5-Substituted 2-aminopyridine C-nucleosides as protonated cytidine equivalents: Increasing efficiency and selectivity in DNA triple-helix formation

Article abstract of DOI:10.1021/ja9704904

The easily accessible C-nucleoside 2-amino-5-(2'-deoxy-β-D- ribofuranosyl)pyridine (P) and its 3-methyl ((Me)P) and 2'-O-methyl (P(OMe)) derivatives were synthesized and incorporated as protonated cytidine equivalents in homopyrimidine oligodeoxynucleotid

Full text of DOI:10.1021/ja9704904

J. Am. Chem. Soc. 1997, 119, 5499-5511
5499
5-Substituted 2-Aminopyridine C-Nucleosides as Protonated
Cytidine Equivalents: Increasing Efficiency and Selectivity in
DNA Triple-Helix Formation
Stefan Hildbrand, Adrian Blaser, Serge P. Parel, and Christian J. Leumann*
Contribution from the Department of Chemistry and Biochemistry of the UniVersity of Bern,
Freiestrasse 3, CH-3012 Bern, Switzerland
ReceiVed February 14, 1997X
Abstract: The easily accessible C-nucleoside 2-amino-5-(2′-deoxy-â-D-ribofuranosyl)pyridine (P) and its 3-methyl
(
MeP) and 2′-O-methyl (POMe) derivatives were synthesized and incorporated as protonated cytidine equivalents in
homopyrimidine oligodeoxynucleotides. Tm measurements indicate that oligonucleotides containing P or MeP have
a higher affinity to double-stranded DNA over the pH range of 6-8 than 5-methylcytidine (MeC) containing
oligonucleotides. This increase in stability is most pronounced above pH 7.0. The average increase in Tm/modification
for the dissociation of oligonucleotide d(TTTTTMePTMePTMePTMePTMePT) from a 21-mer target duplex at pH 7.5 is
2.3 °C relative to oligonucleotide d(TTTTTMeCTMeCTMeCTMeCTMeCT). The pH dependence and sequence
composition effects are much less pronounced for MeP (and also P) containing oligonucleotides than for MeC containing
ones. While oligonucleotide d(TTTMeCMeCMeCMeCTTTTMeCTTT) shows no longer any affinity to the target duplex
above pH 6.5, oligonucleotide d(TTTMePMePMePMePTTTTMePTTT) displays preserved binding with a Tm of 32.5 °C
at pH 7.0 and even binds with a Tm of 23.3 °C at pH 8.0. Oligonucleotides containing POMe show distinctly less
stable triple helices. The average decrease in Tm/modification for oligonucleotide d(TTTTTPOMeTPOMeTPOMeTPOMe
-
TPOMeT) at pH 6.5 is 6.7 °C relative to the MeC containing oligonucleotide. DNase I footprint titration experiments
indicate that d(TTTTTMePTMePTMePTMePTMePT) binds not only five times stronger to a 229 base pair DNA fragment
than d(TTTTTMeCTMeCTMeCTMeCTMeCT) but also with higher selectivity. UV-melting experiments show that
duplexes of d(TTTTTCTXTCTCTCT) (where X ) P, MeP, or POMe) with their antiparallel Watson-Crick complement
are dramatically less stable (∆Tm < -12 °C) at pH 8.0 than the corresponding natural duplex. Thus the new bases
P and MeP show Hoogsteen specific pairing behavior.
Introduction
Triple helix-forming oligonucleotides containing this analog
showed increased affinity to duplex DNA by 3.9 °C per
modification relative to 5-methyldeoxycytidine. A more obvi-
ous way to increase the basicity consists in replacing cytosine
by an intrinsically more basic heterocycle. In this context only
6-amino-2′-O-methylcytidine displaying a pKa of 6.8 was
investigated in triple-helix formation.5 Oligonucleotides con-
taining this analog, however, did not form stable triple helices
with duplex DNA.
Our design of a more readily protonated cytidine equivalent
started from the basic idea, that replacement of the nitrogen
atom N(1) in the pyrimidine ring by carbon and concomitant
removal of the 2-oxo function should result in the more basic
pyridine C-nucleoside P without perturbing the hydrogen-
Oligonucleotide-directed sequence-specific recognition of
double-helical DNA is of interest for the selective control of
gene expression.1 To date two structural motifs of such triple
helices are known: those in which the third strand is primarily
composed of pyrimidine bases (parallel binding motif), and those
in which the third strand is primarily composed of purine bases
(antiparallel binding motif). Triple-helix formation of oligo-
nucleotides with duplex DNA in the parallel (pyrimidine-
purine-pyrimidine) binding motif requires protonation of the
cytosine units in the third strand.2 This requirement, however,
limits the formation of stable triple helices at physiological
conditions due to the relatively low pKa of deoxycytidine (pKa
) 4.3). In the past, attempts to overcome this obstacle mainly
focused on the design and use of nonnatural nucleosides carrying
charge-neutral bases which display the hydrogen-bonding pattern
of protonated cytosine.3
(3) Selected examples: (a) Ono, A.; Ts’o, P. O. P.; Kan, L.-S. J. Am.
Chem. Soc. 1991, 113, 4032-4033. (b) Ono, A.; Ts’o, P. O. P.; Kan, L.-S.
J. Org. Chem. 1992, 57, 3225-3230. (c) Koh, J. S.; Dervan, P. B. J. Am.
Chem. Soc. 1992, 114, 1470-1478. (d) Priestley, E. S.; Dervan, P. B. J.
Am. Chem. Soc. 1995, 117, 4761-4765. (e) Xiang, G.; Soussou, W.;
McLaughlin, L. W. J. Am. Chem. Soc. 1994, 116, 11155-11156. (f)
Berressem, R.; Engels, J. W. Nucleic Acids Res. 1995, 23, 3465-3472. (g)
Xiang, G.; Bogacki, R.; McLaughlin, L. W. Nucleic Acids Res. 1996, 24,
1963-1970. (h) Miller, P. S.; Bhan, P.; Cushman, C. D.; Trapane, T. L.
Biochemistry 1992, 31, 6788-6793. (i) Jetter, M. C.; Hobbs, F. W.
Biochemistry 1993, 32, 3249-3254. (k) Davison, E. C.; Johnsson, K.
Nucleosides Nucleotides 1993, 12, 237-243. (l) Young, S. L.; Krawczyk,
S. H.; Matteucci, M. D.; Toole, J. J. Proc. Natl. Acad. Sci. USA 1991, 88,
10023-10026. (m) Krawczyk, S. H.; Milligan, J. F.; Wadwani, S.; Moulds,
C.; Froehler, B. C.; Matteucci, M. D. Proc. Natl. Acad. Sci. USA 1992, 89,
3761-3764. (n) Hunziker, J.; Priestley, E. S.; Brunar, H.; Dervan, P. B. J.
Am. Chem. Soc. 1995, 117, 2661-2662. (o) von Krosigk, U.; Benner, S.
A. J. Am. Chem. Soc. 1995, 117, 5361-5362.
An alternative approach to increase the affinity of oligo-
nucleotides to target DNA at neutral pH consists in the re-
placement of the cytidine nucleosides by analogs that ideally
were completely protonated at physiological conditions. Car-
bocyclic 5-methyldeoxycytidine for example was found to be
more basic than 5-methyldeoxycytidine by 0.45 pKa units.4
X Abstract published in AdVance ACS Abstracts, June 1, 1997.
(1) (a) Thuong, N. T.; He´le`ne, C. Angew. Chem. 1993, 105, 697-723;
Angew. Chem., Int. Ed. Engl. 1993, 33, 666-690. (b) Soyfer, V. N.;
Potaman, V. N. Triple-Helical Nucleic Acids; Springer-Verlag: New York,
1996.
(2) (a) Felsenfeld, G.; Davies, D. R.; Rich, A. J. Am. Chem. Soc. 1957,
79, 2023-2024. (b) Moser, H. E.; Dervan, P. B. Science 1987, 238, 645-
650. (c) Franc¸ois, J.-C.; Saison-Behmoaras, T.; He´le`ne, C. Nucleic Acids
Res. 1988, 16, 11431-11440.
(4) Froehler, B. C.; Ricca, D. J. J. Am. Chem. Soc. 1992, 114, 8320-
8322.
(5) Pudlo, J. S.; Wadwani, S.; Milligan, J. F.; Matteucci, M. D. Bioorg.
Med. Chem. Lett. 1994, 4, 1025-1028.
S0002-7863(97)00490-3 CCC: $14.00 © 1997 American Chemical Society

5500 J. Am. Chem. Soc., Vol. 119, No. 24, 1997
Hildbrand et al.
4a and 4b (only the â-configured isomers).14 The synthesis of
POMe followed a slightly different way, in which the 2′-O-
methylated precursor 11, which could be readily obtained from
D-ribonolactone 10 in two steps, was used as the sugar
component. The analogous coupling reaction with 2a yielded,
however, after reduction an anomeric mixture 12 (â/R ) 10:1),
the R-anomer of which could be easily removed by column
chromatography after protection of the exocyclic amino function
with phenoxyacetic anhydride (Pac2O). The anomeric configu-
rations in 4a and 6c were assigned by 1H NMR (NOE)
experiments. Irradiation at the H-C(3′) atoms of 4a and 6c
resulted in both cases in an enhancement of the signals of the
base protons H-C(4) and H-C(6). The corresponding N-
protected 2′-deoxynucleosides of 4a and 4b were obtained in
five consecutive steps, representing standard transformations in
nucleoside chemistry. Protection of the amino group with
benzoyl chloride (BzCl) in the case of 4a or phenoxyacetic
anhydride in the case of 4b followed by debenzylation with
BBr3 provided the N-protected ribo-C-nucleosides 5a and 5b.
Selective protection of the 3′- and 5′-hydroxyl groups with 1,3-
dichloro-1,1,3,3-tetraisopropyldisiloxane in pyridine afforded 6a
and 6b. Reaction of 6a with p-tolyl chlorothionoformate and
6b with thiocarbonyl diimidazole15 (f 7a and 7b) followed by
homolytic reductive cleavage of the C-O bond provided 8a
and 8b. Desilylation of 8a and 8b (and also 6c) with tetra-n-
butylammonium fluoride gave the N-protected pyridine C-
nucleosides 9a-c. Compounds 9a-c could then be converted
into the free C-nucleosides P, MeP, and POMe by treatment with
40% aqueous MeNH2 in the case of 9a or 25% aqueous NH3
in the case of 9b and 9c. The pKa values of the cytidine
nucleoside analogs are 6.26 for deoxyribonucleoside 9a and 6.63
for its 3-methyl derivative 9b as well as 6.15 for the ribonucleo-
side 9 (R1 ) R2 ) H, R3 ) OH)7 as determined by potentio-
metric titration and were within the expected range.
Figure 1. (top) The natural occurring base triplet C+GC (pyrimidine
motif) and its isomorphous P+GC triplet. (bottom) The more basic
cytidine analogs P, MeP, and POMe
.
bonding pattern relevant for triple helix formation (Figure 1).
The measured pKa of free 2-aminopyridine is 6.86, thus almost
within the neutral range.6 We recently reported in a preliminary
communication about the use and benefits of oligonucleotides
containing the C-nucleoside P in triple-helix formation.7 In-
dependent of our work, this analog (as well as its R-anomer)
was also prepared and used in triple helix-forming oligonu-
cleotides by Neidle et al. with similar results.8 On the basis of
the positive results obtained with P as a cytidine substitute in
triple-helix formation we looked for possible derivatives thereof,
which were expected to additionally enhance triple-helix forma-
tion. Here we report on our synthesis of the pyridine C-
nucleoside P9 and of its two derivatives MeP and POMe (Figure
1), as well as on their incorporation into oligodeoxynucleotides
and on the properties of these oligonucleotides with respect to
binding to single- and double-stranded DNA.
Synthesis of the Oligonucleotides 16-22. The pyridine
C-nucleosides P, MeP, and POMe were incorporated as cytidine
equivalents in oligodeoxynucleotides. The phosphoramidite
building blocks 14a-c were obtained from the corresponding
N-protected C-nucleosides 9a-c by tritylation (4,4′-dimeth-
oxytrityl chloride, pyridine) followed by reaction of 13a-c with
2-cyanoethyl N,N-diisopropylchlorophosphoramidite and
N,N-diisopropylethylamine (Hu¨nig’s base) (Figure 3).
The oligomers 16-22 (Figure 4) were synthesized on a DNA-
Synthesizer using standard solid-phase â-cyanoethyl phosphora-
midite chemistry.16 Because of the poor lability of the benzoyl
groups in residues of P an efficient deprotection of oligonucle-
otides containing the benzoyl protected C-nucleoside 9a (f 16,
19) could only be achieved by treatment with 40% aqueous
methylamine (70 °C, 26 h). Therefore we decided to change
the protective group strategy and to use the phenoxyacetyl (Pac)
Results
Synthesis of the C-Nucleosides. A scheme for the syntheses
of the pyridine C-nucleosides P, MeP, and POMe is shown in
Figure 2. As starting material for the syntheses of P and MeP
the readily available ribonolactone 311 and the stabase adducts
of 2-amino-5-bromopyridine 2a and its 3-methyl derivative 2b
were used. The coupling essentially followed the syntheses of
C-glycosides by Kraus et al.12 and Krohn et al.13 Bromide-
lithium exchange in 2 with nBuLi at -75 °C and in situ reaction
with lactone 3 furnished a mixture of hemiacetals which were
subsequently reduced with excess of Et3SiH/BF3‚Et2O to provide
as the protective group for the exocyclic amino function in Me
P
P
and POMe. The Pac-protected amino groups in residues of Me
and POMe, however, were found to be partially acetylated during
the capping step of the automated synthesis of the oligonucle-
otides 17, 18, 20, and 21. However these acetyl groups could
then be removed by methylamine treatment (40% in water, 70
(6) Albert, A.; Goldacre, R.; Phillips, J. J. Chem. Soc. 1948, 2240-2249.
(7) Hildbrand, S.; Leumann C. Angew. Chem. 1996, 108, 2100-2102;
Angew. Chem., Int. Ed. Engl. 1996, 35, 1968-1970.
(8) Bates, P. J.; Laughton, C. A.; Jenkins, T. C.; Capaldi, D. C.; Roselt,
P. D.; Reese, C. B.; Neidle, S. Nucleic Acids Res. 1996, 24, 4176-4184.
(9) While a synthesis of P appeared recently,10 MeP and POMe previously
were unknown.
(10) Hsieh, H.-P.; McLaughlin, L. W. J. Org. Chem. 1995, 60, 5356-
5359.
(11) Timpe, W.; Dax, K.; Wolf, N.; Weidmann, H. Carbohydr. Res. 1975,
39, 53-60.
(12) Kraus, G. A.; Molina, M. T. J. Org. Chem. 1988, 53, 752-753.
(13) Krohn, K.; Heins, H.; Wielckens, K. J. Med. Chem. 1992, 35, 511-
517.
(14) Attempts to extend this reaction to the 2-deoxy analog of lactone 3
were not successful, due to enolization at C(2) under the reaction conditions,
as indicated by D2O quenching after the addition step. See also: Piccirilli,
J. A.; Krauch, T.; MacPherson, L. J.; Benner, S. A. HelV. Chim. Acta 1991,
74, 397-406.
(15) This reagent was used, because standard conditions for the introduc-
tion of the p-tolylthioformate residue resulted in a loss of the phenoxyacetyl
group in 6b during the reaction.
(16) (a) Sinha, N. D.; Biernat, J.; Ko¨ster, H. Tetrahedron Lett. 1983,
24, 5843. (b) Users Manual (56-1111-56) “Gene Assembler Special / 4
Primers” from Pharmacia.

5-Substituted 2-Aminopyridine C-Nucleosides
J. Am. Chem. Soc., Vol. 119, No. 24, 1997 5501
Figure 2. Synthesis of the pyridine C-Nucleosides: (a) 1af2a: Cl(Me)2SiCH2CH2Si(Me)2Cl (1 equiv), nBuLi (2 equiv), THF, -75 °C, 2.5 h,
68%. 1bf2b: 1,2-bis[(dimethylamino)dimethylsilyl]ethane (1 equiv), ZnI2 (0.7 mol%), 140 °C, 15 h, 69%. (b) (Cl(iPr)2Si)2O (1.2 equiv), pyridine,
room temperature, 5 h, 39%. (c) Ag2O (3.4 equiv), MeI, 45 °C, 6 h, 67%. (d) 2a (1.8 equiv) or 2b (1.5 equiv), nBuLi (ca. 1.5 equiv), THF, -75
°C, 3 h, -75 °C f 0 °C, 4 h, then Et3SiH (5 equiv), BF3‚Et2O (5 equiv), CH2Cl2, -75°C f room temperature, over night, 64% (4a), 77% (4b)
and 61% (12). (e) 4af5a: BzCl (1.3 equiv), pyridine, CH2Cl2, room temperature, 24 h, then 0.4 M BBr3 in CH2Cl2, -75 °C, 4 h, 70%. 4bf5b:
(i) Pac2O (3 equiv), pyridine, room temperature, 4.5 h, 67%; (ii) 0.4 M BBr3 in CH2Cl2, -75 °C, 5 h, 81%. (f) (Cl(iPr)2Si)2O (1.2 equiv), pyridine,
room temperature, 4 h, 75% (6a) and 84% (6b). (g) Pac2O (3.6 equiv), pyridine, room temperature, 5 h, 85% (+ 8% R-anomer). (h) 6af7a:
p-TolOC(S)Cl (1.6 equiv), N,N-dimethylaminopyridine (2.2 equiv), CH3CN, room temperature, 25 h, 81%. 6bf7b: (Im)2CS (2.5 equiv), DMF, 40
°C, 4 h, 74%. (i) 7af8a: azobisisobutyronitrile (0.1 equiv), nBu3SnH (1.5 equiv), toluene, 80 °C, 4 h, 84%. 7bf8b: (Me3Si)3SiH (1.15 equiv),
toluene, 80 °C, 5 h, 67%. (k) Bu4NF (2 equiv), THF, room temperature, 1.5 h, 86% (9a), 93% (9b) and 87% (9c). (l) 9afP: MeNH2 solution (40%
in H2O), 70 °C, 25 h, 83%. 9bfMeP: NH3 solution (25% in H2O), 55 °C, 15 h, 63%. 9cfPOMe: NH3 solution (25% in H2O), 55 °C, 15 h, 91%.
°C, 24 h). In the subsequent synthesis of oligonucleotide 22,
acetylation was prevented by the use of phenoxy acetic
anhydride as the capping agent. This allowed for efficient
postsynthetic deprotection under standard ammonolysis condi-
tions (25% NH3, 55 °C, ∼18 h) and is therefore the method of
choice. The deprotected crude oligonucleotides were purified
by DEAE () (diethylamino)ethyl) ion exchange HPLC, their
purity controlled by reversed-phase HPLC and their structural
integrity analyzed by matrix-assisted laser desorption ioniza-
tion-time of flight (MALDI-TOF) mass spectrometry. The
oligonucleotides 15, 23, and 24 were prepared for comparison.
Binding of the Oligonucleotides 16-22 to Duplex DNA
(Tm Measurements). To exclude any self-pairing of the pyri-
dine C-nucleoside containing oligonucleotides, a UV-melting
experiment of oligonucleotide 19 in 10 mM NaH2PO4 and 200
mM NaCl at pH 5.0 was performed. As expected no coopera-
tive melting transition could be observed, indicating the absence
of self-association. The binding of the oligonucleotides 16-
22 to duplex DNA was monitored by UV melting curves in a
pH range of 6.0-8.0 and salt concentrations of 140 mM KCl,
7 mM Na2HPO4, and 0.5 mM MgCl2. These conditions were
chosen to approximate the intracellular cationic environment
as closely as possible.17 The UV-melting curves were recorded
for a consecutive heating-cooling-heating cycle with a linear
gradient of 0.5 °C/min. While the heating cycles were super-
imposable the cooling cycles always showed slight hysteresis
for the association of the third strands. Melting transition data
(Tm values) for third strand dissociation of 15-24 from the
chosen 21-mer target duplexes are summarized in Tables 1
and 2 and representative melting curves are reproduced in
Figure 5.
From the data in Table 1 it appears, that the oligomers 19
and 20, in which all cytidines were replaced by the 2′-deoxy-
C-nucleosides P or MeP, show an increase in Tm relative to 15
(natural sequence) and 23 containing 5-methylcytidine (MeC).
The increase in stability is most pronounced above pH 7.0. For
example, the average increase in Tm/modification for MeP at
pH 7.5 is 4.7 °C relative to cytidine and 2.3 °C relative to
5-methylcytidine. From the binding data of 16, 17, 19, and 20
(17) Alberts, B.; Bray, D.; Lewis, J.; Raff, M.; Roberts, K.; Watson, J.
D. Molecular Biology of the Cell, 3rd ed.; Garland: New York, 1994; p
508.

5502 J. Am. Chem. Soc., Vol. 119, No. 24, 1997
Hildbrand et al.
Table 2. Tm Values (°C) and Hyperchromicity (%) (in
Parentheses) of the Dissociation of the Third Strand 22 and 24 from
the Target Duplex as Determined from UV-Melting Curves (λ )
260 nm)a
strand
pH 6.0
pH 6.5
pH 7.0
pH 7.5
pH 8.0
22
24
41.5 (10)
27.2 (8)
37.6 (5)
19.3 (5)
32.5 (7)
b
28.3 (5)
b
23.3 (3)
b
a Triple-helix concentration 1.5-1.7 µM in 140 mM KCl, 7 mM
Na2HPO4, 0.5 mM MgCl2. Tm of the target duplex: 59.5 ( 1 °C. b Only
one Tm value for duplex melting () 59.5 ( 1 °C) was observed.
Figure 3. Synthesis of the phosphoramidites: (a) dimethoxytrityl
chloride (DMTCl; 1.2 equiv), pyridine, room temperature, 3.5-8 h,
84% (13a), 74% (13b), and 81% (13c). (b) (NCCH2CH2O)(iPr2N)PCl
(1.5 equiv), Et(iPr)2N (3 equiv) THF, room temperature, 2 h, 81% (14a),
81% (14b), and 86% (14c).
Figure 4. Sequences of the 21-mer target DNA duplexes (top) and
the triple helix-forming oligonucleotides 15-24 (MeC ) 5-meth-
yldeoxycytidine).
Table 1. Tm values (°C) and Hyperchromicity (%) (in Parentheses)
of the Dissociation of the Third Strand 15-21 and 23 from the
Target Duplex as Determined from UV-Melting Curves (λ ) 260
nm)a
Figure 5. Representative UV-melting curves (λ ) 260 nm) at pH
7.5 (triplex concentration 1.5-1.7 µM in 140 mM KCl, 7 mM
HNa2PO4, 0.5 mM MgCl2). (top) UV-melting curves of 20 (a) and 23
(b) with the corresponding target duplex sequence (Tm of the du-
plex 58 ( 1 °C). (bottom) UV-melting curves of 22 (c) and 24 (d)
with the corresponding target duplex sequence (Tm of the duplex 59 (
1 °C).
strand
pH 6.0
pH 6.5
pH 7.0
pH 7.5
pH 8.0
15
16
17
18
19
20
21
23
39.0 (7)
42.5 (10)
42.4 (8)
33.8 (8)
b
29.1 (4)
33.5 (7)
33.7 (5)
26.1 (5)
44.9 (9)
45.0 (10)
7.0 (2)
18.9 (2)
25.1 (4)
24.7 (3)
16.7 (2)
38.1 (6)
39.2 (8)
c
9.9 (1)
16.3 (1)
16.6 (2)
8.7 (1)
32.3 (5)
33.6 (7)
c
c
9.8 (1)
9.9 (1)
c
24.2 (3)
26.0 (5)
c
b
It is known that triple helix-forming oligonucleotides contain-
ing MeC units display serious sequence composition limitations
with regard to targeting contiguous G-rich purine tracts.18 To
evaluate the sequence composition effect of the C-nucleoside
MeP, oligonucleotide 22 containing four contiguous MeP units
was prepared and compared with oligonucleotide 24 (which
contains four contiguous MeC units) with respect to binding to
the corresponding target duplex (Figure 4, right). From the
binding data in Table 2 it becomes clear, that the affinity of
oligonucleotide 24 to the target duplex decreases much more
with increasing pH than that of oligonucleotide 22, which
contains four contiguous MeP units. While the oligomer 24
shows no longer any affinity to the target above pH 6.5, the
oligomer 22 displays preserved binding with a Tm of 32.5 °C at
pH 7.0 and even still binds at pH 8.0. Figure 6 shows a diagram
7.3 (2)
50.8 (8)
40.6 (6)
30.1 (3)
21.9 (2)
11.5 (1)
a Triple-helix concentration 1.5-1.7 µM in 140 mM KCl, 7 mM
Na2HPO4, 0.5 mM MgCl2. Tm of the target duplex: 58.5 ( 1 °C. b Only
one Tm value representing triplex f single strands transition () 58.5
( 1 °C) was observed. c Only one Tm value for duplex melting ()
58.5 ( 1 °C) was observed.
it can be seen, that the introduction of a methyl group in the
base P in the position equivalent to the methyl group of thymine
and 5-methylcytosine (f MeP) has almost no effect on triplex
stability. The oligomer 21, in which all cytidines were replaced
by the 2′-O-methyl-C-nucleoside POMe, has no longer any
affinity to the target duplex above pH 6.5. Already the
replacement of only one cytidine by POMe results in a loss of
triplex stability relative to the controls (15 and 23) as can be
seen from the binding data of 18.
(18) Brunar, H.; Dervan, P. B. Nucleic Acids Res. 1996, 24, 1987-1991
and references cited therein.

5-Substituted 2-Aminopyridine C-Nucleosides
J. Am. Chem. Soc., Vol. 119, No. 24, 1997 5503
Figure 6. Tm values of third strand dissociation as a function of pH
from UV-melting curves (260 nm) of 20 ([), 23 (b) (top) and 22 ([),
24 (b) (bottom) with the corresponding target duplex sequences. Triplex
concentration 1.5-1.7 µM in 140 mM KCl, 7 mM Na2HPO4, 0.5 mM
MgCl2.
of the triplex stability (Tm values) as a function of the pH for
the oligonucleotides 20 and 23 (top) and 22 and 24 (bottom).
As in the case of MeC also P or MeP containing oligonucleotides
display pH dependent Tm values as expected. This pH
dependence, however, is much less pronounced for MeP (and
also P) than for MeC in the pH range 6.0-8.0. The higher the
pH the larger becomes the relative difference in Tm between
MeC and MeP (and also P) containing oligonucleotides.
Selectivity of the Pyridine C-Nucleosides Containing
Oligonucleotides to Duplex DNA. The selectivity in binding
of oligonucleotides containing the base MeP was monitored by
a DNase I footprint experiment. A 3′-32P-end-labeled (pyrimi-
dine-rich strand) 229 base pair restriction fragment containing
the four triplex target sites outlined in Figure 7, displaying each
one of the four possible canonical base pairs in the center was
incubated at 18 °C with increasing concentrations ranging from
500 pM to 10 µM of the triple helix-forming oligonucleotide
20 (and 23 as a control). The DNase I generated reaction
products were separated by denaturing gel electrophoresis and
visualized by storage phosphor autoradiography.
Figure 7 (bottom) shows a typical example of such an
autoradiogram. On the left MeP-containing oligonucleotide 20
was used as triplex-forming strand and on the right the control
oligonucleotide 23 (Z ) MeC) was used. It can clearly be seen
that oligonucleotide 20 (Z ) MeP) prevents DNase I activity
within the GC box at a concentration approximately five times
lower than the MeC-containing oligonucleotide 23. While the
MeC-containing oligonucleotide 23 also binds to the AT-
containing box at concentrations >1 µM, this is not observed
for the MeP containing oligonucleotide 20. Neither oligonucle-
otide (20 and 23) shows substantial binding to the TA or CG
containing cassette at the concentration range investigated.
Oligonucleotide 19 produced similar results as 20.19
Figure 7. (top) Schematic representation of the oligonucleotides and
duplex target sites used in the DNase footprint titrations. (bottom)
Autoradiogram of a 8% denaturing polyacrylamide gel used to separate
the fragments generated by DNase I digestion. The position of the
binding sites within the 229-mer restriction fragment is indicated on
the right. Lanes 1-10 and 1′-10′, DNA cleavage products produced
by oligonucleotide 20 (left) resp. oligonucleotide 23 (right) at various
concentrations (500 pM, lanes 1 and 1′; 1 nM, lanes 2 and 2′; 5 nM,
lanes 3 and 3′; 10 nM, lanes 4 and 4′; 50 nM, lanes 5 and 5′; 100 nM,
lanes 6 and 6′; 500 nM, lanes 7 and 7′; 1 µM, lanes 8 and 8′; 5 µM,
lanes 9 and 9′; 10 µM, lanes 10 and 10′); lane 11, intact 3′ end-labeled
duplex in the absence of a third strand; lane 12, DNA cleavage products
produced in the absence of a third strand; lane 13, products of a A+G-
specific sequencing reaction.
Affinity of the Pyridine C-Nucleosides Containing Oligo-
nucleotides to Single-Stranded DNA. Binding of the oligo-
nucleotides 15-19 to their antiparallel Watson-Crick comple-
mentary homopurine strand d((AG)5A5) was monitored by UV-
melting curves at two different salt concentrations (200 mM
and 1 M NaCl) and pH 8.0. This pH value was chosen as to
favor the deprotonated form of the 2-aminopyridine residues
that is required for the formation of two hydrogen bonds to
guanine (Figure 9). The duplex melting transition data are
summarized in Table 3 and representative melting curves (200
mM NaCl, pH 8.0) are depicted in Figure 8. From the data in
Table 3 it appears that already the replacement of one deoxy-
cytidine residue by the pyridine C-nucleosides P, MeP, and POMe
results in a dramatic decrease (more than 12 °C) of Tm.
Consequently no cooperative melting transition could be
observed in the case where all deoxycytidines were replaced
by P (f19).
(19) A systematic quantification according to published procedures20 was
not successful due to increased DNase I activity on either edges of the
bound third strand, which may be due to sequence composition effects still
not fully understood.
(20) Brenowitz, M.; Senear, D. F.; Shea, M. A.; Ackers, G. K. Methods
Enzymol. 1986, 130, 132-181.

5504 J. Am. Chem. Soc., Vol. 119, No. 24, 1997
Hildbrand et al.
which a higher Kass would be expected. The differences may
arise from several factors, as e.g. constant- vs variable-
temperature experiments, differences in buffer conditions,
secondary interactions of oligonucleotides with DNase I, and
sequence length effects (21 vs 229 base pairs of the target
duplex). Although it would be in general desirable to be able
to correlate results from UV-melting experiments and DNase
footprint experiments this seems not to be possible so far due
to limited data available.
Effect of Enhanced pKa on Triple-Helix Stability. From
the Tm data in Table 1 it appears that the replacement of the
cytidine (or 5-methylcytidine) residues in the third strand by
the 2-amino-2′-deoxypyridine C-nucleosides P or MeP, which
are by ∼2 pKa units more basic than cytidine, results in an
enhanced affinity of these oligonucleotides to double-stranded
DNA over a pH range of 6.0-8.0. Not only this increase in
triple-helix stability but also the much less expressed pH
dependence of the Tm values (in the pH range 6.0-8.0) for MeP,
and P-containing oligonucleotides compared to MeC-containing
oligonucleotides can be regarded as a direct consequence of
the higher pKa. This is especially pronounced for the sequence
containing contiguous C units (Figure 6, bottom). Interestingly
the replacement of P by MeP (which is by 0.4 pKa units more
basic than P) results in no further enhancement of triple-helix
stability at neutral pH. It seems that the effect of the pKa
difference between MeP and P becomes only operative at pH
values > 8.0.
Figure 8. UV-melting curves (λ ) 260 nm) of 1:1 mixtures of 15 (a),
16 (b) and 19 (c) with the Watson-Crick complement d((AG)5A5) in
10 mM NaH2PO4, 200 mM NaCl at pH 8.0 (concentration 2.2-2.4
µM).
Effect of the Methyl Group on Triple-Helix Stability. It
is well documented that replacement of cytosine by 5-methyl-
cytosine in the third strand extends the pH range of stable triple
helix formation in the parallel binding motif although there exists
no relevant difference in the pKa of the two bases.22
A
thermodynamic analysis (differential scanning calorimetry)
indicated that this enhanced stability imparted by 5-methylcy-
tidine appears to be entropic in origin. A given explanation is
that the methyl group fills a space in the major groove, causing
a release of hydrating water molecules from the double helix
to the bulk upon triplex formation.22c In contrast to this,
oligonucleotides containing P or MeP residues show no differ-
ences in triplex stability as inferred from the almost equal Tm
values for 19 and 20, thus indicating that the “methyl effect” is
not present or operative in this system. Since the two pairs C,
MeC and P, MeP structurally diverge only in the presence or
absence of a carbonyl group on the complexing side (opposite
to the methyl groups) of the base, one is forced to reconsider
the interpretation for the observed entropic effect in the former
system. An alternative interpretation, consistent with the
observed behaviour in both cases, would assign an indirect role
to the methyl group in MeC, in that it favorably changes the
hydration pattern of the carbonyl group by influencing the
torsion around the nucleosidic bond. Clearly, this is speculative
at present and more experiments directed to the determination
of thermodynamic data with P- and MeP-containing oligonu-
cleotides are necessary.
Figure 9. (top) Structures of the Watson-Crick base pairs adenine-
thymine (A-T) and guanine-cytosine (G-C). (bottom) Structures of
the 2-aminopyridine-guanine (P-G; expected steric repulsion between
H-N(2) of G and H-C(6) of P indicated) and the 2-aminopyridine-
inosine (P-I) base pairs.
Table 3. Tm Values (°C) of the Pyridine C-Nucleoside-Containing
Duplexes as Determined from UV-Melting Curves (λ ) 260 nm)a
duplex
200 mM NaCl 1 M NaCl
d((AG)5A5)/d(T5(CT)5) (15)
48.7
36.5
31.2
36.9
b
56.3
43.9
40.3
46.4
b
d((AG)5A5)/d(T5CTPT(CT)3) (16)
d((AG)5A5)/d(T5CTMePT(CT)3) (17)
d((AG)5A5)/d(T5CTPOMeT(CT)3) (18)
d((AG)5A5)/d(T5(PT)5) (19)
a Duplex concentration: 2.2-2.4 µM in 10 mM NaH2PO4; pH 8.0.
b No cooperative melting transition was observed.
Discussion
Tm vs DNase I Footprint Results. In contrast to duplex
formation between two single strands of comparable length and
sequence, where as a rule of thumb an increase in Tm by 3-5
°C corresponds to an increase in the association constant (Kass
)
Effect of the 2′-O-Methyl Group on Triple-Helix Stability.
It is known that triplexes formed with oligo(2′-O-methylribo-
nucleotides) have increased affinity towards duplex DNA
relative to oligo(2′-deoxynucleotides).23 This enhanced stability
by a factor of 10,21 there exists no clear correlation between Tm
differences and Kass in the case of triple-helix formation. From
the data in Table 1 it appears that the replacement of all five
MeC residues by MeP results in an increase of Tm by ∼10 °C (at
pH 7.2). On the other hand from the DNase I footprint results
(Figure 7) it can be seen that the binding affinity of the MeP-
containing oligonucleotide 20 is only five times stronger to the
target than that of the MeC-containing oligonucleotide 23 and
thus seems to contrast the results of the Tm measurements from
(22) (a) Lee, J. S.; Woodsworth, M. L.; Latimer, L. J. P.; Morgan, A. R.
Nucleic Acids Res. 1984, 12, 6603-6614. (b) Povsic, T. J.; Dervan, P. B.
J. Am. Chem. Soc. 1989, 111, 3059-3061. (c) Xodo, L. E.; Manzini, G.;
Quadrifoglio, F.; van der Marel, G. A.; van Boom, J. H. Nucleic Acids
Res. 1991, 19, 5625-5631. (d) Plum, G. E.; Park, Y.-W.; Singleton, S. F.;
Dervan, P. B.; Breslauer, K. J. Proc. Natl. Acad. Sci. USA. 1990, 87, 9436-
9440.
(21) De Mesmaeker, A.; Ha¨ner, R.; Martin, P.; Moser, H. E. Acc. Chem.
Res. 1995, 28, 366-374.
(23) Shimizu, M.; Konishi, A.; Shimada, Y.; Inoue, H.; Ohtsuka, E. FEBS
Lett. 1992, 302, 155-158.

5-Substituted 2-Aminopyridine C-Nucleosides
J. Am. Chem. Soc., Vol. 119, No. 24, 1997 5505
is most probably due to the conformationally uniform sugar
structure of the ribose units. By replacing P by POMe in the
third strand no such stabilizing effect could be observed. The
P f POMe replacement resulted in a distinct loss of triplex
stability although there seems to be no relevant difference in
the pKa of these two nucleosides.24 While the stabilizing effect
of 2′-O-alkylated nucleosides was reported for oligonucleotides
with a uniform ribose backbone there exists only limited data
in the literature with oligonucleotides composed of both deoxy-
and 2′-O-alkyl ribose units (heterogeneous backbone com-
position).3a,f,5 However a clear comparison of the influence of
partial sugar modification on thermal stability has not been done
so far. Our data indicate that heterogeneous oligonucleotides
may behave differently from their pure ribo and deoxyribo
sequences, probably due to differences in the preferred confor-
mation of the third strand or the triplex itself.
disappear in order to be able to form a stable base pair.
Therefore one is tempted to predict the existence of a stable
2-aminopyridine-inosine (P-I) base pair (Figure 9).
Conclusion
The results presented here clearly demonstrate that the easily
accessible pyridine C-nucleosides P and MeP, due to their
enhanced basicity relative to natural cytidine are excellent
replacements for MeC in recognition of double-stranded DNA
at physiological pH. The superiority of P and MeP over MeC
can be summarized as follows: (i) No self-association of
oligonucleotides containing P and MeP could be detected at low
pH. (ii) Oligonucleotides containing P or MeP have a higher
affinity to double-stranded DNA over the pH range of 6-8 than
MeC-containing oligonucleotides. (iii) The Hoogsteen associa-
tion of the C-nucleosides P and MeP is more specific than for
MeC. (iv) Sequence composition effects are much less pro-
nounced for MeP than for MeC. (v) Contrary to MeC there is no
stable Watson-Crick association mode to G possible with P
and MeP.
Thus pyrimidine oligonucleotides containing the 2′-deoxy-
pyridine C-nucleosides P and MeP are structurally preorganized
to bind exclusively in the Hoogsteen association mode. This
extreme selectivity in favor of triplex formation, together with
their easy accessibility and the wide range of applicability in
terms of base sequence composition within the parallel binding
motif are the key features of this new cytidine analog.
Sequence Composition Effects. Recognition of double-
stranded DNA by homopyrimidine oligonucleotides containing
charged cytidine or MeC residues is known to have serious
sequence composition limitations with regard to targeting
contiguous G-rich purine tracts at physiological pH.18 It is
believed that this destabilization is due to an intrastrand
electrostatic repulsion between adjacent protonated C or MeC
units in the third strand. From Table 2 (and Figure 6, bottom)
it clearly appears that this destabilizing sequence effect is much
less pronounced in oligonucleotides containing MeP compared
to those containing MeC. This indicates that at neutral pH,
oligonucleotides containing bases with intrinsically more basic
heterocycles suffer less from intrastrand charge repulsion and
take more energetic advantage from the reduced interstrand
charge repulsion in the triplex.
Experimental Section
General Considerations. 1H NMR (300 or 500 MHz respectively)
spectra and NOE (500 MHz) experiments were recorded on Bruker
AC-300 or DRX500 and 13C NMR (75 MHz) on Bruker AC-300. 31P
NMR (202 MHz) spectra were measured on Bruker DRX500 using
PPh3 () 0 ppm) as external standard. Infrared (IR) spectra were
obtained using a Perkin-Elmer (782 or FT IR 1600) spectrometer.
Optical rotations were determined at room temperature using a Perkin-
Elmer 241 polarimeter. Melting points (mp) were recorded on Bu¨chi
510. UV spectra were performed on a Cary 3E UV/vis spectropho-
tometer (Varian). Electron ionization (EI) mass spectral analyses were
recorded on Varian MAT (CH-7A or 212) and fast-atom bombardment
(FAB) mass spectra on AutoSpec Q VG. pKa values were obtained
by potentiometric titration at 25 °C with 0.1 M aqueous NaOH or HCl
and were measured at Ciba-Geigy AG, Basel. Flash chromatography
was performed using silica gel with an average particle size of 40 µm
(Baker). Thin-layer chromatography was performed on ALUGRAM®
SIL G/UV254 (Macherey-Nagel). Visualization by UV (254 nm) and/
or by dipping into a solution of 10.5 g of cerium(IV) sulfate, 21 g of
phosphomolybdic acid, 60 mL of concentrated sulfuric acid, and 900
mL of H2O, followed by heating with a heat gun. All reactions were
carried out under argon, using purified and distilled solvents. Tetra-
n-butylammonium fluoride (1.1 M in THF) was from Aldrich and
2-cyanoethyl N,N-diisopropylchlorophosphoramidite from Sigma. All
other reagents were obtained from Fluka (butyllithium (nBuLi) as a
1.6 M solution in hexane) and were used without further purification.
1-Aza-1-(5-bromo-2-pyridinyl)-2,2,5,5-tetrametyl-2,5-disilacyclo-
pentane (2a). To a solution of 2-amino-5-bromopyridine 1a (5.00 g,
28.9 mmol) in THF (80 mL) was added at -75 °C nBuLi (18.2 mL,
29.0 mmol). After 1 h at -75 °C a solution of 1,2-bis(chlorodimeth-
ylsilyl)ethane (6.22 g, 28.90 mmol) in THF (15 mL) was added
dropwise. After another 90 min at -75 °C nBuLi (18.2 mL, 29.0
mmol) was added and the mixture was warmed up to room temperature
within 2 h and allowed to stir for additional 2 h at room temperature.
Brine (50 mL) was added, and the mixture extracted with Et2O (2 ×
200 mL). The ether extracts were dried (MgSO4) and concentrated.
Kugelrohr distillation (150 °C/0.05 mbar) afforded 6.20 g (68%) of
Effect of the Missing Carbonyl GroupsSelectivity in
Triple-Helix Formation. While for obvious reasons the 2-oxo
function of C and MeC is not involved in H-bond formation with
its target guanine, it is unclear whether it has any significant
influence on triplex stability and binding selectivity at all. The
bases P and MeP represent excellent “deletion mutants” to assess
the function of the 2-oxo group experimentally. From the data
in Table 1 and 2 it can be deduced that the absence of this
functional group has no vital effect on triplex stability. The
DNase I footprinting studies, however, demonstrating higher
binding selectivity of P- over C-containing oligonucleotides,
underline the important role of this carbonyl group in secondary
target site binding, ultimately being responsible for reduced
pairing selectivity. The fact, that 23 has a much higher affinity
to the AT containing cassette than 20, specifically indicates that
the 2-oxo function of cytidine is involved in H-bonding to
adenine in this sequence context.
Watson-Crick Base-Pairing Properties. In its deproto-
nated form the base 2-aminopyridine can be considered as a
cytosine equivalent that should still be able to bind to guanine
via two hydrogen bonds (Figure 9). On the basis of the duplex
sequence investigated it could be shown that these duplexes
are distinctly less stable than the corresponding natural duplex
(Table 3). Thus the incorporation of P, MeP, or POMe in
oligonucleotides has nearly the same destabilizing consequences
as is expected for a mismatch base pair. This dramatic duplex
destabilization may be ascribed to steric repulsion between the
2-amino group of guanosine and the H-C(6) of P (violation of
the van der Waals radii of the two H-atoms in the base pair;
Figure 9).25 This result suggests that if the 2-oxo function of
cytosine is lost, the 2-amino group of guanine has also to
(24) It can be expected that the 2′-O-methylated derivative POMe displays
approximately the same pKa value than 2-amino-5-(â-D-ribofuranosyl)-
pyridine (pKa ) 6.15).7
(25) Although there exists a possibility to escape the steric repulsion by
propeller twisting of the base pair this would most probably not be tolerated
due to steric clash with the base pairs above and below in the stack.

5506 J. Am. Chem. Soc., Vol. 119, No. 24, 1997
Hildbrand et al.
the stabase adduct 2a as a colorless solid, mp 64-66 °C. 1H NMR
(300 MHz, CDCl3): δ 0.29 (s, 12 H), 0.82 (s, 4 H), 6.46 (dd, J )
8.45, 0.75 Hz, 1 H), 7.46 (dd, J ) 8.85, 2.55 Hz, 1 H), 8.14 (dd, J )
2.6, 0.8 Hz, 1 H). 13C NMR (75 MHz, CDCl3): δ -0.58 (4×), 8.54
(2×), 108.52, 113.26, 139.29, 148.67, 159.43. IR (film): 1465, 1370,
1310, 1250, 950, 815, 785 cm-1. MS (EI): m/z (rel intensity) 317
(19, [M + 1]+), 316 (77, M+), 315 (64, [M + 1]+), 314 (77, M+), 288
(87), 286 (83), 273 (100), 271 (97).
1-Aza-1-(5-bromo-3-methyl-2-pyridinyl)-2,2,5,5-tetramethyl-2,5-
disilacyclopentane (2b). To 2-amino-5-bromo-3-methylpyridine 1b
(2.22 g, 11.9 mmol) and ZnI2 (26 mg, 0.08 mmol) was added dropwise
1,2-bis[(dimethylamino)dimethylsilyl]ethane (3.30 mL, 11.9 mmol) and
the mixture was heated at 140 °C over night. Kugelrohr distillation
(170 °C/0.02 mbar) of the reaction mixture yielded 2.70 g (69%) of
2b as a colorless oil. 1H NMR (300 MHz, CDCl3): δ 0.16 (s, 12 H),
0.88 (s, 4 H), 2.23 (s, 3 H), 7.53 (dd, J ) 2.6, 0.74 Hz, 1 H), 8.18 (dd,
J ) 2.6, 0.74 Hz, 1 H). 13C NMR (75 MHz, CDCl3): δ 0.66 (4×),
9.00 (2×), 19.39, 113.02, 129.12, 141.02, 146.38, 157.93. IR (film):
3040, 2952, 2915, 2891, 1575, 1545, 1461, 1416, 1293, 1273, 1251,
1154, 1144, 956, 927, 898, 886, 866, 812, 786, 663 cm-1. MS (EI):
m/z (rel intensity) 330 (36, M+), 329 (13), 328 (31, M+), 315 (100),
313 (95), 302 (57), 300 (53), 287 (83), 285 (82).
128.40, 136.08, 137.71, 137.98, 138.13, 144.31, 156.98. IR (film):
3482, 3378, 3029, 2864, 1621, 1486, 1454, 1125, 1091, 1048, 1028,
738, 698 cm-1. MS (EI): m/z (rel intensity) 510 (10, M+), 419 (26),
153 (31), 227 (50), 190 (40), 135 (46), 92 (42), 91 (100).
2-(N-Benzoylamino)-5-(â-D-ribofuranosyl)pyridine (5a). To an
ice-cooled solution of 4a (1.60 g, 3.22 mmol) and pyridine (0.30 mL)
in CH2Cl2 (8 mL) was added benzoyl chloride (0.50 mL, 4.30 mmol)
and the resulting mixture was stirred for 24 h at room temperature.
HCl (20 mL, 1M) and H2O (20 mL) was added and the mixture was
extracted with CH2Cl2 (5 × 50 mL). The organic layer was washed
with 2% NaOH (20 mL), dried over MgSO4, and concentrated. The
residue was filtered over silica gel using hexane/ethyl acetate 20:11 to
afford a mixture (1.85 g) of mono- and dibenzoylated nucleoside (Rf
0.29 and 0.19) as a colorless solid, which was dissolved in CH2Cl2 (35
mL), cooled to -75 °C, and treated with a 1 M solution of BBr3 in
CH2Cl2 (13.0 mL). The resulting suspension was stirred for 4 h at
-75 °C, quenched with MeOH (40 mL) and allowed to warm up to
room temperature over night. The solvent was evaporated, the residue
dissolved in H2O (20 mL) and washed with CH2Cl2 (10 mL). NaOH
(10%) was added to make the solution basic. A colorless solid
precipitated slowly at 4 °C, which was filtered off after 48 h and dried
to yield 0.75 g (70%) of C-nucleoside 5a as a colorless solid, mp 192-
194 °C. [R]D ) -56.9 (c ) 0.93, MeOH/0.5 M HCl 7:3). 1H NMR
(300 MHz, CD3OD): δ 3.77 (dd, J ) 11.9, 4.6 Hz, 1 H), 3.85 (dd, J
) 11.9, 3.45 Hz, 1 H), 3.95 (dd, J ) 7.3, 5.5 Hz, 1 H), 4.05 (q, J ≈
3.8 Hz, 1 H), 4.10-4.18 (m, 1 H), 4.78 (d, J ) 7.0 Hz, 1 H), 7.50-
7.68 (m, 3 H), 7.93-8.06 (m, 3 H), 8.25 (d, J ) 8.4 Hz, 1 H), 8.46 (d,
J ) 2.2 Hz, 1 H). 13C NMR (75 MHz, CD3OD): δ 63.86, 73.36,
79.34, 83.07, 87.29, 116.14, 129.01, 130.05, 133.64, 134.36, 135.88,
137.99, 147.64, 153.08, 168.83. IR (KBr): 3260, 1660, 1522, 1530,
1490, 1393, 1310, 1105 cm-1. MS (EI): m/z (rel intensity) 331 (11,
[M + 1]+) 330 (29, M+), 302 (61), 301 (100), 227 (38), 123 (26), 105
(72), 77 (31).
3-Methyl-2-[N-(phenoxyacetyl)amino]-5-(â-D-ribofuranosyl)py-
ridine (5b). A solution of C-nucleoside 4b (4.09 g, 8.02 mmol) and
phenoxyacetic anhydride (6.89 g, 24.0 mmol) in pyridine (40 mL) was
stirred at room temperature for 4.5 h. The reaction mixture was
quenched with H2O (10 mL) and concentrated. The residue was diluted
with ethyl acetate (250 mL), washed with 1 M Na2CO3 (200 mL), dried
(MgSO4), and concentrated. Flash chromatography (hexane/ethyl
acetate 1:1) of the residue afforded 3.45 g (67%) of 3-methyl-2-(N-
phenoxyacetylamino)-5-(2′,3′,5′-tri-O-benzyl-â-D-ribofuranosyl)pyri-
dine as a yellow oil, Rf 0.33 (hexane/ethyl acetate 1:1). [R]D ) -21.9
(c ) 1.72, CHCl3). 1H NMR (300 MHz, CDCl3): δ 2.16 (s, 3 H),
3.62 (dd, J ) 10.3, 3.68 Hz, 1 H), 3.70 (dd, J ) 10.3, 4.05 Hz, 1 H),
3.83 (dd, J ) 7.0, 5.15 Hz, 1 H), 4.06 (dd, J ) 5.14, 4.04 Hz, 1 H),
4.38 (q, J ) 3.68 Hz, 1 H), 4.42-4.66 (m, 6 H), 4.70 (s, 2 H), 5.04 (d,
J ) 6.98 Hz, 1 H), 6.98-7.10 (m, 3 H), 7.18-7.23 (m, 2 H), 7.25-
7.39 (m, 15 H), 7.60 (d, J ) 2.21 Hz, 1 H), 8.31 (d, J ) 2.21, 1 H),
8.61 (broad, NH). 13C NMR (75 MHz, CDCl3): δ 17.81, 67.46, 70.26,
71.95, 72.43, 73.45, 77.23, 79.82, 82.04, 83.43, 114.66, 122.23, 127.51,
127.66, 127.81, 127.88, 128.02, 128.32, 128.36, 128.38, 129.78, 134.41,
137.35, 137.62, 137.71, 137.89, 144.29, 147.78, 157.03, 166.58. IR
(film): 3246, 2918, 2864, 1694, 1599, 1495, 1454, 1128, 1084, 753,
737, 698 cm-1. MS (EI): m/z (rel intensity) 644 (12, M+), 227 (41),
193 (75), 152 (88), 122 (87), 107 (98), 79 (76), 65 (81), 55 (73), 51
(100).
2-Amino-5-(2′,3′,5′-tri-O-benzyl-â-D-ribofuranosyl)pyridine (4a).
To a solution of 2a (4.07 g, 12.91 mmol) in THF (60 mL) was added
at -75 °C nBuLi (7.6 mL, 12.2 mmol). After 1 h at -75 °C a solution
of lactone 311 (3.00 g, 7.17 mmol) in THF (20 mL) was added at -75
°C. The mixture was stirred for 2 h at -75 °C and then allowed to
warm up to 0 °C over 3 h. Saturated NaHCO3 (75 mL) was added
and the mixture extracted with Et2O (4 × 100 mL). The organic layer
was washed with brine (50 mL), dried over MgSO4, and concentrated.
The yellow residue was dried and dissolved in CH2Cl2 (20 mL), and
triethylsilane (5.7 mL, 36 mmol) and BF3‚OEt2 (4.5 mL, 36 mmol)
were added dropwise at -75 °C. The reaction mixture was allowed
to warm up to room temperature over night, quenched with 1 M HCl
(20 mL) and stirred for 1 h at room temperature. The mixture was
neutralized with 2% NaOH and extracted with ethyl acetate (5 × 100
mL). The organic layer was washed with brine (30 mL), dried over
MgSO4 and concentrated. Purification of the residue by flash chro-
matography (ethyl acetate) afforded 2.28 g (64%) of 4a as a slightly
yellow oil, Rf 0.35 (ethyl acetate). [R]D ) -29.3 (c ) 1.35, CHCl3).
1H NMR (300 MHz, CDCl3): δ 3.56 (dd, J ) 10.3, 4.1 Hz, 1 H), 3.61
(dd, J ) 10.3, 4.0 Hz, 1 H), 3.70 (s, broad, NH2), 3.75 (dd, J ) 7.5,
5.1 Hz, 1 H), 3.99 (dd, J ) 5.0, 3.3 Hz, 1 H), 4.29 (q, J ≈ 3.7 Hz, 1
H), 4.45 (AB system, JAB ) 12.0 Hz, 2 H), 4.54 (AB system, JAB
)
12.0 Hz, 2 H), 4.61 (AB system, JAB ) 12.5 Hz, 2 H), 4.86 (d, J ) 7.4
Hz, 1 H), 6.40 (d, J ) 8.8 Hz, 1 H), 7.13-7.38 (m, 15 H), 7.45 (dd,
J ) 8.5, 2.25 Hz, 1 H), 7.95 (d, J ) 2.2 Hz, 1 H). NOE (500 MHz,
CDCl3): irradiation at δ 3.75 (H-C(2′)) produced NOE enhancements
at δ 3.99 (H-C(3′)), 4.86 (H-C(1′)), 7.45 (H-C(4)), 7.95 ((H-C(6));
irradiation at δ 3.99 (H-C(3′)) produced NOE enhancements at δ 3.75
(H-C(2′)), 7.45 (H-C(4)), 7.95 ((H-C(6)). 13C NMR (75 MHz,
CDCl3): δ 70.53, 71.91, 72.33, 73.47, 77.49, 80.40, 81.92, 83.29,
108.51, 125.44, 127.60, 127.66, 127.72, 127.80, 128.11, 128.32, 128.39,
128.41, 136.34, 137.66, 137.90, 138.05, 146.43, 158.30. IR (film):
3360, 3030, 2900, 2870, 1620, 1510, 1500, 1455, 1125, 1085, 1045,
1030, 785, 698 cm-1. UV (CHCl3): λmax nm 243, 295. MS (EI): m/z
(rel intensity) 496 (1, M+), 213 (21), 121 (20), 92 (20), 91 (100), 45
(26), 43 (61), 19 (50).
To a solution of this protected C-nucleoside (3.40 g, 5.27 mmol) in
CH2Cl2 (60 mL) was added dropwise at -75 °C a 1 M solution of
BBr3 in CH2Cl2 (24 mL). After 5 h at -75 °C, MeOH (75 mL) was
added and the mixture was allowed to warm up to room temperature
over night. The mixture was concentrated and the residue diluted with
H2O (40 mL). The aqueous layer was washed with CH2Cl2 (15 mL)
and neutralized with 10% NaOH. A colorless solid precipitated slowly
at 4 °C, which was filtered off and dried to yield 1.60 g (81%) of 5b,
mp 170-171 °C. [R]D ) -34.5 (c ) 0.37, MeOH). 1H NMR (300
MHz, CD3OD): δ 2.32 (s, 3 H), 3.79 (dd, J ) 12.14, 4.42 Hz, 1 H),
3.87 (dd, J ) 12.13, 3.67 Hz, 1 H), 3.99 (dd, J ) 7.35, 5.51 Hz, 1 H),
4.10 (q, J ≈ 3.9 Hz, 1 H), 4.17 (dd, J ) 5.52, 3.31 Hz, 1 H), 4.83 (s,
2 H), 4.86 (d, J ) 7.35 Hz, 1 H), 7.05-7.15 (m, 3 H), 7.35-7.41 (m,
2 H), 7.93 (s, 1 H), 8.41 (s, 1 H). 13C NMR (75 MHz, CD3OD): δ
17.68, 62.14, 66.88, 71.67, 77.68, 80.60, 85.77, 114.86, 121.36, 128.56,
2-Amino-3-methyl-5-(2′,3′,5′-tri-O-benzyl-â-D-ribofuranosyl)py-
ridine (4b). Nucleoside 4b was prepared from 2b (2.50 g, 7.59 mmol),
lactone 3 (2.10 g, 5.02 mmol), and nBuLi (4.8 mL, 7.7 mmol) as
described for 4a. Flash chromatography (ethyl acetate) afforded 1.97
g (77%) of 4b as a slightly yellow oil, Rf 0.36 (ethyl acetate). [R]D
)
-26.7 (c ) 0.89, CHCl3). 1H NMR (300 MHz, CDCl3): δ 2.00 (s, 3
H), 3.60 (dd, J ) 10.3, 3.68 Hz, 1 H), 3.66 (dd, J ) 10.3, 4.05 Hz, 1
H), 3.82 (dd, J ) 7.35, 5.13 Hz, 1 H), 4.04 (dd, J ) 5.13, 3.66 Hz, 1
H), 4.31 (q, J ≈ 3.66 Hz, 1 H), 4.42 (broad, NH2), 4.48 (AB system,
JAB ) 12.5 Hz, 2 H), 4.57 (AB system, JAB ) 4.8 Hz, 2 H), 4.64 (AB
system, JAB ) 3.3 Hz, 2 H), 4.90 (d, J ) 7.35 Hz, 1 H), 7.15-7.39
(m, 16 H), 7.95 (d, J ) 2.22 Hz, 1 H). 13C NMR (75 MHz, CDCl3):
δ 17.00, 70.59, 71.95, 72.30, 73.48, 77.54, 80.50, 81.87, 83.14, 116.46,
125.97, 127.52, 127.63, 127.66, 127.77, 127.86, 128.06, 128.26, 128.39,

5-Substituted 2-Aminopyridine C-Nucleosides
J. Am. Chem. Soc., Vol. 119, No. 24, 1997 5507
129.66, 135.17, 137.32, 144.17, 148.87, 157.99, 167.23. IR (KBr):
3390, 2921, 1716, 1640, 1574, 1490, 1436, 1172, 1115, 1075, 760,
692 cm-1. MS (EI): m/z (rel intensity) 374 (20, M+), 187 (33), 137
(100), 136 (42), 107 (64), 94 (39), 77 (70).
added, and the aqueous layer was extracted with CH2Cl2 (5 × 50 mL).
The organic extracts were dried over MgSO4 and concentrated. Flash
chromatography (hexane/ethyl acetate 5:3) gave 0.97 g (75%) of 6a as
a colorless foam, Rf 0.35 (hexane/ethyl acetate 5:3). [R]D ) -46.0 (c
) 2.05, CHCl3). 1H NMR (300 MHz, CDCl3): δ 1.00-1.15 (m, 28
H), 3.11 (d, J ) 4.0 Hz, OH), 3.90-3.97 (m, 1 H), 3.99-4.16 (m, 3
H), 4.38 (t, J ) 6.25 Hz, 1 H), 4.81 (d, J ) 4.0 Hz, 1 H), 7.44-7.60
(m, 3 H), 7.81 (dd, J ) 8.8, 2.2 Hz, 1 H), 7.91-7.97 (m, 2 H), 8.28 (d,
J ) 2.2 Hz, 1 H), 8.39 (d, J ) 8.1, 1 H), 8.93 (s, NH). 13C NMR (75
MHz, CDCl3): δ 12.67, 12.92, 13.21, 13.43, 16.97, 16.98, 17.08, 17.14,
17.32, 17.36, 17.39, 17.49, 62.46, 71.71, 77.07, 82.91, 83.16, 113.87,
127.28, 128.83, 131.73, 132.25, 134.26, 136.23, 145.64, 151.28, 165.73.
IR (film): 3350, 2940, 2870, 1680, 1525, 1497, 1300, 1125, 1040,
700 cm-1. MS (EI): m/z (rel intensity) 572 (4, M+), 530 (34), 529
(100), 528 (45), 236 (16), 235 (70), 224 (14), 105 (16).
2-O-Methyl-3,5-O-(1,1,3,3-tetraisopropyldisiloxane-1,3-diyl)-D-ri-
bono-1,4-lactone (11). To a solution of D-ribono-1,4-lactone (10) (1.20
g, 8.10 mmol) in pyridine (18 mL) was added dropwise at room
temperature 1,3-dichloro-1,1,3,3-tetraisopropyldisiloxane (3.10 mL, 9.89
mmol) and the resulting solution was stirred for 5 h. Ethyl acetate
(100 mL) was added and the mixture washed with 0.1 M HCl (2 × 50
mL). The aqueous layer was extracted with ethyl acetate (2 × 50 mL),
the combined organic layers were washed with saturated NaHCO3 (1
× 50 mL), dried (MgSO4), and concentrated. Flash chromatography
(hexane/ethyl acetate 5:1) afforded 1.23 g (39%) of 3,5-O-(1,1,3,3-
tetraisopropyldisiloxane-1,3-diyl)-D-ribono-1,4-lactone as a colorless oil,
Rf 0.30 (hexane/ethyl acetate 5:1). [R]D ) +28.9 (c ) 0.5, CHCl3).
1H NMR (300 MHz, CDCl3): δ 1.00-1.13 (m, 28 H), 3.02 (d, J )
2.6, OH), 3.99 (dd, J ) 12.5, 5.9 Hz, 1 H), 4.15 (dd, J ) 12.5, 3.7 Hz,
1 H), 4.25 (dd, J ) 5.9, 2.2 Hz, 1 H), 4.43 (td, J ≈ 6.4, 3.7 Hz, 1 H),
4.51 (dd, J ) 6.6, 5.9 Hz, 1 H). 13C NMR (75 MHz, CDCl3): δ 12.53,
12.81, 13.07, 13.25, 16.75, 16.84, 16.86, 17.02, 17.17, 17.18, 17.22,
17.34, 61.60, 68.45, 69.83, 82.64, 171.74. IR (film): 3500, 2945, 2870,
1792, 1090, 1040, 905, 885, 730 cm-1. MS (EI): m/z (rel intensity)
389 (5, [M - 1]+), 329 (32), 235 (44), 121 (47), 77 (39), 58 (96), 41
(32), 28 (100).
A suspension of this oil (1.05 g, 2.69 mmol) and silver(I) oxide
(2.10 g, 9.06 mmol) in methyl iodide (17 mL) was heated at 45 °C for
6 h. The excess MeI was distilled off, the residue diluted with ethyl
acetate and filtered through a pad of celite. The filtrate was
concentrated and the residue chromatographed (hexane/ethyl acetate
10:1) to yield 0.73 g (67%) of 11 as a clear colorless oil, which
solidified upon standing at -30 °C, Rf 0.30 (hexane/ethyl acetate 10:
1), mp 38-39 °C. [R]D ) +43.9 (c ) 0.54, CHCl3). 1H NMR (300
MHz, CDCl3): δ 0.97-1.10 (m, 28 H), 3.59 (s, 3 H), 3.78 (d, J ≈ 4.8
Hz, 1 H), 4.00 (dd, J ) 13.6, 2.6 Hz, 1 H), 4.13 (dd, J ) 13.6, 1.85
Hz, 1 H), 4.37 (dd, J ) 8.65, 4.6 Hz, 1 H), 4.43 (dt, J ) 8.85, 2.2 Hz,
1 H). 13C NMR (75 MHz, CDCl3): δ 12.47, 12.74, 12.94, 13.37, 16.76,
16.91, 17.01, 17.09, 17.16, 17.18, 17.32, 59.13, 59.29, 69.34, 77.69,
82.14, 170.60. IR (film): 2940, 2895, 2865, 1792, 1465, 1245, 1162,
1130, 1093, 1040, 980, 882, 695 cm-1. MS (EI): m/z (rel intensity)
404 (1, M+), 363 (58), 362 (79), 361 (97), 329 (73), 259 (70), 250
(73), 249 (100), 135 (73), 119 (72), 83 (83), 43 (71).
2-Amino-5-[2′-O-methyl-3′,5′-(1,1,3,3-tetraisopropyldisiloxane-
1,3-diyl)-r,â-D-ribofuranosyl]pyridine (12). C-Nucleoside 12 was
prepared from protected lactone 11 (0.70 g, 1.73 mmol) and 2a (1.12
g, 3.56 mmol) as already described for 4a. Flash chromatography (ethyl
acetate) afforded 0.51 g (61%) of coupling product 12 as an anomeric
mixture (â/R ≈ 10:1) as a slightly yellow oil, Rf 0.36 (ethyl acetate).
1H NMR (300 MHz, CDCl3): δ â-anomer: 0.90-1.12 (m, 28 H), 3.50
(dd, J ) 5.0, 2.0 Hz, 1 H), 3.53 (s, 3 H); 4.92-4.14 (m, 3 H), 4.0
(broad, NH2), 4.32 (dd, J ) 8.8, 5.15 Hz, 1 H), 4.80 (d, J ) 2.2 Hz,
1 H), 6.45 (d, J ) 8.45 Hz, 1 H), 7.51 (dd, J ) 8.45, 2.2 Hz, 1 H),
8.00 (d, J ) 2.55 Hz, 1 H); R-anomer: 0.90-1.12 (m, 28 H), 3.05
(dd, J ) 9.55, 2.55 Hz, 1 H); 3.07 (s, 3 H), 3.71 (d, J ) 8.1 Hz, 2 H),
4.00-4.10 (m, 1 H), 4.0 (broad, NH2), 4.41 (d, J ) 9.55 Hz, 1 H),
4.63 (t, J ) 2.2 Hz, 1 H), 6.47 (d, J ) 8.45 Hz, 1 H), 7.44 (dd, J )
8.65, 2.4 Hz, 1 H), 8.04 (d, J ) 2.2 Hz, 1 H). 13C NMR (75 MHz,
CDCl3): δ â-anomer, 12.58, 12.85, 13.08, 13.50, 16.93, 17.07, 17.13,
17.20, 17.32, 17.43, 17.36, 17.47, 58.96, 60.73, 70.95, 81.08, 82.61,
86.71, 108.44, 126.44, 135.90, 145.74, 157.96; R-anomer, 12.61, 13.18,
13.85, 14.21, 17.08, 17.20, 17.23, 17.29, 17.35, 17.45, 17.55, 17.59,
57.60, 66.07, 70.08, 72.13, 73.55, 81.80, 108.25, 125.25, 137.23, 147.09,
152.73. IR (film): 3360, 2950, 2930, 2860, 1620, 1502, 1135, 1075,
1030, 988, 883, 690 cm-1. MS (EI): m/z (rel intensity) 483 (8, [M +
1]+), 482 (15, M+), 440 (72), 439 (100), 249 (67), 247 (45), 173 (46),
135 (58), 121 (75), 119 (47), 107 (95).
3-Methyl-2-(N-phenoxyacetylamino)-5-[3′,5′-O-(1,1,3,3-tetraiso-
propyldisiloxane-1,3-diyl)-â-D-ribofuranosyl]pyridine (6b). The pro-
tected C-nucleoside 6b was prepared from 5b (1.38 g, 3.7 mmol), 1,3-
dichloro-1,1,3,3-tetraisopropyldisiloxane (1.4 mL, 4.5 mmol) and
pyridine (30 mL) as described for 6a. Flash chromatography (hexane/
ethyl acetate 1:1) gave 1.90 g (84%) of 6b as a colorless foam; Rf 0.34
(hexane/ethyl acetate 1:1). [R]D ) -32.3 (c ) 0.39, CHCl3). 1H NMR
(300 MHz, CDCl3): δ 1.03-1.11 (m, 28 H), 2.28 (s, 3 H), 3.00 (d, J
) 3.68 Hz, OH), 3.91-3.96 (m, 1 H), 4.00-4.06 (m, 1 H), 4.07-4.14
(m, 2 H), 4.36 (t, J ) 6.25 Hz, 1 H), 4.70 (s, 2 H), 4.84 (d, J ) 4.04
Hz, 1 H), 6.98-7.09 (m, 3 H), 7.33-7.39 (m, 2 H), 7.65 (d, J ) 1.47
Hz, 1 H), 8.34 (d, J ) 1.84 Hz, 1 H), 8.55 (s, NH).
13C NMR (75
MHz, CDCl3): δ 12.65, 12.87, 13.19, 13.41, 16.97, 16.99, 17.09, 17.15,
17.35, 17.40, 17.48, 18.00, 62.25, 67.51, 71.53, 76.98, 82.73, 83.00,
114.73, 122.33, 126.13, 129.87, 134.13, 137.37, 143.98, 147.85, 157.08,
166.73. IR (film): 3400, 2944, 2867, 1700, 1684, 1600, 1506, 1496,
1464, 1243, 1129, 1064, 1038, 913, 885, 748, 691 cm-1. MS (EI):
m/z (rel intensity) 617 (14, [M + 1]+), 616 (22, M+), 573 (57), 524
(34), 523 (90), 483 (36), 482 (100), 235 (31), 121 (31).
2-[N-(Phenoxyacetyl)amino]-5-[2′-O-methyl-3′,5′-O-(1,1,3,3-tet-
raisopropyldisiloxane-1,3-diyl)-â-D-ribofuranosyl]pyridine (6c). A
solution of 12 (0.33 g, 0.68 mmol) and phenoxyacetic anhydride (0.70
g, 2.45 mmol) in pyridine (7 mL) was stirred at room temperature for
5 h. The reaction was quenched with H2O (10 mL) and stirred for
another 1.5 h. The mixture was concentrated, and the residue was
diluted with ethyl acetate (150 mL), washed with 0.1 M NaOH (2 ×
20 mL), dried (MgSO4), and concentrated. Flash chromatography
(hexane/ethyl acetate 4:1) of the residue afforded 0.36 g (85%) of
protected nucleoside 6c and 35 mg (8%) of the corresponding R-anomer
as colorless oils. Data of 6c (â-anomer): Rf 0.25 (hexane/ethyl acetate
4:1). [R]D ) -34.8 (c ) 1.0, CHCl3). 1H NMR (300 MHz, CDCl3):
δ 0.90-1.12 (m, 28 H); 3.52 (dd, J ) 4.95, 2.0 Hz, 1 H), 3.56 (s, 3
H), 3.97-4.05 (m, 2 H), 4.14 (dd, J ) 13.4, 3.1 Hz, 1 H), 4.34 (dd, J
) 8.3, 4.95 Hz, 1 H), 4.61 (s, 2 H), 4.92 (d, J ) 2.2 Hz, 1 H), 6.94-
7.07 (m, 3 H), 7.26-7.37 (m, 2 H), 7.80 (dd; J ) 8.5, 2.2 Hz, 1 H),
8.23 (d, J ) 8.8 Hz, 1 H), 8.31 (d, J ) 2.2 Hz, 1 H), 8.95 (s, NH).
NOE (500 MHz, CDCl3): irradiation at δ 3.52 (H-C(2′)) produced
NOE enhancements at δ 4.34 (H-C(3′)), 4.92 (H-C(1′)), 7.80 (H-
C(4)), 8.31 (H-C(6)); irradiation at δ 4.34 (H-C(3′)) produced NOE
enhancements at δ 3.52 (H-C(2′)), 7.80 (H-C(4)), 8.31 (H-C(6));
irradiation at δ 7.80 (H-C(4)) produced NOE enhancements at δ 3.52
(H-C(2′)), 4.34 (H-C(3′)), 4.92 (H-C(1′)), 8.23 (H-C(3)); irradiation
at δ 8.31 (H-C(6)) produced NOE enhancements at δ 3.52 (H-C(2′)),
4.34 (H-C(3′)), 4.92 (H-C(1′)). 13C NMR (75 MHz, CDCl3): δ 12.57,
12.83, 13.08, 13.47, 16.91, 17.05, 17.09, 17.18, 17.30, 17.34 (2×),
17.46, 59.06, 60.59, 67.39, 70.77, 81.31, 82.22, 86.73, 113.79, 114.80,
122.38, 129.83, 133.07, 136.06, 145.65, 149.91, 156.94, 166.79. IR
(film): 3300, 2938, 2868, 1518, 1495, 1463, 1390, 1300, 1240, 1140,
1035, 882, 750, 689 cm-1. MS (EI): m/z (rel intensity) 616 (15, M+),
574 (54), 573 (100), 255 (18), 249 (31). Data of R-anomer: Rf 0.42
(hexane/ethyl acetate 4:1). [R]D ) -22.7 (c ) 1.16, CHCl3). 1H NMR
(500 MHz, CDCl3): δ 0.80-1.08 (m, 28 H); 2.97 (dd, J ) 9.4, 2.0
Hz, 1 H), 3.00 (s, 3 H), 3.67-3.74 (m, 2 H), 4.03 (ddd, J ) 10.0, 6.5,
2.5 Hz, 1 H), 4.48 (d, J ) 9.4 Hz, 1 H), 4.56 (s, 2 H), 4.60 (t, J ≈
2.0 Hz, 1 H), 6.89-7.01 (m, 3 H), 7.23-7.31 (m, 2 H), 7.71 (dd,
J ) 8.7, 2.0 Hz, 1 H), 8.19 (d, J ) 8.7 Hz, 1 H), 8.26 (d, J ) 2.0 Hz,
1 H), 8.89 (s, NH). 13C NMR (75 MHz, CDCl3): δ 12.59, 13.18,
2-(N-Benzoylamino)-5-[3′,5′-O-(1,1,3,3-tetraisopropyldisiloxane-
1,3-diyl)-â-D-ribofuranosyl)]pyridine (6a). To a solution of C-
nucleoside 5a (0.75 g, 2.27 mmol) in pyridine (20 mL) was added
dropwise at 0 °C 1,3-dichloro-1,1,3,3-tetraisopropyldisiloxane (0.85 mL,
2.7 mmol) and the resulting suspension was stirred for 4 h at room
temperature. The solvent was evaporated, 1 M NaOH (20 mL) was

5508 J. Am. Chem. Soc., Vol. 119, No. 24, 1997
Hildbrand et al.
13.84, 14.21, 17.06, 17.21, 17.23, 17.27, 17.35, 17.44, 17.54, 17.57,
57.56, 66.04, 67.42, 69.85, 72.01, 73.19, 82.17, 113.44, 114.80, 122.35,
129.81, 132.13, 137.36, 147.20, 149.97, 156.98, 166.78. IR (film):
3395, 3015, 2940, 2870, 1700, 1590, 1520, 1495, 1462, 1212, 1168,
1118, 1080, 1030, 1000, 883, 750, 690, 665 cm-1. MS (EI): m/z (rel
intensity) 616(10, M+), 575(45), 574(73), 573(100), 249(70), 247(80),
94(34).
133.69, 134.28, 136.26, 145.76, 151.09, 165.66. IR (film): 3100, 2940,
2870, 1680, 1518, 1305, 1085, 1030 cm-1. MS (EI): m/z (rel intensity)
556 (17, M+), 516 (30), 515 (83), 514 (100), 370 (22), 278 (36), 105
(93), 77 (51).
3-Methyl-2-[N-(phenoxyacetyl)amino]-5-[3′,5′-O-(1,1,3,3-tetraiso-
propyldisiloxane-1,3-diyl)-2′-deoxy-â-D-ribofuranosyl]pyridine (8b).
A solution of 7b (1.23 g, 1.69 mmol), azobisisobutyronitrile (51 mg,
0.35 mmol), and tris(trimethylsilyl)silane (0.60 mL, 1.9 mmol) in
toluene (60 mL) was flushed with argon for 45 min and then heated at
80 °C for 5 h. NaOH (40 mL, 1 M) was added, and the aqueous layer
was extracted with ethyl acetate (5 × 80 mL). The organic layer was
dried over MgSO4 and concentrated. Flash chromatography (first
hexane/ethyl acetate 4:1, then 1:1) of the residue yielded 0.68 g (67%)
2-(N-Benzoylamino)-5-[3′,5′-O-(1,1,3,3-tetraisopropyldisiloxane-
1,3-diyl)-2′-O-[(p-tolyloxy)thiocarbonyl]-â-D-ribofuranosyl]pyri-
dine (7a). A solution of 6a (0.93 g, 1.62 mmol) and N,N-
dimethylaminopyridine (0.43 g, 3.52 mmol) in acetonitrile (23 mL)
was treated at room temperature with O-p-tolyl chlorothionoformate
(0.40 mL, 2.59 mmol) and the resulting yellow suspension was stirred
over 3 Å molecular sieves for 25 h. The mixture was concentrated, 1
M NaOH (30 mL) was added, and the aqueous layer was extracted
with ethyl acetate (4 × 70 mL). The organic layer was dried over
MgSO4 and concentrated, and the residue was chromatographed
(hexane/ethyl acetate 5:1) to afford 0.95 g (81%) of 7a as a colorless
foam, Rf 0.30 (hexane/ethyl acetate 5:1). 1H NMR (300 MHz,
CDCl3): δ 0.96-1.19 (m, 28 H), 2.38 (s, 3 H), 4.03-4.14 (m, 2 H),
4.18-4.28 (m, 1 H), 4.58 (dd, J ) 8.6, 5.0 Hz, 1 H), 5.23 (d, J ) 1.1
Hz, 1 H), 5.64 (dd, J ) 5.0, 1.7 Hz, 1 H), 7.00-7.06 (m, 2 H), 7.19-
7.25 (m, 2 H), 7.46-7.60 (m, 3 H), 7.90-7.95 (m, 2 H), 7.97 (dd, J
) 8.6, 2.4 Hz, 1 H), 8.41 (d, J ) 8.5 Hz, 1 H), 8.44 (d, J ) 2.2 Hz,
1 H), 8.87 (s, NH). 13C NMR (75 MHz, CDCl3): δ 12.77, 12.91, 13.08,
13.42, 16.96, 17.06, 17.08, 17.15, 17.30, 17.36, 17.38, 17.49, 20.98,
60.58, 69.02, 80.97, 81.88, 87.55, 113.81, 121.40, 127.26, 128.86,
130.10, 131.94, 132.30, 134.20, 136.43, 145.91, 151.25, 151.48, 165.75,
194.48. IR (film): 3200, 2945, 2925, 2880, 1680, 1505, 1492, 1390,
1300, 1270, 1220, 1195, 1150, 1125, 1038, 882, 700 cm-1. MS (EI):
m/z (rel intensity) 723 (<1, M+), 557 (37), 556 (78), 555 (100), 514
(37), 513 (61), 512 (87), 278 (45), 277 (76), 105 (42).
of 8b as a colorless oil, Rf 0.57 (hexane/ethyl acetate 2:1). [R]D
)
-2.1 (c ) 0.42, CHCl3). 1H NMR (300 MHz, CDCl3) δ 1.03-1.12
(m, 28 H), 2.05 (dt, J ) 15.87, 7.35 Hz, 1 H), 2.28 (s, 3 H), 2.40 (ddd,
J ) 12.87, 6.99, 4.78 Hz, 1 H), 3.85-3.94 (m, 2 H), 4.08-4.18 (m, 1
H), 4.54 (dt, J ) 7.73, 4.78 Hz, 1 H), 4.69 (s, 2 H), 5.11 (t, J ) 7.35
Hz, 1 H), 6.98-7.09 (m, 3 H), 7.32-7.38 (m, 2 H), 7.59 (d, J ) 2.21
Hz, 1 H), 8.25 (d, J ) 2.20 Hz, 1 H), 8.57 (s, NH). 13C NMR (75
MHz, CDCl3): δ 12.54, 12.97, 13.34, 13.46, 16.94, 17.04, 17.07, 17.22,
17.37, 17.39, 17.43, 17.54, 17.96, 42.80, 63.33, 67.49, 72.95, 76.35,
86.49, 114.70, 122.29, 129.83, 136.12, 137.37, 143.97, 147.65, 157.06,
166.69, 171.09. IR (film): 3250, 2944, 2893, 2867, 1698, 1600, 1495,
1464, 1244, 1142, 1119, 1089, 1063, 1040, 886, 754, 691 cm-1. MS
(EI): m/z (rel intensity) 601 (10, [M + 1]+), 600 (23, M+), 558 (40),
557 (97), 507 (20), 482 (18), 466 (19), 135 (16).
2-(N-Benzoylamino)-5-(2′-deoxy-â-D-ribofuranosyl)pyridine (9a).
To a solution of 8a (570 mg, 1.02 mmol) in THF (15 mL) was added
at room temperature Bu4NF (1.85 mL, 2.0 mmol), and the mixture was
stirred for 90 min at room temperature. H2O (0.1 mL) followed by
10% NaOH (0.1 mL) was added, and the mixture was concentrated.
Flash chromatography (ethyl acetate/MeOH 9:1) of the residue afforded
277 mg (86%) of 9a as a colorless solid, Rf 0.34 (ethyl acetate/MeOH
9:1), mp 174-175 °C. [R]D ) +38.0 (c ) 0.85, MeOH). 1H NMR
(300 MHz, CD3OD): δ 2.04 (ddd, J ) 13.0, 10.5, 5.9 Hz, 1 H), 2.28
(ddd, J ) 13.0, 5.3, 1.65 Hz, 1 H), 3.73 (d, J ) 4.8, 2 H), 4.01 (td, J
) 4.9, 2.3 Hz, 1 H), 4.40 (dt, J ) 5.9, 1.9 Hz, 1 H), 5.20 (dd, J )
10.95, 5.35 Hz, 1 H), 7.53-768 (m, 3 H), 7.94 (dd, J ) 8.65, 2.35 Hz,
1 H), 8.99-8.04 (m, 2 H), 8.25 (d, J ) 8.45 Hz, 1 H), 8.42 (d, J ) 2.2
Hz, 1 H). 13C NMR (75 MHz, CD3OD): δ 45.01, 64.25, 74.72, 79.38,
89.67, 116.18, 129.02, 130.05, 133.63, 135.60, 135.90, 137.95, 147.51,
153.00, 168.82. IR (KBr): 3460, 1663, 1531, 1310, 708 cm-1. MS
(EI): m/z (rel intensity) 314 (18, M+), 286 (32), 285 (85), 107 (29),
105 (100), 77 (50), 43 (31).
3-Methyl-2-[N-(phenoxyacetyl)amino]-5-[3′,5′-O-(1,1,3,3-tetraiso-
propyldisiloxane-1,3-diyl)-2′-O-(imidazolyl thiocarbonyl)-â-D-ribo-
furanosyl]pyridine (7b). A solution of C-nucleoside 6b (1.68 g, 2.72
mmol) and thiocarbonyl diimidazole (1.22 g, 6.83 mmol) in DMF (7
mL) was stirred for 1 h at room temperature and then for another 3.5
h at 40 °C. The mixture was diluted with ethyl acetate (100 mL) and
washed with H2O (5 × 25 mL). The organic layer was dried (MgSO4)
and concentrated, and the residue was purified by flash chromatography
(first ethyl acetate/hexane 1:1, then 2:1) to afford 1.47 g (74%) of 7b
as a colorless foam, Rf 0.32 (hexane/ethyl acetate 1:1). [R]D ) -11.2
(c ) 0.48, CHCl3). 1H NMR (300 MHz, CDCl3): δ 0.93-1.13 (m,
28 H), 2.31 (s, 3 H), 4.07-4.14 (m, 2 H), 4.24-4.29 (m, 1 H), 4.62
(q, J ≈ 4.78 Hz, 1 H), 4.71 (s, 2 H), 5.28 (s, 1 H), 5.74 (d, J ) 4.78
Hz, 1 H), 6.98-7.10 (m, 4 H), 7.33-7.39 (m, 2 H), 7.70 (t, J ) 1.48
Hz, 1 H), 7.81 (d, J ) 2.20 Hz, 1 H), 8.40 (d, J ) 1.10 Hz, 1 H), 8.52
(d, J ) 1.83 Hz, 1 H), 8.58 (s, NH). 13C-NMR (75 MHz, CDCl3): δ
12.60, 12.70, 12.95, 13.29, 16.85, 16.93, 17.05, 17.24, 17.26, 17.33,
17.39, 17.99, 60.26, 67.49, 68.99, 80.91, 82.07, 86.86, 114.68, 118.02,
122.31, 129.82, 131.06, 132.61, 136.76, 137.44, 144.26, 148.38, 157.01,
166.70, 183.27. IR (film): 2945, 2867, 1694, 1600, 1494, 1464, 1392,
3-Methyl-2-[N-(phenoxyacetyl)amino]-5-(2′-deoxy-â-D-ribofura-
nosyl)pyridine (9b). Compound 9b was prepared from C-nucleoside
8b (0.63 g, 1.04 mmol) and Bu4NF (1.9 mL, 2.1 mmol) as described
for 9a. Flash chromatography (ethyl acetate/MeOH 10:1) afforded 0.35
g (93%) of nucleoside 9b as a colorless foam, Rf 0.29 (ethyl acetate/
MeOH 10:1). [R]D ) +18.7 (c ) 0.93, CHCl3). 1H NMR (300 MHz,
CD3OD): δ 1.99 (ddd, J ) 13.24, 10.67, 5.88 Hz, 1 H), 2.27 (s, 3 H),
2.28 (ddd, J ) 13.24, 5.51, 1.84 Hz, 1 H), 3.72 (d, J ) 5.15 Hz, 2 H),
4.00 (dt, J ) 4.78, 2.58 Hz, 1 H), 4.39 (dt, J ) 5.51, 1.84 Hz, 1 H),
4.78 (s, 2 H), 5.20 (dd, J ) 10.67, 5.52 Hz, 1 H), 7.02-7.13 (m, 3 H),
7.33-7.40 (m, 2 H), 7.83 (d, J ) 2.21 Hz, 1 H), 8.34 (d, J ) 2.20 Hz,
1 H). 13C NMR (75 MHz, CD3OD): δ 17.93, 45.01, 64.16, 68.64,
74.57, 79.02, 89.66, 116.18, 123.20, 130.95, 131.37, 138.69, 139.45,
145.18, 149.36, 159.39, 170.58. IR (film): 3392, 3011, 2930, 1686,
1324, 1287, 1231, 1154, 1137, 1065, 1038, 994, 884, 772, 691 cm-1
.
MS (EI): m/z (rel intensity) 727 (5, M+), 599 (47), 598 (83), 556 (52),
555 (100), 464 (53), 322 (40), 321 (80), 235 (55).
2-(N-Benzoylamino)-5-[3′,5′-O-(1,1,3,3-tetraisopropyldisiloxane-
1,3-diyl)-2′-deoxy-â-D-ribofuranosyl]pyridine (8a). A solution of 7a
(0.88 g, 1.22 mmol), azobisisobutyronitrile (20 mg, 0.12 mmol) and
nBu3SnH (0.50 mL, 1.89 mmol) in toluene (40 mL) was flushed with
argon for 50 min and then heated at 80 °C for 4 h. NaOH (30 mL, 1
M) was added, and the mixture was extracted with ethyl acetate (4 ×
60 mL). The organic layer was dried over MgSO4 and concentrated.
Flash chromatography (hexane/ethyl acetate 5:1) of the residue yielded
0.57 g (84%) of 8a as a colorless oil, Rf 0.22 (hexane/ethyl acetate
5:1). 1H NMR (300 MHz, CDCl3): δ 0.91-1.20 (m, 28 H), 2.06 (dt,
J ) 12.9, 7.7 Hz, 1 H), 2.39 (ddd, J ) 12.8, 6.7, 4.5 Hz, 1 H), 3.83-
3.95 (m, 2 H), 4.07-4.18 (m, 1 H), 4.55 (dt, J ) 7.7, 4.4 Hz, 1 H),
5.09 (t, J ) 7.35, 1 H), 7.19-7.25 (m, 2 H), 7.46-7.61 (m, 3 H), 7.75
(dd, J ) 8.7, 2.4 Hz, 1 H), 8.22 (d, J ) 1.8 Hz, 1 H), 8.37 (d, J ) 8.8
Hz, 1 H), 8.77 (s, NH). 13C NMR (75 MHz, CDCl3): δ 12.75, 13.04,
13.40, 13.52, 16.98, 17.07, 17.11, 17.27, 17.40, 17.44, 17.45, 17.58,
42.94, 63.55, 73.25, 76.69, 86.59, 113.82, 127.23, 128.84, 132.24,
1599, 1589, 1495, 1427, 1231, 1175, 1084, 1053, 998, 755, 691 cm-1
.
MS (EI): m/z (rel intensity) 359 (36, [M + 1]+), 358 (99, M+), 266
(32), 265 (88), 224 (86), 175 (40), 147 (36), 135 (69), 134 (32), 121
(55), 108 (32).
2-[N-(Phenoxyacetyl)amino]-5-(2′-O-methyl-â-D-ribofuranosyl)-
pyridine (9c). Compound 9c was prepared from C-nucleoside 6c (330
mg, 0.53 mmol) and Bu4NF (1.0 mL, 1.1 mmol) as described for 9a.
Flash chromatography (first ethyl acetate, then ethyl acetate/MeOH 10:
1) yielded 175 mg (87%) of 9c as a colorless solid, mp 117-118 °C,
Rf 0.56 (ethyl acetate/MeOH 20:1). [R]D ) -5.8 (c ) 0.35, CHCl3).
1H NMR (300 MHz, CD3OD): δ 3.44 (s, 3 H), 3.64 (dd, J ) 7.15,
5.35 Hz, 1 H), 3.75 (dd, J ) 12.1, 4.4 Hz, 1 H), 3.83 (dd, J ) 12.1,
3.7 Hz, 1 H), 4.05 (q, J ≈ 3.9 Hz, 1 H), 4.30 (dd, J ) 5.15, 3.7 Hz, 1

5-Substituted 2-Aminopyridine C-Nucleosides
J. Am. Chem. Soc., Vol. 119, No. 24, 1997 5509
H), 4.76 (s, 2 H), 4.85 (d, J ) 7.0 Hz, 1 H), 7.00-7.13 (m, 3 H),
7.29-7.42 (m, 2 H), 7.96 (dd, J ) 8.6, 2.4 Hz, 1 H), 8.21 (d, J ) 8.5
Hz, 1 H), 8.41 (d, J ≈ 2 Hz, 1 H). 13C NMR (75 MHz, CD3OD): δ
58.95, 63.70, 68.68, 71.69, 81.16, 87.68, 88.53, 115.50, 116.21, 123.37,
131.05, 134.63, 138.27, 147.58, 152.02, 159.30, 169.79. IR (film):
3390, 2930, 1692, 1592, 1525, 1492, 1400, 1302, 1232, 1115, 1080,
1058, 750, 688 cm-1. MS (EI): m/z (rel intensity) 375 (17, [M +
1]+), 374 (43, M+), 281 (82), 240 (83), 163 (58), 135 (58), 123 (100),
107 (71), 87 (90), 77 (68).
2-Amino-5-(2′-deoxy-â-D-ribofuranosyl)pyridine (P). A solution
of C-nucleoside 9a (61 mg, 0.19 mmol) in 40% aqueous Me2NH (6
mL) was heated in a closed flask at 70 °C for 25 h. The reaction
mixture was concentrated, and the residue was adsorbed onto silica
gel and chromatographed (CH2Cl2/MeOH 4:1) to afford 34 mg (83%)
of free nucleoside P as a colorless oil, Rf 0.35 (CH2Cl2/MeOH 4:1),
pKa 6.26. 1H NMR (300 MHz, CD3OD): δ 2.01 (ddd, J ) 13.2, 10.7,
5.9 Hz, 1 H), 2.14 (ddd, J ) 13.2, 5.4, 1.8 Hz, 1 H), 3.69 (d, J ) 4.8
Hz, 2 H), 3.94 (td, J ) 4.9, 2.5 Hz, 1 H), 4.37 (dt, J ) 5.9, 1.8 Hz, 1
H), 5.03 (dd, J ) 10.7, 5.5 Hz, 1 H), 6.61 (d, J ) 8.8 Hz, 1 H), 7.57
(dd, J ) 8.8, 2.2 Hz, 1 H), 7.92 (d, J ) 1.8 Hz, 1 H). 13C NMR (75
MHz, CD3OD): δ 44.42, 64.25, 74.73, 79.71, 89.32, 110.51, 127.30,
138.34, 146.46, 160.87. IR (film): 3340, 1630, 1510, 1045, 1025. UV
(H2O): λmax nm (ꢀ) 234 (9300), 260 (650) 292 (3000). MS (EI): m/z
(rel intensity) 210 (54, M+), 179 (32), 123 (24), 121 (72), 120 (35),
107 (100), 94 (36), 31 (22).
2-(N-Benzoylamino)-5-[5′-O-[(4,4′-dimethoxytriphenyl)methyl]-
2′-deoxy-â-D-ribofuranosyl]pyridine (13a). To a solution of 9a (105
mg, 0.33 mmol) in pyridine (1.5 mL) was added at room temperature
(in 3 portions within 2 h) 4,4′-dimethoxytrityl chloride (DMTCl; 136
mg, 0.40 mmol). After another 2 h at room temperature, toluene (3
mL) was added, and the reaction mixture was concentrated. The residue
was dissolved in CH2Cl2 (40 mL) and washed with 0.4 M NaOH (10
mL). The aqueous layer was extracted with CH2Cl2 (5 × 40 mL), the
combined organic layers were dried (Na2SO4) and concentrated. Flash
chromatography (ethyl acetate/hexane 5:1) yielded 173 mg (84%) of
13a as a colorless foam, Rf 0.35 (ethyl acetate/hexane 5:1). 1H NMR
(300 MHz, CDCl3): δ 2.04 (ddd, J ) 13.7, 9.6, 6.0 Hz, 1 H), 2.16
(broad, OH), 2.26 (ddd, J ) 13.2, 5.5, 1.8 Hz, 1 H), 3.26 (dd, J ) 9.7,
5.3 Hz, 1 H), 3.36 (dd, J ) 9.7, 4. Hz, 1 H), 3.79 (s, 6 H), 4.04-4.12
(m, 1 H), 4.41-4.50 (m, 1 H), 5.17 (dd, J ) 10.3, 5.5 Hz, 1 H), 6.79-
6.85 (m, 4 H), 7.17-7.36 (m, 7 H), 7.41-7.60 (m, 5 H), 7.75 (dd, J
) 8.6, 2.4 Hz, 1 H), 7.89-7.95 (m, 2 H), 8.26 (d, J ) 1.8 Hz, 1 H),
8.35 (d, J ) 8.8 Hz, 1 H), 8.74 (broad, NH). 13C NMR (75 MHz,
CDCl3): δ 43.63, 55.24 (2×), 64.41, 74.61, 77.61, 86.33, 86.51, 113.18,
113.81, 126.88, 127.22, 127.88, 128.17, 128.85, 130.05, 130.08, 132.25,
133.45, 134.24, 135.93, 135.95, 136.44, 144.74, 145.87, 151.07, 158.54,
158.56, 165.67. IR (film): 3400, 1672, 1600, 1585, 1510, 1300, 1247,
1170, 1078, 1030, 903, 825, 725, 700 cm-1
. MS (EI): m/z (rel
intensity) 616 (1, M+), 304 (20), 303 (62), 286 (17), 285 (48), 225
(13), 105 (100), 77 (39).
2-Amino-3-methyl-5-(2′-deoxy-â-D-ribofuranosyl)pyridine (MeP).
A solution of 9b (70.6 mg, 0.20 mmol) in 25% aqueous NH3 (3 mL)
was heated in a closed flask at 55 °C overnight. The reaction mixture
was concentrated, and the residue was dissolved in H2O (10 mL) and
made acidic (ca. pH 1) with 1 M HCl. The aqueous layer was washed
with ethyl acetate (5 × 20 mL) and CH2Cl2 (2 × 20 mL), made basic
by adding 1 M NaOH, concentrated, and purified by flash chromatog-
raphy (CH2Cl2/MeOH 4:1) to yield 27.6 mg (63%) of MeP as a colorless
3-Methyl-2-[N-(phenoxyacetyl)amino]-5-[5′-O-[(4,4′-dimeth-
oxytriphenyl)methyl]-2′-deoxy-â-D-ribofuranosyl]pyridine (13b). Nu-
cleoside 13b was prepared from 9b (133 mg, 0.37 mmol), DMTCl
(257 mg, 0.76 mmol), and pyridine (2 mL) as described for 13a. Flash
chromatography (ethyl acetate/hexane 5:1) of the residue yielded 181
mg (74%) of 13b as a slightly yellow foam, Rf 0.18 (ethyl acetate/
hexane 5:1). [R]D ) +17.6 (c ) 1.09, CHCl3). 1H NMR (300 MHz,
CDCl3): δ 2.05 (ddd, J ) 13.24, 10.29, 5.88 Hz, 1 H), 2.21 (s, 3 H),
2.27 (ddd, J ) 13.24, 5.52, 1.84 Hz, 1 H), 3.29 (dd, J ) 9.92, 5.15
Hz, 1 H), 3.36 (dd, J ) 9.92, 4.05 Hz, 1 H), 3.80 (s, 6 H), 4.09 (dt, J
) 5.15, 2.58 Hz, 1 H), 4.41-4.53 (m, 1 H), 4.69 (s, 2 H), 5.19 (dd, J
) 9.92, 5.52 Hz, 1 H), 6.80-6.87 (m, 4 H), 6.96-7.10 (m, 3 H), 7.19-
7.40 (m, 9 H), 7.43-7.49 (m, 2 H), 7.67 (d, J ) 2.20, 1 H), 8.28 (d,
J ) 2.21 Hz, 1 H), 8.55 (s, NH). 13C NMR (75 MHz, CDCl3): δ
17.90, 43.81, 55.21, 64.32, 67.50, 74.58, 77.37, 86.33, 86.54, 113.16,
114.72, 122.31, 126.85, 127.85, 128.14, 129.85, 130.05, 135.92, 137.65,
144.13, 144.74, 147.69, 157.07, 158.53, 166.71. IR (film): 3392, 2931,
1684, 1607, 1508, 1442, 1300, 1249, 1175, 1083, 1034, 829, 754, 692,
668 cm-1. MS (FAB): (rel intensity) 661 (61, [M + 1]+), 304 (27),
303 (100), 269 (10), 135 (10).
2-[N-(Phenoxyacetyl)amino]-5-[5′-O-[(4,4′-dimethoxytriphenyl)-
methyl]-2′-O-methyl-â-D-ribofuranosyl]pyridine (13c). Nucleoside
13c was prepared from 9c (170 mg, 0.45 mmol), DMTCl (180 mg,
0.53 mmol) and pyridine (2.5 mL) as described for 13a. Flash
chromatography (first ethyl acetate/hexane 1:1, then 3:1) of the residue
yielded 294 mg (81%) of DMT-protected nucleoside 13c as a colorless
foam, Rf 0.23 (ethyl acetate/hexane 1:1). [R]D ) +4.0 (c ) 0.6, CHCl3).
1H NMR (300 MHz, CDCl3): δ 2.57 (d, J ) 4.1, OH), 3.24 (dd, J )
10.3, 4.05 Hz, 1 H), 3.36 (dd, J ) 10.3, 3.3 Hz, 1 H), 3.36 (s, 3 H),
3.64 (dd, J ) 7.0, 5.5 Hz, 1 H), 3.72 (s, 6 H), 4.11 (q, J ) 3.7 Hz, 1
H), 4.15-4.21 (m, 1 H), 4.57 (s, 2 H), 4.79 (d, J ) 7.0 Hz, 1 H),
6.73-6.79 (m, 5 H), 6.91-7.03 (m, 4 H), 7.11-7.41 (m, 9 H), 7.78
(dd, J ) 8.6, 2.4 Hz, 1 H), 8.19 (d, J ) 8.8 Hz, 1 H), 8.32 (d, J ) 2.2
Hz, 1 H), 8.90 (s, NH). 13C NMR (75 MHz, CDCl3): δ 55.21 (2×),
58.66, 63.92, 67.41, 70.85, 79.33, 84.39, 86.32, 86.85, 113.15, 113.87,
114.80, 122.42, 126.84, 127.84, 128.17, 129.87, 130.08, 130.10, 132.50,
135.85, 135.89, 136.32, 144.72, 146.02, 158.51, 166.81. IR (film):
3400, 1600, 1510, 1490, 1300, 1248, 1172, 725 cm-1. MS (FAB):
m/z (rel intensity) 676 (3, M+), 304 (40), 303 (100), 285 (12), 135
(14), 109 (12).
1
oil, Rf 0.31 (CH2Cl2/MeOH 4:1). [R]D ) +43.4 (c ) 0.08, H2O). H
NMR (300 MHz, CD3OD): δ 2.00 (ddd, J ) 13.24, 10.66, 5.88 Hz, 1
H), 2.19 (ddd, J ) 13.24, 5.52, 1.84 Hz, 1 H), 2.23 (s, 3 H), 3.70 (d,
J ) 4.78 Hz, 2 H), 3.96 (dt, J ) 4.78, 2.58 Hz, 1 H), 4.37 (dt, J )
5.89, 1.84 Hz, 1 H), 5.04 (dd, J ) 10.30, 5.52 Hz, 1 H), 7.50 (s, 1 H),
7.80 (s, 1 H). 13C NMR (75 MHz, CD3OD): δ 17.34, 44.46, 64.24,
74.69, 79.53, 89.38, 119.61, 128.06, 138.71, 142.43, 158.75. IR
(KBr): 3343, 2918, 1666, 1630, 1489, 1436, 1380, 1094, 1050, 1000,
863 cm-1. UV (H2O): λmax nm (ꢀ) 232 (10 400), 260 (950), 293 (5250).
MS (EI): m/z (rel intensity) 225 (17, [M + 1]+), 224 (100, M+), 223
(22), 193 (22), 137 (21), 135 (97), 134 (32), 121 (80), 108 (28).
A sample of MeP was treated with a saturated solution of HCl in
MeOH to afford after flash chromatography (CH2Cl2/MeOH 4:1) its
hydrochloride salt MeP‚HCl as a colorless foam, pKa 6.63. 1H NMR
(300 MHz, CD3OD): δ 1.99 (ddd, J ) 12.87, 10.306, 5.88 Hz, 1 H),
2.25 (ddd, J ) 12.87, 5.52, 1.84 Hz, 1 H), 2.31 (s, 3 H), 3.69 (dd, J )
12.14, 4.78 Hz, 1 H), 3.74 (dd, J ) 11.77, 4.04 Hz, 1 H), 3.98 (dt, J
) 4.78, 2.58 Hz, 1 H), 4.39 (dt, J ) 5.88, 1.84 Hz, 1 H), 5.08 (dd, J
) 10.30, 5.52 Hz, 1 H), 7.83 (s, 1 H), 7.94 (s, 1 H).
2-Amino-5-[2′-O-methyl-â-D-ribofuranosyl]pyridine (POMe).
A
solution of 9c (75 mg, 0.20 mmol) in 25% NH3 (4 mL) was heated in
a closed flask at 57 °C for 14 h. The reaction mixture was concentrated,
the residue diluted with 0.1 M HCl (10 mL) and washed with ethyl
acetate (5 × 40 mL). The organic layer was extracted with 0.1 M
HCl (10 mL). The combined aqueous layers were made basic (ca. pH
9) with 2 M NaOH and concentrated. Flash chromatography (ethyl
acetate/MeOH 6:1) yielded 42 mg (91%) of free nucleoside POMe as a
colorless oil, Rf 0.20 (ethyl acetate/MeOH 6:1). [R]D ) -41.9 (c )
0.36, H2O). 1H NMR (300 MHz, CD3OD): δ 3.42 (s, 3 H), 3.62 (dd,
J ) 7.35, 5.15 Hz, 1 H), 3.72 (dd, J ) 12.15, 4.4 Hz, 1 H), 4.00 (q, J
≈ 3.7 Hz, 1 H), 4.27 (dd, J ) 5.15, 3.7 Hz, 1 H), 4.70 (d, J ) 7.35
Hz, 1 H), 6.71 (d, J ) 8.8 Hz, 1 H), 7.70 (dd, J ) 8.8, 2.2 Hz, 1 H),
7.93 (d, J ) 2.2 Hz, 1 H). 13C NMR (75 MHz, CD3OD): δ 58.89,
63.73, 71.71, 81.31, 87.39, 87.97, 111.39, 126.41, 139.44, 144.14,
160.03. IR (KBr): 3350, 2920, 2900, 1625, 1570, 1505, 1410, 1349,
2-(N-Benzoylamino)-5-[5′-O-[(4,4′-dimethoxytriphenyl)methyl]-
2′-deoxy-â-D-ribofuranosyl]pyridine-3′-O-(2-cyanoethyl-N,N-diiso-
propyl)phosphoramidite (14a). 2-Cyanoethyl N,N-diisopropylchlo-
rophosphoramidite (75 µL, 0.34 mmol) was added dropwise at room
temperature to a solution of 13a (135 mg, 0.22 mol) and N,N-
diisopropylethylamine (110 µL, 0.65 mmol) in THF (7 mL). After 2
h at room temperature, CH2Cl2 (50 mL) was added and the organic
1129, 1115, 1080, 1050, 1030, 980, 821 cm-1
. UV (H2O): λmax
nm (ꢀ) 235 (14450), 292 (4570), 260 (850). MS (EI): m/z (rel inten-
sity) 240 (28, M+), 208 (15), 150 (17), 124 (19), 123 (100), 107 (26),
87 (45).

5510 J. Am. Chem. Soc., Vol. 119, No. 24, 1997
Hildbrand et al.
Table 4. Purification, Characterization, and Yields of Oligonucleotides 16-22
HPL-chromatography
MALDI-TOF MS
strand
DEAE-HPLCa
RP-HPLCb
calcdc
foundc
ODd (yield (%))
16
17
18
19
20
21
22
35-65% B in 35 min (tR 17.9)
40-50% B in 30 min (tR 15.1)
40-60% B in 40 min (tR 14.5)
35-65% B in 35 min (tR 11.5)
35-40% B in 30 min (tR 11.3)
41-46% B in 40 min (tR 10.4)
35-40% B in 30 min (tR 12.5)
tR 16.0
tR 9.8
tR 14.8
tR 16.9
tR 8.8
tR 16.6
tR 8.9
4407.9
4422.0
4438.0
4340.0
4410.1
4490.1
4410.1
4407.9
4428.0
4439.4
4340.4
4414.7
4489.3
4412.5
51 (33)
68 (45)
59 (39)
37 (32)
55 (47)
34 (30)
39 (33)
a Purification was performed by elution with a linear gradient of B. Mobile phase A: 20 mM NaH2PO4 in H2O/CH3CN 4:1 (pH 3.9 for 16, 19
and 4.1 for 17, 18, 20-22), B: A + 1 M NaCl; flow 1 mL/min (tR in min); UV detection (260 nm). b The purity of the isolated oligomers 16-22
was proved using always the same linear gradient: 10-35% B in 30 min; mobile phase A: 0.1 M triethylammonium acetate in H2O (pH 7.0), B:
0.1 M triethylammonium acetate in H2O/CH3CN 1:4 (pH 7.0); flow 1 mL/min (tR in min); UV detection (260 nm). c Monoanionic form. d Total
optical density measured at 260 nm.
layer was washed with saturated NaHCO3 (2 × 10 mL). The aqueous
layer was extracted with CH2Cl2 (2 × 50 mL), the combined organic
layers were dried (Na2SO4) and concentrated. Flash chromatography
(ethyl acetate/hexane 1:1) gave 145 mg (81%) of 14a as a colorless
foam consisting of a 1:1 mixture of diastereoisomers, Rf 0.64, 0.52
(ethyl acetate/hexane 1:1). 1H NMR (500 MHz, CDCl3): δ 1.10 (d, J
) 6.8, 3 H), 1.16-1.21 (m, 9 H), 1.99-2.09 (m, 1 H), 2.34 (dd, J )
13.0, 5.1, 0.5 H), 2.44 (dd, J ≈ 13.0, 4.9, 0.5 H), 2.46 (t, J ) 6.5, 1
H), 2.62 (t, J ) 6.5, 1 H), 3.23-3.36 (m, 2 H), 3.54-3.67 (m, 2 H),
3.67-3.90 (m, 2 H), 3.784, 3.786, 3.789, 3.792 (4s, 6 H), 4.22-4.29
(m, 1 H), 4.51-4.58 (m, 1 H), 5.14-5.19 (m, 1 H), 6.80-6.85 (m, 4
H), 7.18-7.24 (m, 1 H), 7.26-7.31 (m, 2 H), 7.33-7.37 (m, 4 H),
7.44-7.52 (m, 4 H), 7.54-7.59 (m, 1 H); 7.78-7.82 (m, 1 H), 7.92
(s, 1 H), 7.93 (s, 1 H), 8.29-8.32 (m, 1 H), 8.359, 8.364 (2d, J ) 8.5,
1 H), 8.71, 8.72 (2s, NH). 31P NMR (202 MHz, CDCl3): δ 154.00,
154.12. IR (film): 2970, 2930, 1675, 1605, 1580, 1510, 1692, 1460,
1443, 1390, 1360, 1340, 1300, 1248, 1175, 1070, 1030, 973, 905, 830,
728 cm-1. MS (FAB): m/z (rel intensity) 817 (2, M+), 304 (40), 303
(100), 135 (10), 119 (13), 105 (30).
Hz, 0.5 H), 6.76-6.81 (m, 4 H), 6.95-7.07 (m, 3 H), 7.15-7.37 (m,
9 H), 7.39-7.47 (m, 2 H), 7.83 (dd, J ) 8.8, 2.2 Hz, 0.5 H), 7.87 (dd,
J ) 9.4, 2.0 Hz, 0.5 H), 8.23 (d, J ) 8.5 Hz, 1 H), 8.39 (d, J ) 2.2 Hz,
0.5 H), 8.40 (d, J ) 2.2 Hz, 0.5 H), 8.95 (s, broad, NH). 31P NMR
(202 MHz, CDCl3): δ 150.74, 151.33. IR (film): 3400, 2964, 2928,
1702, 1602, 1510, 1300, 1245, 1175, 1080, 1030, 750 cm-1
. MS
(FAB): m/z (rel intensity) 877(6, M+), 304 (23), 303 (100).
Oligonucleotide Synthesis and UV-Melting Experiments. Oli-
gonucleotides 16-21 were synthesized on a Pharmacia Gene-Assembler
Special (connected to a Compaq ProLinea 3/25 ZS personal computer)
using standard â-cyanoethyl phosphoramidite chemistry. Reagents and
concentrations were as for the synthesis of natural DNA oligomers.
Syntheses were performed on a 1.3 µmol scale, according to the
manufacturer’s protocol.16b The only change made to the usual
synthesis cycle was the prolongation of the coupling time of the
synthetic nucleosides to 6 min. Coupling efficiencies were estimated
from trityl assays and were >97% per step. After chain assembly and
final detritylation (trityl off mode) the oligomers were removed from
the support and deprotected by treatment with ca. 4 mL 25% NH3 (55
°C, 12-22 h) followed by ca. 4 mL 40% MeNH2 (70 °C, 20-26 h).
The crude oligonucleotides were purified by ion exchange HPLC
using a Nucleogen DEAE 60-7 (125 × 4.0 mm) column (Macherey &
Nagel). The isolated oligonucleotides were desalted over SEP-PAK
C-18 cartridges (Waters) according to a standard procedure.26 The
purity of the oligomers were determined by reversed-phase HPLC using
a Aquapore Rp-300 (220 × 4.6 mm) column (Brownlee). The synthesis
of oligonucleotide 22 essentially followed the protocol described,
however, phenoxyacetic anhydride was used instead of acetic anhydride
as the capping agent. This change in the synthesis simplified the depro-
tection step in that the MeNH2 treatment was not necessary. The control
oligomers 15, 23, and 24 were prepared by standard oligonucleotide
chemistry. Details for syntheses, HPLC purification, and characteriza-
tion by MALDI-TOF mass spectrometry are given in Table 4.
UV melting experiments were performed on a Cary 3E UV/vis
spectrophotometer (Varian). Melting curves (λ ) 260 nm) were re-
corded for a consecutive heating (0 f 90 °C) cooling-heating protocol
with a linear gradient of 0.5 °C/min. All measurements were conducted
in buffer solutions. The final pH was always adjusted to the desired
value using 0.1 M HCl or 0.1 M NaOH.
3-Methyl-2-[N-(phenoxyacetyl)amino]-5-[5′-O-[(4,4′-di-
methoxytriphenyl)methyl]-2′-deoxy-â-D-ribofuranosyl]pyridine-3′-
O-(2-cyanoethyl-N,N-diisopropyl)phosphoramidite (14b). Phos-
phoramidite 14b was prepared from 2-cyanoethyl N,N-diisopropyl-
chlorophosphoramidite (95 µL, 0.43 mmol), 13b (135 mg, 0.20 mmol)
and N,N-diisopropylethylamine (110 µL, 0.65 mmol) in THF (8 mL)
as described for 14a. Flash chromatography (ethyl acetate/hexane 1:1)
afforded 143 mg (81%) of 14b as a colorless foam consisting of a 1:1
mixture of diastereoisomers, Rf 0.29 and 0.19 (ethyl acetate/hexane 1:1).
1H NMR (300 MHz, CDCl3): δ 1.10 (d, J ) 6.57 Hz, 3 H), 1.16-
1.22 (m, 9 H), 2.01-2.10 (m, 1 H), 2.20 (s, 3 H), 2.35 (dd, J ) 13.0,
5.3 Hz, 0.55 H), 2.44 (dd, J ) 13.0, 5.3 Hz, 0.45 H), 2.47 (t, J ) 6.57
Hz, 1 H), 2.63 (t, J ) 6.42 Hz, 1 H), 3.23-3.38 (m, 2 H), 3.56-3.66
(m, 2 H), 3.67-3.76 (m, 1 H), 3.79 (s, 3 H), 3.80 (s, 3 H), 3.80-3.88
(m, 1 H), 4.22-4.28 (m, 1 H), 4.53-4.59 (m, 1 H), 4.69 (s, 2 H),
5.16-5.20 (m, 1 H), 6.80-6.84 (m, 4 H), 6.98-7.02 (m, 2 H), 7.04-
7.08 (m, 1 H), 7.19-7.24 (m, 1 H), 7.25-7.31 (m, 2 H), 7.32-7.38
(m, 6 H), 7.45-7.48 (m, 2 H), 7.68-7.74 (m, 1 H), 8.28-8.34 (m, 1
H), 8.53 (s, NH). 31P NMR (202 MHz, CDCl3): δ 164.69, 164.85. IR
(film): 2965, 2930, 1696, 1607, 1508, 1364, 1300, 1249, 1178, 1033,
977, 829, 754 cm-1. MS (FAB): m/z (rel intensity) 861 (6, M+), 304
(23), 303 (100).
Materials for DNase I Footprinting. All aqueous solutions for
DNA manipulations were prepared with Milli-Q water. DNA and cell
manipulations followed standard protocols.26 The construction of
plasmid pJM4C1 was performed by standard methods and is published
elsewhere.27 dGTP, dCTP, sonicated calf thymus DNA, AvaI restriction
enzyme, Klenow fragment of the DNA polymerase I and DNase I were
from Pharmacia. PvuII restriction enzyme was from Boehringer
Mannheim and was used according to the suppliers protocol in the
buffer provided. [R-32P]-dCTP (g3000 Ci/mmol) was obtained from
Amersham. Bis-tris was from Sigma and the salts used for buffers
were from Fluka (MicroSelect grade).
2-[N-(Phenoxyacetyl)amino]-5-[5′-O-[(4,4′-dimethoxytriphenyl)-
methyl]-2′-O-methyl-â-D-ribofuranosyl]pyridine-3′-O-(2-cyanoethyl-
N,N-diisopropyl)phosphoramidite (14c). Phosphoramidite 14c was
prepared from 13c (0.22 g, 0.33 mmol), 2-cyanoethyl N,N-diisopro-
pylchlorophosphoramidite (110 µL, 0.49 mmol) and N,N-diisopropyleth-
ylamine (170 µL, 0.99 mmol) in THF (10 mL) as described for 14a.
Flash chromatography (hexane/ethyl acetate 15:14) yielded 246 mg
(86%) of 14c as a colorless foam consisting of a 1:1 mixture of
diastereoisomers, Rf 0.40 and 0.35 (hexane/ethyl acetate 3:2). 1H NMR
(300 MHz, CDCl3): δ 1.02 (d, J ) 7.0 Hz, 3 H), 1.13-1.22 (m, 9 H),
2.31 (t, J ) 6.6 Hz, 1 H), 2.61-2.69 (m, 1 H), 3.21-3.28 (m, 1 H),
3.33-3.71 (m, 5 H), 3.34 (s, 1.5 H), 3.37 (s, 1.5 H), 3.77 (s, 3 H),
3.78 (s, 3 H), 3.83-3.97 (m, 1 H), 4.23-4.29 (m, 0.5 H), 4.30-4.45
(m, 1.5 H), 4.61 (s, 2 H), 4.85 (d, J ) 8.1 Hz, 0.5 H), 4.86 (d, J ) 8.1
(26) Sambrook, J.; Fritsch, E. F.; Maniatis, T. Molecular Cloning: A
Laboratory Manual; Cold Spring Harbor Laboratory Press: Plainview, NY,
1989.
(27) Marfurt, J.; Parel, S. P.; Leumann, C. Nucleic Acids Res. 1997, 25,
1875-1882.

5-Substituted 2-Aminopyridine C-Nucleosides
J. Am. Chem. Soc., Vol. 119, No. 24, 1997 5511
3′-End-Labeling of the Restriction Fragment. Approximately 2
µg plasmid pJM4C1 were subsequently digested with the AvaI and
PvuII restriction enzymes and the products separated by agarose gel
electrophoresis. The desired band was excised and extracted with a
QiaexII extraction kit (Qiagen). The restriction fragment was ethanol
precipitated and resuspended in water. 3′-End-labeling was performed
with [R-32P]dCTP using the Klenow fragment of the DNA polymerase
I. Excess unlabeled nucleotide triphosphates were added after labeling
to ensure complete fill in. Unincorporated nucleotides were removed
on a gel filtration column (NICK column, Pharmacia). The labeled
fragment was precipitated with ethanol. From 2 µg of plasmid DNA
a total Cerenkov radioactivity of ∼3 500 000 cpm was achieved.
DNase I Footprint Titration. These experiments were performed
essentially according to a published protocol.3d Triplex forming
oligonucleotide was placed in 10 microcentrifuge tubes to span a
concentration range of 25 µM to 2.5 nM (final concentration) and
adjusted to a volume of 18 µL with H2O. Two more tubes containing
only 18 µL H2O were used for the reference lanes. A stock solution
(27 µL) containing 5X physiological association buffer (50 mM NaCl,
700 mM KCl, 2.5 mM MgCl2, 5 mM spermine, 50 mM bis-tris, pH
7.2, 50 µM base pairs sonicated calf thymus DNA; 9 µL), the labeled
restriction fragment, and H2O in appropriate amounts were added to
each tube. The tubes were vortexed, shortly centrifuged and then
allowed to equilibrate for 7 days at 18 °C (final solution conditions,
10 mM NaCl, 140 mM KCl, 0.5 mM MgCl2, 1 mM spermine, 10 mM
bis-tris, 10 µM bp calf thymus DNA, ∼20000 cpm labeled restriction
fragment), and then 5 µL of a solution containing DNase I (5.9 u, 0.45
u/µL), CaCl2 (50 mM), MgCl2 (50 mM), non-specific oligonucleotide
(10 µM, 5′-d(TATAATTTAA)-3′), bis-tris (10 mM, pH 7.0) and
glycerol (5%) were added and digestion was allowed to proceed for 2
min at room temperature. The reaction was stopped by the addition
of 8.3 µL of a solution containing 1.4 M NaCl, 0.14 M EDTA (pH
8.0) and 0.35 µg/µL of glycogen, followed by 120 µL of ethanol.
Precipitation was performed at -20 °C for 3 h, followed by centrifuga-
tion at 16000g for 30 min at 0 °C. The pellets were washed with cold
70% ethanol, resuspended in 20 µL of H2O, lyophilized to dryness,
and dissolved in 7 µL of 80% formamide loading buffer. After
denaturation (10 min at 85 °C), the samples were loaded on a 8%
denaturing polyacrylamide gel and separated by electrophoresis in 1X
TBE buffer at 1800V for 75 min. After drying on a slab drier, the gel
was exposed to a storage Phosphor-Imager screen (Molecular Dynam-
ics) overnight.
Acknowledgment. We are grateful to the Swiss National
Science Foundation and the Wander-Stiftung for financial
support. We thank the analytical Department of Ciba-Geigy
AG, Basel for performing the pKa measurements.
Supporting Information Available: 1H NMR spectra of
all compounds described (2a-14c) (14 pages). See any current
masthead page for ordering and Internet access instructions.
JA9704904