characterization of the dna adducts induced by aristolochic acids in oligonucleotides by...

RAPID COMMUNICATIONS IN MASS SPECTROMETRY

Rapid Commun. Mass Spectrom. 2008; 22: 3735–3742

) DOI: 10.1002/rcm.3791
Published online in Wiley InterScience (www.interscience.wiley.com
Characterization of the DNA adducts induced by

aristolochic acids in oligonucleotides by electrospray

ionization tandem mass spectrometry

Wan Chan1, Hao Yue1, Ricky N. S. Wong2 and Zongwei Cai1*1Department of Chemistry, Hong Kong Baptist University, Kowloon Tong, Kowloon, Hong Kong SAR, China2Department of Biology, Hong Kong Baptist University, Kowloon Tong, Kowloon, Hong Kong SAR, China

Received 5 June 2008; Revised 13 August 2008; Accepted 23 September 2008

*CorrespoKong BapE-mail: zContract/Grants CHKBU24Contract/Baptist UContract/Fund fromgrant num

Metabolic activation of carcinogenic aristolochic acids (AA) produces reactive aristolactam-

nitrenium ion intermediates. Electrophilic attack of the aristolactam-nitrenium ion via its C7 position

to the exocyclic amino group in the purine bases leads to the formation of DNA adducts.

DNA-binding assays have demonstrated that carcinogens show site- and sequence-specificity and

the biological consequence is defined by the nature of binding aswell as their position in the genome.

In this study, electrospray ionization tandem mass spectrometry was applied for the identification

and position mapping of DNA adducts in oligonucleotides (ODNs). The developed method

was successfully applied for the analysis of unmodified and AA-modified ODNs (5(-TTTATT-3(,5(-TTTGTT-3( and 5(-TACATGTGT-3(). The observation of the modified bases (modified adenine

and guanine) together with the complementary product ions ([an-B�n]

S, wS) from the cleavage of the

3( C–O bond adjacent to the modified base in MS/MS analyses readily enabled the identification of

the AA-binding site in ODNs. Copyright # 2008 John Wiley & Sons, Ltd.

The formation of covalently bonded DNA adducts is a

key cellular event in the mechanism of action of many

carcinogens and mutagens. Thus, detection of DNA adducts

has been used as the biomarker for carcinogen exposure

studies. The traditional approach to investigate DNA

adducts involves enzymatic hydrolysis of the modified

DNA prior to analysis at the nucleobase, nucleoside or

nucleotide levels.1–14 Although these approaches can pro-

vide quantitative data, the characterization of DNA adducts

after enzymatic hydrolysis results in the loss of all sequence-

specific information. The sequencing of modified oligonu-

cleotides (ODNs) is important for the therapeutic industry

and for toxicologists because the biological consequences of

carcinogen–DNA adducts are defined by the structure of the

lesion and its position within the genome. Methods based on

molecular biology, such as Maxam–Gilbert sequencing and

polymerase chain reaction, have been used extensively to

determine the site- and sequence-selectivity of carcinogens in

ODN binding assays.15–18 However, these techniques give

no structural information regarding the adducts formed.

Electrospray ionization tandemmass spectrometry (ESI-MS/

ndence to: Z.W. Cai, Department of Chemistry, Hongtist University, Kowloon, Hong [email protected] sponsor: Research Grant Council, Universityommittee of Hong Kong; contract/grant number:59/06M.grant sponsor: Faculty Research Grant of Hong Kongniversity; contract/grant number: FRG/06-07/II-56.grant sponsor: Health and Health Services Researchthe Food andHealth Bureau of Hong Kong; contract/ber: 05060141.

MS) with the capability to generate informative fragment

ions has been used for the characterization of synthetic

ODNs19–26 and for the position mapping of modified bases

within ODNs.27–29 It has been observed that alkylation of

guanine labilizes the glycosidic bond, resulting in the loss

of the alkylated base followed by cleavage of the 30 C–O

bond adjacent to the alkylated residue. The resulting

complementary pairs of sequence ions ([an-B�n]

�, w�) readily

enabled the identification of the binding site.28,29

Aristolochic acid (AA) is a mixture of structurally related

nitrophenanthrene carboxylic acids derived from the herbal

species Aristolochia and Asarum.30–32 Major components of

AA include aristolochic acid I (AAI, 8-methoxy-6-nitrophe-

nanthro(3,4-d)-1,3-dioxolo-5-carboxylic acid) and aristo-

lochic acid II (AAII, 6-nitrophenanthro(3,4-d)-1,3-dioxolo-5-

carboxylic acid) that differ by a methoxy group (Scheme 1).

AA-containing herbs were widely used to treat tumors,

snake bites, obstetrics, rheumatism, small pox and pneumo-

nia33,34 until AA was observed to be a strong carcinogen

in rats.35 Although being banned worldwide, misuse of

AA-containing herbs exists.

AA is a known nephrotoxin36,37 and also one of the most

potent carcinogens in the Carcinogenic Potency Database.38

Upon metabolic activation by mammalian enzymes, AA is

reduced to aristolactam via an aristolactam-nitrenium ion

intermediate.3–14,40,41 Electrophilic attack of the aristolactam-

nitrenium ion via its C7 position to the exocyclic amino

group in the purine bases (dA and dG) leads to the formation

of major adducts (Scheme 1).6–9,42 DNA-AA adducts

were detected in laboratory rodents3–8 and in patients

suffering from aristolochic acid nephropathy.12–14 The

Copyright # 2008 John Wiley & Sons, Ltd.

Scheme 1. Metabolic activation and DNA adduct formation of aristolochic acids.

3736 W. Chan et al.

reported DNA-AA adducts were elucidated spectroscopi-

cally as 7-(deoxyadenosine-N6-yl)aristolactam I (dA-AAI),

7-(deoxyguanosine-N6-yl)aristolactam I (dG-AAI), 7-(deox-

yadenosine-N6-yl)aristolactam II (dA-AAII) and 7-(deoxy-

guanosine-N6-yl)aristolactam II (dG-AAII).8

DNA–AA binding assays at nucleoside/nucleotide levels by

using 32P-postlabelling have been conducted extensively.3–9,11–14

In our previous study, liquid chromatography/tandem mass

spectrometry (LC/MS/MS)was used to study theDNAadducts

induced by AA at the nucleoside level.10 Recently, an LC/

ESI-MS method was developed by Grollman et al. for the

analysis of DNA-AA adducts in the renal tissue of patients

suffering from Balkan endemic nephropathy.39 Multiple-stage

mass spectrometric analyses (MS/MS and MS3) on a two-

dimensional quadrupole ion trap (2D-QIT) mass spectrometer

were performed for the peak identification. Similar to our

observation,10 the AA-DNA adducts showed characteristic

fragmentation loss of a deoxyribose moiety with 116Da.

Characteristic fragment ions at m/z 262 and 292 were detected

for the DNA adducts induced by AAII and AAI, respectively.

The site- and sequence-specificity of AA in base modification

was investigated by using PCR.16–18 Despite the potential

advantages in applying ESI-MS/MS to study the site- and

sequence-specificity of AA inODNs, its application has not been

reported, to the best of our knowledge. In this paper, we report

the first application of ESI-MS/MS for the position mapping of

base modification by AA in ODNs.

EXPERIMENTAL

ChemicalsODNs were obtained from Bio Basic Inc. (Ontario, Canada).

Aristolochic acid (AA), mixture of AAI and AAII (96%, 1:1),

was purchased from Acros (NJ, USA). AAI was obtained by


chromatographic separation of the AA mixture by reversed-

phase high-performance liquid chromatography (RP-HPLC).

Zinc powder (60mesh) was obtained from Aldrich and was

ground using a mortar and pestle before use. Oasis HLB

solid-phase extraction (SPE) cartridges (Plus) were obtained

fromWaters (MA, USA). Acetic acid and ammonium acetate

were obtained from Panreac (Barcelona, Spain). HPLC-grade

acetonitrile was purchased from Tedia (OH, USA). Water

was produced by a Milli-Q Ultrapure water system (Waters,

CT, USA) with the water outlet operating at 18.2MV.

Modification and purification of ODNsThe lyophilized ODNs together with 10 equivalents of AA

(mixture of AAI and AAII for 50-TTTATT-30, 50-TTTGTT-30,

50-TTTCTT-30 and AAI for 50-TACATGTGT-30) were dis-

solved in a mixture of 500 mL of water/acetic acid (99:1). The

reaction was initiated by the addition of 20mg of zinc

powder. The mixtures were vortex-mixed and allowed to

stand at room temperature for 10min before 100 mL of

water/acetic acid (95:5) and 5 equivalents of AAwere added

and incubated for another 10min. To further increase the

yield of the modification to the ODNs, the above processes

were repeated four more times.

The reaction mixture for each of TTTATT, TTTGTT and

TTTCTTwas purified by SPE. The SPE cartridge (Oasis HLB)

after activation by acetonitrile was conditioned by using

5mL of water containing 0.1M ammonium acetate. The

reaction mixtures were then loaded to the cartridges

followed by washing with 1mL of water and eluted by

5mL of water/acetonitrile (50:50). Fraction of the eluents

were evaporated to dryness and reconstituted in water/

acetonitrile (50:50) with 5mM of ammonium acetate

immediately before analysis. The reaction mixture of

TACATGTGT was purified by RP-HPLC with fluorometric


DOI: 10.1002/rcm

ESI-MS/MS characterization of AA-DNA adducts in ODNs 3737

detection (Ex 399; Em 474) on a Waters Alliance 2695 HPLC

system equipped with a 446 fluorescence detector (Milford,

MA, USA). The collected HPLC fraction pools were freeze-

dried, desalted by using SPE as described above, evaporated

to dryness and redissolved in water/acetonitrile (50:50) with

5mM of ammonium acetate immediately before analysis.

Analytical methods and instrumentationESI-MS and MS/MS analyses were performed on a

quadrupole time-of-flight (Qq-TOF) mass spectrometer

equipped with a standard ESI source in negative ionization

mode (API Q-STAR Pulsar i, Applied Biosystems, Foster

City, USA). Prepared samples were infused into the mass

spectrometer at a flow rate of 5mL/min by aHarvard syringe

pump. The electrospray voltage (ES), declustering potential I

(DPI), declustering potential II (DPII) and focusing potential

(FP) used were optimized to be �4.0 kV, �80V, �150V and

15V, respectively. The mass range was from m/z 100–1800.

The collision energy for product ion scans was typically in

the range of 40 to 50 eV for the MS/MS experiments. The ion

source gas I (GSI), curtain gas (CUR) and collision-assisted

dissociation gas (CAD) was set at 40, 25 and 3psi,

respectively. Mass calibration of the TOF analyzer was

performed by infusing 10 pmol/mL of TTTATT at a flow rate

of 5mL/min. Mass spectra were recorded on a personal

computer with the Analyst QS software (service pack 7,

Applied Biosystems). Sixty to ninety scans were summed for

each spectrum.

Optimization of ESI-MS parametersTo optimize the ESI-MS parameters, TTTATT in acetonitrile/

water (1:1) with 5mMof ammonium acetate was infused into

the Qq-TOFMS system. It was found that the use of negative

ionization mode produced a MS signal about four times

stronger than that in the positive ionization mode. Negative

ionization mode was therefore used in the subsequent

analysis. With TTTATT as the model compound, the ESI

source parameters (IS, DPI, DPII and FP)were optimized (see

Analytical Methods and Instrumentation).

Scheme 2. Nomenclature for fragment io

et al.19 Possible fragments along the phosp

-OH group are termed a, b, c and d, while frag

termed w, x, y and z.


To investigate the effect of the solvent system for the ESI-MS

analysis of ODNs, TTTATT at equimolar were prepared in

different buffer solutions and infused into the Qq-TOF

instrument. It was found that the addition of 5mM of

ammonium acetate to the acetonitrile/water mixture (1:1)

increased the MS signal by more than two times. Further

increase in the ammonium acetate concentration, however,

gave no significant signal enhancement. It was also observed

that the addition of ammonium acetate to the acetonitrile/

water (1:1) solvent system did not alter the fragmentation

pattern of the ODNs in MS/MS analysis so 5mM of

ammonium acetate in acetonitrile/water (1:1) mixture was

used to dissolve the unmodified and modified ODNs in the

subsequent ESI-MS/MS analyses.

Preparation of modified ODNsODNs containing single exocyclic amino groups bearing a

base (TTTATT, TTTGTT and TTTCTT) were reacted with

the AA mixture (AAI and AAII) while the ODN containing

multiple potential target sites (TACATGTGT) were reacted

with AAI, by using zinc powder in acetic acid as the

activation system. A modification to the activation system

reported by Broschard et al.16 was used in this study, in

which the phosphate buffer of pH 5.8 was replaced by 1%

acetic acid solution and the addition of AA was made in

several small portions. The increased acid content in the

reaction system allowed the nitroreduction of AA to be

completed in a shorter period of time (from 6 h to 1 h) and at

a lower temperature (from 378C to 218C). Because the

nitroreduction of AA proceeded rapidly in the modified

environment and the adducts were formed only when

the reactive intermediate (aristolactam nitrenium ion)

approached the purine bases within the ODNs, the addition

of AA was also divided into small portions so as to allow a

higher yield of the modified ODNs. The modified ODNs

were purified either by SPE for TTTATT, TTTGTT and

TTTCTT or by off-line HPLC for TACATGTGT. The

purified fractions were then infused individually into the

Qq-TOF mass spectrometer for analyses by ESI-MS and -

MS/MS.

ns of ODNs proposed by McLuckey

hodiester backbone containing the 50

ments containing the 30 -OH group are


DOI: 10.1002/rcm

Figure 1. Typical ESI-MS spectrum of 50-TTTATT-30 modified with 35 equivalents

of AA (mixture of AAI and AAII). Labeled peaks displaym/z values of the [M–2H]2�

ions of the unmodified and AA-modified ODNs.

3738 W. Chan et al.

RESULTS AND DISCUSSION

ESI-MS and ESI-MS/MS analyses of themodified and unmodified ODNs

The challenge in using ESI-MS/MS for the positional

mapping of base modification in ODNs includes: (1) The

complexity of the tandem mass spectrometry of modified

and unmodifiedODNs because of the possible fragmentation

along the entire phosphate linkage (Scheme 2); (2) The

limited availability of modified ODNs for the analysis

because of the low conversion efficiency into DNA adducts

(e.g. 0.1–0.4% of the AAI employed);8 (3) The treatment of

ODNs containing multiple targeted bases for a carcinogen

results in the formation DNA adducts at any of the potential

bases, thus a series of positional isomers can be formed; and

(4) The complexity of themodification induced to the ODN is

increased by the potential multiple adduct formation in the

ODN.

Because ODNs had poor retention on RP-HPLC columns,

ion-pairing LC/MS methods were developed for the analysis

of ODNs.43,44 However, the severe interference caused by the

ion-pairing reagents affects the on-line coupling ofHPLCwith

MS detection. Most studies involving ODNs have therefore

been performed by syringe infusion.19–28 In this study, the

modified and unmodified ODNs were analyzed by ESI-Qq-

TOFMS. The nomenclature introduced byMcLuckey et al. was

used to describe the fragments originated from the modified

and unmodified ODNs.19 While the possible fragments along

the phosphodiester backbone containing the 50 –OH group

were termed a, b, c and d, fragments containing the 30 –OH

group were termed w, x, y and z ions (Scheme 2).

Mechanism studies on the dissociation of unmodified

ODNs in the gas phase showed that the loss of the nucleobase

(B) was the initial dissociation step, leading to the subsequent

cleavage of the 30 C–O bond. [an–Bn]� and w� ions were

observed as the major ions.20,25 It was also observed that

electron-withdrawing modifications to the nucleobase

enhanced the loss of the modified nucleobase (B�) and the


fragmentation of the adjacent 30 C–O bond to yield the

[an–Bn�]� and w� ions.27,28

ODNs containing single site-specific AA lesions are

important for understanding the mutagenicity or carcino-

genicity of AA.16–18,45 In those studies, AA-modified ODNs

were characterized by denaturing gel electrophoresis and32P-postlabelling analyzed after their enzymatic hydrolysis.

These processes are time-consuming (days) and laborious. In

this study, the site-specificity of AA in DNA binding was

investigated by using ODNs containing a single exocyclic

amino group bearing bases (TTTATT, TTTGTT and

TTTCTT). A typical ESI-MS spectrum from the analysis of

the reconstituted SPE eluent of TTTATT is shown in Fig. 1.

[M–2H]2� ions of the modified and unmodified oligonucleo-

tides were detected. Peaks at m/z 884.6, 1015.2 and 1030.2

corresponded to the unmodified, AAII-modified and AAI-

modified ODNs, respectively. The peak intensity of the [M–

2H]2� ions of AAI- and AAII-modified TTTATT were

similar, indicating that AAI and AAII have similar reactivity

in the tested ODN.Adduct formationwas also observed to be

at a higher yield with the A moiety than on the G moiety of

the tested ODNs.

ESI-MS/MS analyses were performed on the hexaODNs to

demonstrate the feasibility of the developed method for the

identification and position mapping of AA modifications on

ODNs. The MS/MS spectra of the [M–2H]2� ions of

(A) unmodified, (B) AAII-modified and (C) AAI-modified

TTTATT are shown in Fig. 2. Both unmodified (Fig. 2(A)) and

modified ODNs (Figs. 2(B) and 2(C)) showed similar

fragmentations, with the [an–Bn]� and [w]� ions observed

as the major fragment ions. Mass calibration was performed

by infusing standard ODN TTTATT solution prior to

analysis. The calibrated TOF analyzer allowed the identifi-

cation of product ions at mass accuracy better than 20ppm.

The high resolution capability of the TOF instrument readily

allowed the determination of multiply charged ions, i.e. ion

peaks in the ion clusters of doubly charged ions are 0.5Da

apart while that for triply charged ions are 0.33Da apart.


DOI: 10.1002/rcm

Table 1. Characteristic product ions observed from the MS/

MS analyses of unmodified and AA-modified TTTGTT

TTTG-TT TTTGa-TT TTTGb-TT

w2� 625 625 625

w42� 629 760 775

w52� 781 912 927

[a2–B2(T)]� 401 401 401

[a4–B4(G)]� 1009 1009 1009[M–B4(G)]2� — 818 818[B4(G)]� — a411 b441

aAAII-modified;bAAI-modified.

Figure 2. MS/MS spectra of the [M–2H]2� ions for (A) the unmodified, (B) AAII-

modified, and (C) AAI-modified 50-TTTATT-30. Doubly charged ions are labeled with

asterisks.


Alkylation of adenine by AA labilizes the glycosidic bond,

resulting in loss of the alkylated base atm/z 395 (Fig. 2(B)) and

425 (Fig. 2(C)) for ODNs modified by AAII and AAI,

respectively. The AA-modified adenine bases were stable

because of the charge delocalization on the adenine base,

which was similar to the charge delocalization arguments of

the negatively charged base ions proposed by Pan et al. who

investigated the initial fragmentation of ODNs.25

The doubly charged ion of the depurinated nucleotide

residue was detected atm/z 818 as a complementary ion. The

fragmentation process was followed by cleavage of the 30

C–O bond adjacent to the alkylated residue. Cleavage of the

30 C–O bond of the depurinated nucleotide residue yielded

two singly charged ODN fragments. One of the fragment

ions were detected at m/z 625, that was the w2� ion,

dinucleotide fragment 30 to the alkylated adenine. Another

ion at m/z 1009, that was the [a4–B4]� ion, tetranucleotide

fragment 50 to the alkylated adenine, was also detected. The


resulting complementary pairs of sequence ions of [a4–B4]�

and w2� together with the modified adenine adduct ion

readily enabled the identification of the AA-binding site in

ODNs.


DOI: 10.1002/rcm

Scheme 3. Proposed resonance structures of the AA-modified guanine fragment ion

formed via cleavage of AA-modified ODNs.

3740 W. Chan et al.

Similarly, the binding site of AA to TTTGTT was

identified. Table 1 lists some of the characteristic fragment

ions from the MS/MS analyses of the modified and

unmodified ODNs. In line with the observation in the

MS/MS analysis of TTTATT (Fig. 2), the [a4–B4]� andw2

� ions

of unmodified and modified TTTGTT were detected at m/z

1009 and 625, respectively. AAII- and AAI-modified

guanines were detected at m/z 411 and 441, respectively.

The possible resonance structures corresponding to the AA-

modified guanine fragment ions are proposed in Scheme 3.

The doubly charged ion of the depurinated nucleotide

residue for both the modified and unmodified ODNs was

detected at m/z 818. The site of binding of AA on the ODN

sequence was unequivocally elucidated from the MS/MS

analysis.

The deoxycytidine adducts of AA (dC-AAI and dC-AAII)

were detected from the previous in vitro studies at

the nucleoside level.10 However, neither modification to

the deoxycytidine could be detected in TTTCTT nor in the

previous in vitro and in vivo studies.3–9 The observed results

Figure 3. Full scan ESI-MS spectra of (A)

(B) the AAI-modified 50-TACATGTGT-30 (iso


might provide helpful information for the low mutagenic

potential of the deoxycytidine by AA.

Upon successful analysis of the ODN modification on

single site-specific AA lesions by MS and MS/MS, the study

was extended to the analysis of ODNs containing multiple

binding sites. The reaction product of AA and TACATGTGT

was separated by reversed-phase HPLC. Three samples were

obtained after the HPLC separation, of which one was

unmodified and two were AAI-modified ODNs (isomer I

and isomer II). The samples were freeze-dried and desalted

by using SPE prior to the analysis by ESI-Qq-TOFMS. The full

scan ESI-MS spectrum of the unmodified ODN revealed ion

peaks at m/z 680.8, 908.5 and 1363.2, representing the

[M–4H]4�, [M–3H]3� and [M–2H]2� ions, respectively

(Fig. 3(A)). Figure 4(A) shows the MS/MS spectrum from

the [M–3H]3� ion peak at m/z 908 of the unmodified ODN.

Major ions of [an–Bn]� and w� were observed.

Figure 3(B) shows the full scan ESI-MS spectrum of the

AAI-modified TACATGTGT. The [M–4H]4�, [M–3H]3� and

[M–2H]2� ions were detected at m/z 753.6, 1005.2 and 1508.3,

the unmodified 50-TACATGTGT-30 and

mer I).


DOI: 10.1002/rcm

Figure 4. MS/MS spectra of (A) the unmodified 50-TAC-ATGTGT-30 at m/z 908,

(B) isomer I, and (C) isomer II of the AAI-modified 50-TACATGTGT-30 at m/z 1005.

The MS/MS analyses revealed the modification at isomer I and isomer II was on the

fourth base (adenine) and the sixth base (guanine) from the 50 end of the ODN,

respectively. Doubly and triply charged ions are labeled with asterisks and #,

respectively.


respectively. The MS/MS analysis on the [M–3H]3� ion of

isomer I revealed the characteristic loss of AAI-modified

adenine atm/z 425 (Fig. 4(B)), indicating themodificationwas

at the adenine in the ODN. The doubly charged ion of the

depurinated nucleotide residue was detected atm/z 1294 as a

complementary ion. A singly charged ion at m/z 1003 was

observed in the MS/MS spectrum of isomer I together with

the observation of a significant intensity enhancement of the

ion peak at m/z 793. This observation revealed the

modification site to be the fourth base from the 50 end of

the ODN. The ion peaks at m/z 1003m/z 793 corresponded to

[a4–B4]� and w5

2�, respectively. Specific cleavage at the

modification site giving [an–Bn]� and w� ions after the

depurination was also observed in the DNA adduct studies

by Iannitti et al. and Marzilli et al.27,28 [M–H]� ion peaks

corresponding to cytosine, thymine, adenine, and guanine


ions were also observed at m/z 110, 125, 134 and 150,

respectively (data not shown).

Similarly, the MS/MS analysis on the [M–3H]3� ion of

isomer II at m/z 1005 revealed the characteristic loss of AAI-

modified guanine at m/z 441 (Fig. 4(C)), indicating that the

modification was at guanine within the ODN. A doubly

charged ion at m/z 810 and a significant intensity enhance-

ment at m/z 954 were also observed, which indicated that

the modification site was the sixth base from the 50 end of the

ODN. The complementary ion peaks at m/z 810 and 954

corresponded to [a6–B6]2� and w3

�, respectively. Modifi-

cations were not detected either in the terminal adenine

(second base) or in the guanine (eighth base) of

TACATGTGT. The geometric effects could probably be the

reason for the non-detection of modifications in those

locations. No cytidine or multiple adduct formation were


DOI: 10.1002/rcm

3742 W. Chan et al.

identified. We believed the low binding efficiency (0.1–0.4%

of the AAI employed in dA-AAI or dG-AAI synthesis)8 of

AA to DNA could be the reason for the non-detection of

multiple adduct formation.

CONCLUSIONS

Electrospray ionization tandem mass spectrometry was

shown here to be an effective approach for the position

mapping of DNA adducts induced by AA in ODNs.

The results presented in this paper confirm that A and G

are the reactive bases towards electrophilic attack by the

aristolactam-nitrenium ion. The reactions were conducted

with single-stranded ODNs, which might affect reactivity

and selectivity of adduction products. The observation of the

modified bases (modified adenine and guanine) together

with the complementary product ions ([an–B�n]

�, w�) from

the cleavage of the 30 C–O bond adjacent to the modified base

readily enabled the identification of the AA-binding site in

ODNs.

AcknowledgementsFinancial support from the Research Grant Council, Univer-

sity Grants Committee of Hong Kong (HKBU2459/06M),

Faculty Research Grant of Hong Kong Baptist University

(FRG/06-07/II-56), and the Health and Health Services

Research Fund from the Food and Health Bureau of Hong

Kong (05060141) is acknowledged.

REFERENCES

1. Koc H, Swenberg JA. J. Chromatogr. B 2002; 778: 323.2. de Kok TMCM, Moonen HJJ, van Delft J, van Schooten FJ.

J. Chromatogr. B 2002; 778: 345.3. Pfau W, Schmeiser HH, Wiessler M. Carcinogenesis 1990; 11:

1627.4. PfauW, Schmeiser HH, Wiessler M. Chem. Res. Toxicol. 1991;

4: 581.5. Bieler CA, Stiborova M, Wiessler M, Cosyns JP, deStrihou

CV, Schmeiser HH. Carcinogenesis 1997; 18: 1063.6. Stiborova M, Frei E, Wiessler M, Schmeiser HH. Chem. Res.

Toxicol. 2001; 14: 1128.7. Stiborova M, Frei E, Breuer A, Wiessler M, Schmeiser HH.

Mutat. Res. 2001; 493: 149.8. Pfau W, Schmeiser HH. Carcinogenesis 1990; 11: 313.9. Stiborova M, Frei E, Sopko B, Sopkova K, Markova V,

Lankova M, Kumstyrova T, Wiessler M, Schmeiser HH.Carcinogenesis 2003; 24: 1695.

10. Chan W, Zheng YF, Cai ZW. J. Am. Soc. Mass Spectrom. 2007;18: 642.

11. Schmeiser HH, Frei E, Wiessler M, Stiborova M. Carcinogen-esis 1997; 18: 1055.

12. Bieler CA, Stiborova M, Wiessler M, Cosyns J-P, van Yperselede Strihou C, Schmeiser HH. Carcinogenesis 1997; 18: 1063.

13. Stiborova M, Frei E, Sopko B, Wiessler M, Schmeiser HH.Carcinogenesis 2002; 23: 617.


14. Arlt VM, Pfohl-Loszkowicz A, Cosyns J-P, Schmeiser HH.Mutat. Res. 2001; 494: 143.

15. Antsypovich S, Quirk-Dorr D, Pitts C, Tretyakova N. Chem.Res. Toxicol. 2007; 20: 641.

16. Broschard TH, Wiessler M, Schmeiser HH. Cancer Lett. 1995;98: 47.

17. Arlt VM, Wiessler M, Schmeiser HH. Carcinogenesis 2000; 21:235.

18. Arlt VM, Schmeiser HH, Pfiefer GP. Carcinogenesis 2001; 22:133.

19. McLuckey SA, Van Berkel QJ, Glish GL. J. Am. Soc. MassSpectrom. 1992; 3: 60.

20. McLuckey SA, Habibi-Goudarzi S. J. Am. Chem. Soc. 1993;115: 12085.

21. Wan KX, Gross J, Hillenkamp F, Gross ML. J. Am. Soc. MassSpectrom. 2001; 12: 193.

22. Wan KX, Gross ML. J. Am. Soc. Mass Spectrom. 2001; 12:580.

23. Monn STM, Schurch S. J. Am. Soc. Mass Spectrom. 2005; 16:370.

24. Tromp JM, Schurch S. J. Am. Soc. Mass Spectrom. 2005; 16:1262.

25. Pan S, Verhoevan K, Lee JHK. J. Am. Soc.Mass Spectrom. 2005;16: 1853.

26. Wu J, McLuckey SA. Int. J. Mass Spectrom. 2004; 237: 197.27. Iannitti P, Sheil MM,WickhamG. J. Am. Chem. Soc. 1997; 119:

1490.28. Marzilli LA,WangD, KobertzWR, Essigmann JM, Vouros P.

J. Am. Soc. Mass Spectrom. 1998; 9: 676.29. Sherman CL, Pierce SE, Brodbelt JS, Tuesuwan B, Kerwin

SM. J. Am. Soc. Mass Spectrom. 2006; 17: 1342.30. Kite GC, Yule MA, Leon C, Simmonds MSJ. Rapid Commun.

Mass Spectrom. 2002; 16: 585.31. Zhou XG, Zheng CY, Sun JY, You TY. J. Chromatogr. A 2006;

1109: 152.32. Chan W, Lee KC, Liu N. J. Chromatogr. A 2007; 1164: 113.33. Kupchan MS, Doskotch RW. J. Med. Pharm. Chem. 1962; 5:

657.34. Rucker G, Chung BS. Plant Med. 1975; 27: 68.35. Mengs U, Lang W, Poch JA. Arch. Toxicol. 1982; 51: 107.36. Debelle FD, Nortier JL, De Prez EG, Garbar CH, Vienne AR,

Salmon IJ, Deschodt-Lanckman MM, Vanherweghem JL.J. Am. Soc. Nephrol. 2002; 13: 431.

37. Vanherweghem JL, Depierreux M, Tielemans C, Abramo-wicz D, Dratwa M, Jadoul M, Richard C, Vandervelde D,Verbeelen D, Vanhaelenfastre R, Vanhaelen M. Lancet 1993;341: 387.

38. Gold LS, Zeiger E. Handbook of Carcinogenic Potency andGenotoxicity Databases. CRC Press: Boca Raton, 1997.

39. GrollmanAP, Shibutani S,MoriyaM,Miller F,Wu L,Moll U,Suzuki N, Femandes A, Rosenquist T,Medverec Z, JakovinalK, Brdar B, Slade N, Turesky RJ, Goodenough AK, Rieger R,Vukelic M, Jelakovic B. Proc. Natl. Acad. Sci. USA 2007; 104:12129.

40. Chan W, Cui L, Xu GW, Cai ZW. Rapid Commun. MassSpectrom. 2006; 20: 1755.

41. Chan W, Luo H-B, Zheng YF, Cheng YK, Cai ZW. DrugMetab. Dispos. 2007; 35: 866.

42. Arlt VM, Stiborova M, Schmeiser HH. Mutagenesis 2002; 17:265.

43. Apffel A, Chakel JA, Fischer S, Lichtenwalter K, HancockWS. Anal. Chem. 1997; 69: 1320.

44. Deguchi K, Ishikawa M, Yokokura T, Ogata I, Ito S,Mimura T, Ostrande C. Rapid Commun. Mass Spectrom.2002; 16: 2133.

45. Debrauwer L, Rathanao E, Couve C, Poulain S, Pouyet C,Jouanin I, Paris A. J. Chromatogr. A. 2002; 976: 123.


DOI: 10.1002/rcm

characterization of the dna adducts induced by aristolochic acids in oligonucleotides by...

Documents