Oligonucleotides with (N-thymin-1-ylacetyl)-1-arylserinol backbone: Chiral acyclic analogs with restricted conformational flexibility

Article abstract of DOI:10.1016/S0040-4020(00)01099-1

All four threoerythro stereoisomers of 2(R/S)-(N-thymin-1-ylacetyl)-amino-1(R/S)-aryl-1,3-propanediol were synthesized from 2(R/S)-amino-1(R/S)-aryl-1,3-propanediol in 45-50% overall yield. The inversion of the C1 hydroxyl group in (1S, 2S), 4a, and (1R, 2R), 4d, was accomplished under Mitsunobu conditions to get (1R, 2S), 4c, and (1S, 2R), 4e isomers, respectively. Compounds 4a-f were individually converted into their respective amidite synthons 5a-f. All these stereoisomers were individually incorporated into oligonucleotides (ODNs) at pre-determined positions and various biophysical studies of their hybrids with complementary DNA were carried out. All the four stereoisomers when present at 3′/5′ terminal positions in the ODNs were almost equally efficient in their binding capacity as the natural oligomers, with (1S, 2S) being marginally favored over other stereoisomers. The incorporation of these chiral acyclic nucleosides also protected the ODN against enzymatic degradation.

Full text of DOI:10.1016/S0040-4020(00)01099-1

TETRAHEDRON
Pergamon
Tetrahedron 57 (2001) 1311±1321
Oligonucleotides with (N-thymin-1-ylacetyl)-1-arylserinol
backbone: chiral acyclic analogs with restricted
conformational ¯exibility
Vipul S. Rana, Vaijayanti A. Kumarp and Krishna N. Ganesh
Division of Organic Chemistry (Synthesis), National Chemical Laboratory, Pune 411008, India
Dedicated to the memory of the late Dr V. N. Gogte
Received 13 July 2000; revised 31 October 2000; accepted 16 November 2000
AbstractÐAll four threo/erythro stereoisomers of 2(R/S)-(N-thymin-1-ylacetyl)-amino-1(R/S)-aryl-1,3-propanediol were synthesized from
2(R/S)-amino-1(R/S)-aryl-1,3-propanediol in 45±50% overall yield. The inversion of the C1 hydroxyl group in (1S, 2S), 4a, and (1R, 2R), 4d,
was accomplished under Mitsunobu conditions to get (1R, 2S), 4c, and (1S, 2R), 4e isomers, respectively. Compounds 4a±f were individually
converted into their respective amidite synthons 5a±f. All these stereoisomers were individually incorporated into oligonucleotides (ODNs)
at pre-determined positions and various biophysical studies of their hybrids with complementary DNA were carried out. All the four
stereoisomers when present at 30/50 terminal positions in the ODNs were almost equally ef®cient in their binding capacity as the natural
oligomers, with (1S, 2S) being marginally favored over other stereoisomers. The incorporation of these chiral acyclic nucleosides also
protected the ODN against enzymatic degradation. q 2001 Elsevier Science Ltd. All rights reserved.
1. Introduction
favorable pre-organization of PNA.10 Although PNA show
tight complexation with DNA at salt concentration much
lower than cellular salt concentrations, their poor aqueous
solubility and low cell uptake has opened scope for further
chemical modi®cations. The progress in this direction gave
way to the hybrid molecules like PHONA,11 ether linked
ODNs,12 AANA,13 PPNA,14 chimeric PNA:DNA sequen-
ces15a and positively charged chiral PNA oligomers15b,c to
include advantages of both the structures like superior enzy-
matic stabilities, aqueous solubility and speci®c binding
ability. The unfavorable entropic factor in acyclic analogs
for duplex formation, seems to be adequately compensated
by the additional rigid amide linkage to the nucleobase,
restricting the degrees of freedom as in 2-deoxy-d-ribose
sugar, assuming that the enthalpic contribution of molecular
recognition remains unaffected. Additional favorable
conformational restrictions16 to natural DNA as in bicyclo-
DNA,17 or 20-substituted 20-deoxy-d-ribose containing
DNA are known to further stabilize the DNA duplexes.
The hybridization properties of oligonucleotides via
Watson±Crick or Hoogsteen base pairing are of funda-
mental importance for their applications in molecular
biology and biotechnology.1 In addition to the base ®delity,
other necessary physico-chemical requirements for employ-
ing oligonucleotides (ODNs) as chemotherapeutic anti-
sense/antigene agents are enzymatic resistance, solubility
in aqueous media and ability to penetrate the cell
membrane.2±4 The naturally occurring phosphodiester±
sugar backbone of DNA largely provides the aqueous solu-
bility but is susceptible to digestion by intracellular
enzymes and hence, these ODNs have too short a half-life
to be useful as therapeutic agents. The replacement of the
sugar ring with open chain structures as in acyclic ODNs
endows enzymatic stability but at the cost of their speci®c
binding to the target sites.5±8 Studies on open chain struc-
tures were almost discouraged till the advent of peptide
nucleic acids (PNA).9 The rigid achiral polyamide structure
of PNA does not resemble the sugar phosphate backbone in
any way but still binds avidly to the natural DNA/RNA
sequences with very high sequence speci®city. The tight
binding of PNA:DNA complexes was attributed to the
absence of charge repulsion and an entropic gain due to
Encouraged by these reports, we thought of imparting back-
bone rigidity into serinol derived acyclic DNA analogs by
the introduction of substituents in the acyclic backbone.
This would give an opportunity to study the effect of a
second chiral center in the backbone, with an intrinsic
synthetic advantage of dealing with one primary and one
secondary hydroxyl function in a precursor instead of two
primary hydroxyl groups as in most other acyclic backbone
modi®cations. Our choice was 2-amino-1-aryl-1,3-propane-
diol, a synthetic precursor for the broad spectrum antibiotic
Keywords: oligonucleotides; acyclic backbone; antisense/antigene thera-
peutics.
Corresponding author. Tel./fax: 191-20-589-3153;
e-mail: vkumar@dna.ncl.res.in
p
0040±4020/01/$ - see front matter q 2001 Elsevier Science Ltd. All rights reserved.
PII: S0040-4020(00)01099-1

1312
V. S. Rana et al. / Tetrahedron 57 (2001) 1311±1321
Figure 1. Stereostructures of modi®ed nucleosides.
chloramphenicol and the preliminary results were reported
in a communication.18 In this paper we present the detailed
synthesis of all four threo/erythro stereoisomers of 2(R/S)-
(N-thymin-1-ylacetyl)-amino-1(R/S)-aryl-1,3-propanediol
(Fig. 1), their incorporation into DNA and the biophysical
studies of their hybrids with complementary DNA. The
phenyl substituent in this backbone may restrict the confor-
mational mobility of the acyclic chain, and also introduces
Scheme 1. Reagents: (i) ClCH2, (90±95%); (ii) Thymine, anhy. K2CO3, DMF, (70±80%); (iii) DMTrCl, Pyridine, (80±85%); (iv) a) PhCOOH, PPh3, DIAD,
dry THF, (80±85%); b) MeOH/NH3, (90%); (vii) {[(CH3)2CH]2N}2POCH2CH2CN, Tetrazole, EDC (80%).

V. S. Rana et al. / Tetrahedron 57 (2001) 1311±1321
1313
Scheme 2. Oligonucleotide sequences.
a second chiral center equivalent to the C30 stereogenic
center of deoxyribose sugar, leading to four possible
stereoisomers. This would provide a very good model for
understanding the importance of the threo d-natural cyclic
sugar ring based DNA backbone compared to the acyclic
analogs with two chiral centers equivalent to C40 and C30 of
the sugar and also the base attachment through a rigid amide
bond instead of C10. Additionally, depending on its stereo-
chemistry, the phenyl substituent at the 30 equivalent
center of 2-deoxy-d-ribose may be suitably inclined to
show favorable stacking interactions as in the case of
aromatic PNA.19 The effect of substituents on the phenyl
ring to determine the electronic effects, incorporation of
these monomeric thymine units in duplex/triplex forming
ODNs at predetermined positions and their biophysical
evaluation including enzymatic stability are all reported.
product 4, obtained after column chromatography, was
characterized by NMR (1H, 13C) spectra, which were
identical for both (1R, 2R) and (1S, 2S) stereoisomers. The
inversion of the C1 hydroxyl in (1S, 2S) 4a and (1R, 2R) 4d
was accomplished under Mitsunobu conditions, using
benzoic acid as a nucleophile followed by hydrolysis to
get (1R, 2S) 4c and (1S, 2R) 4e, respectively. The p-nitro
substituted phenyl ring in 4d was hydrogenated using 10%
Pd±C in methanol/pyridine/water (1:0.5:1, v/v/v) to
generate the corresponding p-aminophenyl derivative in
quantitative yield. The p-amino group was then protected
by benzoylation using benzoyl chloride in pyridine to give
4f in 90% yield. All new compounds were characterized for
structural purity by 1H, 13C NMR, optical rotation and mass
spectrometry. Compounds 4a±f were individually phos-
phitylated at the secondary hydroxyl group by 2-cyano-
ethyl-N,N,N0,N0-tetraisopropyl phosphorodiamidite reagent
in the presence of tetrazole to afford the respective amidite
synthons 5a±f.22 These were individually characterized by
31P NMR spectroscopy, which showed two signals between
148.7 and 150.9 ppm.
1.1. Synthesis of 3-O-dimethoxytrityl-1-O-benzoyl-
2(R/S)-(N-thymin-1-ylacetyl)-amino-1(R/S)-aryl-1,3-
propanediol and their phosphoramidites
The optically active acyclic nucleoside analogs 5a±f bear-
ing thymine were prepared from 2-amino-1-aryl-1,3-pro-
panediol 1 (arylˆphenyl or p-nitrophenyl)20 as shown in
Scheme 1. The ®rst step involved the reaction of 2(S/R)-
1.2. Synthesis and characterization of ODNs containing
chiral acyclic backbone
The unmodi®ed ODNs were synthesized using standard
b-cyanoethyl phosphoramidite monomers. The phosphor-
amidites of 2(R/S)-(N-thymin-1-ylacetyl)-amino-1(R/S)-
aryl-1,3-propanediol (5a±f) were incorporated at various
desired sites in ODNs (Scheme 2) using solid phase (CPG
resin) synthesis on an automated DNA synthesizer. The
modi®ed ODNs were synthesized by a slight modi®cation
in the standard procedure23 with the coupling time of the
modi®ed monomers increased to 15 min, to ensure com-
pletion of the reaction. The coupling ef®ciencies of the
modi®ed amidites were similar (.99%) to those of the
normal amidites. After the completion of the synthesis,
the ODNs were cleaved from the resin and all the base
and phosphate protecting groups were simultaneously
hydrolyzed by aqueous ammonia treatment to yield the
completely deprotected ODNs. The modi®ed ODNs were
puri®ed by polyacrylamide gel electrophoresis using 20%
PAGE and their purity was ascertained by reverse phase
HPLC. The retention of modi®cation in ODNs was
con®rmed by MALDI-TOF mass spectral results for 10.
amino-1(S/R)-aryl-1,3-propanediol
with
chloroacetyl
chloride in the presence of aqueous Na2CO3 and dioxane
to obtain the N-chloroacetyl derivative 2 in more than 80%
yield. This was then alkylated with the nucleobase, thymine,
in the presence of anhydrous K2CO3 to get the thyminyl diol
3, the primary hydroxyl function of which was directly
protected by the 4,40-dimethoxytrityl group.21 The pure
Table 1. UV Tm (8C) of duplexes of ODN 8 with modi®ed ODNs 9±13a,b
Control
1
1S, 2Sc
1R, 2Rc
1R, 2Rd
1S, 2Rd
10 11
Entry
2
3
4
5
6
7
8
9
ODN
Tm
9
58
10 11 12 13 12 13 10
58 49 59 55 58 54 56
13
53
10 13
57 53
a All values ^0.58.
b 10 one acyclic unit at 30-end of the sequence, 11 one acyclic unit at the
center of the sequence, 12 one acyclic unit at the 50-end of the sequence,
13 three acyclic units at the 30-end of the sequence.
c ArˆPh.
d Arˆp-NO2ZPh.

1314
V. S. Rana et al. / Tetrahedron 57 (2001) 1311±1321
4, Fig. 2) has no effect on duplex Tm as compared to the
control (Table 1, entry 1). Increasing the number of modi-
®cations to three at the 30-end (Table 1, entry 5) slightly
destabilized the duplex (DTm/modˆ218C). In contrast,
even a single modi®cation at the center of the duplex
(Table 1, entry 3) caused a large destabilization (DTmˆ
298C) (Fig. 2). In the case of ODNs containing the (1R,
2R) isomer, the duplex Tm showed a similar trend, i.e. the
50-end modi®cation was as good as the control (Table 1,
entry 7). Three modi®cations at the 30-end (Table 1, entry
6) caused a slight destabilization (DTm/modˆ218C).
The incorporation of one unit of the erythro (1S, 2R) isomer
at the 30-end showed Tm (Table 1, entry 10) as good as that
of the control, whereas three modi®cations with this stereo-
isomer (Table 1, entry 11) showed a slight destabilization
(DTm/modˆ21.58C), similar to that observed with duplexes
containing other ODNs (entries 6±11). Changing the substi-
tution at the para position of the phenyl ring in the backbone
(Arˆp-nitrophenyl) also had a negligible effect upon duplex
formation with either a single end modi®cation (DTm/
modˆ218C) or even three modi®ed monomeric units
(DTm/modˆ228C) in the ODNs (Table 1, entries 8, 9).
The UV melting pro®les of duplexes comprising of 8:9,
8:10, 8:11 and 8:12 are shown in Fig. 2.
Figure 2. UV melting pro®les for duplexes containing 1S, 2S (ArˆPh)
isomer (at pH 7.1). 8:9 (ÐÐ); 8:10 (± ± ±), 8:11 (WWW) and 8:12 (XXX).
(Calculated mass for molecular formula C182H234N41O120P17
average 5442.65, observed 5444.)
1.3. UV melting studies
1.3.1. Duplex melting studies
The various ODN duplexes were constituted from the
unmodi®ed ODN 9 and the modi®ed ODNs 10±13, using
the common complementary unmodi®ed 18-mer 8. Table 1
summarizes the Tm data obtained from UV melting experi-
ments of control (unmodi®ed) and modi®ed duplexes. It is
seen that the incorporation of threo (1S, 2S) isomer at single
sites at the 30-end (Table 1, entry 2) or 50-end (Table 1, entry
1.3.2. UV melting studies for triplexes
Triplexes were individually constituted from the unmodi®ed
24-mer duplex 7:6 and individual ODNs 9±13 as third
strand. Since triplex stabilities are sensitive to pH,24 values
Table 2. UV Tm of triplexes from duplex 6:7 and modi®ed ODNs 9±13a,b
Control
1
1S, 2Sc
1R, 2Rc
1R, 2Rd
11
1S, 2Rd
13
1R, 2Re
1R, 2Se
Entry
2
3
4
5
6
7
8
9
10
12
14
15
ODN
pH 5.8
pH 7.1
9
57
39
10
57
35
11
40
17
12
54
31
13
44
22
10
54
34
11
36
17
12
56
31
13
40
22
10
53
33
13
41
23
10
54
35
13
42
22
10
53
33
10
55
33
a All values ^0.58.
b 10 one acyclic unit at 30-end of the sequence, 11 one acyclic unit at the center of the sequence, 12 one acyclic unit at the 50-end of the sequence, 13 three
acyclic units at the 30-end of the sequence.
c ArˆPh.
d Arˆp-NO2ZPh.
e Arˆp-NH2ZPh.
Figure 3. UV melting pro®les for triplexes (at pH 5.8) containing (A) 1R, 2R (ArˆPhZNO2) and insetЮrst derivatives. (B) 1S, 2R (ArˆPhZNO2) 9*7:6
(ÐÐ); 10*7:6 (± ± ±); 13*7:6 (XXX).

V. S. Rana et al. / Tetrahedron 57 (2001) 1311±1321
1315
Figure 4. CD spectra of (A) monomers: (a) 1S, 2S (ArˆPh); (b) 1R, 2S (ArˆPh); (c) 1R, 2R (Arˆp-NO2ZPh); (d) 1S, 2R (Arˆp-NO2ZPh); (B) oligonucleo-
tides: (a) 9; (b) 10; (c) 11; (d) 13; (C) triplexes: (a) 9*7:6; (b) 10*7:6; (c) 13*7:6.
of UV Tm were measured at two different pH values, namely
5.8 and 7.1. All the triplexes showed characteristic biphasic
sigmoidal transitions with the transition in the lower
temperature range corresponding to the dissociation of the
third strand and that in the higher temperature range arising
from duplex denaturation. Accurate Tm values were deter-
mined from the ®rst derivative curves. The results of the Tm
data obtained from the UV melting experiments of various
triplex forming ODNs containing chiral acyclic analogs
(4a±f) are summarized in Table 2.
(1S, 2S) enantiomers showed triplex formation, with (1R,
2R) modi®cations being slightly less stable than (1S, 2S)
modi®cations. The triplex stability was also a function of
the position of the modi®cation in the sequence. For both,
(1R, 2R) isomer 4b and (1S, 2S) isomer 4a, the most stable
triplexes were observed when incorporation of the modi®ed
units was at the 30-end (Table 2, entries 2, 6), followed by a
slightly lower stability for the 50-end modi®cation (Table 2,
entries 4, 8). Increasing the number of modi®cations to three
at the 30-terminus also effected more destabilization (Table
2, entries 5, 9) with a net magnitude of almost 248C per
substitution. The most detrimental effect was observed in
At pH 5.8, ODNs containing both the threo (1R, 2R) and

1316
V. S. Rana et al. / Tetrahedron 57 (2001) 1311±1321
Figure 5. Time course of snake venom phosphodiesterase digestion of oligonucleotides: (a) 9; (b) 11 (1S, 2S; ArˆPh); (c) 10 (1S, 2S; ArˆPh); (d) 15; (e) 14
(1R, 2R; Arˆp-NO2ZPh).
the case where the modi®cation site is in the center of the
sequence (Table 2, entries 3, 7). In general, the stabilization
followed the order: control.30-end.50-end.three modi®-
cations at 30-end.central modi®cation.
differences in values of Tm among different stereoisomers
are more pronounced at lower pH, 5.8, than at neutral pH
7.1. All the complexes formed with the modi®ed backbone
show a linear increase in Tm with increasing salt concen-
tration similar to the control triplexes.
In the case of ODNs containing p-nitro-substituted mono-
mers (1R, 2R) 4d (Table 2), at pH 5.8 a single modi®cation
at the 30-end (Table 2, entry 10) destabilized the triplex
(DTm/modˆ24.58C), whereas three modi®cations at the
30-end caused a signi®cant destabilization of 258C per
modi®cation (Table 2, entry 11) (Fig. 3(A)). The triplexes
with p-aminophenyl substituted analogs (1R, 2R) 4f showed
values of Tm similar (Table 2, entry 14) to those of their
p-nitrophenyl analogs.
1.4. CD spectroscopy
The CD spectra of acyclic nucleosides (4a, 4c, 4d and 4e)
recorded in CHCl3 are shown in Fig. 4(A). Compound 4a
with (1S, 2S) stereochemistry showed a broad positive band
at 270 nm (curve a), whereas the isomer with (1R, 2R)
stereochemistry 4d showed a broad negative band (curve
c) at the same wavelength indicating the mirror image
stereochemistry to that of the (1S, 2S) isomer. The inversion
of the (1S, 2S) isomer at the C1 center of 4a generating the
(1R, 2S) isomer 4c retained the broad positive band at
275 nm (curve b). In a similar way, the spectral pattern of
(1S, 2R) as in 4e showed a negative band at 285 nm (curve
d), as that of the (1R, 2R) isomer. The CD spectra of single
stranded ODNs containing different chiral acyclic analogs
4a±4f are similar to that of the control single strand ODN. A
selected set of such CD spectra containing the (1S, 2S)
isomer 4a, at different positions in ODNs 10, 11 and 13
along with control 9 are shown in Fig. 4(B). The basic
spectral pro®les of all the ODNs are similar, irrespective
of the position (at the 30-end or in the center), or the number
of modi®cations (single or three modi®cations in an ODN).
Fig. 4(C) shows the CD spectra of triplexes constituted from
the duplex (7:6) and the modi®ed third strands containing
the chiral acyclic backbone 4a at different positions along
with the CD spectrum of the control triplex. A characteristic
negative band at around 210 nm, along with a positive band
at 275 nm and negative band around 250 nm in all cases
indicated successful triplex formation25 and no differences
were seen among the triplexes of control and modi®ed
ODNs.
In the erythro geometry, as in the case of the (1S, 2R) isomer
(Arˆp-NO2ZPh), a single modi®cation at the 30-end gives a
melting temperature identical to that of the (1R, 2R) isomer
(Table 2, entry 12) at pH 5.8. Trisubstitutions with the same
isomer at the 30-end also shows no change in the melting
pro®le (Fig. 3(B)). The trend remains the same even at pH
7.1 (Table 2, entry 13). The other erythro isomer, (1R, 2S),
obtained by the inversion of the C1 stereocenter of (1S, 2S)
isomer 4c when Ar is phenyl, also exhibited no major
changes in the binding properties of the triplex at either
pH 5.8 or 7.1 (Table 2, entry 15).
The overall melting pattern was similar but less pronounced
at pH 7.1. In the case of the control as well as the modi®ed
third strand, the values of Tm at this pH decreased by
18±238C, as expected for C containing sequences.24 The
Table 3. Rates of enzymatic digestion of different ODNs
ODN
T1/2 (min)
% Ha
9
4
3
11
5.5
0
0
24
0
11
10b
13b
14
15
.120
.120
3
1.5. Enzymatic stability of the oligomers
.120
The stability of ODNs containing the acyclic nucleosides at
different positions towards snake venom phosphodiesterase
(SVPDE) was studied by following the increase in absor-
bance at 260 nm (hyperchromicity) after addition of
SVPDE.7,26 Fig. 5 shows the time-dependent hydrolysis of
a % Hˆ[((A260 after degradation)2(A260 before addition of SVPDE))/(A260
before addition of SVPDE)]£100.
b Modi®ed oligonucleotides containing isomers 1S, 2S; 1R, 2R; 1R, 2S
(ArˆPh), 1R, 2R; 1S, 2R (Arˆp-NO2ZPh).

V. S. Rana et al. / Tetrahedron 57 (2001) 1311±1321
1317
modi®ed ODNs 10 and 11 in comparison with that of the
unmodi®ed oligomer 9. Since the hyperchromicity upon
digestion of a polypyrimidine sequence is less compared
to that of polypurines which have a better stacking, it
was decided to study the enzymatic degradation of a
mixed base sequence 14 as a typical example along with
that of unmodi®ed control 15.
tolerated better at the end of sequences, as compared to
just one at the center of modi®ed ODN. The stereochemical
preference for any one of the four isomers is not entirely
evident from the UV melting studies, though the (1S, 2S)
isomer 4a shows better stability for both duplex and triplex
formation. Aryl substitution at the C1 center was expected
to provide additional stabilization through stacking inter-
actions. The sequence 13, having three consecutive modi-
®ed monomers, was thus synthesized. The complexes
formed with this sequence in all the cases showed some
destabilization both, for duplexes (DTmˆ21±28C/mod)
and triplexes (DTmˆ24±58C/mod). A marginal preference
for the (1S, 2S) stereoisomer was again observed. The
conformational freedom in the open chain analog may
entropically disfavor the aryl ring stacking interactions
and hence any additional stability due to multiple incorpo-
ration of the modi®ed monomer is not seen. The (1S, 2S) 4a
threo isomer thus showed overall better binding ef®ciency
in open chain structure although the geometry of the corre-
sponding C30 and C40 centers is (3S, 4R) erythro in the
natural furanose ring.
The results of the enzymatic stability studies are shown in
Table 3. It is seen that the unmodi®ed ODN 9 is completely
digested in 8 min, whereas the ODNs with a single acyclic
analog substitution towards the 30-end, 10, is stable to enzy-
matic degradation even after 2 h. An oligomer with an
acyclic nucleoside in the central position of ODN, 11,
shows a hyperchromicity of about 50% of that observed
for the unmodi®ed ODN. The ODN with three consecutive
substitutions at the 30-end, 13, is also stable to enzyme
digestion. These results can be represented graphically as
shown in Fig. 5. Thus, all the different isomers of chiral
acyclic nucleosides protected the ODNs against enzymatic
degradation.
The (1S, 2R) isomer 4e, having the erythro geometry as in
deoxyribose sugar did not show signi®cantly different bind-
ing from the other isomers, 4a±d in its binding properties.
Thus, the presence of erythro geometry and a second
equivalent chiral center at C1 does not favorably assist
complex formation. The base attachment through an
amide linkage in the acyclic backbone does provide some
structural rigidity, but may not be as favorable as when
attached directly to the pentofuranose ring in the b orien-
tation as in natural nucleosides. Different substitutions on
the phenyl ring such as, p-nitro or amino functions caused
negligible change in the stability of the complexes and
hence showed the acceptance of different substitutions in
the 30/50 ends of the backbone. Such a ®nding is of potential
use, as the amino substitution can be used as a handle for the
attachment of useful linkers or ¯uorophores.
2. Discussion
Most of the ODNs reported so far with acyclic units in the
backbone show a signi®cant decrease in the stability of the
complexes formed between the modi®ed ODNs and natural
DNA or RNA.5±7 The reason attributed to this destabili-
zation is that the conformational freedom upon duplex
formation from rigid ODNs favorably overrules the entropy
loss resulting from hybridization of the ODN bearing
¯exible acyclic analogs with complementary unmodi®ed
ODNs. The well-documented chloramphenicol precursor
threo 2(R)-amino-1(R)-aryl-1,3-propanediol and its enantio-
mer 2(S)-amino-1(S)-aryl-1,3-propanediol were very easily
converted into the nucleoside analogs 4a±f in excellent
yields. The C1 center in both the isomers was easily inverted
under Mitsunobu conditions to get the set of the erythro
pair, i.e. 2(R)-amino-1(S)-aryl-1,3-propanediol and 2(S)-
amino-1(R)-aryl-1,3-propanediol. The C2 center in this
analog is equivalent to the C40 center, whereas the C1 center
is treated as analogous to the C30 center of deoxyribonucleo-
sides. Both these centers, C40 and C30 [erythro (30S, 40R)],
play a very important role in dictating the stability of natural
nucleosides when present in a ®ve-membered furanose
sugar ring structure. The substitution on the furanose ring
also affects the sugar ring puckering, which in turn directs
the helix conformation and thus, stability of the resulting
duplexes. The acyclic backbone carries a bulky aryl sub-
stituent at C1, a position equivalent to C30 in the furanose
ring. Any structural perturbation in the duplex/triplex due to
the incorporation of acyclic backbone 4a±f was expected to
be less pronounced at the 30/50 end positions of the
sequences. This is clearly demonstrated by the UV melting
studies (Tables 1 and 2). The complexes of modi®ed ODNs
10 and 12 with target ss/ds DNA were as stable as those
formed by the control ODN 9. The effect of the acyclic
backbone in ODNs on complex formation is expected to
be more evident in the central position of the ODNs and
is clearly re¯ected in the stability of the duplexes/triplexes
formed comprising modi®ed ODNs with target sequences.
Even the presence of three consecutive acyclo analogs is
The CD spectral studies indicated that even though the
monomer spectral pro®les of (1S, 2S) and (1R, 2R) isomers
(4a and 4b) are opposite to each other, upon incorporation
into ODNs, the basic spectral structure is similar to that of
the control ODN. This result reiterates that the intra-strand
base stacking in the acyclic backbone is similar to that in
natural ODNs. The CD spectra of triplexes containing the
modi®ed ODNs showed a pattern similar to that of the
unmodi®ed triplex, with the characteristic negative band
at 210 nm. This further con®rmed the overall similarity in
the structure of both, modi®ed and unmodi®ed complexes.
The overall CD results suggest that the incorporation of
different isomers does not drastically change the basic single
strand conformation of ODNs. This may be the reason for
the insensitivity of the overall binding patterns in duplexes
and triplexes to stereochemistry, since, upon incorporation,
all stereoisomers adopt a similar conformation to that of the
natural nucleosides.
Water plays an important role in the formation of a hydro-
gen-bonding network in the major and minor grooves of the
DNA duplex. Any hydrophobic substitution in this groove
or the addition of salt displaces some of the water molecules
to the bulk. This phenomenon of displacement of water

1318
V. S. Rana et al. / Tetrahedron 57 (2001) 1311±1321
molecules from the ordered hydration network to the dis-
ordered surroundings gives an entropic advantage and
thereby stabilizes the complex. The phenyl substitution in
the described acyclic backbone may cause similar hydro-
phobic desolvation locally in the major groove.27 Although
the acetamide nucleobase linker and the phenyl rings in the
acyclic backbone are hydrophobic in nature, the backbone is
composed of negatively charged phosphate groups as inter-
nucleoside linkages. The aqueous solubility of the oligo-
mers therefore was found to be as good as natural oligomers.
molecules. This is, to the best of our knowledge, the ®rst
example of an acyclic nucleoside analog with two chiral
centers forming stable duplexes and triplexes.
4. Experimental
All the chemicals used were of laboratory or analytical
grades. All the solvents used were puri®ed according to
the literature procedures. TLC was performed on Merck
pre-coated 60 F254 plates and the spots were rendered visible
by UV light and/or as dark spots after spraying with per-
chloric acid in ethanol (60%) followed by charring. Column
chromatography was carried out for puri®cation of
An important requirement to be met by ODNs as potential
therapeutic agents in the antisense or antigene strategy, is
their stability against degradation by nucleases present
within the cell and in serum. The main nuclease activity
present in fetal calf serum is that of 30-exonuclease.28 The
oligomers containing chiral acyclic nucleosides I, with
different isomers at their 30-end, show extremely high
stability towards SVPDE. From the observed results of
enzymatic digestion experiments, it can be concluded that
SVPDE is not able to cleave the phosphodiester linkage
between a natural nucleoside and acyclic nucleoside I. As
a result, ODNs with modi®cation at the 30-end did not show
any hyperchromicity upon enzymatic hydrolysis, whereas,
an unmodi®ed ODN is completely broken down within
8 min. An oligomer with an acyclic nucleoside in the center
showed a hyperchromicity about 50% of that observed for
the unmodi®ed ODN resulting most probably from sequen-
tial degradation of a 19-mer ODN to a 10-mer after which it
is stabilized by the presence of the acyclic nucleoside at its
30-end.
1
compounds on Loba silica gel (100±200 mesh). H NMR
(200 MHz) and 13C NMR (50 MHz) spectra were recorded
on a Bruker ACF 200 spectrometer ®tted with an Aspect
3000 computer. All chemical shifts are referred to internal
TMS unless otherwise mentioned and chemical shifts are
expressed in d scale (ppm). 31P NMR spectra were recorded
at 81 MHz with 85% H3PO4 as external reference. Optical
rotations were measured on a JASCO DIP-181 polarimeter.
4.1. General procedure for N-acetylation
The starting amine (1 mmol) was taken in dioxane (50 ml)
containing 10% aq. Na2CO3 (50 ml, pH,8.5). The mixture
was cooled in an ice-bath and 2-chloroacetyl chloride was
added in two portions (2.5 mmol each). After 5±10 min, the
pH was adjusted to ,8.5 by the addition of aq. Na2CO3. The
reaction mixture was concentrated to half its volume and
extracted with ethyl acetate (80 ml£3). The organic layer
was dried over anhydrous Na2SO4, concentrated and puri-
®ed by column chromatography to obtain a thick colorless
liquid which solidi®ed after keeping.
3. Conclusion
The present study demonstrates that the incorporation of
acyclic nucleosides derived from precursors of chloram-
phenicol with two chiral centers within the normal DNA
backbone, leads to the formation of stable DNA duplexes
and triplexes when present at the 30/50 ends of the ODNs.
Different substitutions on the phenyl ring only slightly
decrease the stability of the complexes, thereby proving
the versatility of this backbone for ¯uorescent labelling.
The structural adaptation of all the four stereoisomers
with C1/C2 stereogenic centers allows the oligomers to
form duplexes/triplexes with target sequences with almost
equal ef®ciency as that of the natural oligomer, the dia-
stereomer (1S, 2S) being marginally favored over other
stereoisomers. The important ®nding is that the incorpo-
ration of this chiral acyclic nucleoside at the 30-end of the
ODN protected the ODN against enzymatic degradation,
without compromising the stability of the complexes.
4.1.1. 2(S)-(N-Chloroacetyl)-amino-1(S)-phenyl-1,3-pro-
panediol, 2a. Mp 1518C;
nmax(Nujol) 3386, 1653,
1537 cm21 1H NMR (200 MHz, CDCl31D2O) d 7.35
;
(brs, 5H, Ph), 5.05 (d, 1H, Jˆ4.0 Hz, PhCH), 4.1 (m, 1H,
C2ZH), 4.0 (d, 1H, Jˆ15.1 Hz, CH2Cl), 3.89 (d, 1H,
Jˆ15.7 Hz, CH2Cl), 3.82 (d, 2H, Jˆ5.0 Hz, CH2OH). 13C
NMR (50 MHz, CDCl3) d 167.0 (CO), 140.8 (Ar), 128.2,
127.6, 125.6 (Ar), 72.0 (C-1), 62.1 (C-3), 56.6 (C-3), 42.3
(CH2Cl). [Found: C, 54.41; H, 5.34; N, 5.75. C11H14NO3Cl
20
D
requires C, 54.7; H, 5.74; N 5.8.] ‰aŠ ˆ136.0 (cˆ0.2,
MeOH).
4.1.2. 2(R)-(N-Chloroacetyl)-amino-1(R)-phenyl-1,3-pro-
panediol, 2b. Mp 1508C;
nmax(Nujol) 3380, 1652,
1537 cm21; H NMR (200 MHz) d 7.4 (bs, 5H, Ph), 5.05
(d, 1H, Jˆ4.8 Hz, PhCH), 4.2 (m, 1H, C2ZH), 4.0 (d, 1H,
Jˆ15.1 Hz, CH2Cl), 3.89 (d, 1H, Jˆ15.7 Hz, CH2Cl), 3.7
(d, 2H, Jˆ5.0 Hz, CH2OH). 13C NMR (50 MHz, CDCl3) d
167.0 (CO), 140.9 (Ar), 128.4, 127.9, 125.4 (Ar), 72.2 (C-1),
62.4 (C-3), 56.3 (C-3), 42.2 (CH2Cl). [Found: C, 54.41; H,
5.34; N, 5.75. C11H14NO3Cl requires C, 54.7; H, 5.74; N
1
The presently observed hybridization properties of the
modi®ed ODNs along with the exhibited enzymatic stability
have potential use in the design of chimeric backbone based
second generation antisense/antigene therapeutic agents.29
A novel class of nucleoside analogs (b-lactam nucleoside
chimeras) has been reported as an example of potential dual
action drugs.30 The presently used modi®cation is a close
substructure of the active pharmacophore chloramphenicol
and the hybrid molecules (monomers and oligomers) as
designed here could be further examples of this class of
20
D
5.8.] ‰aŠ ˆ237.0 (cˆ0.2, MeOH).
4.1.3. 2(R)-(N-Chloroacetyl)-amino-1(R)-p-nitrophenyl-
1,3-propanediol, 2c. Mp 968C; nmax(Nujol) 3420, 1700,
1510 cm21 1H NMR (200 MHz) d 7.5, 8.05 (dd, 4H,
;
Jˆ10.8 Hz, Ar), 7.2 (d, 1H, Jˆ6.5, CHAr), 5.34 (d, 1H,

V. S. Rana et al. / Tetrahedron 57 (2001) 1311±1321
1319
Jˆ6 Hz, C2ZH), 5.15 (Br, 1H, OH), 4.5 (Br, 1H, OH), 4.1
(m, 1H), 3.85 (d, 1H, Jˆ16.2 Hz, CH2Cl), 3.7 (d, 1H,
Jˆ16.2 Hz, CH2Cl), 3.65 (m, 2H, CH2OH). 13C NMR
(50 MHz, CDCl3) d 166.4 (CO), 148.9, 146.9 (Ar), 126.5,
122.9 (Ar), 71.02 (C-1), 61.98 (C-3), 55.95 (C-2), 42.09
(CH2Cl). [Found: C, 45.65; H, 4.85; N, 9.59.
C11H13N2O5Cl requires C, 45.7; H, 4.5; N, 9.7.]
1704, 1688, 1660, 1525, 1505, 1110 cm21
;
1H NMR
(200 MHz, CDCl3) d 8.74 (s, 1H, NH), 6.75±7.5 (m, 19H,
Ar, TZC6), 5.05 (d, 1H, Jˆ5.4 Hz), 4.2 (m, 1H), 4.1 (d, 1H,
Jˆ16.2 Hz), 3.9 (d, 1H, Jˆ16.2 Hz), 3.8 (s, 6H), 3.36 (m,
2H), 1.86 (d, 3H, Jˆ1.5 Hz). 13C NMR (CDCl3, 50 MHz) d
110.4 (C-5), 86.4 (CPh[OMePh]2), 73.3 (C-1), 62.7 (C-3),
55.1 (C-2), 54.8 (OCH3), 49.6 (CH2ZT), 11.7 (TZCH3).
FAB MS: (M1Na) 658.
20
‰aŠ ˆ210.4 (cˆ0.3, MeOH).
D
4.3.2. 3-O-Dimethoxytrityl-2(R)-(N-thymin-1-ylacetyl)-
amino-1(R)-phenyl-1,3-propanediol, 4b. nmax(Nujol) 1700,
4.2. General procedure for the preparation of N-thymin-
1-ylacetylamino-1-aryl-1,3-propanediol
1
1686, 1660, 1515, 1505, 1110 cm21; H NMR (200 MHz,
CDCl3) d 8.74 (s, 1H), 6.85±7.5 (m, 19H), 5.05 (d, 1H,
Jˆ5.4 Hz), 4.25 (m, 1H), 4.1 (d, 1H, Jˆ16.2 Hz), 3.9 (d,
1H, Jˆ16.2 Hz), 3.8 (s, 6H), 3.5 (m, 2H), 1.9 (d, 3H,
Jˆ1.4 Hz). 13C NMR (CDCl3, 50 MHz) d 110.8 (C-5),
86.4 (CPh[OMePh]2), 73.1 (C-1), 63.5 (C-3), 55.5 (C-2),
55.2 (OCH3), 50.5 (CH2ZT), 12.2 (TZCH3). FAB MS:
(M1Na) 658.
The N-chloroacetyl compound (1 mmol) was taken in dry
DMF (25 ml). Thymine (1 mmol) and anhydrous K2CO3
(1.1 mmol) were added and the reaction mixture stirred
overnight at room temperature or 558C. After completion
of the reaction as checked by TLC, the solvent was com-
pletely removed to get the crude product. Puri®cation by
column chromatography gave the compounds 3a±c as
gums which slowly solidi®ed upon keeping. In some cases
the crude material could be directly used for the protection
of the primary hydroxyl group without puri®cation as
thymine nucleobase impurity could not be completely
separated.
4.3.3. 3-O-Dimethoxytrityl-2(R)-(N-thymin-1-ylacetyl)-
amino-1(R)-p-nitrophenyl-1,3 propanediol, 4d. H NMR
1
(200 MHz, CDCl3) d 9.75 (s, 1H), 8.65 (s, 1H), 6.85±8.05
(m, 18H, Arom), 5.05 (d, 1H, Jˆ5.8 Hz), 4.25 (m, 1H), 4.3
(d, 1H, Jˆ17.6 Hz), 4.05 (d, 1H, Jˆ17.6 Hz), 3.8 (s, 6H),
3.35 (m, 2H), 1.85 (s, 3H). 13C NMR (CDCl3, 75.476 MHz)
d 110.8 (C-5), 86.2 (CPh[OMePh]2), 71.6 (C-1), 62.4 (C-3),
55.5 (C-2), 54.7 (OCH3), 49.8 (CH2ZT), 11.5 (TZCH3).
FAB MS: (M1H) 680, (M1Na) 703.
4.2.1. 2(S)-(N-Thymin-1-ylacetyl)-amino-1(S)-phenyl-1,3-
propanediol, 3a. Mp 2138C; nmax(Nujol) 3440, 1695,
1510 cm21 1H NMR (200 MHz, CDCl31DMSO-d6) d
;
11.1 (s, 1H, NH), 7.3 (m, 5H, Ph), 5.0 (d, 1H, Jˆ3.5 Hz,
PhCH), 4.3 (2£d, 2H, Jˆ15.9 Hz, CH2ZT), 4.1 (m, 1H, C-1
H), 3.65 (m, 2H, CH2OH), 1.85 (s, 3H, TZCH3). 13C NMR
(50 MHz, CDCl3) d 167.0, 158.1, 164.3 (CO), 154.8±125.6,
113.4 (Ar), 111.5 (C-5), 73.1 (C-1), 62.1 (C-3), 56.1 (C-2),
49.3 (CH2 T), 12.6 (TZCH3). [Found: C, 57.4; H, 5.67.
C16H19N3O5 requires C, 57.65; H, 5.70.] FAB MS:
(M1H) 334, (M1Na), 356. aDˆ125.0 (cˆ0.4, MeOH).
4.4. General procedure for Mitsunobu reaction and
hydrolysis of resulting benzoyl ester
The starting compound 4 (1 mmol) was taken in dry THF
(10 ml), to which, triphenylphosphine (1.5 mmol) and
benzoic acid (1.5 mmol) were added. The reaction mixture
was stirred for a few min in an ice-bath, DIAD (1.5 mmol)
was added and the reaction was stirred at ambient tempera-
ture for a few hours and monitored by TLC. After com-
pletion of the reaction, the solvent was completely
removed. The pure product was obtained after column
chromatography using 20±70% ethyl acetate with petro-
leum ether containing 0.5% pyridine as eluent. The
compound was taken in methanol (AR, 5 ml), to which,
methanol saturated with ammonia (5 ml) was added. The
reaction mixture was left tightly closed at room temperature
overnight. The solvent was completely removed and the pure
products were obtained as foams using column chroma-
tography in good yields.
4.2.2. 2(R)-(N-Thymin-1-ylacetyl)-amino-1(R)-p-nitro-
phenyl-1,3-propanediol, 3c. Mp 2578C;
nmax(Nujol)
3441, 1700, 1510 cm21 1H NMR (200 MHz, CDCl31
;
DMSO-d6) d 11.2 (s, 1H, NH), 7.65, 8.15 (2£d, 5H,
Jˆ9.6 Hz, Ar, TZC6), 5.85 (d, 1H, Jˆ6.0 Hz), 5.75 (brs,
1H, OH), 5.05 (brs, 1H, OH), 4.25 (dd, 2H, Jˆ16.1 Hz,
CH2T), 3.45 (m, 2H, CH2OH), 1.95 (s, 3H, TZCH3) 13C
NMR (50 MHz, CDCl3) d 166.1, 163.5, 150.2 (CO),
140.5, 126.4, 121.8 (Ar), 107.8 (C-5), 69.2 (C-1), 60.02
(C-3), 55.6 (C-2), 48.48 (CH2T), 10.87 (TZCH3). FAB
MS: (M1H) 378, (M1Na) 401. aDˆ151.2 (cˆ0.3, MeOH).
4.3. General procedure for dimethoxytrityl protection
4.4.1. 3-O-Dimethoxytrityl-2(S)-(N-thymin-1-ylacetyl)-
amino-1(R)-phenyl-1,3-propanediol, 4c. 1H NMR (200
MHz, CDCl3) d 6.75±7.4 (m, 19H), 5.85 (d, 1H,
Jˆ5.4 Hz), 4.2 (2£d, 2H, Jˆ16.2 Hz), 4.05 (m, 1H), 3.75
(s, 6H), 3.36 (m, 2H), 1.85 (d, 3H, Jˆ1.5 Hz). 13C NMR
(CDCl3, 50 Hz) d 110.7 (C-5), 86.5 (CPh[OMePh]2), 74.5
(C-1), 61.5 (C-3), 55.0 (C-2), 54.7 (OCH3), 50.1 (CH2ZT),
12.0 (TZCH3). FAB MS: (M1Na) 658.
The starting material (1 mmol) was co-evaporated twice
with dry pyridine (5 ml) and taken in dry pyridine (10 ml).
4,40-Dimethoxytrityl chloride (1.2 mmol) was added in two
batches in the interval of 0.5±1 h. The reaction mixture was
stirred at ambient temperature and monitored for com-
pletion by TLC. After completion of the reaction, the
solvent was completely removed and the pure product in
the form of a foam was obtained after column chroma-
tography using DCM/methanol as the eluting system.
4.4.2. 3-O-Dimethoxytrityl-2(R)-(N-thymin-1-ylacetyl)-
amino-1(S)-p-nitrophenyl-1,3-propanediol, 4e. H NMR
1
4.3.1. 3-O-Dimethoxytrityl-2(S)-(N-thymin-1-ylacetyl)-
amino-1(S)-phenyl-1,3-propanediol, 4a.
(200 MHz, CDCl3) d 9.75 (s, 1H), 8.65 (s, 1H), 6.80±8.05
(m, 18H), 5.05 (d, 1H, Jˆ5.8 Hz), 4.3 (m, 1H), 4.15 (2£d,
nmax(Nujol)

1320
V. S. Rana et al. / Tetrahedron 57 (2001) 1311±1321
2H, Jˆ16.8 Hz), 3.75 (s, 6H), 3.25 (m, 2H), 1.85 (s, 3H).
13C NMR (CDCl3, 50 MHz)
110.8 (C-5), 86.7
4.7. UV melting experiments
d
(CPh[OMePh]2), 73.7 (C-1), 61.1 (C-3), 55.5 (C-2), 55.0
(OCH3), 50.4 (CH2ZT), 12.0 (TZCH3). FAB MS: (M1H)
680, (M1Na) 703.
Duplex and triplex melting experiments were carried out in
the buffer 100mM sodium cacodylate containing 20mM
MgCl2 and 1M NaCl at the pH mentioned in each case.
Appropriate ODNs, each at a strand concentration of 1mM
based on UV absorbance at 260 nm calculated using molar
extinction coef®cients of dAˆ15.4, dCˆ7.3, dGˆ11.7,
Tˆ8.8 cm2 mmol21 were mixed and heated at 808C for
3 min, cooled to room temperature followed by overnight
storage at 48C. The A260 at various temperatures were
recorded using a Perkin±Elmer lambda 15 UV/VIS spectro-
photometer, ®tted with a water jacketed 5-cell holder and a
Julabo temperature programmer with a heating rate of
0.58C min21 over a range of 5±808C. The spectrophoto-
meter chamber was ¯ushed with dry nitrogen gas to
prevent moisture condensation at temperatures below
158C. The melting temperatures Tm were determined
from the midpoints of the transitions in the plots of
fraction absorbance change versus temperature and
were further con®rmed by differential (dA/dT versus
T ) curves. The Tm values are accurate to ^0.58C over
the reported values and are the average of three sets of
experiments.
4.4.3. 3-O-Dimethoxytrityl-2(R)-(N-thymin-1-ylacetyl)-
amino-1(R)-p-aminophenyl-1,3-propanediol, 4f. To a
solution of 3-O-dimethoxytrityl-2(R)-(N-thymin-1-ylacetyl)-
amino-1(R)-p-nitrophenyl-1,3-propanediol 4c (0.15 g,
0.22 mmol) in pyridine/methanol (6 ml, 1:1) was added
10% Pd/C (0.02 g). The reaction mixture was hydrogenated
under 40 psi hydrogen pressure for 4 h. The reaction
mixture was ®ltered and the ®ltrate was concentrated to
get product 4f (0.14 g, yield 98%), which was used for
further reaction without puri®cation.
1H NMR (200 MHz, CDCl3) d 9.75 (s, 1H), 8.65 (s, 1H),
6.85±8.05 (m, 18H), 5.05 (d, 1H, Jˆ5.8 Hz), 4.25 (m, 1H),
4.18 (2£d, 2H, Jˆ17.6 Hz), 3.8 (s, 6H), 3.35 (m, 2H), 1.85
(s, 3H).
4.5. General procedure for amidite preparation
The starting material 4 (0.5 mmol) and tetrazole (0.5 mmol)
were co-evaporated with dry dichloroethane (5±10 ml)
twice and re-dissolved in dry dichloroethane (5 ml). 2-
Cyanoethyl-N,N,N0,N0-tetraisopropylphosphorodiamidite (0.6
mmol) was added and the reaction mixture was stirred at
ambient temperature for 3 h. It was then diluted with
dichloromethane and washed with 10% aqueous sodium
bicarbonate solution. The organic phase was dried over
anhydrous Na2SO4 and concentrated to a foam. The residue
was dissolved in minimum amount of dichloromethane and
loaded on a column of dry silica gel packed in dry DCM.
The product 5 was eluted using a mixture of ethyl acetate/
dichloromethane/TEA (1:1:0.1). The compound was
obtained as a mixture of diastereomers. 31P NMR
(81 MHz, CDCl3): d 5a 1S, 2S (ArˆPh), 149.7, 149.3; 5b
1R, 2R (ArˆPh), 150.3, 149.6; 5c 1R, 2R (Arˆp-NO2ZPh),
151.2, 150.1; 5d 1R, 2S (ArˆPh), 149.7, 148.7; 5e 1S, 2R
(Arˆp-NO2ZPh), 150.9, 149.3; 5f 1R, 2R (Arˆ
p-NH2ZPh), 150.6, 149.5.
4.8. Circular dichroic spectral studies
Circular dichroism spectra were recorded on a JASCO J-715
spectropolarimeter attached to a Julabo water circulator for
maintaining the temperature. The samples were scanned in
the range of 320±200 nm at a scan speed of 200 nm min21
,
band width 1.0 nm, sensitivity 10 mdeg, resolution 0.1 nm
and response factor 2 s. Each spectrum was recorded as an
accumulation of ®ve scans using a 10 mm cell. The samples
were made in a similar manner to that for UV melting
experiments by taking 1mM of each appropriate strand.
The cell was thermostated using a Julabo water circulator
at 108C for all measurements.
4.9. Enzymatic stability of oligonucleotides
A solution of the ODNs (0.2±0.3 OD) in 2.0 ml of the pH
8.6 buffer (0.1M Tris±HCl; 0.1M NaCl; 14mM MgCl2) was
digested with 1.2 units SVPDE (Sigma) (34 ml of a solution
of the enzyme in the following buffer: 5mM Tris±HCl; pH
7.5; 50% glycerol (v/v)) at 258C. During digestion, the
increase in UV absorbance at 260 nm was followed.
The absorption versus time curve of the digestion was
plotted from which the hyperchromicity and half-life of
the oligomer were evaluated.
4.6. Oligonucleotide synthesis and puri®cation
N-Protected standard nucleoside phosphoramidites (A, T, G
and C) and nucleoside derivatized controlled pore glass
supports were purchased from Cruachem, UK. The
DNA synthesis was carried out on Pharmacia LKB-Gene
Assembler Plus. The crude ODNs were cleaved from the
support, deprotected using aqueous ammonia at 558C for
16 h and desalted using NAP-10 gel ®ltration columns.
The desalted ODNs were puri®ed by 20% PAGE under
denaturing conditions. The purity of the ODNs were
rechecked by reverse phase HPLC using the buffer systems
A: 5% CH3CN in 0.1M triethylammmoniumacetate pH 7.0
(TEAA) and B: 30% CH3CN in 0.1M TEAA using a
gradient A to B of 1.5% min21 at a ¯ow rate of
1.5 ml min21. The absorption spectrum of each peak was
scanned in the range of 200±600 nm using a diode array
detector. 10 (calculated mass for molecular formula
C182H234N41O120P17 average 5442.65, observed 5444).
Acknowledgements
V. S. R. thanks CSIR for fellowship. We thank the Bioor-
ganic Chemistry Department, Centre for Macromolecular
Science, Lodz, Poland for providing the mass spectral
analysis of the ODN. K. N. G. acknowledges the Jawaharlal
Nehru Center for Advanced Scienti®c Research, of which he
is a senior honorary faculty member.

V. S. Rana et al. / Tetrahedron 57 (2001) 1311±1321
1321
1996, 6, 1799±1804. (b) Benheida, R.; Devys, M.; Fourrey,
J.-L.; Lecubin, F.; Sun, J.-S. Tetrahedron Lett. 1998, 39,
6167±6170.
References
1. (a) Uhlmann, E.; Peyman, A. Chem. Rev. 1990, 90, 543±584.
(b) Cook, P. D. Anticancer Drug Res. 1991, 6, 585.
(c) Milligan, J. F.; Matteucci, M. D.; Martin, J. C. J. Med.
Chem. 1993, 36, 1923±1927. (d) Bennett, C. F. In Applied
Antisense Oligonucleotide Technology, Stein, C. A., Craig,
A. M., Eds.; Wiley-Liss: New York, 1998; pp 129±146.
2. Crooke, S. T. In Burgers Medicinal Chemistry and Drug
Discovery, Wolff, M. E., Ed.; John Wiley: New York, 1995.
3. Sun, J.-S.; Garestier, T.; Helene, C. Curr. Opinion Struct. Biol.
1996, 6, 327.
14. (a) Van der Laan, A. C.; Stromberg, R.; Van Boom, J. H.;
Kuyl-Yeheskiely, E.; Ef®mov, V. A.; Chakhmakhcheva,
O. G. Tetrahedron Lett. 1996, 37, 7857±7860. (b) Kehler,
J.; Henriksen, U.; Vejbjerg, H.; Dahl, O. Bioorg. Med.
Chem. 1998, 6, 315±322.
15. (a) Uhlmann, E.; Peymann, A.; Breipohl, G.; Will, D. W.
Angew. Chem., Int. Ed. Engl. 1998, 37, 2796±2823.
(b) Sforza, S.; Corradini, R.; Ghirardi, S.; Dossena, A.;
Marchelli, R. Eur. J. Org. Chem. 2000, 2905±2913.
(c) D'Costa, M.; Kumar, V.; Ganesh, K. N. Organic Lett.
1999, 1, 1513±1516.
4. Crooke, S. T. Chem. Ind. 1996, 90±93.
5. (a) Oglivie, K. K.; Nguyen-Ba, N.; Gillen, M. F.; Radatus,
B. K.; Cheriyan, U. O.; Hanna, H. R.; Smith, K. O.; Galloway,
K. S. Can. J. Chem. 1984, 62, 241±252. (b) Hakimelahi, G. H.;
Zarrinehzad, M.; Jarrahpour, A. A.; Sharghi, H. Helv. Chim.
Acta 1987, 70, 219±231.
16. Herdewijn, P. Leibigs Ann. 1996, 9, 1337±1348.
17. Bolli, M.; Lubini, P.; Leumann, C. Helv. Chim. Acta 1995, 78,
2077±2096.
18. Rana, V. S.; Kumar, V. A.; Ganesh, K. N. Bioorg. Med. Chem.
Lett. 1997, 7, 2837±2840.
6. (a) Schneider, K. C.; Benner, S. A. J. Am. Chem. Soc. 1990,
112, 453±455. (b) Azymah, M.; Chavis, C.; Lucas, M.;
Morvan, F.; Imbach, J.-L. Nucleosides Nucleotides 1992, 11,
1241±1255.
19. Tsantrizos, Y. S.; Lunetta, J. F.; Boyd, M.; Fader, L. D.;
Wilson, N.-C. J. Org. Chem. 1997, 62, 5451±5457.
20. Boehringer, C. F.; Soehne, G.m.b.H. Brit. Patent 741,711,
1955; Chem. Abstr. 1957, 51, 5830h.
7. (a) Augustyns, K.; Van Aerschot, A.; Van Schepdael, A.;
Urbanke, C.; Herdewijn, P. Nucleic Acids Res. 1991, 19,
2587±2593. (b) Vandendriessche, F.; Snoeck, R.; Janssen,
G.; Hoogmartens, J.; Van Aerschot, A.; De Clercq, E.;
Herdewijn, P. J. Med. Chem. 1992, 35, 1458.
21. Jones, R. A. In Oligonucleotide Synthesis: A Practical
Approach, Gait, M. J., Ed.; IRL Press: Oxford, 1984; pp 27.
22. Sinha, N. D.; Biernat, J.; McManus, J.; Koster, H. Nucleic
Acids Res. 1984, 12, 4539±4557.
23. Pharmacia LKB Gene Assembler Plus User's Manual.
24. Ganesh, K. N.; Kumar, V. A.; Barawkar, D. A. In Supra-
molecular Control of Structure and Reactivity, Hamilton,
A. D., Ed.; John Wiley: New York, 1996; Vol. 3, pp 267±315.
25. (a) Rana, V. S.; Ganesh, K. N. Nucleic Acids Res. 2000, 28,
1162±1169. (b) Manzini, G.; Xodo, L. E.; Gasparotto, D.
J. Mol. Biol. 1990, 213, 833±843.
8. (a) Nielsen, P.; Kiepeker, F.; Wengel, J. Nucleic Acids Res.
1994, 22, 703±710. (b) Nielsen, P.; Dreioe, L. H.; Wengel, J.
Bioorg. Med. Chem. Lett. 1995, 3, 19±28.
9. (a) Nielsen, P. E.; Egholm, M.; Berg, R. H.; Buchardt, O.
Science 1991, 254, 1497±1499. (b) Egholm, M.; Buchardt,
O.; Nielsen, P. E.; Berg, R. H. J. Am. Chem. Soc. 1992, 114,
1895±1897.
26. Xiang, G.; McLaughlin, L. W. Tetrahedron 1998, 54, 375±
392.
10. Nielsen, P. E. In Peptide Nucleic Acids, Protocols and Appli-
cations, Nielsen, P. E., Egholm, M., Eds.; Horizon Scienti®c
Press: Norfolk, 1999.
27. Rajeev, K. G.; Jadhav, V. R.; Ganesh, K. N. Nucleic Acids
Res. 1997, 25, 4187±4193.
11. Peyman, A.; Uhlmann, E.; Wagner, K.; Augustin, S.;
Breipohl, G.; Will, D. W.; Schaffer, A.; Wallmeier, H.
Angew. Chem., Int. Ed. Engl. 1996, 35, 2636±2638.
12. Merle, Y.; Bonneil, E.; Merle, L.; Sagi, J.; Szemzo, A. Int. J.
Biol. Macromol. 1995, 17, 239.
28. Shaw, J.-P.; Kent, K.; Bird, J.; Fishback, J.; Froehler, B.
Nucleic Acids Res. 1991, 19, 747±750.
29. Agrawal, S. Trends Biotechnol. 1996, 14, 376.
30. Domling, A.; Starnecker, M.; Ugi, I. Angew. Chem., Int. Ed.
Engl. 1995, 34, 2238±2239.
13. (a) Ramasamy, K. S.; Seifert, W. Bioorg. Med. Chem. Lett.