unmodified natural DNA, it could modify the lysine residues
of DNA-interacting proteins10 andcouldbeusedtomaplysine
residues in DNA-binding proteins as a part of structureꢀ
function investigations. The 1,4-dicarbonyl structure in the
ring-opened form 10, which exists in equilibrium with 1,
might be a key to effective amine modification.11 However,
the ODN containing 1 undergoes β-elimination from the
ring-opened form 10 under basic and heating conditions,
and the resultant cleavage of the binding sequence might
lead to a decrease in the affinity for the target molecules, and
thus a decrease in the modification efficiency.12 Reaction of
the ODN containing 1 with amine-containing molecules
yields molecules bearing lactam but does not result in
conjugation, which is also useful to introduce various func-
tions into DNA13 and to study DNA interacting molecules.
We designed an ODN containing 3, the 2,2-difluorinated
analogue of 1 (Figure 1). The introduction of the difluoro
groups was intended to prevent DNA strand cleavage by
β-elimination. In addition, an increase in the reactivity of the
aldehyde in the ring-opened form 30 was expected owing to
the electron-withdrawing effect of the fluoride. Herein, we
report the synthesis of ODN containing 3 that can carry out
amine modifications to yield ODN products under condi-
tions that are similar to those of biological systems.
The preparation of ODN containing 3 was carried out
by photoirradiating a caged ODN containing 4, which
was similar to generating the ODN containing 1 from its
corresponding caged precursor ODN.9 In this study, ODN
5 was prepared by photoirradiating 6. The synthesis of
caged sugar 4 was carried out using the 4,5-unsaturated
sugar 14 as a key intermediate (Scheme 2). The reaction of
2-deoxy-2,2-difluororibose14 with o-nitrobenzyl bromide
in the presence of DBU resulted in nitrobenzyl ether 9
(30%) and its anomer (40%). After the hydrolysis of
benzoyl groups of 9, 10 was obtained, and its 5-hydroxyl
group was iodinated to yield 11. The attempted conversion
of 11 into 14 by treatment with DBU only led to the
decomposition of 11. On the other hand, the 3-acetylated
12 was successfully converted into 13 by a similar treat-
ment. After hydrolysis of the 3-acetyl group, treatment of
the obtained 14 with m-CPBA in the presence of methanol
resulted in the introduction of a methoxy group at the
4-position, from which 4 was obtained as a major product
along with its C4-epimer. Compound 4 was converted into
phosphoramidite 15, and ODN 6 was synthesized using
the conventional phosphoramidite method by employing
an automated DNA synthesizer: MALDI-TOF MS data:
m/z = 4051 (MW of 6 = 4050). It was anticipated that
the photoirradiation of 6 yielded the ring-opened form 50,
which existed in equilibrium with 5. ODN 5 might be
composed of an equilibrating mixture of isomers at the
anomeric and C4-positions.15
The photoirradiation of 6 was carried out using a 10 μM
solution of 6 in H2O at 365 nm and 0 °C. After 4 h, HPLC
analysis of the reaction mixture revealed the clear conver-
sion of 6 into a new product (13 min; Figure 2). The result
of the MALDI-TOF MS of the photolyzed sample indi-
cated the formation of 5 (m/z = 3904: MW of 5 = 3902).
The yield of 5 was estimated (ca. 80%) on the basis of
a comparison of the area of its peak with that of the peak
of the internal standard. For the amine modification of a
target molecule, 3 must be accessible to the amino group in
the complex. In order to evaluate the reactivity of 3 with a
proximal amine, ODN 5 was subjected to a reaction with
the complementary ODN 7,16 which carried a hexylamine
that could approach 3 in the duplex (Figure 1). ODN 6
(25.4μM in50mMphosphatebuffer; pH valuesof7 and 8)
was hybridizedwithODN 7 (10μM;half the concentration
of 5 after photoirradiation). The decaging of 6 in the
duplex yielded duplex 5:7. The resultant duplex was in-
cubated at 37 °C in a phosphate buffer and was analyzed
Figure 1. ODNs 5 and 6 containing 3 and caged sugar 4 and
amines 7 and 8.
(9) Usui, K.; Aso, M.; Fukuda, M.; Suemune, H. J. Org. Chem. 2008,
8, 241–248.
(10) Jacobs, A. C.; Kreller, C. R.; Greenberg, M. M. Biochemistry
2011, 50, 136–143.
(11) (a) Op de Beeck, M.; Madder, A. J. Am. Chem. Soc. 2011, 133,
796–807. (b) Op de Beeck, M.; Madder, A. J. Am. Chem. Soc. 2012, 134,
10737–10740.
(12) Formation of DNA interstrand cross-link involving cleavage
of the strand containing the C40-OAS was reported by Greenberg.
Fragments (15-mers) formed by cleavage might still have enough affinity
to the complementary 31-mer for cross-link formation. Sczepanski, J. T.;
Jacobs, A. C.; Greenberg, M. M. J. Am. Chem. Soc. 2008, 130, 9646–
9647. Sczepanski, J. T.; Jacobs, A. C.; Majumdar, A.; Greenberg, M. M.
J. Am. Chem. Soc. 2009, 131, 11132–11139.
(14) Chou, T. S.; Heath, P. C.; Patterson, L. E.; Poteet, L. M.; Lakin,
R. E.; Hunt, A. H. Synthesis 1992, 565–570.
(15) The structure of the C40-oxidized abasic site in the duplex was
studied by Stubbe et al.: Chen, J.; Dupradeau, F.-Y.; Case, D. A.;
Turner, C. J.; Stubbe, J. Biochemistry 2007, 46, 3096–3107.
(16) Manoharan, M.; Inamati, G.; Tivel, K. L.; Conklin, B.; Ross,
B. S.; Cook, P. D. Nucleosides Nucleotides 1997, 16, 1141–1143.
(13) Lucas, R.; V.-Climent, E.; G.-Pinto, I.; Avino, A.; Eritja, R.;
Gonzalez, C.; Morales, J. C. Chem. Commun. 2012, 48, 2991–2993.
Org. Lett., Vol. 14, No. 23, 2012
5853