Photohydrolysis of methotrexate produces pteridine, which induces poly-G-specific DNA damage through photoinduced electron transfer

Article abstract of DOI:10.1562/0031-8655(2002)076<0467:POMPPW>2.0.CO;2

Methotrexate (MTX), an antineoplastic agent, demonstrates phototoxicity. The mechanism of damage to biomacromolecules induced by photoirradiated MTX was examined using ³²P-labeled DNA fragments obtained from a human gene. Photoirradiated MTX ca

Full text of DOI:10.1562/0031-8655(2002)076<0467:POMPPW>2.0.CO;2

Photohydrolysis of Methotrexate Produces Pteridine, Which Induces Poly-
G–specific DNA Damage Through Photoinduced Electron Transfer
Author(s): Kazutaka Hirakawa, Masahiro Aoshima, Yusuke Hiraku, and Shosuke Kawanishi
Source: Photochemistry and Photobiology, 76(5):467-472.
Published By: American Society for Photobiology
https://doi.org/10.1562/0031-8655(2002)076<0467:POMPPW>2.0.CO;2
URL: http://www.bioone.org/doi/full/10.1562/0031-8655%282002%29076%3C0467%3APOMPPW
%3E2.0.CO%3B2
BioOne (www.bioone.org) is a nonprofit, online aggregation of core research in the biological, ecological, and
environmental sciences. BioOne provides a sustainable online platform for over 170 journals and books published
by nonprofit societies, associations, museums, institutions, and presses.
Your use of this PDF, the BioOne Web site, and all posted and associated content indicates your acceptance of
BioOne’s Terms of Use, available at www.bioone.org/page/terms_of_use.
Usage of BioOne content is strictly limited to personal, educational, and non-commercial use. Commercial inquiries
or rights and permissions requests should be directed to the individual publisher as copyright holder.
BioOne sees sustainable scholarly publishing as an inherently collaborative enterprise connecting authors, nonprofit publishers, academic institutions, research
libraries, and research funders in the common goal of maximizing access to critical research.

Photochemistry and Photobiology, 2002, 76(5): 467–472
Photohydrolysis of Methotrexate Produces Pteridine, Which Induces
Poly-G–specific DNA Damage Through Photoinduced Electron Transfer
{
1
Kazutaka Hirakawa , Masahiro Aoshima , Yusuke Hiraku and Shosuke Kawanishi
2
2
ꢀ
2
1
Radioisotope Center, Mie University School of Medicine, Edobashi, Tsu, Mie, Japan and
Department of Environmental and Molecular Medicine, Mie University School of Medicine, Edobashi, Tsu, Mie, Japan
2
Received 26 April 2002; accepted 12 August 2002
ABSTRACT
INTRODUCTION
Methotrexate (MTX), an antineoplastic agent, demonstrates
phototoxicity. The mechanism of damage to biomacromolecules
The photochemical stability and the excited state property of
pharmaceutical substances should be related to the occurrence of
phototoxicity as an adverse side effect after drug administration to
patients (1). Methotrexate (MTX), a strong inhibitor of dihydro-
folate reductase, has been widely used for chemotherapy in many
types of cancer (2). MTX also has been used in the treatment of
psoriasis. However, the phototoxic effects observed after admin-
istration of MTX to patients have been reported as adverse side
effects (3–6). Furthermore, after administration, MTX also causes
phototoxic effects as adverse side effects of PUVA therapy (4–6).
Because MTX has an absorption band in the UVA region, the
UVA-induced damage to biomacromolecules may participate in
the phototoxicity of MTX.
3
2
induced by photoirradiated MTX was examined using P-
labeled DNA fragments obtained from a human gene. Photo-
irradiated MTX caused DNA cleavage specifically at the
underlined G in 59-GG and 59-GGG sequences in double-
stranded DNA only when the DNA fragments were treated with
piperidine, which suggests that DNA cleavage was caused by
base modification with little or no strand breakage. With
denatured single-stranded DNA the damage occurred at most
guanine residues. The amount of formation of 8-hydroxy-29-
deoxyguanosine (8-oxodGuo), an oxidative product of 29-
deoxyguanosine, in double-stranded DNA exceeded that in
single-stranded DNA. These results suggest that photoirradi-
ated MTX participates in 8-oxodGuo formation at the under-
lined G in 59-GG and 59-GGG sequences in double-stranded
DNA through electron transfer, and then 8-oxodGuo undergoes
further oxidation into piperidine-labile products. Fluorescence
measurement, high-pressure liquid chromatography and mass
spectrometry have demonstrated that photoexcited MTX is
hydrolyzed into 2,4-diamino-6-(hydroxymethyl)pteridine
Photoinduced damage to biomacromolecules has been investi-
gated using DNA as one of the target molecules (7–25). We have
previously reported the mechanism of DNA damage induced by
photoexcited drugs (20–25). In this study, to approach the
clarification of the mechanism of biomacromolecules damage by
photoirradiated MTX, we examined MTX-mediated photolesions
3
2
of P-labeled DNA fragments obtained from a human gene as
a target molecule. We also measured the content of 8-hydroxy-29-
deoxyguanosine (8-oxodGuo) (26,27), an oxidative product of 29-
deoxyguanosine (dGuo), formed by photoirradiating MTX with an
electrochemical detector coupled with high-performance liquid
chromatography (HPLC-ECD). Photochemical degradation of
MTX was investigated by fluorescence measurements. Further-
more, the photolysis product of MTX was analyzed by HPLC and
mass spectrometry. The DNA-damaging ability of 2,4-diamino-6-
(
DHP). DNA damage induced by DHP was observed in a similar
manner as was the damage induced by MTX. The extent of DNA
damage and the formation of 8-oxodGuo by DHP were much
larger than those induced by MTX. The kinetic analysis, based
on the time course of DNA oxidation by photoirradiated MTX,
suggests that DNA damage is caused by photoexcited DHP
rather than by photoexcited MTX. In conclusion, photoexcited
MTX undergoes hydrolysis through intramolecular electron
transfer, resulting in the formation of DHP, which exhibits
a phototoxic effect caused by oxidation of biomacromolecules
through photoinduced electron transfer.
(hydroxymethyl) pteridine (DHP), which is a photoproduct of
MTX, also has been examined.
MATERIALS AND METHODS
{
Posted on the web site on 28 August 2002.
4
Materials. Restriction enzymes (ApaI and HindIII) and T polynucleotide
ꢀ
To whom correspondence should be addressed at: Department of
Environmental and Molecular Medicine, Mie University School of
Medicine, Edobashi 2-174, Tsu, Mie 514-8507, Japan. Fax: 81-59-231-
kinase were purchased from New England Biolabs (Berverly, MA). A
restriction enzyme (EcoRI) and calf intestine phosphatase were from
Boehringer Mannheim GmbH (Mannheim, Germany). MTX was from
3
2
21
5011; e-mail: kawanisi@doc.medic.mie-u.ac.jp
Aldrich Chemical Co. (Milwaukee, WI). [c- P]-ATP (222 TBq mmol )
Abbreviations: dGuo, 29-deoxyguanosine; DHP, 2,4-diamino-6-(hydrox-
ymethyl) pteridine; DTPA, diethylenetriamine-N,N,N9,N0,N0-pentaace-
tic acid; HPLC-ECD, high-pressure liquid chromatography equipped
was from New England Nuclear (Boston, MA). Diethylenetriamine-
N,N,N9,N0,N0-pentaacetic acid (DTPA) was from Dojin Chemicals Co.
(Kumamoto, Japan). Calf thymus DNA was from Sigma Chemical Co. (St.
1
with an electrochemical detector; MTX, methotrexate;
O
2
, singlet
1
Louis, MO). Nuclease P was from Yamasa Shoyu Co. (Chiba, Japan).
oxygen; 8-oxodGuo, 8-hydroxy-29-deoxyguanosine (also known as 8-
oxo-7,8-dihydro-29-deoxyguanosine).
DHP was from Acros Organics (New Jersey).
Preparation of P-59–end-labeled DNA fragments. The DNA frag-
ment of the human p53 tumor suppressor gene was prepared from pUC18
32
ꢀ
2002 American Society for Photobiology 0031-8655/01
$5.0010.00
4
67

4
68 Kazutaka Hirakawa et al.
7
.8) containing 5 lM DTPA were exposed to UVA light (kmax 5 365 nm).
After ethanol precipitation, DNA was digested to the nucleosides with
nuclease P and calf intestine phosphatase and analyzed with an HPLC-
1
ECD as described previously (23).
Fluorescence measurements of MTX and DHP. The fluorescence
spectra of MTX and DHP were measured with an RF-5300PC
spectrophotometer (Shimadzu, Kyoto, Japan). All samples for the
fluorescence spectra were measured in 10 mM sodium phosphate buffer
(
pH 7.8).
Assay of photolysis of MTX. MTX (200 lM) in 10 mM phosphate
buffer (pH 7.8) was irradiated with UVA light (kmax 5 365 nm, 1.16 mW
2
2
cm ) and analyzed by HPLC consisting of an LC-6A pump (Shimadzu)
and equipped with a Wakopak ODS column (i.d. 4.6 3 150 mm, Wako,
Osaka, Japan). The mobile phase consisted of 89.8% (vol/vol) of water,
1
0.0% (vol/vol) of acetonitrile and 0.2% (vol/vol) of acetic acid. The
analysis was carried out at a column temperature of 258C and a flow rate of
2
.5 mL min . The HPLC eluate was routed directly into a photodiode
1
0
array UV–visible detector (SPD-M10A, Shimadzu), and the spectrum of the
eluate was measured. The concentration of detected MTX and photo-
products were estimated by using absorbance at 340 nm.
Mass spectra measurements. Laser desorption mass spectrometry was
performed on a Voyager MALDI-TOFMS (Applied Biosystems, Foster
City, CA) equipped with a nitrogen laser (337 nm, 3 ns pulse) to determine
the molecular weight of photoproducts of MTX. The fraction isolated from
the UVA-irradiated MTX solution by HPLC (retention time: 2.8 min) was
air-dried on a stainless steel probe tip. No matrix solution was added to the
sample. Mass spectrum was obtained to sum 50 laser shots.
RESULTS
Damage to DNA fragments induced by photoirradiated
MTX or DHP
3
2
Figure 1 shows the autoradiogram of P-59–end-labeled DNA
fragments exposed to UVA light (kmax 5 365 nm) in the presence
of MTX or DHP. UVA irradiation caused DNA damage in the
presence of MTX (Fig. 1A), whereas UVA irradiation alone caused
little or no DNA damage under the conditions used. DHP also
induced DNA photodamage (Fig. 1B), and the extent of DNA
damage induced by DHP was greater than the extent of damage
induced by MTX. DNA photodamage was observed when the
DNA fragment was treated with piperidine, suggesting that the
base modification was induced by photoirradiated MTX or DHP
Figure 1. Autoradiogram of DNA fragments exposed to UVA light in the
presence of MTX or DHP. The reaction mixture contained the P-labeled
32
ꢀ
443-base pair (ApaI 14179–EcoRI 14621) fragment, 5 lM/base of calf
thymus DNA, and the indicated concentration of MTX (A) or DHP (B) in
1
00 lL of 10 mM sodium phosphate buffer (pH 7.8) containing 5 lM
2
2
DTPA. The reaction mixtures were exposed to 7 J cm UVA light (kmax
65 nm). Then, the DNA fragments were treated with piperidine and
analyzed by the method described in Materials and Methods.
5
3
(data not shown).
plasmid, ligated fragments containing exons of the p53 gene (28). The 59–
Sequence specificity of DNA damage by photoirradiated
MTX or DHP
ꢀ
ꢀ
end-labeled 650-base pair (HindIII 13972–EcoRI 14621) fragment was
obtained as described previously (29). This fragment was further digested
with ApaI to obtain the singly labeled 443-base pair (ApaI 14179–
ꢀ
ꢀ
Figure 2 shows the sequence specificity of DNA damage induced
by photoirradiated MTX or DHP. Photoirradiated MTX caused
damage to double-stranded DNA fragments at guanine residues,
especially at the underlined G of 59-GG-39 and 59-GGG-39 (59-
TGGG-39) sequences (Fig. 2A). The difference in reactivity of GG
doublets may be due to the difference in the oxidation potential of
GG, depending on the neighboring sequences. When replacing
MTX with DHP, the result was almost the same (Fig. 2A,C). When
denatured single-stranded DNA fragments were used, MTX or
DHP plus photoirradiation caused DNA damage at single guanines
and consecutive guanine residues (Fig. 2B,D). The sequence
specificity of DNA damage by photoirradiated MTX in double-
stranded DNA was almost consistent with that of DNA damage by
photoexcited riboflavin, a Type-I photosensitizer (data not shown).
EcoRI 14621) and 211-base pair (HindIII 13972–ApaI 14182) fragments.
Detection of DNA damage induced by photoirradiated MTX or
DHP. The standard reaction mixture in a microtube (1.5 mL Eppendorf)
32
contained MTX or DHP, P-labeled DNA fragment and calf thymus DNA
in 100 lL of 10 mM sodium phosphate buffer (pH 7.8). Denatured single-
stranded DNA fragments were prepared by heating DNA fragments at 908C
for 10 min, followed by quick chilling before exposure to UVA light. The
2
2
mixtures were exposed to 6 or 7 J cm UVA light by using 10 W UV
lamps (kmax 5 365 nm, UVP, Inc., California) placed at a distance of 20
cm. After irradiation, the DNA fragments were treated with 1 M piperidine
at 908C for 20 min and treated as described previously (30). The DNA
fragments were subjected to electrophoresis on an 8 M urea and 8%
polyacrylamide gel. The autoradiogram was obtained by exposing an X-ray
film to the gel. The preferred cleavage sites were determined by direct
comparison of the positions of the oligonucleotides with those produced by
the chemical reactions of the Maxam–Gilbert procedure (31) by using
a DNA sequencing system (LKB2010 Macrophor, Pharmacia Biotech,
Uppsala, Sweden). A laser densitometer (LKB 2222 UltroScan XL,
Pharmacia Biotech) was used for measurement of the relative amounts of
oligonucleotides from the treated DNA fragments.
Formation of 8-oxodGuo induced by photoirradiated
MTX or DHP
Measurement of 8-oxodGuo formation induced by photoirradiated
MTX or DHP. The reaction mixtures containing 100 lM/base calf
thymus DNA and MTX or DHP in 4 mM sodium phosphate buffer (pH
Figure 3 shows 8-oxodGuo formation induced by photoirradiation
in the presence of MTX or DHP. It has been reported that

Photochemistry and Photobiology, 2002, 76(5) 469
Figure 2. Sequence-specificity of
DNA damage induced by photoirra-
diation in the presence of MTX or
DHP. The reaction mixture contained
3
2
the
(HindIII 13972–ApaI 14182) frag-
P-labeled 211-base pair
ꢀ
ment, 5 lM/base of calf thymus DNA,
and 100 lM MTX (A,B) or 50 lM
DHP (C,D) in 100 lL of 10 mM
sodium phosphate buffer (pH 7.8)
containing 5 lM DTPA. Where in-
32
dicated, the P-labeled and calf thy-
mus DNA fragments were denatured
by heating at 908C for 5 min, followed
by quick chilling on ice (B,D). The
reaction mixtures were exposed to
2
2
6
J cm
UVA light (kmax 5 365
nm). Subsequently, the DNA frag-
ments were treated with piperidine.
The DNA was analyzed, and the
relative amounts of oligonucleotides
were measured by the methods de-
scribed in Materials and Methods.
Horizontal axis shows the nucleotide
numbers of the p53 tumor suppressor
gene.
formation of 8-oxodGuo can cause DNA misreplication that may
lead to mutation or cancer (32,33). The amount of 8-oxodGuo
formation increased in a dose-dependent manner up to 50 lM of
MTX or DHP and decreased at concentrations of over 50 lM,
suggesting that the 8-oxodGuo formed is converted into further
oxidative products. The level of 8-oxodGuo in double-stranded
DNA exceeded that in single-stranded DNA. The formation of 8-
oxodGuo induced by DHP was about twice as much as that
induced by MTX.
analysis, and the intensity increased in a dose-dependent manner
(data not shown). The absorption spectrum of this product was
similar to that of a p-(N-methylamino)benzoyl compound. These
findings indicate that photoirradiated MTX generates DHP and
a p-(N-methylamino)benzoyl compound, possibly p-(N-methyl-
amino)benzoyl-L-glutamic acid.
Photolysis of MTX into DHP
Fluorescence intensity of the pteridine moiety of MTX was very
weak compared with that of DHP (Fig. 4), though pteridine is
a strong fluorescent molecule. Relative fluorescence quantum yield
r
(U ) of MTX, estimated from the comparison with that of
2
3
DHP, is 4.4 3 10 . When MTX was previously treated with
photoirradiation, the fluorescence intensity increased depend-
ing on the extent of photoirradiation (Fig. 4), suggesting that
UVA light can convert MTX into a stronger fluorescent
molecule. The fluorescence spectrum of UVA-irradiated MTX
was similar to that of DHP, suggesting that the photoexcited
MTX is hydrolyzed into DHP.
Figure 5A shows the chromatogram of MTX. The peak height
of MTX was decreased by photoirradiation, and the peak of
a photolysis product was observed. The retention time and the
absorption spectrum of the photolysis product were similar to those
of DHP (data not shown). Furthermore, the photoproduct of MTX
was analyzed by using a mass spectrometer. Mass spectrum with
Figure 3. Formation of 8-oxodGuo induced by photoirradiated MTX or
DHP. The reaction mixture containing 100 lM/base double-stranded (ds)
or single-stranded (ss) calf thymus DNA, and the indicated concentration of
MTX or DHP in 100 lL of 4 mM sodium phosphate buffer (pH 7.8)
containing 5 lM DTPA was exposed to 6 J cm UVA light (kmax 5 365
nm). Where indicated, the DNA fragment was denatured by heating for
1
a molecular ion at m/e 192 (M ) was obtained, confirming the
formation of DHP. The concentration of MTX decreased, whereas
that of DHP increased, depending on the photoirradiation dose
2
2
5
min at 908C followed by chilling on ice before photoirradiation. After
(Fig. 5B). Another photohydrolysis product (absorption maximum
at 279 nm, retention time of 6.4 min) was detected by HPLC
irradiation, DNA was treated, and the amount of 8-oxodGuo was measured
by the methods described in Materials and Methods.

4
70 Kazutaka Hirakawa et al.
Figure 5. A: Chromatogram of MTX with or without photoirradiation. B:
The time course of photolysis of MTX. MTX (200 lM) in 10 mM
phosphate buffer (pH 7.8) was irradiated with UVA light (kmax 5 365 nm,
Figure 4. Fluorescence spectra of MTX previously treated with photo-
irradiation (solid line) and DHP (dotted line). MTX (200 lM) in 10 mM
phosphate buffer (pH 7.8) was previously irradiated with UVA light
2
2
1
8
.16 mW cm ) and analyzed by HPLC using a mobile phase consisting of
9.8% (vol/vol) of water, 10.0% (vol/vol) of acetonitrile and 0.2% (vol/vol)
2
2
of acetic acid. A: The irradiation time was 57.4 min. B: The symbols are the
concentrations of MTX (ꢃ) and DHP ( ) detected by HPLC.
(kmax 5 365 nm, 1.16 mW cm ) for the time indicated and diluted into
s
2 lM solution for measurement. The concentration of DHP was 1 lM.
These spectra were measured in 10 mM sodium phosphate buffer (pH 7.8)
at 340 nm excitation.
from Eq. 2 using the a 3 kox1 and kox2 values mentioned above.
The value of [DHP] formed from MTX was obtained from the
curve fitting of the plots in Fig. 5. The curve obtained in Fig. 6
clearly fits with the experimental results, supporting the idea that
DNA oxidation is caused by DHP generated from MTX.
Measurement and simulation of the time course of the
8
-oxodGuo formation by photoirradiated MTX or DHP
Formation of 8-oxodGuo by MTX or DHP increased depending on
the photoirradiation time (Fig. 6). 8-oxodGuo formation by DHP
was significantly larger than that by MTX. An induction period
DISCUSSION
(
approximately 15 min) was observed in the 8-oxodGuo formation
by MTX, suggesting that MTX itself could cause little or no DNA
oxidation and that reactive species generated from photoirradiated
MTX induced DNA oxidation.
Guanine-specificoxidationcanbecausedbyexcitedphotosensitizers
through Type-I (electron transfer) mechanism, Type-II (singlet
1
oxygen [ O
2
]) mechanism (7,8,20), or both. Cadet and co-workers
The curves in this figure were obtained by simulation according
to the following procedure. The 8-oxodGuo formation is roughly
expressed as
(9,10) have reported that imidazolone and oxazolone are the major
oxidation products of guanine formed by the Type-I mechanism,
1
whereas 8-oxodGuo was found to be the main O
2
-mediated guanine
oxidation product (8). The present study has demonstrated that
photoirradiated MTX and DHP induce base oxidation, especially at
the underlined G in 59-GG-39 and 59-GGG-39 (59-TGGG-39)
sequencesindouble-strandedDNA.Wepreviouslyobservedasimilar
DNAcleavagepatternwhenaType-Iphotosensitizer, riboflavin, was
used(23).Theamountof8-oxodGuoformedbyphotoirradiatedMTX
in double-strandedDNAwasmuch largerthan thatinsingle-stranded
ꢀ
DHP þ hm ꢁ! DHP and
kox1
kox2
ꢀ
DHP þ dGuo ꢁ! dGuoox ꢁꢁꢁ! 8-oxodGuo
ð1Þ
H2O;ꢁeꢁ
ꢀ
where DHP , dGuoox, and kox1 and kox2 are photoexcited DHP,
oxidized dGuo (intermediate into 8-oxodGuo) and reaction rate
constants, respectively. The rate equation can be practically
expressed as follows:
DNA. We have previously reported that the sequence specificity of
1
DNA damage by O
2
is quite different from that by the Type-I
ꢀ
½
dGuooxꢂ=dt ¼ kox1½DHP ꢂ½dGuoꢂ ¼ kox1a½DHPꢂ½dGuoꢂ and
mechanism (20–22). Therefore, the present results can be reasonably
explained by assuming that nucleobase oxidation is induced by the
activated photosensitizer mainly through electron transfer (Type-I
d½8-oxodGuoꢂ=dt ¼ kox2½dGuoox
ꢂ
ð2Þ
ꢀ
1
2
where [dGuo], [dGuoox], [8-oxodGuo], [DHP] and [DHP ] are
concentrations of dGuo, dGuo , 8-oxodGuo, ground state of DHP
mechanism), although O participates in DNA damage to some
ox
extent.GuanineisthemosteasilyoxidizedamongthefourDNAbases
because the oxidation potential of guanine is lower than those of the
otherDNAbases(7,34). Molecularorbitalcalculationshaverevealed
thatstackingoftwoguaninebasesindouble-strandedDNAlowersthe
ionization potential significantly, and electron-loss centers are
localized on the 59-G of the 59-GG-39 sequence through charge
transfer (16). Recently, it has been reported that the ionization
potential of GGG is lower than that of GG and that the central G is the
mostreactiveinthe59-TGGG-39sequence(19).Thesereportssupport
the proposed mechanism by which the photoirradiated photosensi-
and excited state of DHP, respectively. The nondimensional co-
efficient a depends on the number of photons absorbed by DHP. The
fitting curve of time course of 8-oxodGuo formation has been
obtained by numerical calculation. The values of a 3 kox1 and kox2
were determined by the time courseof 8-oxodGuo formation by DHP
2
1
21
6
21
as shown in Fig. 6 and were 0.12 M
s
and 7.4 3 10 s
,
respectively. Based on the assumption that photoexcited DHP
generated from MTX oxidizes dGuo, although photoexcited MTX
does not, the time course of 8-oxodGuo formation was calculated

Photochemistry and Photobiology, 2002, 76(5) 471
Figure 6. Time course of the formation of 8-oxodGuo induced by
photoirradiated MTX or DHP. The reaction mixture containing 100 lM/
base calf thymus DNA, 50 lM MTX (ꢃ) or DHP (
s
) in 100 lL of 4 mM
sodium phosphate buffer (pH 7.8) containing 5 lM DTPA was exposed to
2
2
UVA light (kmax 5 365 nm, 1.16 mW cm ). After irradiation, DNA was
treated, and the amount of 8-oxodGuo was measured by the methods
described in Materials and Methods. Results represent means 6 SD of four
independent experiments. The curves in this figure were simulated as
described in Results.
tizer causes the poly-G–specific oxidation in double-stranded DNA
through electron transfer. Guaninecation radicalsformed inDNAare
finally localized on the underlined sites of the 59-GG-39 and59-GGG-
39(59-TGGG-39)sequencesandreactwiththewatermoleculetoform
the C-8 OH adduct radical, followed by oxidation, leading to the
formationof8-oxodGuo(7,8,13–15).Althoughthe8-oxodGuositeis
not efficiently cleaved under piperidine treatment (15), 8-oxodGuo is
more easily oxidized than dGuo is and can be converted into
piperidine-labile products (e.g. imidazolone, oxazolone, or both [7–
11]) through further oxidation (7,17,18,35,36).
Pteridine is a strong fluorescent molecule, but the pteridine
moiety of MTX shows weak fluorescence, possibly due to an
intramolecular electron transfer from the p-(N-methylamino)-
benzoyl residue to the photoexcited pteridine moiety. This study
showed that photoirradiation converted MTX into the fluorescent
molecule, with the fluorescence spectrum similar to that of DHP.
HPLC and mass spectrum measurements have also shown that
photoexcited MTX generates DHP depending on the irradia-
tion dose. We proposed the photolysis of MTX into DHP and a
p-(N-methylamino)benzoyl compound, possibly p-(N-methylami-
no)benzoyl-L-glutamic acid, through an intramolecular electron
transfer after hydrolysis (Fig. 7).
Figure 7. Proposed mechanism of DNA damage induced by photo-
irradiated MTX.
2
3
r
estimated that the value of U is 4.4 3 10 , the value of ket1 can
1
be estimated to be almost 10 –10
0
11 21
s
. On the other hand, ket2,
Sequence specificity of DNA damage induced by photoexcited
DHP was similar to that of photoirradiated MTX, and the extent of
DNA damage and 8-oxodGuo formation by DHP was much larger
than in the case of MTX. The time course of 8-oxodGuo formation
by photoirradiated MTX has been reasonably explained by
assuming that DNA damage is caused by photoactivation of
DHP generated from MTX and that MTX itself has little or no
DNA-damaging ability. In the photoexcited MTX the intra-
molecular electron transfer leading to DHP formation competes
with the intermolecular reaction with DNA. The intramolecular
electron transfer rate coefficient (ket1) has been roughly estimated
from the following equation:
the collision rate coefficient of one photoexcited MTX molecule
with DNA, has been estimated from the equation of diffusion
controlled reaction:
k
et2 ¼ 8RT½dGuoꢂ=3g
ð4Þ
where R is the gas constant, T is the absolute temperature, [dGuo]
2
3
21
is 25 lM and g is the viscosity of water (0.891 3 10 kg m
2
1
5 21
s
). The ket2 is calculated to be 1.8 3 10 s and is markedly
smaller than ket1. Therefore, in the photoexcited MTX, intra-
molecular electron transfer preferentially occurs rather than DNA
oxidation by photoexcited MTX itself.
In summary, the present study has demonstrated that photo-
irradiated MTX induces poly-G–specific DNA oxidation. The
reactive species that induces DNA oxidation is photoexcited DHP,
which is generated from photoexcited MTX through hydrolysis
after intramolecular electron transfer. This study suggests that the
k
et1 ¼ ð1=È
r
ꢁ 1Þ 3 1=s
0
ð3Þ
where s
ranges almost from 10 to 10
0
is the lifetime of the photoexcited pteridine moiety
2
9
28 21
s
(
) (37). Because this study has

4
72 Kazutaka Hirakawa et al.
phototoxicity of MTX is due to the oxidation of biomacromole-
cules, mainly through electron transfer by photoexcited DHP.
zolone as a major guanine oxidation product. J. Am. Chem. Soc. 120,
373–7374.
8. Kino, K. and H. Sugiyama (2001) Possible cause of G-CfiC-G
transversion mutation by guanine oxidation product, imidazolone.
Chem. Biol. 8, 369–378.
9. Yoshioka, Y., Y. Kitagawa, Y. Takano, K. Yamaguchi, T. Nakamura
and I. Saito (1999) Experimental and theoretical studies on the
selectivity of GGG triplets toward one-electron oxidation in B-form
DNA. J. Am. Chem. Soc. 121, 8712–8719.
0. Kawanishi, S., Y. Hiraku and S. Oikawa (2001) Mechanism of guanine-
specific DNA damage by oxidative stress and its role in carcinogenesis
and aging. Mutat. Res. 488, 65–76.
7
1
1
Acknowledgement—This work was supported by Grants-in-Aid for Sci-
entific Research granted by the Ministry of Education, Science, Sports
and Culture of Japan.
REFERENCES
2
1
. Elisei, F., L. Latterini, G. G. Aloisi, U. Mazzucato, G. Viola, G. Mioio,
D. Vedaldi and F. Dall’Acqua (2002) Excited-state properties and in
vitro phototoxicity studies of three phenothiazine derivatives. Photo-
chem. Photobiol. 75, 11–21.
21. Kawanishi, S., S. Inoue, S. Sano and H. Aiba (1986) Photodynamic
guanine modification by hematoporphyrin is specific for single-stranded
DNA with singlet oxygen as a mediator. J. Biol. Chem. 261, 6090–
6095.
2. International Agency for Research on Cancer (IARC) Working Group
(
1980) IARC Monographs on the Evaluation of the Carcinogenic
Risk of Chemicals to Humans, Vol. 26. IARC, Lyon, France.
. Okamoto, H., A. Fukuda, K. Mizuno, N. Matsuyoshi, K. Fujii and S.
Imamura (1994) Reactivation of phototoxicity test for psoralens plus
ultraviolet A by low-dose methotrexate. Photodermatol. Photoimmu-
nol. Photomed. 10, 134–136.
22. Hiraku, Y. and S. Kawanishi (2000) Distinct mechanism of guanine-
specific DNA photodamage induced by nalidixic acid and fluoroqui-
nolone antibacterials. Arch. Biochem. Biophys. 382, 211–218.
23. Ito, K., S. Inoue, K. Yamamoto and S. Kawanishi (1993) 8-
Hydroxydeoxyguanosine formation at the 59 site of 59-GG-39 sequences
in double-stranded DNA by UV radiation with riboflavin. J. Biol.
Chem. 268, 13221–13227.
24. Ito, K. and S. Kawanishi (1997) Photoinduced hydroxylation of
deoxyguanosine in DNA by pterins: sequence specificity and
mechanism. Biochemistry 36, 1774–1781.
3
4
5
6
. Morison, W. L., K. Momtaz, J. A. Parrish and T. B. Fitzpatrick (1982)
Combined methotrexate-PUVA therapy in the treatment of psoriasis.
J. Am. Acad. Dermatol. 6, 46–51.
. Andersen, K. E. and R. Lindskov (1984) Recall of UVB-induced
erythema in breast cancer patient receiving multiple drug chemother-
apy. Photodermatology 1, 129–132.
25. Ito, K. and S. Kawanishi (1997) Site-specific DNA damage induced by
UVA radiation in the presence of endogenous photosensitizer. Biol.
Chem. 378, 1307–1312.
. Morison, W. L., S. Marwaha and L. Beck (1997) PUVA-induced
phototoxicity: incidence and causes. J. Am. Acad. Dermatol. 36, 183–
1
85.
26. Kasai, H., P. F. Crain, Y. Kuchino, S. Nishimura, A. Ootsuyama and
H. Tanooka (1986) Formation of 8-hydroxyguanine moiety in cellular
DNA by agents producing oxygen radicals and evidence for its repair.
Carcinogenesis 7, 1849–1851.
27. Nair, U. J., R. A. Floyd, J. Nair, V. Bussachini, M. Friesen and H.
Bartsch (1987) Formation of reactive oxygen species and of 8-
hydroxydeoxyguanosine in DNA in vitro with betel quid ingredients.
Chem.-Biol. Interact. 63, 157–169.
7
. Burrows, C. J. and J. G. Muller (1998) Oxidative nucleobase
modifications leading to strand scission. Chem. Rev. 98, 1109–1151.
. Ravanat, J. L., T. Douki and J. Cadet (2001) Direct and indirect effects
of UV radiation on DNA and its components. J. Photochem.
Photobiol. B: Biol. 63, 88–102.
. Cadet, J., M. Berger, G. W. Buchko, P. C. Joshi, S. Raoul and J.-L.
Ravanat (1994) 2,2-Diamino-4-[(3,5-di-O-acetyl-2-deoxy-b-D-erythro-
pentofuranosyl)amino]-5-(2H)-oxazolone: a novel and predominant
radical oxidation product of 39,59-di-O-acetyl-29-deoxyguanosine. J.
Am. Chem. Soc. 116, 7403–7404.
8
9
28. Chumakov, P. (1990) EMBL Data Library, accession number
X54156. European Molecular Biology Laboratory, Heidelberg, Ger-
many.
1
1
0. Raoul, S., M. Berger, G. W. Buchko, P. C. Joshi, B. Morin, M.
29. Yamashita, N., M. Murata, S. Inoue, Y. Hiraku, T. Yoshinaga and S.
Kawanishi (1998) Superoxide formation and DNA damage induced by
a fragrant furanone in the presence of copper(II). Mutat. Res. 397,
191–201.
30. Yamamoto, K. and S. Kawanishi (1989) Hydroxyl free radical is not
the main active species in site-specific DNA damage induced
by copper(II) ion and hydrogen peroxide. J. Biol. Chem. 264,
15435–15440.
31. Maxam, A. M. and W. Gilbert (1980) Sequencing end-labeled DNA with
base-specific chemical cleavages. Methods Enzymol. 65, 499–560.
32. Bruner, S. D., D. P. G. Norman and G. L. Verdine (2000) Structural
basis for recognition and repair of the endogenous mutagen 8-
oxoguanine in DNA. Nature 403, 859–866.
33. Shibutani, S., M. Takeshita and A. P. Grollman (1991) Insertion of
specific bases during DNA synthesis past the oxidation-damaged base
8-oxodG. Nature 349, 431–434.
1
13
15
Weinfeld and J. Cadet (1996) H, C and N nuclear magnetic
resonance analysis and chemical features of the two main radical
oxidation products of 29-deoxyguanosine: oxazolone and imidazolone
nucleosides. J. Chem. Soc. Perkin Trans. 2, 371–381.
1. Cadet, J., G. Remaud and J. L. Ravanat (2000) Measurement of the
main photooxidation products of 29-deoxyguanosine using chromato-
graphic methods coupled to mass spectrometry. Arch. Biochem.
Biophys. 374, 118–127.
1
1
2. Douki, T., D. Angelov and J. Cadet (2001) UV laser photolysis of
DNA: effect of duplex stability on charge-transfer efficiency. J. Am.
Chem. Soc. 123, 11360–11366.
3. Kasai, H., Z. Yamaizumi, M. Berger and J. Cadet (1992) Photo-
sensitized formation of 7,8-dihydro-8-oxo-29-deoxyguanosine (8-hy-
droxy-29-deoxyguanosine) in DNA by riboflavin: a non singlet oxygen
mediated reaction. J. Am. Chem. Soc. 114, 9692–9694.
1
1
4. Yamamoto, F., S. Nishimura and H. Kasai (1992) Photosensitized
formation of 8-hydroxydeoxyguanosine in cellular DNA by riboflavin.
Biochem. Biophys. Res. Commun. 187, 809–813.
5. Cullis, P. M., M. E. Malone and L. A. Merson-Davies (1996) Guanine
radical cations are precursors of 7,8-dihydro-8-oxo-29-deoxyguanosine
but are not precursors of immediate strand breaks in DNA. J. Am.
Chem. Soc. 118, 2775–2781.
6. Sugiyama, H. and I. Saito (1996) Theoretical studies of GG-specific
photocleavage of DNA via electron transfer: significant lowering of
ionization potential and 59-localization of HOMO of stacked GG bases
in B-form DNA. J. Am. Chem. Soc. 118, 7063–7068.
34. Steenken, S. and S. Jovanovic (1997) How easily oxidizable is DNA?
One-electron reduction potentials of adenosine and guanosine radicals
in aqueous solution. J. Am. Chem. Soc. 119, 617–618.
35. Luo, W., J. G. Muller and C. J. Burrows (2001) The pH-dependent role
of superoxide in riboflavin-catalyzed photooxidation of 8-oxo-7,8-
dihydroguanosine. Org. Lett. 3, 2801–2804.
36. Luo, W., J. G. Muller, E. M. Rachlin and C. J. Burrows (2001)
Characterization of hydantoin products from one-electron oxidation of
8-oxo-7,8-dihydroguanosine in a nucleoside model. Chem. Res.
Toxicol. 14, 927–938.
1
37. Hawkins, M. E., W. Pfleiderer, O. Jungmann and F. M. Balis (2001)
Synthesis and fluorescence characterization of pteridine adenosine nu-
cleoside analogs for DNA incorporation. Anal. Biochem. 298, 231–240.
1
7. Kino, K., I. Saito and H. Sugiyama (1998) Product analysis of GG-
specific photooxidation of DNA via electron transfer: 2-aminoimida-