Novel benzopyrimidines as widened analogues of DNA bases

Article abstract of DOI:10.1021/jo0483973

(Chemical Equation Presented). We report on the synthesis, stacking, and pairing properties of a new structural class of size-expanded pyrimidine nucleosides, abbreviated dyT and dyC. Their bases are benzo-homologated variants of thymine and cytosine and

Full text of DOI:10.1021/jo0483973

Novel Benzopyrimidines as Widened Analogues of DNA Bases
Alex H. F. Lee and Eric T. Kool*
Department of Chemistry, Stanford University, Stanford, California 94305
kool@stanford.edu
Received September 11, 2004
We report on the synthesis, stacking, and pairing properties of a new structural class of
size-expanded pyrimidine nucleosides, abbreviated dyT and dyC. Their bases are benzo-homologated
variants of thymine and cytosine and have a design that is distinct from a previously described
class of size-expanded (xDNA) pyrimidines, with a different vector of expansion relative to the
sugar. We term this new base geometry “yDNA” (a mnemonic for “wide DNA”). Both C-glycosides
were prepared using Pd-mediated coupling of iodinated base derivatives with a deoxyribose
precursor. As free deoxynucleosides, both dyT and dyC displayed robust fluorescence, with emission
maxima at 375 and 390 nm, respectively. Both widened pyrimidines could be incorporated readily
as protected phosphoramidite derivatives into synthetic oligonucleotides. Experiments in “dangling
end” DNA contexts revealed that both yT and yC stack more favorably than their natural
counterparts. When opposite natural bases in the context of Watson-Crick DNA were paired, the
yT nucleotide formed a pair with A that was equally stable as a T-A pair, despite the mismatch
in size with the neighboring natural pairs. The yC nucleotide (paired opposite G) was destabilizing
by a small amount in the same context. Despite the large size of the pairs, both yT and yC were
selective for their Watson-Crick complementary partners A and G, respectively. The pairing
properties and fluorescence of yDNA nucleotides may lead to useful applications in the study of
steric effects in DNA-protein interactions. In addition, the compounds may serve as building blocks
for a large-sized artificial genetic system.
Introduction
biophysical studies of DNA and proteins that interact
with it. In addition, size-expanded bases can be paired
with natural complements to generate base pairs that
are also increased in size.1 Early studies of oligomeric
strands composed of sized-expanded nucleotides have
shown that they can bind complementary nucleic acids
with high affinity. Thus, such DNA homologues might
find useful applications in new biotechnologies.
A number of laboratories have examined properties of
nonnatural nucleosides in which the nucleobases are
larger than natural ones. One advantage of increased size
is extended conjugation, which can lead to useful fluo-
rescence. For example, after early work by Robins,3
Moreau and co-workers have constructed bicyclic and
tricyclic nucleobase analogues in an effort to yield useful
fluorescent reporters of triple helix formation in DNA.4
Because these benzo- and naphthopyrimidines are at-
tached as N-nucleosides, they would not form expanded-
We have recently undertaken a series of studies aimed
at design of new nucleotide analogues in which the bases
are larger than those in natural nucleic acids.1 This
approach was inspired by Leonard’s early studies with a
benzo-fused analogue of adenine ribonucleotide, which
was useful as a tool in studies of ATP-dependent en-
zymes.2 A larger set of size-expanded bases and nucle-
otides may serve more generally as useful tools in
* To whom corresponce should be addressed. Phone: (650) 724-4741.
Fax: (650) 725-0259.
(1) (a) Liu, H.; Gao, J.; Lynch, S. R.; Saito, Y. D.; Maynard, L.; Kool,
E. T. Science 2003, 302, 868-871. (b) Liu, H.; Gao, J.; Maynard, L.;
Saito, Y. D.; Kool, E. T. J. Am. Chem. Soc. 2004, 126, 1102-1109. (c)
Liu, H.; Lynch, S. R.; Kool, E. T. J. Am. Chem. Soc. 2004, 126, 6900-
6905. (d) Gao, J.; Liu, H.; Kool, E. T. J. Am. Chem. Soc. 2004, 126, in
press. (e) Liu, H.; Gao, J.; Kool, E. T. Submitted for publication.
(2) (a) Scopes, D. I. C.; Barrio, J. R.; Leonard, N. J. Science 1977,
195, 296-298. (b) Leonard, N. J. Acc. Chem. Res. 1982, 15, 128-135.
(c) Lessor, R. A.; Gibson, K. J.; Leonard, N. J. Biochemistry 1984, 23,
3868-3873.
(3) Stout, M. G.; Robins, R. K. J. Org. Chem. 1968, 33, 1219-1225.
10.1021/jo0483973 CCC: $30.25 © 2005 American Chemical Society
Published on Web 11/25/2004
132
J. Org. Chem. 2005, 70, 132-140

Benzopyrimidines as Widened DNA Bases
size pairs. Several related heterobicyclic compounds
reported by Pfleiderer and by Hawkins are notable for
their useful fluorescence properties,5,6 although they are
analogues of purines, and so are not appreciably ex-
panded in size.
In addition to the above examples, Seley has recently
reported a thienyl-bridged tricyclic analogue of deoxy-
adenosine in an effort to explore enzyme inhibition
properties.7 The compound has not been incorporated into
oligonucleotides. Finally, Matsuda has recently reported
the synthesis of tricyclic bases capable of forming four
hydrogen bonds and studied pairing properties in oligo-
nucleotides.8 They were designed to pair with other
nonnatural bases and display no pairing selectivity
among the natural bases.
Our initial approach to the design of size-expanded
purines and pyrimidines was the linear extension of
length by insertion of a benzene ring (analogous to the
Leonard approach2), yielding 2.4 Å of added size. This
yielded the xDNA nucleosides (dxT and dxA have been
reported to date) and helices containing them.1 Because
of their larger sizes, the xDNA nucleosides are fluores-
cent, and experiments showed that they stack with high
affinity on neighboring DNA sequences.1c Single substi-
tutions of xDNA base pairs in natural DNAs were
somewhat destabilizing, due to the mismatch in size with
the neighboring DNA. However, when strands were
designed to form helices in which every base pair was
expanded, the resulting duplexes were exceptionally
stable.1a,e
Studies of the properties of the xDNA nucleosides are
continuing. However, examination of models suggested
to us that other geometries might be equally viable for
extension of bases and pairs. An adenine-homologous
nucleoside having a new “yDNA” geometric expansion
was described recently.9 Here we describe the design,
preparation, and properties of a new class of widened
pyrimidine analogues, also having the new yDNA clas-
sification. These compounds are shown to exhibit a
number of potentially useful properties, including strong
stacking affinity in DNA, selective pairing with Watson-
Crick complements, and robust fluorescence. Similarities
and differences with previous expanded pyrimidine ana-
logues are discussed.
the extension vector orientation is altered by 60°. Simple
models of the expected base pairs (Figures 1C and 2B)
shows that the overall effects of these two extensions on
base pair length are similar but are predicted to yield
different groove depths. More specifically, the stacking
geometry with neighboring bases is altered, and yDNA
pairs are expected to project further into the major
groove, while the previous xDNA pairs project further
into the minor groove. Such simple models suggested that
the yDNA-widened pyrimidine design merited testing
both as the free nucleosides or nucleotides and in DNA
helices. A very recent study reported a widened form of
a purine base (yA), which is also benzo-fused and has a
similar vector of extension as yT and yC.9 It remained to
be seen whether pyrimidine-like compounds would be-
have with similarly favorable properties as did the purine
analogue yA.
It may be noted that, formally, the base yT as prepared
here lacks a methyl group and thus is technically a closer
analogue of uracil than thymine. Similarly, the yC base,
with its methyl group, is a close analogue of 5-methyl-C.
The presence or absence of methyl was expected to have
little influence on biophysical properties;11 we chose the
different methylation states to evaluate the two different
synthetic routes, which later were shown to be nearly
equally efficient (see below).
Space-filling models (Figure 1B) illustrate the close
relationships between natural T/C nucleobases and the
yT/yC analogues, which are enlarged by addition of the
benzo group. Electrostatic potentials mapped as colors
on the surfaces are extremely similar (T vs yT and C vs
yC) along the Watson-Crick edges, suggesting similar
hydrogen-bonding ability. However, the benzo groups are
significantly less polar than the rest of the molecules.
Synthesis. The preparation of the yT nucleoside
started with iodoanthranilic acid (3) and proceeded to the
free dyT nucleoside in nine steps and in generally high
yield (Scheme 1). The critical C-glycoside bond formation
was achieved in 50% yield under Heck conditions, giving
the desired â anomer exclusively. The anomeric geometry
was confirmed by proton NOE data. The free nucleoside
was thus obtained in eight steps and 20% yield. Two
subsequent steps following standard conditions were
needed to prepare the 5′-O-dimethoxytrityl, 3′-O-phos-
phoramidite derivative as required for standard oligo-
nucleotide synthesis (see the details in the Supporting
Information). Overall yield for the total of 10 steps was
18%. Based on precedent with a different isomer of this
nucleoside derivative,1a,b we opted not to protect the
oxygens for DNA synthesis; later oligomer synthesis
added support to this decision (see below).
Results
Design. The design of the current widened pyrimidines
(yT and yC, Figure 1) is different from previous extended
pyrimidines (xT and xC)1,10 by their vector of extension
(Figure 2). The length of extension of the bases in yT and
yC is essentially the same as xT and xC, at 2.4 Å, but
The synthesis of the dyC nucleoside followed a similar
strategy, making use of Heck chemistry to install the
glycosidic bond (Scheme 2). The synthesis in this case
began with 5-methylanthranilic acid (13), and the ben-
zopyrimidine heterocycle was protected at the exocyclic
amine prior to glycoside bond formation. This did not add
steps to the overall approach, since the same protecting
group was needed for oligonucleotide synthesis. In gen-
eral, as for dyT, the dyC synthesis proceeded with high
yield and the latter required fewer steps overall as
(4) (a) Michel, J.; Gueguen, G.; Vercauteren, J.; Moreau, S. Tetra-
hedron 1997, 53, 8457-8478. (b) Godde, F.; Toulme, J. J.; Moreau, S.
Biochemistry 1998, 37, 13765-13775. (c) Godde, F.; Toulme, J. J.;
Moreau, S. Nucleic Acids Res. 2000, 28, 2977-2985.
(5) (a) Driscoll, S. L.; Hawkins, M. E.; Balis, F. M.; Pfleiderer, W.;
Laws, W. R. Biophys. J. 1997, 73, 3277-3286;.(b) Lehbauer, J.;
Pfleiderer, W. Helv. Chim. Acta 2001, 84, 2330-2342.
(6) Hawkins, M. E.; Balis, F. M. Nucleic Acids Res. 2004, 32, e62.
(7) Seley, K. L.; Januszczyk, P.; Hagos, A.; Zhang, L.; Dransfield,
D. T. J. Med. Chem. 2000, 43, 4877-4883.
(8) Minakawa, N.; Kojima, N.; Hikishima, S.; Sasaki, T.; Kiyosue,
A.; Atsumi, N.; Ueno, Y.; Matsuda, A. J. Am. Chem. Soc. 2003, 125,
9970-9982.
(9) Lu, H.; He, K.; Kool, E. T. Angew. Chem., Int. Ed. 2004, in press.
(10) Liu, H.; Gao, J.; Kool, E. T. Submitted for publication.
(11) Wang, S.; Xu, Y.; Kool, E. T. BioMed. Chem. 1997, 5, 1043-
1050.
J. Org. Chem, Vol. 70, No. 1, 2005 133

Lee and Kool
FIGURE 1. Structures of widened nucleobases and nucleosides in this study. (A) Nucleosides dyT and dyT shown next to the
natural nucleosides dT and dC. (B) Space-filling models of the yT and yC nucleobases with electrostatic potentials mapped on the
surfaces. Natural T and C bases are shown for comparison (methyl groups are omitted for clarity). Calculated using AM1 with
Spartan (Wavefunction, Inc.). (C) The designed structures of widened base pairs involving yT and yC, with comparison to natural
analogous pairs.
compared to the former. Glycoside bond formation for yC
was accomplished in 59% yield through the use of Heck
chemistry, and if desired, the free nucleoside could be
obtained from amino-protected intermediate 19, thus
yielding this dyC nucleoside 2 in seven steps and 26%
yield overall. The entire synthesis of the protected
phosphoramidite derivative proceeded in 22% yield (eight
steps) from starting compound 13. Details are given in
the Experimental Section and in the Supporting Infor-
mation.
Properties. Because other isomers of benzopyrim-
idines are fluorescent,1b,d,4 we tested the fluorescence of
the dyT and dyC free nucleosides. Excitation and emis-
sion spectra were recorded for the two in methanol
(Figure 3). Emission maxima for dyT and dyC were 375
and 390 nm, respectively, which are nearly the same as
the structurally related dxT and dxC analogues.1d Ab-
sorption maxima for dyT were at 221 nm (ꢀ ) 38 100)
and a long-wavelength band at 315 nm (ꢀ ) 3040); the
corresponding maxima for and dyC were 231 nm (ꢀ )
33 900) and 312 nm (ꢀ ) 2780). Florescence quantum
yields for the two nucleosides were measured to be 0.54
(dyT) and 0.40 (dyC) in water. Thus, both are relatively
efficient violet-colored fluorophores.
We then tested the ability of the ydT and dyC nucle-
otide precursors to be incorporated into DNA by synthesis
of short (trimeric) oligomers, followed by characterization.
Proton NMR spectra of the trimers T-yT-T and T-yC-T
in crude form after deprotection showed clean and intact
incorporation of the new nucleosides. The two oligomers
were also characterized by electrospray mass spectra,
which also confirmed the expected masses (Supporting
Information).
Size-widened nucleobase analogues such as yT and yC
may be useful as fluorescent probes of biophysical
interactions and enzymatic mechanisms involving DNA.
We therefore characterized the basic stacking and pairing
properties of these new size-expanded nucleoside deriva-
tives in the context of natural DNA duplexes. Table 1
shows dangling-end thermal denaturation data,12 mea-
suring stacking tendencies of the compounds as compared
with standard-sized T and C, and Table 2 lists pairing
data, with each of the two new benzopyrimidines sepa-
rately placed opposite natural bases in the center of a
12mer DNA duplex. Results of the stacking studies
showed that both yT and yC in a 5′-dangling position of
the self-complementary duplex formed by the sequence
5′-dCGCGCG stabilized the helices quite considerably
134 J. Org. Chem., Vol. 70, No. 1, 2005

Benzopyrimidines as Widened DNA Bases
mol, which is smaller than the selectivity shown by
natural T in the control experiments (Supporting Infor-
mation). The yC analogue displayed somewhat different
performance in these contexts (Table 2). The yC substitu-
tion, when paired opposite G, was mildly destabilizing,
at approximately 0.6-2.2 kcal/mol relative to C. How-
ever, the pairing selectivity was similar to that seen for
yT; the data for yC mismatches showed a penalty of 1.0-
4.1 kcal/mol relative to the yC-G pair. This is smaller
than the pairing selectivity displayed by natural C in the
control duplexes.
The pairing thermodynamic data established that the
yT-A and yC-G pairs were not strongly destabilizing
and were selective. Although these facts suggested a
hydrogen-bonded geometry in the duplex (see Figure 1C),
confirmation of this must await future structural studies.
Nevertheless, to obtain more information about possible
structural distortions we used CD spectroscopy to evalu-
ate 12mer duplexes containing these pairs. The data are
shown in the Supporting Information. The results showed
that spectra of duplexes containing a central yT-A pair
closely resembled those containing a native T-A pair
(Figure S1, Supporting Information). Similarly, the CD
spectra of duplexes containing yC-G or C-G pairs were
also virtually the same. Thus, the preliminary data
suggest that although the widened pairs must by neces-
sity cause some localized backbone distortion due to the
added size, the overall duplex conformation is not sig-
nificantly distorted.
FIGURE 2. Comparison of the designs of yDNA and previous
xDNA nucleobase designs. (A) Skeletons of yT and xT, showing
the vectors of extension (arrows) by insertion of a benzo-ring.
(B) Illustration comparing the framework of expanded-size
T-A base pairs involving xT (top) and yT (bottom). Dashed
lines connect the glycosidic points of attachment. Note that
total degree of extension is similar; the chief difference is the
shapes of the grooves (the major groove is above the pairs and
the minor groove is below).
over the core duplex lacking a dangling base, suggesting
stacking on the neighboring base pairs that is consider-
ably more favorable in duplex form than in the single-
stranded state. Both yT and yC gave greater stabilization
than their natural smaller counterparts; however, the
advantage of yT over T was larger than that of yC over
C, which showed only a small difference. A comparison
with data for earlier xDNA variants of these widened
compounds (Table 1) shows that the previous expanded
analogues xT and xC stabilize the duplex more dramatic-
ally,1d suggesting more efficient stacking.
In the pairing studies we tested each of the two
widened pyrimidines in both a purine context (5′-
AGAAXGAAAAG) and a pyrimidine context (5′-CTTTT-
CYTTCTT), where X and Y are widened or natural bases.
Data are listed in Table 2, and examples of melting
profiles are shown in Figure 4; additional data are
available in the Supporting Information. In general, the
results were similar regardless of the difference in the
two contexts. The thermal denaturation data showed
(Table 2) that, when paired opposite A, yT showed equal
stability as the control, T paired with A. The margin of
base pairing selectivity of yT for A (as compared to
“mismatches” of yT with C, G, or T) was 0.9-3.1 kcal/
Discussion
The syntheses of the current benzopyrimidine nucleo-
side analogues are in general relatively short and ef-
ficient. One may compare the ease of preparation of dyT
and dyC with the syntheses of the previously described
dxT and dxC nucleoside analogues,1b,10 which are es-
sentially isomers of the present compounds. The new
synthetic data show that the earlier isomers were pre-
pared in similar yields. The free dxT nucleoside was
prepared in eight steps (34% yield), while the dxC
nucleoside was prepared in a longer route starting from
the dxT nucleoside (11 steps overall, 19% total). Thus,
the new dyT nucleoside is synthesized in the same
number of steps with slightly lower overall yield as dxT,
while the dyC nucleoside is prepared in fewer steps and
higher yield than the previous dxC.
A comparison of stacking and pairing properties of dyT
and dyC, which are pyrimidine analogues, with the
recently reported first member of the yDNA class, dyA
(a purine analogue),9 shows that dyA, a larger, tricyclic
compound, stacks considerably more strongly. This is not
surprising, as the size of a nucleobase or base analogue
generally correlates with stacking ability in DNA.12d In
addition, it is consistent with the fact that natural
adenine stacks with higher affinity than thymine and
cytosine do. As for pairing properties of dyA in natural
DNA, the earlier data showed that yA is destabilizing
when paired opposite T to a similar degree as dyC is
here.9 The selectivity of yA for T was also similar in
magnitude to the selectivities of yT and yC, which were
measured in the same sequence context here. This
parallel behavior is not unexpected, as the vector of
widening (by benzo-homologation) is almost the same for
yA as for the new pyrimidines.
(12) (a) Turner, D. H.; Sugimoto, N.; Freier, S. M. Annu. Rev.
Biophys. Biophys. Chem. 1988, 17, 167-192. (b) SantaLucia, J.; Allawi,
H. T.; Seneviratne, P. A. Biochemistry 1996, 35, 3555-3562. (c)
Guckian, K.; Schweitzer, B. A.; Ren, R. X.-F.; Sheils, C. J.; Paris, P.
L.; Tahmassebi, D. C.; Kool, E. T. J. Am. Chem. Soc. 1996, 118, 8182-
8183. (d) Guckian, K. M.; Schweitzer, B. A.; Ren, R. X.-F.; Sheils, C.
J.; Tahmassebi, D. C.; Kool, E. T. J. Am. Chem. Soc. 2000, 122, 2213-
2222. (e) Bommarito, S.; Peyret, N.; SantaLucia, J. Nucleic Acid Res.
2000, 28, 1929-1934. (f) Nakano, S.; Uotani, Y.; Nakashima, S.; Anno,
Y.; Fujii, M.; Sugimoto, N. J. Am. Chem. Soc. 2003, 125, 8086-8087.
(g) Lai, J. S.; Qu, J.; Kool, E. T. Angew. Chem., Int. Ed. 2003, 42, 5973-
5977.
J. Org. Chem, Vol. 70, No. 1, 2005 135

Lee and Kool
SCHEME 1a
a Reagents and conditions: (a) urea, 160 °C, 96%; (b) POCl3, DEA, reflux, 75%; (c) 0.5 M NaOMe/MeOH, reflux, 91%; (d) 1,2-dehydro-
3-O-(tert-butyldiphenylsilyl)-5-hydroxymethylfuran, Pd(OAc)2, Ph3P, Bu3N, CH3CN, 85 °C, 50%; (e) (i) 1.0 M TBAF, THF, cat. HOAc, 0
°C, (ii) NaB(OAc)3H, CH3CN-HOAc, 4:1 v/v, -15 °C, 85%; (f) Ac2O, pyridine, 25 °C, 87%; (g) NaI, HOAc, 60 °C, 83%; (h) NH3, MeOH, 25
°C, 12 h, 99%; (i) DMTrCl, DIPEA, pyridine, 25 °C, 95%; (j) N,N-Diisopropylammonium tetrazole, 2-cyanoethyl tetraisopropylphosphor-
amidite, CH2Cl2, 25 °C, 2 h, 94%.
A comparison of the pairing behavior of the yT/yC
family in DNA with the previous xT/xC isomers (which
have different geometries of glycosidic attachment) shows,
in general, quite similar properties.1d,10 All of these
expanded/widened bases retain the possibility of hydro-
gen bonding that is analogous to that of Watson-Crick
pairs. Because of their size expansions, however, the
bases are expected to be somewhat destabilizing when
surrounded by normally-sized DNA. The new data gener-
ally support this idea. When paired with adenine, the
previous xT was slightly destabilizing, while the current
data show that substitution of yT for T is energetically
neutral. As for the cytosine analogues, yC in the present
study is destabilizing, while the earlier xC was less
destabilizing. In terms of pairing selectivity for Watson-
Crick partners, xT displayed lower selectivity than the
current data show for yT. A study of xC pairing selectivity
showed that it was somewhat higher than that seen for
yC here. Thus, overall, the differences between the yDNA
pyrimidines and the xDNA pyrimidines in the context of
natural DNA are small, with minor variations that are
likely due, at least in part, to differences in stacking
stability.
as tools in the study of size and steric effects in protein-
DNA recognition. If such compounds are to be used as
probes in a base-paired context, it will be useful in future
studies to examine their structures in DNA at high
resolution to test whether they are, in fact, paired as
expected (see Figure 1) and what backbone distortions
result from the mismatch in size with the neighboring
pairs. It is encouraging in this regard that the yDNA
bases are only weakly destabilizing at most and retain
selectivity in pairing. Thus, they likely represent a
minimal and localized perturbation to the overall DNA
structure (this is supported by the CD spectra as well)
and give little alteration to the stability as well.
Beyond the possible use of yT and yC nucleotides as
probes of biophysical interactions, yDNA nucleosides are
also useful building blocks for construction of a new,
artificial genetic system (Lee, A. H. F.; Kool, E. T.
Manuscript in preparation). The yDNA nucleotide ge-
ometry is distinct from that of the earlier xDNA design.1
It remains to be seen how helices fully constructed from
the yDNA design compare to those of the xDNA design.
In addition, work will be needed to explore whether other
properties of a functioning genetic system, such as
transcription or replication, can be exhibited by such
artificial information-encoding systems.
Because of their generally favorable pairing properties
and their fluorescence, yDNA nucleotides may be useful
136 J. Org. Chem., Vol. 70, No. 1, 2005

Benzopyrimidines as Widened DNA Bases
SCHEME 2a
a Reagents and conditions: (a) ICl, 6% HCl, Na2SO3, NaOH, 95%; (b) (i) Ethyl chloroformate, THF, reflux, (ii) PBr3, Et2O, reflux, 95%;
(c) S-methylisothiourea hemisulfate, CH3CN-H2O, Na2CO3, reflux, 90%; (d) isobutyric anhydride, reflux, 69%; (e) 1,2-dehydro-3-O-(tert-
butyldiphenylsilyl)-5-hydroxymethylfuran, Pd(dba)2, AsPh3, Bu3N, CH3CN, 85 °C, 59%; (f) 1.0 M TBAF, THF, cat. HOAc, 0 °C, (ii) NH4OH,
0 °C, (iii) NaB(OAc)3H, CH3CN-HOAc, 4:1 v/v, -15 °C, 84%; (g) DMTrCl, DIPEA, pyridine, 25 °C, 98%; (h) N,N-diisopropylammonium
tetrazole, 2-cyanoethyl tetraisopropylphosphoramidite, CH2Cl2, 25 °C, 82%.
NMR (CDCl3, 400 MHz): δ 8.63 (dd, 1H, J ) 2.4, 0.4 Hz), 8.23
(dd, 1H, 8.8, 2.0 Hz), 7.73 (d, 1H, J ) 8.4 Hz). 13C NMR (CDCl3,
100 MHz): δ 162.3, 155.4, 151.3, 144.9, 134.6, 129.3, 132.6,
94.6. HREIMS m/z: [M]+ calcd for C8H3N2Cl2I 323.8718, found
323.8706.
Experimental Section
6-Iodo-1H-quinazoline-2,4-dione (4). 2-Amino-5-iodoben-
zoic acid 3 (6.52 g, 24.79 mmol) and urea (15.00 g, 0.25 mol)
were heated to 150 °C in a round-bottom flask, and the melt
was stirred overnight. The mixture was then allowed to cool
to 100 °C, and a volume equivalent of water was added. The
resulting mixture was stirred for 10 min to allow unreacted
urea to dissolve. The solid was filtered, resuspended in 150
mL 0.5 M NaOH (aq), and then heated to boiling to allow
formation of the sodium salt. The mixture was neutralized
with 10% diluted hydrochloric acid. The precipitate was
suction-filtered, and the filter cake was washed with a water-
methanol mixture. The product 4 (6.85 g, 96%) was obtained
after vacuum-drying without further purification as a yellow
powder. 1H NMR (DMSO-d6, 400 MHz): δ 11.25 (s, 1H), 8.11
(s, 1H), 7.93 (dd, 1H, J ) 7.2, 2.0 Hz), 6.98 (d, 1H, J ) 8.4
Hz). 13C NMR (DMSO-d6, 100 MHz): δ 161.9, 150.6, 142.8,
141.1, 134.9, 118.2, 116.7, 84.8. HREIMS m/z: [M]+ calcd for
C8H5N2O2I 287.9397, found 287.9402.
6-Iodo-2,4-dimethoxyquinazoline (6). 6-Iodo-2,4-dichlo-
roquinazoline 5 (5.00 g, 15.43 mmol) was suspended in 0.5 M
sodium methoxide in methanol (77.0 mL). The reaction
mixture was heated at reflux under argon atmosphere for 18
h. The reaction mixture was then cooled to room temperature
and neutralized with 10% HCl (aq). The quenched mixture was
concentrated in vacuo. The resulting red solution was dissolved
in EtOAc (200 mL) and washed with water (2 × 70 mL). The
aqueous phase was washed with CH2Cl2 (50 mL) again, and
the combined organics were dried (Na2SO4) and concentrated
in vacuo. The orange residue was purified by silica column
chromatography (hexane/ EtOAc 20:1) to yield 6 (4.45 g, 91%)
as a white solid. 1H NMR (CDCl3, 400 MHz): δ 8.41 (d, 1H, J
) 2.0 Hz), 7.95 (dd, 1H, J ) 8.4, 2.0 Hz), 7.46 (d, 1H, J ) 8.8
Hz), 4.17 (s, 1H), 4.09 (s, 1H). 13C NMR (CDCl3, 100 MHz): δ
168.1, 162.3, 151.4, 142.4, 132.7, 127.9, 115.4, 87.6, 54.9, 54.7.
HREIMS m/z: [M]+ calcd for C10H9N2O2I 315.9709, found
315.9708.
6-Iodo-2,4-dichloroquinazoline (5). A mixture of iodo-
quinazolinedione 4 (7.64 g, 26.52 mmol) and phosphorus
oxychloride (30.0 mL, 323 mmol) was stirred in an oven-dried
round-bottom flask fitted with a condenser. To this solution
was added N,N-diethylaniline (8.8 mL, 68.97 mmol). The
mixture was heated at reflux for 4 h. Then the solution was
allowed to cool to room temperature, and volatiles were
removed in vacuo. The residue was dissolved in CH2Cl2 (150
mL) and washed with water (2 × 120 mL). The organic layer
was dried (Na2SO4) and concentrated in vacuo. The resulting
residue was purified by silica column chromatography (hexane/
2,4-Dimethoxy-6-{2′R-cis-3-[2′3′-dehydro-3′-(tert-butyl-
diphenylsilyloxy)]-5′-hydromethyl-2′-furanyl}quin-
azole (7). The dimethoxyquinazoline 6 (1.67 g, 5.28 mmol) and
1,2-dehydro-3-O-(tert-butyldiphenylsilyl)-5-hydroxymethylfu-
ran13 (1.44 g, 4.07 mmol) were charged in an oven-dried round-
bottom flask equipped with a septum. Freshly distilled CH3CN
(13) Zhang, H. C.; Daves, G. D. J. J. Org. Chem. 1992, 57, 4690-
4696.
1
EtOAc 15:1) to yield 3 (6.44 g, 75%) as an orange powder. H
J. Org. Chem, Vol. 70, No. 1, 2005 137

Lee and Kool
ylquinazoline was characterized by NMR. 1H NMR (CDCl3, 400
MHz): δ 8.10 (d, 1H, J ) 2.0 Hz), 7.80 (m, 2H), 5.33 (dd, 1H,
J ) 16.8, 5.6 Hz), 4.18 (s, 3H), 4.10 (s, 3H), 4.08 (m, 1H), 3.99
(m, 2H), 2.94 (m, 1H), 2.58 (m, 1H). 13C NMR (CDCl3, 100
MHz): δ 213.35, 169.19, 162.41, 152.34, 135.58, 131.87, 126.73,
121.40, 113.32, 82.48, 61.45, 54.81, 54.62, 45.27, 33.16. This
crude product was dissolved in a solution of CH3CN (12 mL)
and glacial acetic acid (3 mL). The resulting mixture was
chilled to -15 °C in an ice-salt bath followed by the addition
of sodium triacetoxyboronhydride (0.82 g, 3.88 mmol) in one
portion. The reaction mixture was allowed to stir at the low
temperature for 10 min and was then concentrated in vacuo.
The residue was purified by silica column chromatography
(hexane/EtOAc 2:1 initially and EtOAc/methanol 10:1 subse-
1
quently) to give 8 (1.01 g, 85%) as a yellow syrup. H NMR
(CDCl3, 400 MHz): δ 7.95 (s, 1H), 7.67 (s, 1H), 7.28 (s,1H),
5.25 (dd, 1H, J ) 10.0, 5.6 Hz), 4.44 (1H, m), 4.15 (m, 1H),
4.13 (s, 3H), 4.07 (s, 3H), 3.79 (m, 1H), 2.29 (m,1H), 2.04 (m,
1H). 13C NMR (CDCl3, 100 MHz): δ 169.1, 162.1, 151.8, 137.3,
132.0, 126.1, 120.8, 113.1, 87.4, 79.7, 73.4, 63.2, 54.7, 54.5, 43.6.
HRFABMS m/z: [M + H]+ calcd for C15H19N2O5 307.1294,
found 307.1300.
1′-â-[6-(2,4-Dimethoxyquinazoline)]-3′,5′-O-(diacetyl)-
2′-deoxy-d-ribofuranose (9). A mixture of 8 (0.41 g, 1.34
mmol), acetic anhydride (0.6 mL, 5.40 mmol), and pyridine (5
mL) was refluxed for 4 h. The mixture was dissolved in EtOAc
(30 mL), and the solution was extracted with saturated sodium
bicarbonate (3 × 50 mL), dried (Na2SO4), and concentrated in
vacuo. The residue was purified by silica column chromatog-
raphy (hexane/EtOAc 1:2) to afford 9 (0.46 g, 87%) as a yellow
1
oil. H NMR (CDCl3, 400 MHz): δ 8.02 (s, 1H), 7.69 (s, 1H),
7.26 (s, 1H), 5.21 (m, 2H), 4.42 (m, 1H), 4.28 (m, 2H), 4.25 (s,
3H), 4.06 (s, 3H), 2.37 (m, 1H), 2.09 (m, 7H). 13C NMR (CDCl3,
100 MHz): δ 170.7, 170.5, 169.1, 162.1, 152.0, 136.6, 131.7,
126.3, 120.8, 113.3, 82.7, 80.2, 76.7, 76.6, 64.3, 54.7, 54.5, 41.3,
21.0, 20.7. HRFABMS m/z: [M + H]+ calcd for C19H23N2O7
391.1505, found 391.1500.
FIGURE 3. Normalized fluorescence excitation and emission
spectra of free deoxynucleosides yT (A) and dyC (B). Spectra
were measured in methanol; excitation for emission spectra
was 315 nm; excitation spectra (dashed lines) were monitored
at 400 nm.
1′-â-[6-(Quinazoline-2,4-dione)]-3′,5′-O-(diacetyl)-2′-
deoxy-d-ribofuranose (10). To a solution of 9 (0.34 g, 0.86
mmol) in glacial acetic acid (10 mL) was added sodium iodide
(0.59 g, 4 mmol). The reaction mixture was heated to 60-65
°C for 45 min, and then the volatiles were removed in vacuo.
The residue was dissolved in EtOAc (50 mL) and extracted
with saturated Na2SO3 (aq) (3 × 30 mL) and saturated sodium
bicarbonate solution (2 × 20 mL). The aqueous layers were
extracted with EtOAc (2 × 30 mL). The combined organics
were dried (Na2SO4) and concentrated in vacuo. The residue
was purified by silica column chromatography (EtOAc) to yield
10 (0.26 g, 83%) as a white solid. 1H NMR (CD3OD, 400
MHz): δ 8.04 (s, 1H), 7.65 (m, 1H), 7.16 (d, 1H, J ) 8.4 Hz),
5.25 (m, 1H), 5.13 (dd, 1H, J ) 10.8, 4.8 Hz), 4.35 (m, 1H),
4.25 (m, 2H), 2.40 (m, 1H), 2.11 (s, 3H), 2.09 (s, 3H), 2.05 (m,
1H). 13C NMR (CD3OD, 100 MHz): δ 172.6, 172.3, 165.0, 152.5,
141.8, 137.5, 134.5, 125.9, 116.7, 115.79, 84.3, 81.3, 78.2, 65.5,
42.2, 20.9, 20.8. HRFABMS m/z: [M + H]+ calcd for C17H19N2O7
363.1192, found 363.1190.
1′-â-[6-(Quinazoline-2,4-dione)]-2′-deoxy-d-ribofuran-
ose (1). A solution of 10 (0.39 g, 1.1 mmol) in saturated NH3/
methanol (15 mL) was stirred at room temperature for 18 h.
The volatiles were removed in vacuo, and the residue was
purified by silica column chromatography (EtOAc/methanol
10:1) to yield 1 (0.33 g, 99%) as a slightly yellow solid. 1H NMR
(CD3OD, 400 MHz): δ 8.02 (s, 1H), 7.70 (m, 1H), 7.16 (d, 1H,
J ) 8.4 Hz), 5.15 (dd, 1H, J ) 10.8, 5.6 Hz), 4.33 (m, 1H), 3.96
(m, 1H), 3.69 (d, 2H, J ) 4.8 Hz), 2.23 (m, 1H), 1.94 (m, 1H).
13C NMR (CD3OD, 100 MHz): δ 165.2, 152.6, 141.6, 138.6,
134.7, 125.8, 116.6, 115.6, 89.4, 80.7, 74.4, 64.0, 44.9. HR-
FABMS m/z: [M + Na]+ calcd for C13H14N2O5Na 301.0800,
found 301.0795.
(50 mL) was added to the flask to form a suspension, and the
mixture was bubbled with argon for 30 min. Pd(OAc)2 (45 mg,
0.21 mmol) and PPh3 (105 mg, 0.41 mmol) were charged into
a separated round-bottom flask and suspended in freshly
distilled CH3CN (30 mL). The catalyst-ligand mixture was
stirred at room temperature and degassed with argon bubbling
before being transferred to the glycal-quinazoline suspension
via hypodermic syringe under inert atmosphere. Tri(n-butyl)-
amine (1.38 mL, 5.81 mmol) was added to the reaction mixture
in a portion. The septum-sealed flask was heated to 85-90 °C
for 24 h and allowed to cool to room temperature. After
removal of the volatiles, the residue was purified with silica
column chromatography (hexane/EtOAc 3:1 initially and 1:1
subsequently) to afford 7 (1.30 g 50%) as yellow foam. 1H NMR
(CDCl3, 400 MHz): δ 7.83 (m, 3H), 7.76 (m, 2H), 7.57 (d, 1H,
J ) 8.4 Hz), 7.43 (m, 7H), 5.67 (dd, 1H, J ) 16.8, 5.6 Hz), 4.79
(m, 1H), 4.37 (m, 1H), 4.15 (s, 3H), 4.07 (s, 3H), 3.88 (bs, 2H),
1.10 (s, 9H). 13C NMR (CDCl3, 100 MHz): δ 169.2, 162.2, 152.2,
151.1, 138.1, 135.9, 135.7, 135.3, 133.2, 133.0, 131.3, 131.0,
130.2, 127.9, 127.8, 126.6, 126.2, 122.5, 121.6, 113.1, 102.6,
84.5, 83.3, 62.7, 54.7, 54.5, 26.3, 19.2, 14.0. HRFABMS m/z:
[M + H]+ calcd for C31H35N2O5Si 543.2315, found 543.2302.
1′-â-[6-(2,4-Dimethoxyquinazoline)]-2′-deoxy-d-ribo-
furanose (8). A solution of compound 7 (0.84 g, 1.55 mmol)
in anhydrous THF (20 mL) was chilled in an ice-water bath.
Glacial acetic acid (0.9 mL) and 1 M TBAF solution in THF
(6.0 mL) were added sequentially via hypodermic syringe. The
reaction mixture was allowed to stir at 0 °C for 5 min, and
the volatiles were removed in vacuo. The crude intermediate
of 2,4-dimethoxy-6-(â-D-glycero-pentofuran-3′-ulos-1′yl)-6-meth-
Phosphoramidite Derivative of Nucleoside Analogue
1. Subsequent synthetic steps and characterization are given
in the Supporting Information.
138 J. Org. Chem., Vol. 70, No. 1, 2005

Benzopyrimidines as Widened DNA Bases
TABLE 1. Thermodynamic Data for Self-Complementary Duplexes Containing a 5′ Dangling Base in the Duplex Core
Sequence 5′-dCGCGCGa
dangling base
Tmb (°C)
∆G°37c (kcal/mol)
∆∆G°37d (kcal/mol)
∆H° c (kcal/mol)
∆S° c (eu)
(none)
T
44.8 ( 0.5
50.0 ( 0.5
47.8 ( 0.5
56.6 ( 0.5
53.4 ( 0.5
57.7 ( 0.5
53.6 ( 0.5
-8.5 ( 0.2
-9.2 ( 0.5
-9.3 ( 0.2
-10.9 ( 0.1
-9.7 ( 0.2
-11.2 ( 0.2
-10.1 ( 0.3
-43 ( 11
-51 ( 9
-52 ( 3
-57 ( 1
-45 ( 1
-113 ( 35
-135 ( 29
-137 ( 9
-150 ( 3
-114 ( 4
-0.4 ( 0.2
-0.4 ( 0.3
-0.7 ( 0.2
-0.6 ( 0.2
-0.9 ( 0.2
-0.8 ( 0.3
C
yT
yC
xTe
XCe
a Conditions: 1 M NaCI, 10 mM phosphate (pH 7.0) with 0.1 mM EDTA. b Tm values are for 5.0 µM oligonucleotide. c Values from
van't Hoff plots with at least five data points. d Values per each dangling base, relative to the core duplex. e Data from ref ld,e.
TABLE 2. Thermodynamic Data for 12mer DNA
Duplexes Containing a Widened Base Pair or Mismatch
(X-Y) in a Central Positiona,b
c
d
base pair
X-Y
Tm
∆G°37
∆H° d
(kcal/mol)
∆S° d
(eu)
(°C)
(kcal/mol)
A-T
40.4 ( 0.5
41.2 ( 0.5
41.5 ( 0.5
28.5 ( 0.5
31.8 ( 0.5
32.3 ( 0.5
25.4 ( 0.5
40.6 ( 0.5
31.9 ( 0.5
36.8 ( 0.5
36.4 ( 0.5
27.3 ( 0.5
43.0 ( 0.5
46.5 ( 0.5
40.6 ( 0.5
28.3 ( 0.5
36.0 ( 0.5
33.1 ( 0.5
26.0 ( 0.5
39.6 ( 0.5
26.5 ( 0.5
31.9 ( 0.5
28.2 ( 0.5
23.7 ( 0.5
-9.4 ( 0.1
-9.4 ( 0.1
-9.4 ( 0.2
-6.3 ( 0.1
-6.9 ( 0.2
-7.0 ( 0.2
-5.7 ( 0.6
-9.1 ( 0.2
-7.2 ( 0.3
-8.2 ( 0.1
-8.1 ( 0.2
-6.2 ( 0.3
-9.7 ( 0.1
-11.1 ( 0.4
-9.1 ( 0.1
-5.8 ( 0.4
-8.1 ( 0.1
-7.4 ( 0.1
-6.1 ( 0.1
-8.9 ( 0.1
-5.8 ( 0.2
-7.1 ( 0.1
-6.1 ( 0.2
-5.2 ( 0.6
-94 ( 12
-84 ( 5
-82 ( 6
-73 ( 3
-82 ( 5
-86 ( 9
-69 ( 16
-78 ( 5
-73 ( 9
-79 ( 2
-73 ( 6
-67 ( 6
-69 ( 6
-95 ( 6
-66 ( 3
-79 ( 7
-74 ( 6
-76 ( 4
-63 ( 4
-44 ( 14
-73 ( 2
-77 ( 2
-77 ( 5
-66 ( 18
274 ( 39
241 ( 15
234 ( 20
215 ( 12
244 ( 17
256 ( 32
207 ( 54
222 ( 20
214 ( 30
231 ( 8
211 ( 20
197 ( 21
192 ( 18
271 ( 18
185 ( 10
237 ( 26
212 ( 19
223 ( 14
185 ( 14
112 ( 44
215 ( 8
T-A
yT-A
yT-G
yT-C
yT-T
yT-φe
A-yT
G-yT
C-yT
T-yT
φ-T
C-G
G-C
yC-G
yC-A
yC-C
yC-T
yC-φ
G-yC
A-yC
C-yC
T-yC
φ-C
224 ( 6
229 ( 16
197 ( 59
a Conditions: 100 mM NaCI, 10 mM MgCl2, 10 mM PIPES-
Na (pH 7.0). b Sequence is 5′-AAGAAXGAAAAG‚5′-CTTTTCYT-
TCTT. c Tm values are for 5.0 µM oligonucleotide. d Averages of
values from van't Hoff and curve fitting methods. e “φ” is tetrahy-
drofuran abasic analogue.
2-Amino-5-iodo-3-methylbenzoic Acid (14). Iodine mono-
chloride (5.1 mL, 0.10 mol) was added to a solution of 2-amino-
3-methylbenzoic acid 13 (5.16 g, 34 mmol) and 6% aqueous
HCl (60 mL) at 50-60 °C. The mixture was stirred at that
temperature for 1 h and cooled to room temperature. Then
the mixture was poured onto sodium metabisulfate (19.0 g,
0.10 mol) in crushed ice and neutralized with potassium
hydroxide pellets cautiously with ice bath cooling. The pale
brown precipitate was filtered, washed with ice-water and
methanol (10 mL), and then dried in vacuo to give 14 (9.34 g
FIGURE 4. Examples of thermal denaturation curves for
12mer duplexes containing single widened bases paired op-
posite each of the four natural bases. (A) Plots for X-yT pairs,
where X ) A, T, G, or C. (B) Plots for X-yC pairs. Thermo-
dynamics are listed in Table 2. Conditions: pH 7.0, 100 mM
NaCl, 10 mM MgCl2, 10 mM Na‚PIPES buffer, [DNA] ) 5 µM;
monitoring at 260 nm.
1
95%) as a brown powder. H NMR (DMSO-d6, 400 MHz): δ
7.86 (d, 1H, J ) 2.0 Hz), 7.44 (d, 1H, J ) 2.4 Hz), 2.08 (s, 3H).
13C NMR (DMSO-d6, 100 MHz): δ 168.8, 149.4, 141.6, 136.8,
126.4, 111.7, 74.5, 17.1. HREIMS m/z: [M]+ calcd for C8H8-
NO2I 276.9600, found 276.9608.
6-Iodo-8-methyl-1H-benzo-1,3-oxazine-2,4-dione (15).
Intermediate 14 (8.10 g, 31.50 mmol) was dissolved in anhy-
drous THF (50 mL), to which was added ethyl chloroformate
(10.0 mL, 97.50 mmol) dropwise at room temperature. After
addition, the mixture was heated at reflux for 22 h under
argon. The mixture was allowed to cool to room temperature,
and the volatiles were removed in vacuo to yield a dark brown
suspension. The residue was redissolved in anhydrous diethyl
ether (50 mL), to which was added PBr3 (3.1 mL, 32.50 mmol)
cautiously. After being refluxed for 24 h under Ar, the mixture
was chilled in ice-water. The yellow precipitate formed was
filtered by suction filtration and washed with petroleum ether
and a solvent mixture of CH2Cl2-methanol (1:1). The filter
cake was dried in vacuo to give 15 (9.20 g, 95%) as a yellowish
brown powder. 1H NMR (DMSO-d6, 400 MHz): δ 11.06 (bs,
1H), 7.99 (s, 1H), 7.92 (s, 1H), 2.29 (s, 3H). 13C NMR (DMSO-
d6, 100 MHz): δ 158.9, 147.0, 145.4, 139.6, 134.2, 127.3, 112.5,
86.3, 16.7. HREIMS m/z: [M]+ calcd for C9H6NO3I 302.9392,
found 302.9399.
J. Org. Chem, Vol. 70, No. 1, 2005 139

Lee and Kool
2-Amino-6-iodo-8-methyl-1H-quinazolin-4-one (16). A
mixture of 15 (8.10 g 26.70 mmol), sodium carbonate (5.00 g,
40.00 mmol), and S-methylisothiourea hemisulfate (4.47 g,
32.00 mmol) was dissolved in a solvent mixture of CH3CN-
H2O (40 mL/10 mL). The mixture was then heated at reflux
for 2 h and cooled to room temperature. The brown precipitate
was filtered with suction filtration, and the filter cake was
washed with a solution of CH3CN-H2O (5:1, 20 mL). As the
vacuum-dried yellow solid was poorly soluble in various
solvents, the crude product 16 (7.25 g, 90%) was subjected to
(EtOAc/methanol 25:1) to afford 19 (0.54 g, 84%) as a white
1
powder. H NMR (CD3OD, 400 MHz): δ 7.85 (s, 1H), 7.52 (s,
1H), 5.12 (dd, 1H, J ) 10.4, 5.2 Hz), 4.34 (m, 1H), 3.97 (m,
1H), 3.71 (d, 2H, J ) 4.8 Hz), 2.72 (m, 1H), 2.38 (s, 3H), 2.23
(m, 1H), 1.96 (m, 1H), 1.23 (s, 6H). 13C NMR (CD3OD, 100
MHz): δ 182.0, 175.5, 139.8, 134.4, 122.1, 120.1, 89.3, 80.9,
74.4, 64.1, 44.8, 36.9, 20.9, 19.4, 19.4, 14.4, 13.9. HRFABMS
m/z: [M + H]+ calcd for C18H24N3O5 362.1716, found 362.1700.
1′-â-6-[2-Amino-(8-methyl-4-oxo-1,4-dihydroquinazolin-
2-yl)]-2′-deoxy-d-ribofuranose (2). A solution of compound
19 (0.25 g, 0.66 mmol) in 5 mL of ammonium hydroxide was
heated to 55 °C overnight. The volatiles were removed in
vacuo, and the residue was purified by washing with CH2Cl2
to afford pure free deoxyriboside dyC (2) (0.18 g, 95%) as a
white to slightly yellow solid. ESIMS m/z: [M + H]+ calcd for
C14H17N3O4 292.3, found 292.3.
1
the next step without further purification. H NMR (DMSO-
d6, 400 MHz): δ 7.98 (s, 1H), 7.70 (s, 1H), 6.63 (bs, 1H), 2.31
(s, 3H). HREIMS m/z: [M]+ calcd for C9H8N3OI 300.9712,
found 300.9722.
N-(6-Iodo-8-methyl-4-oxo-1,4-dihydroquinazolin-2-yl)-
isobutyramide (17). Isobutyric anhydride (7.0 mL, 42.21
mmol) was added to a solution of 16 (4.30 g, 14.30 mmol) in
distilled pyridine (15 mL), and the solution was stirred at room
temperature for 18 h. The resulting mixture was concentrated
in vacuo and purified by silica column chromatography (CH2-
Cl2/EtOAc 6:1 initially, 3:1 subsequently) to give 17 (3.68 g,
69%) a yellow powder. 1H NMR (DMSO-d6, 400 MHz): δ 8.07
(d, 1H, J ) 2.0 Hz), 7.86 (d, 1H, 2.0 Hz), 2.80 (m, 1H), 2.38 (s,
3H), 1.10 (d, 3H, J ) 6.8 Hz), 1.01 (d, 3H, J ) 6.8 Hz). 13C
NMR (DMSO-d6, 100 MHz): δ 180.6, 177.4, 146.6, 142.8, 136.8,
131.9, 121.4, 88.8, 34.7, 34.5, 18.91, 18.8, 16.9. HREIMS m/z:
[M]+ calcd for C13H14N3O2I 371.0130, found 371.0128.
{(2′R)-cis-3-[2′,3′-Dehydro-3′-(tert-butyldiphenylsilyl-
oxy)]-5′-hydroxymethyl-2′-furanyl}-6-[N-(8-methyl-4-oxo-
1,4-dihydroquinazolin-2-yl)]isobutyramide (18). Isobu-
tyramide 17 (2.24 g, 6.04 mmol) and 1,2-dehydro-3-O-(tert-
butyldiphenylsilyl)-5-hydroxymethylfuran 13 (2.56 g, 7.25 mmol)
were charged into an oven-dried round-bottom flask equipped
with a septum. Freshly distilled CH3CN (50 mL) was added
to the flask to form a suspension, and the mixture was bubbled
with argon for 30 min. Pd(dba)2 (580 mg, 0.64 mmol) and
AsPh3 (350 mg, 1.20 mmol) were charged into a separated
round-bottom flask and suspended in freshly distilled CH3-
CN (25 mL). The catalyst-ligand mixture was stirred at room
temperature and degassed with argon bubbling before being
transferred to the glycal-isobutyramide suspension via hy-
podermic syringe under inert atmosphere. Tri(n-butyl)amine
(1.8 mL, 7.85 mmol) was added to the reaction mixture in
portions. The septum-sealed flask was heated to 85-90 °C for
22 h and allowed to cool to room temperature. After the
removal of the volatiles, the residue was purified with silica
column chromatography (hexane/ EtOAc 2:1 initially and 1:1
subsequently) to afford 18 (1.30 g, 59%) as slightly yellow foam.
1H NMR (CDCl3, 400 MHz): δ 7.84 (m, 3H), 7.76 (m, 2H), 7.48
(m, 6H), 7.19 (s, 1H), 5.60 (dd, 1H, J ) 2.0, 1.6 Hz), 4.74 (s,
1H), 4.41 (s, 1H), 3.93 (s, 2H), 2.66 (m, 1H), 2.18 (s, 3 H), 1.29
(dd, 6H, J ) 6.8, 2.4 Hz), 1.11 (s, 9H). 13C NMR (DMSO-d6,
100 MHz): δ 178.6, 160.9, 151.2, 145.6, 137.8, 135.4, 134.6,
134.2, 131.4, 131.1, 130.4, 130.3, 128.0, 128.0, 123.2, 119.4,
101.9, 84.6, 83.4, 62.8, 36.7, 26.4, 21.1, 20.7, 19.3, 19.0, 14.2,
13.5. HRFABMS m/z: [M + H]+ calcd for C34H40N3O5Si
598.2737, found 598.2720.
Protected Phosphoramidite Derivative of Nucleoside
Analogue 2. Subsequent synthetic steps and characterization
are given in the Supporting Information.
Oligonucleotide Synthesis and Characterization. 5′-
O-Tritylated phosphoramidite derivatives of 1 and 2 were
prepared as described. Oligonucleotides were synthesized on
DNA synthesizers using standard â-cyanoethyl phosphor-
amidite chemistry. The coupling time for the modified nucleo-
side phosphoramidites was extended from 100 to 700 s to
increase the likelihood of high yields; shorter coupling times
were not tested. Stepwise coupling yields for the extended
bases were all greater than 98% as determined by trityl cation
response. All extended oligonucleotides were DMT-off on the
5′-end and deprotected from CPG (alkyl-controlled pore glass)
supports in concentrated ammonia overnight at 55 °C. They
were purified by preparative 20% denaturing polyacrylamide
gel electrophoresis (PAGE) and isolated by excision and elution
from gel. The recovered material was subsequently quantified
by absorption at 260 nm with molar extinction coefficients
determined by the nearest neighbor method. All synthesized
oligonucleotides were confirmed by MALDI mass spectroscopy
(5′-TyTT, 5′-TyCT were analyzed with ESI-MS). Data are given
in the Supporting Information.
Thermal Denaturation Studies. Solutions for the ther-
mal denaturation studies contained the indicated concentra-
tions of self-complementary or non-self-complementary strands
containing nucleoside analogues. The buffer contained NaCl/
MgCl2/Na‚PIPES; (100 mM/10 mM/10 mM) at pH 7.0. After
the solutions were prepared they were heated to 95 °C and
annealed in the UV-vis spectrophotometer equipped with a
thermoprogrammer at the rate of 1.0 °C /min. Melting studies
were carried out in Teflon-stoppered 1 cm path length quartz
cuvettes under a steady flow of nitrogen current to prevent
moisture condensation on quartz wall at low temperature.
Absorbance was monitored while the temperature was raised
from 5 to 95 °C or declined from 95 to 5 °C at a rate of 1 °C/
min. All sequences were monitored at 260 nm except the
marked ones. Computer fitting of the melting data with
MeltWin software generated melting temperature (Tm), free
energies (∆G), and other thermodynamic data for the duplexes.
The curve fit values were compared with those calculated from
van’t Hoff plots by plotting 1/Tm verse ln(CT/4) and matched
with good agreement.
1′-â-{6-[N-(8-Methyl-4-oxo-1,4-dihydroquinazolin-2-yl)]-
isobutyramide}-2′-deoxy-d-ribofuranose (19). A solution
of 18 (1.07 g, 1.79 mmol) in anhydrous THF (15 mL) was
chilled in an ice-water bath. Glacial acetic acid (0.6 mL) and
1 M TBAF solution in THF (2.8 mL, 2.69 mmol)) were added
sequentially via hypodermic syringe. The reaction mixture was
allowed to stir at 0 °C for 10 min, ammonium hydroxide (2
mL) was added to quench the reaction, and the volatiles were
removed in vacuo. The intermediate was dissolved in a solution
of CH3CN (15 mL) and glacial acetic acid (3 mL). The resulting
mixture was chilled to -15 °C in an ice-salt bath followed by
the addition of sodium triacetoxyborohydride (0.95 g, 4.48
mmol) in one portion. The reaction mixture was allowed to
stir at low temperature for 7 min and concentrated in vacuo.
The residue was purified by silica column chromatography
Optical Measurements. Solutions, instruments, and meth-
ods for fluorescence and CD spectra are given in the Support-
ing Information.
Acknowledgment. This work was supported by the
U.S. National Institutes of Health (GM63587). A.H.F.L.
was supported by The Croucher Foundation Fellowship.
Supporting Information Available: Experimental de-
tails of nucleoside analogue synthesis, oligonucleotide synthe-
sis and characterization data, and thermodynamics methods.
This material is available free of charge via the Internet at
http://pubs.acs.org.
JO0483973
140 J. Org. Chem., Vol. 70, No. 1, 2005