Synthesis of C-nucleosides designed to participate in triplex formation with native DNA: Specific recognition of an A:T base pair in DNA

Article abstract of DOI:10.1021/jo0511445

We have previously described a system of 2-aminoquinoline- and 2-aminoquinazoline-based C-deoxynucleosides (TRIPsides) that are designed to be incorporated into oligomers that can specifically bind in the major groove via Hoogsteen base pairing to any seq

Full text of DOI:10.1021/jo0511445

Synthesis of C-Nucleosides Designed To Participate in Triplex
Formation with Native DNA: Specific Recognition of an A:T Base
Pair in DNA
Jian-Sen Li† and Barry Gold*,†
Eppley Institute for Research in Cancer and Department of Pharmaceutical Science, University of
Nebraska Medical Center, 986805 Nebraska Medical Center, Omaha, Nebraska 68198-6805
goldbi@pitt.edu
Received June 7, 2005
We have previously described a system of 2-aminoquinoline- and 2-aminoquinazoline-based
C-deoxynucleosides (TRIPsides) that are designed to be incorporated into oligomers that can
specifically bind in the major groove via Hoogsteen base pairing to any sequence of native DNA.
The four TRIPsides are termed antiGC, antiCG, antiTA, and antiAT with respect to the Watson-
Crick base pair targets that they bind. The first three TRIPsides have been prepared, characterized,
and shown to form stable and sequence-specific triplexes. In the present study, we describe the
preparation of two molecules, 2-amino-4-(2′-deoxy-â-D-ribofuranosyl)quinazoline (7) and 2-amino-
6-fluoro-4-(2′-deoxy-â-D-ribofuranosyl)quinoline (14), that can serve as the remaining antiAT
TRIPside. The phosphoramidites of 7 and 14 were prepared, but only the latter was successfully
incorporated into DNA oligomers. It is demonstrated using UV-visible melting experiments that
14 forms sequence-specific intramolecular triplets with A:T base pairs at physiological pH.
Introduction
bases,7 2-amino-4-(2′-deoxy-â-D-ribofuranosyl)quinoline
(antiGC), 2-amino-5-(2′-deoxy-â-D-ribofuranosyl)quino-
line (antiCG), 2-amino-4-(2′-deoxy-â-D-ribofuranosyl)-
quinazoline (antiAT), and 2-amino-5-(2′-deoxy-â-D-ribo-
furanosyl)quinazoline (antiTA), which differentiate
between the four base-pairing schemes in the major
groove (i.e., G:C, C:G, A:T, T:A, respectively; Scheme 1).
The orientation of the triplex-forming oligomer relative
to the duplex is arbitrarily defined as running antipar-
allel to the left strand in the major groove that runs 5′
to 3′ top to bottom. Of the four C-glycoside bases, antiTA,7
antiGC,8 and antiCG9 have been synthesized by the
coupling of a protected ribofuranoid glycal with a halo-
There is intense interest in the design of molecules that
can bind sequence specifically via a triple helix motif to
mixed purine/pyrimidine sequences in native Watson-
Crick DNA.1-6 To achieve this goal, we proposed triplex-
forming oligomers composed of a set of four C-glycoside
† Current address: Department of Pharmaceutical Sciences, Uni-
versity of Pittsburgh, 512 Salk Hall, 3501 Terrace Street, Pittsburgh,
PA 15261.
(1) Griffin, L. C.; Dervan, P. B. Science 1989, 245, 967-970.
(2) Horne, D. A.; Dervan, P. B. J. Am. Chem. Soc. 1990, 112, 2435-
2437.
(3) Jayasena, S. D.; Johnston, B. H. Nucleic Acids Res. 1992, 20,
5279-5288.
(4) Gowers, D. M.; Fox, K. R. Nucleic Acids Res. 1999, 27, 1569-
1577.
(7) Li, J.-S.; Fan, Y.-H.; Zhang, Y.; Marky, L. A.; Gold, B. J. Am.
Chem. Soc. 2003, 125, 2084-2093.
(5) Buchini, S.; Leumann, C. J. Angew. Chem., Int. Ed. 2004, 43,
3925-3928.
(8) Li, J.-S.; Shikiya, R.; Marky, L. A.; Gold, B. Biochemistry 2004,
43, 1440-1448.
(6) Craynest, N. V.; Guianvarc’h, D.; Peyron, C.; Benhida, R.
Tetrahedron Lett. 2004, 45, 6243-6247.
(9) Li, J.-S.; Chen, F.-X.; Shikiya, R.; Marky, L. A.; Gold, B. J. Am.
Chem. Soc. 2005, 127, 12657-12665.
10.1021/jo0511445 CCC: $30.25 © 2005 American Chemical Society
Published on Web 09/10/2005
8764
J. Org. Chem. 2005, 70, 8764-8771

Specific Recognition of an A:T Base Pair in DNA
SCHEME 1. Structures of the Four TRIPsides
Associated with Their Four Potential
Watson-Crick Base Pair Partners
2a and 2b were then converted into their corresponding
acid chlorides (3a and 3b, respectively) by treatment with
an excess of SOCl2 in DMF and Et2O at refluxing
temperature. 3a and 3b underwent a Stille coupling with
N-(tert-butoxycarbonyl)-2-(trimethylstannyl)aniline17 to
afford 4a and 4b, respectively. Removal of the protecting
Boc group from 4a and 4b gave 5a and 5b, respectively,
in high yield. The key step in the synthesis was the
elaboration of the 2-aminoquinazoline heterocycle through
a low-temperature (50 °C) fusion reaction18 of the HCl
salts of 5a and 5b with an excess of cyanamide to yield
the O-protected C-glycosides 6a and 6b, respectively.
Deprotection of 6a and 6b using NaOMe in MeOH at
room temperature (RT) afforded the desired 7 in 70%
yield.
NOESY-NMR experiments on 7 in 10 mM Na-
phosphate buffer (pH 7.0) containing 50 mM NaCl in D2O
confirmed that it adopts the expected anti conformation
(data not shown). This conformational preference is
critical for 7 to differentiate A:T from T:A base pairs.7,8
The UV-vis spectrum of 7 shows absorption peaks at
230 and 352 nm. The absorption peak at 352 nm is typical
for 2-aminoquinazoline heterocycles.7 As expected, 7
showed a characteristic fluorescence emission peak at 430
nm when excited at 350 nm.7,8
Attempts To Incorporate antiAT (7) into Oligo-
mers. The amino group in 7 was selectively protected
by benzoylation to give 8 using standard procedures.7 It
is worth noting that attempts to protect this amino group
with isobutyric anhydride or Fmoc-Cl resulted in the
decomposition of the starting material and a complex
mixture. Using standard methods, we sequentially con-
verted compound 8 into the 5′-DMTr-protected 9 and the
3′-phosphoramidite derivative 10 (Scheme 2).
genated heterocycle using a Pd-mediated Heck-type
reaction.10-13 However, antiAT could not be synthesized
by this method, presumably due to the electron-rich
aminoquinazoline ring system. A literature survey indi-
cated that antiTA was the first example of quinazoline-
based C-glycoside.7 Therefore, it was necessary to develop
a new method to prepare quinazoline C-glycosides.
Herein, we report the synthesis of the last of the four
originally proposed C-glycosides, antiAT (7). Although 7
was prepared, it proved unsuitable for solid-phase syn-
thesis using standard phosphoramidite chemistry. To
mimic the H-bonding scheme of the antiAT‚A:T triplet
(Scheme 1), a 2-amino-6-fluoroquinoline-based C-glyco-
side (antiAT-F, 14) was synthesized and efficiently
incorporated in oligomers employing standard conditions.
Characterization of antiAT-F shows that it forms stable
intramolecular triplexes at physiological pH and with
excellent specificity; that is, mismatches are highly
destabilizing.
It was determined that treatment of 8 in concentrated
NH4OH at 55 °C for 8 h, a standard method for the
deprotection of oligomeric deoxynucleotides, resulted in
partial decomposition of the desired product, 7. However,
it was found that 8 could be quantitatively converted into
7 in concentrated NH4OH at 23 °C for 24 h, which
provided an acceptable post-DNA synthesis deprotection
procedure to remove the N2-benzoyl group. Preliminary
tests indicated that the dimethoxytrityl group in 5′-
DMTr-protected 9 could not be removed by 2% DCA in
CH2Cl2 without affecting the 2-aminoquinazoline hetero-
cycle. This was confirmed in an oligomer synthesis, and
therefore standard solid-phase synthesis, which employs
2% DCA in CH2Cl2 as the detritylation reagent to
generate the free 5′-hydroxy group on the growing
oligomer, is not suitable for preparing oligomers contain-
ing 7. Since Lewis acids can catalyze detritylation, we
Results and Discussion
Synthesis of 7 (antiAT). The known starting materi-
als, O-protected 2′-deoxyribonic-1-â-cyanide 1a14 and
1b,15,16 were hydrolyzed to the corresponding acids 2a and
2b in a refluxing mixture of HOAc and HCl (Scheme 2).
20
successfully used a saturated ZnBr2 solution in CH3NO2
to remove the dimethoxytrityl group in 5′-DMTr-protect-
ed 9 without affecting the 2-aminoquinazoline hetero-
cycle. However, attempts to incorporate a saturated
ZnBr2 solution of CH3NO2 into the solid-phase synthesis
of oligomers composed of 7 proved unsatisfactory because
(10) Cheng, J. C.-Y.; Hacksell, U.; Davies, G. D., Jr. J. Org. Chem.
1985, 50, 2778-2780.
(11) Davies, G. D., Jr. J. Org. Chem. 1992, 57, 4690-4696.
(12) Farr, R. N.; Kwok, D.-I.; Davies, G. D., Jr. J. Org. Chem. 1992,
57, 2093-2100.
(13) Farr, R. N.; Davies, G. D., Jr. Carbohydr. Chem. 1990, 9, 653-
660.
(17) Salituro, F. G.; McDonald, I. A. J. Org. Chem. 1988, 53, 6138-
6139.
(14) Giese, B.; Gilges, S.; Groninger, K. S.; Lamberth, C.; Witzel, T.
Liebigs Ann. Chem. 1988, 615-617.
(18) Fujiwara, N.; Fujita, H.; Iwai, K.; Kurimoto, A.; Murata, S.;
Kawakami, H. Bioorg. Med. Chem. Lett. 2000, 10, 1317-1320.
(19) Theiling, L. F.; McKee, R. L. J. Am. Chem. Soc. 1952, 74, 1834-
1836.
(15) Jazouli, M.; Guianvarc’h, D.; Soufiaoui, M.; Bougrin, K.; Vier-
ling, P.; Benhida, R. Tetrahedron Lett. 2003, 44, 5807-5810.
(16) Hashmi, S. A. N.; Hu, X.; Immoos, C. E.; Lee, S. J.; Grinstaff,
M. W. Org. Lett. 2002, 4, 4571-4574.
(20) Matteucci, M. D.; Caruthers, M. H. Tetrahedron Lett. 1980, 21,
3243-3246.
J. Org. Chem, Vol. 70, No. 22, 2005 8765

Li and Gold
SCHEME 2. Synthesis of antiATa
a (a) HCl/HOAc, reflux, 15 min, 2a (75% yield), 2b (78% yield); (b) SOCl2, DMF, Et2O, reflux, 1.5 h; (c) N-(tert-butoxycarbonyl)-2-
(trimethylstannyl)aniline, Pd2(dba)3, toluene, 85 °C, 4 h, 4a (68% yield), 4b (70% yield); (d) TFA, CH2Cl2, RT, 1 h, 5a (92% yield), 5b (90%
yield); (e) HCl, Et2O, RT, 5 min/NH2CN, 50 °C, 1.5 h, 6a (72% yield), 6b (75% yield); (f) NaOMe, MeOH, RT, 1 h, 70% yield; (g) Me3SiCl,
pyridine, 0 °C, 30 min/PhCOCl, pyridine, RT, 2 h/NH4OH, 0 °C, 30 min, 78% yield; (h) DMTrCl, pyridine, Et3N, DMAP, RT, 72 h, 55%
yield; (i) 2-cyanoethyl-N,N′-diisopropylchlorophosphoramidite, N,N-diisopropylethylamine, CH2Cl2, RT, 1 h, 80% yield.
SCHEME 3. Effect of pH on Hoogsteen Binding of antiAT-F to A:T or antiAT-F+ to G:C Watson-Crick Base
Pair
of limited solubility of ZnBr2 and the required prolonged
deprotection times.
fusion of 4-fluoroaniline salt with ethyl cyanoacetate. The
hydroxy group of 11 was converted into the 4-bromo
compound 12 that was needed in the subsequent Heck
coupling. The approach to couple 12 with the ribofuranoid
glycal has been used previously by others as a route to
C-glycosides10-13 and by us to prepare the antiTA, an-
tiGC, and antiCG TRIPsides.7-9 The yield for the conver-
sion of 12 to 13 is relatively poor (33%) because of the
formation of significant amounts of the quinoline-
quinoline coupling dimer, and efforts are underway to
improve the procedure. Stereoselective reduction of ke-
tone 13 gives the desired deoxynucleoside 14. The pKa of
antiAT-F was calculated as 6.7 using either UV-vis or
fluorescence versus pH experiments (data not shown).
Design and Synthesis of an Alternative to Quin-
azoline-Based antiAT. The key difference between the
antiAT aminoquinazoline (7) and the antiGC amino-
quinoline (Scheme 1) is that the former is less basic (pKa,
4.8 versus 7.2, respectively).7,8 Therefore, antiAT is not
protonated at physiological pH and can form the two
required H-bonds with an A:T base pair via its H-bond
acceptor and H-donator atoms (Scheme 1). In contrast,
antiGC is protonated at neutral pH and binds to G:C
using its two H-bond donors.8 We proposed that a fluoro
substituent would reduce the pKa of the 2-aminoquinoline
system and thus mimic quinazoline 7 (Scheme 3). The
fluoro group, based upon DNA models, was not antici-
pated to present any steric barriers to triplex formation.
The synthesis of antiAT-F is shown in Scheme 4. The
formation of the 2-amino-6-fluoroquinoline (11) ring
system was accomplished, albeit in low yield, by the
1
The conformational analysis using NOE H NMR con-
firmed that 14 adopted the anti conformation with re-
spect to the glycosidic bond (data not shown). Incorpora-
tion of the antiAT-F phosphoramidite (17) into oligomers
followed standard procedures and gave >98% coupling
8766 J. Org. Chem., Vol. 70, No. 22, 2005

Specific Recognition of an A:T Base Pair in DNA
SCHEME 4. Synthesis of antiAT-F (14)a
a (a) NCCH2CO2Et, 210-260 °C, 1.25 h (15% yield); (b) POBr3/PBr3, 140-160 °C (57% yield); (c) Pd2(dba)3, 1,4-anhydro-2-deoxy-D-
erthro-pent-1-enitol, dioxane, reflux/TBAF, AcOH, 0 °C, 5 min (33% yield); (d) NaBH(OAc)3, -22 °C, 1 h (70% yield); (e) (i-PrCO)2O,
pyridine, RT, 2 h (47% yield); (f) DMTrCl, pyridine, Et3N, DMAP, RT, 72 h, 58% yield; (g) 2-cyanoethyl-N,N′-diisopropylchlorophosphora-
midite, N,N-diisopropylethylamine, CH2Cl2, RT, 1 h, 84% yield.
FIGURE 1. Sequences of oligomers (OL-1-4), potential triplex structures (solid vertical bonds indicate two Hoogsteen H-bonds
can be formed; horizontal open bonds indicate a mismatch), and TM’s in 10 mM sodium phosphate buffer containing 200 mM
NaCl as a function of pH; n.d. ) not detected. W ) antiAT-F (14).
yields. After deprotection, HPLC purification, and de-
salting, the purity and structure of the oligomers were
confirmed by analytical HPLC and MALDI-TOF MS.
Intramolecular Triplex Formation with antiAT-
F. The ability of antiAT-F (14) to participate in a triplex
structure with A:T base pairs was evaluated by deter-
mining the stability of a series of putative intramolecular
triplexes (see Figure 1 for potential structures) using
UV-visible absorbance versus temperature profiles. The
antiAT-F deoxynucleoside has a λmax at 325 nm, which
makes it convenient to monitor absorbance increases
commonly observed for aromatic systems when they
unstack from a triplex or duplex into a random coil. In
addition, by monitoring the absorbance at 260 nm it is
possible to observe the natural nucleobases as the Wat-
son-Crick duplex segment melts. In OL-1 and OL-2,
antiAT-F is correctly matched with A:T base pairs,
although the directionality for triplex formation is re-
versed (Figure 1). The melting profile at 325 nm for OL-1
at pH 7.0 indicates a triplex structure with a TM near 32
°C (Figure 2b). At pH 8.0, the triplex to random coil
transition is broadened but the TM remains unchanged
(Figure 2b). At both pH 7.0 and 8.0 the duplex to random
coil transition has a TM of 31-34 °C (Figure 2a).
J. Org. Chem, Vol. 70, No. 22, 2005 8767

Li and Gold
FIGURE 2. TM curves at 260 and 325 nm in 10 mM sodium phosphate buffer containing 200 mM NaCl at pH 6.0 (black), 7.0
(red), and 8.0 (green) for: (a) OL-1 at 260 nm, (b) OL-1 at 325 nm, (c) OL-2 at 260 nm, (d) OL-2 at 325 nm, (e) OL-3 (with a G:C
mismatch) at 260 nm, (f) OL-3 at 325 nm, (g) OL-4 (with a T:A mismatch) at 260 nm, and (h) OL-4 at 325 nm.
In contrast, OL-2 melts with a TM of 34.9 and 47.8 at
pH 7.0 and 8.0, respectively, and with similar hyper-
chromicities (Figure 2d). The TM for the duplex stem of
OL-2 is near 34-35 °C at all pH values (Figure 2c).
Significantly, at pH 6.0 there remains only residual
evidence for a triplex in the melting curves for either
OL-1 or OL-2 (Figure 2). This is consistent with increased
protonation of antiAT-F at pH 6.0, which would convert
antiAT-F to antiAT-F+ and from an A:T to a G:C binder
(Scheme 3). The hyperchromicity and TM at 325 nm for
the melting of the third strand decreases at lower salt
concentration for both OL-1 and OL-2 (data not shown).
We have previously reported that an intramolecular
triplex using antiTA TRIPsides, 5′-d(antiTA6-C5-T6-C5-
A6), has a TM of 25 °C at pH 7.0.7 This is substantially
lower than the >30 °C transition temperature we report
here for the antiAT-F-based OL-2, which has the same
folding orientation and same number of triplets. For
comparison, a natural pyrimidine-motif intramolecular
triplex, 5′-T6-C5-T6-C5-A6, under the same conditions has
a TM of 22.7 °C.7
Therefore, the presence of a Watson-Crick-stabilized
major groove is required for the binding of the third
strand. Given the opportunity to form an antiparallel
Watson-Crick or Hoogsteen structure, oligomers com-
posed of complementary sequences will invariably as-
sociate as the former in the absence of DNA distorting
ligands or mismatches. A notable exception is 5′-
d(ATATAT), which, in the crystal structure, does associ-
ate via Hoogsteen base pairing.22 However, even this
sequence in solution NMR studies adopts a Watson-
Crick duplex.23
In summary, we have successfully developed a syn-
thetic route to a C-nucleoside, antiAT-F, that specifically
binds to A:T base pairs via a triplex structure. The
antiAT-F glycoside provides us with the last of the four
necessary monomers needed to form a stable triplex at
any sequence of native DNA. In addition, we have
reported a synthetic pathway for access to aminoquinazo-
line C-glycosides that cannot be prepared by existing
methods.
Another approach to determine the binding specificity
of antiAT-F is to replace a single A:T base pair in the
duplex segment of OL-2 with either a G:C (OL-3) or T:A
(OL-4) base pair. These changes in the stem should
introduce mismatches between the antiAT-F segment
and the duplex causing a reduction in triplex stability.
The results corroborate this expectation (Figure 2f,h):
both OL-3 and OL-4 show no evidence of temperature-
dependent hyperchromicity when monitored at 325 nm.
This suggests that, at neutral pH, antiAT-F only binds
stably to A:T.
We explored the possibility that antiAT-F could form
an antiparallel Hoogsteen duplex with dA by synthesiz-
ing 5′-(antiAT-F)6-C5-(A)6. Modeling shows that antiAT-F
and dA can also associate via a parallel pseudo-Watson-
Crick duplex. Because of the hairpin arrangement, only
the former is possible in 5′-(antiAT-F)6-C5-(A)6. The UV-
visible melt for 5′-(antiAT-F)6-C5-(A)6 showed no evidence
for duplex formation at 250 or 325 nm (data not shown).
Experimental Section
All reagents were of the highest grade commercially avail-
able and used without any purification. NMR studies were
performed on a 500 MHz instrument.
Synthesis of 3,5-Di-O-benzoyl-2-deoxy-â-D-threo-pento-
furanosyl Chloride (3a) and 3,5-Di-O-toluoyl-2-deoxy-â-
D-threo-pentofuranosyl Chloride (3b). O-Benzoyl-protected
2′-â-deoxyribocyanide 1a (12.0 g, 34.2 mmol) was refluxed in
concentrated HCl (30 mL) and HOAc (120 mL) for 15 min.
Volatiles were removed under vacuum, and the residue was
partitioned in 100 mL of EtOAc and 200 mL of brine. The
organic portion was separated, dried over MgSO4, and filtered.
Removal of the solvent gave 10 g of the O-benzoyl-protected
2′-â-deoxyribonic acid 2a as a viscous syrup, which was ca.
(21) Miller, W. K.; Knight, S. B.; Roe, A. J. Am. Chem. Soc. 1950,
72, 1629-1633.
(22) Abrescia, N. G. A.; Gonza´lez, C.; Gouyette, C.; Subirana, J. A.
Proc. Natl. Acad. Sci. U.S.A. 2002, 99, 2806-2811.
(23) Abrescia, N. G. A.; Gonza´lez, C.; Gouyette, C.; Subirana, J. A.
Biochemistry 2004, 43, 4092-4100.
8768 J. Org. Chem., Vol. 70, No. 22, 2005

Specific Recognition of an A:T Base Pair in DNA
95% pure (based upon NMR) and used without further
purification. The O-tolulyl-protected 2′-â-deoxyribonic acid 2b
was prepared using the same procedure. 2a (10.0 g, 23.5 mmol)
was mixed with SOCl2 (100 mL), DMF (0.3 mL), and anhy-
drous Et2O (200 mL) and refluxed for 1 h. The volatiles were
removed under reduced pressure. Toluene (30 mL) was added
to the residue, and the volatiles were removed under reduced
pressure. The acid chloride (3a) was ready to use for the next
step without any further purification. The O-tolulyl-protected
2′-â-deoxyribonic acid 3b was prepared using the same pro-
cedure and also used without any further purification.
min. The Et2O was removed in vacuo, and the mixture was
stirred at 50 °C for 1.5 h. CHCl3 (200 mL) was added followed
by 1 M NaOH (200 mL) solution. The organic layer was
separated, dried over MgSO4, filtered, and concentrated. The
residue was purified by silica gel column chromatography
(CH2Cl2 to 80% CH2Cl2-Et2O). The product, 6a, was obtained
as a solid in 72% yield: 1H NMR (CDCl3) δ 2.61-2.65 (m, 1
H), 3.07-3.13 (m, 1 H), 4.62-4.71 (m, 3 H), 5.22 (s, 2 H), 5.77
(d, 1 H, J ) 6.5 Hz), 5.90 (dd, 1 H, J ) 6.0, 10.0 Hz), 7.22 (t,
1 H, J ) 6.5 Hz), 7.42-7.70 (m, 8 H), 8.04 (d, 2 H, J ) 7.0
Hz), 8.15 (d, 2 H, J ) 6.5 Hz), 8.22 (d, 1 H, J ) 9.0 Hz); HRMS
(M + Na+) calcd for 470.1717, found 470.1732. 6b was
synthesized by the same procedure and obtained as a white
solid (75% yield): 1H NMR (CDCl3) δ 2.42 (s, 3 H), 2.47 (s, 3
H), 2.60-2.64 (m, 1 H), 3.03-3.09 (m, 1 H), 4.56-4.69 (m, 3
H), 5.17 (s, 2 H), 5.74 (d, 1 H, J ) 6.5 Hz), 5.90 (dd, 1 H, J )
6.0, 10.0 Hz), 7.21-7.5 (m, 3 H), 7.32 (d, 1 H, J ) 8.5 Hz),
7.62 (d, 1 H, J ) 8.5 Hz), 7.69 (t, 1 H, J ) 8.0 Hz), 7.95 (d, 2
H, J ) 8.5 Hz), 8.03 (d, 2 H, J ) 8.5 Hz), 8.22 (d, 1 H, J ) 8.5
Hz); HRMS (M + Na+) calcd for 498.2030, found 498.2018.
Synthesis of (3′,5′-Di-O-benzoyl-2′-deoxy-â-D-threo-
pentofuranosyl)-[2-N-(tert-butoxycarbonyl)]phenyl
Ketone (4a) and (3′,5′-Di-O-toluoyl-2′-deoxy-â-D-threo-
pentofuranosyl)-[2-(N-tert-butoxycarbonylamino)]phe-
nyl Ketone (4b). 3a (10.0 g, 24 mmol) and N-(tert-butoxy-
carbonyl)-2-(trimethylstannyl)aniline (8.55 g, 24 mmol) were
dissolved in toluene (300 mL). To this solution was added Pd2-
(dba)3 (200 mg, 0.5 mmol), and the mixture was heated to 85
°C for 4 h. The reaction was cooled, the volatiles were removed
under reduced pressure, and the residue was purified by silica
gel column chromatography (hexane to 75% hexane-Et2O).
The product 4a was obtained as a syrup in 68% yield: 1H NMR
(CDCl3) δ 1.55 (s, 9 H), 2.51-2.55 (m, 1 H), 2.79-2.85 (m, 1
H), 4.51 (dd, 1 H, J ) 11.5, 5.0 Hz,), 4.61 (dd, 1 H, J ) 11.5,
5.0 Hz), 4.65 4.66 (m, 1 H), 5.60 (dd, 1 H, J ) 9.5, 6.5 Hz),
5.63-5.65 (m, 1 H), 7.05 (t, 1 H, J ) 6.5 Hz), 7.43 7.59 (m, 6
H), 7.63 (t, 1 H, J ) 7.5 Hz), 8.03-8.07 (m, 4 H), 8.11 (d, 1 H,
J ) 9.0 Hz), 8.53 (d, 1 H, J ) 9.0 Hz), 10.82 (s, 1 H); HRMS
(M + Na+) calcd for 568.1948, found 568.1931. 4b was
synthesized by the same procedure and obtained as a white
solid (70% yield): 1H NMR (CDCl3) δ 1.55 (s, 9 H), 2.42 (s, 3
H), 2.46 (s, 3 H), 2.49-2.54 (m, 1 H), 2.76-2.82 (m, 1 H), 4.48
(dd, 1 H, J ) 12.0, 5.0 Hz), 4.61 (dd, 1 H, J ) 12.0, 5.0 Hz),
4.62-4.65 (m, 1 H), 5.57-5.61 (m, 2 H), 7.05 (t, 1 H, J ) 7.5
Hz), 7.25 (d, 2 H, J ) 8.0 Hz), 7.30 (d, 2 H, J ) 8.0 Hz), 7.56
(t, 1 H, J ) 7.5 Hz), 7.95 (d, 2 H, J ) 8.0 Hz), 7.99 (d, 2 H, J
) 8.0 Hz), 8.02 (d, 1 H, J ) 8.0 Hz), 8.53 (d, 1 H, J ) 8.0 Hz),
10.83 (s, 1 H); HRMS (M + Na+) calcd for 596.2261, found
596.2243.
Synthesis of (3′,5′-Di-O-benzoyl-2′-deoxy-â-D-threo-
pentofuranosyl)(2-amino-phenyl) Ketone (5a) and (3′,5′-
Di-O-toluoyl-2′-deoxy-â-D-threo-pentofuranosyl)(2-amino-
phenyl) Ketone (5b). 4a (9.5 g) was dissolved in CH2Cl2 (150
mL) and treated with CF3CO2H (10 mL) at RT for 1 h. Then
the solution was basified with 1 M NaOH (200 mL), the organic
portion was separated and dried over MgSO4, and the volatiles
were removed. The residue was purified by silica gel column
chromatography (hexanes-Et2O from 3:1 to 1:1). The product
5a was obtained as a syrup in 92% yield: 1H NMR (CDCl3) δ
2.49-2.54 (m, 1 H, H2′′), 2.77-2.83 (m, 1 H, H2′), 4.50-4.53
(m, 1 H), 4.60-4.66 (m, 2 H), 5.59 (dd, 1 H, J ) 9.0, 6.0 Hz,
H1′), 5.63-5.64 (m, 1 H), 6.30 (bs, 2 H), 6.67-6.69 (m, 2 H),
7.32 (m, 1 H), 7.42-7.64 (m, 6 H), 7.88 (d, 1 H, J ) 7.0 Hz),
8.06-8.11 (m, 4 H); HRMS (M + Na+) calcd for 468.1423, found
468.1410. 5b was synthesized by the same procedure and
obtained as a white solid (90% yield): 1H NMR (CDCl3) δ 2.42
(s, 3 H), 2.45 (s, 3 H), 2.48-2.52 (m, 1 H), 2.75-2.81 (m, 1 H),
4.48 (dd, 1 H, J ) 12.0, 4.5 Hz), 4.61 (dd, 1 H, J ) 12.0, 4.5
Hz), 4.62-4.64 (m, 1 H), 5.57 (dd, 1 H, J ) 6.5, 9.5 Hz), 5.61
(d, 1 H, J ) 6.5 Hz,), 6.32 (bs, 2 H), 6.65-6.70 (m, 2 H), 7.24
(d, 2 H, J ) 8.0 Hz), 7.28-7.31 (m, 3 H), 7.86 (d, 1 H, J ) 8.5
Hz), 7.96 (d, 2 H, J ) 8.0 Hz), 7.99 (d, 2 H, J ) 8.0 Hz); HRMS
(M + Na+) calcd for 496.1736, found 496.1721.
Synthesis of 2-Amino-4-(2′-deoxy-â-D-threo-pento-
furanosyl)quinazoline (7). 6a or 6b (5.5 g) was dissolved
in CHCl3 (10 mL) and MeOH (100 mL). To this solution was
added NaOMe (1.8 g), and the mixture was stirred at RT for
1 h. The volatiles were removed under reduced pressure, and
the residue was purified by silica gel column chromatography
(CH2Cl2 to 90% CH2Cl2-Et2O). The product was recrystallized
from Et2O-EtOH as white crystals (2.02 g, 70% yield): mp
1
95-97 °C; H NMR (DMSO-d6) δ 2.09-2.13 (m, 1 H), 2.48-
2.53 (m, 1 H), 3.43-3.46 (m, 2 H), 3.87-3.90 (m, 1 H), 4.25-
4.30 (m, 1 H), 4.78 (t, 1 H, J ) 6.0 Hz), 5.15 (d, 1 H, J ) 4.5
Hz), 5.61 (dd, 1 H, J ) 6.5, 9.5 Hz), 6.74 (s, 2 H), 7.20 (t, 1 H,
J ) 8.5 Hz), 7.44 (d, 1 H, J ) 8.0 Hz), 7.66 (d, 1 H, J ) 8.5
Hz), 8.17 (d, 1 H, J ) 8.0 Hz); HRMS (M + H+) calcd for
262.1192, found 262.1179.
Synthesis of N2-Benzoyl-4-(2′-deoxy-â-D-threo-pento-
furanosyl)quinazoline (8). 7 (1.2 g) was dissolved in dry
pyridine (100 mL) was cooled in an ice bath and then (CH3)3-
SiCl (10 mL) was added. The reaction was stirred for 0.5 h
before the ice bath was removed. Then benzoyl chloride (8.6
mL) was added, the reaction was stirred at RT for 2.5 h, and
then H2O (20 mL) was added at 0 °C. After 20 min, NH4OH
(20 mL) was added at 0 °C, and the reaction was stirred for
0.5 h. The volatiles were removed under reduced pressure, and
the residue was purified by silica gel column chromatography
(CH2Cl2 to 95% CH2Cl2-MeOH). The product was recrystal-
lized from CH2Cl2-EtOH as an off-white solid (1.31 g, 78%
1
yield): mp 88-90 °C; H NMR (DMSO-d6) δ 2.16-2.21 (m, 1
H), 2.55-2.64 (m, 1 H), 3.40-3.50 (m, 2 H), 3.91-3.94 (m, 1
H), 4.30-4.33 (m, 1 H), 4.72 (t, 1 H, J ) 5.5 Hz), 5.17 (d, 1 H,
J ) 4.0 Hz), 5.82 (dd, 1 H, J ) 6.5, 9.0 Hz), 7.5-7.54 (m, 2 H),
7.60-7.65 (m, 2 H), 7.88 (d, 1 H, J ) 8.0 Hz), 7.92-7.98 (m, 1
H), 8.00 (d, 2 H, J ) 7.5 Hz), 8.48 (d, 1 H, J ) 8.0 Hz), 11.12
(s, 1 H); HRMS (M + H+) calcd for 388.1273, found 388.1283.
Synthesis of N2-Benzoyl-4-[2′-deoxy-â-D-threo-pento-
furanosyl-5′-O-(4,4′-dimethoxytrityl)]quinazoline (9). 8
(1.3 g) was dissolved in pyridine (60 mL), and 4,4′-dimethox-
ytrityl chloride (1.4 g), Et3N (5 mL), and a catalytic amount of
DMAP were added. The reaction was stirred at RT for 72 h,
after which time the volatiles were removed under reduced
pressure. The residue was purified by silica gel column
chromatography (CH2Cl2 to 97% CH2Cl2-MeOH). The product
was obtained as a light yellow solid (1.3 g, 55% yield): 1H NMR
(CDCl3) δ 2.41-2.45 (m, 1 H), 2.87-2.93 (m, 1 H), 3.20 (dd, 1
H, J ) 8.5, 10.0 Hz,), 3.46 (dd, 1 H, J ) 8.5, 10.0 Hz), 3.68 (s,
3 H), 3.69 (s, 3 H), 4.21-4.24 (m, 1 H), 4.67 (m, 1 H), 6.05 (dd,
1 H, J ) 3.0, 9.0 Hz), 6.72 (d, 2 H, J ) 1.0 Hz), 6.74 (d, 2 H,
J ) 1.0 Hz), 7.11 (t, 1 H, J ) 7.5 Hz), 7.20 (t, 2 H, J ) 8.0 Hz),
7.26 (t, 2 H, J ) 7.5 Hz), 7.40-7.55 (m, 10 H), 7.89 (m, 1 H),
8.10 (d, 1 H, J ) 8.5 Hz), 8.37 (d, 1 H, J ) 8.5 Hz), 8.55 (s, 1
H). HRMS (M + H+) calcd for 668.2761, found 668.2758.
Synthesis of 2-Amino-4-(3′,5′-di-O-benzoyl-2′-deoxy-â-
D-threo-pentofuranosyl)quinazoline (6a) and 2-Amino-
4-(3′,5′-di-O-toluoyl-2′-deoxy-â-D-threo-pentofuranosyl)-
quinazoline (6b). To 5a (7.0 g) in Et2O (100 mL) was added
2 M HCl in Et2O (20 mL) at RT. The mixture was stirred for
5 min, the volatiles were removed, NH2CN (3.8 g) in Et2O (30
mL) was added, and the reaction was stirred for another 5
J. Org. Chem, Vol. 70, No. 22, 2005 8769

Li and Gold
Synthesis of N2-Benzoyl-4-[2′-deoxy-â-D-threo-pento-
furanosyl3′-O-(2-cyanoethoxy)(diisopropylamino)-
phosphino-5′-O-(4,4′-dimethoxytrityl)]quinazoline (10).
9 (0.5 g) was dissolved in dry CH2Cl2 (30 mL) and cooled in
ice, and Hunig’s base (400 µL) and 2-cyanoethyl-N,N′-diiso-
propylchlorophosphoramidite (280 µL) were added. The reac-
tion was stirred at RT for 1 h, the solvent was removed, and
the residue was purified by silica gel column chromatography
(CH2Cl2-hexanes-Et2O-Et3N, 33:66:100:1). The final product
was obtained as a light yellow solid (0.53 g, 80% yield): 1H
NMR (CDCl3) δ 1.09-1.22 (m, 14 H), 2.49-2.67 (m, 2 H), 2.90-
2.96 (m, 1 H), 3.08-3.12 (m, 1 H), 3.46-3.55 (m, 1 H), 3.59-
3.64 (m, 1 H), 3.67 (s, 3 H), 3.68 (s, 3 H), 3.70-3.82 (m, 1 H),
3.88-3.91 (m, 1 H), 4.33-4.39 (m, 1 H), 4.75 (m, 1 H), 6.03
(m, 1 H), 6.71 (m, 4 H), 7.12 (m, 1 H), 7.16-7.28 (m, 4 H),
7.41-7.56 (m, 10 H), 7.91 (m, 1 H), 8.13 (dd, 1 H, J ) 2.5, 8.5
Hz), 8.41 (t, 1 H, J ) 8.0 Hz), 8.54 (s, 0.5 H), 8.63 (s, 0.5 H);
HRMS-FAB (M + H+) calcd for 867.3761, found 867.3772.
Synthesis of 2-Amino-4-hydroxy-6-fluoroquinoline (11).
4-Fluoroanilinium toluene-p-sulfonate (76 g) was heated at 260
°C for 5 min under N2, and ethyl cyanoacetate (30 g) was added
dropwise through an air condenser within 5 min. Then the
temperature of the reaction was allowed to drop to 220-250
°C. After 90 min of heating, the orange slurry was cooled,
CHCl3 (100 mL) was added, and the resulting mixture was-
refluxed overnight to dissolve the residue. This solution was
vigorously mixed with H2O (100 mL), EtOH (40 mL), and
saturated Na2CO3 solution (100 mL). The resulting slurry was
filtered, and the product, a pale-yellow solid, was washed with
H2O and dried in vacuo over P2O5 (7.4 g, 15% yield): 1H NMR
(DMSO-d6) δ 5.26 (bs, 1 H), 6.22 (s, 2 H), 7.32 (m, 2 H), 7.55
(d, 1 H, J ) 8.5), 10.79 (bs, 1 H, OH); HRMS-FAB (M+) calcd
for 178.0542, found 178.0545.
Synthesis of 2-Amino-4-bromo-6-fluoroquinoline (12).
11 (7.2 g) was heated with POBr3 (22.0 g) in PBr3 (20 mL) at
140-160 °C under N2 for 19 h. The reaction was cooled,
carefully basified with 2 M NaOH (250 mL), and thrice
extracted with CHCl3 (30 mL). The organic layer was concen-
trated, and the residue was purified by silica gel chromatog-
raphy using CHCl3-acetone-Et3N (75:25:1) to afford the
product, which was crystallized from benzene-CH2Cl2-hexane
to furnish off-white crystals (5.5 g, 57% yield): 1H NMR
(CDCl3) δ 4.70 (bs, 2 H), 7.10 (s, 1 H), 7.37 (m, 1 H), 7.66 (m,
2 H); HRMS-FAB (M+) calcd for 239.9698, found 239.9707.
Synthesis of 4-[â-D-glyceropentofuran-3′-ulos-1′-yl]-2-
amino-6-fluoroquinoline (13). 1,4-Anhydro-3-O-(tert-butyl-
diphenylsilyl)-2-deoxy-D-erythro-pent-1-enitol13 (2.13 g) and 12
(1.21 g) were dissolved in dioxane (100 mL) under N2. Then
Pd2(dba)3 (0.81 g) and (t-Bu)3P (0.71 mL) were added to the
reaction. After the reaction had been purged with N2 for 15
min, dicyclohexylmethylamine (1.35 mL) was added and the
mixture was refluxed under N2 for 20 h. The reaction was
cooled and filtered, and the filtrate was concentrated in vacuo.
The residue was subjected to silica gel chromatography with
CH2Cl2 to CH2Cl2-MeOH (gradient from 100 to 95%) to give
the desired nucleoside and the quinoline-quinoline dimer side
product in a ratio 2:1 (total 1.95 g). This mixture was dissolved
in THF (40 mL) containing HOAc (0.4 mL) and Bu4NF (5 mL,
1 M in THF) and stirred at 0 °C for 50 min. Then NH4OH (2
mL) was added, and the solution was concentrated in vacuo.
The residue was subjected to silica gel column chromatography
with CH2Cl2 to 91% CH2Cl2-MeOH to give the desired
desilylated product (0.46 g, 33% overall yield): 1H NMR
(CDCl3) δ 2.23 (dd, 1 H, J ) 13.0, 11.0 Hz), 3.08 (dd, 1 H, J )
13.0, 11.0 Hz), 4.01-4.17 (m, 2 H), 4.20 (m, 1 H), 5.63 (dd, 1
H, J ) 7.5, 5.5 Hz), 6.15 (bs, 2 H), 7.25 (s, 1 H), 7.28 (m, 2 H),
7.67 (m, 1 H); HRMS-FAB (M+) calcd for 276.0910, found
276.0914.
solution, and the mixture was stirred at -23 °C for 30 min.
The reaction was concentrated, and the residue was chromato-
graphed on silica gel using CH2Cl2-MeOH (gradient from 10:1
to 2:1). The crude product was obtained as a pale yellow
powder, which was crystallized from Et2O-EtOH to yield an
1
off-white powder (0.77 g, 70% yield): mp 69-70 °C; H NMR
(DMSO-d6) δ 1.73-1.78 (m, 1 H, C2′-H), 2.33-2.37 (m, 1 H,
C2′′-H), 3.45 (dd, 1 H, J ) 11.5, 6.0, C5′-H), 3.56 (dd, 1 H, J
) 11.5, 6.0, C5′′-H), 3.85-3.88 (m, 1 H, C4′-H), 4.18 (m, 1
H, C3′-H), 4.77 (bs, 1 H, OH3′), 5.15 (bs, 1 H, OH5′), 5.45
(dd, 1 H, J ) 10.0, 5.5, C1′-H), 6.37 (s, 2 H, NH2), 6.96 (s, 1
H), 7.32-7.36 (m, 1 H), 7.41-7.49 (m, 2 H); 19F NMR (DMSO-
d6) δ -90.26 (one single peak with multiple splits, referenced
by TFA in H2O); HRMS-FAB (M + H+) calcd for 278.1067,
found 278.1070.
Synthesis of N2-Isobutyryl-4-[2′-deoxy-â-D-threo-pento-
furanosyl]-2-amino-6-fluoroquinoline (15). 14 (0.88 g) in
dry pyridine (50 mL) was cooled in an ice bath and treated
with TMS-Cl (4.8 mL) under N2. The reaction was stirred at
0 °C for 30 min, then isobutyric anhydride (2.62 mL) was
added, and the reaction was stirred for an additional 2 h at
RT under N2. The reaction was then cooled in an ice bath, and
cold H2O (6.25 mL) was added. After 15 min, concentrated
NH4OH (6.25 mL) was added to give a solution approximately
2 M in NH3. This mixture was stirred for another 30 min and
then concentrated under reduced pressure to afford an oil that
was purified on silica gel with CH2Cl2-MeOH (10:1) to furnish
the desired product as a pale yellow solid (0.53 g, 47% yield):
1H NMR (CDCl3) δ 1.31 (dd, J ) 6.0, 0.5, 6 H), 2.13-2.18 (m,
1 H, C2′-H), 2.57-2.64 (m, 2 H, including C2′′-H), 3.86 (dd,
1 H, J ) 17.5, 4.5, C5′-H), 4.01 (dd, 1 H, J ) 17.5, 4.5, C5′′-
H), 4.21 (dd, 1 H, J ) 7.5, 3.5, C4′-H), 4.57 (m, 1 H, C3′-H),
5.77 (dd, 1 H, J ) 9.0, 8.5, C1′-H), 7.43 (m, 1 H), 7.53 (m, 1
H), 7.84 (m, 1 H), 8.13 (s, 1 H), 8.68 (s, 1 H); HRMS-FAB (M+)
calcd for 348.1485, found 348.1492.
Synthesis of N2-Isobutyryl-4-[2′-deoxy-â-D-threo-pento-
furanosyl-5′-O-(4,4′-dimethoxytrityl)]-2-amino-6-fluoro-
quinoline (16). 15 (0.4 g) was dissolved in dry pyridine (10
mL) under a N2 atmosphere, and 4,4′-dimethoxytrityl chloride
(0.5 g) and DMAP (10 mg) were added. Another aliquot of trityl
chloride was added after 3 h, and the reaction was stirred
under N2 at RT for a total of 6 h. After concentration, the
residue was dissolved in CHCl3 (30 mL) and washed thrice
with saturated NaHCO3 (30 mL) and thrice with H2O (30 mL).
The solution was concentrated, and the residue was chromato-
graphed on silica gel with CH2Cl2-MeOH-Et3N (gradient
from 100:0:1 to 100:1:1). The final product was a white solid
(0.43 g, 58% yield): 1H NMR (CDCl3) δ 1.28 (dd, J ) 7.5, 6.5,
6 H), 2.20-2.25 (m, 1 H, C2′-H), 2.46-2.51 (m, 1 H, C2′′-H),
2.55-2.61 (m, 1 H), 3.36 (dd, 1 H, J ) 9.5, 5.5, C5′-H), 3.51
(dd, 1 H, J ) 9.5, 5.5, C5′′-H), 3.79 (s, 6 H), 4.09-4.12 (m,
C4′-H), 4.43-4.45 (m,1 H, C3′-H), 5.65 (dd, 1 H, J ) 10.0,
6.0, C1′-H), 6.80-6.82 (m, 4 H), 7.18-7.48 (m, 10 H), 7.68
(dd, 1 H, J ) 9.5, 2.5), 7.83 (dd, 1 H, J ) 9.5, 5.5), 8.00 (bs, 1
H), 8.61 (s, 1 H); HRMS-FAB (M + H+) calcd for 651.2871,
found 651.2862.
Synthesis of N2-Isobutyryl-4-[2′-deoxy-â-D-threo-pento-
furanosyl-3′-O-(2-cyanoethoxy)(diisopropylamino)-
phosphino-5′-O-(4,4′-dimethoxytrityl)]-2-amino-6-fluoro-
quinoline (17). 16 (0.1 g) dissolved in CH2Cl2 (10 mL) and
cooled in an ice bath under N2 atmosphere was treated with
Hunig’s base (200 µL) followed by 2-cyanoethyl-N,N-diisopro-
pylphosphoramidite (90 µL). The reaction was stirred at 0 °C
for 10 min and then at RT for 45 min. After concentration in
vacuo, the residue was purified by column chromatography
on silica gel with CH2Cl2-hexanes-Et2O-Et3N (50:100:150:
1). The final product was a white powder (0.11 g, 84% yield):
1H NMR (CDCl3) δ 1.12-1.30 (m, 20 H), 2.18-2.28 (m, 1 H,
C2′-H), 2.47-2.67 (m, 3 H, including C2′′-H), 3.38-3.42 (m,
2 H, C5′-H, C5′′-H), 3.31-3.57 (m, 1 H), 3.60-3.70 (m, 1 H),
3.78 (s, 6 H), 4.31 (bs, 1 H, C4′-H), 4.55 (m,1 H, C3′-H), 5.65
(dd, 1 H, J ) 10.5, 5.0, C1′-H), 6.77-6.81 (m, 4 H), 7.16-
Synthesis of 4-[2′-Deoxy-â-D-threo-pentofuranosyl]-2-
amino-6-fluoroquinoline (antiAT-F) (14). 13 (1.1 g) was
dissolved in AcOH and CH3CN (100 mL, 1:1) and stirred under
N2 at -23 °C. NaHB(OAc)3 (1.4 g) was added to the cooled
8770 J. Org. Chem., Vol. 70, No. 22, 2005

Specific Recognition of an A:T Base Pair in DNA
7.45 (m, 10 H), 7.74 (dd, 1 H, J ) 10.5, 3.0), 7.83 (dd, 1 H, J
) 9.5, 5.5), 7.96 (s, 1 H), 8.64 (s, 1 H); HRMS-FAB (M + H+)
calcd for 851.3950, found 851.3967.
purified by reverse-phase HPLC, the trityl group was removed
with 80% HOAc, and the oligomers were repurified and
desalted on a Sephadex G20 column. The incorporation of the
modified nucleotides was determined by UV-visible spectros-
copy and MALDI-TOF MS.
UV-Visible Melting Studies. Absorbance versus temper-
ature profiles were measured simultaneously at 260 and 325
nm with a thermoelectrically controlled UV-vis spectropho-
tometer. The absorbance was monitored in the temperature
range of 0-80 °C, by increasing the temperature at a rate of
∼0.4 °C/min. In these experiments, the 325-nm wavelength
was used to monitor the absorbance changes of the antiAT-F
C-glycosides, while the 260 nm follows the changes of the
nucleobases. First derivative and shape analysis of the result-
ing melting curves affords transition temperatures, TM.
pKa of antiAT-F (14). To 14 (1.62 mg) dissolved in H2O
was added NaOAc buffer to give solutions (1.24 µM final
concentration) with measured pH values of 4.6, 5.5, 5.9, 6.2,
6.6, 6.8, 7.2, 7.5, and 8.3. Two methods were used to measure
the pKa. First, the UV spectra of these solutions were recorded
from 200 to 400 nm. The UV spectra at the different pH values
were analyzed by plotting absorbance at 235 nm as a function
of pH to calculate the pKa. Second, the fluorescence emission
spectra of these solutions were recorded from 300 to 600 nm
with excitation at 330 nm. The fluorescence spectra at different
pH values were analyzed by plotting I (391 nm), I (396 nm), I
(402 nm), or the shifts of maximum wavelength, respectively,
as a function of pH to calculate the pKa. Both methods gave a
calculated pKa of 6.7.
NMR Conformational Studies. The conformational analy-
sis of the antiAT (7) and antiAT-F (14) TRIPsides was
determined by 1H NMR on a 500-MHz spectrometer using
NOESY with presaturation field strength γB1 of 50 Hz and a
mixing time of 0.4 s at 25 °C in 10 mM sodium phosphate
buffer (pH 7.0) in D2O containing 50 mM NaCl. For antiAT
(7), the sample concentration was 15.6 mM, and the relaxation
delay for the NOESY experiment was 2.0 s. For antiAT-F (14),
the sample concentration was 14.8 mM, and the relaxation
delay was 2.1 s.
Synthesis and Purification of Oligomers. The oligomers
were synthesized on a 200 nmol scale using standard phos-
phoramidite solid-phase methods with coupling yields greater
than 98%. After the synthesis, the resin was collected and
treated with NH4OH at 55 °C for 8 h to cleave the oligomer
off the column and remove the acyl protecting groups. The
crude oligomers (as the 5′-dimethoxytrityl derivatives) were
Acknowledgment. This work was supported by
NIH Grant RO1 GM068430 and Cancer Center Support
Grant P30 CA36727 from the National Cancer Institute.
We thank Paul Keifer and Gregory Kubik of the UNMC
Eppley Cancer Center NMR and Molecular Biology
Share Resources for help with the NMR studies and
oligomer synthesis, respectively.
Supporting Information Available: NMR spectra and
HRMS for compounds 4a-17, effect of salt concentration (50
versus 200 mM NaCl) on UV-visible melting curves of OL-
1-4, plots of the effect of pH on UV-visible and fluorescence
spectra of antiAT-F (14). This material is available free of
charge via the Internet at http://pubs.acs.org.
JO0511445
J. Org. Chem, Vol. 70, No. 22, 2005 8771