Synthesis of novel exocyclic amino nucleosides by parallel solid-phase combinatorial strategy

Article abstract of DOI:10.1016/S0040-4020(03)00211-4

A versatile parallel solid-phase combinatorial strategy was developed for the synthesis of large nucleoside libraries. Twelve libraries L1-12 of 1152 novel exocyclic triazinylamino nucleosides and one library L13 of 82 new substituted clitocine derivatives were synthesized in high quality as natural product mimic nucleosides on the semi-automated synthesizer. The polystyrene MMT-Cl resin was selected and utilized. The key intermediate resins 5 and 9 loaded with the corresponding scaffolds were prepared and validated with various amines before parallel synthesis. After a variety of amino building blocks were validated, 56 primary amines in 12 groups (building block set A) and 24 secondary amines in 3 groups (building block set B) were selected and utilized to combinatorialize the first and the second reactive sites on scaffold 5 for the synthesis of libraries L1-12. Eighty-two amines (building block set C) were utilized for the synthesis of clitocine library L13. Thirteen libraries of 1234 novel exocyclic amino nucleosides were all analyzed and characterized by high throughput LC-MS. 81.3-100% of the library members in 13 libraries show more than 60% purity, and 65.7-92.7% of the library members in these libraries show 80-100% purity. The strategy can be widely used for the synthesis of other diverse nucleoside libraries.

Full text of DOI:10.1016/S0040-4020(03)00211-4

Tetrahedron 59 (2003) 2297–2307
Synthesis of novel exocyclic amino nucleosides by parallel
solid-phase combinatorial strategy
Chamakura V. Varaprasad, Qazi Habib, David Y. Li, Jingfan Huang, Jeffrey W. Abt, Frank Rong,
*
Zhi Hong and Haoyun An
Ribapharm, Inc., 3300 Hyland Avenue, Costa Mesa, CA 92626, USA
Received 5 November 2002; revised 6 February 2003; accepted 6 February 2003
Abstract—A versatile parallel solid-phase combinatorial strategy was developed for the synthesis of large nucleoside libraries. Twelve
libraries L1-12 of 1152 novel exocyclic triazinylamino nucleosides and one library L13 of 82 new substituted clitocine derivatives were
synthesized in high quality as natural product mimic nucleosides on the semi-automated synthesizer. The polystyrene MMT-Cl resin was
selected and utilized. The key intermediate resins 5 and 9 loaded with the corresponding scaffolds were prepared and validated with various
amines before parallel synthesis. After a variety of amino building blocks were validated, 56 primary amines in 12 groups (building block set
A) and 24 secondary amines in 3 groups (building block set B) were selected and utilized to combinatorialize the first and the second reactive
sites on scaffold 5 for the synthesis of libraries L1-12. Eighty-two amines (building block set C) were utilized for the synthesis of clitocine
library L13. Thirteen libraries of 1234 novel exocyclic amino nucleosides were all analyzed and characterized by high throughput LC-MS.
81.3–100% of the library members in 13 libraries show more than 60% purity, and 65.7–92.7% of the library members in these libraries
show 80–100% purity. The strategy can be widely used for the synthesis of other diverse nucleoside libraries. q 2003 Elsevier Science Ltd.
All rights reserved.
1. Introduction
nucleoside chemistry for small molecule drug discovery
yet although di/tri-7 and oligonucleotides8,9 were synthe-
sized on solid support, and the modified solid supports have
been used as acylating agent to acylate nucleoside
derivatives.10 There are tremendous difficulties to synthe-
size a large number of nucleoside analogues in a short
period of time utilizing solid-phase combinatorial strategies
to explore wide biological activities: (a) most applicable
linkers do not meet the general requirements for the solid-
phase synthesis of nucleoside libraries: stable enough under
the required reaction conditions during synthesis and labile
enough to be cleaved easily from solid support without
affecting nucleoside products; (b) limited positions on
nucleoside can be attached on solid support; (c) limited
types of reactions can be utilized and limited number of sites
can be combinatorialized on the nucleoside skeletons, which
prevent the generation of large diverse nucleoside libraries;
(d) it is more difficult to get high quality and purity
nucleoside libraries without purification compared to other
small molecule libraries.
After decades of drug discovery and development effort
from hundreds of academic laboratories and pharmaceutical
companies, approximately 50 nucleoside- and nucleotide-
related drugs have been pushed to the market and over 80
nucleosides/tides are in the (pre) clinical studies for various
therapeutic indications including antiviral, anticancer, and
others.1 However, some drugs such as AZT, ddI, 3TC etc
rapidly develop drug resistance and show mitochondrial,
bone-marrow, and other toxicity. Therefore, it is essential to
discover novel nucleoside drugs for critical medical needs.
A variety of different modified nucleoside derivatives2 were
synthesized by the classical approaches in solution, which is
expensive and time-consuming. Combinatorial synthesis of
large diverse libraries and high-throughput screening
technologies have recently emerged as powerful drug
discovery paradigms. Various solid-phase,3 solution-
phase,4 liquid-phase,5 and third-phase6 combinatorial
approaches have been successfully utilized for the genera-
tion of different oligomeric and small molecule libraries for
a wide range of biological screenings. Unfortunately, these
powerful technologies, especially parallel solid-phase
combinatorial strategies, have not been applied to the
Although hundreds of modified purine- and pyrimidine-
based nucleoside derivatives have been reported,1,11,12 very
little work related to exocyclic nucleosides was reported.
Clitocine, 6-amino-5-nitro-4-(b-D-ribofuranosylamino)-
pyrimidine, the compound L13 when R4¼R5¼H (Fig. 1),
is a natural exocyclic amino nucleoside isolated from the
mushroom Clitocybe inversa,13 and it was then synthesized
from 4,6-diamino-5-nitropyrimidine.14 Clitocine exhibits
Keywords: exocyclic amino nucleoside; triazinyl; clitocine mimic; parallel
solid-phase combinatorial; library; automated synthesis.
*
Corresponding author. Tel.: þ1-714-427-6236x2241; fax: þ1-714-641-
7222; e-mail: haoyunan@ribapharm.com
0040–4020/03/$ - see front matter q 2003 Elsevier Science Ltd. All rights reserved.
doi:10.1016/S0040-4020(03)00211-4

2298
C. V. Varaprasad et al. / Tetrahedron 59 (2003) 2297–2307
the high-throughput LC-MS. High quality libraries (plates)
were obtained with the successful rates18 ranging from
75–100%. 65.7–92.7% of the library members in 13 plates
show 80–100% purity. Totally, 1234 novel exocyclic amino
nucleosides were synthesized for the first time utilizing
the parallel solid-phase combinatorial approach. Average
20–30 mg of sample was obtained for each library member
and ready for a wide range of biological studies.
2. Results and discussion
Figure 1. Novel exocyclic amino nucleoside libraries L1-13.
In order to widely explore the biological activity of the
natural product mimic exocyclic amino nucleosides, we
decided to generate large diverse novel nucleoside libraries
by the parallel solid-phase combinatorial approach on the
semi-automated synthesizer. As described above, there are
various difficulties to be overcome, and a lot of new area
needs to be explored. Choosing solid support with a suitable
linker became our first challenge even though over a
hundred different types of solid supports with various
linkers are available and have been utilized for the synthesis
of oligomeric and other small molecule libraries.19 The
carbamate resin has been successfully used for the synthesis
of five carbocyclic L-nucleosides20 and eight unsaturated
carbocyclic adenosine analogues21 based on well-
established reaction conditions in solution. The non-
glycosidic bond of these carbocyclic nucleosides is much
more stable than glycosidic bond of ribofuranosyl nucleo-
sides, which contributed to the synthetic possibility of these
carbocyclic nucleosides on solid support although the
method has not been used for the generation of large
diverse libraries yet. 6-Chloropurine ribose was reacted with
diamine-modified trityl resins providing several 6-amino-
alkyl adenosine derivatives.22 However, the isolation of
dimers from the reaction mixture indicated the cleavage of
diamines from the resins during the reaction even in the
presence of triethylamine. Based on the commonly used
p0rotecting groups at 50-position and the reactivity of
5 -hydroxyl group, we have studied the applicability of
carboxylic acid, trityl chloride, dimethoxytrityl chloride,
monomethoxytrityl chloride (MMT-Cl), THP, Wang, and
o0ther resins to attach the ribose onto resins through
5 -hydroxyl group of the ribofuranosyl nucleosides. Most
of these resins are either too labile during the synthesis or
not labile enough to cleave the product from solid support
after synthesis (data not shown). Trityl chloride and
monomethoxytrityl chloride (MMT-Cl) resins are applic-
able for the synthesis of nucleoside libraries. However, the
cleavage of products from trityl resin requires stronger acid
that caused partial deglycosylation of the base from
nucleoside sugar ring. MMT-Cl resin was more applicable
than trityl chloride resin, therefore, it was utilized for the
synthesis of libraries L1-13 (Fig. 1). The products were
cleaved from the resin by 1.5–2% trifluoroacetic acid (TFA)
in dichloromethane.
potent cytostatic effects against several leukemia cell lines
through the inhibition of adenosine kinase. It also shows
strong insecticidal activity against the pink bollworm
Pectinophora gossypiella.13,14 Other exocyclic amino
pyrimido[5.4-d]pyrimidine15 and pyrimidine nucleosides16
with different bases were also reported. It will be valuable to
explore clitocine mimic libraries with substituted triazine as
the heterocyclic base that has one more nitrogen in the
heterocyclic ring compared to that of clitocine. Although
triazine derivatives without sugar moieties have been
synthesized to study their corticotropin-releasing factor1
receptor antagonists and other biological effects,17 triazine-
based exocyclic nucleosides, the clitocine mimic, have not
been studied up to today. To efficiently explore the potential
antitumor, antiviral, and other biological activities of
clitocine mimic exocyclic nucleoside derivatives, we
designed and synthesized novel triazine- and pyrimidine-
based exocyclic nucleoside libraries.
In this paper, we describe the exploration, development,
synthesis, and characterization of novel disubstituted
exocyclic triazinylamino nucleoside libraries L1-12 and
substituted clitocine library L13 of 1234 novel nucleoside
analogues by the parallel solid-phase combinatorial
approach that, for the first time, was applied to the synthesis
of nucleoside libraries. Polystyrene monomethoxytrityl
chloride resin was selected from various available solid
supports with different linkers. Resin 5, the key intermediate
resin for the synthesis of triazine-based exocyclic amino
nucleoside libraries L1-12, was developed and validated by
different amino building blocks in different solvent systems
without prior synthesis in solution. Fifty-six selected
primary amines in 12 groups were utilized for the first
combinatorialization step, and 24 selected secondary
amines in 3 groups were used for the second nucleophilic
substitution step after a variety of amino building blocks
with different sizes, lipophilicity, diversity and reactivity
were validated. Twelve groups of 56 primary amines
(building block set A) and twenty-four secondary amines
(building block set B) were selected for the combina-
torialization of the first and the second reactive sites of the
triazinyl nucleoside scaffold. Twelve plates of libraries
L1-12 containing 1152 (12£96) triazine-based exocyclic
amino nucleoside derivatives were synthesized in high
quality by the parallel array combinatorial strategy on the
ACT Vanguard semi-automated synthesizer. Eighty-two
substituted clitocine derivatives in library L13 were
synthesized by our newly developed reaction conditions
and strategy from 82 selected amines as building blocks (set
C). All library members were analyzed and characterized by
b-D-Ribofuranosyl-1-azide (2)23 was synthesized by the
reaction of 10-acetyl-20,30,50-tribenzoyl ribofuranose (1) with
sodium azide,24,25 followed by the deprotection of the
resultant triprotected azido intermediate with sodium
cyanide (Scheme 1). The polystyrene monomethoxytrityl
chloride (MMT-Cl) resin was treated with 2 equiv. of

C. V. Varaprasad et al. / Tetrahedron 59 (2003) 2297–2307
2299
Scheme 1. Development of exocyclic triazinylamino nucleoside scaffold.
azidoribose 2 in pyridine in the presence of 4-N,N-di-
methylaminopyridine (DMAP). Compound 2 was attached
o0nto MMT resin through 50-position because the
5 -hydroxyl 0 group is more reactive than the hydroxyl
groups at 2 - and 30-positions. The loading efficiency of
85% was estimated based on the consumption of compound
2. The resultant resin was further treated with t-butyl-
dimethylsilyl chloride (TBDMS-Cl) to give the loaded resin
3 protected by TBDMS protecting groups. Resin 3 showed
strong peak at 2109 cm21 (n, N3) as illustrated by its IR
spectrum b comparing to that of polystyrene MMT-Cl resin
itself (IR spectrum a, Figure 2). A portion of resin 3 was
cleaved by 2% of TFA in dichloromethane and the resultant
20,30-O-bis(t-butyldimethylsilyl)-b-D-ribofuranosyl-1-azide
was confirmed by its MS and NMR spectrometric analyses.
This result further confirmed the structure of resin 3.
Trimethylphosphine was used to reduce the azido group of
resin 3 resulting in the amino resin 4. The structure of resin 4
was also confirmed by the disappearance of the azido peak
at 2109 cm21 (see IR spectrum c, Figure 2). The three
chloro groups of cyanuric chloride (2,4,6-trichloro-1,3,5-
triazine) have different electrophilic reactivity. We have
efficiently utilized the characteristic property of cyanuric
chloride for the synthesis of exocyclic triazine amino
nucleoside libraries. Resin 4 was reacted with cyanuric
chloride at 0–58C for 1 h. The hydrochloride generated
from the reaction was neutralized by adding N,N-diiso-
propylethylamine in the reaction mixture to avoid the acidic
cleavage of the resulted scaffold from MMT resin. A portion
of resin 5 was treated with TFA, and the resultant N-[(4,6-
dichloro)triazin-2-yl]-1-amino-20,30-O-bis(t-butyldimethyl-
silyl)-b-D-ribofuranose was confirmed by NMR spectro-
metry, which also verified the structure of resin 5.
The reaction conditions from resin 5 to products 7, building
blocks, and solvent systems have to be validated before the
synthesis of final libraries on synthesizer. Eight sets of
amino building blocks (two in each set, Table 1) were used
for the validation of reaction conditions. Amines H2NR1
were reacted with resin 5 at 08C to room temperature for 2 h
to ensure the completion of the first nucleophilic substi-
tution but not to react with the second reactive site on the
triazine moiety. Amines HNR2R3 were then reacted with the
last electrophilic site of the resulted resin at 808C for 6 h to
ensure the completion of the reaction and the quality of the
final libraries. Resins 6a–h were treated with tetrabutyl-
ammonium fluoride (TBAF) to remove the TBDMS
protecting group and then treated with 2% TFA to provide
final products 7. During the validation, we noticed that the
excess tetrabutylammonium salt resulted from the deprotec-
tion step was very difficult to be removed completely by the
usual washing procedures. Therefore, the resin was treated

2300
C. V. Varaprasad et al. / Tetrahedron 59 (2003) 2297–2307
Figure 2. FT-IR spectral comparison of different resins.
with DMF–AcOH–H2O (8:1:1) for 10 min, followed by
usual washing procedures to completely remove the excess
amount of tetrabutylammonium salt. The early stage
validation results also indicated that the high quality
products were obtained when the primary amines were
used for the first reaction site of resin 5, and the secondary
amines were reacted with the last reactive site. Therefore,
appropriate combinations of different building blocks were
used to validate acetonitrile, N,N-dimethylformamide
(DMF), and 1-methyl-2-pyrrolidinone (NMP) solvent
systems for the two substitution steps (Table 1). The high
quality of product 7 (Table 1) verified the protocol and the
quality of resin 5. The reaction of resin 5 with isopropyl-
amine and then piperazine did not provide the desired
product. The similar results were also observed for other
nucleoside scaffolds when piperazine or other diamine-
types of compounds were used as nucleophilic building
blocks although the reason is still not clear (data not shown).
Therefore, the diamine-types of building blocks were
excluded for the final library synthesis. The solutions of
building blocks needed to be freshly prepared and used
within 2 days when DMF was used as the solvent in order to
reach the desired quality. N,N-Dimethylamino substitution
was observed during previous validation when DMF
solutions of building blocks were kept for more than 2
days. NMP solutions can be used for weeks and still
provided more pure products than other two solvents.
Therefore, NMP was selected as the solvent for the two
substitution steps on the synthesizer.
Based on the size, liphophilicity, diversity, and expected
reactivity of amino building blocks, 64 primary amines were

C. V. Varaprasad et al. / Tetrahedron 59 (2003) 2297–2307
Table 1. Validation of solvents on loaded resin 5
2301
Entry
H2N–R1
LC-MS purity of 7 in MeCN (%)
LC-MS purity of 7 in NMP (%)
LC-MS purity of 7 in DMF (%)
a
b
c
d
0
83
70
90
93
85
79
85
95
67
79
99
e
f
94
86
82
88
89
90
g
h
87
0
82
0
89
0
selected for the first substitution step by fixing one building
block for the second step, and 32 secondary amines were
selected for the second substitution step by fixing one
building block for the first step (Scheme 2). Thus, 96 wells
of reactions were validated in one reaction block on the
ACT Vanguard semi-automated synthesizer. Based on the
LC-MS results of the validated plate (data not shown), we
selected 56 primary amines out of sixty-four as the building
block set A (H2NR1) (Fig. 3) to react with resin 5 first, and
24 secondary amines out of 32 as the building block set B
(HNR2R3) for the last substitution step (Fig. 4). Fifty-six
primary amines in building block set A were arranged into
12 groups. Twenty-four secondary amines in building block
set B were arranged into three groups, and each group
contained eight amines. Figure 5 shows the arrangement of
the building blocks and 96-well plates. Therefore, an array
combination of building block set A as columns with
building blocks B1-24 as rows generated twelve plates
representing 12 libraries L1-12, which represent 1152
(24£48 or 12£96) novel exocyclic triazinylamino nucleo-
sides. Building blocks B1-8, B9-16, and B17-24 were used
as rows for plates L1-4, L5-8, and L9-12, respectively.
A1-12, A13-24, A25-36, and A37-48 were used as columns
for plates L1-4. A49-50, A2-8, A56, and A10-11 were used
as columns for plates L5 and L9. A12, A14-21, and A23-25
were used as columns for plates L6 and L10. A27-30,
A51-52, A31-33, A53, and A35-36 were used as columns
for plates L7 and L11. A37-48 were used as columns for
plate L8. A37-38, A40-42, A44-48, and A54-55 were used
as columns for plate L12. The 96 compounds in each plate
were enumerated by Afferent TeamWorks software to
generate molecular weights, structural, and other related
information for the desired library members and the key
impurities in each reaction vessel. Afferent TeamWorks
transferred the data to Label Automador for automated
sample bar code labeling and weighing. Libraries L1-12
were synthesized in a parallel fashion utilizing the
optimized procedures and conditions as well as using
NMP as solvent. The first set of building blocks (set A)
substituted the first chloro group on the scaffold at room
temperature for 2 h, and the second set of building blocks
(set B) were reacted with the second reactive site at 808C for
5 h to ensure the completion of the nucleophilic substitution
Scheme 2. Parallel synthesis of exocyclic amino trazinyl nucleoside
libraries L1-12.

2302
C. V. Varaprasad et al. / Tetrahedron 59 (2003) 2297–2307
Figure 3. Building block set A: 56 primary amines.
because the last reactive site is less reactive than the first one
after the substitution of the first reactive site. The washing
processes were automatically conducted and controlled by
the pre-designed program. After deprotection by TBAF, the
final library members were cleaved from resins by 1.5% of
TFA in dichloromethane. The synthesized plates were dried
using Savant SpeedVac and then analyzed by high-
throughput LC-MS. The Afferent TeamWorks software
generated compound ID numbers and molecular weights
which were input into the sample list for LC-MS data
acquisition. The LC-MS data were acquired using
MassLynx software, and the LC-MS data were processed
using OpenLynx software. The LC-MS data and the
compound purity summary were input into Afferent
Figure 5. Building block and plate arrangement for the synthesis of
libraries L1-L12.
Figure 4. Building block set B: 24 secondary amines.

C. V. Varaprasad et al. / Tetrahedron 59 (2003) 2297–2307
2303
Table 2. Quality of libraries L1-12
Library no
Percentage of products 80–100%
Percentage of products .60%
L1
L2
L3
L4
L5
L6
L7
L8
70.8
65.7
74.0
76.0
84.3
92.7
81.2
81.3
76.0
76.0
80.3
75.0
77.8
92.7
88.5
96.9
90.6
95.8
97.9
91.7
100
86.5
81.3
85.4
82.3
90.8
L9
L10
L11
L12
Overall
96 compounds for each library.
TeamWorks as a record for the database. Table 2 shows the
quality of 12 libraries L1-12. The LC-MS results indicated
that 65.7–92.7% of the library members shown 80–100%
purity. 81.3–100% of the library members have more than
60% purity. The percentage of the library members with
more than 60% LC-MS purity is defined as the successful
rate of a library/plate. Those library members passing 60%
purity have been selected by the programmed software and
automatically registered to database for biological screen-
ing. For example, library L6 has high successful rate of
97.9%. Therefore, 94 compounds with over 60% purity out
of 96 wells of reactions were registered. Eighty-nine
compounds out of 96, 92.7% of the library members,
shown 80–100% purity, which represented very high
quality for eight-step reactions on solid support and the
breakthrough in nucleoside chemistry. Some representative
LC chromatograms and mass spectra (see Supporting
Information) further verified the high quality of these
libraries. An average of 20–30 mg of sample for each
library member was obtained for a wide range of biological
screenings. Therefore, 1152 novel exocyclic triazine amino
nucleoside derivatives were successfully synthesized by the
array parallel solid-phase combinatorial strategy on the
semi-automated synthesizer. All library members were
obtained in sufficient quantity, excellent yields, and high
purity as indicated by LC-MS analysis.
This developed strategy for the synthesis of exocyclic
triazinyl amino nucleoside libraries L1-12 was utilized
further for the synthesis of substituted clitocine library
L13 (Scheme 3). Resin 4 loaded with 1-amino-2,3-O-bis-
(t-butyldimethylsilyl)ribose was treated with 4,6-dichloro-
5-nitropyrimidine in NMP for 4 h using N,N-diisopropyl-
ethylamine as a base providing resin 9. The clitocine
scaffold on resin 9 was validated under the similar reaction
Table 3. Validation of clitocine scaffold on resin 9
Entry
LC-MS purity of 11
a
89%
98%
b
c
d
e
100%
No
65%
f
80%
98%
g
h
i
79%
99%
j
No
Scheme 3. Synthesis of clitocine mimic exocyclic amino nucleoside
library.

2304
C. V. Varaprasad et al. / Tetrahedron 59 (2003) 2297–2307
conditions as described above utilizing eleven representa-
tive amino building blocks (Table 3). The results indicated
that primary, secondary, and cyclic amines as well as aniline
all worked efficiently with LC-MS purity of 80–100%.
Hydroxyl group did not effect the reactions and the quality
of the products (see entries c and e). The nucleophilicity of
the amino groups played an important role for the success.
The electron-withdrawing group COOEt (entry d) signifi-
cantly reduced the nucleophilicity of the amino group
adjacent to the ester group, therefore, it did not produce the
desired product. Although aniline worked very well and
provided desired product with 99% purity (entry i), 2-tri-
fluoromethylaniline (entry j) did not generate products
because the steric effect and electron-withdrawing property
of trifluoromethyl significantly reduced the nucleophilicity
of the amino group. After validating the scaffold and
building blocks, we selected 82 amines HNR4R5 as the
building block set C (see Supporting Information) consider-
ing diversity, reactivity, spatial, and other properties. Resin
9 was reacted with the eighty-two selected building blocks
at room temperature for 16 h in a parallel fashion utilizing
the similar procedures described above (Scheme 3). After
removing the TBDMS protecting groups by TBAF, the
products L13 were cleaved from resins by 1.5% TFA in
dichloromethane. 72% compounds out of eighty-two shown
the LC-MS purity of 80–100%.
on solid support were obtained on a Perkin–Elmer FT-IR
spectrometer. The libraries were enumerated by Afferent
TeamWorks 3.0, labeled and weighted by Label Automador,
and synthesized on the ACT Vanguard semi-automated
synthesizer. Libraries were analyzed on a LC-MS system.
The LC-MS system consists of Waters 2790 HPLC, Waters
996 photodiode array (PDA) detector, and Micromass/
Waters ZQ mass spectrometer. Luna C18 column from
Phenomenex was used for compound separation. The mass
spectra at m/z 100–1000 were acquired using electrospray
ionization with both positive and negative ion detections.
UV spectra were recorded at 200–400 nm by the PDA, and
the compound purity was monitored based on the UV
absorbency at 220 nm. The LC-MS operation was con-
trolled by MassLynx software, and the LC-MS data were
processed by OpenLynx software. Polystyrene mono-
methoxytrityl chloride resin was purchased from Novabio-
chem. Other starting materials, building blocks, and
reagents were purchased from Aldrich and other companies,
and used directly. b-D-Ribofuranosyl-1-azide (2) was
synthesized based on the reported procedures.23–25
3.1.1. 2,3-O-Bis(t-butyldimethylsily)-5-O-(monometh-
oxytrityl-polystyrene resin)-b-D-ribofuranosyl-1-azide
(3). To a suspension of polystyrene, monomethoxytrityl
chloride resin (1.0 g, 1.73 mmol/g) in 4 mL of pyridine was
added a solution of b-D-ribofuranosyl-1-azide (2) (0.60 g,
3.46 mmol) in 4 mL of pyridine, followed by the addition of
4-N,N-dimethylaminopyridine (DMAP) (0.122 g, 1.0 mmol).
The reaction mixture was shaken well at room temperature
for 48 h. The resin was filtered and washed sequentially with
CH2Cl2 (3£25 mL), a mixture of CH2Cl2–MeOH–N,N-di-
isopropylethylamine (8.5:1:0.5, 2£20 mL). The resulted
resin was then dried under vacuum over KOH for 16 h. The
loading efficiency was 85% (1.46 mmol/g) calculated based
on the starting material 2 recovered and the specified
loading capacity of the resin. FT-IR (KBr) 2107.3 cm21 (N3
group). A small portion (50 mg) of the resin was treated
with a solution of TFA in CH2Cl2 (1.5%) for 60 s, filtered,
and washed with CH2Cl2 (2£2 mL). The combined filtrate
was concentrated providing the starting azido compound 2,
which was confirmed by 1H NMR (CD3OD) d 5.19 (s, 1H),
4.04 (dt, 1H, J¼6.6, 4.8 Hz), 3.97 (m, 1H), 3.81 (d, 1H, J¼
5.0 Hz), 3.76 (dd, 1H, J¼12.0, 3.0 Hz), 3.56 (dd, 1H, J¼
12.0, 5.7 Hz); MS (ESI) m/z 176 (M)þ.
In conclusion, we have for the first time developed a
versatile parallel solid-phase combinatorial strategy for the
synthesis of large nucleoside libraries. The polystyrene
monomethoxytrityl chloride resin and this solid-phase
synthetic process can be widely utilized for the rapid
synthesis of other large diverse nucleoside libraries to
accelerate drug discovery process. The key intermediate
resins 5 and 9 were validated and FT-IR spectrometry was
used to monitor the reactions on solid support. Twelve
plates of 1152 novel exocyclic trazinylamino nucleoside
derivatives L1-12, the natural product clitocine mimic
nucleosides, were successfully synthesized utilizing this
developed strategy. Eighty-two of substituted clitocine
derivatives L13 were synthesized by the same strategy.
1234 novel exocyclic amino nucleosides were all analyzed
and characterized in a parallel fashion by Waters HT
LC-MS. 76–100% successful rates and 65.7–92.7% of
library members having 80–100% purity demonstrated
the high quality of these new libraries and the advantage
of the parallel solid-phase combinatorial strategy. There-
fore, more than 1108 high quality, novel exocyclic nucleo-
sides out of 1234 reaction wells synthesized were
registered for biological screening with sufficient
quantity (20–30 mg). High throughput screening of these
novel nucleosides in searching antiviral and anticancer
active agents is in progress and will be reported in due
course.
To a suspension of the resulted resin (1.2 g), 5-O-(mono-
methoxytrityl-polystyrene resin)-b-D-ribofuranosyl-1-
azide, in 10 mL of anhydrous DMF were added excess
amount of t-butyldimethylsilyl chloride (TBDMS-Cl)
(1.29 g, 8.65 mmol) and imidazole (1.17 g, 17.3 mmol).
The reaction mixture was shaken well at room temperature
for 16 h. The resin was filtered and washed sequentially
with DMF (3£10 mL), MeOH (3£10 mL), and CH2Cl2
(3£10 mL). The resin was then dried under vacuum over
KOH for 16 h. FT-IR (KBr) 2109.9 cm21 (N3 group). A
small portion (0.10 g) of resin 3 was treated with a solution
of TFA in CH2Cl2 (1.5%) for 60 s, filtered, and washed with
CH2Cl2 (2£2 mL). The combined filtrate was concentrated
providing 20,30-O-bis(t-butyldimethylsilyl)-b-D-ribofurano-
syl-1-azide (30 mg), which was confirmed by 1H NMR
(CDCl3) d 5.13 (d, 1H, J¼1.2 Hz), 4.17 (dd, 1H, J¼7.2,
4.0 Hz), 4.05 (m, 1H), 3.88 (dd, 1H, J¼12.3, 2.7 Hz), 3.83
3. Experimental
3.1. General methods
NMR spectra were recorded at 300 MHz and the chemical
shifts are expressed relative to the added tetramethylsilane.
Fourier-transform Infrared (FT-IR) spectra of the samples

C. V. Varaprasad et al. / Tetrahedron 59 (2003) 2297–2307
2305
(dd, 1H, J¼3.9, 1.5 Hz), 3.60 (dd, 1H, J¼12.3, 3.6 Hz), 0.90
(s, 9H), 0.89 (s, 9H), 0.10 (s, 3H), 0.098 (s, 3H), 0.091 (s,
3H), 0.08 (s, 3H).
for 10 min to remove excess amount of tetrabutylammo-
nium salt. The resin was filtered and washed sequentially
with DMF–H2O mixture (9:1, 3£10 mL), MeOH
(3£10 mL), and CH2Cl2 (3£20 mL). After dried over
KOH under vacuum for 16 h, the resultant resin (50 mg)
was suspended in 1.5 mL of TFA solution in CH2Cl2 (1.5%)
and shaken well at room temperature for 60 s. The resin was
filtered and further washed with MeOH (2£1 mL). The
combined filtrate was concentrated under high vacuum to
provide compound 7.
3.1.2. 1-(R,S)-Amino-2,3-O-bis-(t-butyldimethylsilyl)-50-
O-(monomethoxytrityl-polystyrene resin)-b-D-ribofura-
nose (4). To a suspension of the azido resin 3 (1.1 g) in a
mixture of THF and water (9:1, 7.5 mL) was added a
solution of trimethylphosphine (PMe3) in THF (2.5 mL,
1.0 M). The reaction mixture was shaken well at room
temperature for 6 h. The resultant resin was filtered and then
washed sequentially with THF–H2O (1:1, 3£10 mL),
MeOH (3£10 mL), and CH2Cl2 (3£10 mL). The resin was
then dried over KOH under vacuum for 16 h. FT-IR (KBr)
2109.9 cm21 (N3 group) peak disappeared (see spectrum c,
Figure 2).
0
3.1.6. N 1 -[(5-Nitro-6-chloro)pyrimidin-4-yl]-10-(R,S)-
amino-20,30-O-bis-(t-butyldimethylsilyl)-50-O-(mono-
methoxytrityl-polystyrene resin)-b-D-ribofuranose (9).
Resin 4 (1.0 g) was suspended in a solution of N,N-di-
isopropylethylamine in NMP (8 mL, 20% v/v) and treated
with 4,6-dichloro-5-nitropyrimidine (1.0 g, 5.18 mmol).
The reaction mixture was shaken well at room temperature
for 4 h. The resultant brown suspension was filtered, and the
resin was washed with NMP (3£25 mL) and CH2Cl2
(3£25 mL). After dried under vacuum over P2O5 for 16 h,
resin 9 was obtained.
0
3.1.3. N 1 -[(4,6-Dichloro)triazin-2-yl]-10-(R,S)-amino-
20,30-O-bis-(t-butyldimethylsilyl)-50-O-(monomethoxytri-
tyl-polystyrene resin)-b-D-ribofuranose (5). The amino
resin 4 (1.0 g) was suspended in a solution of N,N-di-
isopropylethylamine in CH2Cl2 (5 mL, 20% v/v) and cooled
to 0–58C. A solution of cyanuric chloride in CH2Cl2 (5 mL,
1.0 M) was added. The resultant suspension was shaken
well at room temperature for 1 h and filtered using a sintered
funnel. The resin was washed with CH2Cl2 (3£25 mL) and
dried over KOH under vacuum for 16 h. A small portion
(0.10 g) of resin 5 was treated with 1.5 mL of TFA solution
in CH2Cl2 (1.5%) for 60 s, filtered, and washed with CH2Cl2
0
3.1.7. N 1 -[(5-Nitro-N-isopropyl)pyrimidin-4-yl]-10-
(R,S)-amino-b-D-ribofuranose (11b). To a suspension of
resin 9 (50 mg) in a solution of N,N-diisopropylethylamine
in NMP (0.75 mL, 20% v/v) was added a solution of
isopropylamine in NMP (0.75 mL, 1 M). The suspension
was shaken well at room temperature for 16 h. The resin was
filtered and washed sequentially with MeOH (3£10 mL),
CH2Cl2 (3£10 mL), an NMP–H2O mixture (3:1, 3£10 mL),
MeOH (3£10 mL), and CH2Cl2 (3£10 mL). The resulted
resin 10 (50 mg) was suspended in a solution of tetra-
butylammonium fluoride (1.5 mL, 1 M) in THF and shaken
well at room temperature for 16 h. The resin was filtered
and washed sequentially with THF (3£10 mL), MeOH
(3£10 mL), an NMP–H2O mixture (3:1, 3£10 mL), MeOH
(3£10 mL), and CH2Cl2 (3£10 mL). The resin was treated
with a mixture of DMF–H2O–AcOH (8:1:1, 2 mL) and
shaken well for 15 min to remove the excess amount of
tetrabutylammonium salt. The resin was filtered and washed
with MeOH (3£10 mL), an NMP–H2O mixture (3:1,
3£10 mL), MeOH (3£10 mL), and CH2Cl2 (3£10 mL).
The resulted clean resin was treated with a solution of TFA
in CH2Cl2 (1.5 mL, 1.5%) for 60 s. The resin was filtered
and washed with MeOH (2£2.5 mL). The combined filtrate
(2£2 mL). The combined filtrate was concentrated pro-
0
viding 30 0mg of N 1 -[(4,6-dichloro)triazin-2-yl]-10-(R,S)-
amino-20,3 -O-bis-(t-butyldimethylsilyl)-b-D-ribofuranose,
which was confirmed by 1H NMR (CDCl3) d 7.32 (d, 0.6H,
J¼8.7 Hz, ex. D2O), 6.87 (d, 0.4H, J¼6.6 Hz, ex. D2O),
5.84 (dd, 0.55H, J¼5.4, 8.7 Hz), 5.52 (d, 0.45H, J¼6.9 Hz),
4.26 (m, 1H), 4.09 (m, 1H), 3.93–3.60 (m, 3H), 3.83 (dd,
1H, J¼1.5, 3.9 Hz), 0.94, 0.92, 0.89, 0.87 (4 s, 18H), 0.10–
0.06 (ms, 12H).
0
3.1.4. N 1 -[(4-N-Alkylamino-6-N,N-dialkylamino)tri-
azin-2-yl]-10-(R,S)-amino-20,30-O-bis-(t-butyl-dimethyl-
silyl)-50-O-(monomethoxytrityl-polystyrene resin)-b-D-
ribofuranose (6). To a suspension of resin 5 (50 mg) in a
solution of N,N-diisopropylethylemine in NMP (0.75 mL,
20% v/v) was added a solution of a primary amine in NMP
(0.75 mL, 1 M). The reaction mixture was shaken well at
room temperature for 2 h. The resin was then washed
sequentially with NMP (3£10 mL), MeOH (3£10 mL), and
CH2Cl2 (3£10 mL). A suspension of the resultant resin
(0.05 g) in a solution of N,N-diisopropylethylamine in NMP
(0.75 mL, 20% v/v) was treated with a solution of a
secondary amine in NMP (0.75 mL, 1 M). The reaction
mixture was shaken well at 808C for 6 h. The resin was then
washed with NMP (3£10 mL) and CH2Cl2 (3£10 mL). The
fully protected and substituted resin was obtained after dried
over KOH under vacuum for 16 h.
1
was concentrated to provide product 11b (15–20 mg). H
NMR (CD3OD) d 8.09, 8.07 (ss, 1H), 6.10 (d, 0.55H,
J¼4.8 Hz), 5.92 (d, 0.45H, J¼3 Hz), 4.24 (m, 1H), 4.15 (m,
0.55H), 4.09 (m, 0.45H), 3.98 (m, 1H), 3.68 (m, 2H), 1.30
(dd, 6H, J¼2.7, 6.6 Hz).
3.2. General procedures for the synthesis of exocyclic
nucleoside libraries L1-12
3.2.1. First amine substitution. Approximately 70 mg of
starting dichloro triazine resin 5 was dispensed in each of
the 96 reaction wells using a dispensing spatula and funnel.
A pre-cooled solution (2208C) of N,N-diisopropylethyl-
amine in NMP (20% v/v, 0.75 mL) was added to each well
with a repetitive pipette, followed by the addition of 12
primary amines (building block set A) (0.75 mL, 1.0 M in
NMP, pre-cooled to 2208C) in the respective columns. The
0
3.1.5. N 1 -[(4-N-Alkyl-amino-6-N,N-dialkylamino)tri-
azin-2-yl]-10-(R,S)-amino-b-D-ribofuranose (7). Resin 6
(50 mg) was suspended in a solution of tetrabutylammo-
nium fluoride in THF (1.5 mL, 1 M) and shaken well at
room temperature for 16 h. The resin was filtered and
treated with a DMF–AcOH–H2O mixture (8:1:1, 1.5 mL)

2306
C. V. Varaprasad et al. / Tetrahedron 59 (2003) 2297–2307
reaction block was covered and shaken at room temperature
for 2 h. The reaction mixtures were filtered, and the resins
were washed with DMF (£3), alternatively with MeOH and
CH2Cl2 (£3), and finally with CH2Cl2 (£2), and then dried
under nitrogen.
weighed vials in the 96-well format. The resins were washed
with 1 mL of MeOH and filtered into the corresponding 96
wells. Toluene (0.25 mL) was added to each well, and the
resultant solutions were concentrated under vacuum using
Savant SpeedVac vacuum evaporator to dryness.
3.2.2. Second amine substitution. To the resulting mono-
chloro triazine resins were added a solution of N,N-di-
isipropylethylamine in NMP (20%, 0.75 mL), followed by
the addition of eight secondary amines (building block set
B) in NMP (0.75 mL, 1.0 M) in the respective rows. The
reaction block was covered and shaken at 808C for 5 h. The
reaction mixtures were filtered, and the resins were washed
with DMF (£3), alternatively with MeOH and CH2Cl2 (£3),
and finally with CH2Cl2 (£2), and then dried under nitrogen.
Acknowledgements
The authors thank Maja Stojiljkovic, Vesna Stoisavljevic,
Roberto B. Amador, and Paul Diaz for helpful discussion.
References
3.2.3. Deprotection. A solution of tetrabutylammonium
fluoride (TBAF) in THF (1.5 mL, 1.0 M) was added to the
resin in each well. The reaction block was shaken at room
temperature overnight. After emptied and washed with
DMF, the resins were washed with 10% acetic acid in DMF
three times to remove the excess amount of tetrabutyl-
ammonium salt, then washed as usual with DMF (£3),
alternatively with MeOH and CH2Cl2 (£3), and finally with
CH2Cl2 (£2), and then dried under nitrogen.
1. The estimation was made based on authors’ personal data
collection as well as the following websites: (a) http://www.
fda.gov/cder/; (b) http://www.actis.org/; (c) http://merck.com/
pubs/mmanual/; (d) http://www.pharmalicensing.com/.
2. (a) In The Handbook of Nucleoside Synthesis; Vorbruggen,
H., Ruh-Pohlenz, C., Eds.; Wiley: New York, 2001.
(b) Chemistry of Nucleosides and Nucleotides; Townsend,
L. B., Ed.; Plenum: New York, 1988; Vol. 1. (c) Chemistry of
Nucleosides and Nucleotides; Townsend, L. B., Ed.; Plenum:
New York, 1991; Vol. 2. (d) Chemistry of Nucleosides and
Nucleotides; Townsend, L. B., Ed.; Plenum: New York, 1994;
Vol. 3.
3.2.4. Cleavage from the resin. A solution of trifluoro-
acetic acid in CH2Cl2 (1.5%) was added to the resin in each
well, and then the reaction block was shaken for 2 min. The
96 reaction wells were filtered into the 96 pre-labeled and
pre-weighed vials in the 96-well format. The resins were
washed with 1 mL of MeOH and filtered into the corre-
sponding 96 wells. Toluene (0.25 mL) was added to each well,
and the resultant solutions were concentrated under vacuum
using Savant SpeedVac vacuum evaporator to dryness.
3. (a) See the following special issues: Chem. Rev. 1997, 92(2);
Chem. Rev. 1996, 96(1); Acc. Chem. Res. 1996, 29(3).
(b) Franzen, R. G. J. Comb. Chem. 2000, 2, 195. (c) Andres,
C. J.; Denhart, D. J.; Deshpande, M. S.; Gillman, K. W. Comb.
Chem. High Throughput Screening 1999, 2, 191. (d) Kundu,
B.; Khare, S. K.; Rastogi, S. K. Prog. Drug Res. 1999, 53, 90.
(e) Sofia, M. J. Med. Chem. Res. 1998, 8, 362. (f) Houghten,
R. A.; Pinilla, C.; Appel, J. R.; Blondelle, S. E.; Dooley, C. T.;
Eichler, J.; Nefzi, A.; Ostresh, J. M. J. Med. Chem. 1999, 42,
3743. (g) Rinnova, M.; Lebl, M. Coll. Czech. Chem. Commun.
1996, 61, 171. (h) Stankova, M.; Lebl, M. Mol. Div. 1996, 2,
75.
3.2.5. Synthesis of clitocine library L13. Substitution.
Approximately 70 mg of starting resin 9 was dispensed in
each of the 96 reaction wells using a dispensing spatula and
funnel. A solution of N,N-diisopropylethylamine in NMP
(20%, 0.75 mL) was added to each well, followed by the
addition of 82 amines (building block set C) in NMP
(0.75 mL, 1.0 M). The reaction block was covered and
shaken at room temperature for 16 h. The reaction mixtures
were filtered, and the resins were washed with DMF (£3),
alternatively with MeOH and CH2Cl2 (£3), and finally with
CH2Cl2 (£2), and then dried under nitrogen.
4. (a) An, H.; Cook, P. D. Chem. Rev. 2000, 100, 3311. (b) An,
H.; Cummins, L. L.; Griffey, R. H.; Bharadwaj, R.; Haly, B. D.;
Fraser, A. S.; Wilson-Lingardo, L.; Risen, L. M.; Wyatt, J. R.;
Cook, P. D. J. Am. Chem. Soc. 1997, 119, 3696. (c) Barvian,
M. R. Chemtracts—Org. Chem. 1998, 11, 639. (d) Merritt,
A. T. Comb. Chem. High Throughput Screening 1998, 1, 57.
5. (a) Gravert, D. J.; Janda, K. D. Chem. Rev. 1997, 97, 489.
(b) Sieber, F.; Wentworth, Jr. P.; Toker, J. D.; Wentworth,
A. D.; Metz, W. A.; Reed, N. N.; Janda, K. D. J. Org. Chem.
1999, 64, 5188. (c) Reed, N. N.; Janda, K. D. Org. Lett. 2000,
2, 1311.
3.2.6. Deprotection. A solution of tetrabutylammonium
fluoride in THF (1.5 mL, 1.0 M) was added to the resin in
each well, and the reaction block was shaken at room
temperature overnight. The resins were filtered and washed
with DMF. A mixture of DMF–AcOH–H2O (7:2:1, 1.5 mL)
was added to the resins, which were shaken at room
temperature for 1 h to remove excess amount of tetrabutyl-
ammonium salt. The resins were filtered and washed with
DMF (£3), alternatively with MeOH and CH2Cl2 (£3), and
finally with CH2Cl2 (£2), and then dried under nitrogen.
6. (a) Nakamura, H.; Linclau, B.; Curran, D. P. J. Am. Chem. Soc.
2001, 123, 10119. (b) Luo, Z. Y.; Zhang, Q. S.; Oderaotoshi,
Y.; Curran, D. P. Science 2001, 291, 1766. (c) Studer, A.;
Hadida, S.; Ferritto, R.; Kim, S. Y.; Jeger, P.; Wipf, P.; Curran,
D. P. Science 1997, 275, 823. (d) Studer, A.; Jeger, P.; Wipf,
P.; Curran, D. P. J. Org. Chem. 1997, 62, 2917.
7. (a) Zhou, W.-Q.; Roland, A.; Jin, Y.; Iyer, R. P. Tetrahedron
Lett. 2000, 41, 441. (b) Jin, Y.; Chen, X.-L.; Cote, M.-E.;
Roland, A.; Korba, B.; Mounir, S.; Iyer, R. P. Bioorg. Med.
Chem. Lett. 2001, 11, 2057. (c) Jin, Y.; Roland, A.; Zhou,
W.-Q.; Fauchon, M.; Lyaku, J.; Iyer, R. P. Bioorg. Med. Chem.
Lett. 2000, 10, 1921.
3.2.7. Cleavage from the resin. A solution of trifluoro-
acetic acid in CH2Cl2 (1.5%) was added to the resin in each
well, and the reaction block was shaken for 2 min. The 96
reaction wells were filtered into the 96 pre-labeled and pre-

C. V. Varaprasad et al. / Tetrahedron 59 (2003) 2297–2307
2307
8. (a) Beaucage, S. L.; Iyer, R. P. Tetrahedron 1992, 48,
2223–2311. (b) Beaucage, S. L. Methods Mol. Biol. (Totowa,
N. J.) 1993, 20(Protocols for Oligonucleotides and Analogs),
33.
16. Townsend, L. B.; Robins, R. K. J. Am. Chem. Soc. 1963, 85,
242.
17. (a) Whitten, J. P.; Xie, Y. F.; Erickson, P. E.; Webb, T. R.; De
Souza, E. B.; Grigoriadis, D. E.; McCarthy, J. R. J. Med.
Chem. 1996, 39, 4354. (b) Gustafson, G. R.; Baldino, C. M.;
O’Donnell, M.-M. E.; Sheldon, A.; Tarsa, R. J.; Veni, C. J.;
Coffen, D. L. Tetrahedron 1998, 54, 4501. (c) Hajduk, P. J.;
Dinges, J.; Schkeryantz, J. M.; Janowick, D.; Kaminski, M.;
Tufano, M.; Augeri, D. J.; Petros, A.; Nienaber, V.; Zhong, P.;
Hammond, R.; Coen, M.; Beutel, B.; Katz, L.; Fesik, S. W.
J. Med. Chem. 1999, 42, 3852.
9. (a) Atkinson, T.; Smith, M. In Oligonucleotide Synthesis, A
Practical Approach; Gait, M. J., Ed.; IRL, 1990; pp 35–81.
(b) Sproat, B.; Gait, M. J. In Oligonucleotide Synthesis, A
Practical Approach; Gait, M. J., Ed.; IRL, 1990; pp 83–115.
10. (a) Link, A.; van Calenbergh, S.; Herdewijn, P. Tetrahedron
Lett. 1998, 39, 5175. (b) van Calenbergh, S.; Link, A.;
Fujikawa, S.; de Ligt, R. A. F.; Vanheusden, V.; Golisade, A.;
Blaton, N. M.; Razenski, J.; Ijzerman, A. P.; Herdewijn, P.
J. Med. Chem. 2002, 45, 1845.
18. ‘Successful rate’ for a plate of 96 compounds was defined as
the percentage of the library members with more than 60%
LC-MS purity comparing with all 96 compounds in the plate.
Therefore, it represents the quality of a plate synthesized.
Library members with more than 60% LC-MS purity are
registered to database for the biological screening to reduce
false-positive and false-negative results. The higher of the
successful rate is, the higher quality of the plate has.
11. (a) Ferrero, M.; Gotor, V. Chem. Rev. 2000, 100, 4319.
(b) Ichikawa, E.; Kato, K. Curr. Med. Chem. 2001, 8, 385.
(c) Pathak, T. Chem. Rev. 2002, 102, 1623.
12. (a) Miles, R. W.; Samano, V.; Robins, M. J. J. Am. Chem. Soc.
1995, 117, 5951. (b) De Clercq, E.; Balzarini, J.; Madej, D.;
Hansske, F.; Robins, M. J. J. Med. Chem. 1987, 30, 481.
(c) Riley, T. A.; Larson, S. B.; Avery, T. L.; Finch, R. A.;
Robins, R. K. J. Med. Chem. 1990, 33, 572. (d) Riley, T. A.;
Larson, S. B.; Avery, T. L.; Finch, R. A.; Robins, R. K. J. Med.
Chem. 1990, 33, 572. (e) Girgis, N. S.; Micheal, M. A.; Smee,
D. F.; Alaghamandan, H. A.; Robins, R. K. J. Med. Chem.
1990, 33, 2750. (f) Harada, H.; Asano, O.; Hoshino, Y.;
Yoshikawa, S.; Matsukura, M.; Kabasawa, Y.; Niijima, J.;
Kotake, Y.; Watanabe, N.; Kawata, T.; Inoue, T.; Horizoe, T.;
Yasuda, N.; Minami, H.; Nagata, K.; Murakami, M.; Nagaoka,
J.; Kobayashi, S.; Tanaka, I.; Abe, S. J. Med. Chem. 2001, 44,
170.
19. (a) Yan, B.; Nguyen, N.; Liu, L.; Holland, G.; Raju, B.
J. Comb. Chem. 2000, 2, 66. (b) Booth, S.; Hermkens, P. H. H.;
Ottenheijm, H. C. J.; Rees, D. C. Tetrahedron 1998, 54, 15385.
(c) Hermkens, P. H. H.; Ottenheijm, H. C. J.; Rees, D. C.
Tetrahedron 1997, 53, 5643. (d) Hermkens, P. H. H.;
Ottenheijm, H. C. J.; Rees, D. C. Tetrahedron 1996, 52, 4527.
20. Choo, H.; Chong, Y.; Chu, C. K. Org. Lett. 2001, 3, 1471.
21. Crimmins, M. T.; Zuercher, W. J. Org. Lett. 2000, 2, 1065.
22. Bressi, J. C.; Choe, J.; Hough, M. T.; Buckner, F. S.; Van
Voorhis, W. C.; Verlinde, C. L. M. J.; Hol, W. G. J.; Gelb,
M. H. J. Med. Chem. 2000, 43, 4135.
13. Kubo, I.; Kim, M.; Wood, W. F.; Naoki, H. Tetrahedron Lett.
1986, 27, 4277.
23. Stimac, A.; Kobe, J. Carbohydr. Res. 2000, 324, 149.
24. Stimac, A.; Kobe, J. Carbohydr. Res. 1992, 232, 359.
25. Logue, M. W.; Han, B. H. Carbohydr. Res. 1983, 121, 287.
14. (a) Moss, R. J.; Petrie, C. R.; Meyer, R. B., Jr.; Nord, L. D.;
Willis, R. C.; Smith, R. A.; Larson, S. B.; Kini, G. D.; Robins,
R. K. J. Med. Chem. 1988, 31, 786. (b) Larson, S. B.; Moss,
R. J.; Kini, G. D.; Robins, R. K. Acta Crystallogr. 1988, C44,
1076. (c) Kamikawa, T.; Fujie, S.; Yamagiwa, Y.; Kim, M.;
Kawaguchi, H. J. Chem. Soc., Chem. Commun. 1988, 195.
15. (a) Berman, H. M.; Rousseau, R. J.; Mancuso, R. W.;
Kreishman, G. P.; Robins, R. K. Tetrahedron Lett. 1973,
3099. (b) Ghose, A. K.; Viswanadhan, V. N.; Sanghvi, Y. S.;
Nord, L. D.; Willis, R. C.; Revankar, G. R.; Robins, R. K.
Proc. Natl Acad. Sci. USA 1989, 86, 8242.
Further Reading
Building block set C, LC-MS profiles of 10 representative
library members, structural and quality data of the
representative plate L8 (Table 2) as well as NMR and MS
spectral data of 12 representative library members (Table 3)
(18 pages)