels of 5Cl-Cyt
in untreated calf thymus DNA and human placental DNA were determined as 0.6 and 6.6
adducts per 107 normal cytosine, respectively. Peroxynitrite also contributed to 5Cl-Cyt
formation in DNA. Level of 5Cl-Cyt in DNA treated with peroxynitrite in the presence of
chloride was higher than that without addition of chloride. Thus, quantification of 5Cl-Cyt in
DNA by this isotope dilution GC/NICI/MS assay may facilitate research on the role of DNA
chlorination in carcinogenesis and in cancer development.
human neutrophils (16), and it is capable of chlorination,
nitration, and oxidation of DNA, protein, and low-density
lipoproteins in vitro (17-21). The level of 5Cl-Ura in
HOCl-treated DNA increased in the presence of nitrite
compared to HOCl alone, indicating that NO2Cl was a
better chlorinating species than HOCl (17). Thus, 5Cl-
Cyt is not a specific biomarker for HOCl, but it could be
considered as a biomarker for “reactive chlorine species”.
Chlorination of adenine was also found in DNA exposed
to HOCl (21), but the amount of 8-chloroadenine was
much lower than 5Cl-Cyt. Furthermore, only 5Cl-Cyt, not
8-chloroadenine, formation was significantly increased
when HOCl was added in the presence of nitrite (17).
Detection of 5Cl-Cyt in vivo was first reported in
salmon sperm DNA (22). It was not clear whether 5Cl-
Cyt was formed endogenously or from molecular chlorine
(Cl2) in processing of water (23). Exposure of human
respiratory epithelial cells to both HOCl and nitrite
significantly increased the yield of 5Cl-Cyt compared to
HOCl treatment alone (24). Henderson et al. detected
5Cl-Cyt in RNA, but not DNA, of intact Escherichia coli
exposed to the myeloperoxidase/H2O2/Cl- system (25).
However, they cannot conclude that the chlorinating
agent did not reach the nucleus and react with DNA since
the limit of quantification was approximately 1 in 104
nucleobases. Apparently, a highly sensitive and specific
assay for 5Cl-Cyt in cellular or tissue DNA is demanded.
In tr od u ction
Upon stimulation, neutrophils and phagocytes excrete
hypochlorous acid (HOCl)1 as a defense mechanism for
killing the invading microorganisms. Formation of HOCl
at the sites of infection and inflammation is through
oxidation of chloride (Cl-) by hydrogen peroxide (H2O2)
catalyzed by the heme enzyme myeloperoxidase (MPO)
(1, 2). Being a bactericidal agent, HOCl also damages the
host cells by reaction with various biological molecules,
including protein, DNA, carbohydrates, lipids, ascorbate,
GSH, and NADH (3-11). It also leads to formation of
cross-links between protein and DNA (12). The major
reactions of DNA with HOCl include oxidation of pyri-
midines, but not purines, and chlorination of cytosine.
5-Chlorocytosine (5Cl-Cyt) was referred as a “fingerprint”
for damage of DNA by HOCl (13). Reaction of HOCl with
nitrite, of which the concentration is elevated during
inflammation (14, 15), forms nitryl chloride (NO2Cl).
Nitryl chloride has been shown to form in activated
* To whom correspondence should be addressed. Fax: (886) 5-272-
1040. E-mail: chehjc@ccunix.ccu.edu.tw.
1
Abbreviations: TBDMS, t-butyldimethylsilyl; 5Cl-Cyt, 5-chloro-
cytosine; 5Cl-dCyd, 5-chloro-2′-deoxycytidine; 5Cl-dUrd, 5-chloro-2′-
deoxycytidine; 5Cl-Ura, 5-chlorouracil; HOCl, hypochlorous acid; MPO,
myeloperoxidase; NICI, negative ion chemical ionization; NO2Cl, nitryl
chloride; PFB, pentafluorobenzyl; SIM, selective ion monitoring; SPE,
solid-phase extraction; TMS, trimethysilyl.
10.1021/tx015578g CCC: $22.00 © 2002 American Chemical Society
Published on Web 02/01/2002

Analysis of 5Cl-Cyt in DNA by GC/ NICI/ MS
Chem. Res. Toxicol., Vol. 15, No. 2, 2002 263
Syn th esis of Isotop e-La beled 5Cl-Ur a In ter n a l Sta n -
d a r d . To a solution containing [13C4,15N2]Ura (1.3 mM) was
added NaCl (200 mM) and NaOCl (2.1 mM, final concentration).
The reaction mixture was adjusted to pH 2.0, shaked at 37 °C
for 1 h, and quenched with methionine (2.1 mM). The reaction
mixture was purified by a C18-OH SPE (500 mg, 3 mL) precon-
dition with 13.5 mL of methanol, followed by 13.5 mL of water.
The SPE column was washed with 2.7 mL of water and the
fraction containing [13C4,15N2]5Cl-Ura was eluted with 2.7 mL
of water. The eluant was dried and analyzed by HPLC using
system 1 for its chemical purity. The yield (59%) was quantified
based on the molar UV absorbance of 5Cl-Ura at 280 nm. The
isotopic purity was confirmed by GC/NICI/MS after pentafluoro-
benzylation of 1.0 ng of [13C4,15N2]5Cl-Ura. No peak corresponding
to the pentafluorobenzylated 5Cl-Ura at m/z 325 was detected.
In this study, an isotope dilution GC negative ion
chemical ionization (NICI) mass spectroscopic assay is
developed for quantification of 5Cl-Cyt in DNA. With
incorporation of stable isotope of the analyte as internal
standard and labeling the analyte with electrophore,
followed by GC/NICI/MS analysis, this assay offers high
sensitivity and specificity. The level of 5Cl-Cyt in un-
treated calf thymus DNA is accurately quantified. Per-
oxynitrite is another strong oxidant released in inflamed
tissues (26, 27), and 5Cl-Cyt is first found in peroxy-
nitrite-treated DNA in the presence of Cl-. The present
work provides a useful tool in investigating the role of
cytosine chlorination in DNA in in vivo experiments.
Syn th esis of N1,N3-Bis(p en ta flu or oben zyl)-5-ch lor ou r a -
cil Der iva tive. To a 20 mL vial containing dried 5Cl-Ura (3.0
mg, 21 µmol) was added a solution of diisopropylethylamine (280
µL, 1.6 mol) and PFB-Br (50 µL, 310 mmol) in 0.6 mL anhydrous
methanol and the reaction mixture was incubated at 42 °C for
2 h. The reaction mixture was evaporated under vacuum and
collected using HPLC system 3 monitoring at 270 nm to afford
2.7 mg of (PFB)2-5Cl-Ura (77% yield) as a white solid. 1H NMR
(CDCl3, 400 MHz) δ 5.01 (s, 2H, benzyl), 5.23 (s, 2H, benzyl),
7.49 (s, 1H, C6-H). NICI-MS (assignment and relative abun-
dance in parentheses) 291 ([MH - PFB - Cl]-, 56%), 325 ([M
- PFB]-, 100%), 327 (39%), 486 ([M - HF]-, 5.6%), 506 (M-,
2.3%).
Mod ifica tion of Ca lf Th ym u s DNA w ith HOCl. Two
conditions were performed to obtain 5Cl-Cyt-containing calf
thymus DNA. Condition A. Calf thymus DNA (2.48 mg/mL) in
potassium phosphate buffer (0.2 M, pH 7.4) was incubated with
NaOCl (1.74 mM) and NaCl (100 mM) at 37 °C for 20 min,
followed by addition of methionine (17.4 mM) and stand at room
temperature for 30 min to quench the reaction. Condition B.
Calf thymus DNA (0.5 mg/mL) in potassium phosphate buffer
(0.2 M, pH 7.4) was incubated with HOCl (0.35 mM) and NaCl
(100 mM) at 37 °C for 60 min, followed by addition of methionine
(3.5 mM).
Exp er im en ta l P r oced u r es
Ma ter ia ls. Calf thymus DNA, human placental DNA, 5-chlo-
rouracil (5Cl-Ura), cytosine, 2′-deoxycytidine, and methionine
were from Sigma Chemical Co. (St. Louis, MO). [13C4,5N2]Uracil
(U-13C4, 99%, U-5N2, 98%) was purchased from Cambridge
Isotope Laboratories (Andover, MA). Diisopropylethylamine,
2,3,4,5,6-pentafluorobenzyl bromide (PFB-Br), anhydrous metha-
nol, and anhydrous phosphorus pentoxide were obtained from
Aldrich Chemical Co. (Milwaukee, WI). Bond Elut C18-OH and
Si solid-phase extraction (SPE) columns were from Varian
(Harbor City, CA). The concentration of NaOCl was determined
by the absorbance at 292 nm (pH 12, ꢀ ) 350 M-1 cm-1) (28).
Peroxynitrite was synthesized according to the previously
described procedures using isoamylnitrite and hydrogen per-
oxide (29) and was stored at -80 °C. The concentration of
peroxynitrite was determined by the absorbance at 302 nm in
1 N NaOH (ꢀ ) 1670 M-1 cm-1) (30).
In str u m en ts. NMR spectra were recorded on a Bruker DPX
400 MHz (Billerica, MA) instrument. GC/NICI/MS experiments
were conducted using a Hewlett-Packard 6890 GC with 5973
MSD mass detector (Agilent, Taiwan).
Liqu id Ch r om a togr a p h y. HPLC system was equipped with
Hitachi L-7000 pump system with D-7000 interface, a Rheodyne
injector, and L-7450A photodiode array detector (Hitachi,
Taiwan). System 1. A Prodigy ODS (3) 250 × 4.6 mm 5 µm
column was eluted with a linear gradient of solvents A and B:
0-20 min, 0% B; 20-40 min, 0 to 100% B (solvent A, 50 mM
ammonium formate, pH 4.0; solvent B, methanol) at a flow rate
of 1.0 mL/min. System 2. A Prodigy ODS (3) 250 × 4.6 mm 5
µm column (Phenomenex, Torrance, CA) was eluted at a linear
gradient of water and methanol: 0-10 min, 5% methanol; 10-
20 min, 5 to 25% methanol; 20-30 min, 25 to 100% methanol
at a flow rate of 1.0 mL/min. System 3. A Prodigy ODS (3) 250
× 4.6 mm 5 µm column was eluted with a linear gradient of
water and methanol: 0-35 min: 0 to 100% methanol at a flow
rate of 1.0 mL/min. System 4. A Prodigy ODS (3) 250 × 4.6 mm
5 µm column was eluted with 50 mM ammonium formate (pH
4.0) at a flow rate of 1.0 mL/min.
P er oxyn itr ite-Tr ea ted DNA. Calf thymus DNA (0.5 mg/
mL) in potassium phosphate buffer (0.2 M, pH 7.25) with or
without NaCl (100 mM) was added peroxynitrite (final concen-
tration 0.35 mM, final pH 7.4) and incubated at 37 °C for 60
min, followed by addition of methionine (3.5 mM) and stand at
room temperature for 30 min.
DNA Hyd r olysis a n d Ad d u ct En r ich m en t by C18-OH
SP E. DNA sample was evaporated, added [13C4,15N2]5Cl-Ura
(1.0 ng) and 60% formic acid (0.5 mL) and heated at 150 °C for
45 min. The hydrolysate was evaporated to dryness, reconsti-
tuted in a 1.0 mL aqueous solution (pH 4.0), and enriched by a
C18-OH SPE column preconditioned with 13.5 mL of methanol,
followed by 13.5 mL of ammonium formate (50 mM, pH 4.0)
solution. The column was washed with 2.7 mL of ammonium
formate (50 mM, pH 4.0) and collected the second 2.7 mL of
ammonium formate solution.
F or m a t ion of 5Cl-Cyt . To a solution containing Cyt (1.0
mM) in potassium phosphate buffer (0.2 M) was added NaOCl
(1.0 mM) or peroxynitrite (20 mM) in the presence or absence
of NaCl (100 mM) with a final pH of 7.4. The reaction mixture
was incubated at 37 °C for 20 min, followed by immediate
analysis by HPLC using system 1 monitoring at 270 and 280
nm. The reaction of Cyt with H2O2 was performed under similar
conditions except that the concentration of H2O2 was 100 mM
and the incubation time was 1 day.
P en ta flu or oben zyla tion a n d Si SP E Colu m n . The eluant
was evaporated and dried over phosphorus pentoxide, followed
by addition of an anhydrous methanolic solution (0.3 mL)
containing pentafluorobenzyl bromide (40 µL) and diisopropy-
lethylamine (57 µL) under argon atmosphere. The reaction
mixture was incubated at 42 °C for 2 h, followed by evaporation
to dryness. The residue was added dichloromethane (0.4 mL)
and purified by Si SPE column preconditioned with 15 mL of
dichloromethane. The column was washed with 1 mL of dichlo-
romethane and eluted with 2 mL of dichloromethane. The eluant
was evaporated, transferred to an insert, and evaporated to
dryness. The residue was dissolved in 10 µL of acetone and 1
µL aliquot was analyzed by GC/NICI/MS with selective ion
monitoring (SIM) at m/z 325 and 331 for PFB2-5Cl-Ura and
Syn th esis of 5Cl-Cyt a n d 5Cl-d Cyd . To a solution contain-
ing Cyt or dCyd (1.0 mM) and NaCl (100 mM) at pH 2.0 was
added NaOCl (1.5 mM), and the pH was adjusted to 2.0. The
reaction was incubated at 37 °C for 1 h in a shaking water bath
and quenched with methionine (1.5 mM). The reaction mixture
was collected by HPLC using system 2 monitoring at 280 nm.
5-Chlorocytosine and 5Cl-dCyd was collected and evaporated
to dryness.
[
13C4,15N2]PFB2-5Cl-Ura, respectively.
GC/NICI/MS An a lysis of (P F B)2-5Cl-Ur a . GC/NICI/MS
studies used a Hewlett-Packard 6890 GC with 5973 MSD mass

264 Chem. Res. Toxicol., Vol. 15, No. 2, 2002
Chen et al.
Ta ble 1. F or m a tion of 5Cl-Cyt fr om Rea ction of Cyt w ith
Oxid a n ts
detector with the negative ion chemical ionization source. In
the SIM mode, the filament was operated at 120 eV with the
ion source at 150 °C. The analyses were carried out with a cool-
on-column inlet, a precolumn (J &W, 1.0 m, 0.53 mm, deactivated
silica), and a HP-5MS capillary column (Hewlett-Packard, 30
m × 0.25 mm, 0.25 µm film thickness) inserted into the ion
source. Methane (99.999% pure) was the moderating gas with
a flow rate of 2.0 mL/min and the pressure at the ion gauge
was 2.2 × 10-4 Torr. Helium was used as the carrier gas
(99.999% pure) at a flow rate of 1.2 mL/min. The oven temper-
ature was held at 50 °C for 2 min and then raised to 300 °C at
10 °C/min. Selective ion monitoring at m/z 325 and 331 was used
to detect and quantify the [M - 181]- fragment ions of PFB2-
5Cl-Ura and [13C4,15N2]PFB2-5Cl-Ura, respectively. The quan-
tification was based on intrapolation of the ratio of the peak
area of PFB2-5Cl-Ura versus [13C4,15N2]PFB2-5Cl-Ura and the
calibration curve. The level of 5Cl-Cyt in DNA was obtained
from dividing the amount of 5Cl-Cyt by the amount of cytosine
in the DNA sample (31).
yield of
yield of
reaction and time
5Cl-Cyt (%) 5Cl-Ura (%)
Cyt + HOCl (1:1), 20 mina
6.74 ( 0.08
NDc
ND
ND
ND
Cyt + ONOO- (1:20), 20 mina
NDd
Cyt + ONOO- (1:20) + Cl-, 20 minb ND
Cyt + H2O2 (1:100) + Cl-, 1 dayb
ND
a
The incubation was performed with Cyt (1.0 mM) with HOCl
or peroxynitrite in potassium phosphate buffer (0.2 M, final pH
7.4) at 37 °C, followed by quench with methionine and HPLC
analysis using system 1. The percentage yields are presented as
mean ( standard deviation from at least duplicated experiments.
b
The incubation was performed in the presence of 100 mM of Cl-.
c Not detectable. The limit of detection for 5Cl-Ura is ca. 1.5 ng.
d
Not detectable. The limit of detection for 5Cl-Cyt is ca. 1.5 ng.
Assa y Ca libr a tion . The stock solutions of 5Cl-Cyt and
[
13C4,15N2]5Cl-Ura (1.0 mg/mL) in water were stored at -80 °C.
Sample solutions for calibration were freshly prepared by
diluting the stock solutions in H2O for each analysis. The
internal standard [13C4,15N2]5Cl-Ura (1.0 ng) was added to
samples containing various amounts of 5Cl-Cyt ranging from
0, 1, 2, 5, 10, 20, 50, 100, and 150 pg. The samples were
processed through the entire procedures, i.e., formic acid hy-
drolysis, C18-OH SPE enrichment, pentafluorobenzylation, Si
SPE purification, and GC/NICI/MS analysis.
HP LC/UV An a lysis of 5Cl-Cyt in DNA. DNA (0.52 mg)
treated with HOCl under condition A was hydrolyzed with 60%
formic acid (0.5 mL) at 150 °C for 45 min. The hydrolysate was
evaporated to dryness, reconstituted in water (0.2 mL), and
analyzed by HPLC using system 3 monitoring at 280 nm.
Resu lts a n d Discu ssion
F or m a tion a n d Sta bility of 5Cl-Cyt. Reaction of Cyt
with equal molar of HOCl under physiological tempera-
ture and pH led to formation of 5Cl-Cyt in 6.7% yield,
respectively, as determined by HPLC analysis with
photodiode array detection. The reaction was quenched
with a scavenger methionine, the fastest HOCl-reacting
amino acid residue of protein (32). Since HOCl is gener-
ated via oxidation of Cl- by H2O2, we examine if other
physiologically relevant oxidants are also capable of Cl-
oxidation. Like HOCl, peroxynitrite is also found in
inflamed tissues. It is formed by the rapid reaction of
superoxide anion with nitric oxide, both of which are
released from macrophages and neutrophils (26, 27).
Peroxynitrite plays an important role in carcinogenesis
related to chronic infections and inflammation (33). In
the presence of plasma concentration of Cl- (100 mM),
small amount of 5Cl-Cyt (0.015%) was detected in the
incubation mixture of Cyt with excess peroxynitrite.
Under similar conditions, no 5Cl-Cyt was detected with
peroxynitrite alone. (Table 1). These results show that
peroxynitrite is capable of oxidizing Cl- forming an
intermediate that is a much weaker chlorinating agent
than HOCl. The identity of this intermediate will be
investigated, but it is not the focus of this study. In
contrast, large excess of H2O2 did not oxidize Cl- during
prolonged incubation in the absence of a peroxidase, such
as myeloperoxidase or eosinophil peroxidase (25), indicat-
ing the importance of peroxidases in generation of HOCl
in vivo.
F igu r e 1. Decomposition of (a) 5Cl-Cyt to 5Cl-Ura and (b) 5Cl-
dCyd to 5Cl-dUrd upon storage at room temperature for 1 week.
solid samples of 5Cl-Cyt and 5Cl-dCyd decomposed
during storage at room temperature for 1 week without
protection from the air or light. The decomposed products
were their deaminated analogues, 5Cl-Ura and 5Cl-2′-
deoxyuridine. Under the conditions described above, 23%
of 5-chlorouracil (5Cl-Ura) derived from solid 5Cl-Cyt and
9% of 5Cl-dUrd derived from solid 5Cl-dCyd as deter-
mined by HPLC analysis with photodiode array detection
(Figure 1). Incubation of purified 5Cl-Cyt at 37 °C for 20
min led to formation of 35% of 5Cl-Ura, whereas 5Cl-
Ura was stable under similar conditions.
GC/NICI/MS An a lysis of 5Cl-Cyt in DNA. The
procedures of this new GC/NICI/MS assay for 5Cl-Cyt
in DNA are modified from the isotope dilution GC/NICI/
MS methods developed in this laboratory for 3,N 4-
ethenocytosine and 1,N 6-ethenoadenine in DNA (31, 34,
35). The entire assay involves addition of a stable isotope
labeled internal standard, acid hydrolysis, adduct enrich-
ment, derivatization with an electrophore, postderivati-
zation cleanup, and GC/NICI/MS analysis as illustrated
in Scheme 1.
Under the hydrolysis condition of DNA with 60%
formic acid at 150 °C for 90 min, 5Cl-dCyd converted to
5Cl-Ura quantitatively, while 98% formic acid at 150 °C
or 1 N HCl at 100 °C for 2 h led to incomplete deamina-
tion (data not shown). Thus, hydrolysis of DNA with 60%
formic acid was used and 5Cl-Ura was measured as the
5-Chlorocytosine and its nucleoside 5Cl-dCyd are not
stable adducts. After collection from the reaction mixture
by reversed-phase HPLC followed by evaporation, the dry

Analysis of 5Cl-Cyt in DNA by GC/ NICI/ MS
Chem. Res. Toxicol., Vol. 15, No. 2, 2002 265
Sch em e 1. P r oced u r es In volved in th e Assa ys of
5Cl-Cyt in DNA
F igu r e 2. (a) Full scan NICI/MS spectrum of PFB2-5Cl-Ura.
(b) Detection of 10 fg (20 amol) of PFB2-5Cl-Ura by GC/NICI/
MS with selective ion monitoring at m/z 325.
pentafluorobenzyl group attached at both N1 and N3 of
5Cl-Ura, yielding N1,N3-bis(pentafluorobenzyl)-5-chlor-
ouracil (PFB2-5Cl-Ura), analogous to other pentafluo-
robenzyl derivatives of uracil (36). The 1H NMR mea-
sured in CDCl3 showed two sets of benzyl protons as two
singlets with chemical shifts at 5.01 and 5.23 ppm in
addition to C6-H at 7.49 ppm. The identity of the
derivative was also confirmed by the full scan mass
spectrum in NICI mode. The NICI mass spectrum
showed a small molecular ion peak of m/z 506. The [M -
181]- ion of 325 was the base peak, indicating that the
compound tends to lose one, but not two, pentafluoroben-
zyl moiety. The relative abundance of the ion at m/z 327
supported that the compound contained a chlorine atom
(Figure 2a). The major fragment ion at m/z 325 was used
for monitoring PFB2-5Cl-Ura in the GC/NICI/MS assay.
The limit of detection for PFB2-5Cl-Ura by GC/NICI/MS
was 10 fg (20 amol) with a S/N ) 8 injected on column
with selective ion monitoring (SIM) mode at m/z 325
(Figure 2b).
end product for quantification of 5Cl-Cyt in DNA hy-
drolysate (13). The isotope analogue having six mass
units higher than the analyte, [13C4,15N2]5Cl-Ura, was
added to the DNA samples as internal standard. The
isotope standard [13C4,15N2]5Cl-Ura used in this assay
was synthesized from [13C4,15N2]uracil reacting with
HOCl in the presence of Cl- under acidic pH to form
[13C4,15N2]5Cl-Ura in optimum yield (25). The isotope
standard was obtained in good yield (59%) after purifica-
tion by a disposable C18-OH solid-phase extraction (SPE)
column, which was sufficient to provide high chemical
purity, determined by reversed-phase HPLC with pho-
todiode array detection. The isotopic purity of [13C4,15N2]-
5Cl-Ura used in the assay was confirmed by derivatiza-
tion with pentafluorobenzyl bromide (PFB-Br) followed
by analysis using GC/NICI/MS at m/z 325 and 331 for
pentafluorobenzylated 5Cl-Ura and [13C4,15N2]5Cl-Ura,
respectively, and no peak corresponding to the pentafluo-
robenzylated 5Cl-Ura at m/z 325 was detected. The purity
of [13C4,15N2]Ura purchased was U-13C4, 99%, U-15N2, 98%.
Assuming there is no contamination during synthesis of
[13C4,15N2]5Cl-Ura, the content of [12C4,14N2]5Cl-Ura should
be (0.01)4 × (0.02)2 ) 4 × 10-10 % or 0.026 zmol in 1.0 ng
of [13C4,15N2]5Cl-Ura, which is far below the detection
limit of (PFB)2-5Cl-Ura (20 amol). Thus, the contribution
from natural abundance of this isotope standard can be
neglected in sample quantification.
The isotope standard [13C4,15N2]5Cl-Ura was stable
under the hydrolysis condition, and it was added before
formic acid hydrolysis. The C18-OH SPE condition was
optimized using a buffer of pH 4 to enrich 5Cl-Ura from
the mixture of normal and adducted bases released by
acid hydrolysis of DNA. Elution of 5Cl-Ura is not affected
by pH ranging from 3 to 7 since it does not contain a
free amino group. The optimum condition for its enrich-
ment from the DNA hydrolysate was for removing
maximum amount of the normal bases. Under this
condition, 5Cl-Ura was collected with 85% of thymine,
49% of guanine, and 4% of cytosine. To ensure complete
derivatization of 5Cl-Ura in the presence of large excess
of normal bases, copious amount of derivatization agent
Standard pentafluorobenzylated 5Cl-Ura was synthe-
sized under mild derivatization conditions (31). The

266 Chem. Res. Toxicol., Vol. 15, No. 2, 2002
Chen et al.
Ta ble 2. Levels of 5Cl-Cyt in DNA
DNA sample (amount)
5Cl-Cyt/Cyta
method
treated calf thymus DNA
HOCl/Cl- (A) (520 µg)b
HOCl/Cl- (A) (0.1 µg)b
HOCl/Cl- (B) (0.1 µg)e
HOCl (B) (0.1 µg)e
ONO2-/Cl- (500 µg)e
ONO2- (500 µg)e
untreated calf thymus DNA
(500 µg)
human placental DNA
(500 µg)
1.1 ( 0.1/102
1.2 ( 0.1/102
8.9 ( 0.4/103
7.9 ( 0.2/103
2.2 ( 0.1/106
5.3 ( 0.4/107
6.0 ( 0.4/108
HPLC/UVc
GC/NICI/MSd
GC/NICI/MS
GC/NICI/MS
GC/NICI/MS
GC/NICI/MS
GC/NICI/MS
6.6 ( 0.4/107
GC/NICI/MS
a
Values are presented as mean ( standard deviation from at
b
F igu r e 3. Calibration curve for the GC/NICI/MS analysis for
PFB2-5Cl-Ura at m/z 325 and [13C4,15N2]PFB2-5Cl-Ura at m/z
331. Samples containing various amounts (0-150 pg) of 5Cl-
Cyt was added a fix amount of [13C4,15N2]5Cl-Ura (1.0 ng) and
subjected to the assay procedures described in Experimental
Procedures (r2) 0.9968). The data are combined from at least
separate experiments in duplicates. The ratio of each analyte
to the internal standard was calculated based on the peak areas.
least duplicated experiments. A solution containing calf thymus
DNA (2.48 mg/mL) in potassium phosphate buffer (0.2 M, final
pH 7.4) and Cl- (100 mM) was added HOCl (1.74 mM) and
incubated at 37 °C for 20 min. The reaction was quenched with
methionine (17.4 mM). c Reversed phase HPLC analysis with UV
d
detection at 280 nm using system 1. Described in Materials and
Methods. e Calf thymus DNA (0.5 mg/mL) in potassium phosphate
buffer (0.2 M) and in the presence or absence of Cl- (100 mM)
was added HOCl (0.35 mM) or peroxynitrite (0.35 mM) and
incubated at 37 °C for 60 min, followed by addition of methionine
(3.5 mM). The final pH of the reaction was 7.4.
was used. After derivatization, the pentafluorobenzylated
normal bases were mostly removed by Si SPE chroma-
tography.
Cyt formation. The yield of 5Cl-Cyt in DNA treated with
HOCl in the presence of 100 mM of Cl- was only 11%
more than that with HOCl alone. In the latter sample,
incubation was carried out in phosphate buffer dissolved
in deionized water without special treatment to remove
Cl-. Contamination of Cl- from the buffer was as little
as 20 µM. This result indicates that HOCl plays a major
role in 5Cl-Cyt formation in DNA under physiological pH,
and the contribution of Cl- in forming molecular chlorine
is minor. Thus, 5Cl-Cyt formation in DNA is likely to
occur at a low concentration of Cl-, such as the intrac-
ellular concentration of 5-15 mM. The concentration of
HOCl was reported to be approximately 100 µM in the
vicinity of activated neutrophils (37). Furthermore, HOCl
is a neutral and stable molecule, and it should have long
enough half-life to reach and react with cellular DNA.
This new GC/NICI/MS assay is sensitive enough to
accurately quantify background levels of 5Cl-Cyt in
untreated calf thymus DNA and in human placental
DNA. In 500 µg each of commercially available calf
thymus DNA and in human placental DNA, the levels
of 5Cl-Cyt were determined to be 0.6 and 6.6/107 cytosine,
respectively. The GC/NICI/MS chromatogram of un-
treated calf thymus DNA showed coelution of the peak
of 23.48 min at m/z 325 and at 331 (Figure 4) injecting
one-tenth of the sample. The peak of 23.48 min at m/z
325 represented 3.7 fmol of PFB2-5Cl-Ura, which was
about 2.6 times of the limit of quantification (14 fmol in
the entire sample). This level of 5Cl-Cyt in untreated calf
thymus DNA was much lower than what was reported
in the control DNA samples using the TMS derivatization
method in in vitro experiments of calf thymus DNA
treated with HOCl (13, 17). It is possible that derivati-
zation by the TMS reagent at elevated temperature could
give increased level of 5Cl-Ura, as that was the case for
8-hydroxyguanine (38, 39), due to artifact formation.
Although the involvement of chlorinating agents from
drinking water cannot be excluded, the presence of 5Cl-
Cyt in untreated calf thymus DNA suggests that chlori-
nation of DNA might be a biological process.
GC/NICI/MS Assa y Ca libr a tion . Calibration of the
assay was performed by addition of a fixed amount of
[13C4,15N2]5Cl-Ura (1.0 ng) with various quantities of 5Cl-
Cyt ranging from 0 to 150 pg and processed through the
assay procedures. The lowest amount detected with
quantitative linearity, i.e., the limit of quantification
(LOQ), was 2.0 pg (14 fmol) with a correlation coefficient
(γ2) of 0.9968 (Figure 3). The recovery of the entire assay
was 44 ( 3%. Since only one-tenth of the processed
sample was injected on GC/MS and the recovery of 44%,
a 23-fold higher LOQ compared to LOD is apparent. This
calibration curve did not pass the origin. Background
peaks corresponding to the peak area ratio of 0.0026 was
observed, which might come from evaporation by centri-
fuge concentrator since high levels of standards such as
those in the standard curve were also evaporated in the
same concentrator. The peak area ratio of 0.0026 was
constantly obtained even after extensive cleaning of the
concentrator. Since the calibration curve was linear and
the correlation coefficient was high, this background level
was subtracted. Once this control ratio is decreased, the
limit of quantification can be lowered. It is possible, but
remains to be proven in the future, that tandem mass
spectrometry with selective reaction monitoring of frag-
mentation of [M - PFB]- ion at m/z 325 to [M - PFB -
Cl + H]- ion at m/z 291 could achieve a lower LOQ than
using GC/MS with SIM.
GC/NICI/MS An a lysis of 5Cl-Cyt in Ca lf Th ym u s
DNA. This assay was used to analyze 5Cl-Cyt in calf
thymus DNA treated with equimolar HOCl (relative to
cytosine in DNA) in the presence of plasma concentration
of Cl- (100 mM). In the GC/NICI/MS chromatogram, the
peaks of m/z 325 and 331 coeluted (not shown) and thus
confirmed the identity of the analyte using as little as
0.1 µg of DNA. The level of 5Cl-Cyt in HOCl/Cl--treated
calf thymus DNA was 1.2 adducts/102 Cyt from dupli-
cated experiments after interpolation of the calibration
curve. This level was further verified by reversed-phase
HPLC analysis with photodiode array detection of 5Cl-
Ura in hydrolysate of 520 µg of the same DNA (Table 2).
Another preparation of HOCl/Cl--treated calf thymus
DNA was analyzed to examine the effect of Cl- on 5Cl-
On the other hand, the level of 5Cl-Cyt in DNA treated
with equimolar peroxynitrite (relative to cytosine in
DNA) in the presence of 100 mM of Cl- was determined

Analysis of 5Cl-Cyt in DNA by GC/ NICI/ MS
Chem. Res. Toxicol., Vol. 15, No. 2, 2002 267
gram quantity of the isotope standard also serves as
carrier for trace amount of analyte in the sample, which
might have poor recovery. In addition, the presence of
the isotope standard assists in locating small amount of
the analyte peak in the complicated chromatogram since
the peak of the derivatized standard has virtually the
same retention time (0-0.02 min deviation) as that for
the analyte monitored at different m/z channels. Since
quantification of the analyte is calculated from the ratio
of the peak areas obtained from the GC/NICI/MS analy-
sis, variation in sample recovery can be adjusted. Col-
lectively, the use of this isotope standard allows un-
equivocal detection and accurate quantification of the
analyte in the complex mixture of DNA hydrolysate.
Electrophore labeling coupled with NICI/MS analysis
offers high assay sensitivity. Our results demonstrate
that this new isotope dilution GC/NICI/MS assay should
be useful for 5Cl-Cyt analysis in DNA in in vivo studies
since less amount of DNA sample is needed due to
increase in sensitivity.
Little is known about the mutagenicity of 5Cl-Cyt and
its repair mechanism. It was reported that tumor cells
exposed to 5-chloro-2′-deoxycytidine (5Cl-dCyd), a radi-
osensitizer, incorporate 5-chloro-2′-deoxycytidine (5Cl-
dUrd) into their DNA (40) and delayed tumor growth
(41). The mechanism might be that 5Cl-dCyd was con-
verted to the proximate radiosensitizer 5Cl-dUrd by
dCMP deaminase and thus disrupted pyrimidine me-
tabolism and DNA synthesis (40). It is highly possible
that similar mechanisms apply to the bactericidal action
of HOCl. Since drinking water contains Cl2 that is in
equilibrium with HOCl, humans are constantly exposed
to exogenous chlorinating agents in addition to those
formed endogenously. Obviously, more study is needed
in order to understand the biological significance of 5Cl-
Cyt. With this sensitive and specific assay of 5Cl-Cyt in
hand, it is now feasible to perform in vivo investigations,
such as the correlation between cytosine chlorination of
DNA and carcinogenesis associated with chronic infec-
tions and inflammation, or the effect of chlorine process-
ing in drinking water to human health.
F igu r e 4. GC/NICI/MS SIM chromatogram of PFB2-5Cl-Ura
in untreated calf thymus DNA. Calf thymus DNA (500 µg) was
hydrolyzed and subjected to the assay procedures described in
Experimental Procedures. The peak of 23.48 min at m/z 325
represents 3.7 fmol (1.9 pg) of PFB2-5Cl-Ura.
Ack n ow led gm en t. This work was supported by
grants from National Science Council of Taiwan (NSC
89-2113-M-194-028) and from National Chung Cheng
University (to H.-J . C. C.). The authors thank Mr. Chia-
Ming Chang for helpful technical support.
F igu r e 5. GC/NICI/MS analysis with selective ion monitoring
chromatogram of PFB2-5Cl-Cyt in peroxynitrite/chloride-treated
calf thymus DNA (500 µg). The peak of 23.48 min at m/z 325
represents 79 fmol (40 pg) of PFB2-5Cl-Ura.
Refer en ces
(1) Klebanoff, S. J ., and Clark, R. A. (1978) The Neutrophil: Function
and Clinical Disorders, North-Holland Biochemical Press, Am-
sterdam.
(2) Hurst, J . K., and Barrette, W. C. (1989) Leukocytic oxygen
activation and microbicidal oxidative toxins. CRC Crit. Rev.
Biochem. Mol. Biol. 24, 271-328.
(3) Kettle, A. J . (1996) Neutrophils convert tyrosyl residues in
albumin to chlorotyrosine. FEBS Lett. 379, 103-106.
(4) Schraufstatter, I. U., Browne, K., Harris, A., Hyslop, P. A.,
J ackson, J . H., Quehenberger, O., and Cochrane, C. G. (1990)
Mechanisms of hypochlorite injury of target cells. J . Clin. Invest.
85, 554-562.
(5) Pru¨tz, W. A. (1996) Hypochlorous acid interactions with thiols,
nucleotides, DNA, and other biological substrates. Arch. Biochem.
Biophys. 332, 110-120.
to be 2.2/106 cytosine, more than 4 times higher than that
without addition of Cl-. The peak of 23.48 min at m/z
325 represents 79 fmol of PFB2-5Cl-Ura (Figure 5). The
level of 5Cl-Cyt in DNA treated with peroxynitrite in the
presence of 20 µM of Cl- was 5.3/107 cytosine. The
accuracy of the assay was 5% in average ranging from 2
to 9% depending on the adduct levels (Table 2). To the
best of our knowledge, this is the first report of 5Cl-Cyt
formation in DNA by peroxynitrite through oxidation of
Cl-.
Isotope dilution mass spectrometry has several advan-
tages over mass spectrometric methods without using the
isotope as internal standard. Since the isotope has the
same chemical properties as the analyte, it can monitor
the recovery during each step of the assay and ensure
accurate quantification of the analyte. Addition of nano-
(6) Hawkins, C. L., and Davies, M. J . (1998) Degradation of hyalu-
ronic acid, poly- and monosaccharides, and model compounds by
hypochlorite: Evidence for radical intermediates and fragmenta-
tion. Free Radical Biol. Med. 24, 1396-1410.
(7) Carr, A. C., Winterbourn, C. C., Blunt, J . W., Phillips, A. J ., and
Abell, A. D. (1997) Nuclear magnetic resonance characterization

268 Chem. Res. Toxicol., Vol. 15, No. 2, 2002
Chen et al.
of 6 alpha-chloro-5 beta-cholestane-3 beta,5-diol formed from the
reaction of hypochlorous acid with cholesterol. Lipids 32, 363-
367.
of DNA in intact human respiratory tract epithelial cells. Free
Radical Biol. Med. 28, 1039-1050.
(25) Henderson, J . P., Byun, J ., and Heinecke J . W. (1999) Molecular
chlorine generated by the myeloperoxidase-hydrogen peroxide-
chloride system of phagocytes produces 5-chlorocytosine in bacte-
rial RNA. J . Biol. Chem. 274, 33440-33448.
(26) Koppenol, W. H., Moreno, J . J ., Pryor, W. A., Ischiropoulos, H.,
and Beckman, J . S. (1992) Peroxynitrite, a cloaked oxidant formed
by nitric oxide and superoxide. Chem. Res. Toxicol. 5, 834-842.
(27) Ischiropoulos, H., Zhu, L., and Beckman, J . S., (1992) Peroxyni-
trite formation from macrophage-derived nitric oxide. Arch.
Biochem., Biophys. 298, 446-451.
(8) Volf, I., Roth, A., Cooper, J ., Moeslinger, T., and Koller, E. (2000)
Hypochlorite modified LDL are a stronger agonist for platelets
than copper oxidized LDL. FEBS Lett. 483, 155-159.
(9) Halliwell, B., Wasil, M., and Grootveld, M. (1987) Biologically
significant scavenging of the myeloperoxidase-derived oxidant
hypochlorous acid by ascorbic acid: Implications for antioxidant
protection in the inflamed rheumatoid joint. FEBS Lett. 213, 15-
18.
(10) Prutz, W. A. (1998) Interactions of hypochlorous acid with
pyrimidine nucleotides, and secondary reactions of chlorinated
pyrimidines with GSH, NADH, and other substrates. Arch.
Biochem. Biophys. 349, 183-191.
(11) Pru¨tz, W. A., Kissner, R., Koppenol, W. H., Ru¨egger, H. (2000)
On the irreversible destruction of reduced nicotinamide nucle-
otides by hypohalous acids. Arch. Biochem. Biophys. 380, 181-
191.
(12) Kulcharyk, P. A., and Heinecke, J . W. (2001) Hypochlorous acid
produced by the myeloperoxidase system of human phagocytes
induces covalent cross-links between DNA and protein. Biochem-
istry 40, 3648-3656.
(13) Whiteman, M., J enner, A., and Halliwell, B. (1997) Hypochlorous
acid-induced base modification of DNA. Chem. Res. Toxicol. 10,
1240-1246.
(14) Leone, A. M., Francis, P. L., Rhodes, P., and Moncada, S. (1994)
A rapid and simple method for the measurement of nitrite and
nitrate in plasma by high performance capillary electrophoresis.
Biochem. Biophys. Res. Commun. 200, 951-957.
(15) Green, L. C., Wagner, D. A., Glogowski, J ., Skipper, P. L.,
Wishnok, J . S., and Tannenbaum, S. R. (1982) Analysis of nitrate,
nitrite, and [15N]nitrate in biological fluids. Anal. Biochem. 126,
131-138.
(16) Eiserich, J . P., Hristova, M., Cross, C. E., J ones, A. D. Freeman,
B. A., Halliwell, B., and van der Vliet, A. (1998) Formation of
nitric oxide-derived inflammatory oxidants by myeloperoxidase
in neutrophils. Nature 391, 393-397.
(17) Whiteman, M., Spencer, J . P., J enner, A., and Halliwell, B. (1999)
Hypochlorous acid-induced DNA base modification: Potentiation
by nitrite: Biomarkers of DNA damage by reactive oxygen species.
Biochem. Biophys. Res. Commun. 257, 572-576.
(18) Chen, H.-J . C., Chen, Y.-M., Wang, T.-F., Wang, K.-S., and Shiea,
J . (2001) 8-Nitroxanthine, an adduct derived from 2′-deoxygua-
nosine or DNA reaction with nitryl chloride. Chem. Res. Toxicol.
14, 536-546.
(19) Eiserich, J . P., Cross, C. E., J ones, A. D., Halliwell, B., and van
der Vliet, A. (1996) Formation of nitrating and chlorinating
species by reaction of nitrite with hypochlorous acid. A novel
mechanism for nitric oxide-mediated protein modification. J . Biol.
Chem. 271, 19199-19208.
(20) Panasenko, O. M., Briviba, K., Klotz, L. O., and Sies, H. (1997)
Oxidative modification and nitration of human low-density lipo-
proteins by the reaction of hypochlorous acid with nitrite. Arch.
Biochem. Biophys. 343, 254-259.
(21) Whiteman, M., J enner, A., and Halliwell, B. (1999) 8-Chloroad-
enine: A novel product formed from hypochlorous acid-induced
damage to calf thymus DNA. Biomarkers 4, 303-310.
(22) Lis, A. W., McLaughlin, R. K., McLaughlin, D. L., Daves, G. D.,
and Anderson, W. R. (1973) 5-Chlorocytosine. Occurrence in
salmon sperm deoxyribonucleic acid. J . Am. Chem. Soc. 95, 5789-
5790.
(23) Lis, A. W., McLaughlin, R. K., and McLaughlin, D. I. (1974)
Preliminary characterization of minor bases from DNA fraction-
ation, with discussion on 5-chlorocytosine as artifact or naturally
occurring base. Physiol. Chem. Phys. 6, 527-536.
(28) Morris, J . C. (1966) The acid ionization constant of HOC1 from
5-35 degrees. J . Phys. Chem. 70, 3798-3805.
(29) Uppu, R. M., and Pryor, W. A. (1996) Biphasic synthesis of high
concentrations of peroxynitrite using water-insoluble alkyl nitrite
and hydrogen peroxide. Method Enzymol. 269, 322-329.
(30) Beckman, J . S., Chen, J ., Ischiropoulos, H., and Crow, J . P. (1994)
Oxidative chemistry of peroxynitrite. Methods Enzymol. 233,
229-240.
(31) Chen, H.-J . C., Lin, T. C., Hong, C.-L., and Chiang, L.-C. (2002)
Analysis of 3,N 4-ethenocytosine in DNA and in human urine by
isotope dilution gas chromatography/negative ion chemical ioniza-
tion/mass spectrometry. Chem. Res. Toxicol. 14, 1612-1619.
(32) Pattison, D. I., and Davies, M. J . (2001) Absolute rate constants
for the reaction of hypochlorous acid with protein side chains and
peptide bonds. Chem. Res. Toxicol. 14, 1453-1464.
(33) Ohshima, H., and Bartsch, H. (1994) Chronic infections and
inflammatory processed as cancer risk factors: Possible role of
nitric oxide in carcinogenesis. Mutat. Res. 305, 253-264.
(34) Chen, H.-J . C., Zhang, L., Cox, J ., Cunningham, J . A., and Chung,
F.-L. (1998) DNA adducts of 2,3-epoxy-4-hydroxynonanal: detec-
tion of 7-(1′,2′-dihydroxy-heptyl)-3H-imidazo[2,1,i]purine and
1,N 6-ethenoadenine by gas chromatography/negative ion chemi-
cal ionization/mass spectrometry. Chem. Res. Toxicol. 11, 1474-
1480.
(35) Chen, H.-J . C, Chiang, L.-C. Tseng, M.-C., Zhang, L. L., Ni, J .,
and Chung, F.-L. (1999) Detection and quantification of 1,N 6-
ethenoadenine in human placental DNA by mass spectrometry.
Chem. Res. Toxicol. 12, 1119-1126.
(36) Saha, M., Kresbach, G. M., Giese, R. W., Annan, R. S., and Vouros,
P. (1989) Preparation and mass spectral characterization of
pentafluorobenzyl derivatives of alkyl and hydroxyalkyl-nucleo-
base DNA adducts. Biomed. Environ. Mass Spectrom. 18, 958-
972.
(37) Foote, C. S., Goyne, T. E., and Lehrer, R. I. (1983) Assessment of
chlorination by human neutrophils. Nature 301, 715-716.
(38) England, T. G., J enner, A., Aruoma, O. I., and Halliwell, B. (1998)
Determination of oxidative DNA base damage by gas chromatog-
raphy-mass spectrometry. Effect of derivatization conditions on
artifactual formation of certain base oxidation products. Free
Radical Res. 29, 321-330.
(39) Senturker, S., and Dizdaroglu, M. (1999) The effect of experi-
mental conditions on the levels of oxidatively modified bases in
DNA as measured by gas chromatography-mass spectrometry:
How many modified bases are involved? Prepurification or not?
Free Radical Biol. Med. 27, 370-380.
(40) Santos, O., Perez, L. M., Briggle, T. V., Boothman, D. A., and
Greer, S. B. (1990) Radiation, pool size and incorporation studies
in mice with 5-chloro-2′-deoxycytidine. Int. J . Radiat. Oncol. Biol.
Phys. 19, 357-365.
(41) Greer, S., Santos, O., Gottlieb, C., Schwade, J ., and Marion, H.
S. (1992) 5-chlorodeoxycytidine, a radiosensitizer effective against
RIF-1 and Lewis lung carcinoma, is also effective against
a
DMBA-induced mammary adenocarcinoma and the EMT-6 tumor
in BALB/c mice. Int. J . Radiat. Oncol. Biol. Phys. 22, 505-510.
(24) Spencer, J . P., Whiteman, M., J enner, A., and Halliwell, B. (2000)
Nitrite-induced deamination and hypochlorite-induced oxidation
TX015578G