stability of triplexes. The results of the Tm experiments are
also shown in Table 1.
complementary HP 4, 5′-CAAAGAAGAAGAAGACTTT-
TGTCTTCTTCTTCTTTG. The significant decrease in the
Tm value could be explained in terms of the increasing
electronic repulsion between the protonated cytosine bases.
Similar phenomena have been observed in earlier studies by
Roberts8 and recent studies by James.9 In contrast, the Tm
values of TFOs 7 and 8 having s8A and o8A residues,
respectively, as the protonated cytosine base analogs in-
creased compared with the unmodified TFOs (entries 2 and
3). In particular, the Tm value of TFO 7 having four
consecutive s8A bases was significantly higher (by 27.1 °C)
than that of TFO 6. These results strongly support our
proposal that the hybridization ability of TFO by incorporat-
ing a sequence of consecutive s8A bases would be increased.
In summary, we synthesized the s8A phosphoramidite unit
6 and TFOs having s8A residues for the first time. The
selectivity of a TFO having an s8A residue was higher than
that of the unmodified TFO, whereas the hybridization ability
of a TFO having an s8A was lower than that of the
unmodified TFO. Moreover, it was also found that the
hybridization ability of TFO 7 having four consecutive s8A
bases was significantly higher than those of the corresponding
unmodified TFO 6 or TFO 8 containing four consecutive
o8A as a well-known modified base of a protonated cytosine.
These results indicate that it might be necessary to use the
cytosine analogs properly in sequences of discontinuous and
consecutive cytosine bases. In the case of TFOs containing
discontinuous cytosine bases, the use of 5-methylcytosine
accessible to the protonation5,6 of the cytosine ring in a TFO
under neutral conditions might be better because of its
stronger hydrogen bonding ability. On the other hand, s8A
should be used in place of cytosine to avoid electronic
repulsion and enhance the stacking interaction when a series
of cytosine bases are arranged in a consecutive sequence.
Further studies are now under way in this direction.
The Tm value of TFO 2 containing one s8A in entry 2 was
lower than that of unmodified TFO 1 in entry 1 (29.0 vs
30.5 °C), although the Tm value of TFO 2 containing one
s8A was higher than that of TFO 3 containing one o8A in
entry 3 (29.0 vs 28.2 °C). It was also found that the Tm value
of the TFO decreased further by the addition of a discontinu-
ous s8A as shown in entry 3 (27.4 vs 30.5 °C) though the
Tm values of TFO 3 obtained at various pHs showed the pH-
independence of TFO 3 in Figure S2 in Supporting Informa-
tion. In entry 5, the Tm value of TFO 5 containing two
discontinuous o8As was significantly lower (by 8.4 °C) than
that of unmodified TFO 1. Because the backbone structure
was somewhat disturbed by the presence of a bigger s8A
base in place of the protonated C, the distance between the
C1′ atoms in the neighboring mononucleotide units became
longer only at the modified bases so that the constant
structure could not be preserved around the modified sites.
Subsequently, we examined the selectivity of modified
DNA oligomers to study whether the s8A base can actually
form a third Hoogsteen-type base pair with the guanine base.
In the hairpin duplex (HP 2), we replaced the central C-G
base pair with a base pair of G-C that cannot form hydrogen
bonds with s8A or o8A, as shown in entries 6-11 of Table
1. We carried out the Tm experiments at pH 6.0 since the
triplexes formed between TFO 1 or 2 and HP 2 were very
unstable at pH 7.0. It should be noted that the selectivity of
s8A with ∆Tm of 30.8 °C (difference in the Tm values between
entry 8 and entry 9) was superior to that of the unmodified
cytosine with ∆Tm of 25.6 °C (difference in the Tm values
between entry 6 and entry 7). On the other hand, the o8A
base has a poorer selectivity with ∆Tm of 18.5 °C (difference
in the Tm values between entry 10 and entry 11). Because
the s8A base has a higher selectivity than the protonated C
base, it is likely that the former can form two hydrogen bonds
with G of G-C. The predominance of s8A over o8A could
be explained in terms of the higher contribution of the syn
form around the glycosyl bond of the former.
Acknowledgment. This study was supported by a grant
from CREST of Japan Science and Technology (JST) and a
Grant-in-Aid for Scientific Research from the Ministry of
Education, Culture, Sports, Science and Technology, Japan
and Research for Promoting Technological Seeds program
of JST. This study was also supported in part by a grant of
the Genome Network Project from the Ministry of Education,
Culture, Sports, Science and Technology, Japan and by the
global COE project.
Next, we synthesized oligodeoxyribonucleotides having
four s8A or o8A bases in a consecutive sequence and studied
their hybridization property to examine if our proposal is
correct. It was expected that when a consecutive sequence
of four s8A bases is incorporated into the DNA oligomer
TFO 7, the stacking effect of two s8A bases was enhanced,
as reported in the case of poly-2-thiouridylates.21 The Tm
values of the oligomers having four consecutive unmodified
and modified (o8A and s8A) bases are summarized in Table
2. The Tm value of TFO 6 with a consecutive sequence of
four cytosines to HP 3 was very low (Tm ) 9.7 °C) compared
with the Tm value (24.4 °C) of TFO 9 having four
discontinuous cytosine bases, 5′-TTTCTTCTTCTTCT, to the
Supporting Information Available: Experimental pro-
cedures and full spectroscopic data for all new compounds.
This material is available free of charge via the Internet at
OL802622S
608
Org. Lett., Vol. 11, No. 3, 2009