Nucleosides. Part LXIV. Base-Labile Protecting Groups for the Oligoribonucleotide Synthesis

Article abstract of DOI:10.1002/1522-2675(20010613)84:6<1504::AID-HLCA1504>3.0.CO;2-G

New labile protecting groups for the anticipated synthesis of oligoribonucleotides were developed and introduced via their carbonochloridates 8-11 at the 5'-O position of thymidine (15) to form 16, 18, 21, and 24 in good yields (Schemes 2 and 3). Similarly, the 5'-O-diphenylphosphinoyl(dpp)-protected thymidine derivative 27 was synthesized with diphenylphosphinoyl chloride 14 as the reactive reagent. With the help of the model compounds 16, 18, 21, 24, and 27, the deprotection rates of these functions towards base treatment were recorded to evaluate their usefulness as temporary protecting groups in RNA assembly (Table). Finally, the relative stability of the 2-(4-nitrophenyl)ethoxycarbonyl (npeoc) protecting group towards bases confirmed its use as a permanent blocking group in our npe/npeoc strategy.

Full text of DOI:10.1002/1522-2675(20010613)84:6<1504::AID-HLCA1504>3.0.CO;2-G

1504
Helvetica Chimica Acta ± Vol. 84 (2001)
Nucleosides
Part LXIV1)
Base-Labile Protecting Groups for the Oligoribonucleotide Synthesis
by Ursula Münch and Wolfgang Pfleiderer*
Fakultät für Chemie, Universität Konstanz, Postfach 5560, D-78434 Konstanz
Dedicated to Prof. Dr. Edgar Heilbronner on the occasion of his 80th birthday with best personal wishes
New labile protecting groups for the anticipated synthesis of oligoribonucleotides were developed and
introduced via their carbonochloridates 8 ± 11 at the 5'-O position of thymidine (15) to form 16, 18, 21, and 24 in
good yields (Schemes 2 and 3). Similarly, the 5'-O-diphenylphosphinoyl(dpp)-protected thymidine derivative 27
was synthesized with diphenylphosphinoyl chloride 14 as the reactive reagent. With the help of the model
compounds 16, 18, 21, 24, and 27, the deprotection rates of these functions towards base treatment were recorded
to evaluate their usefulness as temporary protecting groups in RNA assembly (Table). Finally, the relative
stability of the 2-(4-nitrophenyl)ethoxycarbonyl (npeoc) protecting group towards bases confirmed its use as a
permanent blocking group in our npe/npeoc strategy.
1. Introduction. ± Discoveries in the past several years considering RNA, e.g. its role
as ribozymes [2][3], its application in antisense technology [4][5], or its high binding
affinity to specific targets [6 ± 8] increased the demand for synthetic oligoribonucleo-
tides.
The solid-phase synthesis of oligonucleotides has been improved immensely in
recent years with the development of phosphoramidite chemistry [9 ± 13]. Compared to
DNA, however, the progress in the chemical build-up of RNA has been less fast and
remains still less successful since the additional 2'-OH function extends the complexity
of an appropriate protecting group strategy. The 2'-O-protecting group must remain
intact during all the various steps of automated RNA assembly, but should be removed
at the end of synthesis under mild conditions to avoid isomerization or cleavage of the
internucleotide linkages. Further problems are a reduction in the stepwise condensa-
tion yields of the phosphoramidites due to the steric constraints imposed by the
additional 2'-O-protecting group and the liability of free RNA to enzymatic
degradation.
The 2'-OH function of commercially available phosphoramidites is mostly
protected with the (tert-butyl)dimethylsilyl (tbdms) group, which was introduced by
Ogilvie et al. [14][15]. Since this group is removed by F ions, it is completely
compatible with the dimethoxytrityl group, the most common protecting group for the
5'-OH function. However, under the ammoniacal conditions used for the cleavage of
the acyl blocking groups of the nucleobases, there is a danger of partial loss of the (tert-
1
)
Part LXIII: [1].

Helvetica Chimica Acta ± Vol. 84 (2001)
1505
butyl)dimethylsilyl group, which can lead to cleavage of the nucleotide chain [16].
Additional problems are, e.g., the build-up of salt by-products or the steric hindrance of
the bulky (tert-butyl)dimethylsilyl group.
Reese and coworkers introduced acetal protecting groups like the tetrahydro-2H-
pyranyl [17] or the achiral tetrahydro-4-methoxy-2H-pyran-4-yl group [18] for the
blocking of the 2'-OH function, which are removed under mild acidic conditions. But
especially for the synthesis of longer oligoribonucleotides, the acid-labile acetal
function is not sufficiently compatible with the traditional 5'-O-trityl groups. As an
improvement to this situation, Reese and coworkers, furthermore, applied the more
stable 1-(2-chloro-4-methylphenyl)-4-methoxypiperidin-4-yl (ctmp) [19][20] and 1-(2-
fluorophenyl)-4-methoxypiperidin-4-yl (fpmp) groups [21][22] in combination with
the more acid-sensitive 5'-O-pixyl function, and just recently the introduction of the 2'-
O-bis(2-acetoxyethoxy)methyl (Ace) orthoester function in combination with a 5'-O-
alkoxy-bis(trimethylsilyloxy)silyl group [23] as well as the 2'-O-[(triisopropylsilyl)oxy]-
methyl (Tom) group [24][25] have also been recommended . Another alternative
strategy replaces the acid-labile 5'-O-protecting group by base-labile functions to
achieve complete compatibility with the 2'-O-acetal group. For example, Gait and co-
workers employed the tetrahydro-4-methoxy-2H-pyran-4-yl (thmp) group together
with the b-eliminating (9H-fluoren-9-ylmethoxy)carbonyl (fmoc) function [26], which
is, however, not compatible with the 2-cyanoethyl phosphate protecting group, while
Bergmann and Pfleiderer combined the same acetal group with the 5'-O-[2-(dansyl)-
ethoxy]carbonyl (danseoc) group [27][28] and the more stable 2-(4-nitrophenyl)ethyl
group at the internucleotidic linkage.
In this paper, we report the search of new base-labile protecting groups for the
temporary 5'-O-protection in oligoribonucleotide synthesis. Their introduction into the
5'-OH function of thymidine led to model substances, which were evaluated regarding
their usefulness as new potential protecting groups for RNA assembly.
2. Syntheses. ± The syntheses of the new reactive protecting groups were performed
from the respective hydroxy compounds, i.e., from 1-phenyl-2-cyanoethanol (ˆ b-
hydroxybenzenepropanenitrile; 1pce; 1), 2-phenyl-2-cyanoethanol (ˆa-(hydroxyme-
thyl)benzeneacetonitrile; 2pce; 2), and 2,2-bis(4-nitrophenyl)ethanol ((np)2e; 3)
according to known literature procedures [29 ± 31]. The synthesis of 1,2-bis(4-nitro-
phenyl)ethanol (ˆ1,3-bis(4-nitrophenyl)propan-2-ol; 7) began with the nitration of 1-
benzyl-2-phenylethyl acetate (4) [32] at 0 ± 108 leading to a mixture of nitrated
products, whereby the anticipated compound 5 could only partially be crystallized from
the reaction mixture by Et2O (Scheme 1). Subsequent acid-catalyzed methanolysis of
both the nitrated acetate 5 ( ! 7; 69%) and the remaining oily product mixture 6 gave
alcohol 7 in a total yield of 35% over the two steps.
The following reaction of the alcohols 1 ± 3 and 7, respectively, with an excess of
diphosgene (1.5 ± 2.5 equiv.) in presence of Et3N under ice-bath cooling gave the
corresponding carbonochloridates 8 ± 11 in good yields (73 ± 100%). All efforts to get
pure compounds 8 and 9 failed since the carbonochloridate 8 was contaminated with b-
chlorobenzenepropanenitrile (12; ca. 10%) [33] and the starting alcohol 1 (ca. 10%),
and product 9 contained 20% of a-methylidenebenzeneacetonitrile (13) [34 ± 36],
formed from 9 by a b-elimination process. Fortunately, these by-products, identified by

1506
Helvetica Chimica Acta ± Vol. 84 (2001)
Scheme 1
O2N
O
O
MeOH/HCl
69%
O2N
5
HNO3/H2SO4
O2N
O
O
+
OH
MeOH/HCl
mixture
O2N
4
6
7
35%
over two steps
their 1H-NMR data, did not interfere in the following reaction step and thus, no effort
was spent on the purification of 8 and 9, respectively. Since we could not find 1H-NMR
data for b-chlorobenzenepropanenitrile (12) in the literature, this compound was
synthesized from hydroxy derivative 1 by reaction with SOCl2 in a yield of 37% ([33]:
47%) to insure positive identification of compound 12.
Scheme 2
R3
R3
O
Ph
CCl3COCl
R1
R1
NC
OH
O
Cl
Cl
R2
R2
H
12
1-3, 7
8-11
R1
R2
R3
%
CN
Ph
H
H
1
CN
CN
H
Ph
H
8
9
78
73
2
3
7
Ph
13
4-NO2-C6H4 4-NO2-C6H4
4-NO2-C6H4
H
10 quant. ([30]: quant.)
H
4-NO2-C6H4CH2 11 95
The introduction of the protecting groups at the 5'-O-position of thymidine (15) was
carried out with a slight excess (1.3 equiv.) of carbonochloridate 8 ± 11 and the
commercially available diphenylphosphinoyl chloride (14), respectively, at low
temperature ( 558 (iPrOH/N2 bath)) in CH2Cl2/pyridine, to give the desired 5'-O-
substituted thymidines 16, 18, 21, 24, and 27 in yields between 53 and 82%. The 3',5'-di-
O-substituted nucleosides 17, 20, 23, 26, and 28 were formed as by-products in 6 ± 16%
yield, and the 3'-O-isomers could occassionally also be isolated in small amounts (19
(5%), 22 (4%), and 25 (3%)). In the case of the (1-phenyl-2-cyanoethoxy)carbonyl
(1pceoc) group, it became necessary to neutralize the reaction mixture with acid before
workup, since the pyridine concentrated during the evaporation of the solvent caused
already cleavage of the 5'-O-protecting group.

Helvetica Chimica Acta ± Vol. 84 (2001)
1507
Scheme 3
RO
Thy
HO
Thy
HO
Thy
O
O
O
8-11
+
O
Ph
HO
16, 18, 21, 24, 27
HO
RO
P
or
15
Cl
19, 22, 25
Ph
O
14
RO
Thy
O
HN
Thy =
+
O
N
RO
17, 20, 23, 26, 28
R
5'-O-subst. 3'-O-subst. 3',5'-di-O-subst.
R3
O
R1
R1
O
R2
R2
R3
CN
CN
H
Ph
H
16 (71%)
18 (68%)
21 (74%)
24 (82%)
17 (14%)
20 (12%)
23 (7%)
26 (6%)
Ph
19 (5%)
22 (4%)
25 (3%)
4-NO2-C6H4 4-NO2-C6H4
H
4-NO2-C6H4
H
4-NO2-C6H4CH2
O
27 (53%)
28 (16%)
Ph
P
Ph
3. Kinetic Studies. ± All new protecting groups were tested regarding their cleavage
ability towards different bases, mainly 1,8-diazabicyclo[5.4.0]undec-7-ene (DBU) as
well as piperidine and Et3N (Table). The [2,2-bis(4-nitrophenyl)ethoxy]carbonyl
((np)2eoc) function had been developed by Ramage et al. [30] as a protecting group in
peptide synthesis. Comparatively, we also evaluated the liability of the [2-(4-nitro-
phenyl)ethoxy]carbonyl (npeoc) group in nucleoside 29 [37], because we use this
group together with the 2-(4-nitrophenyl)ethyl (npe) protecting group in our npe/
npeoc strategy for the permanent blocking of base and phosphate functions [37 ± 40].
The advantage of this strategy lies in the opportunity to synthesize crude oligomers in a
very pure form, since the permanent protecting groups are removed by DBU in an
aprotic solvent via b-elimination and are washed out while the oligonucleotide is still
bound to the solid support.
For the kinetic studies, the appropriate nucleoside was treated with a 10-fold excess
of base, usually in MeCN. The only exception was the deprotection reaction of the 5'-O-
(diphenylphosphinoyl)thymidine (27), which had to be run in a protic solvent (e.g.,
MeOH). In contrast to all other protecting groups, which were removed in a

1508
Helvetica Chimica Acta ± Vol. 84 (2001)
Table. Half-Lives of Protected Thymidine Nucleosides in Presence of Base
HO
Thy
RO
Thy
O
O
Base
HO
HO
15
16, 18, 21, 24, 27, 29
R
Abbreviation Base Concentration [M]
Half-life [t1/2
]
R3
O
R1
R1
O
R2
R2
R3
Ph
16 CN
H
1pceoc
DBU
12 sa)
0.1
0.5
0.5
0.01
0.5
0.1
0.5
0.5
0.1
1
piperidine
Et3N
47 mina)
15.2 ha)
< 15 sc)
< 12 sc)
12 sa)
9.4 mina)
9.8 ha)
38.6 minb)
5.3 minb)
18 CN
Ph
H
H
2pceoc
DBU
Et3N
21 4-NO2-C6H4 4-NO2-C6H4
(np)2eoc
DBU
piperidine
Et3N
24 4-NO2-C6H4
H
4-NO2-C6H4CH2 nbnpeoc
DBU
O
dpp
1
7.3 hb)
27
29
DBU
DBU
Ph
P
Ph
12.2 ha)
94 mina)
8.7 mina)
npeoc
0.01
0.1
1
O
4-NO2-C6H4
O
a) Calculation of t1/2 according to Method A. b) Calculation of t1/2 according to Method B. c) Complete
cleavage within t.
b-elimination process by releasing an alkene derivative (see Scheme 4), the diphenyl-
phosphinoyl function was cleaved by transesterification in an addition-elimination
mechanism initiated by a nucleophilic attack of the MeO ion at the P-atom.
Aliquots were quenched in AcOH/MeCN at defined time intervals, and the
mixtures analyzed by HPLC. From the obtained data, we determined the correspond-
ing half-lives (t1/2) for the pseudo-first-order reactions. For these calculations, we used
two different methods.

Helvetica Chimica Acta ± Vol. 84 (2001)
1509
Scheme 4
Method A considers the relative amount of protected nucleoside and cleaved
protecting group, whereby the difference in extinction coefficient of starting material
and resulting alkene derivative 32 ± 34 has to be taken into account. For this reason, and
to identify the deprotection products clearly by their retention time, compounds 32 ± 34
were synthesized from their corresponding carbonochloridates 8 and 31 [37] or acetate
30 [30], respectively, by treatment with DBU according to Ramage et al. [30]
(Scheme 5). Thereby, 3-phenylprop-2-enenitrile (32) was obtained as a mixture of its
(Z)-isomer 32a and (E)-isomer 32b in a ratio of 3 :7 (by 1H-NMR). The calculation of
the half-lives were carried out with the extinction coefficients of this mixture, although
the e-values differ considerably for the (Z)-isomer 32a (log e273 ˆ 4.22) and the (E)-
isomer 32b (log e272 ˆ 4.36) [41]. But since the calculated kinetic data shall only serve
for evaluation, this simplification should be acceptable. The reaction of the 5'-O-{[2,2-
bis(4-nitrophenyl)ethoxy]carbonyl}thymidine (21) with piperidine led to the alkene 33
and a second cleavage product, the ene-piperidine adduct 35 [30]. For an unequivocal
identification of compound 35, reference material was synthesized according to
Ramage et al. [30] by reaction of 1,1-bis(4-nitrophenyl)ethene (33) with piperidine to
reveal the same retention time as the product obtained during the kinetic evaluation.
Scheme 5
R3
O
R1
R2
R3
H
CN
H
Ph
H
DBU
R1
O
R4
R2
32a
32b
8, 30, 31
32-34
R1
CN
R2
H
R3 R4
Ph Cl
CN
H
H
%
Ph
8
32 76 (32a/32b 3:7)
33 70 ([30]: 69)
34 54
30 4-NO2-C6H4 4-NO2-C6H4
H
H
CH3
Cl
31 4-NO2-C6H4
H
4-NO2-C6H4
HN
H
4-NO2-C6H4
4-NO2-C6H4
N
4-NO2-C6H4
H
88%
35
33

1510
Helvetica Chimica Acta ± Vol. 84 (2001)
In those cases, where Method A could not be employed for calculation, Method B
was used, whereby the ratio of the amount of starting nucleoside at a time t and the
amount of starting nucleoside at the 0 min time point t0 was determined.
According to the results listed in the Table, both the 1pceoc and the (np)2eoc groups
are suitable for the temporary protection of the 5'-OH function for the anticipated
oligoribonucleotide synthesis. The evaluated half-lives for the deprotection of these
functions with DBU (0.1m) are, accordingly, small and considerably shorter than for the
npeoc group, which we wish to use for the permanent blocking during RNA assembly.
For cleavages with 0.1m DBU, the half-lives of both functions were calculated
identically at 12 sec, while the deprotection with piperidine and Et3N is faster for the
(np)2eoc than for the 1pceoc group. However, with half-lives of ca. 9 min (0.5m
piperidine) and 10 h (0.5m Et3N), these bases are not useful for the cleavage of the
(np)2eoc function as a 5'-O-protecting group.
The b-proton of the 1pceoc group is mainly activated only by the cyano function,
however, with both the cyano and the phenyl function at C(b), the 2pceoc group
becomes extremely base-labile. Thus, even with very dilute DBU (0.01m) or with the
weaker Et3N (0.5m), the deprotection reaction is too fast to record any kinetics.
Because of the fact, that some reactions during the synthesis of the monomeric building
blocks as well as during the RNA assembly are catalyzed by base, the extreme base
liability of the 2pceoc group might cause problems and can not be recommended.
In the case of the nbnpeoc group, the proton at C(b) is activated by only one
nitrophenyl function. In this way, the deprotection rate is immensely decreased
compared to the (np)2eoc function, where two nitrophenyl groups contribute to the
reactivity of H C(b). With half-lives, recorded in the order of minutes and thus
approximately half of the value as for the npeoc group, the nbnpeoc function is not
useful for 5'-O protection.
This unsuitability also applies to the diphenylphosphinoyl (dpp) function, the half-
life of which, in the presence of 1m DBU in MeOH, was evaluated to ca. 7 h. However, the
dpp group was also tried for the protection of the 2'-OH function in combination with
the dimethoxytrityl group as the temporary 5'-O-protecting group. As expected, this 2'-
O-protecting group has the tendency for intramolecular 2' ! 3'-O-migration, as found
with acyl functions [42 ± 44] which excludes their application as 2'-O-protecting groups.
Nevertheless, to examine the migration tendency of the dpp group, it was
introduced at the 2'-O-position of the 1,1,3,3-tetraisopropyl-1,3-disiloxane-1,3-diyl-
(tipds)-protected uridine (36) [45] by reaction with diphenylphosphinoyl chloride (14)
to give 37 (Scheme 6). The subsequent cleavage of the tipds function from compound
37 was carried out with Bu4NF in THF. But even addition of AcOH in order to decrease
the basicity of the F-ion could not avoid 2' ! 3'-O-migration of the dpp function. The
reaction led to a mixture of the 2'-O-dpp- and the 3'-O-dpp-isomer 38 and 39,
1
respectively, in a ratio of 1:2 according to H-NMR data. Since both isomers showed
identical chromatographic properties, the mixture was subsequently tritylated with
2.25 equiv. of monomethoxytrityl chloride, thus achieving better resolution of the
product mixture. Indeed, the resulting products were easily separated by flash
chromatography, giving the 2'-O-dpp-substituted product 40 in 31% yield and the 3'-O-
dpp-substituted compound 41 in 40% yield, respectively. These findings exclude the
dpp group as a sugar protecting group in RNA assembly.

Helvetica Chimica Acta ± Vol. 84 (2001)
1511
Scheme 6
O
Ph
P
O
Ura
Cl
O
Ura
Ph
O
O
Bu4NF · 3 H2O
Si
Si
O
14
AcOH
65%
80%
O
O
Odpp
O
OH
Si
Si
36
37
O
HO
Ura
MeOTrO
Ura
O
O
HN
Ura =
MeOTrCl
O
N
R1O
OR2
R1O
OR2
R2
R2
ratio
%
R1
H
R1
H
40
dpp
H
38
dpp
H
31
40
1
2
41 dpp
39 dpp
MeOTr = monomethoxytrityl; dpp= diphenylphosphinoyl
From our results, we can conclude that the 1pceoc and (np)2eoc groups can be
considered as potential alternatives for 5'-OH protection in conjunction with 2'-O-
acetal and npe/npeoc groups in oligoribonucleotide synthesis. The use of the 1pceoc
group, however, is advantageous since its carbonochloridate 8 is prepared in only two
steps with very good yields. Further studies showing the application of the 1pceoc group
in RNA synthesis will be presented in due time.
Experimental Part
General. Products were dried under high vacuum. TLC: precoated silica gel thin-layer sheets 60 F254 from
Merck. Flash chromatography (FC): silica gel (Baker, 30 ± 60 mm); 0.2 ± 0.3 bar. HPLC: pump L 6000,
autosampler AS 4000, Merck-Hitachi, UV-detector Uvikon 730 SLC, Kontron; column RP 18, LiChrospher
125 Â 4 mm, 5 mm, Merck; elution: A; 0.1m (Et3NH)OAc buffer (pH7); B; MeCN; gradient: 0% B (0 ± 3 min),
0 ± 50% B (3 ± 10 min), 50 ± 100% B (10 ± 25 min), 100 ± 50% B (25 ± 30 min), 50 ± 0% B (30 ± 33 min), 0% B
(33 ± 35 min); flow rate, 1 ml/min. M. p.: Gallenkamp melting-point apparatus; no corrections. UV/VIS: Perkin-
Elmer Lambda 5; lmax in nm (log e). 1H-NMR: Bruker AC 250; d in ppm rel. to SiMe4 or CDCl3 ((D6)DMSO) as
internal standard.
1. b-Hydroxybenzenepropanenitrile (1) [29]. A soln. of 1.6m BuLi in hexane (34.5 ml, 55 mmol) in anh.
THF (34.5 ml) was cooled under N2 to 808 (iPrOH/N2 bath). A soln. of MeCN (2.65 ml, 50 mmol) in dry THF
(50 ml) was added dropwise within 8 min and the resulting suspension stirred for 2 h at 808. After dropwise
addition of benzaldehyde (5.31 g, 50 mmol) in anh. THF (50 ml) within 6 min, the cooling bath was removed
and the resulting clear soln. stirred for additional 25 min and then poured onto ice/H2O/conc. HCl soln. (100 g,
60 ml, and 17 ml, resp.). The aq. layer was extracted with Et2O (3 Â 100 ml), the combined org. phase dried
(Na2SO4) and evaporated, and the yellowish residue purified (removal of benzaldehyde) by FC (silica gel
(150 g), 14 Â 5.5 cm, petroleum ether/AcOEt 7:1 (640 ml), 6 :1 (210 ml), 5 :1 (300 ml), 4 :1 (250 ml), 3 :1
(200 ml), 2 :1 (300 ml), AcOEt (600 ml)): 7.06 g (96%) ([29]: 78% after distillation) of 1. Colorless oil. nD20
1.5394 ([46]: n2D0 1.5388). TLC (petroleum ether/AcOEt 2 :1): Rf 0.33. UV (MeOH): 256 (2.53), 212 (sh, 3.74).
1H-NMR (CDCl3): 7.55 ± 7.29 (m, Ph); 5.01 (t, CH); 2.75 (d, CH2); 2.54 (br. s, OH).

1512
Helvetica Chimica Acta ± Vol. 84 (2001)
2. 1-(4-Nitrobenzyl)-2-(4-nitrophenyl)ethyl Acetate (5). Nitrating acid was prepared by adding 98% sulfuric
acid (2 ml) to 60% nitric acid (2 ml) under ice-bath cooling. This mixture was added dropwise to the stirred and
ice-cooled oil of 1-benzyl-2-phenylethyl acetate [32] (4; 1.27 g, 5.0 mmol) within 15 min, while the temp. was
maintained between 0 and 108 ( ! slightly yellow mixture). After stirring for further 2 h at r.t., the mixture was
poured onto ice (10 g) and the product extracted with Et2O (25 ml, 2 Â 10 ml). The combined org. phase was
washed with sat. NaHCO3 soln. (2 Â 30 ml), H2O (3 Â 30 ml), and sat. NaCl soln. (30 ml), dried (Na2SO4), and
evaporated to a light yellow oil. After cooling in the refrigerator, the oil was crystallized from Et2O in a
sonicating bath, and the product was isolated by filtration to give 0.48 g (28%) of 5. Upon evaporation, the
filtrate afforded a yellow oily mixture 6 (1.19 g), containing 5 and other nitrated by-products.
Data of 5: Pale yellow solid. M. p. 139 ± 1408. TLC (petroleum ether/AcOEt 7 :3): Rf 0.39. UV (MeOH):
271 (4.33), 211 (sh, 4.20). 1H-NMR ((D6)DMSO): 8.15 (d, 4 H o to NO2); 7.50 (d, 4 H m to NO2); 5.38 ± 5.26 (m,
CH); 3.15 ± 2.93 (m, 2 CH2); 1.80 (s, Me). Anal. calc. for C17H16N2O6 (344.3): C 59.30, H 4.68, N 8.14; found: C
59.20, H 4.73, N 8.07.
3. 1-(4-Nitrobenzyl)-2-(4-nitrophenyl)ethanol (ˆ1,3-Bis(4-nitrophenyl)propan-3-ol; 7). a) From 5: By
heating up, a suspension of 5 (0.48 g, 1.39 mmol) in MeOH (5 ml) and conc. HCl soln. (0.1 ml) was dissolved,
and the resulting soln. was refluxed for 6 h. The yellow mixture was then poured onto ice (10 g) and the org.
phase extracted with AcOEt (3 Â 20 ml). The combined org. phase was washed with H2O (2 Â 40 ml) and sat.
NaCl soln. (40 ml), dried (Na2SO4), and evaporated. The solid residue was crystallized from MeOH (15 ml) to
give 292 mg (69%) of 7.
b) From Mixture 6: A yellow soln. of mixture 6 (1.14 g) in MeOH (5 ml) and conc. HCl soln. (0.1 ml) was
heated under reflux for 6 h. The mixture was then poured onto ice (10 g), the org. phase extracted with AcOEt
(30 ml, 2 Â 20 ml), the combined org. phase washed with H2O (2 Â 40 ml) and sat. NaCl soln. (40 ml), dried
(Na2SO4), and evaporated, and the brown oil purified by FC (silica gel (25 g), 10.5 Â 2.6 cm, toluene (200 ml),
toluene/AcOEt 9 :1 (200 ml), 8 :1 (90 ml), 7:1 (80 ml), 6 :1 (70 ml), 5 :1 (60 ml)): 152 mg of 7. Mixture
fractions containing 7 were evaporated separately and purified once more by FC (silica gel (15 g), 9.5 Â 2.3 cm,
toluene/AcOEt 10 :1 (220 ml), 9 :1 (200 ml)): 81 mg of 7. Overall yield from 1-benzyl-2-phenylethyl acetate (4)
35% of 7. Colorless needles. M. p. 143 ± 1458. TLC (petroleum ether/AcOEt 2 :1): Rf 0.38. UV (MeOH): 276
1
(4.33), 212 (sh, 4.18). H-NMR ((D6)DMSO): 8.15 (d, 4 H o to NO2); 7.50 (d, 4 H m to NO2); 4.92 (d, OH);
4.08 ± 3.92 (m, CH); 2.94 ± 2.71 (m, 2 CH2). Anal. calc. for C15H14N2O5 (302.3): C 59.60, H 4.67, N 9.28; found: C
59.76, H 4.73, N 9.29.
4. 2-Cyano-1-phenylethyl Carbonochloridate (8). Under N2, a soln. of 1 (1.51 g, 10 mmol) and Et3N
(1.41 ml, 1.03 g, 10 mmol) in dry THF (5 ml) was added dropwise to an ice-cooled, stirred soln. of
trichloromethyl carbonochloridate (4.92 g, 3.0 ml, 25 mmol) in dry THF (10 ml) within 10 min. After 10 min,
the ice-bath was removed, and stirring was continued at r.t. for 6 h 30 min. The mixture was filtered through a
sintered glass funnel, washed with dry THF, and evaporated: 1.91 g of a slightly yellow oil. 1H-NMR: ca. 82% of
8 (yield 78%), 10% of b-chlorobenzenepropanenitrile (12), and 8% of 1. 8: TLC (petroleum ether/AcOEt 2 :1):
1
Rf 0.82. H-NMR (CDCl3): 7.45 ± 7.37 (m, Ph); 5.93 (dd, CH); 3.08 ± 2.90 (m, CH2).
5. 2-Cyano-2-phenylethyl Carbonochloridate (9). As described in Exper. 4, with 2 (1.30 g, 8.85 mmol), Et3N
(1.23 ml, 90 mg, 8.85 mmol), THF (10 ml), trichloromethyl carbonochloridate (2.65 ml, 4.35 g, 22 mmol), and
THF (10 ml); addition within 15 min, stirring for 15 min at 08 and 4 h 45 min at r.t.: 1.56 g of a slightly yellow oil,
1
which, according to H-NMR data, consists of ca. 80% of 9 (yield 73%) and 20% of a-methylidenebenzene-
acetonitrile (13) [34 ± 36], cleaved from 9 by b-elimination. 9: TLC (petroleum ether/AcOEt 2 :1): Rf 0.82.
1H-NMR (CDCl3): 7.44 ± 7.35 (m, Ph); 4.53 (d, CH2); 4.23 (t, CH). Anal. calc. for C9.8H7.8Cl0.8NO1.6 (9/13 80 :20)
(193.5): C 60.82, H 4.06, N 7.24; found: C 60.61, H 4.22, N 7.50.
6. 2,2-Bis(4-nitrophenyl)ethyl Carbonochloridate (10) [30]. As described in Exper. 4, with 3 (1.30 g,
4.51 mmol), Et3N (0.63 ml, 460 mg, 4.51 mmol), THF (10 ml), trichloromethyl carbonochloridate (0.82 ml,
1.35 g, 6.80 mmol), and THF (8 ml), addition within 10 min, stirring for 10 min at 08 and 3 h 30 min at r.t.: 1.59 g
(quant.) ([30]: quant.) of 10. Pale yellow, cloudy oil which crystallized on cooling in the refrigerator. Almost
colorless solid. M.p. 95 ± 978 ([30]: 97 ± 988). TLC (petroleum ether/AcOEt 2 :1): Rf 0.70. UV (MeOH): 271
(4.34), 210 (sh, 4.23). 1H-NMR (CDCl3): 8.21 (d, 4 H o to NO2); 7.40 (d, 4 H m to NO2); 4.88 (d, CH2); 4.67 (t,
CH). Anal. calc. for C15H11ClN2O6 (350.7): C 51.37, H 3.16, N 7.99; found: C 51.37, H 3.30, N 7.94.
7. 1-(4-Nitrobenzyl)-2-(4-nitrophenyl)ethyl Carbonochloridate (11). As described in Exper. 4, with 7 (1.00 g,
3.31 mmol), Et3N (0.46 ml, 340 mg, 3.32 mmol), THF (9 ml), trichloromethyl carbonochloridate (0.8 ml, 1.31 g,
6.63 mmol), and THF (6 ml); addition within 10 min, stirring for 10 min at 08 and 4 h 10 min at r.t.: 1.15 g (95%)
of 11. Colorless solid. M. p. 140 ± 1448 (dec.). TLC (petroleum ether/AcOEt 2 :1): Rf 0.52. UV (MeOH): 271

Helvetica Chimica Acta ± Vol. 84 (2001)
1513
(4.33), 211 (sh, 4.20). 1H-NMR (CDCl3): 8.19 (d, 4 H o to NO2); 7.37 (d, 4 H m to NO2); 5.31 (quint., CH); 3.09
(d, 2 CH2). Anal. calc. for C16H13ClN2O6 (364.7): C 52.69, H 3.59, N 7.68; found: C 52.84, H 3.75, N 7.71.
8. b-Chlorobenzenepropanenitrile (12) [33]. A soln. of thionyl chloride (1.2 ml, 1.96 g, 16 mmol) in dry
toluene (1.6 ml) was added quickly and dropwise to a soln. of b-hydroxybenzenepropanenitrile (1; 804 mg,
5.46 mmol) and pyridine (0.32 ml) in dry toluene (15 ml), and the mixture was refluxed for 1.5 h. After cooling
to r.t., the mixture was poured onto ice (19 g), the org. phase was extracted with CH2Cl2 (3 Â 20 ml), the
combined org. phase washed with sat. NaHCO3 soln. (2 Â 60 ml), dried (Na2SO4), and evaporated, and the oil
purified by FC (silica gel (40 g), 14.5 Â 2.9 cm, petroleum ether/AcOEt 9 :1 (200 ml), 8 :1 (180 ml), 7:1
(160 ml), 6 :1 (140 ml)): 339 mg (37%) ([33]: 47%) of 12. Colorless oil. TLC (petroleum ether/AcOEt 5 :1): Rf
0.58. 1H-NMR (CDCl3): 7.43 ± 7.37 (m, Ph); 5.10 (t, CH); 3.10, 3.09 (2d, CH2). Anal. calc. for C9H8ClN (165.6): C
65.27, H 4.87, N 8.46; found: C 64.96, H 4.98, N 8.33.
9. 5'-O-[(2-Cyano-1-phenylethoxy)carbonyl]thymidine (16) and 3',5'-Bis-O-[(2-cyano-1-phenylethoxy)car-
bonyl]thymidine (17). After co-evaporation with dry pyridine (3 Â 3 ml), thymidine (15; 204 mg, 0.84 mmol)
was dissolved in dry pyridine (4 ml) and cooled under N2 to 558 (iPrOH/N2 bath). A soln. of an oil, containing
ca. 80% of 8 (230 mg, 1.10 mmol), in dry CH2Cl2 (4 ml) was added dropwise within 15 min, and stirring was
continued for 6 h at 55 to 208. Then the mixture was diluted with CH2Cl2 (20 ml) and washed with H2O
(20 ml), the aq. layer extracted with CH2Cl2 (2 Â 20 ml), the combined org. phase dried (Na2SO4), evaporated,
and co-evaporated with toluene (4 Â 10 ml), and the residue submitted to FC (silica gel (20 g), 14 Â 2.3 cm,
CH2Cl2/MeOH 100 :1 (101 ml), 100 :2 (204 ml), 100 :3 (309 ml), 100 :4 (208 ml), 100 :5 (105 ml)): 250 mg
(71%) of 16. Combined fractions containing 17 were still contaminated with b-hydroxybenzenepropanenitrile
(1) and purified by another FC (silica gel (10 g), 14.5 Â 1.4 cm, CH2Cl2 (100 ml), CH2Cl2/MeOH 100 :1 (101 ml),
100 :2 (102 ml)): 69 mg (14%) of 17.
Data of 16: Colorless foam. TLC (toluene/AcOEt/MeOH 5 :4 :1): Rf 0.25. UV (MeOH): 264 (3.98), 213
(sh, 4.25). 1H-NMR (CDCl3): 8.37 (br. s, NH); 7.39 ± 7.35 (m, Ph (1pceoc)); 7.28, 7.24 (2s, H C(6)); 6.29, 6.26 (2t,
H
C(1')); 5.91 ± 5.81 (m, CH (1pceoc)); 4.51 ± 4.32 (m, H C(3'), 2 H C(5')); 4.13 ± 4.09 (m, H C(4')); 2.96 ±
2.92 (m, CH2 (1pceoc)); 2.45 ± 2.31 (m, H C(2'), OH C(3')); 2.26 ± 2.11 (m, H C(2')); 1.90, 1.81 (2s, Me).
Anal. calc. for C20H21N3O7 (415.4): C 57.83, H 5.10, N 10.12; found: C 57.21, H 5.16, N 9.55.
Data of 17: Colorless foam. TLC (toluene/AcOEt/MeOH 5 :4 :1): Rf 0.49. UV (MeOH): 262 (3.92), 213
(sh, 4.19). 1H-NMR (CDCl3): 8.03 (br. s, NH); 7.40 ± 7.35 (m, 2 Ph (1cpeoc)); 7.21 (s, H C(6)); 6.33 ± 6.28 (m,
H
H
C(1')); 5.89 ± 5.73 (m, 2 CH (1pceoc)); 5.20 ± 5.09 (m, H C(3')); 4.49 ± 4.29 (m, 2 H C(5')); 4.19 (m,
C(4')); 3.03 ± 2.88 (m, 2 CH2 (1cpeoc)); 2.52 ± 2.42 (m, H C(2')); 2.31 ± 2.26 (m, H C(2')); 1.91, 1.73 (2s,
Me). Anal. calc. for C30H28N4O9 ´ 0.5 H2O (597.6): C 60.30, H 4.89, N 9.38; found: C 60.65, H 4.89, N 9.25.
10. 5'-O-[(2-Cyano-2-phenylethoxy)carbonyl]thymidine (18), 3'-O-[(2-Cyano-2-phenylethoxy)carbon-
yl]thymidine (19), and 3',5'-Bis-O-[(2-cyano-2-phenylethoxy)carbonyl]thymidine (20). As described in Exper. 9,
with 15 (201 mg, 0.83 mmol), pyridine (5 ml), 9 (230 mg, 1.10 mmol; oil of ca. 80% purity), and CH2Cl2 (5 ml);
addition within 5 min, stirring for 6 h 30 min at 55 to 108. Washing with a mixture of H2O (20 ml) and conc.
HCl (3.7 ml). Co-evaporation with H2O (4 Â 10 ml), MeOH (3 Â 10 ml), and CH2Cl2 (2 Â 10 ml). FC (silica gel
(20 g), 14 Â 2.1 cm, CH2Cl2/MeOH 100 :1 (101 ml), 100 :2 (204 ml), 100 :3 (206 ml), 100 :4 (104 ml)) gave
233 mg (68%) of 18, 16 mg (5%) of 19, and 57 mg (12%) of 20.
Data of 18: Colorless foam. TLC: Rf 0.23 (toluene/AcOEt/MeOH 5 :4 :1). UV (MeOH): 264 (4.01), 213
(sh, 4.24). 1H-NMR (CDCl3): 8.60 (br. s, NH); 7.43 ± 7.33 (m, Ph (2pceoc)); 7.31, 7.28 (2s, H C(6)); 6.31 (t,
H
H
C(1')); 4.50 ± 4.33 (m, CH2 (2pceoc), H C(3'), 2 H C(5')); 4.25 ± 4.17 (m, CH (2pceoc)); 4.14 ± 4.12 (m,
C(4')); 2.60 (br. s, OH C(3')); 2.44 ± 2.35 (m, H C(2')); 2.24 ± 2.13 (m, H C(2')); 1.92, 1.87 (2s, Me). Anal.
calc. for C20H21N3O7 ´ 0.5 H2O (424.4): C 56.60, H 5.23, N 9.90; found: C 56.89, H 5.14, N 9.88.
Data of 19: Colorless foam. TLC: Rf 0.30 (toluene/AcOEt/MeOH 5 :4 :1). UV (MeOH): 263 (3.99), 213
(sh, 4.15). 1H-NMR (CDCl3): 8.93 (br. s, NH); 7.42 ± 7.38 (m, Ph (2pceoc), H C(6)); 6.18 ± 6.09 (m, H C(1'));
5.28 (m, H C(3')); 4.41 (d, CH2 (2pceoc)); 4.23 ± 4.14 (m, CH (2pceoc), H C(4')); 3.96 ± 3.83 (m, 2 H C(5'));
2.88 (br. s, OH C(5')); 2.59 ± 2.48 (m, H C(2')); 2.48 ± 2.37 (m, H C(2')); 1.90 (s, Me).
Data of 20: Colorless foam. TLC (toluene/AcOEt/MeOH 5 :4 :1): Rf 0.47. UV (MeOH): 263 (3.98), 212
(sh, 4.32). 1H-NMR (CDCl3): 8.96 (br. s, NH); 7.47 ± 7.34 (m, 2 Ph (2pceoc), H C(6)); 6.32 (t, H C(1')); 5.21 ±
5.14 (m, H C(3')); 4.48 ± 4.41 (m, 2 CH2 (2pceoc), H C(5')); 4.28 ± 4.19 (m, 2 CH (2pceoc), H C(5')); 4.00 ±
3.91 (m, H C(4')); 2.50 ± 2.45 (m, H C(2')); 2.32 ± 2.19 (m, H C(2')); 1.91, 1.86 (2s, Me). Anal. calc. for
C30H28N4O9 (588.6): C 61.22, H 4.80, N 9.52; found: C 61.32, H 5.13, N 9.07.
11. 5'-O-{[2,2-Bis(4-nitrophenyl)ethoxy]carbonyl}thymidine (21), 3'-O-{[2,2-Bis(4-nitrophenyl)ethoxy]car-
bonyl}thymidine (22), and 3',5'-Bis-O-{[2,2-bis(4-nitrophenyl)ethoxy]carbonyl}thymidine (23). As described in
Exper. 9, with 15 (200 mg, 0.83 mmol), pyridine (5 ml), 10 (376 mg, 1.07 mmol), and CH2Cl2 (5 ml); addition

1514
Helvetica Chimica Acta ± Vol. 84 (2001)
within 20 min, stirring for 6 h 10 min at 55 to 108. FC (20 g silica gel, 14 Â 2.1 cm, CH2Cl2/MeOH 100 :1
(101 ml), 100 :2 (102 ml), 100 :3 (103 ml), 100 :4 (104 ml), 100 :5 (105 ml)) gave 342 mg (74%) of 21 and 19 mg
(4%) of 22. Combined fractions containing product 23 were still contaminated and purified by another FC
(silica gel (7 g), 14.5 Â 1.4 cm, CH2Cl2 (100 ml), CH2Cl2/MeOH 100 :0.5 (100.5 ml), 100 :1 (101 ml), 100 :2
(102 ml)): 51 mg (7%) of 23.
Data of 21: Colorless foam. TLC (toluene/AcOEt/MeOH 5 :4 :1): Rf 0.39. UV (MeOH): 268 (4.43), 212
(sh, 4.36). 1H-NMR (CDCl3): 8.20 (d, 4 H o to NO2); 8.00 (br. s, NH); 7.39 (d, 4 H m to NO2); 7.18 (s, H C(6));
6.20 (t, H C(1')); 4.79 ± 4.73 (m, CH2 (np)2eoc); 4.68 ± 4.61 (m, CH (np)2eoc); 4.36 ± 4.34 (m, H C(3'),
2 H C(5')); 4.08 ± 4.05 (m, H C(4')); 2.40 ± 2.30 (m, H C(2')); 2.20 ± 2.12 (m, H C(2'), OH C(3')); 1.81 (s,
Me). Anal. calc. for C25H24N4O11 ´ 0.5 H2O (565.5): C 53.10, H 4.46, N 9.91; found: C 53.26, H 4.41, N 9.84.
Data of 22: Colorless foam. TLC (toluene/AcOEt/MeOH 5 :4 :1): Rf 0.50. UV (MeOH): 268 (4.42), 211
(sh, 4.36). 1H-NMR (CDCl3): 9.10 (br. s, NH); 8.20 (d, 4 H o to NO2); 7.42 (s, H C(6)); 7.41 (d, 4 H m to NO2);
6.13 (dd, H C(1')); 5.23 ± 5.21 (m, H C(3')); 4.75 ± 4.72 (m, CH2 (np)2eoc)); 4.64 ± 4.59 (t, CH (np)2eoc); 4.07
(m, H C(4')); 3.87 ± 3.84 (m, 2 H C(5')); 2.80 (br. s, OH C(5')); 2.43 ± 2.33 (m, 2 H C(2')); 1.86 (s, Me).
Anal. calc. for C25H24N4O11 ´ H2O (574.5): C 52.27, H 4.74, N 9.75; found: C 52.82, H 4.87, N 9.30.
Data of 23: Colorless foam. TLC (toluene/AcOEt/MeOH 5 :4 :1): Rf 0.67. UV (MeOH): 268 (4.58), 211
(sh, 4.47). 1H-NMR (CDCl3): 8.88 (br. s, NH); 8.20, 8.19 (2d, 8 H o to NO2); 8.39, 7.41 (2d, 8 H m to NO2); 7.13
(s, H C(6)); 6.18 (dd, H C(1')); 5.00 ± 4.97 (m, H C(3')); 4.84 ± 4.72 (m, 2 CH2 (np)2eoc); 4.70 ± 4.57 (m,
2 CH (np)2eoc); 4.33 ± 4.31 (m, 2 H C(5')); 4.15 ± 4.13 (m, H C(4')); 2.36 ± 2.33 (m, H C(2')); 2.14 ± 2.11 (m,
H
C(2')); 1.78 (s, Me). Anal. calc. for C40H34N6O17 ´ 0.5 H2O (879.7): C 54.61, H 4.01, N 9.55; found: C 54.56, H
4.27, N 9.25.
12. 5'-O-{[1-(4-Nitrobenzyl)-2-(4-nitrophenyl)ethoxy]carbonyl}thymidine (24), 3'-O-{[1-(4-Nitrobenzyl)-2-
(4-nitrophenyl)ethoxy]carbonyl}thymidine (25), and 3',5'-Bis-O-{[1-(4-nitrobenzyl)-2-(4-nitrophenyl)ethoxy]-
carbonyl}thymidine (26). As described in Exper. 9, with 15 (401 mg, 1.66 mmol), co-evaporated with pyridine
(3 Â 5 ml), pyridine (8 ml), 11 (787 mg, 2.16 mmol), and CH2Cl2 (7 ml); addition within 15 min at 458, stirring
for 6 h 10 min at 45 to 208. FC (silica gel (40 g), 14.5 Â 2.7 cm, CH2Cl2/MeOH 100 :1 (202 ml), 100 :2
(408 ml), 100 :3 (309 ml), 100 :4 (104 ml)) gave 770 mg (82%) of 24 and 25 mg (3%) of 25. Combined fractions
containing product 26 were still contaminated and purified by another FC (silica gel (13 g), 17 Â 1.3 cm, CH2Cl2/
MeOH 100 :1 (101 ml), 100 :2 (102 ml), 100 :3 (103 ml)): 83 mg (6%) of 26.
Data of 24: Colorless foam. TLC (toluene/AcOEt/MeOH 5 :4 :1): Rf 0.37. UV (MeOH): 267 (4.44), 210
(sh, 4.34). 1H-NMR (CDCl3): 8.96 (br. s, NH); 8.15, 8.11 (2d, 4 H o to NO2); 7.37, 7.33 (2d, 4 H m to NO2); 7.22 (s,
H
H
H
C(6)); 6.19 (t, H C(1')); 5.23 (quint., CH (nbnpeoc)); 4.31 ± 4.15 (m, H C(3'), 2 H C(5')); 4.04 ± 3.99 (m,
C(4')); 3.07 ± 2.97 (m, 2 CH2 (nbnpeoc)); 2.75 (br. s, OH C(3')); 2.40 ± 2.31 (m, H C(2')); 2.12 ± 2.02 (m,
C(2')); 1.87 (s, Me). Anal. calc. for C26H26N4O11 ´ H2O (588.5): C 53.06, H 4.80, N 9.52; found: C 53.00, H
4.53, N 9.36.
Data of 25: Colorless foam. TLC (toluene/AcOEt/MeOH 5 :4 :1): Rf 0.50. UV (MeOH): 267 (4.45), 211
(sh, 4.40). 1H-NMR (CDCl3): 8.87 (br. s, NH); 8.18, 8.15 (2d, 4 H o to NO2); 7.41 (s, H C(6)); 7.41, 7.38 (2d, 4 H
m to NO2); 6.16 (dd, H C(1')); 5.18 (quint., CH (nbnpeoc)); 5.06 ± 5.03 (m, H C(3')); 3.87 ± 3.85 (m,
H
H
C(4')); 3.80 ± 3.69 (m, 2 H C(5')); 3.04 (d, 2 CH2 (nbnpeoc)); 2.63 (br. s, OH C(5')); 2.39 ± 2.27 (m,
C(2')); 2.22 ± 2.15 (m, H C(2')); 1.87 (s, Me).
Data of 26: Colorless foam. TLC (toluene/AcOEt/MeOH 5 :4 :1): Rf 0.68. UV (MeOH): 269 (4.64), 212
(sh, 4.54). 1H-NMR (CDCl3): 8.47 (br. s, NH); 8.20 ± 8.09 (m, 8 H o to NO2); 7.41 ± 7.29 (m, 8 H m to NO2); 7.12
(s, H C(6)); 6.15 (dd, H C(1')); 5.17 (quint., 2 CH (nbnpeoc)); 4.63 (m, H C(3')); 4.12 ± 3.98 (m,
2 H C(5')); 3.82 (m, H C(4')); 3.07 ± 2.98 (2d, 4 CH2 (nbnpeoc)); 2.23 ± 2.16 (m, H C(2')); 2.01 ± 1.92 (m,
H
C(2')); 1.87 (s, Me). Anal. calc. for C42H38N6O17 (898.8): C 56.13, H 4.26, N 9.35; found: C 55.80, H 4.54, N
8.99.
13. 5'-O-(Diphenylphosphinoyl)thymidine (27) and 3',5'-Bis-O-(diphenylphosphinoyl)thymidine (28). As
described in Exper. 9, with 15 (202 mg, 0.83 mmol), pyridine (4 ml), diphenylphosphinoyl chloride (14; 0.21 ml,
260 mg, 1.10 mmol), and CH2Cl2 (3 ml); addition within 10 min at 558, stirring for 6 h 30 min at 50 to 208.
FC (silica gel (20 g), 14 Â 2.1 cm, CH2Cl2/MeOH 100 :1 (101 ml), 100 :2 (204 ml), 100 :3 (309 ml), 100 :4
(208 ml), 100 :5 (105 ml)) gave 197 mg (53%) of 27 and 87 mg (16%) of 28.
Data of 27: Colorless foam. TLC (toluene/AcOEt/MeOH 5 :4 :1): Rf 0.22. UV (MeOH): 270 (sh, 3.96), 265
(4.00), 261 (sh, 3.98), 219 (sh, 4.39). 1H-NMR (CDCl3): 9.11 (br. s, NH); 7.84 ± 7.73 (m, 4 H (dpp)); 7.56 ± 7.39 (m,
6 H (dpp)); 7.20 (s, H C(6)); 6.29 (t, H C(1')); 5.72 (br. s, OH C(3')); 4.62 ± 4.59 (m, H C(3')); 4.25 (m,
2 H C(5')); 4.02 (m, H C(4')); 2.47 ± 2.35 (m, H C(2')); 2.16 ± 2.05 (m, H C(2')); 1.57 (s, Me). Anal. calc. for
C22H23N2O6P (442.4): C 59.73, H 5.24, N 6.33; found: C 59.81, H 5.29, N 5.81.

Helvetica Chimica Acta ± Vol. 84 (2001)
1515
Data of 28: Colorless foam. TLC (toluene/AcOEt/MeOH 5 :4 :1): Rf 0.32. UV (MeOH): 271 (sh, 4.64),
265 (4.14), 261 (sh, 4.11), 222 (4.60). 1H-NMR (CDCl3): 8.33 (br. s, NH); 7.84 ± 7.68 (m, 8 H (dpp)); 7.57 ± 7.38
(m, 12 H (dpp), H C(6)); 6.40 (dd, H C(1')); 5.24 (m, H C(3')); 4.43 (m, H C(4')); 4.24 (m, H C(5'));
4.05 ± 3.96 (m, H C(5')); 2.57 (m, H C(2')); 2.17 ± 2.03 (m, H C(2')); 1.65 (s, Me). Anal. calc. for
C34H32N2O7P2 ´ H2O (660.6): C 61.82, H 5.19, N 4.24; found: C 61.56, H 5.14, N 4.39.
14. 3-Phenylprop-2-enenitrile (32): (Z)-Isomer 32a and (E)-Isomer 32b [41][47][48]. To a soln. of an oil,
containing ca. 80% of 8 (3.06 g, 15 mmol), in dry CH2Cl2 (50 ml), DBU (6.2 ml, 6.32 g, 41 mmol) was added
under CO2 release in an exothermic reaction. After stirring for 30 min, the mixture was diluted with CH2Cl2
(20 ml) and washed subsequently with 2m HCl (2 Â 80 ml), H2O (80 ml), 1m NaOH (2 Â 80 ml), and sat. NaCl
soln. (2 Â 80 ml). The org. layer was dried (Na2SO4) and evaporated and the yellow oil submitted to distillation
at 119 ± 1228/12 ± 13 mbar ([44]: 121 ± 1238/9 Torr): 1.42 g (76%) of 32 in a ratio 32a/32b 3 :7.
Data of 32: Colorless oil. TLC (petroleum ether/AcOEt 2 :1): Rf 0.80. UV (MeOH): 273 (4.29), 215 (sh,
4.16) ([45]: 273 (4.22) for 32a, 272 (4.36) for 32b). Anal. calc. for C9H7N ´ 0.125 H2O (131.4): C 82.26, H 5.47, N
10.66; found: C 82.35, H 5.56, N 10.9.
Data of 32a: 1H-NMR (CDCl3): 7.81 ± 7.75 (m, 2 H (Ph)); 7.47 ± 7.35 (m, 3 H (Ph)); 7.12 (d, J ˆ 12.0,
PhCHCHCN); 5.44 (d, J ˆ 12.1, PhCHCHCN).
Data of 32b: 1H-NMR (CDCl3): 7.47 ± 7.35 (m, Ph, PhCHCHCN); 5.87 (d, J ˆ 16.6, PhCHCHCN).
15. 1,1-Bis(4-nitrophenylethene) (ˆ1,1'-Ethenylidenebis[4-nitrobenzene]; 33) [30]. On adding DBU
(0.456 ml, 465 mg, 3.05 mmol) to a yellow soln. of compound 30 [30] (505 mg, 1.53 mmol) in dry CH2Cl2
(4.5 ml), the colour turned immediately dark blue. After stirring for 30 min, the mixture was diluted with CH2Cl2
(10 ml) and washed subsequently with 1m HCl (2 Â 10 ml), H2O (10 ml), 2m NaOH (2 Â 10 ml), and sat. NaCl
soln. (3 Â 10 ml). The org. layer was dried (Na2SO4) and evaporated and the yellow solid recrystallized from
CH2Cl2/petroleum ether 6 :5 (20 ml): 290 mg (70%) ([30]: 69%) of 33. Slightly yellow crystals. TLC (petroleum
ether/AcOEt 2 :1): Rf 0.72. UV (MeOH): 300 (3.78), 216 (sh, 3.75) ([27]: 304). 1H-NMR (CDCl3): 8.22 (d, 4 H
o to NO2); 7.45 (d, 4 H m to NO2); 5.75 (s, CH2). Anal. calc. for C14H10N2O4 (270.2): C 62.22, H 3.73, N 10.37;
found: C 62.50, H 3.81, N 10.37.
16. (4-Nitrophenyl)ethene (ˆ4-Nitro-1-ethenylbenzene; 34) [49]. To a soln. of 31 [37] (215 mg, 0.94 mmol)
in dry MeCN (94 ml), DBU (1.4 ml, 1.43 g, 9.37 mmol) was added and the mixture stirred overnight for 20 h.
The reaction was stopped by adding AcOH (0.536 ml, 9.37 mmol), the mixture diluted with CH2Cl2 (20 ml) and
washed with H2O (20 ml), the aq. layer reextracted with CH2Cl2 (2 Â 20 ml), the combined org. phase washed
with sat. NaHCO3 soln. (60 ml), the aq. layer extracted with CH2Cl2 (2 Â 40 ml), the combined org. phase dried
(Na2SO4) and evaporated, and the residue purified by FC (silica gel (10 g), 6.5 Â 2.1 cm, CH2Cl2/MeOH 100 :1
(101 ml)): 76 mg (54%) of 34. Colorless oil. TLC (toluene/AcOEt/MeOH 5 :4 :1): Rf 0.90. UV (MeOH): 300
1
(4.14), 219 (3.99) ([46]: 298 (4.14), 218 (4.00)). H-NMR (CDCl3): 8.17 (d, 2 H o to NO2); 7.52 (d, 2 H m to
NO2); 6.77 (dd, CH); 5.92, 5.48 (2d, CH2).
17. 1-[2,2-Bis(4-nitrophenyl)ethyl]piperidine (35) [30]. To a yellow suspension of 33 (100 mg, 0.37 mmol) in
MeCN (3 ml), piperidine (1 ml, 860 mg, 10.1 mmol) was added ( ! soln. and colour change to blue). After
5 min stirring, the colour turned yellow again. After 15 min, the mixture was diluted with CH2Cl2 (20 ml) and
washed with H2O (20 ml), the aq. layer extracted with CH2Cl2 (2 Â 20 ml), the combined org. phase dried
(Na2SO4) and evaporated, and the product purified by FC (silica gel (7 g), 14.5 Â 1.3 cm, petroleum ether/
AcOEt 9 :1 (100 ml), 8 :1 (90 ml), 7:1 (80 ml), 6 :1 (70 ml), 5 :1 (60 ml)): 116 mg (88%) of 35. Orange-red oil.
TLC (petroleum ether/AcOEt 2 :1): Rf 0.40. UV (MeOH): 275 (4.31), 211 (sh, 4.28) ([27]: 275). 1H-NMR
(CDCl3): 8.15 (d, 4 H o to NO2); 7.35 (d, 4 H m to NO2); 4.40 (t, CH (bnpe)); 2.91 (d, CH2 (bnpe)); 2.38 (m,
2 CH2N); 1.46 ± 1.23 (m, 3 CH2). Anal. calc. for C19H21N3O4 ´ 0.25 H2O (359.9): C 63.41, H 5.95, N 11.68; found:
C 63.54, H 5.96, N 11.39.
18. Kinetic Studies. 18.1 Method A: Nucleoside 16, 18, 21, or 29, resp., was treated with 10 equiv. of either
DBU/MeCN (0.01, 0.1, and 1m), piperidine/MeCN (0.5m), or Et3N/MeCN (0.5m) in an attached cap Eppendorf
tube at r.t. with stirring. At defined time intervals, aliquots (15 ± 350 ml) were quenched in AcOH/MeCN (0.05 ±
0.25m, at least 3 equiv. AcOH per base, end volume of sample 500 ml), and analysed by reversed-phase HPLC.
The relative ratio of starting material and cleaved protecting group was detected at 268 or 266 nm. For each
experiment, 5 ± 10 time points were taken, plus an assumed 0 min time point. The half-lives (t1/2) for the pseudo-
first-order reactions were obtained from the graphical plots of ln(1 ‡ esmApg/epgAsm) vs. time, where e is the
extinction coefficient of starting material (esm) or cleaved protecting group (epg), respectively, at 268 or 266 nm,
and A is the area under the peak of starting material (Asm) or cleaved protecting group (Apg), respectively. 15: tR
8.25 min. 16: tR 12.11 ‡ 12.18 min, e268 9014. 18: tR 12.51 min. 21: tR 13.56 min, e268 24995. 29: tR 13.44 min, e266

1516
Helvetica Chimica Acta ± Vol. 84 (2001)
18373. 32: tR 14.97 ‡ 15.17 min, e268 19336 (for 32a/32b 3 :7). 33: tR 19.82 min, e268 11772. 34: tR 16.76 min, e266
5131. 35: tR 20.96 min, e268 19039.
18.2. Method B: Nucleoside 24 or 27 was dissolved in MeCN (for 24) or MeOH (for 27), respectively, in an
attached cap Eppendorf tube. Then, 2 ± 3 aliquots (15 ± 40 ml) were quenched in AcOH/MeCN (0.05 ± 0.25m, end
volume of sample 500 ml), and analysed by HPLC for the determination of the area under the peak of starting
material at time t0 (A0). Then the cleavage reaction was started by addition of 10 equiv. of either DBU/MeCN
(for 24, 0.2 and 2m, end concentration of reaction soln. 0.1 and 1m) or DBU/MeOH (for 27, 2m, end
concentration of reaction soln. 1m) and stirred at r.t.. At defined time intervals, aliquots (15 ± 40 ml) were
quenched in AcOH/MeCN (0.05 ± 0.25m, at least 3 equiv. AcOH per base, end volume of sample 500 ml), and
analysed by reversed-phase HPLC. The decrease of starting material was detected by 268 nm. For each
experiment, 5 ± 10 aliquots were taken. The half-lives (t1/2) for the pseudo-first-order reactions were obtained
from the graphical plots of ln(At/A0) vs. time, where At is the area under the peak of starting material at time t.
24: tR 15.51 min. 29: tR 11.97 min.
19. 2'-O-(Diphenylphosphinoyl)-3',5'-O-(1,1,3,3-tetraisopropyldisiloxane-1,3-diyl)uridine (37). 3',5'-O-
(1,1,3,3-Tetraisopropyldisiloxane-1,3-diyl)uridine (36) [45] (608 mg, 1.25 mmol) and diphenylphosphinoyl
chloride (14; 0.485 ml, 600 mg, 2.54 mmol) were stirred in dry pyridine (7.5 ml) overnight for 16 h. Then the
mixture was diluted with CH2Cl2 (20 ml) and washed with H2O (20 ml), the aq. phase extracted with CH2Cl2
(2 Â 20 ml), the combined org. layer dried (Na2SO4), evaporated, and co-evaporated with toluene (4 Â 10 ml),
and the residue purified by FC (silica gel (30 g), 15 Â 2.7 cm, CH2Cl2/MeOH 100 :1 (101 ml), 100 :2 (204 ml),
100 :3 (206 ml), 100 :4 (104 ml)): 690 mg (80%) of 37. Colorless foam. TLC (toluene/AcOEt/MeOH 5 :4 :1): Rf
0.38. UV (MeOH): 270 (sh, 3.93), 263 (sh, 4.04), 259 (4.05), 220 (sh, 4.27). 1H-NMR (CDCl3): 8.20 (br. s, NH);
7.83 (m, 4 H (dpp)); 7.67 (d, H C(6)); 7.53 ± 7.40 (m, 6 H (dpp)); 5.86 (s, H C(1')); 5.62 (d, H C(5)); 5.01 (dd,
H
C(2')); 4.42 (dd, H C(3')); 4.32 ± 4.14 (m, H C(4'), H C(5')); 3.98 (m, H C(5')); 1.06 ± 0.92 (m,
4 Me2CH). Anal. calc. for C33H47N2O8PSi2 (686.9): C 57.70, H 6.90, N 4.08; found: C 57.54, H 6.98, N 4.36.
20. 2'-O-(Diphenylphosphinoyl)uridine (38) and 3'-O-(Diphenylphosphinoyl)uridine (39). To a soln. of 37
(201 mg, 0.29 mmol) in dry THF (3 ml), AcOH (80 ml, 84 mg, 1.40 mmol) and Bu4NF ´ 3H2O (230 mg,
0.73 mmol) were added. The mixture was stirred for 2 h, then diluted with CH2Cl2 (20 ml), and washed with H2O
(20 ml). The aq. phase was extracted with CH2Cl2 (2 Â 20 ml) and the combined org. layer dried (Na2SO4),
evaporated, and co-evaporated with H2O (2 Â 10 ml), MeOH (3 Â 10 ml), and CH2Cl2 (2 Â 10 ml). The residue
was purified by FC (3 g silica gel, 6 Â 1.3 cm, CH2Cl2/MeOH 100 :1 (50.5 ml), 100 :2 (51 ml), 100 :3 (51.5 ml),
100 :4 (52 ml), 100 :5 (105 ml)): 84 mg (65%) of 38/39 1 :2. Colorless foam. TLC (toluene/AcOEt/MeOH
5 :4 :1): Rf 0.21. Anal. calc. for C21H21N2O7P ´ H2O (462.4): C 54.55, H 5.01, N 6.06; found: C 54.64, H 4.86, N
6.07.
Data of 38: 1H-NMR ((D6)DMSO): 11.27 (br. s, NH); 7.91 ± 7.75 (m, 4 H (dpp)); 7.67 ± 7.41 (m, 6 H (dpp),
H
C(6)); 6.16 (d, H C(1')); 5.79 (d, OH C(3')); 5.48 (d, H C(5)); 5.19 (t, OH C(5')); 4.63 (m, H C(2'));
4.10 (m, H C(3')); 3.98 (m, H C(4')); 3.53 ± 3.30 (m, 2 H C(5')).
Data of 39: 1H-NMR ((D6)DMSO): 11.39 (br. s, NH); 7.91 ± 7.75 (m, 4 H (dpp)); 7.67 ± 7.41 (m, 6 H (dpp),
H
C(6)); 5.94 (2d, H C(1'), OH C(2')); 5.70 (d, H C(5)); 5.19 (t, OH C(5')); 4.60 (m, H C(3')); 4.26 ±
4.21 (m, H C(2'), H C(4')); 3.53 ± 3.30 (m, 2 H C(5')).
21. 2'-O-(Diphenylphosphinoyl)-5'-O-(monomethoxytrityl)uridine (40) and 3'-O-(Diphenylphosphinoyl)-
5'-O-(monomethoxytrityl)uridine (41). A mixture 38/39 1:2 (80 mg, 0.18 mmol) and MeOTrCl (79 mg,
0.26 mmol) was stirred in dry pyridine (2 ml) overnight for 18 h. After addition of more MeOTrCl (48 mg,
0.15 mmol) and stirring for another 4 h, the mixture was diluted with CH2Cl2 (20 ml) and washed with H2O
(20 ml). The aq. phase was extracted with CH2Cl2 (2 Â 20 ml), the combined org. layer dried (Na2SO4),
evaporated, and co-evaporated with toluene (4 Â 10 ml) and CH2Cl2 (2 Â 10 ml), and the residue separated by
FC (7.5 g silica gel, 14 Â 1.5 cm, CH2Cl2/MeOH 100 :1 (101 ml), 100 :2 (204 ml), 100 :3 (51.5 ml)): 40 mg (31%)
of 40 and 52 mg of 41 (40%).
Data of 40: Colorless foam. TLC: Rf 0.40 (toluene/AcOEt/MeOH 5 :4 :1). UV (MeOH): 271 (sh, 3.93),
265 (sh, 4.01), 259 (4.02), 223 (sh, 4.50). 1H-NMR (CDCl3): 9.37 (br. s, NH); 7.94 ± 7.78 (m, 4 H (dpp), H C(6));
7.59 ± 7.47 (m, 6 H (dpp)); 7.31 (m, 10 H (MeOTr), 2 H o to MeO); 6.86 (d, 2 H m to MeO); 6.32 (d, H C(1'));
5.18 (d, H C(5)); 4.78 (m, H C(2'), OH C(3')); 4.64 (m, H C(3')); 4.28 (m, H C(4')); 3.82 (s, MeO); 3.53
(m, 2 H C(5')). Anal. calc. for C41H37N2O8P (716.7): C 68.71, H 5.20, N 3.91; found: C 68.40, H 5.33, N 4.45.
Data of 41: Colorless foam. TLC (toluene/AcOEt/MeOH 5 :4 :1): Rf 0.28. UV (MeOH): 270 (sh, 3.97), 264
1
(sh, 4.06), 260 (4.06), 222 (sh, 4.49). H-NMR (CDCl3): 9.66 (br. s, NH); 7.83 ± 7.70 (m, 4 H (dpp), H C(6));
7.59 ± 7.37 (m, 6 H (dpp)); 7.28 (m, 10 H (MeOTr)); 7.22 (d, 2 H o to MeO); 6.84 (d, 2 H m to MeO); 6.10 (d,
H
C(1')); 5.26 (d, H C(5)); 5.03 (br. s, OH C(2')); 4.82 (m, H C(3')); 4.48 (m, H C(2'), H C(4')); 3.82 (s,

Helvetica Chimica Acta ± Vol. 84 (2001)
1517
MeO); 3.59 (m, H C(5')); 3.42 ± 3.38 (m, H C(5')). Anal. calc. for C41H37N2O8P ´ H2O (734.7): C 67.02, H 5.35,
N 3.81; found: C 67.24, H 5.26, N 4.46.
REFERENCES
[1] S. Matysiak, H.-P. Fitznar, R. Schnell, W. Pfleiderer, Helv. Chim. Acta 1998, 81, 1545.
[2] T. R. Cech, Science (Washington, D.C.) 1987, 236, 1532.
[3] S. Altman, M. Baer, C. Guerrier-Takada, A. Vioque, Trends Biochem. Sci. 1988, 11, 515.
[4] E. Uhlmann, A. Peyman, Chem. Rev. 1990, 90, 543.
[5] Y. Eguchi, T. Itoh, J. Tomizawa, Annu. Rev. Biochem. 1991, 60, 631.
[6] C. Tuerk, L. Gold, Science (Washington, D.C.) 1990, 249, 505.
[7] A. D. Ellington, J. W. Szostak, Nature (London) 1990, 346, 818.
[8] R. D. Jenison, S. C. Gill, A. Pardi, B. Polsky, Science (Washington, D.C.) 1994, 263, 1425.
[9] M. D. Matteucci, M. H. Caruthers, Tetrahedron Lett. 1980, 21, 719.
[10] S. L. Beaucage, M. H. Caruthers, Tetrahedron Lett. 1981, 22, 1859.
[11] M. A. Dorman, S. A. Noble, C. J. McBride, M. H. Caruthers, Tetrahedron 1984, 40, 95.
[12] M. D. Matteucci, M. H. Caruthers, J. Am. Chem. Soc. 1981, 103, 3185.
[13] N. D. Sinha, J. Biernat, H. Köster, Nucleic Acids Res. 1984, 12, 4539.
[14] K. K. Ogilvie, K. L. Sadana, E. A. Thompson, M. A. Quilliam, J. B. Westmore, Tetrahedron Lett. 1974, 15, 2861.
[15] K. K. Ogilvie, N. Theriault, K. L. Sadana, J. Am. Chem. Soc. 1977, 99, 7741.
[16] T. Wu, K. K. Ogilvie, R. T. Pon, Nucleic Acids Res. 1989, 17, 3501.
[17] B. E. Griffin, C. B. Reese, Tetrahedron Lett. 1964, 5, 2925.
[18] C. B. Reese, R. Saffhill, J. E. Sulston, J. Am. Chem. Soc. 1967, 89, 3366.
[19] T. S. Rao, C. B. Reese, H. T. Serafinowska, H. Takaku, G. Zappia, Tetrahedron Lett. 1987, 28, 4897.
[20] O. Sakatsume, M. Ohtsuki, H. Takaku, C. B. Reese, Nucleic Acids Res. 1989, 17, 3689.
[21] C. B. Reese, E. A. Thompson, J. Chem. Soc., Perkin Trans. 1 1988, 2881.
[22] M. V. Rao, C. B. Reese, V. Schehlmann, P. S. Yu, J. Chem. Soc., Perkin Trans. 1 1993, 43.
[23] S. A. Scaringe, F. E. Wincott, M. H. Caruthers, J. Am. Chem. Soc. 1998, 120, 11820.
[24] X. Wu, S. Pitsch, Nucleic Acids Res. 1998, 19, 4315.
[25] S. Pitsch, P. A. Weiss, X. Wu, D. Ackermann, T. Honegger, Helv. Chim. Acta 1999, 82, 1753.
[26] C. Lehmann, Y.-Z. Xu, C. Christodoulou, Z.-K. Tan, M. J. Gait, Nucleic Acids Res. 1989, 17, 2379.
[27] F. Bergmann, W. Pfleiderer, Helv. Chim. Acta 1994, 77, 481.
[28] F. Bergmann, W. Pfleiderer, Helv. Chim. Acta 1994, 77, 988.
[29] M. Kaiser, C. R. Hauser, J. Org. Chem. 1968, 33, 3402.
[30] R. Ramage, A. J. Blake, M. R. Florence, T. Gray, G. Raphy, P. L. Roach, Tetrahedron 1991, 47, 8001.
[31] R. Adams, H. O. Calvary, Org. Synth. Collect. Vol. II 1943, 287.
[32] J. R. Cannon, K. T. Potts, C. L. Raston, A. F. Sierakowski, A. H. White, Aust. J. Chem. 1978, 31, 297.
[33] T. N. Majid, M. C. P. Yeh, P. Knochel, Tetrahedron Lett. 1989, 30, 5069.
[34] J. M. Steward, C. H. Chang, J. Org. Chem. 1956, 21, 635.
[35] J. Colonge, J. Dreux, J.-P. Regeaud, Bull. Soc. Chim. Fr. 1959, 1244.
Â
[36] M. Bremond, G. J. Martin, M. Cariou, Org. Magn. Reson. 1978, 9, 433.
[37] F. Himmelsbach, B. S. Schulz, T. Trichtinger, R. Charubala, W. Pfleiderer, Tetrahedron 1984, 40, 59.
[38] K. P. Stengele, W. Pfleiderer, Tetrahedron Lett. 1990, 31, 2549.
[39] E. Uhlmann, W. Pfleiderer, Helv. Chim. Acta 1981, 64, 1688.
[40] H. Schirmeister, F. Himmelsbach, W. Pfleiderer, Helv. Chim. Acta 1993, 76, 385.
[41] I. Matsuda, S. Murata, Y. Ishii, J. Chem. Soc., Perkin Trans. 1 1979, 26.
[42] B. E. Griffin, M. Jarman, C. B. Reese, J. E. Sulston, Tetrahedron 1967, 23, 3201.
[43] H. P. M. Fromageot, B. E. Griffin, C. B. Reese, J. E. Sulston, Tetrahedron 1967, 23, 3215.
[44] H. P. M. Fromageot, C. B. Reese, J. E. Sulston, Tetrahedron 1968, 24, 3533.
[45] W. T. Markiewicz, J. Chem. Res. (M) 1979, 181.
[46] N. S. Vulꢀfson, L. C. Vinograd, Dokl. Akad. Nauk SSSR 1956, 106, 669; Chem. Abstr. 1960, 54, 10940h.
[47] H. Takahashi, F. Fujiwara, M. Ohta, Bull. Chem. Soc. Jpn. 1962, 35, 1498.
[48] W. E. Parham, W. N. Moulton, A. Zuckerbraun, J. Org. Chem. 1956, 21, 72.
[49] C. Buschhaus, Ph. D. Thesis, University of Konstanz, 1988.
Received April 2, 2001