Independent Generation of 5-(2′-Deoxycytidinyl)methyl Radical and the Formation of a Novel Cross-Link Lesion between 5-Methylcytosine and Guanine

Article abstract of DOI:10.1021/ja034866r

Reactive oxygen species (ROS) can damage DNA. Although a number of single nucleobase lesions induced by ROS have been structurally characterized, only a few intrastrand cross-link lesions have been identified and characterized, and all of them involve adj

Full text of DOI:10.1021/ja034866r

Published on Web 09/18/2003
Independent Generation of 5-(2′-Deoxycytidinyl)methyl
Radical and the Formation of a Novel Cross-Link Lesion
between 5-Methylcytosine and Guanine
Qibin Zhang and Yinsheng Wang*
Contribution from the Department of Chemistry-027, UniVersity of California at RiVerside,
RiVerside, California 92521-0403
Received February 25, 2003; E-mail: yinsheng.wang@ucr.edu
Abstract: Reactive oxygen species (ROS) can damage DNA. Although a number of single nucleobase
lesions induced by ROS have been structurally characterized, only a few intrastrand cross-link lesions
have been identified and characterized, and all of them involve adjacent thymine and guanine or adenine.
In mammalian cells, the cytosines at CpG sites are methylated. On the basis of the similar reactivity of
5-methylcytosine and thymine toward hydroxyl radical and the similar orientation of adjacent thymine guanine
(TG) and 5-methylcytosine guanine (mCG) in B-DNA, we predict that the cross-link lesion, which was
identified in TG and has a covalent bond formed between the 5-methyl carbon atom of T and the C8
carbon atom of G, should also form at mCG site. Here, we report for the first time the independent generation
of 5-(2′-deoxycytidinyl)methyl radical, and our results demonstrate that this radical can give rise to the
predicted novel intrastrand cross-link lesion in dinucleoside monophosphates d(mCG) and d(GmC).
Furthermore, we show that the cross-link lesion can also form in d(mCG) from γ irradiation under anaerobic
conditions.
Introduction
It is known that, upon γ irradiation, the three major sites for
hydroxyl radical attack on both thymine and mC are the C5,
C6, and the 5-methyl carbon atoms, and the yields for the
hydroxyl radical attack at these three different carbon atoms
are similar for the two nucleobases.15 Furthermore, in B-DNA,
the distance between the methyl carbon atom of thymine and
C8 carbon atom of the vicinal guanine is 6.36 Å, whereas the
corresponding distance in mCG is 5.51 Å.16 On the basis of the
similar reactivity between mC and T toward hydroxyl radical
and the similar orientation of the adjacent pyrimidine-purine
nucleobases at mCG and TG sites in B-DNA, we predict that a
similar cross-link lesion should form between the 5-methyl
carbon atom of mC and the C8 carbon atom of its 3′ adjacent
guanine. In addition, Pfeifer and co-workers17 recently observed
mCG f TT tandem double mutation in nucleotide excision
repair (NER)-deficient XP-A cells, indicating that cross-link
lesion might form at mCG site.
Reactive oxygen species (ROS) are generated by either
aerobic metabolism or exogenous oxidizing agents.1 ROS can
damage nucleic acids and a multitude of damaged products
involving a single nucleobase are known.2 Those adducts may
contribute to a number of pathological conditions including
cancer,2 neurodegeneration,3 and natural processes of aging.1
In mammalian cells, the cytosines at CpG sites are methylated.
Although CpGs are under-represented by 5-fold of their expected
frequency in mammalian DNA, methylated CpGs are mutational
hotspots and the most common mutations observed are C to T
transition mutations.4 In addition to single-nucleobase lesions,
several intrastrand cross-link lesions formed at adjacent TG and
GT sites have been structurally characterized.5-14 However, no
cross-link lesion between mC and G has been reported.
(1) Finkel, T.; Holbrook, N. J. Nature 2000, 408, 239-247.
(2) Dizdaroglu, M. NATO ASI Ser., Ser. A 1999, 302, 67-87.
(3) Rolig, R. L.; McKinnon, P. J. Trends Neurosci. 2000, 23, 417-424.
(4) Pfeifer, G. P. Mutat. Res. 2000, 450, 155-166.
(5) Romieu, A.; Bellon, S.; Gasparutto, D.; Cadet, J. Org. Lett. 2000, 2, 1085-
1088.
To address the mechanism of ROS-induced nucleic acid
damage, Greenberg,18-24 Giese,25-28 and Cadet et al.5,6 reported
(6) Bellon, S.; Ravanat, J. L.; Gasparutto, D.; Cadet, J. Chem. Res. Toxicol.
2002, 15, 598-606.
(14) Box, H. C.; Dawidzik, J. B.; Budzinski, E. E. Free Radic. Biol. Med. 2001,
31, 856-868.
(7) Delatour, T.; Douki, T.; Gasparutto, D.; Brochier, M. C.; Cadet, J. Chem.
Res. Toxicol. 1998, 11, 1005-1013.
(15) von Sonntag, C. The Chemical Basis of Radiation Biology; Taylor &
Francis: London, 1987.
(8) Box, H. C.; Freund, H. G.; Budzinski, E. E.; Wallace, J. C.; Maccubbin,
A. E. Radiat. Res. 1995, 141, 91-94.
(16) The distance between the C8 of G and the 5-methyl carbon of mC was the
average value determined from the X-ray structures of duplex DNA
(bdlb72.ndb, bdlb73.ndb, and bdlb74.ndb from http://ndbserver.rutgers.edu).
The corresponding distance in TG was determined from structures created
in software package InsightII using standard B-DNA geometry.
(17) Lee, D. H.; O’Connor, T. R.; Pfeifer, G. P. Nucleic Acids Res. 2002, 30,
3566-3573.
(18) Barvian, M. R.; Greenberg, M. M. J. Am. Chem. Soc. 1995, 117, 7, 8291-
8292.
(19) Barvian, M. R.; Greenberg, M. M. J. Org. Chem. 1995, 60, 1916-1917.
(9) Box, H. C.; Budzinski, E. E.; Dawidzik, J. D.; Wallace, J. C.; Evans, M.
S.; Gobey, J. S. Radiat. Res. 1996, 145, 641-643.
(10) Box, H. C.; Budzinski, E. E.; Dawidzik, J. B.; Gobey, J. S.; Freund, H. G.
Free Radic. Biol. Med. 1997, 23, 1021-1030.
(11) Budzinski, E. E.; Dawidzik, J. B.; Rajecki, M. J.; Wallace, J. C.; Schroder,
E. A.; Box, H. C. Int. J. Radiat. Biol. 1997, 71, 327-336.
(12) Box, H. C.; Budzinski, E. E.; Dawidzik, J. B.; Wallace, J. C.; Iijima, H.
Radiat. Res. 1998, 149, 433-439.
(13) Box, H. C. Radiat. Res. 2001, 155, 634-636.
9
10.1021/ja034866r CCC: $25.00 © 2003 American Chemical Society
J. AM. CHEM. SOC. 2003, 125, 12795-12802
12795

A R T I C L E S
Zhang and Wang
Scheme 1
Reagents: (a) (i) (CH2O)n/TEA/H2O/60 °C. (ii) (CH3CO)2O, pyridine. (iii) PhSH/TEA/DMF/70 °C. (iv) NH3/MeOH. (b) DMTrCl/DMAP/TEA/pyridine.
(c) NC(CH2)2OP(Cl)N(iPr)2/DIEA/CH2Cl2. (d) POCl3/1,2,4-triazole/TEA/CH3CN. (e) (i) N2-isobutyryl-dG/tetrazole/DMF/CH3CN. (ii) I2/THF/H2O/pyridine.
(f) (i) 1% TFA/CH2Cl2. (ii) 29% NH4OH.
the synthesis and characterization of a number of photolabile
precursors of radical intermediates involved in radiation-induced
nucleic acid damage. One of the precursors, 5-(phenylthio-
methyl)-2′-deoxyuridine, which yields a 5-methyl radical of
thymine upon 254-nm irradiation, can give rise to a known
cross-link lesion where the 5-methyl carbon atom of thymine
and the C8 carbon atom of an adjacent guanine or adenine are
covalently bonded.5,6 This and earlier work by Box et al.11
demonstrated that the intrastrand cross-link lesion can initiate
from a single radical event. Therefore, we reason that a similar
photolabile radical precursor of mC may facilitate us to obtain
the predicted cross-link lesion between mC and guanine.
Here, in this paper, we report for the first time the independent
generation of the 5-(2′-deoxycytidinyl)methyl radical and the
formation of a novel intrastrand cross-link lesion between the
5-methyl carbon atom of 5-methylcytosine and the C8 carbon
atom of the adjacent guanine in dinucleoside monophosphates.
by Romieu and co-workers5 and Anderson et al.24 using
phenylthio photochemistry and Norrish type I photocleavage,
respectively.
We adopted the phenylthio photochemistry and employed a
nucleoside conversion29 step to prepare the desired phosphor-
amidite building block 5 from building block 4 (Scheme 1).
Compound 4 was prepared according to that reported by Cadet
and co-workers5,6 with some modifications to improve the yield.
The key modification was in step (a), where we added more
paraformaldehyde during the course of the reaction followed
by addition of triethylamine to keep the reaction solution basic,
which afforded a 51% yield for step a, compared to a 13% yield
reported in the previous study.5
Under standard nucleobase deprotection conditions, i.e.,
incubation with 30% ammonium hydroxide at 55 °C for 12 h,
the phenylthio moiety fully decomposed.5 We found that the
phenylthio moiety is stable upon prolonged treatment with 30%
ammonium hydroxide at room temperature. Such treatment for
48 h is sufficient for the removal of standard nucleobase
protecting groups. Therefore, we employed standard nucleobase
protection for the synthesis of dinucleoside monophosphates
containing 5-(phenylthiomethyl)-2′-deoxycytidine. Furthermore,
the base deprotection with ammonium hydroxide also converts
the 1,2,4-triazol-1-yl group to an amino group,30 which gives
us dinucleoside monophosphates with the desired 5-(phenyl-
thiomethyl)-2′-deoxycytidine.
Results and Discussion
1. Synthesis of Phosphoramidite Building Block of 4-(1,2,4-
Triazol-1-yl)-5-(phenylthiomethyl)-2′-deoxyuridine. Because
the 5-methyl radical of mC is expected to give rise to a cross-
link lesion where the 5-methyl carbon atom of mC and the C8
carbon atom of the adjacent G are covalently bonded, we
decided to first synthesize a photolabile precursor for this radical.
A phosphoramidite building block containing a photolabile
precursor of the 5-methyl radical of thymidine has been reported
With building block 5, we synthesized dinucleoside mono-
phosphate d(mCSPhG) (7), where mCSPh is 5-(phenylthiomethyl)-
2′-deoxycytidine, using solution-phase phosphoramidite chem-
istry (Scheme 1).31 We also prepared d(GmCSPh) (12) using
solution-phase chemistry starting from compound 3 (Scheme
2).
2. Demonstration of the Formation of the Intrastrand
Cross-Link Lesions from d(mCSPhG) and d(GmCSPh). We
then irradiated d(mCSPhG) and d(GmCSPh) with 254-nm UV
(20) Greenberg, M. M.; Yoo, D. J.; Goodman, B. K. Nucleosides & Nucleotides
1997, 16, 33-40.
(21) Greenberg, M. M.; Barvian, M. R.; Cook, G. P.; Goodman, B. K.; Matray,
T. J.; Tronche, C.; Venkatesan, H. J. Am. Chem. Soc. 1997, 119, 1828-
1839.
(22) Greenberg, M. M. Chem. Res. Toxicol. 1998, 11, 1235-1248.
(23) Tronche, C.; Goodman, B. K.; Greenberg, M. M. Chem. & Biol. 1998, 5,
263-271.
(24) Anderson, A. S.; Hwang, J. T.; Greenberg, M. M. J. Org. Chem. 2000, 65,
4648-4654.
(25) Giese, B.; Dussy, A.; Elie, C.; Erdmann, P.; Schwitter, U. Angew. Chem.,
Int. Ed. Engl. 1994, 33, 1941-1944.
(26) Giese, B.; Beyrich-Graf, X.; Erdmann, P.; Petretta, M.; Schwitter, U. Chem.
& Biol. 1995, 2, 367-375.
(29) Webb, T. R.; Matteucci, M. D. J. Am. Chem. Soc. 1986, 108, 2764-2765.
(27) Giese, B.; Dussy, A.; Meggers, E.; Petretta, M.; Schwitter, U. J. Am. Chem.
Soc. 1997, 119, 11 130-11 131.
(30) Divakar, K. J.; Reese, C. B. J. Chem. Soc., Perkin Trans. 1 1982, 1171-
1176.
(28) Gugger, A.; Batra, R.; Rzadek, P.; Rist, G.; Giese, B. J. Am. Chem. Soc.
1997, 119, 8740-8741.
(31) Tallman, K. A.; Greenberg, M. M. J. Am. Chem. Soc. 2001, 123, 5181-
5187.
9
12796 J. AM. CHEM. SOC. VOL. 125, NO. 42, 2003

Generation of 5-(2′-Deoxycytidinyl)methyl Radical
A R T I C L E S
Scheme 2
Reagents: (a) Ac2O/pyridine. (b) POCl3/1,2,4-triazole/TEA/CH3CN. (c) 80% AcOH. (d) (i) dG-CE phosphoramidite/DMF/CH3CN (ii) I2/THF/H2O/
pyridine. (e) (i) 80% AcOH. (ii) 29% NH4OH.
Figure 1. HPLC traces for the separation of the 254-nm irradiation mixtures
of d(mCSPhG) (a) and d(GmCSPh) (b).
Figure 2. Product-ion spectra of the ESI-produced [M + H]+ ions of
light, and indeed the cross-link lesion, in which the 5-methyl
d(mC∧G) (a) and d(G∧mC) (b).
carbon atom of mC and the C8 carbon atom of the adjacent
guanine are cross-linked, forms from both dinucleoside mono-
phosphates. The 22.9-min fraction from the HPLC of the
irradiation mixture of d(mCSPhG) is the anticipated cross-link
product [labeled as d(mC∧G) in Figure 1].
ESI-MS and tandem MS (MS/MS) indicate the presence of
the cross-link lesion. The cross-link lesion has a molecular
weight that is 2 amu less than that of the unmodified d(mCG),
which is what we observed in the positive-ion ESI-MS (ion
of m/z 569, data not shown). Product-ion spectrum of the [M
+ H]+ ion of d(mC∧G) showed the formation of an abundant
fragment ion of m/z 275, which is attributed to the protonated
ion of the cross-linked nucleobase moiety (Figure 2a). In
addition, we observed an ion corresponding to the loss of a
2-deoxyribose (ion of m/z 471). The assignments of the fragment
ions are supported by exact mass measurements on a Fourier
transform ion cyclotron resonance mass spectrometer (Table 1).
Similar fragment ions have also been observed for the cross-
link products between T and G or A.5,6 In contrast to the product-
ion spectrum of d(mCG) (see the Supporting Information),
which shows a facile neutral loss of 5-methylcytosine, we did
not observe that loss in the product-ion spectrum of d(mC∧G).
The result is again in accord with that the 5-methylcytosine
is being cross-linked. ESI-MS/MS of the [M + H]+ ion of
d(G∧mC) gave similar results (Figure 2b).
9
J. AM. CHEM. SOC. VOL. 125, NO. 42, 2003 12797

A R T I C L E S
Zhang and Wang
Table 1. Results for Exact Mass Measurement of Fragment Ions
Observed in the Product-Ion Spectra of the [M + H]+ Ions of
d(mC∧G) and d(G∧mC)
ion
identities
calculated
mass
measured
mass
deviation
(ppm)
d(G∧mC)
551.1404
H2O loss
551.1410
551.1407
551.1408
471.1140
471.1134
471.1137
275.1000
275.1000
275.1000
1.1
0.5
0.7
2-deoxyribose loss
471.1142
275.1005
-0.4
-1.7
-1.1
-1.8
-1.8
-1.8
[nucleobase + H]+
d(mC∧G)
471.1142
2-deoxyribose loss
471.1137
471.1142
471.1145
275.1004
275.1002
275.1004
-1.1
0.0
0.6
Figure 4. Time-dependent study of 254-nm irradiation of d(GmCSPh), which
shows the %yield of d(G∧mC), the combined %yield of d(GHMC) and
d(GmC), and the decay of starting material d(GX) as a function of time.
The experiment was done by taking aliquots out at different irradiation times,
and the aliquots were dried and subjected to HPLC analysis. The yield was
based on the peak areas in the HPLC traces.
[nucleobase + H]+
275.1005
-0.4
-1.1
-0.4
of mC. In Figure 1, the fractions eluting at 34.5, 38.7, and 43.2
min are the dinucleoside monophosphates with the mC moiety
being oxidized to 5-(hydroxymethyl)cytosine [d(HMCG)], 5-me-
thylcytosine [d(mCG)], and 5-formylcytosine [d(FCG)], respec-
tively. Similarly, we observed the formation of d(GmC) and
d(GHMC) upon the 254-nm irradiation of d(GmCSPh). The
formation of d(HMCG), d(FCG), and d(GHMC) indicates that there
might be residual amount of oxygen present in the irradiation
solution.32 If that is the case and the amount of molecular oxygen
is not in large excess, we would expect to see that the formation
of these oxidation products and cross-link lesion both increase
with time at short irradiation time. However, these oxidation
products are expected to reach a plateau sooner than the cross-
link lesion. Our time-dependent irradiation experiment with
d(GmCSPh) (Figure 4) demonstrates that the yield of d(G∧mC)
and the overall yield of d(GHMC) and d(GmC) reach the plateau
at a similar time. The above results were obtained with
continuous argon bubbling during the 254-nm irradiation. We
did similar experiment except that the solution was not bubbled
with argon during the irradiation, and it turned out that the two
yields again reach the plateau at a similar time (data not shown).
Therefore, it remains unclear about the reason for the formation
of the oxidation products where the 5-methyl group is oxidized
to 5-formyl or 5-hydroxymethyl group.
By using the peak areas and assuming that the extinction
coefficients at 260 nm are similar for different products, we
estimated that the yield for the formation of d(mC∧G) is 21.4%,
which is lower than that for the formation of d(G∧mC) from
d(GmCSPh) (37.4%, HPLC trace shown in Figure 1b) under
similar irradiation conditions. This result is consistent with the
distances between the methyl carbon atom of mC and the C8
carbon atom of the adjacent G in these two dinucleoside
monophosphates, which, in standard B-DNA geometry, are 3.95
and 5.51 Å for adjacent GmC and mCG, respectively.16
Although a higher cross-linking yield was obtained when G is
the 5′ nucleobase, the difference between those two is not as
drastic as that reported for the similar cross-link lesions between
T and purines.6 The different observation with adjacent mCG/
Figure 3. 1H NMR spectra of d(mCG) (a), d(mC∧G) (b), and d(G∧mC)
(c). D2O was used as solvent, and impurity peaks were marked with a “X”.
1H NMR spectra of d(mC∧G) and d(G∧mC) further demon-
strate the presence of the cross-link lesion. Most strikingly, the
1H NMR spectra of both d(mC∧G) and d(G∧mC) showed the
absence of the H8 proton of guanine (Figure 3b&c) compared
to that of unmodified d(mCG) (Figure 3a). In addition, the
5-methylene protons in d(mC∧G) and d(G∧mC) occur as two
doublets [δ 3.86, 4.07 for d(mC∧G) and δ 4.03, 4.26 for
d(G∧mC)], whereas the methyl proton resonance shows as a
singlet in d(mCG). The NMR results are consistent with the
proposed structure and with previous results for the similar
cross-link lesion formed between T and G.5
In addition to the novel cross-link lesion, we observed the
formation of other products initiated from the 5-methyl radical
(32) Bienvenu, C.; Wagner, J. R.; Cadet, J. J. Am. Chem. Soc. 1996, 118,
11 406-11 411.
9
12798 J. AM. CHEM. SOC. VOL. 125, NO. 42, 2003

Generation of 5-(2′-Deoxycytidinyl)methyl Radical
A R T I C L E S
Figure 6. Product-ion spectra of the m/z 569 ion for the fractions with
retention times of 35.3 min (a) and 38.3 min (b) from the γ irradiation
mixture.
Conclusions
The 5-(2′-deoxycytidinyl)methyl radical can form from
5-methyl-2′-deoxycytidine upon either γ irradiation,15 during
which the hydroxyl radical can attack the 5-methyl carbon atom
of 5-methylcytosine via hydrogen abstraction, or one-electron
photooxidation with 2-methyl-1,4-naphthoquinone as a photo-
sensitizer.32 The resulting cation radical from the latter process
can deprotonate to give the 5-methyl radical.32 Here, we
demonstrate that the 5-(2′-deoxycytidinyl)methyl radical can be
specifically generated by 254-nm irradiation of 5-(phenyl-
thiomethyl)-2′-deoxycytidine in dinucleoside monophosphates
d(mCSPhG) and d(GmCSPh). Moreover, the resultant radical can
attack the C8 carbon atom of the adjacent guanine to yield a
novel cross-link lesion where the 5-methyl carbon atom of mC
and the C8 carbon atom of guanine are covalently bonded. In
addition, we demonstrated for the first time that the cross-link
lesion can form in dinucleoside monophosphate d(mCG) via γ
irradiation under anaerobic conditions.
Figure 5. LC-MS/MS analysis of 125-pmol authentic d(mC∧G) (a) and
40-nmol γ irradiation mixture of d(mCG) under anaerobic conditions (b).
The mass spectrometer was set to monitor the fragmentation of the ion of
m/z 569. Shown in the insets are the MS/MS for the ion of m/z 569.
GmC and TG/GT might be due to different sequences studied.
Here, we studied dinucleoside monophosphates, whereas 3-mer
and 15-mer oligodeoxynucleotides were investigated in the
previous study.6
3. Demonstration of the Formation of the Cross-Link
Lesion in d(mCG) Induced by γ Irradiation under Anaero-
bic Conditions. Next we confirmed, by LC-MS/MS experi-
ment, that γ irradiation of d(mCG) in the absence of oxygen
can also give rise to the cross-link lesion. In this experiment,
the MS/MS was set to monitor the fragmentation of the m/z
569 ion, which is the protonated ion of the cross-link lesion
d(mC∧G) (Vide supra), in the positive-ion mode. The total-ion
current of the γ irradiation mixture (Figure 5b) showed a peak
eluting at a very similar time as that of the standard cross-link
lesion d(mC∧G) (Figure 5a). Furthermore, MS/MS of the 18.5-
min peak for the γ irradiation sample gives identical fragmenta-
tion as that of the authentic standard (MS/MS are shown as
insets in the Figure 5). In addition to the cross-link lesions, we
observed other two fractions eluting at 35.3 and 38.3 min, which
are very likely the dinucleoside monophosphates with the 2′-
deoxyguanosine being oxidized to two stereoisomers of the 5′,8-
cyclo-2′-deoxyguanosine,33 which have the same molecular
weight as the d(mC∧G). Furthermore, MS/MS of the two cyclo
adducts showed that the most abundant fragment ions are due
to the loss of the unmodified mC and 5-methyl-2′-deoxycytidine
(Figure 6).
To our knowledge, this is the first report about the formation
of a cross-link lesion between 5-methylcytosine and guanine.
Next, we will quantify the formation of the novel cross-link
lesions in isolated and cellular DNA and investigate the
biological implications of this type of cross-link lesion.
Materials and Methods
All chemicals, unless otherwise specified, were from Sigma-Aldrich
(St. Louis, MO). The reagents for solid-phase DNA synthesis were
purchased from Glen Research (Sterling, VA). All other solvents, silica
gel, and TLC plates were obtained from EM Science (Gibbstown, NJ).
Mass Spectrometry. ESI-MS and MS/MS experiments were
carried out on an LCQ Deca XP ion-trap mass spectrometer (Ther-
moFinnigan, San Jose, CA). An equal-volume solvent mixture of
acetonitrile and water was used as solvent for electrospray, and a 2-µL
aliquot of a ∼5 µM sample solution was injected in each run. The
spray voltage was 3.4 kV, and the capillary temperature was maintained
at 200 °C.
(33) Dizdaroglu, M. Biochem J. 1986, 238, 247-254.
9
J. AM. CHEM. SOC. VOL. 125, NO. 42, 2003 12799

A R T I C L E S
Zhang and Wang
An IonSpec HiResESI external ion source FTICR mass spectrometer
(IonSpec Co., Lake Forest, CA) equipped with a 4.7 T unshielded
superconducting magnet was used for exact mass measurements. The
instrument was equipped with an Analytica (Branford, CT) electrospray
ion source. Samples were dissolved in CH3CN/H2O (1/1, v/v) solution
at a concentration of 5 µM and infused by a syringe pump at a flow
rate of 1 µL/min. The source end plate was kept at 3.5 and 3.0 kV for
experiments in positive- and negative-ion modes, respectively. Ions were
accumulated in a hexapole for 1000-1500 ms and transported through
a quadrupole ion guide to the analyzer cell by standard pulse sequences.
Gy/min for 3 h (total dose: ∼430 Gy). The resulting solution was dried
and subjected directly to LC-MS analysis.
A 2*250 mm Microsorb C18 column with a particle size of 5 µm
and a pore size of 100 Å (Varian, Walnut Creek, CA) was used for
separation, and a 50-min gradient of 0-28% acetonitrile in 10 mM
ammonium acetate was employed. The flow rate was 100 µL/min, and
the effluent was coupled to the LCQ Deca XP ion-trap mass
spectrometer without splitting. The mass spectrometer was set up to
monitor the fragmentation of the ion of m/z 569 in the positive-ion
mode. Approximately 40-nmol irradiation mixture was injected in each
run. As a control, standard d(mC∧G) (125 pmol) was also injected to
the column in a separate LC-MS/MS run with identical experimental
setup as for the γ-irradiation mixture.
Exact mass measurement of the molecular ions were done by using
the [M + 2H]2+ and [M + H]+ ions of a peptide PRCGVPDVA and
the [M - H]- and [M - Ade]- ions of ApA as internal standards in
the positive- and negative-ion modes, respectively. All spectra were
acquired in the broadband mode and, with the internal standards, the
mass accuracy for all mass measurements reported in this work was
within 5 ppm for the monoisotopic ion peaks.
For exact mass measurements in MS/MS, the [M + H]+ ions of the
analyte of interest and 2′-deoxyguanosine were first isolated in the ICR
cell by ejection of all other species. Sustained off-resonance irradiation/
collisionally activated dissociation (SORI-CAD) of the analyte ions
were performed for 1000 ms by a r.f. burst with an amplitude of 1.2 to
2.5 V depending on the analytes (i.e., at a frequency around 1000 Hz
from the cyclotron frequency of the isolated ions). Ions were excited
as a result of multiple collisions with pulsed nitrogen gas. A 400-ms
pulse of nitrogen was introduced at the beginning of the irradiation
period (the maximum pressure read-out on the ion gauge was 3 × 10-6
Torr). After a delay of approximately 2 s, the resulting fragment ions
were accelerated for detection by a r.f. sweep excitation waveform (100
V p-p). The image current was amplified, digitized at an acquisition
rate of 2 MHz with 2048 k data points, and Fourier transformed to
yield a mass spectrum.
Synthesis and Characterization of Compounds. Preparation of
2. We followed the procedures described by Ueda et al.34 with some
modifications. Paraformaldehyde (3.1 g) and 2′-deoxyuridine (1) (5.25
g, 23 mmol) were dissolved in 0.5 M triethylamine (TEA) in H2O (80
mL) and the solution was stirred at 60 °C. Additional paraformaldehyde
in two portions (4.5 g, 3.2 g) was added after 1 d and 2 d, immediately
after which TEA (1 mL) and H2O (10 mL) were also added to keep
the solution basic. After 4 d, the solution was concentrated. The residue
was coevaporated three times with benzene and once with pyridine,
and dried under vacuum overnight. The final residue was dissolved in
pyridine (60 mL), to which acetic anhydride (21 mL) was added. The
solution was stirred at room-temperature overnight. After adding a small
volume of methanol (5 mL), the solution was concentrated in Vacuo.
The residue was then partitioned between CHCl3 and H2O, and the
organic layer was washed sequentially with saturated NaHCO3, saturated
NaCl, and H2O. The dried residue was dissolved in DMF (10 mL), to
which PhSH (2.6 mL, 25.3 mmol) and TEA (5 mL) were added, and
the mixture was stirred overnight at 70 °C. The whole mixture was
partitioned between EtOAc and H2O, and the organic layer was dried
over anhydrous Na2SO4, and the solvent was removed in Vacuo. The
latter dried residue was dissolved in methanolic ammonia (50 mL) and
kept at room-temperature overnight. The solvent was then removed
and the residue was dissolved in CH3OH, to which silica gel was added.
The gel was dried and placed on top of a silica gel column, which was
eluted with CH3OH/CHCl3 (1/5, v/v). Appropriate fractions were pooled
and solvent evaporated to leave compound 2 as a white foam (4.08 g,
51%). ESI-MS: m/z 350.9 [M + H]+, 373.1 [M + Na]+; 349.1 [M -
NMR Measurements. All NMR spectra were recorded on a Varian
Unity Inova 300 MHz instrument (Palo Alto, CA). The residual proton
signal of the solvent serves as internal reference.
UVC Irradiation. An aqueous solution of dinucleoside monophos-
phate d(mCSPhG) or d(GmCSPh) with AU260 of 0.4 in a quartz tube was
degassed by argon bubbling for 30 min and irradiated for 15-20 min
with 254-nm light generated by a Rayonet photochemical reactor that
is equipped with 16 light tubes (The Southern New England Ultraviolet
Company, CT). The irradiation was done under continuous argon
bubbling. The resulting solution was then lyophilized to dryness.
1
H]-. H NMR (DMSO-d6): δ 11.47 (s, 1H, NH), 7.69 (s, 1H, H-6),
7.35 (m, 5H, Ar-H of PhS), 6.12 (t, JH1′-H2′ ) 6.2 Hz, JH1′-H2′′ ) 7.2
Hz, 1H, H-1′), 5.23 (d, JOH-H3′ ) 4.6 Hz, 1H, 3′-OH), 5.00 (t, JOH-H5′
) JOH-H5′′ ) 5.1 Hz, 1H, 5′-OH), 4.16 (m, 1H, H-3′), 3.82 (s, 2H,
CH2-SPh), 3.75 (m, 1H, H-4′), 3.51 (m, 2H, H-5′ and H-5”), 2.00 (m,
1H, H-2′), 1.82 (m, 1H, H-2′′).
HPLC. The HPLC separation was performed on a system composed
of a Hitachi L-6200A pump (Hitachi Ltd, Tokyo, Japan), a SSI 500
variable wavelength UV Detector (Scientific System Inc., State College,
PA), and a Peak Simple Chromatography Data System (SRI Instruments
Inc., Las Vegas, NV). A 4.6 × 250 mm Apollo C18 column (5 µm in
particle size and 300 Å in pore size, Alltech Associates Inc., Deerfield,
IL) was used for the separation. A solution of 50 mM TEAA (solution
A) and a mixture of 50 mM TEAA and acetonitrile (70/30, v/v) (solution
B) were used as the mobile phases. A gradient of 40-min 0-30%
solution B, 5-min 30-100% B, and 5-min 100-0% B was used for
the separation of the 254-nm irradiation products of d(mCSPhG) and
d(GmCSPh). Under this condition, d(GmC) and d(GHMC) were not well
resolved. Therefore, the fraction containing the two components was
further separated by using a gradient of 5-min 0-15% B, 15-min 15-
28% B, and 2-min 28-0% B, and the column was heated to 45 °C. In
all separations, the flow rate was 0.8 mL/min, and the fractions were
monitored at 260 nm. The HPLC condition for LC-MS experiments
was different and described separately in a latter section.
Preparation of 3. Compound 2 (2.26 g, 6.46 mmol) was dissolved
in dry pyridine (15 mL) and the resulting solution was evaporated to
dryness. The procedure was repeated twice. The resulting residue was
again dissolved in dry pyridine (40 mL), to which 4,4′-dimethoxytrityl
chloride (DMTrCl) (2.63 g, 7.76 mmol) and 4-(dimethylamino)pyridine
(DMAP) (40 mg, 0.33 mmol) were quickly added, followed by slow
addition of TEA (1.5 mL, 10.76 mmol). The reaction mixture was stirred
at room temperature for 3 h. After the reaction was complete as
monitored by TLC (95/5, CHCl3/CH3OH, v/v), the solution was cooled
to 5 °C, and a small volume of methanol (5 mL) was added. After 10
min, the mixture was evaporated to dryness in Vacuo. The resulting
residue was purified by silica gel chromatography with a step gradient
of methanol (0-5%) in CHCl3/TEA (99/1, v/v). Appropriate fractions
were pooled and concentrated to dryness to give compound 3 as a white
foam (2.91 g, 69%). ESI-MS: m/z 652.7 [M + H]+, 675.1 [M +
γ-Irradiation of d(mCG) and LC-MS Identification of d(mC∧G)
from the Irradiation Mixture. A 3-mL solution of d(mCG) (300 nmol)
in a 50-mL round-bottom flask was degassed by three freeze-pump-
thaw cycles, filled with argon, and exposed to a Mark I 137Cs Irradiator
(JL Shepherd and Associates, San Fernando, CA) at a dose rate of 2.4
1
Na]+; m/z 651.2 [M - H]-. H NMR (DMSO-d6): δ 11.53 (s, 1H,
NH), 7.44-6.82 (m, 19H, H-6 and aromatic H), 6.10 (t, JH1′-H2′ ) 6.2
(34) Suzuki, Y.; Matsuda, A.; Ueda, T. Chem. Pharm. Bull. 1987, 35, 1085-
1092.
9
12800 J. AM. CHEM. SOC. VOL. 125, NO. 42, 2003

Generation of 5-(2′-Deoxycytidinyl)methyl Radical
A R T I C L E S
Hz, JH1′-H2′′ ) 6.7 Hz, 1H, H-1′), 5.35 (d, JOH-H3′ ) 4.1 Hz, 1H, 3′-
OH), 4.14 (m, 1H, H-3′), 3.86 (m, 1H, H-4′), 3.74 (s, 6H, O-CH3),
3.63 (d, JHa-Hb ) 13.3 Hz, 1H, CH2a-SPh), 3.50 (d, JHb-Ha ) 13.3 Hz,
1H, CH2b-SPh), 3.14 (m, 2H, H-5′ and H-5′′), 2.10 (m, 1H, H-2′), 1.96
(m, 1H, H-2′′).
evaporated off under reduced pressure to give a yellow oil (compound
6). ESI-MS: m/z 1178.1 [M + Na]+.
Detritylation was done by dissolving 6 in 1% TFA solution in CH2-
Cl2. The solution was stirred for 30 min, and the solvent was removed
by evaporation under reduced pressure. The residue was purified by
silica gel chromatography using a step gradient of CH3OH (0-10%)
in CH2Cl2. Appropriate fractions were combined and solvent was
evaporated off under vacuum to give a glass (detritylation product).
ESI-MS: m/z 854.0 [M + H]+, 876.1 [M + Na]+.
Preparation of 4. Compound 3 (1.26 g, 1.93 mmol) was dissolved
in dry CH2Cl2 (15 mL) and the resulting solution was evaporated to
dryness. The operation was repeated twice. The resulting residue was
dissolved in dry CH2Cl2 (14 mL) and kept under an argon atmosphere.
Dry diisopropylethylamine (DIEA) (0.84 mL, 4.83 mmol) was then
added, followed by dropwise addition of 2-cyanoethyl-N,N-diisopro-
pylchlorophosphoramidite (0.65 mL, 2.91 mmol). The reaction mixture
was stirred for 15 min at room temperature, then another portion of
DIEA (0.84 mL) was added, and the reaction mixture was stirred for
another 15 min. TLC (toluene/EtOAc, 1/1, v/v) check found two
diastereomers (Rf ) 0.65 and 0.54). Workup was done by cooling the
reaction mixture in an ice bath followed by addition of CH3OH (3 mL).
The solution was then quickly extracted with EtOAc (70 mL). The
EtOAc layer was washed twice with saturated NaHCO3 (30 mL) and
once with saturated NaCl (30 mL). The organic layer was dried with
anhydrous Na2SO4, and solvent was evaporated off to give 4 in white
foam. ESI-MS: m/z 852.9 [M + H]+, 875.3 [M + Na]+.
Preparation of 5. Compound 5 was synthesized from compound 4
by nucleoside conversion using Reese Reagent 29,30,35. 1,2,4-triazole
(3.4 g, 49 mmol) was suspended in dry acetonitrile (70 mL) which
was cooled in an ice bath, then POCl3 (0.8 mL, 8.75 mmol) was slowly
added with rapid stirring. TEA (9 mL, 64.5 mmol) was then added
dropwise and the suspension was stirred for 30 min. Compound 4 was
dissolved in dry acetonitrile (20 mL) and added over 20 min to the
solution, and the solution was continuously stirred for another 90 min.
The reaction was stopped by adding saturated NaHCO3 solution (100
mL) and extracted with CH2Cl2 (200 mL). The organic layer was
washed sequentially with saturated NaHCO3 (50 mL) and brine (50
mL), dried with Na2SO4, and evaporated under reduced pressure to a
small volume. The resulting residue was loaded onto a silica gel column
and eluted with EtOAc:CH2Cl2:TEA (7:5:1, v/v), and appropriate
fractions were pooled and solvent evaporated under reduced pressure
to give a white foam (1.1 g, overall yield for the above two steps 62%).
ESI-MS: m/z 904.1 [M + H]+, m/z 926.3 [M + Na]+. 1H NMR
(DMSO-d6) (Figure S1): δ 9.36 (s, 1H, H-triazolyl), 8.41 (s, 1H,
H-triazolyl), 8.14 (s, 1H, H-6), 7.40-6.83 (m, 18H, Ar-H of DMTr
and PhS), 6.03 (t, JH1′-H2′ ) 5.6 Hz, JH1′-H2′′ ) 6.2 Hz, 1H, H-1′), 4.44
(m, 1H, H-3′), 4.39 (d, JHa-Hb ) 13.3 Hz, 1H, CH2a-SPh), 4.16 (m,
1H, H-4′), 3.92 (d, JHb-Ha ) 13.3 Hz, 1H, CH2b-SPh), 3.73 (s, 6H,
O-CH3), 3.68-3.51 (m, 4H, CH2-OP and CH-iPr), 3.41-3.24 (m,
2H, H-5′), 2.67 (t, JCH2-CH2CN ) 5.6 Hz, 2H, CH2CN), 2.45-2.18 (m,
2H, H-2′ and H-2′′), 1.22-1.09 (m, 12H, CH3-iPr). 31P NMR (DMSO-
d6): δ 151.61, 152.15.
After detritylation, the compound was further deprotected under a
mild condition using aqueous ammonium hydroxide (29%) for 60 h at
room-temperature instead of using standard base deprotection conditions
(29% ammonium hydroxide, 55 °C, 12 h). Under the mild condition,
no degradation of the phenylthio moiety was observed. The final product
was obtained by lyophilization. Overall yield (steps e-f in Scheme 1)
of compound 7 (153 mg, 37%). HRMS (ESI-FTICR) calcd 679.1700
1
[M + H]+, found 679.1683. H NMR (D2O) (Figure S2): δ 8.06 (s,
1H), 7.30 (m, 5H), 7.00 (s, 1H), 6.29 (t, J ) 6.7 Hz, 1H), 6.00 (t, J )
6.6, 7.2 Hz, 1H), 4.83 (m, 1H), 4.48 (m, 1H), 4.22 (m, 1H), 4.09 (m,
2H), 3.98 (m, 1H), 3.83 (s, 2H), 3.46 (d, J ) 4.1 Hz, 2H), 2.54 (m,
2H), 2.25 (m, 2H).
Preparation of 8 (Scheme 2). Compound 3 (1.58 g, 2.42 mmol)
was dried twice by coevaporation with dry pyridine and dissolved in
dry pyridine (40 mL). After stirring at room temperature for 10 min,
acetic anhydride (0.5 mL, 5.29 mmol) was added. Reaction was kept
at room temperature and stirred for overnight. The workup was done
by evaporating the solvent off under reduced pressure, and the resulting
residue was coevaporated with toluene (20 mL) for three times. After
drying under vacuum, the product was obtained as a white/light yellow
foam almost quantitatively. ESI-MS: m/z 717.1 [M + Na]+; 693.1
[M - H]-.
Preparation of 9. Compound 9 was synthesized from compound 8
by a nucleoside conversion procedure using Reese Reagent.30 1,2,4-
triazole (3.2 g, 46 mmol) was suspended in dry acetonitrile (75 mL) in
an ice bath, to which POCl3 (0.9 mL) was slowly added with rapid
stirring. Then TEA (7.5 mL) was added slowly and the suspension
was stirred for another 30 min. Compound 8 (1.67 g, 2.41 mmol) in
dry acetonitrile (10 mL) was added slowly and the solution was stirred
for 80 min. The reaction was terminated by pouring the mixture into
cold saturated NaHCO3 solution (80 mL), and the resulting solution
was extracted with CH2Cl2 (250 mL). The organic layer was washed
with brine (100 mL) and dried with anhydrous Na2SO4. The solvent
was evaporated off under reduced pressure to afford 9 in a white foam
(1.62 g, 90%). ESI-MS: m/z 768.1 [M + Na]+, 784.13 [M + K]+.
1H NMR (DMSO-d6): δ 9.40 (s, 1H, H-triazolyl), 8.43 (s, 1H,
H-triazolyl), 8.12 (s, 1H, H-6), 7.40-6.85 (m, 18H, Ar-H of DMTr
and PhS), 6.06 (t, JH1′-H2′ ) 6.2 Hz, JH1′-H2′′ ) 6.7 Hz, 1H, H-1′); 5.21
(m, 1H, H-3′), 4.43 (d, JHa-Hb ) 13.3 Hz, 1H, CH2a-SPh), 4.20 (m,
1H, H-4′), 4.05 (d, JHb-Ha ) 13.3 Hz, 1H, CH2b-SPh), 3.71 (s, 6H,
O-CH3), 3.24 (m, 2H, H-5′ and H-5′′), 2.28 (m, 2H, H-2′ and H-2′′),
2.05 (s, 3H, C(O)CH3).
Synthesis of d(mCSPhG) (7, Scheme 1). Dinucleoside monophos-
phates were synthesized using the procedures described by Cadet and
co-workers.5 N2-isobutyryl-2′-deoxyguanosine was synthesized follow-
ing a transient protection procedure developed by Ti and co-workers.36
Compound 5 (505 mg, 0.61 mmol) was dissolved in dry CH3CN (27
mL), then N2-isobutyryl-2′-deoxyguanosine (308 mg, 0.91 mmol) in 6
mL DMF was added followed by 0.45 M tetrazole in CH3CN (6.7 mL).
The resulting solution was stirred at room temperature for 25 min, and
the volume of the reaction mixture was reduced to 15 mL. A 0.1-M
iodine solution in THF/H2O/pyridine (78/20/2, v/v) was added until
the dark brown color did not dissipate, and the reaction mixture was
stirred for another 50 min. The reaction was then quenched by addition
of 1 M Na2S2O3 (5 mL) and CHCl3 (50 mL). The organic layer was
washed with water, dried with anhydrous Na2SO4, and the solvent was
Preparation of 10. Compound 9 (1.52 g, 2.04 mmol) was detrity-
lated with 80% acetic acid for 3 h and neutralized with saturated
NaHCO3 solution. The solution was then extracted with CHCl3, and
the organic layer was dried with anhydrous Na2SO4. The solvent was
evaporated under reduced pressure, and the residue was purified on
silica gel column using 5% CH3OH in EtOAc. Appropriate fractions
were combined and solvent evaporated under reduced pressure.
Compound 10 was obtained as a white foam (0.82 g, 91%). ESI-MS:
m/z 443.9 [M + H]+, 466.1 [M + Na]+, 482.1 [M + K]+; 442.1 [M -
1
H]-. H NMR (DMSO-d6) (Figure S3): δ 9.39 (s, 1H, H-triazolyl),
8.45 (s, 1H, H-triazolyl), 8.43 (s, 1H, H-6), 7.29 (m, 5H, Ar-H of
PhS), 6.08 (t, JH1′-H2′ ) 7.7 Hz, JH1′-H2′′ ) 6.2 Hz, 1H, H-1′), 5.27 (t,
JOH-H5′ ) 5.1 Hz, JOH-H5′′ ) 4.6 Hz, 1H, 5′-OH), 5.20 (m, 1H, H-3′),
4.54 (d, JHa-Hb ) 13.8 Hz, 1H, CH2a-SPh), 4.46 (d, JHb-Ha ) 13.8 Hz,
(35) Xu, Y. Z.; Zheng, Q.; Swann, P. F. J. Org. Chem. 1992, 57, 3839-3845.
(36) Ti, G. S.; Gaffney, B. L.; Jones, R. A. J. Am. Chem. Soc. 1982, 104, 1316-
1319.
9
J. AM. CHEM. SOC. VOL. 125, NO. 42, 2003 12801

A R T I C L E S
Zhang and Wang
1H, CH2b-SPh), 4.15 (m, 1H, H-4′), 3.63 (m, 2H, H-5′ and H-5′′), 2.46
(m, 1H, H-2′), 2.02 (m, 1H, H-2′′), 2.09 (s, 3H, C(O)CH3).
Preparation of d(GmCSPh) (12, Scheme 2). Compound 11 was
prepared using the same procedure as for the preparation of compound
6, which was obtained as a foam. ESI-MS: m/z 1198.0 [M + H]+,
1220.3 [M + Na]+; 1195.9 [M - H]-.
Acknowledgment. The authors thank Prof. Lawrence C.
Sowers at Loma Linda University for using a DNA synthesizer
in his laboratory. This work was supported in part by the
National Institute of Health.
Compound 11 was detritylated with 80% acetic acid for 3 h at room
temperature, then deprotected in a similar way to give 12 as white
powder (95 mg, overall yield for steps d-e in Scheme 2: 53%). HRMS
Supporting Information Available: 1H NMR spectra of
synthetic compounds, and H NMR and mass spectrometric
characterization of the 254-nm irradiation products of d(mCSPhG)
and d(GmCSPh). This material is available free of charge via
the Internet at http://pubs.acs.org.
1
1
(ESI-FTICR) calcd 679.1700 [M + H]+, found 679.1682. H NMR
(D2O) (Figure S4): δ 7.97 (s, 1H), 7.36 (m, 5H), 7.19 (s, 1H), 6.19 (t,
J ) 6.2 Hz, 1H), 6.10 (t, J ) 5.1, 6.7 Hz, 1H), 4.89 (m, 1H), 4.27 (m,
1H), 4.17 (m, 2H), 4.02 (m, 2H), 3.78 (d, JHa-Hb ) 14.3 Hz, 1H), 3.66
(d, JHb-Ha ) 14.3 Hz, 1H), 2.74 (m, 2H), 2.25 (m, 1H), 1.75 (m, 1H).
JA034866R
9
12802 J. AM. CHEM. SOC. VOL. 125, NO. 42, 2003