Oligonucleotides Composed of 2′-Deoxy-1′,5′-anhydro-D-mannitol Nucleosides with a Purine Base Moiety

Article abstract of DOI:10.1021/jo9718511

2′-Deoxy-D-mannitol nucleosides with a purine base moiety have been conveniently synthesized starting from 1,5-anhydro-4,6-O-benzylidene-D-glucitol. The 3-OH function of 1,5-anhydro-4,6-O-benzylidene-D-glucitol was selectively protected with tert-butyldimethylsilyl group, and the 2′-OH function was subsequently converted to the corresponding O-triflate derivative for the introduction of the nucleobase moieties. These nucleoside derivatives were transformed to 1,5-anhydro-4-O-(P-(2-cyanoethyl)-P-(N,N-diisopropylamino)phosphinyl)-2-deoxy-6- O-monomethoxytrityl-3-O-(tert-butyldimethylsilyD-D-mannitol with either a 2-(N⁶-benzoyladenin-9-yl) or a 2-(N²-isobutyrylguanin-9-yl) substituent as the building blocks for oligonucleotide synthesis. The corresponding fully modified oligonucleotides afford considerably less stable duplexes with RNA as compared to the 3-deoxy hexitol nucleic acid analogues described previously. The reason for the lower stability was investigated using molecular modeling. MD simulations of single strand MNA(GCGTAGCG) and MNA(GCGTAGCG) complexed with RNA(CGCAUCGC) in aqueous solution were performed by use of AMBER 4.1 with the particle mesh Ewald (PME) method for the treatment of long-range electrostatic interactions. Frequent hydrogen bonds between the 3′-hydroxyl and the 6′-O of the phosphate backbone of the following base changed the conformation of the single strand as well as the MNA:RNA complex. The MNA:RNA backbone widens up and shows partial unwinding and disruption of base pair hydrogen bonds consistent with their low hybridization potential.

Full text of DOI:10.1021/jo9718511

1574
J . Org. Chem. 1998, 63, 1574-1582
Oligon u cleotid es Com p osed of 2′-Deoxy-1′,5′-a n h yd r o-D-m a n n itol
Nu cleosid es w ith a P u r in e Ba se Moiety
Nafizal Hossain, Berthold Wroblowski, Arthur Van Aerschot, J ef Rozenski,
Andre De Bruyn, and Piet Herdewijn*
Laboratory of Medicinal Chemistry, Rega Institute for Medical Research, K.U. Leuven,
Minderbroedersstraat 10, B-3000, Leuven, Belgium
Received October 7, 1997
2′-Deoxy-D-mannitol nucleosides with a purine base moiety have been conveniently synthesized
starting from 1,5-anhydro-4,6-O-benzylidene-D-glucitol. The 3-OH function of 1,5-anhydro-4,6-O-
benzylidene-D-glucitol was selectively protected with tert-butyldimethylsilyl group, and the 2′-OH
function was subsequently converted to the corresponding O-triflate derivative for the introduction
of the nucleobase moieties. These nucleoside derivatives were transformed to 1,5-anhydro-4-O-
(P-(2-cyanoethyl)-P-(N,N-diisopropylamino)phosphinyl)-2-deoxy-6-O-monomethoxytrityl-3-O-(tert-
butyldimethylsilyl)-D-mannitol with either a 2-(N6-benzoyladenin-9-yl) or a 2-(N2-isobutyrylguanin-
9-yl) substituent as the building blocks for oligonucleotide synthesis. The corresponding fully
modified oligonucleotides afford considerably less stable duplexes with RNA as compared to the
3-deoxy hexitol nucleic acid analogues described previously. The reason for the lower stability
was investigated using molecular modeling. MD simulations of single strand MNA(GCGTAGCG)
and MNA(GCGTAGCG) complexed with RNA(CGCAUCGC) in aqueous solution were performed
by use of AMBER 4.1 with the particle mesh Ewald (PME) method for the treatment of long-range
electrostatic interactions. Frequent hydrogen bonds between the 3′-hydroxyl and the 6′-O of the
phosphate backbone of the following base changed the conformation of the single strand as well as
the MNA:RNA complex. The MNA:RNA backbone widens up and shows partial unwinding and
disruption of base pair hydrogen bonds consistent with their low hybridization potential.
In tr od u ction
An intriguing finding in the field of sugar-modified
oligonucleotides is the observation that oligonucleotides
composed of 1,5-anhydro-2,3-dideoxy-D-arabino-hexitol
nucleosides1 form very stable duplexes with natural
nucleic acids.1-4 The hexitol nucleic acid (HNA) obeys
the same binding rules as found in nature and with even
higher specificity.4 This is due to the positioning of the
base moiety in the 2′-position, instead of the anomeric
position, allowing the oligomer to fold in a helix-like
structure with the same geometry as found in the A-form
of dsRNA.5 The anhydrohexitol moiety of 1 (Figure 1)
can be considered as a mimic of 2-deoxy-ribo-furanose
frozen in its 2′-exo/3′-endo conformation. Studies of the
duplex stability of HNA with DNA and RNA4 reveal that
the former duplexes (HNA-RNA) are more stable than
the latter associations (HNA-DNA). This may be due
to a better complementarity of HNA with RNA (in an
A-form duplex) and/or to different degrees of hydration.
To study further this structure-function relationship, we
synthesized 2-deoxy-1,5-anhydro-D-mannitol nucleosides
(2a ,b) (Figure 1) and incorporated them into oligonucle-
otides (mannitol nucleic acids, MNA). These nucleoside
analogues have a supplementary hydroxyl group in the
F igu r e 1.
3′-â position, which we expect not to interfere with the
conformational preference of the nucleoside. The syn-
thesis of 2-deoxy-1,5-anhydro-D-mannitol nucleosides
with a pyrimidine base moiety was described previously,
starting from the D-altritol analogues.6 The configuration
in the 3′-position was inverted, making use of a pyrimi-
dine O2 neighboring group participation reaction. This
reaction cannot be used for synthesis of the purine
derivatives. Moreover, the synthetic strategy should take
into account an easy way to selectively introduce a 3′-
hydroxyl protecting group which would facilitate oligo-
merization of the new hexitol nucleoside. Therefore, we
started synthesis from 1,5-anhydro-4,6-O-benzylidene-D-
glucitol7 and introduced the 3′-O-protecting group before
attachment of the base moiety. The structure of the final
compounds was proven by chemical and spectroscopical
means before their conversion to phosphoramidites and
their use for oligonucleotide synthesis. The influence of
(1) Herdewijn, P.; Dosboszewski, B.; De Winter, H.; De Ranter, C.;
Verheggen, I.; Augustijns, K.; Hendrix, C.; Van Aerschot, A. Am. Chem.
Soc. Symp. Ser. 1994, 580, 80-99.
(2) Van Aerschot, A.; Verheggen, I.; Hendrix, C.; Herdewijn, P.
Angew. Chem., Int. Ed. Engl. 1995, 34, 1338-1339.
(3) Hendrix, C.; Verheggen, I.; Rosemeyer, H.; Seela, F.; Van
Aerschot, A.; Herdewijn, P. Chem. Eur. J . 1997, 3, 110-120.
(4) Hendrix, C.; Rosemeyer, H.; De Bouvere, B.; Van Aerschot, A.;
Seela, F.; Herdewijn P. Chem. Eur. J . 1997, 3, 1513-1520.
(5) Herdewijn, P. Liebigs Ann. 1996, 1337-1348.
(6) Pe´rez-Pe´rez, M.-J .; De Clercq, E.; Herdewijn, P. Bioorg. Med.
Chem. Lett. 1996, 6, 1457-1460.
(7) Verheggen, I.; Van Aerschot, A.; Toppet, S.; Snoeck, R.; J anssen,
G.; Claes, P.; Balzarini, J .; De Clercq, E.; Herdewijn, P. J . Med. Chem.
1993, 36, 2033-2040.
S0022-3263(97)01851-3 CCC: $15.00 © 1998 American Chemical Society
Published on Web 02/14/1998

2′-Deoxy-1′,5′-anhydro-D-mannitol Nucleosides
J . Org. Chem., Vol. 63, No. 5, 1998 1575
Sch em e 1
the 3′-hydroxyl group on the conformation of the oligo-
nucleotide was investigated through molecular modeling
and free molecular dynamics simulations. The modeling
of the complex of an octamer of mannitol nucleotides 6′-
GCGTAGCG-4′ complexes with RNA (CGCAUCGC) was
done by restraining the nucleobase positions on the
positions of the nucleobases in an ideal A-DNA:RNA
helix. Then free molecular dynamics simulations in
aqueous solution of the duplex MNA:RNA in an A-RNA
like conformation as well as of the single strand MNA,
departing from the same conformation, were performed
and the results compared with the conformation of
natural A-RNA.
To introduce the nucleobase, 4 was converted to the
corresponding mesyl derivative (5) upon treatment with
mesyl chloride in pyridine. When 5 was treated with the
sodium salt of adenine in DMF at 110 °C, compounds 9
(20%) and 10 (30%) were obtained. This reaction was
found to be not very reproducible (yields range from 10
to 20% for 9 and 30-50% for 10). Therefore, we moved
to the use of the triflate leaving group for the preparation
of 9.9,10 In this approach, 4 was converted to the
corresponding triflate derivative 6 upon treatment with
trifluoromethanesulfonic anhydride in the presence of
pyridine in CH2Cl2 at -5 °C. After standard workup 6
was reacted with tetrabutylammonium salt of adenine
in CH2Cl2 to give 9 in 48% yield. Treatment of 9 with
benzoyl chloride in pyridine at room temperature af-
forded 11, which was treated with CF3COOH in dry CH2-
Cl2 at room temperature to give 12 (85%). Compound
12 was treated with MMTrCl in pyridine to give 13 (74%).
Complete deprotection of 9 with aqueous CF3COOH gave
2a in 52% yield.
Resu lts a n d Discu ssion s
1,5-Anhydro-4,6-O-benzylidene-D-glucitol (3) was pre-
pared from commercially available bromo acetyl-R-D-
glucose, in three steps.7 Treatment of 3 with tert-
butyldimethylsilyl chloride (TBDMSCl) in pyridine at
room temperature gave 4 (31%) and 7 (40%) (Scheme 1).
The TBDMS group preferentially approaches from the
R-face leading to 7 as a major product. However, as we
need compound 4 to be the major product, other reaction
circumstances were investigated. When 3 was treated
with TBDMSCl in the presence of imidazole in DMF8 at
room temperature, 4 (58%) was formed as a major
product along with 7 (27%) in a combined yield of 85%.
Likewise, the triflate approach was used for the
synthesis of the guanine analogue (Scheme 2). This
approach gave reproducible yields both for 9 and 14.
Compound 6 was treated with tetrabutylammonium salt
of 6-iodo-2-aminopurine9,10 in CH2Cl2 to give 14 (70%).
(9) Bisacchi, G. S.; Singh, J .; Godfrey, J . D. J r.; Kissick, T. P.; Mitt,
T.; Malley, M. F.; Di Marco, J . D.; Gougoutas, J . Z.; Mueller, R. H.;
Zahler, R. J . Org. Chem. 1995, 60, 2902-2905.
(10) De Bouvere, B.; Kerremans, L.; Rozenski, J . J anssen, G.; Van
Aerschot, A.; Claes, P.; Busson, R.; Herdewijn P. Liebigs Ann. 1997,
1453-1461.
(8) Greene, T. W.; Wuts, P. G. M. Protective Groups in Organic
Synthesis, 2nd ed.; J ohn Wiley & Sons: Inc.: New York, 1991; pp 77-
83.

1576 J . Org. Chem., Vol. 63, No. 5, 1998
Hossain et al.
Sch em e 2
Sch em e 3
Compound 14 was treated with aqueous NaOH in diox-
ane at 55 °C to give 15 (96%) which can easily be
precipitated out from the reaction mixture. These alka-
line reaction conditions also removed the 3′-O-TBDMS
group. Therefore, it was necessary to reprotect the 3′-
OH group by treating 15 with TBDMSCl in the presence
of imidazole in DMF. Compound 16 (70%) was obtained
by simple filtration of the reaction mixture. Treatment
of 16 with isobutyryl chloride in pyridine followed by
removal of the benzylidene protecting group using CF3-
COOH in CH2Cl2 gave 18 in 80% yield. Compound 18,
likewise, was easily crystallized from MeOH/CH2Cl2 in
pure form. Finally, 18 was treated with MMTrCl in
pyridine to give 19 (75%) after crystallization. It is
worthwhile to mention that purification of guanine
intermediates is very easy because all intermediates
could be precipitated out or crystallized in pure form. The
fully deprotected nucleoside, 2b, was obtained from 14
in 94% yield using aqueous CF3COOH. To assign the
configuration of 4 and 7 together with the silylation site
in an unambiguous chemical way, we further trans-
formed 10 to 22 and 23 and also synthesized 25 from 8
(Scheme 3). Thus, 10 was treated with CS2 and MeI in
the presence of NaH in THF to give intermediate 20
which was converted to 21 upon treatment with n-Bu3-
SnH in the presence of AIBN in toluene. Finally, 21 was
treated with aqueous CF3COOH at room temperature to
give 22. The benzylidene protecting group in 10 was
removed upon treatment with aqueous CF3COOH to give
23. The introduction of the tert-butyldimethylsilyl pro-
tecting group in the 2-position of 7 and the 3-position of
4 was proven by chemical means.

2′-Deoxy-1′,5′-anhydro-D-mannitol Nucleosides
J . Org. Chem., Vol. 63, No. 5, 1998 1577
Ta ble 1. 1H NMR Ch em ica l Sh ifts (δ) a n d Cou p lin g
Con sta n ts (J ) of 2a a n d 2b
2a
2b
coupled
to proton
δ
J
δ
J
proton
(ppm)
(Hz)
(ppm)
(Hz)
1′ax (dd)
4.164
4.118
1′eq
2′
-13.2
-13.2
2.4
2.4
1′eq (dd)
2′ (m)
4.315
5.093
4.187
3.694
3.588
4.282
4.923
4.130
3.690
3.548
2′
3′
4′
5′
1.0
5.1
9.8
9.8
1.5
5.3
9.8
9.8
3′ (dd)
4′ (t)
5′ (dddd)
6′A
6′B
5.4
2.0
5.4
2.4
6′A (dd)
3.920
3.906
6′B
-12.2
-12.7
6′B (dd)
A2 (s)
A8 (s)
G8 (s)
4.018
8.534
8.243
-
4.005
-
-
F igu r e 2. Deconvoluted electrospray mass spectrum of the
mannitol nucleic acid 6′-AGGAGA-4′-propanediol.
8.208
The obtained building blocks were used for oligonucle-
otide assembly at a 1 µmol scale following the standard
protocol (ABI 392) except for a prolonged coupling time
of 10 min to ensure adequate coupling yields (95%
according to trityl analysis) and a prolonged acid treat-
ment (80 s) to fully deprotect the monomethoxytrityl
group. Two hexamers and a dodecamer were assembled
on a 1,3-propanediol functionalized support to avoid
synthesis of the modified supports.1-3 Following depro-
tection with ammonia and lyophylization, the residue was
treated with a 1 M TBAF solution for 22 h at rt as for
standard RNA synthesis procedures. Purification by ion
exchange chromatography yielded the desired hexamers
in moderate yield. Both displayed only a single sharp
peak on RP-HPLC analysis (chromatogram not shown).
Further treatment of the isolated peak with 1 M TBAF,
however, yielded a pattern of breakdown products. Elec-
trospray analysis of the obtained hexamers after conver-
sion to their triethylammonium salts15 went sluggishly,
but the correct mass was determined after smoothing and
deconvolution using the maximum entropy algorithm
(Figure 2).
Compound 4 was converted to the corresponding mesyl
derivative 5 which was treated with the sodium salt of
adenine in DMF. This reaction yields a mixture of 9 and
10 (compound 10 cannot be formed from the 2-O-silylated
compound 8). Compound 10 must be formed from the
(â)2,3-epoxide intermediate. Deoxygenation of the 2-hy-
droxyl group of 10, followed by removal of the benzylidene
protecting group gave compound 22. This compound
shows a clearly distinct NMR spectrum from the previ-
ously synthesized R- and â-analogues of 2-substituted 1,5-
anhydro-2,3-dideoxy-D-erythro-hexitol nucleosides7,11 which
demonstrate that opening of the â-epoxide occurred by
nucleophilic attack at the 3-position of the carbohydrate
moiety. Compound 7 was converted to the corresponding
mesyl derivative 8 which was treated with adenine in
the presence of NaH in DMF to give 25 via the epoxide
24. Compound 25 is a known product.12 These chemical
transformations unambiguously prove the location of the
silyl protecting group of the starting materials 4 and 7.
The conformation of the final compounds 2a and 2b
1
were deduced from H NMR analysis. Table 1 shows the
chemical shifts and coupling constant for all protons. The
Recently it was claimed that deprotection of tBDMSi
groups of oligonucleotides was improved using triethyl-
amine trihydrofluoride, maximizing the yield of synthetic
RNA.16 Therefore, the dodecamer after ammonia depro-
tection was divided in two parts and treated with a 1 M
TBAF solution or TEA, 3HF, respectively. The latter
method, however, only yielded fully degraded product,
while TBAF deprotection afforded a product mixture
displaying a very broad peak on ion exchange chroma-
tography at the expected position. However, isolation
and RP-HPLC analysis of the obtained material dis-
played several peaks suggesting incomplete deprotection
of the oligomer. The present methodology therefore did
not allow isolation of pure dodecamers. Only oligonucle-
otides of short length (i.e. hexamers) could be obtained.
The obtained hexamers (mannitol nucleic acids, MNA)
were analyzed for their hybridization properties via
large coupling constant obtained for J 3′,4′ and J 4′,5′ to-
gether with the lower values obtained for J 2′,3′ and J 1′eq,2′
,
J 1′ax,2′ may be indicative for the axial position of the
purine base moiety as was also observed for the 1,5-
anhydro-2,3-dideoxy-D-arabino hexitol nucleosides.7,13
The monomethoxytritylated precursors 13 and 19 were
phosphitylated under standard conditions14 to afford the
phosphoramidites 26 and 27 in 86% and 91% yield,
respectively. 13C NMR analysis displayed the signals
characteristic for cyanoethyl phosphoramidites, but com-
plete assignment proved difficult. MS confirmed the
expected structure.
(11) Hossain, N.; Rozenski, J .; De Clercq, E.; Herdewijn, P. J . Org.
Chem. 1997, 62, 2442-2447.
(12) Verheggen, I.; Van Aerschot, A.; Pillet, N.; van der Wenden, E.
M.; IJ zerman, A.; Herdewijn, P. Nucleosides Nucleotides 1995, 14, 321-
324.
(13) Verheggen, I.; Van Aerschot, A.; Van Meervelt, L.; Rozenski,
J .; Wiebe, L.; Snoeck, R.; Andrei, G.; Balzarini, J .; Claes, P.; De Clercq,
E.; Herdewijn, P. J . Med. Chem. 1995, 38, 826-835.
(14) Methods in Molecular Biology; Agrawal, S., Ed.; Humana
Press: New J ersey, 1993; Vol. 20.
(15) Ceulemans, G.; Van Aerschot, A.; Wroblowski, B.; Rozenski, J .;
Hendrix, C.; Herdewijn, P. Chem. Eur. J . 1997, 3, 1997-2010.
(16) Westman, E.; Stro¨mberg, R. Nucleic Acid Res. 1994, 22, 2430-
2431.

1578 J . Org. Chem., Vol. 63, No. 5, 1998
Ta ble 2: Tm of Hexa m er ic Ma n n itol
Hossain et al.
Va lu esa
Oligon u cleotid e An a logu es (MNA) w ith Eith er DNA,
RNA, or HNA Com p lem en ta r y Sequ en ces in Com p a r ison
w ith th e Hybr id iza tion P r op er ties of Ar a bin o Hexitol
Oligon u cleotid es (HNA)
MNA
0.1 M NaCl
1 M NaCl
1. 6′-AGGAGA-4′/RNA
2. 6′-AGGAGA-4′/DNA
3. 6′-AGGAGA-4′/HNA
4. 6′-GAGAGA-4′/RNA
5. 6′-GAGAGA-4′/DNA
6. 6′-GAGAGA-4′/HNA
-
-
14
b
23
14
b
18
-
-
15
20
HNA
7. 6′-AGGAGA-4′/RNA
8. 6′-AGGAGA-4′/DNA
9. 6′-AGGAGA-4′/HNA
10. 6′-GAGAGA-4′/RNA
11. 6′-GAGAGA-4′/DNA
12. 6′-GAGAGA-4′/HNA
45c
b
55
43c
b
55
31c
69
53
15c
69
54
a
F igu r e 3. Melting profiles for the hexamer sequence 6′-
AGGAGA (NMA and HNA) with the RNA complement 5′-
(UCUCCU) and with the HNA sequence 6′-TCTCCT as the
complement, at 4 µM in 1 M NaCl buffer (all normalized)
(MNA-RNA (s); HNA-RNA (- - -); MNA−HNA (O); HNA-
HNA (0)).
Tm (°C) were determined in
a buffer containing 0.02 M
KH2PO4, pH 7.5, 0.1 mM EDTA with 0.1 or 1 M NaCl and 4 µM
b
of each oligonucleotide (only 2 µM for line 1). No hypochromicity
detectable. c Data from ref 3.
melting experiments, which proved disappointing. Only
very weak hybridization was noticed for the mannitol
oligonucleotide analogues with complementary RNA as
well as with HNA sequences (Table 2, lines 1-6). No
hypochromicity was detected with DNA as the comple-
mentary partner while the hexitol oligonucleotides (HNA)
afford very good hybridization with RNA and display
exceptionally strong self-pairing with a complementary
HNA strand (lines 7-12). The very large discrepancy
in stability for duplexes containing the mannitol hexa-
mers versus the duplexes with arabinohexitol oligonucle-
otides (HNA) is shown in Figure 3 (with a RNA sequence
and with a HNA sequence as the complementary part-
ners). The very broad transitions for duplexes with MNA
as a partner do not allow accurate Tm determination and
suggest multiple transitions during melting. The pres-
ence of the extra hydroxyl group at the 3′â-position in
the MNA analogues has a deleterious effect on their
hybridization potential.
The free molecular dynamics simulations of the MNA:
RNA duplex as well as the MNA single strand showed a
strong influence of the axial 3′-hydroxyl group on the
conformation, which led to a considerable amount of
instability in both simulations. Both simulations exhib-
ited large conformational changes which let them drift
away from their A-RNA like starting conformations
(Figure 4). The single strand MNA partially unwound
during 300 ps of MD forming a kink at the position of
MNA (T4A5) in the center of the strand. The structure
is stabilized by frequent hydrogen bonds between the 3′
hydroxy and the 6′O in the phosphate backbone of the
following nucleotide (Figure 4). Under the same condi-
tions, single strand RNA totally unwinds, without any
remaining intramolecular hydrogen bonds or base stack-
ing interactions. During 1 ns of molecular dynamics
simulation the complex of MNA:RNA showed a similar
behavior as the single strand MNA (Figure 4). The
frequent formation of 3′ hydroxy-6′O hydrogen bonds is
the driving cause of the conformational change. The
conformational transition consists of a widening of the
nucleotide backbone with an accompanied partial un-
winding of the MNA:RNA strands. The widening is most
pronounced in the last three 4′-terminal base pairs,
where the distance between the P atoms of the two
strands increased from 17.7 to 18.5 Å. The base pair
hydrogen bonds are partially broken during MD simula-
tion, as well as the base stacking interactions are
weakened, again most pronounced in the last three 4′-
terminal base pairs. The minor groove becomes wider
(from 11.0 to 13.8 Å) and less deep as in natural A-RNA
(Figure 4). These findings suggest that the MNA is
conformational restricted and the more flexible RNA
strand has to fit to the preformed MNA conformation,
inducing conformational stress. This picture of instabil-
ity and unwinding and widening of the MNA:RNA
strands is consistent with the experimentally determined
low hybridization potential of MNA with RNA as well as
with MD simulations of natural RNA,17 HNA:RNA, HNA:
DNA18 and Altritol Nucleic Acids19 which, also consistent
with their experimentally determined hybridization po-
tentials, remain stable during MD simulation. These
findings suggest that MD simulation by use of PME may
be a useful tool for determining the stability of complexes
of modified nucleic acids with RNA or DNA. The lower
stability of the MNA:RNA duplex when compared with
the HNA:RNA duplex may primarily be due to the
presence of an intramolecular hydrogen bond and, con-
sequently, unfavorable entropy and enthalpy for binding.
Exp er im en ta l Section
Analytical instruments and general methods were described
previously.11 1H NMR spectra for 2a and 2b were recorded at
33 °C in D2O on a Varian Unity-500 spectrometer (500 MHz
for 1H). The instrument was equipped with a 5-mm i.d. PFG
probe. The FIDs were acquired in 32 K data points, with a
spectral width of 8000 Hz. Data processing involved resolution
enhancement by a Gaussian-to-Lorentzian window multiplica-
tion and zero filling to 64 K, resulting in a digital resolution
of ∼0.12 Hz/pt. Chemical shifts are listen in ppm, relative to
internal DSS (measured against the residual HDO signal as
secondary reference, δ 4.691 ppm at 33 °C). Coupling constant
(17) Cheatham, T. E. III; Miller, J . L.; Fox, T.; Darden, T. A.;
Kollman, P. A. J . Am. Chem. Soc. 1995, 117, 4193-4194. Cheatham,
T. E.III; Kollman, P. A. J . Am. Chem. Soc. 1997, 119, 4805-4825.
(18) De Winter, H.; Lescrinier, E.; Van Aerschot, A.; Herdewijn, P.
J . Am. Chem. Soc. (submitted).
(19) Wroblowski, B.; Herdewijn, P. Chem. Eur. J . (submitted).

2′-Deoxy-1′,5′-anhydro-D-mannitol Nucleosides
J . Org. Chem., Vol. 63, No. 5, 1998 1579
F igu r e 4. Initial structure (A) and average structure of the last 50 ps of single strand MNA (B) as well as initial structure (C,
E) and average structure of the last 100 ps of the MNA:RNA duplex (D, F). The phosphate-sugar backbone of MNA is highlighted
as sticks. The place of each base and the hydrogen bonds are indicated (A-D). A part of the solvent accessible surface forming the
minor groove is shown with dots (E, F). Figures (A-D) were generated using BobScript,26,27 and (E, F) by use of RasMol.28
values were derived by first-order spectral analysis. UV-
melting experiments and mass spectrometric analysis of
oligonucleotides were done as described.15
1,5-An h yd r o-4,6-O-b en zylid en e-3-O-TBDMS-D-glu ci-
tol (4) a n d 1,5-An h yd r o-4,6-O-ben zylid en e-2-O-TBDMS-
D-glu citol (7). To a solution of 3 (7.0 g, 27.7 mmol) in DMF

1580 J . Org. Chem., Vol. 63, No. 5, 1998
Hossain et al.
(200 mL) was added imidazole (2.08 g, 30.5 mmol), followed
by TBDMSCl (4.6 g, 30.6 mmol), and kept at room temperature
for 24 h. The reaction mixture was concentrated. The residue
was dissolved in CH2Cl2 (300 mL). After addition of saturated
aqueous NaHCO3 (50 mL), the two layers were separated. The
aqueous layer was extracted with CH2Cl2 (100 mL). The
combined organic layer was washed with H2O (25 mL). The
organic layer was dried, filtered, and concentrated. The
residue was purified by silica gel column chromatography (0-
15% EtOAc in hexane) to give 4 (5.9 g, 58%) and 7 (2.8 g, 27%).
When pyridine was used both as base and solvent the reactiv-
ity was altered, giving 7 as major product. Com p ou n d 4: 1H
NMR (DMSO-d6) 7.49-7.29 (m, 5H); 5.58 (s, 1H); 5.10 (d, J )
5.8 Hz, 1H); 4.16 (dd, J ) 4.3, 10.1 Hz, 1H); 3.78 (dd, J ) 5.0,
10.8 Hz, 1H); 3.69-3.13 (m, 6H); 0.86 (s, 9H); 0.5 (s, 3H); 0.1
(s, 3H). 13C NMR (DMSO-d6) 137.8, 128.8, 128.0, 126.1, 100.6,
81.4, 76.3, 70.9, 70.6, 70.4, 68.0, 25.9, 18.2. HRMS calcd for
J ) 5.6, 9.7 Hz, 1H); 4.29-3.94 (m, 5H); 3.58 (m, 1H); 1.00 (s,
9H); 0.00 (s, 3H); -0.1 (s, 3H). 13C NMR (DMSO-d6) 156.0,
152.3, 151.0, 139.9, 137.8, 128.9, 128.0, 126.1, 117.8, 100.9,
78.9, 72.1, 69.8, 69.6, 67.5, 54.6, 25.3, 17.8. HRMS calcd for
C24H34N5O4Si1 (M + H)+ 484.2379, found 484.2374. Mp: 241-
242 °C. Com p ou n d 10: 1H NMR (DMSO-d6) 8.51 (s, 1H);
8.18 (s, 1H); 7.33-7.16 (m, 5H); 5.88 (d, J ) 3.7 Hz, 1H); 5.78
(s, 1H); 5.06 (br d, J ) 3.9 Hz, 1H); 4.46 (dd, J ) 5.3, 10.0 Hz,
1H); 4.25 (dd, J ) 4.8, 9.8 Hz, 1H); 4.20-4.10 (m, 3H); 3.78
(m, 2H). 13C NMR (DMSO-d6) 156.2, 152.5, 150.6, 140.1, 137.6,
129.0, 128.1, 126.0, 118.2, 101.1, 74.3, 68.8, 68.1, 67.9, 67.0,
55.1. HRMS calcd for C18H20N5O4 (M + H)+ 370.1515, found
370.1534.
1,5-An h yd r o-2-(N6-b en zoyla d en in -9-yl)-2-d eoxy-3-O-
TBDMS-D-m a n n itol (12). To a cold (ice-water) solution of
9 (330 mg, 0.68 mmol) in pyridine (5 mL) was added benzoyl
chloride (236 µL, 2.04 mmol) and kept at room-temperature
overnight. The reaction mixture was cooled to 0 °C, and
saturated aqueous NaHCO3 (2 mL) was added and extracted
with CH2Cl2 (3 × 30 mL). The combined organic layer was
washed with H2O (10 mL), concentrated, and coevaporated
with toluene. The residue was dissolved in CH2Cl2 (50 mL)
and washed with saturated aqueous NaHCO3 (3 × 10 mL) and
H2O (10 mL). The organic layer was concentrated to give 11
which was directly treated with CF3COOH (5 mL) in CH2Cl2
(5 mL) at room temperature for 22 h. The solvent was removed
and coevaporated with MeOH and toluene. The residue was
dissolved in MeOH (10 mL) and treated with NH4OH (3 mL).
The reaction mixture was concentrated and coevaporated with
MeOH, and the residue was purified by silica gel column
chromatography (0-4% MeOH in CH2Cl2) to give 12 (290 mg,
85%). 1H NMR (DMSO-d6) 11.08 (br s, 1H); 8.78 (s, 1H); 8.76
(s, 1H); 8.16-7.50 (m, 5H); 5.20 (br s, 1H); 5.10 (d, J ) 5.1
Hz, 1H); 4.11 (m, 4H); 3.62 (m, 3H); 3.30 (m, 1H); 0.6 (s, 9H);
0.5 (s, 6H). 13C NMR (DMSO-d6) 165.7, 153.7, 151.1, 150.0,
144.0, 133.6, 132.4, 128.5, 124.4, 81.7, 73.3, 67.5, 66.7, 60.0,
54.9, 25.5, 17.7. HRMS calcd for C24H34N5O5Si (M + H)+
500.2329, found 500.2317.
C
19H31O5Si1 (M + H)+ 367.1940, found 367.1970. Com p ou n d
7: 1H NMR (DMSO-d6) 7.49-7.30 (m, 5H); 5.55 (s, 1H); 4.17
(dd, J ) 3.4, 10.1 Hz, 1H), 3.77 (dd, J ) 5.4, 10.6 Hz, 1H);
3.67-3.16 (m, 6H); 0.88 (s, 9H); 0.1 (s, 6H). 13C NMR (DMSO-
d6) 137.9, 128.9, 128.1, 126.5, 100.8, 81.1, 74.4, 72.2, 71.0, 70.2,
68.0, 25.9, 18.0. HRMS calcd for C19H31O5Si1 (M + H)+
367.1940, found 367.1972.
1,5-An h ydr o-4,6-O-ben zylid en e-3-O-TBDMS-2-O-m esyl-
D-glu citol (5). To a cold (ice-water) solution of 4 (770 mg,
2.10 mmol) in CH2Cl2 (20 mL) was added Et3N (465 µL, 3.34
mmol) followed by MsCl (243 µL, 3.14 mmol), and the reaction
mixture was kept at 0 °C for 2 h. The reaction was quenched
with ice-water and diluted to 100 mL by addition of CH2Cl2.
The organic layer was separated, washed with saturated
aqueous NaHCO3 (10 mL) and H2O (5 mL), dried, filtered, and
concentrated to give 5 (934 mg, 100%) which was used directly
in the next step. 1H NMR (DMSO-d6) 7.50-7.30 (m, 5H); 5.61
(s, 1H); 4.49 (m, 1H); 4.17 (m, 2H); 3.96 (t, J ) 8.5 Hz, 1H);
3.72-3.35 (m, 4H); 3.24 (s, 3H); 0.84 (s, 9H); 0.1 (s, 3H); -0.02
(s, 3H). 13C NMR (DMSO-d6) 137.4, 129.0, 128.0, 126.3, 101.0,
80.9, 78.5, 72.9, 70.5, 67.7, 66.9, 37.9, 25.7, 18.0. HRMS calcd
for C20H33O7S1Si1 (M + H)+ 445.1716, found 445.1797.
1,5-An h ydr o-2-(aden in -9-yl)-4,6-O-ben zyliden e-2-deoxy-
3-O-TBDMS-D-m a n n itol (9) a n d 1,5-An h yd r o-3-(a d en in -
9-yl)-4,6-O-ben zylid en e-3-d eoxy-D-a ltr itol (10). Meth od
A. A mixture of adenine (564 mg, 4.17 mmol) and NaH (80%,
188 mg, 6.27 mmol) in DMF was heated at 100 °C for 30 min.
Compound 5 (934 mg, 2.10 mmol) in DMF (2 mL) was added
and the reaction mixture kept at 110 °C for 20 h. The solvent
was removed. After addition of EtOAc (100 mL) and saturated
aqueous NaHCO3 (20 mL), two layers were separated. The
aqueous layer was extracted with EtOAc (50 mL), and the
combined organic layer was washed with H2O (20 mL), dried,
filtered, and concentrated. The residue was purified by silica
gel column chromatography (0-7% MeOH in CH2Cl2) to give
9 (200 mg, 20%) and 10 (300 mg, 30%) as a white solid. The
yield of 9 varied from 10 to 20%.
Meth od B. To a cold (ice-NaCl) solution of 4 (1.03 g, 2.81
mmol) in CH2Cl2 (30 mL), pyridine (450 µL) was added,
followed by slow addition of trifluoromethanesulfonic anhy-
dride (705 µL in 515 µL CH2Cl2) and kept at -5 °C for 2 h.
The reaction was quenched with ice-water, and the volume
was increased to 60 mL by addition of CH2Cl2. Two layers
were separated. The aqueous layer was extracted with CH2-
Cl2 (30 mL). The combined organic layer was washed succes-
sively with prechilled (at -5 °C) aqueous saturated NaH2PO4
(3 × 10 mL) and H2O (10 mL). The organic layer was dried
over MgSO4, filtered, and concentrated at 18 °C to give light
yellow solid 6 (1.4 g, 100%). To a solution of tetrabutylam-
monium salt of adenine (1.37 g, 3.64 mmol) in CH2Cl2 (36 mL)
at room temperature, 6 (1.4 g, 2.80 mmol) in CH2Cl2 (15 mL)
was added and left at room temperature for 40 h. The reaction
mixture was filtered, the precipitate washed with CH2Cl2, and
the combined organic layer was concentrated. The residue was
purified by silica gel column chromatography (0-3% MeOH
in CH2Cl2) to give 9 (650 mg, 48%). Com p ou n d 9: 1H NMR
(DMSO-d6) 8.38 (s, 1H); 8.15 (s, 1H); 7.49-7.30 (m, 5H); 7.20
(br s, 2H); 5.78 (s, 1H); 5.08 (br d, J ) 5.4 Hz, 1H); 4.40 (dd,
1,5-An h yd r o-2-(N6-b en zoyla d en in -9-yl)-2-d eoxy-6-O-
MMTr -3-O-TBDMS-D-m a n n itol (13). Compound 12 (290
mg, 0.58 mmol) was treated with MMTrCl (268 mg, 0.87 mmol)
in pyridine (5 mL) at room-temperature overnight. The
reaction mixture was cooled to 0 °C, and saturated aqueous
NaHCO3 (5 mL) was added and extracted with CH2Cl2 (3 ×
30 mL). The combined organic layer was washed with H2O
(5 mL), dried, filtered, and concentrated. The residue was
purified by silica gel column chromatography (0-1.5% MeOH
in CH2Cl2) to give 13 (330 mg, 74%). 1H NMR (CDCl3) 9.26
(s, 1H); 8.86 (s, 1H); 8.70 (s, 1H); 8.08-6.89 (m, 19H); 5.18
(m, 1H); 4.22 (d, J ) 12.8 Hz, 1H); 4.08-3.83 (m, 3H); 3.79 (s,
3H); 3.63 (m, 1H), 3.46 (m, 2H); 1.99 (d; J ) 2.2 Hz, 1H); 0.61
(s, 9H); 0.1 (s, 6H). 13C NMR (CDCl3) 164.5, 158.6, 152.9,
152.4, 149.3, 144.1, 144.0, 143.3, 135.3, 134.0, 132.6, 130.3,
128.8, 128.4, 128.0, 127.8, 127.0, 121.6, 113.3, 86.9, 80.0, 73.7,
69.0, 68.8, 62.8, 55.2, 55.0, 25.4, 17.8. HRMS calcd for
C
44H49N5O6SiNa (M + Na)+ 794.3349, found 794.3361.
1,5-An h yd r o-2-(a d en in -9-yl)-2-d eoxy-D-m a n n itol (2a ).
Compound 9 (200 mg, 0.41 mmol) was treated with 80%
aqueous CF3COOH (10 mL) at room-temperature overnight.
The solvent was removed and coevaporated with MeOH and
toluene. The residue was dissolved in MeOH (5 mL), treated
with NH4OH (3 mL), concentrated, coevaporated with MeOH
and toluene, and crystallized from MeOH/CH2Cl2 to give 2a
(60 mg, 52%). 1H NMR (D2O, 33 °C at 500 MHz). 8.53 (s, 1H);
8.24 (s, 1H); 5.09 (m, 1H); 4.31 (dd, J ) 1.0, 13.2 Hz, 1H); 4.18
(dd, J ) 9.8, 5.1 Hz, 1H); 4.16 (dd, J ) 2.4, 13.2 Hz, 1H); 4.01
(dd, J ) 2.0, 12.2 Hz, 1H); 3.92 (dd, J ) 5.4, 12.2 Hz, 1H);
3.69 (t, J ) 9.8 Hz, 1H); 3.58 (ddd, J ) 2.0, 5.4, 9.8 Hz, 1H).
13C NMR (DMSO-d6) 156.0, 152.2, 150.9, 140.4, 117.9, 81.7,
71.9, 68.0, 66.6, 60.2, 54.3. HRMS calcd for C11H16N5O4 (M +
H)+ 282.1202, found 282.1241. Anal. Calcd for C11H15N5O4‚
0.6CH3OH C, 46.37, H, 5.84, N, 23.31. Found C, 46.33, H,
5.55, N, 22.92. Mp: 232-233 °C.
1,5-An h yd r o-4,6-O-b en zylid en e-2-d eoxy-2-(N6-iod o-2-
a m in op u r in e-9-yl)-3-O-TBDMS-D-m a n n itol (14). To a so-

2′-Deoxy-1′,5′-anhydro-D-mannitol Nucleosides
J . Org. Chem., Vol. 63, No. 5, 1998 1581
lution of tetrabutylammonium salt of 6-iodo-2-aminopurine
(3.6 g, 7.16 mmol) in CH2Cl2 (30 mL) was added 6 (2.59 g, 5.19
mmol) in CH2Cl2 (5 mL) and the reaction mixture kept at room
temperature for 24 h. The precipitate was filtered off and
washed with CH2Cl2 (10 mL). The combined filtrate was
concentrated and purified by silica gel column chromatography
(0-50% EtOAc in hexane) to give 14 (2.20 g, 70%). 1H NMR
(DMSO-d6) 8.32 (s, 1H); 7.49-7.30 (m, 5H); 6.69 (br s, 2H);
5.78 (s, 1H); 4.85 (br d, J ) 5.5 Hz, 1H); 4.37 (dd, J ) 5.5, 9.5
Hz, 1H); 4.23 (dd, J ) 4.8, 9.9 Hz, 1H); 4.17-3.86 (m, 4H);
3.55 (m, 1H); 0.55 (s, 9H); 0.00 (s, 3H); -0.01 (s, 3H). 13C NMR
(DMSO-d6) 159.4, 151.5, 141.2, 137.7, 129.6, 128.9, 128.0,
126.0, 122.9, 100.9, 78.8, 72.0, 69.6, 69.1, 67.4, 54.7, 25.2, 17.8.
HRMS calcd for C24H33N5O4ISi (M + H)+ 610.1348, found
610.1398. Mp: 129-131 °C.
1,5-An h yd r o-4,6-O-b en zylid en e-2-d eoxy-2-(gu a n in -9-
yl)-D-m a n n itol (15) a n d 1,5-An h yd r o-4,6-O-ben zylid en e-
2-deoxy-2-(gu an in -9-yl)-3-O-TBDMS-D-m an n itol (16). Com-
pound 14 (2.0 g, 3.28 mmol) was treated with aqueous NaOH
(N) (50 mL) in dioxane (50 mL) at 55 °C for 60 h. The reaction
mixture was cooled to 0 °C, and the pH was adjusted to 7.0 by
addition of dilute aqueous HCl. The solvent was removed to
give a white solid. After addition of H2O (20 mL), the solid
was crushed to powder with a spatula, filtered, washed with
H2O (10 mL), and dried to give 15 (1.21 g, 96%). 1H NMR
(DMSO-d6) 7.94 (s, 1H); 7.53-7.30 (m, 5H); 6.50 (br s, 2H);
5.80 (s, 1H); 5.48 (d, J ) 5.1 Hz, 1H); 4.78 (m, 1H); 4.23 (dd,
J ) 5.0, 10.2 Hz, 1H); 4.14-3.83 (m, 5H); 3.48 (m, 1H). 13C
NMR (DMSO-d6) 157.0, 153.4, 152.6, 137.9, 136.7, 129.0, 128.1,
126.4, 115.7, 101.1, 79.0, 72.3, 69.7, 68.0, 67.6, 54.4. HRMS
calcd for C18H20N5O5 (M + H)+ 386.1464, found 386.1457. This
compound was directly used in the next reaction without
further purification.
To a suspension of 15 (1.10 g, 2.85 mmol) in DMF (45 mL)
was added imidazole (327 mg, 4.80 mmol), followed by
TBDMSCl (683 mg, 4.53 mmol), and the reaction mixture kept
at room temperature for 40 h. The solvent was removed, and
H2O (20 mL) was added, followed by EtOAc (50 mL), to give a
white precipitate. The precipitate was filtered off, washed
successively with H2O and EtOAc, and dried to give 16 (1.0 g,
70%). 1H NMR (DMSO-d6) 10.59 (br s, 1H); 7.97 (s, 1H); 7.50-
7.28 (m, 5H); 6.30 (br s, 2H); 5.77 (s, 1H); 4.78 (m, 1H); 4.38-
3.88 (m, 6H); 3.52 (m, 1H); 0.66 (s, 9H); -0.1 (s, 6H). 13C NMR
(DMSO-d6) 157.2, 153.2, 152.6, 137.8, 136.7, 128.8, 128.0,
126.0, 115.5, 100.8, 78.8, 72.0, 69.6, 69.5, 67.5, 54.2, 25.3, 17.8.
HRMS calcd for C24H34N5O5Si (M + H)+ 500.2329, found
500.2284. Anal. Calcd for C24H33N5O5Si‚1.0H2O C, 55.69, H,
6.81, N, 13.53. Found C, 55.62, H, 6.75, N, 13.84.
1,5-An h yd r o-2-d eoxy-2-(N2-isobu tyr ylgu a n in -9-yl)-6-O-
MMTr -2-d eoxy-3-O-TBDMS-D-m a n n itol (19). The reaction
was performed using a reaction condition described for 13
using 18 (340 mg, 0.70 mmol) and MMTrCl (432 mg, 1.40
mmol) in pyridine (14 mL) to give 19 (400 mg, 75%). This
product was crystallized from CH3OH/CH2Cl2. 1H NMR
(DMSO-d6) 12.05 (br s, 1H); 11.50 (br s, 1H); 8.18 (s, 1H); 7.55-
7.20 (m, 12H); 6.86 (m, 2H); 5.18 (d, J ) 7.0 Hz, 1H); 4.77 (m,
1H); 4.16-3.90 (m, 3H); 3.78 (s, 3H); 3.76-3.38 (m, 4H); 2.80
(m, 1H); 1.12 (d, J ) 2.1 Hz, 3H); 1.10 (d, J ) 2.3 Hz, 3H);
0.58 (s, 9H); 0.00 (s, 6H). 13C NMR (DMSO-d6) 180.1, 158.2,
155.0, 149.8, 147.6, 144.6, 139.0, 135.2, 130.2, 128.2, 127.9,
126.9, 119.0, 113.2, 85.8, 80.2, 73.3, 67.5, 67.1, 62.8, 55.1, 54.6,
34.7, 25.4, 19.3, 18.6, 17.6. HRMS calcd for C41H51N5O7SiNa
(M + H)+ 776.3455, found 776.3402.
1,5-An h yd r o-2-d eoxy-2-(gu a n in -9-yl)-D-m a n n itol (2b).
Compound 14 (305 mg, 0.50 mmol) was treated with 80% CF3-
COOH in H2O (20 mL) at room temperature for 60 h. The
solvent was removed and coevaporated with MeOH and
toluene. The residue was dissolved in MeOH (10 mL) and
treated with NH4OH (2 mL). The reaction mixture was
concentrated, and 2b (140 mg, 94%) was crystallized from
MeOH/CH2Cl2. 1H NMR (D2O, 33 °C at 500 MHz) 8.20 (s, 1H);
4.92 (m, 1H); 4.28 (dd, J ) 1.5, 13.2 Hz, 1H); 4.13 (dd, J )
5.3, 9.8 Hz, 1H); 4.11 (dd, J ) 2.4, 13.2 Hz, 1H); 4.00 (dd, J )
2.4, 12.7 Hz, 1H); 3.90 (dd, J ) 5.4, 12.7 Hz, 1H); 3.69 (t, J )
9.8 Hz, 1H); 3.54 (ddd, J ) 9.8, 2.4, 5.4 Hz, 1H). 13C NMR
(DMSO-d6) 157.0, 153.2, 152.4, 137.2, 115.6, 81.6, 71.8, 67.9,
66.8, 60.2, 53.8. HRMS calcd for C11H16N5O5 (M + H)+
298.1151, found 298.1168. Anal. Calcd for C11H15N5O5‚2.5H2O
C, 38.60, H, 5.89, N, 20.46. Found C, 38.80, H, 5.82, N, 20.36.
1,5-An h yd r o-3-(a d en in -9-yl)-4,6-O-ben zylid en e-2,3-d i-
d eoxy-D-a ltr itol (21). A mixture of 10 (350 mg, 0.95 mmol),
imidazole (25 mg), and NaH (80%) (120 mg) in THF (10 mL)
was stirred at 0 °C for 30 min and then at room temperature
for 30 min. After addition of CS2 (570 µL) and stirring at room
temperature for 60 min, MeI (234 µL) was added and kept at
room temperature for 3 h. After addition of H2O (2 mL), the
mixture was concentrated, diluted with CH2Cl2 (50 mL),
washed with saturated aqueous NaHCO3 (3 × 10 mL), H2O
(10 mL), dried, filtered, and concentrated to give 20 which was
directly treated with n-Bu3SnH (505 µL, 1.88 mmol) and AIBN
(38 mg, 0.23 mmol) in toluene (10 mL) at 110 °C overnight.
The solvent was removed, and the residue was purified by
silica gel column chromatography (0-4% MeOH in CH2Cl2)
1
to give 21 (150 mg, 45%). H NMR (CDCl3) 8.50 (s, 1H); 8.38
(s, 1H); 7.48-7.27 (m, 5H); 6.12 (br s, 2H); 5.67 (s, 1H); 5.19
(m, 1H); 4.40 (dd, J ) 4.6, 10.5 Hz, 1H); 4.22-3.68 (m, 5H);
2.93 (m, 1H); 2.39 (m, 1H). 13C NMR(CDCl3) 155.6, 152.6,
150.9, 140.9, 136.6, 129.2, 128.3, 125.9, 119.3, 102.5, 78.3, 69.7,
67.7, 63.3, 51.4, 29.7. HRMS calcd for C18H20N5O3 (M + H)+
354.1566, found 354.1577.
1,5-An h yd r o-2-d eoxy-2-(N2-isobu tyr ylgu a n in -9-yl)-3-O-
TBDMS-D-m a n n itol (18). To a cold (0 °C) suspension of 16
(660 mg, 1.32 mmol) in pyridine (20 mL) was slowly added
isobutyryl chloride (414 µL, 3.96 mmol). The ice bath was
removed and the reaction mixture kept at room temperature
for 5 h. The reaction mixture was quenched with ice-water
and extracted with EtOAc (3 × 25 mL). The combined organic
layer was concentrated and coevaporated with toluene to
remove the traces of pyridine. The residue was dissolved in
EtOAc (100 mL) and washed with saturated aqueous NaHCO3
(4 × 10 mL) and H2O (10 mL). The organic layer was
concentrated to give 17 which was directly treated with CF3-
COOH (10 mL) in CH2Cl2 (10 mL) at room temperature for 22
h. The solvent was removed, coevaporated with MeOH/
toluene. The residue was dissolved in MeOH (30 mL) and
treated with NH4OH (4 mL). The solvent was removed and
product was crystallized from CH2Cl2 in MeOH to give 18 (510
mg, 80%). 1H NMR (DMSO-d6) 12.0 (s, 1H); 11.50 (s, 1H); 8.18
(s, 1H); 5.23 (d, J ) 6.3 Hz, 1H); 4.88 (br s, 1H); 4.73 (br d, J
) 5.4 Hz, 1H); 3.98 (m, 3H); 3.77-3.51 (m, 3H); 3.25 (m, 1H);
2.80 (m, 1H); 1.15 (d, J ) 3.9 Hz, 3H); 1.07 (d, J ) 3.9 Hz,
3H); 0.62 (s, 9H); 0.05 (s, 3H); 0.01 (s, 3H). 13C NMR (DMSO-
d6) 180.1, 155.0, 149.9, 147.4, 139.5, 119.0; 81.7, 73.2, 67.1,
66.8, 60.0, 54.7, 34.7, 25.4, 19.3, 18.5, 17.6. HRMS calcd for
1,5-An h yd r o-3-(a d en in -9-yl)-2,3-d id eoxy-D-a ltr itol (22).
Compound 21 (80 mg, 0.22 mmol) was treated with 80% CF3-
COOH in H2O (5 mL) at room-temperature overnight. The
solvent was removed and coevaporated with MeOH/toluene.
The residue was dissolved in MeOH (10 mL), treated with NH4-
OH (2 mL), concentrated, and purified by silica gel column
chromatography (0-12% MeOH in CH2Cl2) to give 22 (50 mg,
83%). 1H NMR (DMSO-d6) 8.60 (s, 1H), 8.52 (s, 1H); 7.60 (br
s, 2H); 5.70 (br s, 1H); 5.29 (m, 2H); 4.40-4.08 (m, 6H); 2.82
(m, 1H); 2.19 (m, 1H). 13C NMR (DMSO-d6) 156.0, 152.3,
149.6, 140.5, 118.5, 79.7, 65.5, 61.2, 60.0, 51.3, 26.7. HRMS
calcd for C11H16N5O3 (M + H)+ 266.1253, found 266.1267.
1,5-An h ydr o-3-(aden in -9-yl)-3-deoxy-D-altr itol (23). Com-
pound 10 (150 mg, 0.40 mmol) was treated with 80% CF3-
COOH in H2O (5 mL) at room-temperature overnight. The
solvent was removed and coevaporated with MeOH/toluene.
The residue was dissolved in MeOH (5 mL) and treated with
NH4OH (2 mL), concentrated, and precipitated from MeOH/
CH2Cl2 to give 23 (70 mg, 61%). 1H NMR (DMSO-d6) 8.15 (s,
1H); 8.14 (s, 1H); 7.25 (br s, 2H); 5.47 (br s, 1H); 5.17 (br d, J
) 5.9 Hz, 1H); 4.92 (br s, 1H); 4.64 (dd, J ) 2.9, 10.2 Hz, 1H);
4.34 (br s, 1H); 3.94-3.33 (m, 6H). 13C NMR (DMSO-d6) 155.8,
152.0, 150.0, 140.6, 118.4, 79.9, 67.0, 66.6, 63.6, 59.3, 57.8.
C
21H36N5O6Si (M + H)+ 482.2434, found 482.2432. Anal.
Calcd for C21H35N5O6Si‚0.3CH2Cl2 C, 50.45, H, 7.08, N, 13.81.
Found C, 50.28, H, 6.98, N, 13.55.

1582 J . Org. Chem., Vol. 63, No. 5, 1998
Hossain et al.
HRMS calcd for C11H16N5O4 (M + H)+ 282.1202, found
282.1225. Anal. Calcd for C11H15N5O4‚0.9CH2Cl2 C, 39.96, H,
4.73, N, 19.58. Found C, 39.90, H, 4.87 and N, 19.54.
1,5-An h ydr o-2-(aden in -9-yl)-4,6-O-ben zyliden e-2-deoxy-
D-a ltr itol (25). Compound 8 was prepared as described for
5, using 7 (184 mg, 0.5 mmol), Et3N (125 µL, (0.9 mmol), and
MsCl (54 µL, (0.7 mmol) in CH2Cl2 (5 mL) to give 8 (223 mg,
100%) which was directly used for the next step. A mixture
of adenine (135 mg, 1.0 mmol) and NaH (80%, 60 mg, 2.0
mmol) in DMF (5 mL) was heated at 95 °C for 60 min.
Compound 8 (223 mg, 0.5 mmol) in DMF (2 mL) was added
and the reaction mixture kept at 95 °C for 20 h. The solvent
was removed, and EtOAc (50 mL) was added followed by
saturated aqueous NaHCO3 (15 mL). The two layers were
separated. The aqueous layer was back extracted with EtOAc
(2 × 20 mL). The combined organic layer was washed with
H2O (10 mL), dried, filtered, concentrated, and purified by
silica gel column chromatography (0-7% MeOH/CH2Cl2) to
give 25 (100 mg, 51%). The spectroscopic properties of 25 are
identical as reported previously.12
P h osp h or a m id ite Syn th esis. The phosphitylation reac-
tion was carried out on 0.45 mmol of 13 or 19, respectively, in
8 mL of dichloromethane using 3 equiv of N,N-diisopropyl-
ethylamine and 1.5 equiv of N,N-diisopropylchlorophosphora-
midite. Reactions were complicated as only a slight change
in Rf value was seen on TLC analysis. Therefore, the reaction
was stirred for 2 h at room temperature and quenched by
addition of 2 mL of ethanol and stirring for 10 min more. The
mixture was partitioned between 30 mL of dichloromethane
and 30 mL of saturated aqueous NaHCO3. The organic phase
was washed with brine (2 × 25 mL), dried on Na2SO4, and
evaporated. The obtained residue was flash purified on 30 g
of silica gel (hexane/acetone/triethylamine 50/48/2) and pre-
cipitated in cold hexane.
procedure (RESP21) of the charges to the 6-31G*-derived
electrostatic potential from a GAMESS calculation.22 The
modeling of the MNA:RNA complex in A-RNA geometry was
done by modeling d(GCGTAGCG):r(CGCUACGC) in Arnott’s
canonical A-RNA geometry and subsequent docking of the
p(GCGTAGCG) strand onto the DNA strand. The MNA strand
was then minimized by restraining the base atoms onto the
corresponding positions of the DNA strand by a force constant
of 10 kcal/mol Å2 per atom. After deletion of the DNA strand,
the MNA:RNA complex was energy minimized to a rms
gradient of 0.1 kcal/mol Å using restraints on the base
hydrogen bonds (10 kcal/mol Å2). The resulting duplex was
put in a rectangular box of TIP3P water extending ap-
proximately 12 Å in each direction from the nucleic acid atoms.
The 14 water molecules with the highest negative electrostatic
potential were replaced by sodium ions which results in 3529
remaining waters and an initial box size of 50.9 × 43.7 × 45.9
Å. The single strand MNA was cut out of the duplex MNA:
RNA, soaked in a water box and the 7 waters with highest
electrostatic potential replaced by sodium ions, resulting in a
rectangular box of 50.0 × 40.6 × 36.3 Å with 2347 water
molecules. The subsequent molecular dynamics simulations
were applied using constant temperature (300 K) with a
separate but equal temperature coupling time constant of 0.4
ps-1 for solute and solvent atoms,23 and constant pressure (1
bar) conditions with a compressibility of 44.6 × 10-6 bar-1 and
a pressure relaxation time of 0.4 ps-1. The integration time
step was 2 fs and SHAKE24 was applied on all bond lengths.
A 9 Å cutoff for the Lennard-J ones interactions was used
while updating the nonbonded pairlist every 10 steps. Long-
range electrostatic interactions were treated by the particle
mesh Ewald (PME) method25 with a PME charge grid spacing
of approximately 1 Å in each direction, an order of 4 of the
âspline interpolarion and a direct sum tolerance of 10-4
.
26: yield 86%, LSIMS (neg. mode, NBA) [M - H]- 970.
27: yield 91%, LSIMS (neg. mode, NBA) [M - H]- 952.
Oligon u cleotid e Syn th esis a n d P u r ifica tion . Standard
operations for oligonucleotide assembly were as described.15
Following assembly (trityl off) the support was treated with
NH4OH concentrated/EtOH 3:1 for 16 h at 50 °C, and the
supernatants was lyophylized. The residue was taken up in
1 mL of TBAF/THF 1 M and left in the dark for 22 h at rt.
After dilution with 1 mL of 0.03 M KH2PO4, the mixture was
gel filtered in 2 portions over a NAP-25 column. Following
concentration a second gel filtration was done eluting with
buffer A (0.03 M KH2PO4 pH 6.7, 20% CH3CN), and the eluate
was purified by ion exchange chromatography on a Mono Q
HR 10/10 column with a KH2PO4 gradient. Product containing
fractions were desalted on Sephadex G50 and analyzed on a
polystyrene reverse phase column (PLRP-S, Polymer Labora-
tories) with a CH3CN gradient in aqueous 0.1 M TEAA.
For the dodecamer, the residue after ammonia deprotection
was divided in two, one part treated as above, while the other
part was dissolved in 1 mL of TEA, 3HF/DMF 3:1 and heated
for 1 h 30 at 55 °C. After addition of 1 mL aqueous 0.1 M
TEAA the mxiture was worked-up as above.
While analysis of the first part showed a broad peak on ion
exchange chromatography at the expected position, the depro-
tection mixture by TEA, 3HF only displayed peaks with short
retention times.
Molecu la r Mod elin g. The force field parameters and
partial charges for RNA were taken from the AMBER 4.1 force
field parameter database.20 The force field parameters for
MNA were taken from Amber 4.1 without modification. The
partial charges of MNA were obtained by a 2 stage fitting
Equilibrium was achieved by energy minimization of the whole
system until the gradient of the energy dropped below 0.1 kcal/
mol Å. Then the systems were heated to 300 K over a period
of 10 ps after which the production runs over 1 ns for the
duplex MNA:RNA and 300 ps for the single strand MNA were
initiated, saving the conformations every 0.2 ps. The energy
minimizations and the MD simulations were performed with
the SANDER module and the results analyzed with the
CARNAL module of AMBER 4.1.20
Ack n ow led gm en t. We thank Dr. Herman van Hal-
beek for high resolution NMR for 2a and 2b and G.
Schepers for assistance with oligonucleotide synthesis.
We also thank M. Vandekinderen for excellent editorial
help and Prof. W. Pfleiderer for elemental analyses. We
are grateful to the K.U. Leuven for financial support
(GOA97/11).
J O9718511
(21) Bayly, C. I.; Cieplak, P.; Cornell, W. D.; Kollman, P. A. J . Phys.
Chem. 1993, 97, 10269-10280.
(22) Schmidt, M. W.; Baldridge, K. K.; Boatz, J . A.; Elbert, S. T.;
Gordon, G. S.; J ensen, J . H.; Koseki, S.; Matsunaga, N.; Nguyen, K.
A.; Su, S.; Windus, T. L.; Dupuis, M.; Montgomery, J . A. J . Comput.
Chem. 1993, 14, 1347-1363.
(23) Berendsen, H. J . C.; Postma, J . P. M.; van Gunsteren, W. F.;
DiNola, A.; Haak, J . R. J . Chem. Phys. 1984, 8, 3684-3690.
(24) van Gunsteren, W. F.; Berendsen, H. J . C. Mol. Phys. 1977, 5,
1311-1327.
(25) Essmann, U.; Perera, L.; Berkowitz, M. L.; Darden, T.; Lee, H.;
Pedersen, L. G. J . Chem. Phys. 1995, 103, 8577-8593.
(26) Kraulis, P. J . J . Appl. Crystallogr. 1991, 24, 946-950.
(27) Esnouf, R. M. J . Mol. Graphics 1997, 15, 132-134.
(28) Sayle, R.; Milner-White, E. J . Trends Biochem. Sci. 1995, 20,
374.
(20) Cornell, C. D.; Cieplak, P.; Bayly, C. I.; Gould, I. R.; Merz, K.
M., J r.; Ferguson, D. M.; Spellmeyer, D. C.; Fox, T.; Caldwell, J . W.;
Kollman, P. A. J . Am. Chem. Soc. 1995, 117, 5179-5197.