The synthesis and stability of
oligodeoxyribonucleotides containing the deoxyadenosine mimic 1-(2
'
- deoxy-
[beta]-D- ribofuranosyl)imidazole-4-carboxamide
The synthesis and stability of oligodeoxyribonucleotides containing the deoxyadenosine mimic 1-(2 ' - deoxy- [beta]-D- ribofuranosyl)imidazole-4-carboxamide
W. Travis
Johnson
1
,
Peiming
Zhang
2
and
Donald E.
Bergstrom
1,2,
*
1
Department of Medicinal Chemistry and Molecular Pharmacology, Purdue University,
West Lafayette
, IN 47907,
USA
and
2
Walther Cancer Institute,
Indianapolis
, IN 46208,
USA
Received October 8, 1996;
Accepted November 25, 1996
ABSTRACT
Oligodeoxyribonucleotides containing the nucleoside analog 1-(2
'
-deoxy-
[beta]-D-ribofuranosyl) imidazole-4- carboxamide (1) were synthesized by solid phase
phosphoramidite technology. Nucleoside 1, which contains a reactive exocyclic
amide moiety, was incorporated into synthetic oligodeoxyribonucleotides with the use of a benzoyl protecting group. The corresponding
oligodeoxyribonucleotides with dI or dA in the same position in the sequence
were synthesized for UV comparison of helix-coil transitions. The thermal melting studies indicate that 1, which could conceptually
adopt either a dA- or a dI-like hydrogen bond configuration, pairs with significantly higher
affinity to T than to dC. Nucleoside 1 further resembles dA in the relative
order of its base pairing preferences (T > dG > dA > dC), but may be less
discriminating than dA in its bias for base pairing with T over dG.
INTRODUCTION
Recent studies to find a more effective universal nucleoside than 2'-deoxyinosine (dI) point towards two types of nucleobase analogs,
the nitroazoles and related derivatives (
1
-
3
) and analogs in which the purine ring is replaced by a carboxamide substituted
azole (
4
-
7
). Although 2'-deoxyinosine (dI) has found broad utility as a `universal
nucleoside' in synthetic hybridization probes (
8
,
9
) and degenerate PCR primers (
10
,
11
), it shows a preference for base pairing with dC and is recognized by both DNA
and RNA polymerases as a dG analog when present in a DNA template (
12
). Like dG-dA mismatches (
13
-
15
), dI-dA base pairs are not favored, but neither do they severely disrupt the
DNA double helix (
16
,
17
). Significant helix distortion occurs when dI base pairs with T (
18
) and dI-dG base pairs are almost as disruptive to the helix as dG-dG mismatches (
18
). As a result of these biased base pairing characteristics, the utility of
deoxyinosine as a universal nucleoside for many
in vitro
DNA manipulations is seriously limited.
Azole carboxamides have been proposed as degenerate nucleobase analogs on the basis of their potential to assume multiple conformations which
present different hydrogen bonding configurations (
4
,
5
). An analysis of pyrrole 3-carboxamide, pyrazole 3-carboxamide and imidazole 4-carboxamide by NMR and semi- empirical AM-1 calculations led to the proposal that because
the difference in energy between the two possible co-planar conformations (
syn
and
anti
) about the bond linking the heterocycle to the amide is relatively small,
pyrrole 3-carboxamide may be able to assume either conformation necessary to
hydrogen bond within a duplex to T or C. In contrast, it appeared that the
energy difference between the
syn
and
anti
conformation of imidazole 4-carboxamide was substantial and strongly favored the
anti
conformation. The
anti
conformation would allow base pairing to G or T, but precludes pairing to A or C. As illustrated in Figure
1
, with the amide group locked in the
anti
conformation, rotation about the glycosidic bond provides the means for
1
to adopt a hydrogen bonding configuration and spatial positioning which mimics either dA or dC.
MATERIALS AND METHODS
General information
NMR spectra were obtained on a Bruker AC 250 spectrometer.
1
H and
13
C were referenced to TMS and 85% phosphoric acid was used as an external
standard for
31
P. Chemical shifts are reported in parts per million (p.p.m.). FAB and MALDI
mass spectra were recorded by the mass spectroscopy laboratories, Department of Medicinal Chemistry and Molecular Pharmacology or Department of
Biochemistry respectively, Purdue University. Elemental analyses were performed by the Microanalysis Laboratory, Department of Chemistry, Purdue University. Analytical TLC was carried out on
pre-coated Whatman 60 F
254
plates. Chromatotron preparative chromatography plates were prepared using
silica gel 60 PF
254
containing a gypsum binding agent manufactured by Merck. Anhydrous solvents
were freshly distilled from the appropriate drying agents or purchased from
Aldrich Chemical; all other chemicals were of reagent grade or better quality
and were used as received.
Synthesis and characterization of oligodeoxyribonucleotides
Oligodeoxyribonucleotides were prepared from commercially available dI, dA, dC, dG and T phosphoramidites (Glen Research) on a Milligen/BioSearch 8700 DNA synthesizer by standard solid phase
phosphoramidite chemistry. The oligonucleotides were purified using 20%
polyacrylamide-8 M urea preparative gel electrophoresis. The desired oligonucleotides
were extracted from the gels and desalted with Waters C
18
SepPakstm as per the manufacturer's instructions. Oligodeoxyribonucleotides were
characterized by MALDI mass spectrometry and/or by HPLC analysis of the
constitutent nucleosides obtained by digestion with snake venom
phosphodiesterase and bacterial alkaline phosphatase (
19
). An analytical Phenomenex C
18
column on a Beckman Goldtm HPLC system equipped with a diode array UV scanning device and a dual
channel UV detector set to 254 and 234 nm was used for the HPLC analysis.
1-(2
'
-Deoxy-5
'
-O-dimethoxytrityl-
[beta]
-D-ribofuranosyl)imidazole-4- carboxamide (
2
).
1-(2'-Deoxy-[beta]-D-ribofuranosyl)imidazole-4- carboxamide (
1
) (
5
) (177 mg, 0.780 mmol) was dried by repeated co-evaporations with anhydrous pyridine (2 * 1.0 ml) and dissolved in pyridine (7.0 ml). 4,4'-Dimethoxytrityl chloride (333 mg, 0.936 mmol) was added and the solution was stirred at room temperature overnight. The reaction was quenched by
the dropwise addition of methanol and evaporated under reduced pressure to an oily residue.
The residue was separated by chromatography on silica gel (CH
2
Cl
2
/ethanol). Compound
2
was obtained as a white foam (367 mg, 86.0%):
R
f
= 0.48 (CHCl
3
/methanol, 5:1);
1
H NMR (acetone-
d
6
): [delta] 7.81 (d,
J
2,5
=
1.36 Hz, 1H, H-2), 7.78 (d,
J
5,2
= 1.36 Hz, 1 H, H-5), 7.48-7.19 (m, 9H, aromatic DMTr), 6.91-6.85 (m, 4H, aromatic DMTr), 7.07 (s, CONH
a
, exchangeable H), 6.32 (s, CONH
b
, exchangeable H), 6.18 (pseudotriplet,
J =
6.31 Hz, 1H, H-1'), 4.53 (m, 1H, H-3'), 4.17-4.09 (m, 1H, H-4'), 3.79 (s, 6H, OC
H
3
), 3.34-3.22 (m, 2H, H-5' and H-5''), 2.56-2.47 (m, 2H, H-2' and H-2''); FAB
m
/
z
303.0 (DMTr
+
), 530.2 (MH
+
); high resolution FAB
m
/
z
MH
+
(calc. 530.2291, found 530.2285); anal.: calc. for C
30
H
31
N
3
O
6
[middot]H
2
O: C, 65.80, H, 6.07, N, 7.67; found: C, 66.03, H, 5.86, N, 7.57.
1-(2
'
-Deoxy-5
'
-O-dimethoxytrityl-
[beta]
-D-ribofuranosyl)imidazole-4- carboxamide-3
'
-O-(2-cyanoethyl-N,N-diisopropylphosphoramidite) (
3
) and 1-(2
'
-deoxy-5
'
-O-dimethoxytrityl-
[beta]
-D-ribofuranosyl) imidazole-4-[N-(2-cyanoethyl-N,N-diisopropylphosphoramidite)]
carboxamide-3
'
-O-(2-cyanoethyl-N,N-diisopropylphosphoramidite) (
4
).
Compound
2
(235 mg, 0.44 mmol) was co-evaporated with anhydrous pyridine (3 * 1.0 ml) and dried
in vacuo
(38oC, 250 mTorr) to an oily residue. The residue was dissolved in anhydrous CH
2
Cl
2
and then diisopropylammonium tetrazolide (28 mg, 0.22 mmol) and 2-cyanoethyl
N
,
N
,
N
',
N
'-tetraisopropylphosphorodiamidite (161 mg, 0.53 mmol) were added.
The solution was gently swirled and allowed to stand under nitrogen at room
temperature for 75 min. Analysis by TLC (CH
2
Cl
2
/ethyl acetate/TEA, 45:45:10) indicated the complete absence of starting
material and the presence of two new components in the crude reaction mixture.
The reaction was cooled to 0oC, quenched by the dropwise addition of methanol (0.5% triethylamine added) and evaporated under reduced pressure to an oil. The residue was
dissolved in CH
2
Cl
2
,washed with saturated NaHCO
3
(twice), dried over Na
2
SO
4
and evaporated under reduced pressure to an oil. The products was separated by
chromatography on a silica gel-based chromatotron plate using a mixed solvent of hexane/ethyl acetate/triethylamine as an eluent solvent. Compound
4
was obtained as a thick yellow oil (50 mg, 11%):
R
f
= 0.71 (CH
2
Cl
2
/ethyl acetate/triethylamine, 45:45:10);
31
P NMR (acetone-
d
6
) [delta] 149.99 and 149.88 (P-3'-O), 117.50 and 117.33 (P-NCO) (phosphoramidite diastereomers);
1
H NMR (acetone-
d
6
) d 7.85-7.79 (m, 2H, H-2, H-5), 7.45-7.16 (m, 13 H aromatic DMTr), 6.93-6.82 (m, 4H, aromatic DMTr), 6.22
(pseudotriplet, 1H, H-1'), 4.70-4.62 (m, 1H, H-3'), 4.27-4.18 (m, 1H H-4'), 3.79 (2 s, 6H, OC
H
3
), 3.76-3.56 (m, 8H, OC
H
2
and C
H
), 3.34-3.32, (m, 2H, H-5'and H-5''), 3.00-2.79 (m, 4H, CH
2
CN), 2.68-2.57 (m, 2 H, H-2'and H-2''), 1.31-1.12 (m, 24H, CH
3
); FAB
m
/
z
930.0 (MH
+
), 303.0 (DMTr
+
), 828.8 (M-N[
ip
]
2
+
). Compound
3
was obtained as a white foam (180 mg, 57%):
R
f
= 0.33 (CH
2
Cl
2
/ethyl acetate/triethylamine, 45:45:10);
31
P NMR (acetone-
d
6
) [delta] 149.79 and 149.69 (phosphoramidite diastereomers);
1
H NMR (acetone-
d
6
) [delta] 7.84 and 7.82 (d,
J
2,5
= 1.36 Hz, 1H, H-2), 7.79 and 7.77 (d,
J
5,2
= 1.36 Hz, 1H, H-5), 7.50-7.21 (m, 9H, aromatic DMTr), 6.92-6.87 (m, 4H, aromatic DMTr), 7.09 (s, amide proton), 6.38 (s, amide proton), 6.23
(pseudotriplet,
J
= 6.26 Hz, 1H, H-1'), 4.76-4.62 (m, 1H, H-3'), 4.27-4.17 (m, 1H, H-4'), 3.79 (2 s, 6H, OCH
3
), 3.76-3.56 (m, 4H, OCH
2
and CH), 3.35-3.29 (m, 2H, H-5'and H-5''), 2.80-2.75 (m, 2H, CH
2
CN), 2.68-2.56 (m, 2H, H-2'and H-2''), 1.26-1.17 (m, 12H, CH
3
); high resolution FAB mass MH
+
(calc. 730.3370, found 730.3407).
1-Methylimidazole-4-[N-(2-cyanoethyl-N,N-diisopropylphosphoramidite)]
carboxamide (
6
).
To a solution of 1-methylimidazole-4- carboxamide (
5
) (10 mg, 0.078 mmol) in anhydrous CH
2
Cl
2
(0.5 ml) under nitrogen were added diisopropylammonium tetrazolide (28 mg, 0.22 mmol)
and 2-cyanoethyl
N
,
N
,
N
',
N
'-tetraisopropylphosphorodiamidite (20 mg, 0.53 mmol). The solution
was gently swirled and allowed to stand at room temperature for 5 h. The crude
reaction mixture was filtered through glass wool and purified by chromatography on a silica gel-based chromatotron plate using a mixed solvent of hexane/ethyl acetate/triethylamine as
eluent. Compound
6
was obtained as an oil (10 mg, 25%) which rapidly deteriorates in the absence of triethylamine:
R
f
= 0.43 (CH
2
Cl
2
/ethyl acetate/triethylamine, 45:45:10);
31
P NMR (dichloromethane -
d
2
) [delta] 116.30;
1
H NMR (dichloromethane-
d
2
) [delta] 7.40 (d,
J
2,5
= 1.33 Hz, 1H, H-2), 7.28 (d,
J
5,2
= 1.33 Hz, 1H, H-5), 3.76-3.4 (m, 4H, OCH
2
and CH), 2.60-2.56 (m, 2H, CH
2
CN), 1.23 (2 s, 12H, CH
3
).
1-Methylimidazole-4-(N-benzoyl)carboxamide (
7
).
Benzoyl chloride (187 mg, 1.33 mmol) was slowly added to a stirred solution of
1-methylimidazole-4-carboxamide (
5
) (
20
) (85 mg, 0.664 mmol) in anhydrous pyridine (4.0 ml) at 0oC. The reaction was slowly warmed to room temperature and allowed to stand
for 2 h. The reaction mixture was evaporated to dryness under reduced pressure
and purified by chromatography on silica gel (CHCl
3
/methanol). Compound
7
was obtained as a white powder (145 mg, 95%):
R
f
= 0.83 (CHCl
3
/methanol, 5:1),
1
H NMR (acetone-
d
6
) d 9.06 (br s, amide proton), 7.97-7.93 (m, 2H, Bz
ortho
), 7.87 (d,
J
2,5
= 1.22 Hz, 1H, H-2), 7.73 (d,
J
5,2
= 1.22 Hz, 1H, H-5), 7.68-7.56 (m, 3H, Bz
meta
,
para
), 3.88 (s, 3H, CH
3
); FAB
m
/
z
230.2 (MH
+
); anal.: calc. for C
12
H
11
N
3
O
2
: C, 62.87, H, 4.84, N, 18.33; found: C, 62.66, H, 4.67, N, 18.05.
1-(2
'
-Deoxy-
[beta]
-D-ribofuranosyl)imidazole-4-(N-benzoyl)carboxamide (
8
).
Nucleoside
1
(278 mg, 1.22 mmol) was dried by repeated co-evaporations (3 * 2 ml) with anhydrous pyridine and dissolved in pyridine (10 ml).
Trimethylsilyl chloride (1.712 g, 15.75 mmol) was added slowly to the pyridine
solution at 0oC and the solution stirred under nitrogen at this temperature for 45 min. Benzoyl chloride (343 mg, 2.44 mmol) was added slowly and the solution
allowed to stand at 0oC for 1 h, after which time it was allowed to warm to room temperature, stirred for an additional 2 h, then re-cooled to 0oC. Selective aminolysis of the silyl ether groups was achieved
by the slow addition of water (3 ml) followed, after 1.5 h, by 16 M ammonium
hydroxide (4.0 ml). The solution was allowed to stand for 15 min and then was
evaporated under reduced pressure to a crusty residue. The residue was purified
by column chromatography on silica gel (CHCl
3
/methanol) to give
8
(300 mg, 76%) as a hygroscopic white foam:
R
f
= 0.35 (CHCl
3
/methanol, 5:1);
1
H NMR (acetone-
d
6
) [delta] 10.69 (s, amide proton), 8.32 (d,
J
2,5
= 1.17 Hz, 1H, H-2), 8.18 (d,
J
5,2
= 1.17 Hz, 1H, H-5), 8.03-7.92 (m, 2H, Bz
ortho
), 7.76-7.53 (m, 3H, Bz
meta
,
para
), 6.20 (pseudotriplet,
J
= 6.39 Hz, 1H, H-1'), 5.58-4.80 (2 br s, 3' and 5' O
H
), 4.41-4.37 (m, 1H, H-3'), 3.94-3.89 (m, 1H, H-4'), 3.61-3.58 (m, 2H, H-5'and H-5''), 2.48-2.34 (m, 2H, H-2'and H-2''); high resolution
FAB mass MH
+
(calc. 332.1246, found 332.1239); anal.: calc. for C
16
H
17
N
3
O
5
[middot]2/3 H
2
O: C, 55.97, H, 5.38, N, 12.24; found: C, 56.10, H, 5.22, N, 12.13.
1-(2
'
-Deoxy-5
'
-O-dimethoxytrityl-
[beta]
-D-ribofuranosyl)imidazole-4- (N-benzoyl) carboxamide (
9
).
Nucleoside derivative
8
(280 mg, 0.85 mmol) was dried by repeated co-evaporations with anhydrous pyridine (3 * 2 ml) and then dissolved in pyridine (10 ml). 4,4'-Dimethoxytrityl chloride (344 mg, 1.02 mmol) was added
and the solution was stirred at room temperature under nitrogen. Additional
small portions of 4,4'-dimethoxytrityl chloride were added until all starting material
disappeared by TLC analysis (CHCl
3
/methanol, 5:1). After 6 h, the reaction mixture was cooled to 0oC, quenched by the addition of cold water (20 ml) and evaporated under
reduced pressure to an oily residue. The residue was separated by
chromatography on silica gel (CH
2
Cl
2
/ethanol) to give
9
as a hygroscopic white foam (387 mg, 73%):
R
f
= 0.62 (CHCl
3
/methanol, 10:1);
1
H NMR (acetone-
d
6
) [delta] 10.55 (s, amide proton), 8.03 (d,
J
2,5
= 1.28 Hz, 1H, H-2), 7.97 (d,
J
5,2
= 1.28 Hz, 1H, H-5), 7.96-7.92 (m, 2H, Bz
ortho
), 7.67-7.55 (m, 3H, Bz
meta
,
para
), 7.47-7.20 (m, 9 H, aromatic DMTr), 6.90-6.83 (m, 4H, aromatic DMTr), 6.25 (pseudotriplet,
J
= 6.33 Hz, 1H, H-1'), 4.58 (m, 1H, H-3'), 4.14 (m, 1 H, H-4'), 3.76 (s, 6H, OC
H
3
), 3.32-3.28 (m, 2H, H-5'and H-5''), 2.68-2.44 (m, 2H, H-2'and H-2''); FAB
m
/
z
303.0 (DMTr
+
), 634.0 (MH
+
); high resolution FAB
m
/
z
MH
+
(calc. 634.2553, found 634.2521); anal.: calc. for C
30
H
31
N
3
O
6
[middot]H
2
O: C, 65.80, H, 6.07, N, 7.67; found: C, 66.03, H, 5.86, N, 7.57.
1-(2
'
-Deoxy-5
'
-O-dimethoxytrityl-
[beta]
-D-ribofuranosyl)imidazole-4- (N-benzoyl)carboxamide-3
'
-O-(2-cyanoethyl-N,N-diisopropylphosphoramidite) (
10
).
Trityl derivative
9
(103 mg, 0.163 mmol) was co-evaporated with anhydrous acetonitrile (3 * 1 ml), dried
in vacuo
(room temperature, 100 mTorr) and dissolved in anhydrous CH
2
Cl
2
(2.5 ml). Diisopropylammonium tetrazolide (40 mg, 0.23 mmol) and 2-cyanoethyl
N
,
N
,
N
',
N
'-tetraisopropylphosphorodiamidite (190 mg, 0.63 mmol) were added to
this solution and the mixture was swirled gently and then allowed to stand
under nitrogen at room temperature for 2 h. The reaction was then cooled to 0oC, quenched by the dropwise addition of methanol (0.5% triethylamine added)
and evaporated under reduced pressure to an oily residue which was dissolved in
CH
2
Cl
2
, washed with saturated NaHCO
3
(twice), dried over Na
2
SO
4
and purified by chromatography on a silica gel-based chromatotron plate using a mixed solvent of hexane/ethyl acetate/TEA
as the eluent solvent. Compound
10
was obtained as a yellow foam (33 mg, 25%) which rapidly deteriorated in the absence of triethylamine:
R
f
= 0.25 (CH
2
Cl
2
/ethyl acetate/triethylamine, 45:45:10);
31
P NMR (acetone-
d
6
) [delta] 149.96, 149.85 (phosphoramidite diastereomers);
1
H NMR (acetone-
d
6
) [delta] 10.49 (s, amide proton), 8.08-7.92 (m, 4H, H-2, H-5, Bz
ortho
), 7.68-7.55 (m, 3H, Bz
meta
,
para
), 7.45-7.21 (m, 9H, aromatic DMTr), 6.91-6.82 (m, 4H, aromatic DMTr), 6.30 (pseudotriplet,
J
= 6.50 Hz, 1H, H-1'), 4.80-4.65 (m, 1H, H-3'), 4.37-4.24 (m, 1H, H-4'), 3.77 (s, 6H, OC
H
3
), 3.72-3.59 (m, 4H, OC
H
2
and C
H
), 3.36 (m, 2H, H-5'and H-5''), 2.79-2.61 (m, 4H, C
H
2
CN and H-2'and H-2''), 1.25-1.18 (m, 12H, C
H
3
); FAB
m
/
z
303.2 (DMTr
+
), 834.5 (MH
+
); high resolution FAB mass MH
+
(calc. 834.3632, found 834.3658).
Incorporation of nucleoside
1
into oligodeoxyribonucleotides.
Phosphoramidite
3
(90 mg, 0.123 mmol) was dissolved in anhydrous acetonitrile (3 ml) and loaded
on the DNA synthesizer. This solution was used for the synthesis of several oligodeoxyribonucleotides on a 1 [mu]mol scale. Spectrophotometric trityl assay (A
500
) of each coupling cycle showed that
3
coupled with a coupling yield of only 62%, while additional couplings, which utilized commercially available deoxyribonucleoside phosphoramidite derivatives, demonstrated an average stepwise coupling efficiency of >98%.
Preparation of phosphoramidite
10
in situ and synthesis of oligodeoxyribonucleotides containing
1
.
Trityl derivative
9
(0.185 g, 0.29 mmol) was dried by repeated co-evaporations (3 * 2 ml) with anhydrous acetonitrile and dissolved in acetonitrile (2
ml) under nitrogen. A solution of tetrazole in anhydrous acetonitrile (0.46 M, 0.635 ml) along with 2-cyanoethyl
N
,
N
,
N
',
N
'-tetraisopropylphosphorodiamidite (84 mg, 0.28 mmol) were added to
this solution. The reaction mixture was allowed to stand, with occasional
gentle swirling, at room temperature for 2 h. TLC analysis showed the complete
absence of starting material and the presence of a diastereomeric mixture of
products (
R
f
= 0.25, CH
2
Cl
2
/ethyl acetate/TEA, 45:45:10). The crude reaction mixture was filtered through a nylon syringe filter to remove organic salts. The
concentration of the solution was adjusted to 0.063 M (based on the theoretical
yield of
10
) by the addition of anhydrous acetonitrile and the solution loaded onto a
commercial DNA synthesizer to be used for the synthesis of
oligodeoxyribonucleotides on a 1 [mu]mol scale by otherwise standard solid phase phosphoramidite methodology. Spectrophotometric trityl assay analysis (A
500
) of each coupling cycle revealed that phosphoramidite
10
coupled in 95% yield. Additional couplings, which utilized commercially available deoxyribonucleoside phosphoramidite
derivatives, proceeded with an average stepwise coupling efficiency of >98%.
Thermal denaturation studies
Solution preparation
. The stock solutions were prepared by dissolving each oligonucleotide in a pH 7 buffer consisting of 1.0 M NaCl, 10 mM sodium phosphate and 0.1 mM EDTA. The concentration of oligonucleotides was determined by UV spectroscopy, based on an assumption that at high temperature oligonucleotides are unpaired and unstacked. .An aliquot of the stock solution (50-70 [mu]l) was diluted to 2.7 ml and the UV melting curve was recorded. The
upper baseline in each UV melting curve was fitted to a straight line (
y
= a + b
x
). The absorbance at 25oC was then calculated using this equation. The extinction coefficient of an
oligonucleotide at 25oC was taken as the sum of the mononucleotides in the strand. Beer's law was
applied to calculate the concentration of oligonucleotide for each solution. All stock solutions were
refrigerated (4oC) during the course of the experiments.
UV melting measurements.
Absorbance versus temperature profiles were recorded in 1 cm path length fused
quartz cuvettes at 260 nm on a Cary 3tm UV-visible spectrophotometer equipped with a Peltier temperature
control device and thermal software. All the samples were degassed in a vacuum
desiccator before use. The oligodeoxyribonucleotide strand concentration was ~15 [mu]M. A layer of silicone oil (Dow 200 fluid, 100 CSTKS) was placed on the
surface of the aqueous solution to prevent solvent evaporation. Nitrogen was
continuously run through the measurement chamber to prevent condensation of
water vapor at low temperature. Samples were heated and cooled at a rate of 0.5oC/min. Data was collected in 1oC increments.
Data analysis
. The data from the Cary 3 spectrometer was exported in ASCII format to the
software application Igor (Wavemetrics Inc.) for curve fitting. The mathematical expression for the curve was then calculated in the software application Mathematica
(Wolfram Research). Analysis of melting curves was accomplished using the
equation developed by Gralla and Crothers (
21
),
[part][alpha]/[part](1/
T
) = -[alpha](1 - [alpha])[Delta]
H
/(1 + [alpha])
R
,
where [alpha] is the fraction of strands bonded in a helix. From this equation, at the
maximum of the derivative curve, [alpha] = 0.414 and [part][alpha]/[part](1/
T
)
max
= 0.172[Delta]
H
/
R
. When [alpha] = 0.5, [part][alpha]/[part](1/T)
[alpha] = 0.5
= 0.167[Delta]
H
/
R
. Since
T
m
is defined as the temperature at which [alpha] = 0.5, the value for the
T
m
can be found by multiplying [[part][alpha]/[part](1/
T
)
max
] by [0.167/0.172 ]. Note that
T
m
is always smaller than
T
max
.
The transition enthalpy was calculated from the equation,
[Delta]
H
= 4.37/(1/
T
max
- 1/
T
3/4
),
where
T
max
is the temperature (in Kelvin) at the maximum of the differential melting
curves and
T
3/4
is the temperature at the upper half-height of the differential melting curves. In this analysis [Delta]
C
p
was assumed to be 0 (
22
), thus enthalpy and entropy are independent of temperature.
For self-complementary sequences, the equilibrium for duplex formation is expressed
as,
K
= [alpha] /2(1 - [alpha])
2
C
t
,
where
C
t
is the total strand concentration (
23
). At
T
=
T
m
, where [alpha] = 0.5, it can be seen that
K
Tm
= 1/
C
t
.
The standard free energy at
T
=
T
m
is calculated from the equation,
[Delta]
G
Tm
=
RT
m
ln
K
Tm
.
Entropy values are then calculated from the relationship,
[Delta]
S
= ([Delta]
H
- [Delta]
G
Tm
)/
T
m
Free energies are reported under standard conditions (298 K, 15 [mu]M strand concentration, 1 M salt).
Molecular modeling
B-DNA was generated on a Silicon Graphics IRIS 4D/120GTX using the software application QUANTA. The lowest energy conformation was calculated using the Newton-Raphson minimization equation performed in CHARMm. Nucleoside
1
was placed opposite each of the natural bases in a duplex created from d(5'-CCT TTT T
1
T TTT TGG-3') and d(5'-CCA AAA AXA AAA AGG-3'), where X = A, C, G or T. Structures
were examined in four different conformational arrangements achieved by
rotating about the bond connecting the carboxamide group to the imidazole
(rotation a) and the glycosidic bond (rotation b). The values of C-1' to C-1' and the angle [lambda] were obtained by direct measurement of the model
following minimization.
RESULTS AND DISCUSSION
Synthesis
Unmodified and dI-containing oligodeoxyribonucleotides were prepared from commercially available dI, dA, dC and T phosphoramidite derivatives by standard solid phase phosphoramidite chemistry (
24
,
25
). Synthesis of oligonucleotides containing nucleoside
1
could be achieved by use of phosphoramidite
3
(Fig.
2
), but both the yields for the preparation of phosphoramidite
3
from nucleoside
1
and the incorporation of
3
into an oligonucleotide were relatively low. These problems appear to stem from
unusual reactivity of the amide group towards phosphitylating reagents.
Thermal denaturation studies
UV thermal melting studies were pursued in order to assess the effects of
1
on duplex stability relative to the natural base pairs and base pairs between
deoxyinosine and the natural bases. For these studies, a modified version of the Dickerson deoxydodecamer (
39
) d(CGCXAATTYGCG)
2
(X =
1
, I, A, C or G; Y = A, C, G or T) was chosen. The structures of many variations of this sequence, containing
either matched or severely mismatched base pairs, have been studied extensively by X-ray crystallography, NMR spectroscopy and optical thermal melting (
14
,
39
-
43
). The thermal studies have demonstrated that helix-coil transitions involving the Dickerson deoxydodecamer, especially at high concentrations, are commonly bimolecular in nature. However, it is possible that self-complementary sequences may form hairpin structures (
44
).
For comparative purposes the UV melting curves of self-complementary duplexes containing nucleoside
1
as well as the duplexes containing deoxyinosine at the same position are shown
in Figures
5
and
6
. The melting profiles show that sequences containing nucleoside
1
associate and dissociate with a cooperativity like that of sequences containing natural nucleobases.
Molecular modeling
It was of interest to determine how well imidazole 4-carboxamide fits opposite each of the natural bases in a B-DNA model in comparison with the natural base pairs. The association
of nucleoside
1
with dG and T is illustrated in Figure
7
.
CONCLUSION
The imidazole carboxamide nucleoside analog
1
was successfully incorporated into oligodeoxyribonucleotides. The apparent reactivity of the amide functionality of
1
necessitated an acyl protecting group in order to obtain acceptable DNA
synthesis coupling yields. Thermal denaturation studies of oligodeoxyribonucleotides containing
1
opposite dA, dC, dG or T reveal that this nucleoside, like the prototypical
universal nucleoside dI, destabilizes the double helix and is highly biased in its base pairing characteristics. Due to the structural similarity to dA and the parallel relative stabilities
of base pairs involving
1
to those previously observed for dA base pairs, this compound can be considered
as a dA mimic when utilized as a component of DNA. Base pairs involving dI,
especially dI[middot]dG and dI[middot]T base pairs, can contribute to significant distortion of the
double helix. This seriously limits the utility of dI in many
in vitro
DNA/RNA applications. Therefore, synthetic oligonucleotides which contain
1
, or contain
1
in conjunction with dI, might be useful in various
in vitro
assay procedures. Although
1
does not fulfill all of the requirements of a universal nucleoside, the results
of this study are being used in the design of other less discriminate
nucleosides with more favorable base pairing properties. Comparison of the
results of the thermal denaturation study with the results of DNA replication
studies demonstrate that there is not a clear relationship between the ability
of polymerase to recognize nucleoside
1
and the stability of base pairs containing
1
. Thermodynamically
1
prefers to pair with either T or G. As its triphosphate, it appears that
1
can replace dATP and to a lesser extent dGTP as a substrate for
Taq
DNA polymerase (
7
). Additional studies utilizing oligodeoxyribonucleotides containing
1
and structurally related azole nucleoside analogs to clarify the relationship
between base structure and DNA replication are currently in progress.
ACKNOWLEDGEMENTS
The National Institutes of Health and the National Cancer Institute are
gratefully acknowledged for support of this research. Helpful discussion with
V.J.Davisson, G.Hoops and N.Paul are also appreciated.
REFERENCES
1 Bergstrom,D.E., Zhang,P., Toma,P.H., Andrews,C.A. and Nichols,R. (1995) J. Am. Chem. Soc., 117, 1201-1209.
*To whom correspondence should be addressed at: Department of Medicinal
Chemistry and Molecular Pharmacology, 401 Hansen Building, Purdue University, West Lafayette, IN 47907, USA. Tel: +1 317 494 6275; Fax: +1 317 494 9193;
Email: bergstrom@pharmacy.purdue.edu
