Improving an antitrypanosomal lead applying nucleophilic substitution on a safety catch linker

Article abstract of DOI:10.1016/S0968-0896(01)00253-X

In a joint effort with various laboratories we have been aiming at the structure-based design of glycolysis inhibitors as anti-trypanosomal drugs. 2′-Deoxy-2′-(3-methoxybenzamido)-N⁶-(1-naphtylmethyl)adenosine (1a) was thus revealed as a promis

Full text of DOI:10.1016/S0968-0896(01)00253-X

Bioorganic & Medicinal Chemistry 10 (2002) 159–165
Improving an Antitrypanosomal Lead Applying Nucleophilic
Substitution on a Safety Catch Linker
Abolfasl Golisade,a Claudia Herforth,a Ludo Quirijnen,b Louis Maesb and
Andreas Linka,*
aInstitut fur Pharmazie, Abteilung fur Pharmazeutische Chemie, Universitat Hamburg, Bundesstrasse 45,
D-20146 Hamburg, Germany
¨
¨
¨
bTibotec Group N.V., Intercity Business Park, Mechelen Noord, Zone L, Generaal de Wittelaan 11B 3,
B-2800 Mechelen, Belgium
Dedicated to Prof. Dr. J. Thiemon the occasion of his 60th birthday
Received 26 March 2001; accepted 10 July 2001
Abstract—In a joint effort with various laboratories we have been aiming at the structure-based design of glycolysis inhibitors as
anti-trypanosomal drugs. 20-Deoxy-20-(3-methoxybenzamido)-N6-(1-naphtylmethyl)adenosine (1a) was thus revealed as a promising
lead structure for the development of selective agents against protozoan parasites. Here we describe the polymer-assisted synthesis
of novel amido derivatives of the scaffold 20-amino-20-deoxy-N6-(1-naphtylmethyl)adenosine (5a) we reported recently. This build-
ing block synthesized in solution was treated with an excess of polymer-supported carboxylic acids leading to chemoselective,
practically quantitative conversion of the amine to the desired analogous amides. The best compound (1h) fromthis series was
obtained after on-bead nucleophilic substitution of the carboxylic acid equivalent attached to the Kenner safety catch linker and
exhibited an improved inhibitory effect on T. b. brucei blood streamforms with an IC 50 of 0.85 mM in vitro # 2001 Elsevier Science
Ltd. All rights reserved.
Introduction
fromthe adenosine scaffold. Introducing diversity into
either the carbohydrate and/or base subunits of adeno-
sine represents promising strategies to identify inhibi-
tors of enzymes that are dependent on binding
nucleoside derivatives to attain activity.
The prevalence of African trypanosomiasis, that seemed
to decrease steadily in the 1960s, is now greater than
ever in sub-Saharan regions of the African continent.
Despite many promising reports on novel drugs against
trypanosomes under development, research in this area
is still highly demanded.1ꢀ4
Recent reports have demonstrated that several potent
and selective inhibitors of glycolytic enzymes in Kineto-
plastidae can be discovered by this approach and that
these compounds decrease survival of trypanosomes in
vitro.8 One of these seven glycolytic enzymes involved
in the conversion of glucose to 3-phosphoglycerate is
glyceraldehyde-3-phosphate dehydrogenase GAPDH.9
Ideally, the quest for selective drugs for the therapy of
infectious diseases starts fromthe identification of tar-
get structures that are absent in the host while exhibit-
ing essential functions for the infective agent.5
Alternatively, targeting the binding sites of coenzyms
rather than the often well-conserved active sites of
enzymes offers a rational starting point for the genera-
tion of selective inhibitors of protozoan parasites.6,7 In
this respect, a broad array of targets that utilize adenosyl-
containing substrates like ATP or NAD+ bear the
opportunity for the rational design of ligands starting
Chemistry
Using information derived from theoretical calculations
and experimental data from X-ray crystallography on
the binding orientation of an N6-benzylated NAD+-
derivative to the coenzyme-pocket of protozoal
GAPDH,
naphtylmethyl)adenosine (1a) was synthesized and
shown to exhibit a marked inhibitory potency and a
20-deoxy-20-(3-methoxybenzamido)-N6-(1-
*Corresponding author. Tel.: +49-40-42838-3467; fax: +49-40-
42838-6573; e-mail: link@uni-hamburg.de
0968-0896/02/$ - see front matter # 2001 Elsevier Science Ltd. All rights reserved.
PII: S0968-0896(01)00253-X

160
A. Golisade et al. / Bioorg. Med. Chem. 10 (2002) 159–165
synthetic strategies applying polymer supports as solu-
bility control auxiliaries in order to facilitate the gen-
eration of molecular libraries have proven to be a
powerful tool in drug discovery, we decided to synthe-
size a set of lead structure analogues by polymer-sup-
ported or polymer-assisted protocols.12ꢀ18 Thus, we
applied 20-amino-20-deoxy-N6-(1-naphthylmethyl)ade-
nosine (5a) as a core building block to access a series of
20-amido-20-deoxy-N6-(1-naphthylmethyl)adenosine de-
rivatives (1b–l).
The synthesis of the structurally demanding common
scaffold 5a was performed from the antiviral drug
vidarabine (2) following reported procedures.19,20
Figure 1. Lead structure 1a.
We investigated the versatility of this classical synthesis
via 6-chloropurine for the synthesis of alternatively N6-
substituted 20-amino-20-deoxyadenosine intermediates
because of the general interest in the chemistry of N6-
substituted adenosine derivatives that apparently
increased lately.21 Pd-catalyzed hydrogenation had been
reported to proceed slowly and incomplete, for example
for 20-azido-20-deoxy-N6-dimethyladenosine by Ikehara
and Takatsuka.22 Consequently we felt the necessity to
evaluate the effectiveness of this synthetic step for the
preparation of analogous amino nucleoside scaffolds.
The synthesis of model compound 5b was carried out as
outlined in Scheme 1 and proceeded well.
For fast and simple introduction of different acid resi-
dues in parallel, a polymer-assisted acylation protocol
was applied. Therefore commercially available car-
boxylic acids were coupled to the Kenner safety catch
linker improved by Backes et al., by established amide
bond forming procedures.23ꢀ30
Picture 1. N6-benzyl NAD+-derivative inside L. mexicana GAPDH.
considerable selectivity for the protozoal enzymes as
opposed to the human counterpart.10 The development
of this compound resulted from the rational attempt to
combine structural elements that demonstrated bene-
ficial effects for binding and inhibition, separately (Fig. 1,
Picture 1).7
Prior to the polymer-assisted reaction with scaffold 5a,
the Kenner linker has to be activated by N-alkylation to
forma good leaving group as described by Backes et al.
or by use of trimethylsilyldiazomethane.26,29 The poly-
mer bound acids subsequently form the desired amides
upon agitation with a solution of an appropriate amine
at slightly elevated temperatures.
The benzyl substituent (Picture 1, upper right corner) of
a N6-benzyl NAD+-derivative fills a narrow hydro-
phobic pocket in the non-functional adenosine binding
site that may lead to development of ligands with high
binding affinity (coordinates fromthe homologous
Leishmania mexicana GAPDH; PDB-identifier: 1GYQ,
authors: S. Suresh, W. Hol10). At the same time,
another lipophilic cleft at the 20-region (lower right cor-
ner), that is occluded in the human counterpart of
GAPDH, opens the way for the design of selective
inhibitors of trypanosomes.7 Generally, amido sub-
stitution of deoxyadenosines leads to compounds with
retained hydrogen bond donor capabilities on the
substitution site in contrast to O-acylated adenosine
derivatives.11
With the intention of widening the range of structural
modifications achievable on the 20-amino position
beyond the scope of commercially available carboxylic
acids, intermediate nucleophilic substitution of poly-
mer-bound residues as a source for diversity was incor-
porated into the sequence. Because the range of
available amines is exceptionally broad, this drug dis-
covery approach might be regarded as especially attrac-
tive.31 4-Fluoro-3-nitro-benzoic acid was thus attached
to the sulfamoyl linker yielding intermediate 6a and
subsequently converted to the corresponding aniline
derivatives 6g–l by treatment with excess amines
(Scheme 2).
As to be anticipated, the measured effects of substitu-
tion in the 20-region and on the N6-nitrogen atomcan-
not easily be correlated.11 Therefore, for optimization of
the lead structure 1a, a combinatorial approach allow-
ing for the synthesis of arrays of selected modifications
is an interesting objective to be addressed. Because
Alkylation with bromoacetonitrile as activation step for
the Kenner linker is reported to give poor results for
aromatic residues adjacent to the sulfamoyl attachment
site.26 For that reason we used prolonged reaction times
for this activation protocol and employed commercially

A. Golisade et al. / Bioorg. Med. Chem. 10 (2002) 159–165
161
available trimethylsilyl diazomethane as an alternative
activation reagent to ensure optimal results and for
means of comparison. However, in both cases, the
yields of the compounds obtained were generally good.
The predicted less effective protocol using bromoaceto-
nitrile turned out to be slightly superior within this ser-
ies of compounds. Apparently, the advantage of a
putative higher ratio of activation with trimethylsilyl-
diazomethan is overcompensated by the resulting less
effective activation as compared to the electron-with-
drawing variant (Table 1).
A novel interesting alternative activation procedure via
Mitsunobu reaction of polymer bound acyl sulfona-
mides using pentafluorobenzyl alcohol, triphenylpho-
sphine, and diisopropyl azodicarboxylate was added to
this list of techniques only recently.32 Alkylation of the
aniline-nitrogen in 6g–l was observed in neither case.
Biological evaluation
The series of 20-amido-20-deoxy-N6-(1-naphthylmethyl)-
adenosines 1b–l as well as the related compounds 4a, 4b
and 5a, 5b were tested for in vitro activity upon trypo-
mastigotes (T. b. brucei) as described by Hirumi et al.33
For means of comparison, the most active compound of
this class 1a reported so far was included in the evalua-
tion as well as N6-(1-naphthylmethyl)adenosine (7) itself
(structure not shown). The results obtained are sum-
marized below (Table 2).
Scheme 1. Synthesis of compounds 5a and 5b. (a) (i) TIPDSCl, pyri-
dine; (ii) TfCl, DMAP, CH2Cl2; (iii) NaN3, DMF; (iv) CCl4, isoamyl
nitrite, 60 ꢁC; (v) TBAF, THF, 50 ꢁC, 5 h; (b) 4a: 1-naphthylmethyla-
mine, 1-propanol, 50 ꢁC, 16 h, 4b: 2-(2-methoxyphenyl)ethylamine, 1-
propanol, 50 ꢁC, 16 h; (c) 5a: InN/H4Cl, EtOH ÁT; (d) 5b: Pd/C 10%,
H2 (2 bar), dioxane, 6 h.
Discussion
Because the Kenner linker modified by Ellman and his
group tolerates strongly basic or acidic conditions prior
to activation, on-bead modifications of carboxylic acids
attached to the linker permit the construction of poly-
mer bound carboxylic acid equivalent libraries. We
demonstrated its value as a tool for the simple on-bead
modification by nucleophilic substitution as source of
diversity. Subsequent parallel modification of amino-
deoxynucleosides was shown to be a straightforward
concept to obtain desired 20-amido-20-deoxy-N6-(1-
naphthylmethyl)adenosines. The first series of com-
pounds obtained by this approach showed marked
inhibitory activity in vitro. Compounds 1b and 1h–l
displayed inhibitory concentrations comparable to the
analogue with the lowest IC50 versus T. b. brucei
GAPDH described before (1a), all in the same order of
Table 1. Results of alternative activation protocols
Entry 1
From 6h-–l, R1=CH3
From 6h–l, R1=CH2CN
Puritya (%)
Yieldb (%)
Puritya (%)
Yieldb (%)
h
i
k
l
92
94
92
95
88
85
96
94
94
96
97
95
96
92
95
96
aPurity of crude reaction filtrate determined via semi-preparative
MPLC, 100% method, detection at 254 nm.
bIsolated material obtained by evaporation of purified product con-
taining fractions.
Scheme 2. Formation of aniline derivatives 6g–l fromfluoro-analogue
6a. (a) Appropriate amine (excess), DMF. R1=H; (b) Activation by
treatment with trimethylsilyldiazomethan in hexanes for 24
(R1=CH3) or BrCH2CN, DIPEA in NMP 48 h (R1=CH2CN).
h

162
A. Golisade et al. / Bioorg. Med. Chem. 10 (2002) 159–165
Table 2. In vitro activity upon trypomastigotes (T. b. brucei)
Compd
Yield (%)a
Residue R in structures 1
IC50b (mM)
1a
1b
1c
1d
1e
95
97
96
93
96
95
3-Methoxyphenyl
2-(3-Indolyl)ethyl
3-(3-Indolyl)propyl
5
4
19
>32
>32
17
3-(3,5-Dichlorophenoxy)propyl
(3-Thienyl)methyl
3,5-Difluorophenyl
1f
1g
1h
1i
1k
1l
85
4-[2-(4-Methoxyphenyl)ethyl]amino-3-nitrophenyl
4-Cyclopropylamino-3-nitrophenyl
3-Nitro-[4-(piperid-1-yl)]phenyl
3-Nitro-[4-(morpholin-1-yl)]phenyl
(4-Chlorobenzyl)amino-3-nitrophenyl
16
88–96
85–92
95–96
94–96
0.85
3.5
4
8
4a
4b
5a
5b
7
>32
>32
>32
>32
4
Suramin
0.059
aIsolated material obtained by evaporation of purified product containing fractions.
bIC50 values were determined by measuring growth inhibition at 3–5 concentrations. IC50 values for 1h and 1i were confirmed by repeated experi-
ments (identical conditions, 5 concentrations) and are reported as arithmetic mean of repeated experiments.
magnitude. The IC50 of the apparently most active
compound 1h of 0.85 mM was confirmed in repeated
experiments, ranging from 0.7 mM to 1.0 mM.
case one should not consider this approach a PASP
sequence but more or less a SPOS or PASP/SPOS
hybride technique. Because acid residues can be trans-
ferred that otherwise would have had to be synthesized
and purified separately, the additional operations of
attachment and activation become much more economical
and this approach represents a considerable advantage in
the process of ‘high-speed-analogueing’.
Surprisingly even the 20-unsubstituted adenosine deri-
vative N6-(1-naphthylmethyl)adenosine (7) included for
means of comparison, showed a marked inhibition of T.
b. brucei trypomastigotes of IC50=4 mM in vitro. This
result is not in accordance with results fromenzyme
based assays. The IC50 for the highly analogous
GAPDH of L. mexicana has been determined to be
much higher (150 mM).8 Therefore the question arises,
whether the activity of 7 versus trypomastigotes is not
attributed to inhibition of GAPDH alone. This finding
will be addressed by future investigations. Therefore
compounds 1h and 7 will be immobilized on agarose
beads and surface plasmon resonance detection devices
and used to screen for hitherto not identified binding
partners from cell fractions and an array of commer-
cially available enzymes. This well established technique
was shown to be a valuable tool in a project for the
evaluation of target structures for the novel class of cyclin-
dependent kinase inhibitors called paullones we reported
recently.34 The increased inhibitory activity of compound
1h in comparison to the lead 1a will most likely become
helpful in this approach because weak affinity binders
usually do not lead to straightforward target identification.
Conclusion
We were successful in the identification of a lead struc-
ture analogue with improved biological activity using
on-bead nucleophilic substitution on a safety catch lin-
ker as source for diversity. Additional modifications of
the improved lead structure 1h in the region of the 50-
OH seemto be beneficial in cases where the binding
region of the nicotinamide moiety of NAD+ can be
targeted as well.37 Because structural modifications in
different molecular regions of this class of compounds
are not easy to correlate due to conformational flex-
ibility, a library of polymer-supported acids obtained as
described above might become a valuable tool for the
synthesis of amide libraries based on multi-substituted
nucleoside scaffolds in the future.
In our initial attempts unchanged acid equivalents were
transferred to building blocks originating fromsolution-
phase chemistry.30 Because the coupling to the resin and
subsequent transfer to building blocks in solution after
activation is just an advanced use of coupling reagents,
this approach cannot be regarded as a solid-phase
synthesis (SPOS) technique. Due to this reason, we
consider this application of the Kenner linker for the
construction of chemoselective acylating reagents a spe-
cial kind of polymer-assisted solution phase (PASP)
synthesis.35,36 In the series 1g–l reported here, acid resi-
dues are altered by on-bead modification prior to
activation of the linker and subsequent transfer. In this
Experimental
Identity of all compounds was assigned by NMR spec-
1
troscopy. Sample purity was deduced from H NMR
data as well as evaluated by MPLC. Purity is reported
as purity of crude products prior to purification (per-
centage of target compound contained in residue from
evaporated reaction mixture). Yields are reported as
isolated material obtained by evaporation of purified
product containing fractions. 1H NMR spectra were
recorded on a Bruker AMX 400 spectrometer, using tet-
ramethylsilane as internal standard. MPLC simultaneous
purity analyses/purifications were performed using a

A. Golisade et al. / Bioorg. Med. Chem. 10 (2002) 159–165
163
Buchi 681 pump (flow rate 10 mL, MeOH/H2O 70:30),
and UV-detector (254 nm) with Merck 310-25 Lobar-
LiChroprep1-RP-18 columns. TLC was performed on
Macherey-Nagel Polygram1 Sil G/UV254 precoated
microplates, spots were visualized under UV-illumina-
tion. Infrared spectra were recorded using KBr pellets
on a Perkin Elmer 1660 FT-IR spectrometer. MS data
(FAB) was obtained on a Finnigan MAT 311A instru-
ment with m-nitrobenzylic alcohol as matrix. Solvents
were purified according to standard procedures and
freshly distilled prior to use. Standard glass ware was
oven dried at 150 ꢁC and kept in a desiccator.
supported activated acids 6h–l were transferred to the
amino group of 4.06 mg (10 mmol) of 20-amino-20-
deoxy-N6-(1-naphthylmethyl)adenosine (5a) by shaking
at 55 ꢁC in 5 mL THF. The reaction was monitored by
TLC and terminated when the starting material was
quenched (6–24 h). Polymer beads and particulates were
removed by filtration, the beads where washed exhaus-
tively with dry THF and the combined THF fractions
were evaporated. 1H NMR experiments of the crude
material obtained as well as MPLC analyses revealed
high purity of the compounds and the absence of
impurities in detectable quantities other than traces of
starting material or acid hydrolyzed from the resin.
General procedure A for the synthesis of the polymer
supported acids 6h–l. To a flask containing 52.0 g of
dry 4-sulfamylbenzamidomethyl polystyrene with an
initial loading level of 1.25 mmol/g as determined by
elemental analysis (prepared from very high load ami-
nomethylated polystyrene batch number A20540, pur-
chased fromNovabiochem, Switzerland) was added 500
mL of THF. The resin was allowed to swell at room
temperature for 2 h. In another flask, 44.4 g (240 mmol)
of 4-fluoro-3-nitro benzoic acid was dissolved in 500 mL
dry THF and preactivated via in situ anhydride formation
by adding 18.6 mL (120 mmol) N,N-diisopropylcarbodii-
mide overnight. CAUTION: N,N-diisopropylcarbo-
diimide may lead to severe allergic reactions, strictly
avoid skin contact. After addition of 15 mL Hunig’s
base (88 mmol) and 500 mg (4.1 mmol) 4-(dimethyla-
mino)pyridine as catalyst, to the swollen resin, the cou-
pling mixture was added. The resulting reaction mixture
was agitated at roomtemperature for 48 h. The resin
beads were filtered off and washed exhaustively with
THF (two times), methanol (two times), and THF (two
times) resulting in 53.48 g resin after careful drying in
vacuo. This weight increase of 11.48 g corresponds to a
practically quantitative conversion of 4-sulfamylbenza-
midomethyl polystyrene to [(4-fluoro-3-nitrobenzoyl)4-
sulfamyl]benzamidomethyl polystyrene 6a resulting in a
loading level of 1.00 mmol/g. For the derivatization of
6a to resins 6h–l, 2.0 g 6a (2.0 mmol) each was treated
with 20 mmol of the appropriate amine in 25 mL DMF
overnight and subsequently washed with DMF (three
times), methanol (two times), and THF (two times).
20 -[(4-Cyclopropylamino-3-nitro)benzamido]-20 -deoxy-
N6-(1-naphthylmethyl)adenosine (1h). Compound 1h
was prepared fromresin 6h synthesized by general pro-
cedure A using cyclopropyl amine. Following general
procedure B, activation method A1 88%, activation
method A2 96% yellow product was obtained. 1H
NMR (400 MHz, DMSO-d6) d (ppm)=8.77 (bs, 1H,),
8.66 (s, 1H), 8.58 (d, 1H, J=8.14 Hz), 8.37 (s, 1H),
8.29–8.14 (m, 3H), 8.05–8.02 (m, 1H), 7.96–7.94 (m,
1H), 7.83–7.81 (m, 1H), 7.59–7.52 (m, 2H), 7.46–7.34
(m, 3H), 6.28 (d, 1H, J=8.14 Hz), 5.40–5.31 (m, 1H),
5.18 (bs, 2H), 4.35 (d, 1H, J=5.09 Hz), 4.14–4.08 (m,
1H), 3.77–3.69 (m, 1H), 3.65–3.58 (m, 1H), 2.72–2.63
(m, 1H), 0.91–0.87 (m,2H), 0.68–0.59 (m, 2H), HRFAB-
MS [M+H]+ calcd=611.2367 found=611.2351,
MPLC purity: 92% for A1, 94% for A2.
20-Deoxy-N6-(1-naphthylmethyl)-20-[3-nitro-4-(piperid-1-
yl)benzamido]adenosine (1i). Compound 1i was pre-
pared fromresin 6i synthesized by general procedure A
using piperidine. Following general procedure B, acti-
vation method A1 85%, activation method A2 92%
orange product was obtained. 1H NMR (400 MHz,
CDCl3) d (ppm)=8.46 (bs, 1H,), 8.38 (d, 1H, J=2.03
Hz), 8.10–8.07 (m, 1H), 7.99–7.97 (dd, 1H, J=2.03, 6.61
Hz), 7.90–7.88 (m, 1H), 7.85–7.83 (d, 1H, J=8.14 Hz),
7.78 (bs, 1H), 7.56–7.51 (m, 3H), 7.46–7.43 (t, 1H,
J=8.14, 7.12 Hz), 7.10–7.08 (d, 1H, J=8.65 Hz), 6.21
(bs, 1H), 6.07 (bs, 1H), 5.83 (d, 1H, J=4.07 Hz), 5.70
(dd, 1H, J=3.06, 5.59), 5.30 (bs and s overlapped, 3H),
5.21 (dd, 1H, J=4.07, 4.58), 4.59–4.52 (m, 1H), 4.14–
4.11 (m, 1H), 3.99–3.92 (m, 1H), 3.17–3.14 (m, 4H),
Activation method A1. 400 mg of resin 6h–l were subse-
quently swollen in THF and treated with 4.0 mL (tri-
methylsilyl)diazomethane 1 M solution in hexanes (4.0
mmol) purchased from Aldrich for 24 h and washed
with THF (10 mL), methanol (three times 10 mL), and
THF (10 mL).
1.78–1.61
(m,
6H),
HRFAB-MS
[M+H]+
calcd=638.2601 found=638.2588, MPLC purity: 94%
for A1, 96% for A2.
20-Deoxy-20-[4-(morphol-1-yl)-3-nitro]benzamido-N6-(1-
naphthylmethyl)adenosine (1k). Compound 1k was pre-
pared fromresin 6k synthesized by general procedure A
using morpholine. Following general procedure B, acti-
vation method A1 96%, activation method A2 95%
yellow product was obtained. 1H NMR (400 MHz,
CDCl3) d (ppm)=8.46 (bs, 1H), 8.42 (d, 1H, J=2.03
Hz), 8.10–8.03 (m, 2H), 7.90–7.87 (m, 1H), 7.85 (d, 1H,
J=8.65 Hz), 7.78 (bs, 1H), 7.56–7.49 (m, 1H), 7.44 (t,
1H, J=8.14, 7.12 Hz), 7.13 (d, 1H, J=9.16 Hz), 6.20
(bs, 1H), 6.04 (bs, 1H), 5.84 (d, 1H, J=4.07 Hz), 5.71
(dd, 1H, J=2.54, 6.11 Hz), 5.30 (bs and s overlapped,
3H), 5.22 (dd, 1H, J=4.07, 4.58 Hz), 4.60–4.54 (m, 1H),
Activation method A2. The sulfonamide linker of 400
mg 6h–l was activated for cleavage by alkylation with
640 mL (9 mmol) bromoacetonitrile CAUTION: alky-
lating agent, strictly avoid skin contact, and 340 mL (2
mmol) Hunig’s base in 3 mL 1-methylpyrrolidone for 48
h. The dark brown slurry was washed with dry dime-
thylsulfoxide (5ꢂ5 mL) and THF (5ꢂ10 mL) yielding
off-white to yellowish resin particles.
General procedure B for the synthesis of 20-amido-20-deoxy-
N6-(1-naphthylmethyl)adenosines 1h–l. The polymer

164
A. Golisade et al. / Bioorg. Med. Chem. 10 (2002) 159–165
4.14 (dd, 1H, J=1.02, 11.70), 3.98–3.96 (m, 1H), 3.86 (t,
4H, J=4.58 Hz), 3.17 (t, 4H, J=4.58), HRFAB-MS
[M+H]+ calcd=640.2394, found=640.2395, MPLC
purity: 92% for A1, 97% for A2.
HMI-9 medium as described by Hirumi et al.33 In a 96-
well microplate, 10,000 haemoflagellates were incubated
at different drug concentrations for 4 days. Parasite
multiplication was evaluated microscopically. IC50
values were determined by measuring growth inhibition
at 3–5 concentrations. Suramin was included as standard
agent.
20-[4-(4-Chlorobenzyl)amino-3-nitro]benzamido-20-deoxy-
N6-(1-naphthylmethyl)adenosine (1l). Compound 1l was
prepared fromresin 6l synthesized by general procedure
A using 4-chlorobenzyl amine. Following general pro-
cedure B, activation method A1 94%, activation
method A2 96% yellow product was obtained. 1H
NMR (400 MHz, CDCl3) d (ppm)=8.83 (d, 1H,
J=1.52 Hz), 8.70 (t, 1H, J=5.09 Hz), 8.46 (bs, 1H),
8.10–8.07 (m, 1H), 8.00–7.97 (m, 1H), 7.90–7.88 (m,
1H), 7.85 (d, 1H, J=8.14 Hz), 7.79 (bs, 1H), 7.56–7.51
(m, 4H), 7.44 (t, 1H, J=4.14, 7.12 Hz), 7.37 (d, 2H,
J=8.65 Hz), 7.29 (d, 2H, J=8.65 Hz), 6.84 (d, 1H,
J=9.16 Hz), 6.19 (bs, 1H), 6.05 (bs, 1H), 5.83 (d, 1H,
J=4.58 Hz), 5.69 (dd, 1H, J=3.05, 4.59 Hz), 5.30 (bs
and s overlapped, 3H), 5.19 (dd, 1H, J=4.07, 4.58 Hz),
4.60 (d, 2H, J=5.59 Hz), 4.57–4.55 (m, 1H), 4.14 (d,
1H, J=11.19 Hz), 3.98–3.96 (m, 1H), HRFAB-MS
[M+H]+ calcd=694.2055, found=694.2074, MPLC
purity: 95% for A1 and A2.
Acknowledgements
This work was supported by the Fonds der Chemischen
Industrie FCI, the Deutsche Pharmazeutische Gesell-
schaft
DPhG,
and
the
Deutsche
For-
schungsgemeinschaft DFG (Graduiertenkolleg 464).
Helpful discussion and encouragement by Prof. Dr. D.
Geffken is gratefully acknowledged.
References and Notes
1. Control and Surveillance of African Trypanosomiasis Report
of a WHO Expert Committee; WHO: Geneva, 1998; Vol. 881.
2. Pecoul, B.; Gastellu, M. Lancet 1999, 354, 955.
3. Kaminsky, R.; Brun, R. Antimicrob. Agents Chemother.
1998, 42, 2858.
20-Azido-20-deoxy-N6-[2-(2-methoxyphenyl)ethyl]adeno-
sine (4b). An amount of 330 mg (1.06 mmol) 3 was dis-
solved in 20 mL 1-propanol. To this solution were
4. Zhou, W.; Xin, Z.; Zhang, X.; Shen, J.; Qiu, Q. Yao Hsueh
Hsueh Pao 1996, 31, 823.
added
200
mg
(1.35
mmol)
2-(2-methoxy-
phenyl)ethylamine and 250 mL (1.50 mmol) Hunig’s
base. The reaction mixture was stirred at 50 ꢁC for 16 h.
Evaporation of the solvent and purification via column
chromatography (4.5ꢂ15 cm, n-hexane/EtOAc 1:1) gave
430 mg (97%) of a pale yellow foam. ir (KBr): 2110
5. Jomaa, H.; Wiesner, J.; Sanderbrand, S.; Altincicek, B.;
Weidemeyer, C.; Hintz, M.; Turbachova, I.; Eberl, M.; Zei-
dler, J.; Lichtenthaler, H.; Soldati, D.; Beck, E. Science 1999,
285, 1573.
6. Opperdoes, F. R. Annu. Rev. Microbiol. 1987, 41, 127.
7. Verlinde, C. M. L. J.; Callens, M.; Van Calenbergh, S.; Van
Aerschot, A.; Herdewijn, P.; Hannaert, V.; Michels, P. A. M.;
Opperdoes, F. R.; Hol, W. G. J. J. Med. Chem. 1994, 37, 3605.
8. Aronov, A. M.; Verlinde, C. L. M. J.; Hol, W. G. J.; Gelb,
M. H. J. Med. Chem. 1998, 41, 4790.
9. Verlinde, C. M. L. J.; Hannaert, V.; Blonski, C.; Willson,
M.; Perie, J. J.; Fothergill-Gillmore, L. A.; Opperdoes, F. R.;
Hol, W. G. J.; Michels, P. A. M. Drug Res. Updates 2001, 4, 1.
10. Aronov, A. M.; Suresh, S.; Buckner, F. S.; Voorhis,
W. C. V.; Verlinde, C. L. M. J.; Opperdoes, F. R.; Hol,
W. G. J.; Gelb, M. H. Proc. Natl. Acad. Sci. U.S.A. 1999, 96,
4273.
11. Van Calenbergh, S.; Verlinde, C. M. L. J.; Soenens, J.; De
Bruyn, A.; Callens, M.; Blaton, N. M.; Peeters, O. M.;
Rozenski, J.; Hol, W. G.; Herdewijn, P. J. Med. Chem. 1995,
38, 3838.
12. Fruchtel, J. S.; Jung, G. Angew. Chem. Int. Ed. 1996, 35,
17.
13. Hall, S. Mol. Diversity 1999, 4, 131.
14. James, I. W. Mol. Diversity 1998, 3, 181.
15. Nefzi, A.; Dooley, C. T.; Ostresh, J. M.; Houghten, R. A.
Bioorg. Med. Chem. Lett. 1998, 8, 2273.
16. Smith, J. R. Biomed. Health Res. 1999, 22, 171.
17. Flynn, D. L. Med. Res. Rev. 1999, 19, 409.
18. Link, A. Angew. Chem. Int. Ed. 2000, 39, 4039.
19. Robins, M.; Wnuk, S.; Hernandez-Thirring, A.; Samano,
M. J. Amer. Chem. Soc. 1996, 118, 11341.
cmꢀ1 (N3). 1H NMR (400 MHz, DMSO-d6)
d
(ppm)=8.38 (s, 1H), 8.24 (bs, 1H), 7.89 (s, 1H), 7.21–
7.15 (m, 2H), 6.99 (d, 2H, J=7.63 Hz), 6.85 (t, 1H,
J=7.63, 7.12 Hz), 6.07–6.04 (m, 2H, 10H and 30OH over-
lapped), 5.30 (t, 1H, J=5.60 Hz), 4.64 (t, 1H, J=6.10, 5.60
Hz), 4.57–4.51 (m, 1H), 4.01 (m, 1H), 3.78 (s, 3H); 3.72–
3.64 (m, 3H), 3.60–3.53 (m, 1H), 2.91 (t, 2H, J=7.12 Hz)
HRFAB-MS [M+H]+ calcd: 433.1737 found: 433.1742.
20-Amino-20-deoxy-N6-[2-(2-methoxyphenyl)ethyl]adeno-
sine (5b). A solution of 300 mg (0.74 mmol) 4b, in
dioxane (150 mL) was purged with nitrogen and
charged with 10% Pd-C (100 mg). The reaction mixture
was shaken under hydrogen (pressure 2 bar) for 6 h.
The catalyst was removed by filtration through silica
gel. Evaporation of the solvent and purification over
silica gel (1.5ꢂ15 cm, EtOAc) and DOWEXTM 2ꢂOH
(4.5ꢂ15 cm, MeOH/H2O 70:30) sequentially gave 290
1
mg (97%) of the title compound. H NMR (400 MHz,
DMSO-d6) d (ppm)=8.29 (bs, 1H) 8.21 (bs, 1H), 7.84
(bs, 1H), 7.21–7.15 (m, 1H), 6.97 (d, 2H, J=8.14 Hz),
6.85 (t, 1H, 10H, J=7.63, 7.12 Hz), 5.68 (d, 1H, J=8.14
Hz), 5.56–5.46 (m, 2H), 4.04–3.94 (m, 3H), 3.78 (s, 3H),
3.71–3.63 (m,3H), 3.57–3.48 (m, 1H); 2.91 (t, 2H,
J=7.63, 7.12 Hz), HRFAB-MS [M+H]+ calcd:
401.1937 found: 401.1913.
20. Golisade, A.; Van Calenbergh, S.; Link, A. Tetrahedron
2000, 56, 3167.
21. Lin, X.; Robins, M. Org. Lett. 2000, 2, 3497.
22. Ikehara, M.; Takatsuka, Y. Chem. Pharm. Bull. 1978, 26,
985.
In vitro activity upon trypomastigotes (T. b. brucei).
Blood streamforms of T. b. brucei were cultivated in

A. Golisade et al. / Bioorg. Med. Chem. 10 (2002) 159–165
165
23. Kenner, G. W.; McDermott, J. R.; Sheppard, R. C. J.
Chem. Soc., Chem. Commun. 1971, 636.
31. Zaragoza, F.; Stephensen, H. Angew. Chem. Int. Ed. 2000,
39, 554.
24. Backes, B. J.; Ellman, J. A. J. Am. Chem. Soc. 1994, 116,
11171.
32. Shen, D.-M.; Shu, M.; Chapman, K. Org. Lett. 2000, 2,
2789.
25. Backes, B. J.; Harris, J. L.; Leonetti, F.; Craik, C. S.; Ell-
man, J. A. Nat. Biotechnol. 2000, 18, 187.
26. Backes, B. J.; Virgilio, A. A.; Ellman, J. A. J. Am. Chem.
Soc. 1996, 118, 3055.
27. Backes, B. J.; Ellman, J. A. J. Org. Chem. 1999, 64, 2322.
28. Yang, L.; Morriello, G. Tetrahedron Lett. 1999, 40, 8197.
29. Ingenito, R.; Bianchi, E.; Fattori, D.; Pessi, A. J. Am.
Chem. Soc. 1999, 121, 11369.
30. Link, A.; Van Calenbergh, S.; Herdewijn, P. Tetrahedron
Lett. 1998, 39, 5175.
33. Hirumi, H.; Hirumi, K. J. Parasitology 1989, 75, 985.
34. Leost, M.; Schultz, C.; Link, A.; Wu, Y.; Biernat, J.;
Mandelkow, E.; Bibb, J.; Snyder, G.; Greengard, P.; Zahar-
evitz, D.; Gussio, R.; Senderowicz, A.; Sausville, E.; Kunick,
C.; Meijer, L. Eur. J. Biochem. 2000, 267, 5983.
35. Flynn, D. L.; Devraj, R.; Naing, W.; Parlow, J.; Weidner,
J.; Yang, S. Med. Chem. Res. 1998, 8, 219.
36. Curran, D. P. Angew. Chem. Int. Ed. 1998, 37, 1174.
37. Kennedy, K. J.; Bressi, J. C.; Gelb, M. H. Bioorg. Med.
Chem. Lett. 2001, 11, 95.