Thermolytic Carbonates for Potential 5′-Hydroxyl Protection of Deoxyribonucleosides

Article abstract of DOI:10.1021/jo035089g

Thermolytic groups structurally related to well-studied heat-sensitive phosphate/thiophosphate protecting groups have been evaluated for 5′-hydroxyl protection of deoxyribonucleosides as carbonates and for potential use in solid-phase oligonucleotide synthesis. The spatial arrangement of selected functional groups forming an asymmetric nucleosidic 5′-O-carbonic acid ester has been designed to enable heat-induced cyclodecarbonation reactions, which would result in the release of carbon dioxide and the generation of a nucleosidic 5′-hydroxyl group. The nucleosidic 5′-O-carbonates 3-8, 10-15, and 19-21 were prepared and were isolated in yields ranging from 45 to 83%. Thermolytic deprotection of these carbonates is preferably performed in aqueous organic solvent at 90 °C under near neutral conditions. The rates of carbonate deprotection are dependent on the nucleophilicity of the functional group involved in the postulated cyclodecarbonation reaction and on solvent polarity. Deprotection kinetics increase according to the following order: 4 < 5 < 10 < 6 < 12 < 7 < 13 < 8 < 14 ? 19-21 and CC14 < dioxane < MeCN < t-BuOH < MeCN:phosphate buffer (3:1 v/v, pH 7.0) < EtOH:phosphate buffer (1:1 v/v, pH 7.0). Complete thermolytic deprotection of carbonates 7, 8, 13, and 14 is achieved within 20 min to 2 h under optimal conditions in phosphate buffer-MeCN. The 2-(2-pyridyl)amino-1-phenylethyl and 2-[N-methyl-N-(2-pyridyl)]aminoethyl groups are particularly promising for 5′-hydroxyl protection of deoxyribonucleosides as thermolytic carbonates.

Full text of DOI:10.1021/jo035089g

Th er m olytic Ca r bon a tes for P oten tia l 5′-Hyd r oxyl P r otection of
Deoxyr ibon u cleosid es
Marcin K. Chmielewski, Vicente Marcha´n, J acek Cies´lak, Andrzej Grajkowski,
Victor Livengood,† Ursula Mu¨nch,‡ Andrzej Wilk,§ and Serge L. Beaucage*
Division of Therapeutic Proteins, Center for Biologics Evaluation and Research, Food and Drug
Administration, 8800 Rockville Pike, Bethesda, Maryland 20892, and Laboratory of Bioorganic
Chemistry, NIDDK, National Institutes of Health, 9000 Rockville Pike, Bethesda, Maryland 20892.
beaucage@cber.fda.gov
Received J uly 26, 2003
Thermolytic groups structurally related to well-studied heat-sensitive phosphate/thiophosphate
protecting groups have been evaluated for 5′-hydroxyl protection of deoxyribonucleosides as
carbonates and for potential use in solid-phase oligonucleotide synthesis. The spatial arrangement
of selected functional groups forming an asymmetric nucleosidic 5′-O-carbonic acid ester has been
designed to enable heat-induced cyclodecarbonation reactions, which would result in the release of
carbon dioxide and the generation of a nucleosidic 5′-hydroxyl group. The nucleosidic 5′-O-carbonates
3-8, 10-15, and 19-21 were prepared and were isolated in yields ranging from 45 to 83%.
Thermolytic deprotection of these carbonates is preferably performed in aqueous organic solvent
at 90 °C under near neutral conditions. The rates of carbonate deprotection are dependent on the
nucleophilicity of the functional group involved in the postulated cyclodecarbonation reaction and
on solvent polarity. Deprotection kinetics increase according to the following order: 4 < 5 < 10 <
6 < 12 < 7 < 13 < 8 < 14 = 19-21 and CCl4 < dioxane < MeCN < t-BuOH < MeCN:phosphate
buffer (3:1 v/v, pH 7.0) < EtOH:phosphate buffer (1:1 v/v, pH 7.0). Complete thermolytic deprotection
of carbonates 7, 8, 13, and 14 is achieved within 20 min to 2 h under optimal conditions in phosphate
buffer-MeCN. The 2-(2-pyridyl)amino-1-phenylethyl and 2-[N-methyl-N-(2-pyridyl)]aminoethyl
groups are particularly promising for 5′-hydroxyl protection of deoxyribonucleosides as thermolytic
carbonates.
In tr od u ction
being compatible with the traditional nucleobase N-acyl
protecting groups. These carbonates include the 9-fluo-
renylmethoxycarbonyl,3 2-(phenylsulfonyl)ethoxycarbo-
nyl,4 2-(4-nitrophenyl)ethoxycarbonyl,5 2-dansylethoxy-
carbonyl,6 and 1-phenyl-2-cyanoethoxycarbonyl groups.7
These carbonates are sensitive to nonnucleophilic basic
conditions and are generally compatible with 2′-O-acetal
groups in the synthesis of oligoribonucleotides. Carbon-
Since the orthogonal protection of hydroxyl groups still
presents a formidable challenge in the preparation of
either natural or synthetic bioactive molecules, alkyl
carbonates have found wide applications in such synthe-
ses as protecting groups for alcohols.1 In the context of
nucleic acid synthesis, the isobutyloxycarbonyl group has
been used to regioselectively protect the 5′-hydroxyl of
thymidine to permit phosphorylation of the 3′-hydroxy
function.2 However, because of the strongly nucleophilic
conditions that are required for its cleavage, the isobu-
tyloxycarbonyl group is incompatible with the conven-
tional N-acyl groups used for nucleobase protection, and
has found limited applications as a hydroxyl protecting
group in oligonucleotide synthesis. Carbonates have since
been designed to provide 5′-hydroxyl protection while
(3) (a) Balgobin, N.; Chattopadhyaya, J . Nucleosides Nucleotides
1987, 6, 461-463. (b) Pathak, T.; Chattopadhyaya, J . Acta Chem.
Scand. 1985, B39, 799-806. (c) Gioeli, C.; Chattopadhyaya, J . J . Chem.
Soc., Chem. Commun. 1982, 672-674. (d) Brown, T.; Pritchard, C. E.;
Turner, G.; Salisbury, S. A. J . Chem. Soc. Chem. Commun. 1989, 891-
893. (e) Ma, Y.-X.; Sonveaux, E. Nucleosides Nucleotides 1987, 6, 491-
493. (f) Ma, Y.-X.; Sonveaux, E. Biopolymers 1989, 28, 965-973. (g)
Lehmann, C.; Xu, Y.-Z.; Christodoulou, C.; Tan, Z.-K.; Gait, M. J .
Nucleic Acids Res. 1989, 17, 2379-2390. (h) Lehmann, C.; Xu, Y.-Z.;
Christodoulou, C.; Gait, M. J .; Van Meervelt, L.; Moore, M.; Kennard,
O. Nucleosides Nucleotides 1991, 10, 1599-1614. (i) Gait, M. J .;
Lehmann, C. In Current Protocols in Nucleic Acid Chemistry; Beaucage,
S. L., Bergstrom, D. E., Glick, G. D., J ones, R. A., Eds.; J ohn Wiley &
Sons: New York, 2000; pp 2.4.1-2.4.22.
(4) J osephson, S.; Balgobin, N.; Chattopadhyaya, J . Tetrahedron
Lett. 1981, 22, 4537-4540. See also: Balgobin, N.; J osephson, S.;
Chattopadhyaya, J . B. Tetrahedron Lett. 1981, 22, 3667-3670.
(5) Schirmeister, H.; Himmelsbach, F.; Pfleiderer, W. Helv. Chim.
Acta 1993, 76, 385-401.
(6) (a) Bergmann, F.; Pfleiderer, W. Helv. Chim. Acta 1994, 77, 481-
501. (b) Bergmann, F.; Pfleiderer, W. Helv. Chim. Acta 1994, 77, 988-
998.
* To whom correspondence should be addressed at the Food and
Drug Administration. Phone: (301)-827-5162. Fax: (301)-480-3256.
On leave from the Institute of Bioorganic Chemistry, Polish
Academy of Sciences, Poznan, Poland.
† National Institutes of Health.
‡ Present address: Fluka GmbH, Industriestrasse 25, CH-9471
Buchs, Switzerland.
§ Present address: U.S. Pharmacopeia, 12601 Twinbrook Parkway,
Rockville, MD 20852.
(1) For reviews see: (a) Shaikh, A. G.; Sivaram, S. Chem. Rev. 1996,
96, 951-976. (b) Parrish, J . P.; Salvatore, R. N.; J ung, K. W.
Tetrahedron 2000, 56, 8207-8237.
(2) Ogilvie, K. K.; Letsinger, R. L. J . Org. Chem. 1967, 32, 2366.
(7) Mu¨nch, U.; Pfleiderer, W. Helv. Chim. Acta 2001, 84, 1504-1517.
10.1021/jo035089g This article not subject to U.S. Copyright. Published 2003 American Chemical Society
Published on Web 11/25/2003
J . Org. Chem. 2003, 68, 10003-10012 10003

Chmielewski et al.
ates not requiring strong basic conditions for subsequent
deprotection have also been developed for 5′-hydroxyl
protection of nucleosides. Specifically, the (p-chlorophe-
noxy)carbonyl group has been successfully used in the
solid-phase synthesis of an hexadecaoligodeoxyribonucle-
otide.8 The iterative removal of the 5′-O-(p-chlorophe-
noxy)carbonate group throughout oligonucleotide syn-
thesis is effected in less than 10 min upon treatment with
a buffered peroxyanion solution.8,9 The photosensitive
R-methyl-6-nitropiperonyloxycarbonyl10 and 2-(2-nitro-
phenyl)propyloxycarbonyl11 groups have been employed
for 5′-/3′-hydroxyl protection of deoxyribonucleoside phos-
phoramidites in the photolithographic in situ preparation
of oligonucleotides on glass surfaces. These photolabile
carbonate groups are typically cleaved under UV irradia-
tion at 365 nm. The 3′,5′-dimethoxybenzoinyloxycarbonyl
group for 5′-O-protection of deoxyribonucleoside phos-
phoramidites has similarly been tested in the photo-
lithographic synthesis of oligonucleotides on arrays.
Photochemical deprotection of this carbonate proceeded
rapidly at 310 nm in a nonpolar solvent or without
solvent.12 While the photodirected approaches to the
synthesis of oligonucleotide on microarrays exhibit at-
tractive features, these methods have the potential to
form undesirable side products or lead to unwanted
secondary photoreactions with oligonucleotides. In this
regard, the photogeneration of strong acids13 for cleaving
the terminal 5′-O-dimethoxytrityl group of growing oli-
gonucleotide chains that are assembled via standard
phosphoramidite chemistry14 has also the propensity of
being harsh to oligonucleotides and to glass surfaces.
Given our interest at improving the chemical synthesis
of oligonucleotides on arrays, our strategy is to avoid
photodeprotection reactions. We are now reporting our
findings on the development of thermolabile carbonates
for 5′-hydroxyl protection of deoxyribonucleosides, and
evaluating the suitability of these groups for iterative
deprotection reactions and for potential applications to
the synthesis of oligonucleotides on arrayable glass
surfaces.
Resu lts a n d Discu ssion
The use of deoxyribonucleoside cyclic N-acylphosphora-
midites in the P-stereodefined solid-phase synthesis of
selected oligodeoxyribonucleoside phosphorothioates led
to the formation of internucleotidic 2-[N-(2-fluoroacetyl)-
amino]-1-phenylethyl thiophosphotriester linkages.15 These
phosphotriesters are thermolabile, as they are cleanly
and efficiently converted to phosphorothioate diesters
within 1 h upon heating at 85 °C in MeCN:H2O (1:1 v/v).
It was postulated that thiophosphate deprotection pro-
ceeded through a cyclodeesterification mechanism involv-
ing the participation of the amidic carbonyl group with
concomitant formation of an oxazoline side product.15
This novel approach to phosphate/thiophosphate depro-
tection led to a search for new thermolabile phosphate/
thiophosphate protecting groups. Aside from the 2-[N-
(2-fluoroacetyl)amino]-1-phenylethyl group, the 2-(N-
acetylamino)ethyl,16 2-(N-acetyl-N-methylamino)ethyl,16
2-(N-formyl-N-methylamino)ethyl,16 2-(N,N-dimethylami-
nocarbonyloxy)ethyl,16 4-oxopentyl,16,17 3-(N-tert-butyl-
carboxamido)-1-propyl,16,17b,18 3-(N,N-dimethylcarboxa-
mido)-1-propyl,19 3-(2-pyridyl)-1-propyl,20 and several
2-benzamidoethyl groups21 have also demonstrated ther-
molytic properties under neutral aqueous conditions.
Given the relative ease with which these phosphate/
thiophosphate protecting groups are removed from oli-
gonucleotides, we decided to use structurally similar
groups for the 5′-hydroxyl protection of deoxyribonucleo-
sides. The thermolytic properties of these groups will be
evaluated as carbonates so that release of carbon dioxide
under the deprotection conditions may mimic phosphate/
thiophosphate leaving groups. Several 3′-O-acetylthymi-
dine 5′-O-carbonates were prepared to determine the
structural parameters influencing the deprotection kinet-
ics of each carbonate in the production of 3′-O-acetylthy-
midine.
P r ep a r a tion of 3′-O-Acetylth ym id in e 5′-O-Ca r -
bon a tes. The synthesis of 3′-O-acetylthymidine 5′-O-
carbonates begins with the condensation of 1,1′-carbon-
yldiimidazole (1.1 equiv) with 3′-O-acetylthymidine (1,
1.0 equiv) in anhydrous MeCN to afford the correspond-
ing 3′-O-acetylthymidine 5′-O-(1-imidazolyl)carbamate 2
(Figure 1A) in yields exceeding 90% (TLC). To this
solution is added a primary alcohol (1.5 equiv) followed
by anhydrous 1,1,3,3-tetramethylguanidine (6.0 equiv).
TLC analysis of the reaction shows that formation of the
desired 3′-O-acetylthymidine 5′-O-carbonate (3-8) does
not further progress beyond a 2 h reaction time. When
secondary alcohols are employed in the preparation of
3′-O-acetylthymidine 5′-O-carbonates, 1,1′-carbonyldi-
(8) Dellinger, D. J .; Caruthers, M. H.; Betley, J . R. U.S. Patent
6,222,030, April 24, 2001.
(9) Caruthers, M. H.; Wyrzkiewicz, T. K.; Dellinger, D. J . In
Innovation and Perspectives in Solid-Phase Synthesis; Epton, R., Ed.;
Mayflower Worldwide Ltd: Birmingham, UK, 1994; p 39.
(10) McGall, G. H.; Barone, A. D.; Diggelman, M.; Fodor, S. P. A.;
Gentalen, E.; Ngo, N. J . Am. Chem. Soc. 1997, 119, 5081-5090.
(11) (a) Giegrich, H.; Eisele-Bu¨hler, S.; Hermann, C.; Kvasyuk, E.;
Charubala, R.; Pfleiderer, W. Nucleosides Nucleotides 1998, 17, 1987-
1996. (b) Beier, M.; Hoheisel, J . D. Nucleic Acids Res. 2000, 28, e11.
(c) Beier, M.; Stephan, A.; Hoheisel, J . D. Helv. Chim. Acta 2001, 84,
2089-2095. (d) Pirrung, M. C.; Wang, L.; Montague-Smith, M. P. Org.
Lett. 2001, 3, 1105-1108. (e) Beier, M.; Hoheisel, J . D. J . Biotechnol.
2002, 94, 15-22. (f) Albert, T. J .; Norton, J .; Ott, M.; Richmond, T.;
Nuwaysir, K.; Nuwaysir, E. F.; Stengele, K.-P.; Green, R. D. Nucleic
Acids Res. 2003, 31, e35.
(15) Wilk, A.; Grajkowski, A.; Phillips, L. R.; Beaucage, S. L. J . Am.
Chem. Soc. 2000, 122, 2149-2156.
(16) Grajkowski, A.; Wilk, A.; Chmielewski, M. K.; Phillips, L. R.;
Beaucage, S. L. Org. Lett. 2001, 3, 1287-1290.
(12) (a) Pirrung, M. C.; Bradley, J .-C. J . Org. Chem. 1995, 60, 6270-
6276. (b) Pirrung, M. C.; Fallon, L.; McGall, G. J . Org. Chem. 1998,
63, 241-246.
(13) (a) Gao, X.; Yu, P.; LeProust, E.; Sonigo, L.; Pellois, J . P.; Zhang,
H. J . Am. Chem. Soc. 1998, 120, 12698-12699. (b) LeProust, E.; Pellois,
J . P.; Yu, P.; Zhang, H.; Gao, X.; Srivannavit, O.; Gulari, E.; Zhou, X.
J . Comb. Chem. 2000, 2, 349-354. (c) Serafinowski, P. J .; Garland, P.
B. J . Am. Chem. Soc. 2003, 125, 962-965. (d) Kla´n, P.; Zabadal, M.;
Heger, D. Org. Lett. 2000, 2, 1569-1571.
(14) (a) Beaucage, S. L.; Caruthers, M. H. Tetrahedron Lett. 1981,
22, 1859-1862. (b) McBride, L. J .; Caruthers, M. H. Tetrahedron Lett.
1983, 24, 245-248. (c) Adams, S. P.; Kavka, K. S.; Wykes, E. J .; Holder,
S. B.; Galuppi, G. R. J . Am. Chem. Soc. 1983, 105, 661-663. (d) Sinha,
N. D.; Biernat, J .; McManus, J .; Ko¨ster, H. Nucleic Acids Res. 1984,
12, 4539-4557.
(17) (a) Wilk, A.; Chmielewski, M. K.; Grajkowski, A.; Phillips, L.
R.; Beaucage, S. L. Tetrahedron Lett. 2001, 42, 5635-5639. (b) Wilk,
A.; Chmielewski, M. K.; Grajkowski, A.; Phillips, L. R.; Beaucage, S.
L. In Current Protocols in Nucleic Acid Chemistry; Beaucage, S. L.,
Bergstrom, D. E., Glick, G. D., J ones, R. A., Eds.; J ohn Wiley & Sons:
New York, 2002; pp 3.9.1-3.9.15.
(18) Wilk, A.; Chmielewski, M. K.; Grajkowski, A.; Phillips, L. R.;
Beaucage, S. L. J . Org. Chem. 2002, 67, 6430-6438.
(19) Wilk, A.; Beaucage, S. L. Unpublished results.
(20) Cies´lak, J .; Beaucage, S. L. Unpublished results.
(21) Guzaev, A. P.; Manoharan, M. J . Am. Chem. Soc. 2001, 123,
783-793.
10004 J . Org. Chem., Vol. 68, No. 26, 2003

Thermolytic Carbonate Protecting Groups
SCHEME 1a ,b
a
Reaction conditions: (i) MeCN:phosphate buffer, pH 7.0, (3:1
v/v), 90 °C, 3 h. bKey: Thy ) thymin-1-yl; R ) acetyl and H.
the thermolytic stability of nucleosidic 5′-O-carbonate
protecting groups, the carbonate 3 was dissolved in
MeCN:phosphate buffer, pH 7.0 (3:1 v/v), and heated to
90 ( 2 °C in a tightly stoppered screw-cap glass vial. The
thermostability of 3 was assessed by reverse phase (RP)
HPLC to serve as a reference in a series of experiments.
The 5′-O-butyl carbonate group was cleaved to the extent
of ca. 2% within 6 h, as evidenced by the presence of 1
and thymidine (Table 1). In this experiment, 3′-O-
deacetylation of 3 is more predominant than the cleavage
of the 5′-O-butyl carbonate group. Incidentally, the 3′-
O-acetyl group serves as an internal standard in estimat-
ing the stability of 5′-O-carbonates to potential cleavage
caused by the buffer and the nucleophilic functional
groups endogenous to the carbonates at elevated tem-
peratures. In this context, the thermostability of 5′-O-
(4-oxopentyl) carbonate 4 is very similar to that of 3
under identical conditions (Table 1). In contrast to the
relatively rapid thermolytic cleavage of the 4-oxopentyl
phosphate protecting group,17a the 4-oxo group of carbon-
ate 4 was not sufficiently nucleophilic to induce signifi-
cant release of carbon dioxide through a cyclodecarbon-
ation mechanism analogous to the cyclodeesterification
of heat-sensitive phosphate/thiophosphate protecting
groups. To further assess the participation of nucleophilic
groups to the putative intramolecular cyclodecarbonation
mechanism, the 5′-O-amidoethyl carbonates 5 and 6 were
selected for such an investigation. Under the conditions
studied, the amidic carbonyl group of carbonate 5 is
significantly more nucleophilic than the 4-oxo group of
carbonate 4, and results in more efficient 5′-O-decarbon-
ation than 3′-O-deacetylation (Table 1). Furthermore, the
significant decrease in thermostability of carbonate 6 is
reflective of an increased nucleophilicity of the amidic
carbonyl group, which led to a ca. 30-fold increased
production of 1 and thymidine within 3 h relative to
that produced from carbonate 4 (Table 1) within 6 h.
Given the modest 3′-O-deacetylation detected in this
experiment, it is likely that 1 is generated through the
proposed intramolecular cyclodecarbonation mechanism
(Scheme 1).
F IGURE 1. Synthesis of 3′-O-acetylthymidine 5′-O-carbon-
ates. Key: CDI ) 1,1′-carbonyldiimidazole; TMG ) 1,1,3,3-
tetramethylguanidine; Thy ) thymin-1-yl.
imidazole is reacted first with the secondary alcohol (1.0
equiv) in dry MeCN (Figure 1B).
Formation of the corresponding (1-imidazolyl)carbam-
ate intermediate 9 is near complete (TLC) within 6 h at
25 °C. Addition of solid 3′-O-acetylthymidine (1.0 equiv)
to the solution followed by dry 1,1,3,3-tetramethylguani-
dine (6.0 equiv) produced the 3′-O-acetylthymidine 5′-O-
carbonate product (10-14) within 2 h. The nucleosidic
carbonates 3-8 were purified by silica gel chromatogra-
phy and were isolated in yields ranging from 70 to 83%,
whereas carbonates 10-14 were isolated in yields vary-
ing from 45 to 73%.
Str u ctu r a l P a r a m eter s Affectin g th e Th er m osta -
bility of 3′-O-Acetylth ym id in e 5′-O-Ca r bon a tes. To
systematically address the structural features affecting
To further validate the postulated cyclodecarbonation
mechanism operating in the thermolytic deprotection of
selected 3′-O-acetylthymidine 5′-O-carbonates, the ther-
J . Org. Chem, Vol. 68, No. 26, 2003 10005

Chmielewski et al.
TABLE 1. RP -HP LC An a lysis of 3′-O-Acetylth ym id in e 5′-O-Ca r bon a tes u n d er Th er m olytic Con d ition sa
recovered
carbonate
(area %)
3′-OH
carbonate
(area %)
5′-O-carbonate
5′-OH
time
(min)
1
thymidine
(area %)
speciesb
(area %)
speciesc
(area%)
carbonate
(area %)
3
4
5
6
7
8
10
11
12
13
14
360
360
360
180
120
30
360
360
120
60
93.3
93.8
79.6
18.9
0.4
5.0
3.5
3.4
1.3
0.0
0.0
2.3
5.7
0.9
0.0
0.0
1.6
2.5
0.1
0.2
0.8
5.4
7.0
6.8
5.9
0.2
5.4
7.2
0.9
98.3
97.3
83.0
20.2
0.4
1.7
2.7
16.2
74.4
92.6
92.9
62.1
1.9
79.3
92.5
98.8
17.0
79.8
99.6
99.7
68.0
2.1
84.7
99.7
99.7
0.3
0.3
29.7
92.2
14.4
0.3
32.0
97.9
15.3
0.3
15
0.2
0.2
a
The carbonates (∼2 mg) were dissolved in MeCN:phosphate buffer (3:1 v/v, pH 7.0, 0.5 mL) and were heated to 90 ( 2 °C. RP-HPLC
analyses of the deprotection reactions were performed with a 5 µm Supelcosil LC-18S column (25 cm × 4.6 mm) under the following
conditions: starting from 0.1 M triethylammonium acetate pH 7.0, a linear gradient of 1% MeCN/min is pumped at a flow rate of 1
mL/min for 40 min and, then, held isocratically for 20 min. Retention times (RT) of the 3′-O-acetylthymidine 5′-O-carbonates and related
b
deprotection species are reported in the Supporting Information. 5′-O-Carbonate species is the sum of recovered carbonate and 3′-OH
carbonate expressed in area %. c 5′-OH species is the sum of 1 and thymidine expressed in area %.
SCHEME 2
mostability of carbonates 10-12 was determined (Table
1). Addition of a phenyl group to the carbon atom
adjacent to the carbonate function generates an electro-
philic benzylic center, which should favor a nucleophilic
amidic carbonyl attack and release of carbon dioxide in
retention time (RT) of 58 min.23 Mass spectral analysis
of RP-HPLC-purified 16 is consistent with its proposed
accordance with the proposed cyclodecarbonation mech-
anism (see Scheme 1). As a consequence of this modifica-
structure. The identity of 16 was further corroborated
tion, the thermostability of carbonate 10 is considerably
by its synthesis (Scheme 2), which was adapted from a
lower than that of carbonate 5 under identical conditions,
and resulted in a 4-fold increase in the production of 1
published method.24
The oxazoline 16 was isolated by silica gel chromatog-
raphy and was shown to have the same RP-HPLC
retention time as the oxazoline generated from the
thermolytic cyclodecarbonation of 15 when using two
different chromatographic gradients.23 In addition, mass
spectral analysis of these oxazolines revealed peaks
corresponding to the correct mass expected for 16. These
findings strongly support the cyclodecarbonation of 3′-
O-acetylthymidine 5′-O-amidoethyl carbonates under
thermolytic conditions.
Even though the thermolytic cyclodecarbonation of 12
is 85% complete within 2 h, the use of such amidoethyl
carbonates for 5′-hydroxyl protection toward oligonucle-
otide synthesis is not practical given the sluggish depro-
tection kinetics of these carbonates. Structural modifi-
cations of the amidoethyl carbonates were required to
significantly improve their deprotection rates under
thermolytic conditions. In this regard, the 2-(2-pyridyl)-
aminoethyl carbonates 7 and 8 were designed to increase
the electronic density on the nitrogen atom of the
2-pyridyl function in a manner analogous to that of the
amidic carbonyl group of carbonates 5 and 6, respectively.
Syntheses of 2-(2-pyridyl)aminoethanol and 2-[N-methyl-
N-(2-pyridyl)]aminoethanol, which are prerequisite to the
preparation of carbonates 7 and 8 (Figure 1A), were
and thymidine (Table 1). The carbonate 11 was subjected
to the same thermolytic conditions to assess whether the
creation of the benzylic center alone accounted for the
decreased thermostability of 10. Since the thermostability
of 11 is similar to that of carbonate 3 (Table 1), the
benzylic function is necessary but not sufficient to ac-
count for the lower thermostability of 10 relative to that
of 5, and further supports an operative cyclodecarbon-
ation mechanism in the thermolysis of selected 3′-O-
acetylthymidine 5′-O-amidoethyl carbonates. On the
basis of the significant difference in thermostability
observed between carbonates 5 and 6, one can predict
that the thermostability of 12 would be considerably
lower than that of 10 under the same conditions. Indeed,
the thermolytic release of 1 and thymidine from 12
proceeds to the extent of 85% within 2 h when compared
with that of 68% in the case of 10 over a 6 h deprotection
time (Table 1).
The carbonate 15 (Scheme 1) was prepared as shown
in Figure 1B to unequivocally confirm the cyclodecar-
bonation mechanism operating in the thermolytic cleav-
age of 3′-O-acetylthymidine 5′-O-amidoethyl carbonates.22
The carbonate 15 was designed to (i) produce the oxazo-
line 16 upon thermolytic cyclodecarbonation conditions,
(ii) enable UV detection of 16 at 254 nm, and (iii) provide
lipophilicity to 16 to facilitate its isolation by RP-HPLC.
As expected, RP-HPLC analysis of the thermolytic
cyclodecarbonation of 15 indicated the formation of the
oxazoline 16, which is revealed as a peak having a
(23) RP-HPLC analysis was performed with a 5 µm Supelcosil LC-
18S column (25 cm × 4.6 mm) and the following gradient pumped at
a flow rate of 1 mL/min: linear from 0% B to 50% B in 50 min, isocratic
50% B for 5 min, and linear from 50% B to 0% B in 5 min. However,
when using the following gradient at a flow rate of 1 mL/minslinear
from 0% B to 70% B in 50 min, isocratic 70% B for 5 min, and linear
from 70% B to 0% B in 5 minsthe retention time of 16 is 44 min. A:
0.1 M triethylammonium acetate, pH 7.0; B: MeCN.
(22) The amido alcohol required for the preparation of 15 was
obtained from the condensation of 4-(dimethylamino)benzoic acid with
1,1′-carbonyldiimidazole in THF followed by addition of 2-methylamino-
1-phenylethanol. Details of the synthesis can be found in the Experi-
mental Section.
(24) Vorbru¨ggen, H.; Krolikiewicz, K. Tetrahedron 1993, 49, 9353-
9372.
10006 J . Org. Chem., Vol. 68, No. 26, 2003

Thermolytic Carbonate Protecting Groups
performed as recommended in the literature25 by reacting
2-bromopyridine with 2-aminoethanol and 2-(methylami-
no)ethanol, respectively.
SCHEME 3a ,b
The thermostability of 7 and 8 was tested under
conditions identical with those used for amidoethyl
carbonates. As expected, the increased nucleophilicity of
the 2-N-alkylaminopyridyl group in carbonate 7 resulted
in the formation of 1 and thymidine to the extent of 99.6%
within 2 h on the basis of RP-HPLC analysis of the
deprotection reaction (Table 1). The deprotection kinetics
of carbonate 7 is thus considerably faster than that of
the analogous amidoethyl carbonate 5, which produced
1 and thymidine to the extent of 17% within 6 h at 90 °C
(Table 1). It is likely that the thermolytic deprotection
of 7 proceeds through a cyclodecarbonation mechanism
similar to that proposed for amidoethyl carbonates.
Indeed, RP-HPLC analysis of the thermolytic deprotec-
tion reaction revealed a peak (RT of 22 min) correspond-
ing to the expected cyclodecarbonation side product 17
(Scheme 3) on the basis of NMR data (see the Experi-
mental Section).
a
Reaction conditions: (i) MeCN:phosphate buffer, pH 7.0, (3:1
v/v), 90 °C, 2 h. bKey: Thy ) thymin-1-yl; R ) acetyl and H.
Consistent with the corresponding amidoethyl carbon-
ate, thermolytic deprotection of carbonate 8 was rapid
and produced 1 and thymidine to the extent of 99.7%
within 30 min. To further improve the deprotection
kinetics of carbonates 7 and 8, the 1-phenyl-2-(2-pyridyl)-
aminoethyl carbonate 13 and its N-methyl derivative 14
were prepared as outlined in Figure 1B. RP-HPLC
analysis of the thermolytic deprotection of carbonates 13
and 14, under the conditions described in Scheme 3,
indicated that 1 and thymidine were both produced to
the extent of 99.7% within 60 and 15 min, respectively
(Table 1). The thermolysis of 13 also led to the formation
of the cyclodecarbonation side product 18, which exhib-
TABLE 2. RP -HP LC An a lysis of 3′-O-Acetylth ym id in e
5′-O-Ca r bon a te 14 u n d er Th er m olytic Con d ition s in
Va r iou s Solven tsa
recovered
carbonate
(area %)
1
Thy
(area %)
solvent
(area %)
CCl4
65.6
56.5
10.0
7.1
ND
ND
34.4
43.5
90.0
92.9
93.4
98.6
ND
ND
ND
ND
6.6
dioxane
MeCN
t-BuOH
Ab
Bc
1.4
a
The carbonate (∼2 mg) was heated to 90 ( 2 °C in the specified
solvent (0.5 mL) for 20 min or as indicated. RP-HPLC analyses of
the deprotection reactions were performed under conditions identi-
b
cal with those described in Table 1. A: EtOH:phosphate buffer
(1:1 v/v, pH 7.0), 10 min. c B: MeCN:phosphate buffer (3:1 v/v,
pH 7.0). ND, not detected.
for the synthesis of 14 to assess the generality of the
thermolytic deprotection method.
ited a RP-HPLC retention time of 47 min due to increased
lipophilicity relative to that of 17 (RT of 22 min under
identical chromatographic conditions). NMR and mass
spectral analyses of 18 are in accordance with its
proposed structure (see the Experimental Section).
Solven t Effects on th e Th er m olytic Dep r otection
of 3′-O-Acetylth ym id in e 5′-O-Ca r bon a te 14. While
the thermolytic deprotection of 14 has been investigated
with use of phosphate buffer-MeCN, it became impera-
tive to evaluate its deprotection kinetics in other solvents.
Thus, phosphate buffer-MeCN was replaced by CCl4,
dioxane, MeCN, t-BuOH, and phosphate buffer-EtOH
in separate thermolytic experiments. The results of this
study are presented in Table 2 and indicate that the rates
of deprotection increase with solvent polarity, as the
fastest rates being observed are those obtained in phos-
phate buffer-EtOH.
Removal of the 5′-O-carbonate group from 19-21 is,
like that of 14, generally complete within 20 min in
phosphate buffer-MeCN, and within 10 min in phos-
phate buffer-EtOH, as determined by RP-HPLC analysis
of the deprotection reactions (Table 3). A more extensive
3′-O-deacetylation occurred in experiments performed in
phosphate buffer-EtOH, presumably because of the
buffer’s propensity to induce transesterification at el-
evated temperature (Table 3). Noteworthy is the ther-
molytic deprotection of carbonate 20, which produced
The N-protected-3′-O-acetyldeoxyribonucleoside 5′-O-
carbonates 19-21 were then prepared as recommended
(25) Weiner, N.; Kaye, I. A. J . Org. Chem. 1949, 14, 868-872.
J . Org. Chem, Vol. 68, No. 26, 2003 10007

Chmielewski et al.
TABLE 3. RP -HP LC An a lysis of N-P r otected 3′-O-Acetyl-2′-d eoxyr ibon u cleosid e 5′-O-Ca r bon a tes 19-21 u n d er
Th er m olytic Con d ition sa
recovered
carbonate
(area %)
N-protected
3′-OAc dN
(area %)
N-protected
dN
(area %)
time
(min)
dN
(area %)
carbonate
bufferb
19
A
A
B
B
A
A
B
B
A
A
B
B
5
10
15
20
5
10
15
20
5
2.1
ND
0.8
ND
3.7
ND
0.2
ND
2.7
ND
0.3
ND
95.3
95.2
97.3
97.3
93.3
93.8
99.1
99.4
95.2
95.1
98.1
97.6
2.6
4.8
1.9
2.7
2.8
6.1
0.7
0.6
2.2
4.9
1.6
2.4
ND
ND
ND
ND
<0.1
0.1
ND
ND
ND
ND
ND
ND
20
21
10
15
20
a
The carbonates (∼2 mg) were dissolved in buffer A or B (0.5 mL) and were heated to 90 ( 2 °C for specified periods of time. RP-HPLC
analyses were performed with a 5 µm Supelcosil LC-18S column (25 cm × 4.6 mm) under the following conditions: Starting from 0.1 M
triethylammonium acetate pH 7.0, a linear gradient of 1% MeCN/min is pumped at a flow rate of 1 mL/min for 40 min, and is held
isocratically for 5 min. Then, a linear gradient of 2% MeCN/min is set for 10 min at 1 mL/min, and finally held isocratically for 20 min.
Retention times (RT) of the 3′-O-acetyl-2′-deoxyribonucleoside 5′-O-carbonates and related deprotection species are reported in the
b
Supporting Information. A: EtOH:phosphate buffer (1:1 v/v, pH 7.0). B: MeCN:phosphate buffer (3:1 v/v, pH 7.0). ND, not detected; dN
) 2′-deoxyribonucleoside.
small amounts of 6-N-benzoyladenine (e0.1%) and
induced the loss of N6-benzoyl protection (e0.1%) only
when phosphate buffer-EtOH was used as the solvent
(Table 3).
Con clu sion
On the basis of our findings on thermolabile phosphate/
thiophosphate protecting groups we have developed novel
thermolytic carbonate groups for the 5′-hydroxyl protec-
tion of deoxyribonucleosides. Most 5′-O-carbonates are
stable at ambient temperature in anhydrous aprotic
solvents, such as MeCN, except for 5′-O-carbonates 14,
and 19-21, which show a ∼10% deprotection over 24 h.
With the exception of 5′-O-carbonates 3, 4, and 11,
thermolytic deprotection of the carbonates studied pro-
ceeds predominantly through a cyclodecarbonation mech-
anism (see Schemes 1 and 3) within 10 min to more than
6 h at 90 °C (see Tables 1 and 3). 3′-O-Deacetylation of
nucleoside 5′-O-carbonates occurring during thermolytic
deprotection indicates that hydrolysis and/or nucleophilic
attack caused by endogenous functional groups may
likely contribute to carbonate cleavage at elevated tem-
perature. The nucleoside 5′-O-carbonates 8 and 13 are
representative examples of 5′-hydroxyl protection. These
compounds are easily purified by silica gel chromatog-
raphy and yet undergo near complete thermolytic 5′-O-
deprotection in less than 60 min at 90 °C under neutral
conditions. The use of carbonates 8 and 13 in automated
oligonucleotide synthesis on microarrays is, however,
impractical for such an iterative process given the time
required for thermolytic cleavage of the 5′-O-carbonates.
The 2-(2-pyridyl)amino-1-phenylethyl and 2-[N-methyl-
N-(2-pyridyl)]aminoethyl groups may nonetheless be
useful for hydroxyl protection of nucleosides and carbo-
hydrates, and that of any natural or synthetic products
as carbonates, whenever orthogonal protection/deprotec-
tion of functional groups may be required.
Sta bility of th e 3′-O-Acetyl-2′-d eoxyr ibon u cleo-
sid e 5′-O-Ca r bon a tes 14 a n d 19-21 in An h yd r ou s
Solven ts. The purification of 14 and N-protected-3′-O-
acetyl-2′-deoxyribonucleoside 5′-O-carbonates 19-21 by
silica gel chromatography was difficult. The purified
carbonates were contaminated with variable amounts of
3′-O-acetylthymidine 1 or N-protected-3′-O-acetyl-2′-
deoxyribonucleosides that were produced during chro-
matography. Thus, the carbonates 14 and 19-21 were
not completely stable to the chromatographic conditions
used for purification and raised questions on the stability
of these carbonates in anhydrous MeCN, which is the
preferred solvent used for deoxyribonucleoside phos-
phoramidites in the synthesis of oligonucleotide on solid
supports or on planar glass surfaces. The stability of
carbonates 14 and 19-21, as 0.1 M solutions in dry
MeCN, was tested at 25 °C. RP-HPLC analysis of the
solutions revealed the presence of 1 and N-protected-3′-
O-acetyl deoxyribonucleosides contaminating the carbon-
ates to the extent of 10% after 24 h. Conversion of
carbonate 14 to 1 in MeCN was reduced considerably
when the solution was stored at 5 °C. At this tempera-
ture, the formation of 3′-O-acetylthymidine over a period
of 24 h was minimal (1.4%). The formation of 1 was also
minimized when 14 was dissolved in dioxane as a 0.1 M
solution to be kept at 25 °C. Under these conditions, RP-
HPLC analysis of the solution indicated that the forma-
tion of 3′-O-acetylthymidine reached the level of 2.5%
within 24 h. While the 3′-O-phosphoramidite derivatives
of 14 and that of 19-21 may not be as stable in MeCN
as standard 5′-O-dimethoxytrityl deoxyribonucleoside
phosphoramidites, it is, however, anticipated that these
phosphoramidite derivatives, when dissolved in dry di-
oxane, may still find applications in the synthesis of short
oligonucleotides (20-25 mers) given that most syntheses
are complete within 5 h.
The lesser stability of the 2-[N-methyl-N-(2-pyridyl)]-
amino-1-phenylethyl group for 5′-hydroxyl protection of
deoxyribonucleosides as a carbonate prompted its modi-
fication to improve its stability at 25 °C while retaining
its rapid thermolytic deprotection kinetics. Preliminary
results show that the placement of an electron-withdraw-
ing group at C-4 of the pyridyl moiety provides adequate
stability to the carbonate at ambient temperature to
10008 J . Org. Chem., Vol. 68, No. 26, 2003

Thermolytic Carbonate Protecting Groups
(()-2-[N -Me t h yl-N -(2-p yr id yl)a m in o]-1-p h e n yle t h a -
n ol. (()-2-[N-Methyl-N-(2-pyridyl)]amino-1-phenylethanol was
prepared in a manner identical with that of 1-phenyl-2-
(pyridyl)aminoethanol, and was isolated as an oil in 55% yield.
The crude oil was redistilled as recommended in the litera-
ture.27 1H NMR (300 MHz, DMSO-d6) δ 8.08 (ddd, J ) 1.0,
2.0, 5.0 Hz, 1H), 7.47 (ddd, J ) 2.0, 7.0, 8.8 Hz, 1H), 7.32 (m,
5H), 6.60 (m, 1H), 6.54 (ddd, J ) 2.0, 5.0, 7.0 Hz, 1H), 4.85
(dd, J ) 4.8, 7.8 Hz, 1H), 3.62 (ddd, J ) 4.8, 7.8, 14.1 Hz, 2H),
2.90 (s, 3H). 13C NMR (75 MHz, DMSO-d6) δ 37.5, 58.0, 70.8,
105.6, 111.0, 125.8, 126.8, 127.8, 136.9, 144.1, 147.3, 158.1.
FAB-HRMS: calcd for C14H17N2O (M + H)+ 229.1341, found
229.1346.
Gen er a l P r oced u r e for th e P r ep a r a tion of th e 3′-O-
Acetylth ym id in e 5′-O-Ca r bon a tes 3-8. Dry 3′-O-acetylthy-
midine (1, 29.6 mg, 0.10 mmol) and 1,1′-carbonyldiimidazole
(18 mg, 0.11 mmol) were placed in a flame-dried 4-mL glass
vial, which was immediately stoppered with a rubber septum.
Anhydrous MeCN (0.5 mL) was added to the solids by the use
of a syringe through the rubber septum. The solution, which
was obtained 5 min after adding MeCN, was left standing at
25 °C. Thin-layer chromatography (TLC) analysis of the
reaction indicated near complete conversion of 1 to the
corresponding 5′-O-carbonylimidazolide 2 within 3 h. Then,
the selected primary alcohol (0.15 mmol) was added to the
reaction mixture followed by 1,1,3,3-tetramethylguanidine (70
µL, 0.6 mmol). The solution was left at room temperature until
the 5′-O-carbonylimidazolide 2 was completely consumed (∼2
h) and a new product was formed as evidenced by TLC. The
reaction mixture was then applied onto one preparative TLC
plate, which was developed once with dichloromethane/
methanol (9:1 v/v) as the mobile phase. The desired UV-
absorbing band was cut out and the product was eluted from
silica gel with dichloromethane/acetonitrile (8:2 v/v). The
solvent was removed by evaporation under reduced pressure28
affording the 3′-O-acetylthymidine 5′-O-carbonates 3-8 in
yields ranging from 70 to 83%.
permit oligonucleotide synthesis on glass surfaces. Ef-
ficient conversion of the electron-withdrawing group to
an electron-donating group led to speedy thermolytic
deprotection of the carbonate protecting group at tem-
peratures much lower than 90 °C. The details on the use
of such a versatile thermolytic carbonate for 5′-hydroxyl
protection of deoxyribonucleosides will be reported else-
where in due course.
Exp er im en ta l Section
(()-2-[(4-Dim eth yla m in o)ben zoyl]a m in o-1-p h en yleth a -
n ol. 4-(Dimethylamino)benzoic acid (3.30 g, 20.0 mmol) and
1,1′-carbonyldiimidazole (3.24 g, 20.0 mmol) were dissolved in
anhydrous tetrahydrofuran (20 mL), and the magnetically
stirred solution was refluxed for 1 h. (()-2-Amino-1-phenyle-
thanol (2.70 g, 20.0 mmol) and triethylamine (2.8 mL, 20
mmol) were then added to the stirred solution, which was kept
under reflux for an additional 2.5 h. The solution was cooled
to ambient temperature and evaporated to dryness under
reduced pressure. The material was dissolved in chloroform
(100 mL) and the solution was washed with a saturated
solution of sodium bicarbonate (3 × 100 mL) followed by water
(100 mL). The organic phase was dried over anhydrous
magnesium sulfate, and then evaporated to dryness under low
pressure affording a white amorphous solid (4.50 g, 15.8 mmol,
79%) that is pure enough (TLC) to be used without further
purification.1H NMR (300 MHz, DMSO-d6:CDCl3) δ 8.10 (m,
1H), 7.72 (d, J ) 9.0 Hz, 2H), 7.38-7.19 (m, 5H), 6.65 (d, J )
9.0 Hz, 2H), 5.50 (d, J ) 4.4 Hz, 1H), 4.77 (ddd, J ) 4.4, 4.4,
8.0 Hz, 1H), 3.50 (ddd, J ) 4.4, 6.2, 13.2 Hz, 1H), 3.28 (ddd, J
) 5.2, 8.0, 13.2 Hz, 1H), 2.95 (s, 6H). 13C NMR (75 MHz,
DMSO-d6:CDCl3) δ 39.5, 47.6, 71.5, 110.5, 121.0, 125.7, 126.6,
127.7, 128.3, 143.7, 151.8, 166.5. FAB-HRMS: calcd for
C
17H21N2O2 (M + H)+ 285.1603, found 285.1610.
(()-2-(N-Acetyl)a m in o-1-p h en yleth a n ol. This compound
5′-O-(1-Bu tyl)oxyca r bon yl-3′-O-a cetylth ym id in e (3). 1H
NMR (300 MHz, CDCl3) δ 7.42 (q, J ) 1.3 Hz, 1H), 6.44 (dd,
J ) 5.6, 8.7 Hz, 1H), 5.27 (ddd, J ) 1.9, 4.3, 6.5 Hz, 1H), 4.45
(dd, J ) 2.9, 11.9 Hz, 1H), 4.38 (dd, J ) 2.7, 11.9 Hz, 1H),
4.20 (m, 3H), 2.42 (ddd, J ) 1.9, 5.6, 14.1 Hz, 1H), 2.23 (ddd,
J ) 6.5, 8.7, 14.1 Hz, 1H), 2.12 (s, 3H), 1.93 (d, J ) 1.3 Hz,
3H), 1.67 (m, 2H), 1.40 (m, 2H), 0.95 (t, J ) 7.3 Hz, 3H). 13C
NMR (75 MHz, CDCl3): δ 12.5, 13.5, 18.8, 20.8, 30.6, 37.3,
67.0, 68.5, 74.4, 82.2, 84.5, 111.7, 134.9, 150.5, 154.7, 163.5,
170.4. FAB-HRMS: calcd for C17H25N2O8 (M + H)+ 385.1611,
found 385.1613.
5′-O-(4-Oxo-1-p en t yl)oxyca r b on yl-3′-O-a cet ylt h ym i-
d in e (4). 1H NMR (300 MHz, CDCl3) δ 7.38 (q, J ) 1.3 Hz,
1H), 6.40 (dd, J ) 5.7, 8.7 Hz, 1H), 5.27 (ddd, J ) 2.1, 4.3, 6.5
Hz, 1H), 4.45 (dd, J ) 3.2, 11.9 Hz, 1H), 4.39 (dd, J ) 2.9,
11.9 Hz, 1H), 4.21 (m, 3H), 2.55 (t, J ) 7.1 Hz, 2H), 2.44 (ddd,
J ) 2.1, 5.7, 14.3 Hz, 1H), 2.24 (ddd, J ) 6.5, 8.7, 14.3 Hz,
1H), 2.15 (s, 3H), 2.11 (s, 3H), 1.96 (m, 2H), 1.93 (d, J ) 1.3
Hz, 3H). 13C NMR (75 MHz, CDCl3) δ 12.5, 20.8, 22.6, 29.8,
37.3, 39.3, 67.1, 67.8, 74.2, 82.1, 84.6, 111.6, 134.8, 150.4, 154.6,
163.5, 170.4, 207.0. FAB-HRMS: calcd for C18H25N2O9 (M +
H)+ 413.1560, found 413.1565.
5′-O-(2-N-Acetyl)a m in oeth yloxyca r bon yl-3′-O-a cetylth -
ym id in e (5). 1H NMR (300 MHz, DMSO-d6) δ 8.05 (t, J ) 5.6
Hz, 1H), 7.51 (m, 1H), 6.19 (dd, J ) 6.2, 8.4 Hz, 1H), 5.20 (ddd,
J ) 2.6, 6.7, 9.5 Hz, 1H), 4.36 (dd, J ) 4.3, 11.5 Hz, 1H), 4.31
(dd, J ) 5.5, 11.5 Hz, 1H), 4.17 (ddd, J ) 4.3, 5.5, 9.5 Hz, 1H),
4.09 (m, 2H), 3.29 (q, J ) 5.5 Hz, 2H), 2.40 (ddd, J ) 6.7, 8.4,
14.4 Hz, 1H), 2.28 (ddd, J ) 2.6, 6.2, 14.4 Hz, 1H), 2.07 (s,
3H), 1.80 (s, 3H), 1.79 (d, J ) 1.4 Hz, 3H). 13C NMR (75 MHz,
DMSO-d6) δ 12.0, 20.7, 22.4, 35.3, 37.5, 66.5, 67.1, 73.6, 80.7,
was prepared in a manner identical with that described for
the synthesis of (()-2-[(4-dimethylamino)benzoyl]amino-1-
phenylethanol. Characterization data were comparable to
those reported in the literature.26
2-[N-Meth yl-N-(2-p yr id yl)]a m in oeth a n ol. 2-Bromopyri-
dine (7.6 mL, 80 mmol) and 2-methylaminoethanol (12.8 mL,
160 mmol) were heated for 24 h in an oil bath kept at 160 °C.
The reaction mixture was then allowed to cool to ambient
temperature and chloroform (300 mL) was added. The solution
was shaken with a saturated solution of potassium carbonate
(200 mL). The organic layer was dried over anhydrous
magnesium sulfate and evaporated to dryness. The material
left was distilled under reduced pressure affording a liquid
1
(bp 112-115 °C at 0.8 Torr) (6.4 g, 46 mmol, 58%). H NMR
(300 MHz, DMSO-d6) δ 8.04 (ddd, J ) 1.0, 2.0, 4.9 Hz, 1H),
7.46 (ddd J ) 2.0, 7.0, 8.7 Hz, 1H), 6.58 (ddd J ) 1.0, 2.0, 8.7
Hz, 1H), 6.51 (ddd J ) 2.0, 4.9, 7.0 Hz, 1H), 3.56 (t, J ) 1.9
Hz, 4H), 3.01 (s, 3H). 13C NMR (75 MHz, DMSO-d6) δ 36.7,
51.7, 58.5, 105.5, 110.9, 137.0, 147.3, 158.2. FAB-HRMS: calcd
for C8H13N2O (M + H)+ 153.1028, found 153.1034.
(()-1-P h en yl-2-(2-p yr id yl)a m in oeth a n ol. (()-1-Phenyl-
2-(2-pyridyl)aminoethanol was prepared according to the
procedure of Gray et al.27 and was isolated as a solid in 60%
yield. 1H NMR (300 MHz, DMSO-d6) δ 7.96 (ddd, J ) 1.0, 2.0,
5.0 Hz, 1H), 7.35 (m, 6H), 6.52 (ddd, J ) 1.0, 2.0, 8.5 Hz, 1H),
6.47 (ddd, J ) 2.0, 5.0, 7.0 Hz, 1H), 5.65 (d, J ) 4.2 Hz, 1H),
4.75 (m, 1H), 3.51 (ddd, J ) 4.3, 6.7, 13.4 Hz, 1H), 3.26 (ddd,
J ) 5.1, 7.9, 13.4 Hz, 1H). 13C NMR (75 MHz, DMSO-d6) δ
49.2, 71.5, 108.4, 111.5, 125.9, 126.7, 127.9, 136.5, 144.2, 147.2,
158.8. FAB-HRMS: calcd for C13H15N2O (M + H)+ 215.1184,
found 215.1192.
(26) La¨ıb, T.; Ouazzani, J .; Zhu, J . Tetrahedron: Asymmetry 1998,
9, 169-178.
(27) Gray, A. P.; Heitmeier, D. E.; Spinner, E. E. J . Am. Chem. Soc.
1959, 81, 4351-4355.
(28) Given the thermolytic properties of these carbonates it is
recommended to remove the solvent without external source of heat
to prevent premature 5′-O-deprotection.
J . Org. Chem, Vol. 68, No. 26, 2003 10009

Chmielewski et al.
84.0, 109.9, 135.7, 150.4, 154.2, 163.5, 169.4, 170.0. FAB-
HRMS: calcd for C17H24N3O9 (M + H)+ 414.1513, found
414.1524.
138.4, 150.2, 159.8, 163.5, 170.1, 170.3. FAB-HRMS: calcd for
C23H28N3O9 (M + H)+ 490.1826, found 490.1828.
5′-O-(1-P h en yl)-1-bu tyloxyca r bon yl-3′-O-a cetylth ym i-
1
5′-O-[2-(N-Acet yl-N-m et h yl)]a m in oet h yloxyca r b on yl-
3′-O-a cetylth ym id in e (6). 1H NMR (300 MHz, CDCl3) δ 7.28
(q, J ) 1.3 Hz, 1H), 6.33 (dd, J ) 5.6, 8.6 Hz, 1H), 5.22 (ddd,
J ) 2.1, 4.5, 6.5 Hz, 1H), 4.39 (dd, J ) 4.0, 12.2 Hz, 1H), 4.33
(dd, J ) 3.3, 12.2 Hz, 1H), 4.25 (m, 1H), 4.22 (t, J ) 5.6 Hz,
1H), 4.21 (t, J ) 5.6 Hz, 1H), 3.62 (t, J ) 5.6 Hz, 1H), 3.57 (t,
J ) 5.6 Hz, 1H), 3.07 (s, 3H), 2.48 (ddd, J ) 2.1, 5.6, 14.2 Hz,
1H), 2.18 (ddd, J ) 6.5, 8.6, 14.2 Hz, 1H), 2.14 (s, 3H), 2.12 (s,
3H), 1.94 (d, J ) 1.3 Hz, 3H). 13C NMR (75 MHz, CDCl3) δ
12.6, 20.8, 21.7, 29.6, 37.3, 37.5, 61.3, 62.2, 63.8, 74.0, 82.1,
84.8, 111.5, 134.4, 150.1, 163.2, 170.2, 170.3. FAB-HRMS:
calcd for C18H26N3O9 (M + H)+ 428.1669, found 428.1679.
d in e (11). H NMR (300 MHz, CDCl3) δ 7.38 (q, J ) 1.3 Hz,
1H), 7.36-7.30 (m, 5H), 6.38 (dd, J ) 5.6, 8.9 Hz, 1H), 5.59
(dd, J ) 6.5, 7.7 Hz, 1H), 5.18 (ddd J ) 2.0, 4.2, 6.5 Hz, 1H),
4.41 (dd, J ) 3.2, 11.9 Hz, 1H), 4.35 (dd, J ) 2.8, 11.9 Hz,
1H), 4.20 (m, 1H), 2.38 (ddd, J ) 2.0, 5.6, 14.2 Hz, 1H), 2.13
(ddd, J ) 6.5, 8.9, 14.2 Hz, 1H), 2.09 (s, 3H), 1.96 (m, 1H),
1.94 (d, J ) 1.3 Hz, 3H), 1.80 (m, 1H), 1.34 (m, 2H), 0.92 (t, J
) 7.4 Hz, 3H). 13CNMR (75 MHz, CDCl3) δ 12.5, 13.6, 18.7,
20.8, 29.7, 37.3, 38.2, 67.0, 74.4, 81.0, 82.2, 84.5, 111.6, 126.4,
128.5, 134.9, 139.5, 150.2, 154.2, 163.3, 170.3. FAB-HRMS:
calcd for C23H29N2O8 (M + H)+ 460.1924, found 461.1908.
5′-O-[2-(N-Acetyl-N-m eth yl)]a m in o-1-p h en yleth yloxy-
ca r bon yl-3′-O-a cetylth ym id in e (12). 1H NMR (300 MHz,
CDCl3) δ 7.40-7.30 (m, 6H), 6.34 (dd, J ) 5.7, 8.8 Hz, 1H),
5.80 (dd, J ) 5.1, 8.4 Hz, 0.5H), 5.75 (dd, J ) 5.1, 8.4 Hz, 0.5H),
5.27 (ddd, J ) 2.0, 4.4, 6.7 Hz, 0.4H), 5.15 (ddd, J ) 2.0, 4.4,
6.7 Hz, 0.6H), 4.45 (ddd, J ) 2.6, 3.1, 11.8 Hz, 1H), 4.35 (ddd,
J ) 2.6, 3.1, 11.8 Hz, 1H), 4.21 (m, 1H), 3.80-3.60 (m, 2H),
2.97 (s, 1.6H), 2.95 (s, 1.4H), 2.45 (ddd, J ) 2.0, 5.7, 14.3 Hz,
0.5H), 2.38 (ddd, J ) 2.0, 5.7, 14.3 Hz, 0.5H), 2.25 (ddd, J )
6.7, 8.8, 14.3 Hz, 0.5H), 2.17 (ddd, J ) 6.7, 8.8, 14.3 Hz, 0.5H),
2.11 (s, 1.6H), 2.10 (s, 1.4H), 2.07 (s, 3H), 1.88 (d, J ) 1.3 Hz,
2H), 1.80 (d, J ) 1.3 Hz, 1H). 13C NMR (75 MHz, CDCl3) δ
12.46, 12.51, 20.8, 29.6, 37.7, 37.9, 67.1, 67.3, 74.0, 74.4, 78.5,
78.7, 82.0, 82.2, 84.6, 111.5, 111.6, 126.2, 126.3, 128.7, 128.8,
134.7, 134.8, 150.2, 150.3, 153.9, 154.0, 163.2, 163.4, 170.3,
170.4, 171.1, 171.2. FAB-HRMS: calcd for C24H30N3O9 (M +
H)+ 504.1982, found 504.1996.
5′-O-[2-(2-P yr id yl)]a m in o-1-p h en ylet h yloxyca r b on yl-
3′-O-a cetylth ym id in e (13). 1H NMR (300 MHz, CD3CN) δ
8.04 (ddd, J ) 1.0, 2.0, 5.0 Hz, 1H), 7.40 (m, 6H), 7.33 (q, J )
1.4 Hz, 1H), 6.55 (ddd, J ) 2.0, 5.0, 7.0 Hz, 1H), 6.47 (ddd, J
) 1.0, 2.0, 8.4 Hz, 1H), 6.13 (dd, J ) 6.9, 7.2 Hz, 1H), 5.80
(dd, J ) 4.3, 8.2 Hz, 1H), 5.23 (m, 1H), 4.32 (ddd, J ) 3.3, 4.6,
12.0 Hz, 2H), 4.17 (ddd, J ) 3.3, 4.6, 8.0 Hz, 1H), 3.75 (ddd, J
) 4.3, 8.2, 14.3 Hz, 1H), 3.65 (ddd, J ) 4.3, 8.2, 14.3 Hz, 1H),
2.32 (m, 2H), 2.04 (s, 3H), 1.78 (d, J ) 1.4 Hz, 3H). 13C NMR
(75 MHz, CD3CN) δ 12.5, 21.1, 37.2, 47.1, 67.9, 74.7, 79.9, 82.6,
86.2, 108.9, 111.6, 113.7, 127.3, 129.4, 129.5, 136.8, 138.1,
139.2, 148.6, 151.5, 155.2, 159.5, 164.5, 171.3. FAB-HRMS:
calcd for C26H29N4O8 (M + H)+ 525.1985, found 525.1998.
5′-O-2-[N-Met h yl-N-(2-p yr id yl)]a m in o-1-p h en ylet h yl-
oxycar bon yl-3′-O-acetylth ym idin e (14). 1H NMR (300 MHz,
CDCl3) δ 8.17 (ddd, J ) 1.0, 2.0, 4.9 Hz, 0.5H), 8.15 (ddd, J )
1.0, 2.0, 4.9 Hz, 0.5H), 7.46 (ddd, J ) 2.0, 7.1, 8.8 Hz, 0.5H),
7.43 (ddd, J ) 2.0, 7.1, 8.8 Hz, 0.5H), 6.58 (ddd, J ) 2.0, 4.9,
7.1 Hz, 0.5H), 7.35 (m, 6H), 6.56 (ddd, J ) 2.0, 4.9, 7.1 Hz,
0.5H), 6.49 (ddd, J ) 1.0, 2.0, 8.8 Hz, 0.5H), 6.47 (ddd, J )
1.0, 2.0, 8.8 Hz, 0.5H), 6.36 (dd, J ) 5.6, 9.0 Hz, 0.5H), 6.33
(dd, J ) 5.6, 9.0 Hz, 0.5H), 6.00 (dd, J ) 4.5, 7.6 Hz, 0.5H),
5.95 (dd, J ) 4.5, 7.6 Hz, 0.5H), 5.13 (ddd, J ) 1.8, 6.4, 8.0
Hz, 0.5H), 5.10 (ddd, J ) 1.8, 6.4, 8.0 Hz, 0.5H), 4.38 (dd, J )
5.1, 11.9 Hz, 1H), 4.28 (dd, J ) 2.7, 11.9 Hz, 1H), 4.17 (m,
1H), 3.98 (dd, J ) 7.6, 14.5 Hz, 1H), 3.91 (dd, J ) 4.5, 14.5
Hz, 1H), 2.95 (s, 1.5H), 2.90 (s, 1.5H), 2.39 (ddd, J ) 1.8, 5.6,
14.1 Hz, 0.5H), 2.34 (ddd, J ) 1.8, 5.6, 14.1 Hz, 0.5H), 2.10 (s,
1.5H), 2.09 (s, 1.5H), 2.07 (ddd, J ) 6.4, 9.0, 14.1 Hz, 0.5H),
2.02 (ddd, J ) 6.4, 9.0, 14.1 Hz, 0.5H), 1.87 (d, J ) 1.3 Hz,
1.5H), 1.78 (d, J ) 1.3 Hz, 1.5H). 13C NMR (75 MHz, CDCl3)
δ 12.4, 12.5, 20.8, 37.2, 37.5, 37.8, 38.5, 55.1, 55.5, 67.1, 67.2,
74.1, 74.5, 78.8, 78.9, 82.0, 82.1, 84.5, 105.5, 111.6, 112.1, 126.2,
126.4, 128.5, 128.6, 128.7 134.6, 134.7, 137.2, 137.5, 137.9,
147.8, 148.0, 150.6, 154.0, 158.0, 158.2, 163.8, 170.3. FAB-
HRMS: calcd for C27H31N4O8 (M + H)+ 539.2142, found
539.2169.
5′-O-[2-N -(2-P yr id yl)]a m in oe t h yloxyca r b on yl-3′-O-
1
a cetylth ym id in e (7). H NMR (300 MHz, DMSO-d6) δ 7.94
(m, 1H), 7.46 (m, 1H), 7.34 (ddd, J ) 2.0, 7.2, 8.4 Hz, 1H),
6.47 (m, 2H), 6.17 (dd, J ) 6.2, 8.1 Hz, 1H), 5.19 (ddd, J )
2.9, 6.8, 9.7 Hz, 1H), 4.32 (m, 2H), 4.20 (m, 2H), 4.16 (m, 1H),
3.50 (m, 2H), 2.37 (ddd, J ) 6.8, 8.1, 14.4 Hz, 1H), 2.26 (ddd,
J ) 2.9, 6.2, 14.4 Hz, 1H), 2.05 (s, 3H), 1.76 (d, J ) 1.0 Hz,
3H). 13C NMR (75 MHz, DMSO-d6) δ 11.7, 20.5, 35.4, 66.6,
66.9, 73.5, 80.7, 84.0, 108.2, 109.8, 111.7, 135.5, 136.4, 147.2,
150.1, 154.1, 158.2, 158.3, 163.3, 169.8. FAB-HRMS: calcd for
C
20H25N4O8 (M + H)+ 449.1672, found 449.1662.
5′-O-[2-N-Meth yl-N-(2-pyr idyl)]am in oeth yloxycar bon yl-
1
3′-O-a cetylth ym id in e (8). H NMR (300 MHz, DMSO-d6) δ
8.04 (ddd, J ) 1.0, 2.0, 4.9 Hz, 1H), 7.48 (ddd, J ) 2.0, 7.1, 8.6
Hz, 1H), 7.41 (m, 1H), 6.61 (ddd, J ) 1.0, 2.0, 8.6 Hz, 1H),
6.55 (ddd, J ) 2.0, 4.9, 7.1 Hz, 1H), 6.19 (dd, J ) 6.1, 8.4 Hz,
1H), 5.15 (ddd, J ) 2.7, 6.5, 9.3 Hz, 1H), 4.28 (m, 3H), 4.12
(ddd, J ) 4.8, 7.6, 9.3 Hz, 1H), 3.81 (m, 3H), 2.99 (s, 3H), 2.32
(ddd, J ) 6.5, 8.4, 14.4 Hz, 1H), 2.22 (ddd, J ) 2.7, 6.1, 14.4
Hz, 1H), 2.06 (s, 3H), 1.74 (d, J ) 1.1 Hz, 3H). 13C NMR (75
MHz, DMSO-d6) δ 12.5, 20.6, 35.3, 36.5, 47.6, 49.4, 49.8, 65.6,
67.1, 83.9, 105.5, 109.8, 111.5, 137.2, 137.8, 144.4, 147.3, 154.1,
154.6, 157.9, 169.9. FAB-HRMS: calcd for C21H27N4O8 (M +
H)+ 463.1829, found 463.1855.
Gen er a l P r oced u r e for th e P r ep a r a tion of th e 3′-O-
Acetylth ym id in e 5′-O-Ca r bon a tes 10-15. A selected sec-
ondary alcohol (0.10 mmol) and 1,1′-carbonyldiimidazole (18
mg, 0.11 mmol) were placed in a flame-dried 4-mL glass vial,
which was immediately stoppered with a rubber septum.
Anhydrous MeCN (0.5 mL) was added to the reactants with a
syringe through the rubber septum. An immediate solution
was obtained and was left standing at 25 °C. TLC analysis of
the reaction indicated near complete conversion of the starting
alcohol to the corresponding carbonylimidazolide 9 within
6 h. 3′-O-Acetylthymidine (1, 29.6 mg, 0.10 mmol) was then
added to the reaction mixture followed by 1,1,3,3-tetrameth-
ylguanidine (70 µL, 0.6 mmol). The solution was left at room
temperature until 9 was completely consumed (∼2 h) and a
new product was formed as evidenced by TLC. The reaction
mixture was then applied onto one preparative TLC plate,
which was developed once with dichloromethane/methanol (9:1
v/v) as the mobile phase. The desired UV-absorbing band was
cut out and the product was eluted from silica gel with
dichloromethane/acetonitrile (8:2 v/v). The solvent was re-
moved by evaporation under reduced pressure28 affording the
3′-O-acetylthymidine 5′-O-carbonates 10-14 in yields ranging
from 45 to 73%.
5′-O-(2-N-Acetyl)a m in o-1-p h en yleth yloxyca r bon yl-3′-
1
O-a cetylth ym id in e (10). H NMR (300 MHz, CDCl3) δ 7.40
(m, 5H), 7.28 (q, J ) 1.3 Hz, 1H), 6.32 (dd, J ) 5.7, 8.6 Hz,
1H), 5.62 (t, J ) 8.7 Hz, 1H), 5.22 (ddd, J ) 2.1, 4.3, 6.7 Hz,
1H), 4.38 (dd, J ) 4.3, 12.1 Hz, 1H), 4.33 (dd, J ) 3.4, 12.1
Hz, 1H), 4.24 (m, 1H), 3.98 (t, J ) 8.7 Hz, 1H), 3.54 (t, J ) 8.7
Hz, 1H), 2.47 (ddd, J ) 2.1, 5.7, 14.2 Hz, 1H), 2.17 (ddd, J )
6.7, 8.6, 14.2 Hz, 1H), 2.13 (s, 3H), 2.11 (s, 3H), 1.94 (d, J )
1.3 Hz, 3H). 13C NMR (75 MHz, CDCl3) δ 12.5, 20.7, 20.8, 29.6,
37.5, 48.3, 63.8, 74.1, 77.8, 82.1, 84.8, 111.4, 125.6, 128.8, 134.5,
5′-O-[2-N-(4-Dim eth yla m in o)ben zoyl]a m in o-1-p h en yl-
eth yloxycar bon yl-3′-O-acetylth ym idin e (15). 1H NMR (300
MHz, CDCl3) δ 7.62 (d, J ) 9.1 Hz, 2H), 7.41-7.34 (m, 5H),
7.30 (q, J ) 1.3 Hz, 0.3H), 7.23 (q, J ) 1.3 Hz, 0.7H), 6.64 (d,
10010 J . Org. Chem., Vol. 68, No. 26, 2003

Thermolytic Carbonate Protecting Groups
J ) 9.1 Hz, 1.2H), 6.63 (d, J ) 9.1 Hz, 0.8H), 6.35 (m, 1H),
6.30 (dd, J ) 5.5, 8.8 Hz, 1H), 5.88 (dd, J ) 4.2, 8.2 Hz, 0.3H),
5.84 (dd, J ) 4.2, 8.2 Hz, 0.7H), 5.17 (ddd, J ) 2.0, 6.5, 8.0
Hz, 1H), 4.49 (dd, J ) 3.6. 11.8 Hz, 1H), 4.31 (dd, J ) 2.9.
11.8 Hz, 1H), 4.19 (m, 1H), 3.96 (ddd, J ) 4.2, 6.5, 14.3 Hz,
1H), 3.76 (ddd, J ) 5.5, 8.2, 14.3 Hz, 1H), 3.01 (s, 6H), 2.39
(ddd, J ) 2.0, 5.5, 14.2 Hz, 0.5H), 2.35 (ddd, J ) 2.0, 5.5, 14.2
Hz, 0.5H), 2.18 (ddd, J ) 6.5, 8.8, 14.2 Hz, 0.5H), 2.10 (ddd, J
) 6.5, 8.8, 14.2 Hz, 0.5H), 2.08 (s, 2H), 2.07 (s, 1H), 1.86 (d, J
) 1.3 Hz, 2H), 1.76 (d, J ) 1.3 Hz, 1H). 13C NMR (75 MHz,
CDCl3) δ 12.4, 12.5, 20.8, 36.9, 37.1, 40.0, 44.5, 44.7, 67.0, 67.3,
74.0, 74.4, 77.2, 79.4, 79.5, 82.0, 82.1, 84.7, 111.0, 111.6, 120.6,
120.7, 126.2, 126.4, 128.3, 128.4, 128.7, 128.8, 128.9, 134.8,
136.8, 137.1, 150.2, 152.6, 154.1, 163.3, 167.3, 167.4, 170.3.
FAB-HRMS: calcd for C30H35N4O9 (M + H)+ 595.2404, found
595.2377.
d6) δ 43.7, 62.0, 112.0, 118.9, 137.8, 147.6, 154.6. FAB-
HRMS: calcd for C7H9N2 (M + H)+ 121.0766, found 121.0771.
Isola tion a n d Ch a r a cter iza tion of 3-P h en yl-2,3-d ih y-
d r oim id a zo[1,2-a ]p yr id in e (18). This compound was iso-
lated from the thermolytic deprotection of carbonate 13 by RP-
HPLC. The peak corresponding to 18 (RT of 47 min, vide supra)
was collected and processed in a manner identical with that
1
described for the isolation of 17. H NMR (300 MHz, DMSO-
d6) δ 8.36 (ddd, J ) 1.0, 2.0, 4.9 Hz, 1H), 8.11 (ddd, J ) 1.0,
2.0, 8.5 Hz, 1H), 7.86 (ddd, J ) 2.0, 7.3, 8.5 Hz, 1H), 7.46 (m,
5H), 7.15 (ddd, J ) 2.0, 4.9, 7.3 Hz, 1H), 5.77 (dd, J ) 7.7, 8.6
Hz, 1H), 4.63 (dd J ) 8.6, 10.4 Hz, 1H), 4.04 (dd, J ) 7.7, 10.4
Hz, 1H). 13C NMR (75 MHz, DMSO-d6) δ 50.7, 74.3, 112.2,
119.1, 126.2, 128.7, 128.8, 138.0, 147.7, 150.6, 153.8. FAB-
HRMS: calcd for C13H13N2 (M + H)+ 197.1079, found 197.1070.
P r ep a r a tion of th e N-P r otected 3′-O-Acetyl-2′-d eoxy-
r ibon u cleosid e 5′-O-Ca r bon a tes 19-21. These carbonates
were prepared and purified in a manner similar to that
described for the preparation of the 3′-O-acetylthymidine 5′-
O-carbonates 10-15, and were isolated in similar yields.
Gen er a l P r oced u r e for th e Th er m olytic Dep r otection
of 3′-O-Acetyl-2′-d eoxyr ibon u cleosid e 5′-O-Ca r bon a tes.
3′-O-Acetyl-2′-deoxyribonucleoside 5′-O-carbonates (∼2 mg)
were placed in a 4-mL screw-cap glass vial. The material was
dissolved by adding either MeCN:phosphate buffer pH 7.0 (3:1
v/v) or EtOH:phosphate buffer pH 7.0 (1:1 v/v) (0.5 mL) as
indicated in Tables 1-3. The glass vial containing the solution
was securely capped and then heated for a predetermined
period of time (see Tables 1-3) to 90 ( 2 °C, using a heating
block. The progress of the reaction was analyzed by RP-HPLC,
using a 5 µm Supelcosil LC-18S column (25 cm × 4.6 mm) and
a linear gradient of 1% MeCN/min, starting from 0.1 M
triethylammonium acetate, pH 7.0, at a flow rate of 1 mL/
min. The MeCN gradient was created by pumping a solution
of 40% MeCN in 0.1 M triethylammonium acetate, pH 7.0. RP-
HPLC retention times of the 5′-O-carbonates under study are
reported in the Supporting Information.
N4-Ben zoyl-5′-O-[2-N-m et h yl-N-(2-p yr id yl)]a m in o-1-
ph en yleth yloxycar bon yl-3′-O-acetyl-2′-deoxycytidin e (19).
1H NMR (300 MHz, DMSO-d6) δ 8.12-7.99 (m, 4H), 7.62 (m,
1H), 7.54-7.47 (m, 3H), 7.46 (ddd, J ) 2.0, 4.9, 8.8 Hz, 1H),
7.40-7.30 (m, 5H), 6.59 (dd, J ) 4.9, 6.8 Hz, 1H), 6.54 (dd, J
) 2.0, 8.8 Hz, 1H), 6.17 (dd, J ) 6.2, 7.4 Hz, 0.5H), 6.13 (dd,
J ) 6.2, 7.4 Hz, 0.5H), 5.87 (dd, J ) 4.6, 8.4 Hz, 0.5H), 5.84
(dd, J ) 4.6, 8.4 Hz, 0.5H), 5.16 (ddd, J ) 2.5, 6.6, 9.2 Hz,
1H), 4.29 (m, 3H), 3.94 (dd, J ) 4.6, 14.5 Hz, 1H), 3.89 (dd, J
) 8.4, 14.5 Hz, 1H), 2.93 (s, 1.5H), 2.92 (s, 1.5H), 2.49 (m, 1H),
2.29 (ddd, J ) 2.5, 7.4, 14.4 Hz, 1H), 2.06 (s, 3H). 13C NMR
(75 MHz, DMSO-d6) δ 20.6, 32.3, 37.1, 37.3, 38.1, 54.3, 57.3,
64.4, 67.1, 77.9, 78.1, 105.6, 108.2, 111.7, 113.1, 126.07, 126.1,
127.7, 128.2, 128.4, 128.5, 129.2, 129.6, 136.5, 137.0, 137.19,
137.24, 137.8, 144.7, 147.4, 147.5, 153.5, 153.6, 154.3, 157.71,
157.75, 169.8. FAB-HRMS: calcd for C33H33N5O8 (M + Cs)+
760.1383, found 760.1384.
Syn th esis a n d Ch a r a cter iza tion of 2-(4-Dim eth yla m i-
n o)p h en yl-5-p h en yl- ∆2-oxa zolin e (16). To a suspension of
4-(dimethylamino)benzoic acid (332 mg, 2.01 mmol) and
2-amino-1-phenylethanol (276 mg, 2.01 mmol) in MeCN (6 mL)
was added, under an inert atmosphere, pyridine (7 mL), Et3N
(840 µL, 6.06 mmol), and CCl4 (780 µL, 8.08 mmol). A solution
of Ph3P (1.57 g, 5.99 mmol) in MeCN:pyridine (1:1 v/v, 16 mL)
was then added to the reaction mixture, dropwise, over a
period of 5 min. The resulting clear yellow solution turned
brown and cloudy within 1 h. The suspension was left stirring
overnight and was then evaporated under reduced pressure
to a brown residue. The material was dissolved in CH2Cl2 (60
mL) and was washed with ice-cold 2 N NaOH (50 mL). The
aqueous layer was extracted with CH2Cl2 (2 × 60 mL). The
combined organic phases were dried over anhydrous sodium
sulfate and rotoevaporated to dryness. The residue was
purified by silica gel chromatography with use of a linear
gradient of ethyl acetate (2.5% to 15%) in toluene affording
N6-Ben zoyl-5′-O-[2-N-m et h yl-N-(2-p yr id yl)]a m in o-1-
p h e n yle t h yloxyca r b on yl-3′-O-a ce t yl-2′-d e oxya d e n os-
1
in e (20). H NMR (300 MHz, DMSO-d6) δ 8.74 (s, 0.5H), 8.73
(s, 0.5H), 8.64 (s, 0.5H), 8.63 (s, 0.5H), 8.11-8.05 (m, 2H), 8.03
(dd, J ) 2.0, 4.9 Hz, 1H), 7.68-7.51 (m, 3H), 7.45 (ddd, J )
2.0, 7.1, 8.8 Hz, 1H), 7.38-7.30 (m, 6H), 6.61-6.47 (m, 3H),
5.82 (dd, J ) 5.0, 7.9 Hz, 1H), 5.40 (ddd, J ) 2.7, 6.2, 8.6 Hz,
1H), 4.33 (dd, J ) 7.7, 12.0 Hz, 1H), 4.25 (dd, J ) 4.5, 12.0
Hz, 1H), 4.23 (m, 1H), 3.95-3.79 (m, 2H), 3.12 (ddd, J ) 6.2,
7.8, 14.4 Hz, 0.5H), 3.09 (ddd, J ) 6.2, 7.8, 14.4 Hz, 0.5H),
2.89 (s, 1.5H), 2.86 (s, 1.5H), 2.60 (ddd, J ) 2.7, 6.8, 14.4 Hz,
1H), 2.10 (s, 1.5H), 2.09 (s, 1.5H). 13C NMR (75 MHz, DMSO-
d6) δ 20.7, 35.3, 37.1, 37.2, 54.3, 54.4, 57.3, 64.4, 67.0. 67.1,
73.9, 74.0, 77.8, 77.9, 81.2, 81.3, 83.5, 83.8, 85.2, 105.62, 105.64,
108.2, 111.7, 113.1, 125.8, 126.1, 127.7, 128.17, 128.19, 128.3,
128.4, 129.2, 129.5, 136.5, 136.6, 137.0, 137.2, 137.7, 137.8,
144.7, 147.4, 147.5, 151.5, 151.6, 153.50, 153.54, 157.7, 169.8.
FAB-HRMS: calcd for C34H33N7O7 (M + Cs)+ 784.1496, found
784.1505.
1
16 (327 mg, 61%) as a light-brown solid (mp 106-108 °C). H
NMR (300 MHz, CDCl3) δ 7.89 (d, J ) 9.0 Hz, 2H), 7.37-7.32
(m, 5H), 6.69 (d, J ) 9.0 Hz, 2H), 5.61 (dd, J ) 7.8, 9.9 Hz,
1H), 4.44 (dd, J ) 9.9, 14.4 Hz, 1H), 3.93 (dd, J ) 7.8, 14.4
Hz, 1H), 3.03 (s, 6H). CI-MS: calcd for C17H17N2O (M - H)-
265, found 265.
N2-Isobu tyr yl-5′-O-[2-N-m eth yl-N-(2-p yr id yl)]a m in o-1-
p h e n y le t h y lo x y c a r b o n y l-3′-O -a c e t y l-2′-d e o x y g u a -
1
Isola tion a n d Ch a r a cter iza tion of 2,3-Dih yd r oim i-
d a zo[1,2-a ]p yr id in e (17). This compound was isolated from
the thermal deprotection of carbonate 7 with RP-HPLC. The
peak exhibiting a RT of 22 min under the chromatographic
conditions used for analysis of the deprotection reaction (vide
supra) was collected and the eluate was evaporated to dryness
under reduced pressure. The residue, generated from multiple
RP-HPLC runs, was coevaporated several times with aqueous
acetonitrile to remove residual triethylammonium acetate, and
was then left exposed to high vacuum overnight prior to
n osin e (21). H NMR (300 MHz, DMSO-d6) δ 8.17 (s, 0.5H),
8.16 (s, 0.5H), 8.09 (dd, J ) 2.0, 5.0 Hz, 0.5H), 8.07 (dd, J )
2.0, 5.0 Hz, 0.5H), 7.48 (ddd, J ) 2.0, 6.9, 8.8 Hz, 1H), 7.38-
7.27 (m, 5H), 6.58 (dd, J ) 2.0, 6.9 Hz, 1H), 6.55 (ddd, J )
2.0, 5.0, 8.8 Hz, 1H), 6.23 (dd, J ) 6.4, 8.5 Hz, 0.5 H), 6.20
(dd, J ) 6.4, 8.5 Hz, 0.5 H), 5.82 (dd, J ) 4.5, 8.1 Hz, 0.5H),
5.80 (dd, J ) 4.5, 8.1 Hz, 0.5H), 5.26 (m, 1H), 4.23 (m, 3H),
3.91 (ddd, J ) 4.5, 8.1, 14.6 Hz, 1H), 3.85 (ddd, J ) 4.5, 8.1,
14.6 Hz, 1H), 2.92 (s, 1.5H), 2.89 (s, 1.5 H), 2.85 (m, 1H), 2.74
(qt, J ) 6.7 Hz, 1H), 2.50 (ddd, J ) 2.1, 6.2, 14.5 Hz, 1H), 2.07
(s, 3H), 1.11 (d, J ) 6.7 Hz, 3H), 1.10 (d, J ) 6.7 Hz, 3H). 13C
NMR (75 MHz, DMSO-d6) δ 20.7, 32.3, 36.6, 37.1, 37.3, 38.1,
54.3, 54.4, 57.3, 64.4, 77.9, 78.0, 85.0, 105.6, 105.7, 108.3, 111.7,
113.1, 126.1, 127.7, 128.2, 128.4, 128.5, 129.2, 129.6, 136.5,
1
characterization. H NMR (300 MHz, DMSO-d6) δ 8.36 (ddd,
J ) 1.0, 2.0, 4.9 Hz, 1H), 8.07 (ddd, J ) 1.0, 2.0, 8.6 Hz, 1H),
7.83 (ddd, J ) 2.0, 7.2, 8.6 Hz, 1H), 7.13 (ddd, J ) 2.0, 4.9, 7.2
Hz, 1H), 4.45 (m, 2H), 4.16 (m, 2H). 13C NMR (75 MHz, DMSO-
J . Org. Chem, Vol. 68, No. 26, 2003 10011

Chmielewski et al.
137.0, 137.21, 137.24, 137.7, 137.8, 144.7, 147.4, 147.5, 153.51,
153.54, 154.4, 157.7, 169.8, 169.9. FAB-HRMS: calcd for
C
Development Agreement with the Center for Biologics
Evaluation and Research.
31H35N7O8 (M + Cs)+ 766.1601, found 766.1566.
Su p por tin g In for m a tion Ava ila ble: Materials and meth-
ods, RP-HPLC retention times of 3′-O-acetyl-2′-deoxyribo-
nucleoside 5′-O-carbonates and that of related deprotection
species, and 1H NMR spectra of (()-2-[(4-dimethylamino)-
benzoyl]amino-1-phenylethanol, 2-[N-methyl-N-(2-pyridyl)]-
aminoethanol, (()-1-phenyl-2-(2-pyridyl)aminoethanol, (()-2-
[N-methyl-N-(2-pyridyl)amino]-1-phenylethanol, 3-8, 10-15,
and 17-21. This material is available free of charge via the
Internet at http://pubs.acs.org.
Ack n ow led gm en t. This research was supported in
part by an appointment to the Postgraduate Research
Participation Program at the Center for Biologics Evalu-
ation and Research administered by the Oak Ridge
Institute for Science and Education through an inter-
agency agreement between the U.S. Department of
Energy and the U.S. Food and Drug Administration. We
thank Agilent Technologies for generous equipment and
financial support through a Cooperative Research and
J O035089G
10012 J . Org. Chem., Vol. 68, No. 26, 2003