Synthesis and biological evaluation of substrate-based inhibitors of 6-phosphogluconate dehydrogenase as potential drugs against African trypanosomiasis

Article abstract of DOI:10.1016/S0968-0896(03)00191-3

The synthesis and biological evaluation of three series of 6-phosphogluconate (6PG) analogues is described. (2R)-2-Methyl-4,5-dideoxy, (2R)-2-methyl-4-deoxy and 2,4-dideoxy analogues of 6PG were tested as inhibitors of 6-phosphogluconate dehydrogenase (6P

Full text of DOI:10.1016/S0968-0896(03)00191-3

Bioorganic & Medicinal Chemistry 11 (2003) 3205–3214
Synthesis and Biological Evaluation of Substrate-Based Inhibitors
of 6-Phosphogluconate Dehydrogenase as Potential Drugs Against
African Trypanosomiasis
Christophe Dardonville,a Eliana Rinaldi,b Stefania Hanau,b Michael P. Barrett,c
Reto Brund and Ian H. Gilberta,*
aWelsh School of Pharmacy, Redwood building, Cardiff University, King Edward VII Avenue, Cardiff CF10 3XF, UK
bDipartimento di Biochimica e Biologia Molecolare, Universita’ di Ferrara, Via Luigi Borsari 46, 44100 Ferrara, Italy
cIBLS, Division of Infection & Immunity, University of Glasgow, Glasgow G12 8QQ, UK
dSwiss Tropical Institute, Socinstrasse 57, CH-4002 Basel, Switzerland
Received 10 January 2003; accepted 21 March 2003
Abstract—The synthesis and biological evaluation of three series of 6-phosphogluconate (6PG) analogues is described. (2R)-2-
Methyl-4,5-dideoxy, (2R)-2-methyl-4-deoxy and 2,4-dideoxy analogues of 6PG were tested as inhibitors of 6-phosphogluconate
dehydrogenase (6PGDH) from sheep liver and also Trypanosoma brucei where the enzyme is a validated drug target. Among the
three series of analogues, seven compounds were found to competitively inhibit 6PGDH from T. brucei and sheep liver enzymes at
micromolar concentrations. Six inhibitors belong to the (2R)-2-methyl-4-deoxy series (6, 8, 10, 12, 21, 24) and one is a (2R)-2-
methyl-4,5-dideoxy analogue (29b). The 2,4-dideoxy analogues of 6PG did not inhibit both enzymes. The trypanocidal effect of the
compounds was also evaluated in vitro against T. brucei rhodesiense as well as other related trypanosomatid parasites (i.e., Trypano-
soma cruzi and Leishmania donovani).
# 2003 Elsevier Science Ltd. All rights reserved.
Introduction
phogluconate dehydrogenase (6PGDH), the third
enzyme of the pentose phosphate pathway7 (PPP) which
generates NADPH and ribulose-5-phosphate (Fig. 1).
The PPP plays a crucial role in the host–parasite rela-
tionship because it maintains a pool of NADPH, which
amongst other roles is involved in protecting the para-
site against oxidative stress, and it generates carbohy-
drate intermediates used in nucleotide and other
biosynthetic pathways.
Human African trypanosomiasis is a major health
problem in sub-Saharan Africa, caused by, Trypano-
soma brucei gambiense in West and Central Africa and
T.b. rhodesiense in East Africa.1,2 The drugs currently
used to treat sleeping sickness are far from satisfactory
because of major side effects,3 the parenteral mode of
administration, increasing resistance4 and the unafford-
able price for African countries. Moreover, some of
these drugs (suramin, pentamidine) are unable to cross
the blood–brain barrier in sufficient quantity to treat
late-stage cases of HAT.5
The gene encoding the T. brucei 6PGDH has been
cloned,8 and the enzyme purified9 and crystallized. The
enzyme is remarkable for its degree of divergence with
regard to other eukaryotic 6PGDHs. This can be
Bloodstream forms of T. brucei produce ATP exclu-
sively through glycolysis, making the inhibition of any
of the glycolytic enzymes a potential therapeutic
approach.6 We decided to target the enzyme 6-phos-
*Corresponding author. Fax.: +44-(0)29-2087-4149; e-mail: gilber-
tih@cf.ac.uk
Figure 1.
0968-0896/03/$ - see front matter # 2003 Elsevier Science Ltd. All rights reserved.
doi:10.1016/S0968-0896(03)00191-3

3206
C. Dardonville et al. / Bioorg. Med. Chem. 11 (2003) 3205–3214
attributed to the fact that the trypanosomatid 6PGDH
has been derived from a cyanobacterial precursor.10
However, despite sharing only 35% amino acid
sequence identity with the sheep liver enzyme, residues
which make direct contact with the substrate in the
active site appear to be conserved between the two
enzymes. Similarly, most of the residues contacting the
co-enzyme are conserved.11
RNA interference has recently demonstrated that
6PGDH is essential for growth of bloodstream form T.
brucei.12 One possible mechanism by which cell death
results from inhibition of 6PGDH involves an accumu-
lation of 6PG, which in turn is a potent inhibitor of
6-phosphoglucose isomerase13 and thus the glycolytic
pathway, which leads to increased passage of glucose
6-phosphate through the PPP and hence increased levels
of 6PG, initiating a toxic positive feedback loop. This
mode of action has been supported in yeast14 and
Drosophila.15
Crystallographic studies of the sheep liver16 and the T.
brucei enzyme have shown that, in spite of the con-
servation of the amino-acid residues, spatial differences
at the active site exist and they may be exploited for
selective drug design. Of particular note is the fact that
the trypanosomal enzyme has a difference in the binding
pocket in the region of the 2-OH group of the substrate.
Selective inhibitors of the trypanosomal enzyme have
been identified, including sugar derivatives17 and 2-
Figure 2.
Figure 3. Compounds tested as inhibitors of 6PGDH.

C. Dardonville et al. / Bioorg. Med. Chem. 11 (2003) 3205–3214
3207
deoxy-6-phosphogluconate (Fig. 2) which shows several
orders of magnitude more potency in inhibition of the
T. brucei enzyme (Ki=4.4 mM) compared to that of
sheep liver (Ki=770 mM).18a Investigation of the
mechanism of action of 6PGDH from different species by
Rippa et al.18b suggested that the C2 hydroxyl of 6PG was
important for the binding of the substrate, for communi-
cation between subunits in the 6PGDH homodimer and
for the keto-enol tautomerization. However there were
important differences between species as evidenced by the
differences in binding of 2-deoxy-6-phosphogluconate.
In designing inhibitors of 6PGDH that might have
efficacy as anti-trypanosomal drugs, a number of con-
siderations need to be made. Phosphorylated carbohy-
drates are unlikely to possess good drug-like
characteristics due to metabolic instability and imper-
meability at biological membranes. However, deoxy
and/or protected analogues of 6PG may represent
inhibitors with better pharmacokinetic properties.
Introduction of a methyl group in C2 was considered
potentially capable of forming hydrophobic interactions
with T. brucei enzyme due to the differences in binding
pocket found juxtaposed to the 2-position in this
variant of the enzyme.
Molecular modelling studies on the interaction between
6PG and the T. brucei enzyme revealed a number of
possibly important functionalities involved in the
interaction.19 Modelling indicates that there is little
scope for interaction between the enzyme and the 4-OH,
although studies using the Candida enzyme did indicate
that 6-phosphogalactonate (the 40-epimer of 6PG) was
not bound by the enzyme, perhaps indicating steric
hindrance if the 4-OH was in the 4S-conformation.20
Considering these facts, we designed three series of 6PG
analogues: (2R)-2-methyl-4,5-dideoxy, (2R)-2-methyl-4-
deoxy and 2,4-dideoxy analogues (Fig. 2) that were used
to probe the active site of T. brucei and sheep 6PGDH
around the 2-, 4- and 5-positions. To assess the impor-
tance of the individual free OH groups, we also prepared
some derivatives protected on the 3-OH and 5-OH
Scheme 1. (a) Bu2BOTf, CH2Cl2, Et3N, 0 ꢀC, then 2, À78 ꢀC; (b) Bu3SnH, 1,10-azobis(cyclohexanecarbonitrile), benzene, reflux, 2 h; (c) LiOH,
ꢀ
ꢀ
.
H2O2, 0 C; (d) BnBr, K2CO3, DMF, rt; (e) CuCl2 2H2O, iPrOH; (f) (BnO)3P, I2, À78 C, CH2Cl2; (g) CCl3NHOBn, TfOH, CH2Cl2, rt; (h) Ac2O,
DMAP, pyridine, rt; (i) MeO3BF4, DTBMP, CH2Cl2, rt; (j) PhCH(OMe)2, PPTS, DMF; (k) DHP, PPTS, CH2Cl2; (l) H2, Pd/C, MeOH then HCl
1M; (m) MeOH/HCl, rt.

3208
C. Dardonville et al. / Bioorg. Med. Chem. 11 (2003) 3205–3214
groups (i.e., Ac, Me, Bn, benzylidene). The three series
of compounds were tested as inhibitors of 6PGDH of T.
brucei and sheep liver and their trypanocidal effect was
also evaluated in vitro against T. brucei sub-species as
well as other related trypanosomatids parasites (i.e.,
Trypanosoma cruzi and Leishmania donovani).
phorylating agent were added at once, the diphos-
phorylated compound 25 was obtained as major
product (61%). The second phosphorylation in 25 was
thought to occur on the 3-OH, based on the downfield
chemical shift observed in the 13C NMR spectra for the
CH (+0.1ppm) and CH 3 (+0.5 ppm) in b-position and
the up-field chemical shift for the carbonyl of the ester
group (À1ppm) in the g-position.
Results and Discussion
Chemistry
The free hydroxyl groups of phosphates 12 and 13 were
protected with acetyl (15), methyl (16), benzyl (14),
benzylidene (17) and THP groups (18). Phosphate 20
was obtained by hydrogenolysis of 15 with 5% Pd/C.
Compounds 20 and 22 were obtained from 18 and 19,
respectively, by hydrogenolysis followed by mild acidic
hydrolysis of the THP protecting groups.27
Synthesis of (2R)-2-methyl-4-deoxy and 2,4-dideoxy ana-
logues (Scheme 1). Our synthetic strategy21 for the
synthesis of the (2R)-2-methyl-deoxy analogues of 6PG
relied on the Evans aldol reaction for the installation of
the chiral centres in the 2- and 3-positions.22 We have
previously reported the synthesis of compounds 26–28,
29a,b, 30–32 (Fig. 3) and now this approach was extended
to the synthesis of the 2,4-dideoxy analogues (4, 5, 7, 9,
11, 13, 19, 22) as described below.
Biological results
All compounds were tested for their potency as compe-
titive inhibitors with respect to substrate (6PG) against
6-phosphogluconate dehydrogenases of T. brucei and
sheep liver (Table 1 and Fig. 3). The compounds were
also assayed for their in vitro toxicity against blood-
stream form trypomastigotes (the clinically relevant
stage) of T. brucei rhodesiense. The compounds were
also assayed against two related parasites, T. cruzi
(intracellular amastigotes) and L. donovani (intracellular
amastigotes), responsible for Chagas’ disease and visc-
eral Leishmaniasis, respectively (Table 2).
Diastereoselective boron aldol reaction of aldehyde
(3S)-223 with the chiral oxazolidinones 1a and 1b affor-
ded (2R,3S)-3 (60%) and (2R,3S)-4 (47%), respectively.
Reductive desulfurization23,24 of 4 with Bu3SnH in
refluxing toluene afforded 5 in high yield (90%).
The benzyl esters 8 and 9 were prepared in two steps
from 4 and 5, respectively: the chiral auxiliary was first
removed with LiOH/H2O2,25 affording the pure acids 6
and 7 in good yield (86 and 78%, respectively) by acid-
base workup. The acids were subsequently esterified
with benzyl bromide and K2CO3 in DMF affording 8
and 9 (78 and 89%, respectively).
Table 2. In vitro antiparasitic activity
Compd T. brucei
T. cruzi
L. donovani Cytotoxicity L6-cells
IC50 (mM)
IC50 (mM)a IC50 (mM)b IC50 (mM)c
3
6
7
8
55.4
415
96.6
254
>306
51.1
354
146.4
91.4
11.6
8.2
156
80
81.2
336
>263
>369
>331
>239
>413
179.4
153.2
>306
>79.5
>138
>147
>97
>102
97.8
72
31.4
47.1
>42
35.9
Following the removal of the isopropylidene protecting
group with CuCl2/iPrOH, triol 10 and 11 were selec-
tively phosphorylated with dibenzylphosphoiodinate
[prepared from (BnO)3P and I2 in CH2Cl2]26 working at
À78 ꢀC. It is noteworthy that when the reaction was
carried out at higher temperature (i.e., À10 ꢀC to
+15 ꢀC over 2 h), the lactone product 24 was obtained
instead (13%). Besides, when 2.5 equivalents of phos-
9
10
11
12
13
14
15
16
17
19
20
21
22
23
24
25
26
27
28
29a
29b
30
31
32
>333
240
89.6
49.2
98.3
25.4
22.3
24.4
9.5
>54
dnp
dnp
Table 1. 6PGDH inhibitiona
24.9
>349
>116
>88
>123
>110
>71
dnp
>263
>369
>331
>214
8.5
>357
43.8
Compd
Sheep 6PGDH
Ki (mM)
T. brucei 6PGDH
Ki (mM)
6PGb
16
770
65
650
3.5
4.4
>1000
800
800
40.5
37.8
20.2
2-Deoxy-6PGc
38e
204.8
65
dnpd
31.4
29b
6
8
7.46
650
9
59.11 dnp
45
31.9
10
12
21
24
>1000
100
400
500
174
>1000
>1000
91.6
14
>372
>317
234
>57
dnp
>124
>106
>117
39.4
324.8
>372
>317
196
600
aAll compounds were assayed for inhibition of 6PGDH. Only com-
pounds displaying inhibition are shown; Lineweaver–Burk plots where
the standard errors of linear regressions were below 0.05 are an esti-
mate of the accuracy of the enzyme inhibition assays.
bKm values from ref 18b.
aControl: melarsoprol, IC50=0.0055 mM.
bControl: benznidazole, IC50=1.076 mM.
cControl: pentamidine, IC50=9.3 mM.
dDetermination not possible due to cytotoxicity on the host cell.
cKi values from ref 18a.
eMinimum inhibitory concentration (mg/mL).

C. Dardonville et al. / Bioorg. Med. Chem. 11 (2003) 3205–3214
3209
Inhibition of 6PGDH of T. brucei and sheep liver
enzymes
enzyme active site that is filled by substituents in the
(2R)-position, and substituents at this position should
increase the affinity of inhibitors [e.g., this is exempli-
fied in the T. brucei enzyme by comparison of the Km
values for 6PG (Km=3.5 mM) and 2-deoxy-6PG
(Km=9.6 mM)].18b
Among all the compounds assayed against sheep liver
and T. brucei 6PGDH enzymes, seven had a competitive
inhibitory activity, with a Ki in the micromolar range.
These results are reported in Table 1. Four compounds
selectively inhibited either the sheep 6PGDH (21, 24
and 29b) or the T. brucei enzyme (10). Moreover, three
compounds (6, 8 and 12) inhibited both enzymes with a
low selectivity index in favour of the sheep enzyme (1.2–
1.5). Three molecules (20, 22 and 30), which are the
most structurally related to 6PG, were not inhibitors of
the enzyme. Therefore, they were also tested as sub-
strates of the enzyme, but none of them displayed any
activity.
C3-Position
All the compounds tested in this study had the 3S ste-
reochemistry as in 6PG. The 3S-configuration is pre-
sumably preferred as previously reported for the
Candida enzyme in which the Ki for 6-phosphoallonate
(in which the stereochemistry at the C3 position is
inverted) is about 20-fold higher than Km for 6PG.20
More recently, Pasti et al. concluded that an oxygen at
C3 of 6PG, independent of the stereochemistry, was
fundamental for binding to the active site.17 Also the
presence of blocking groups on the 3-OH seemed to
reduce activity, implying the H-bonding interaction of
the 3-OH with the enzyme active site is important
(compare 12 with 14–18). However, the presence of an
acetate blocking group (H-bond acceptor) did not
always reduce activity (compare 15 and 12; 21 and 20).
Discussion
Comparison of data allows several important conclu-
sions to be drawn. However care has to be taken when
trying to draw structure–activity relationships, as inhi-
bitors may be binding in different orientations.
C4-Position
C5- and C6-Positions
The lack of inhibitory activity of the fully deprotected
compounds 20, 22 and 30, which are particularly close
analogues of 6PG is informative. The lack of inhibitory
activity of 22 is particularly interesting when consider-
ing that 2-deoxy-6-phosphogluconate is an excellent
inhibitor of the T. brucei enzyme [Ki (T. brucei)=4.4 mM]
and a moderate inhibitor of the sheep one [Ki
(sheep)=770 mM].18a The presence of a hydroxyl group at
C4 appears to be essential for the observed inhibitory
activity of 2-deoxy-6-phosphogluconate in the case of
the T. brucei enzyme. However the 4-OH may not be
strictly essential in the sheep enzyme as seen by the
inhibition given by the 3,5-acetate protected compound
21 (Ki=400 mM) in the sheep enzyme. The presence of
acetate protecting groups (H-bond acceptor) in C3 and
C5 instead of OH groups (H-bond acceptor/donor)
could account for the affinity and selectivity of this
molecule for the sheep 6PGDH active site.
Either a hydroxyl group or a large lipophilic group
seems to give activity at the 5-position. Compare the
activity of the (2R)-methyl-4-deoxy analogues 10, 12
and 21 with their non-active (2R)-methyl-4,5-dideoxy
counterparts 26, 27 and 31). Similarly, there is an
improvement of inhibition when adding the benzyl-pro-
tected phosphate moiety (compound 12, Ki
=174 mM) to the triol 10 (Ki=500 mM). These results
may reflect different binding modes compared to 6PG.
(T. brucei)
C1-Position
There is no clear data on SAR of our compounds at
position C1. Thus compounds 6 and 8 show similar
inhibition, despite one being a carboxylic acid (6) and
the other a benzyl ester (8). Compound 29b which car-
ries a methyl ester protecting group is highly selective
towards the sheep enzyme (Ki (sheep)=65 mM). Com-
pound 27, however, which is similar to compound 29b
but carries a bulky benzyl ester rather than a methyl
ester, has no inhibitory effect indicating that the binding
pocket is not large enough to accommodate the bulkier
substituents. Intriguingly, the presence of the 5-OH in
compound 12 allows this compound to participate in
some important interactions in both enzymes in spite of
the bulky benzyl ester at position C1. It is possible that
analogues do not bind in the manner predicted based on
their homology to 6PG.
Our results indicate that while the 4-OH is important
for the binding of the phosphate analogues of 6PG, it is
not absolutely essential in the case of protected analo-
gues since high micromolar level inhibition is seen with
several inhibitors of the 2-methyl-4-deoxy series (6, 8,
10, 12, 24 and 29b), which either have no phosphate or a
protected phosphate. This latter result may reflect a
different binding mode of the compounds in the enzyme
active site.
In vitro toxicity. The results of the in vitro toxicity on
three related trypanosomatid parasites (T. brucei, T.
cruzi and L. donovani) are reported in Table 2. Among
the 27 compounds that were assayed, three of them (27,
14 and 15) showed a moderate in vitro antiparasitic
activity against T.b. rhodesiense trypomastigotes
(<12 mM), two compounds (19, 25) had a moderate
C2-Position
No 2,4-dideoxy analogues (7, 9, 11, 13, 19, 22) inhibited
the enzyme, whilst the inhibitors have a (2R)-methyl
group at the 2-position (compare compounds 6 and 7; 8
and 9; 10 and 11; 12 and 13). These results may confirm
the hypothesis that there is a binding pocket in the

3210
C. Dardonville et al. / Bioorg. Med. Chem. 11 (2003) 3205–3214
toxicity against T. cruzi amastigotes (<10 mM) but none
had significant anti-leishmanial activity.
tids has made development of a single entity that treats
diseases caused by all three groups difficult, and also
makes interpretation on modes of action of toxicity
difficult.
The most active compounds against T.b. rhodesiense
were the (2R)-2-methyl protected analogues of 6PG.
Benzyl and especially acetate protection of the hydroxyl
groups in positions 3 and 5 (compounds 14 and 15,
respectively) appeared to be the most effective, with a
12- and 18-fold increase in activity, respectively, if
compared with the methyl protection (compound 16).
The benzylidene protection (compound 17) gave only a
moderate improvement of activity (IC50=80 mM) com-
pared to the unprotected 12 (IC50=146.4 mM). The fact
that the corresponding deprotected (2R)-2-methyl ana-
logues of 6PG (20 and 30) were totally inactive in vitro
probably reflects the highly charged character of these
molecules (i.e., phosphate and carboxylate moieties) at
physiological pH that would prevent them crossing the
cell membranes of the parasite. As a whole, all com-
pounds bearing unprotected phosphates (20–23, and
30–32) and free carboxylic acids (6, 7) showed poor in
vitro cytotoxicity.
Conclusion
Based on the results of inhibition of this series of com-
pounds several conclusions regarding interactions cru-
cial for binding could be made. First of all, there may be
a possibility of esters binding to the active site. Sec-
ondly, the 2-position appears more important to the
sheep enzyme than the T. brucei. Regarding the C3
position, blocking the hydroxyl groups probably redu-
ces activity. Regarding the 4-OH and 5-OH, they appear
important in binding to the T. brucei enzyme although
the 4-OH is probably less important in the case of the
sheep 6PGDH. Finally, blocked phosphate appeared to
show good activity compared to the unblocked analo-
gues. This could be due to a different mode of binding,
increased lipophilicity or a change in conformation of
the active site.
An interesting feature is the increase in activity of the
(2R)-2-methyl-4,5-dideoxy analogue 27 (IC50=7.5 mM)
compared with its corresponding acetylated analogue 28
(IC50=59 mM). On the contrary, the (2R)-2-methyl-4-
deoxy analogue 12 (146 mM) is less potent than its
acetylated counterpart 15 (IC50=8.2 mM); the only dif-
ference between the compounds 27, 28, 12, and 15 is the
presence of the 5-hydroxyl group. Variability in lipo-
philicity, facilitating uptake at the plasma membrane
may explain this difference.
Some compounds had trypanocidal activity although
correlations between anti-parasitic activity and enzyme
inhibition are not evident. Criteria involving drug-
uptake and other pharmacological events such as possi-
ble metabolic conversion limit interpretations that can
be made on cytotoxic effect until further studies are
performed.
To conclude, this work described the rational design of
substrate-based inhibitors of 6PGDH. The results
reported here open the route to the design of more effi-
cient inhibitors.
All the trypanosomatids have structurally similar
6PGDH enzymes and it is likely that they would
respond similarly to most inhibitors, although this
remains to be verified experimentally. Although there is
little correlation between anti-parasitic activity and
enzyme inhibition, the toxicity of the compounds could
be at least in part due to inhibition of 6PGDH,
although this needs to be verified experimentally. In
fact, the two most potent compounds against T. brucei
(14 and 15) are both protected analogues of 12 which is
the most potent 6PGDH inhibitor of the series. In order
to show anti-parasitic activity, the compounds must
cross the T. brucei plasma membrane and fully pro-
tected analogues of 12 could meet the pharmacokinetic
criteria to cross the membrane. Hence, metabolic con-
version of different products within the cell (for example
removal of protecting groups) could yield products with
differing inhibitory activity than the parents, thus lack
of potent 6PGDH inhibitory activity by a given com-
pound might not necessarily mean that the cytotoxic
effect is unrelated to inhibition of 6PGDH by a meta-
bolite of the parent compound.
Experimental
Chemistry
All reaction solvents were purchased anhydrous from
Aldrich Chemical Co. and used as received. Reactions
were monitored by TLC using pre-coated silica gel 60
F254 plates. Chromatography was performed with silica
gel (220–240 mesh) or with IsoluteTM pre-packed col-
umns. All reactions requiring anhydrous conditions or
an inert atmosphere were performed under a positive
pressure of N2. 1H NMR, 13C NMR and 31P NMR
spectra were recorded at 300, 75 and 121 MHz, respec-
tively.
The synthesis of compounds 2, 3, 6, 8, 10, 12, 14–18, 20,
21, 24 and 26–32 has been described elsewhere.21
(4R)-4-Benzyl-3-(2-methylsulfanyl-acetyl)-oxazolidin-2-
one (1b). Thionyl chloride (2 mL, 27 mmol) was added
drop wise to a solution of (methylthio)acetic acid
(2.38 g, 22.4 mmol) in Et2O (10 mL) and the reaction
was heated at ca. 40 ꢀC for 75 min. The volatiles were
removed in vacuo affording the (methylthio)acetic acid
chloride as a greenish oil that was used without further
For T. cruzi amastigotes, compounds must cross the
host cell membrane and then that of the parasites. In
the case of Leishmania the compounds must cross
three membranes, with that of the parasitophorous
vacuole joining those of the host and parasite sur-
faces. The variant biology of the different kinetoplas-

C. Dardonville et al. / Bioorg. Med. Chem. 11 (2003) 3205–3214
3211
purification. A 2.5 M solution of nBuLi in hexanes
(9 mL, 22.4 mmol) was added at À65 ꢀC to a solution of
(R)-4-benzyl-2-oxazolidinone (3.17 g, 22.4 mmol) in
THF (50 mL). After 5 min, (methylthio)acetic acid
chloride was added. The reaction mixture was stirred
30 min at the same temperature and 90 min at room
temperature. The reaction was quenched with 5%
NaHCO3 and the solvent was removed in vacuo. The
crude product was partitioned between CH2Cl2 and
water. The aqueous phase was extracted with CH2Cl2
(2Â) and the organic extracts were dried (MgSO4)
and concentrated. Chromatography (isolute 50 g SI)
with hexane/EtOAc: 0!30% afforded 1b as a yellow
oil (3.1g, 65% for two steps). Rf: 0.6 (50% EtOAc/
were eluted first with 100% hexane and the product 5
was eluted with hexane/EtOAc 25%. Colourless oil
(285 mg, 84%). H NMR (CDCl3) d 7.42–7.5 (m, 5H);
1
4.75 (m, 1H); 4.4 (m, 2H); 4.35–4.1 (m, 3H); 3.65 (t,
1H); 3.32 (m, 2H); 3.17 (m, 2H); 2.84 (dd, 1H,
J=10.4 Hz, J=13.5 Hz); 1.82 (t, 2H); 1.47 (s, 3H); 1.41
(s, 3H); 13C NMR (CDCl3) d 172.8 (Cq); 153.8 (Cq);
135.4 (Cq); 129.9 (CH); 129.4 (CH); 127.9 (CH); 109.2
(Cq); 73.7 (CH); 70.1(CH 2); 66.8 (CH2); 65.8 (CH); 55.4
(CH); 43.5 (CH2); 40.5 (CH2); 38.2 (CH2); 27.4 (CH3);
22
26.2 (CH3); ½aꢁD =À13.5 (c 0.74, CH2Cl2); MS (ES+)
m/z 381[M+NH 4]; ESHRMS m/z 364.1756 [M+H]
(C19H26NO6 requires: 364.1760).
1
hexane); H NMR (CDCl3) d 7.45–7.25 (m, 5H); 4.77
(3S)-4-[(4S)-2,2-Dimethyl-1,3-dioxolan-4-yl]-3-hydroxy-
butanoic acid (7). A 0.05 M solution of aldol 5 (1.06 g,
2.9 mmol) in a 3/1THF/H 2O mixture (45 mL/15 mL)
was treated at 0 ꢀC with a 30% H2O2 (1.5 mL,
(m, 1H); 4.35–4.2 (m, 2H); 3.9 (dd, 2H, J=14.2 Hz,
J=22.8 Hz); 3.38 (dd, 1H, J=3.4 Hz, J=13.4 Hz); 2.85
(dd, 1H, J=9.6 Hz, J=13.4 Hz); 2.27 (s, 3H). Anal.
calcd for C13H15NO3S: C, 58.85; H, 5.70; N, 5.28; S,
12.08; found: C, 58.72; H, 5.54; N, 4.95; S, 11.90.
.
14.6 mmol) and LiOH H2O (185 mg, 4.4 mmol). After
2 h at 0 ꢀC, the reaction was quenched with 1.5 M
Na2SO3 aqueous solution (11.5 mL, 17 mmol) and buf-
fered with a 5% NaHCO3 aqueous solution (5 mL).
THF was removed in vacuo and the aqueous solution
was diluted with water and extracted with CH2Cl2
(2Â50 mL). The aqueous phase was acidified to pH ꢂ2
with 2% HCl and extracted with EtOAc (3Â50 mL).
Combined ethyl acetate extracts were dried (Na2SO4)
and concentrated to yield the pure acid 17 as a colour-
less oil (466 mg, 78%). 1H NMR (CDCl3) d 7.0 (br, 2H);
4.3–4.15 (m, 2H); 4.02 (dd, 1H, J=6.2 Hz, J=8.4 Hz);
3.5 (t, 1H, J=7.7 Hz); 2.5 (m, 2H); 1.65 (t, 2H); 1.32 (s,
3H); 1.29 (s, 3H); 13C NMR (CDCl3) d 176.9 (Cq); 109.5
(Cq); 73.6 (CH); 69.9 (CH2); 65.9 (CH); 41.9 (CH2); 40.2
(4R)-4-Benzyl-3-(2R,3S)-4-[(4S)-2,2-dimethyl-1,3-dioxo-
lan-4-yl)-3-hydroxy-(2R)-2-methylsulfanylbutanoyl]-1,3-
oxazolan-2-one (4). A solution of 1b (2.33 g, 8.8 mmol)
in dry CH2Cl2 (15 mL) cooled to À10 ꢀC was treated
sequentially with dibutylboron triflate (2.4 mL,
9.6 mmol) and DIPEA (2.1mL, 12 mmol). The yellow
reaction mixture was stirred for 2 h at ca. À5 ꢀC. The
clear orange reaction mixture was cooled to À78 ꢀC and
a solution of aldehyde 12 (1.53 g, 8.6 mmol) in CH2Cl2
(10 mL) was added drop wise over a 5-min period. The
reaction was allowed to gradually warm up to room
temperature overnight and was quenched with phos-
phate buffer (pH 7.2, 30 mL). The resulting mixture was
stirred 4 h at room temperature. The organic phase was
separated and the aqueous phase was extracted with
CH2Cl2 (2Â). The combined organic phases were dried
(MgSO4) and concentrated to give a crude orange oil.
Chromatography with hexane/EtOAc: 25!30% affor-
ded 4 as a yellow oil (1.54 g, 47%) and some recovered
22
(CH2); 27.3 (CH3); 26.0 (CH3); ½aꢁD =À7.97 (c 1.38,
MeOH); MS (ES-) m/z 203 [MÀH]; ESHRMS m/z
222.1340 [M+NH4] (C9H20NO5 requires: 222.1341).
Benzyl (3S)-4-[(4S)-2,2-dimethyl-1,3-dioxolan-4-yl]-3-hy-
droxybutanoate (9). A solution of acid 7 (390 mg,
1.9 mmol), benzyl bromide (0.34 mL, 2.8 mmol) and
K2CO3 (340 mg, 2.3 mmol in DMF (12 mL) was stirred
23 h at room temperature. The reaction was diluted with
CH2Cl2 and filtered on a path of Celite. The filtrate was
concentrated in vacuo and the crude residue was pur-
ified by flash chromatography (Isolute 10 g SI) with
hexane/EtOAc: 0!100%, affording the ester 9 as a col-
ourless oil (500 mg, 89%). Rf : 0.45 (50% EtOAc in
1
oxazolidinone 1b (1g); H NMR (CDCl3) d 7.4–7.2 (m,
5H); 4.72 (m, 2H); 4.4–4.0 (m, 5H); 3.55 (dd, 1H,
J=7 Hz, J=8.2 Hz); 3.4 (s, 1H); 3.2 (dd, 1H,=3.4 Hz,
J=13.4 Hz); 2.8 (dd, 1H, J=9.4 Hz, J=13.4 Hz); 2.17
(s, 3H); 1.8 (t, 2H); 1.37 (s, 3H); 1.3 (s, 3H); 13C NMR
(CDCl3) d 171.4 (Cq); 170.4 (Cq); 153.2 (Cq); 135.5 (Cq);
129.9 (CH); 129.3 (CH); 127.8 (CH); 109.1 (Cq); 73.9
1
(CH); 70.1(CH ); 66.7 (CH); 66.3 (CH2); 55.2 (CH);
2
hexane); H NMR (CDCl3) d 7.4 (s, 5H); 5.2 (s, 2H);
50.4 (CH); 38.9 (CH2); 37.9 (CH2); 27.5 (CH3); 26.0
4.45–4.25 (m, 2H); 4.13 (dd, 1H, J=6 Hz, J=8.1Hz);
3.61(t, 1H, J=7.2 Hz); 3.3 (m, 1H); 2.75–2.55 (m, 2H);
1.77 (t, 1H, J=6.5 Hz); 1.45 (s, 3H); 1.41 (s, 3H); 13C
NMR (CDCl3) d 172.8 (Cq); 135.9 (Cq); 129.1 (CH);
128.9 (CH); 128.8 (CH); 109.3 (Cq); 73.7 (CH); 70.0
19
(CH3); 13.5 (CH3); ½aꢁD =À44.8 (c 0.58, CH2Cl2); MS
(ES+) m/z 427 [M+NH4]; ESHRMS m/z 410.1634
[M+H] (C20H28NO6S requires: 410.1637). Anal. calcd
for C20H27NO6S: C, 58.66; H, 6.65; N, 3.42; S, 7.83;
found: C, 58.43; H, 6.80; N, 3.32; S, 7.45.
(CH2); 67.0 (CH2); 66.0 (CH); 42.1(CH ); 40.3 (CH2);
2
23
27.4 (CH3); 26.1(CH ); ½aꢁ =À4.58 (c 2.4, MeOH);
3
D
(4R)-4-Benzyl-3-(3S)-4-[(4S)-2,2-dimethyl-1,3-dioxolan-
4-yl]-3-hydroxybutanoyl-1,3-oxazolan-2-one
MS (ES+) m/z 295 [M+H]; ESHRMS m/z 295.1545
[M+H] (C16H23O5 requires: 295.1545).
(5).
A
solution of 4 (380 mg, 0.9 mmol), tributyltin hydride
(0.37 mL, 1.4 mmol) and 1,10-azobis(cyclohexane-
carbonitrile) (93 mg, 0.4 mmol) in benzene (18 mL) was
refluxed (bath temperature 85 ꢀC) for 3 h. The solvent
was removed in vacuo and the residue was purified by
flash chromatography (Isolute 5 g SI): mobile impurities
Benzyl (3S,5S)-3,5,6-trihydroxyhexanoate (11). A 0.1M
solution of 9 (400 mg, 1.36 mmol) in iPrOH (14 mL) was
.
treated with CuCl2 2H2O (1.15 g, 6.8 mmol). The result-
ing green reaction mixture was stirred at room tem-
perature for 2 h 45 min. The reaction was quenched with

3212
C. Dardonville et al. / Bioorg. Med. Chem. 11 (2003) 3205–3214
1 M Na CO3 (7.5 mL). After 10 min stirring, iPrOH was
2
(20 mg) in MeOH (5 mL) was hydrogenated 30 min at rt.
The catalyst was filtered off and the solvent was
removed in vacuo. The crude residue was dissolved in
THF (2 mL) and 1M aqueous HCl (2 mL) was added.
The reaction was stirred 2 h at room temperature and
added and the reaction mixture was filtered on a path of
Celite. The filter cake was rinsed with iPrOH. The fil-
trate was concentrated in vacuo and the resulting resi-
due was dissolved in EtOAc. Insoluble inorganic salts
were filtered off. The filtrate was dried (MgSO4) and
concentrated. Chromatography (Isolute 5 g SI) with
hexane/EtOAc: 0!100% yielded 11 as a colourless oil
(130 mg, 38%); Rf : 0.23 (100% EtOAc); 1H NMR
(CDCl3) d 7.4 (br s, 5H); 5.16 (s, 2H); 4.5–4.3 (m, 3H);
4.15 (m, 1H); 3.61 (m, 1H); 3.49 (m, 1H); 2.7–2.47 (m,
2H); 1.56 (t, 2H); 13C NMR (CDCl3) d 172.9; 136.0;
129.0; 128.8; 128.7; 69.4; 67.3; 67.0; 65.3; 42.4; 39.5;
1
the solvent was evaporated to give a brownish oil. H
NMR (D2O) d 4.3–3.5 (m, 4H); 2.7–2.2 (m, 2H); 1.7–1.2
(m, 2H); 13C NMR (D2O) d 176.2 (Cq); 70.4 (CH2);
66.95 (d, CH, J=7.5 Hz); 64.8 (CH); 42.5 (CH2); 39.0
(CH2); 31P NMR (D2O) d+1.39; MS (ES+) m/z 267
[M+Na];
ESHRMS
m/z
267.0257
[M+Na]
(C6H13O8NaP requires: 267.0246).
22
½aꢁD =À6.67 (c 0.9, MeOH); MS (ES+) m/z 272
(3S,5S)-3,5-Dihydroxy-(2R)-2-methyl-6-phosphonooxy-
hexanoic acid methyl ester (23). A suspension of 18
(50 mg) and 5% Pd/C (10 mg) in MeOH (5 mL) was
hydrogenated 30 min at rt. The catalyst was filtered
off and the solvent was removed in vacuo. The
crude residue was dissolved in MeOH/HCl (5 mL)
and the reaction was stirred 2 h at rt. The solvent was
evaporated and the residue was dried under high-
vacuum overnight. The residue was dissolved in
MeOH and the white solid that precipitated was fil-
tered off. Concentration of the filtrate gave the
[M+NH4]; ESHRMS m/z 255.1234 [M+H] (C13H19O5
requires: 255.1232). Anal. calcd for C13H18O5/ 0.6H2O:
C, 58.90; H, 7.30; found: C, 59.01; H, 7.31.
6-(Bis-benzyloxy-phosphoryloxy)-(3S,5S)-3,5-dihydroxy-
hexanoic acid benzyl ester (13). A solution of tribenzyl-
phosphite (360 mg, 1mmol) and iodine (236 mg,
0.9 mmol) in CH2Cl2 (2 mL) was stirred 10 min at 0 ꢀC
and 15 min at room temperature. A part of this yellow
solution (1.2 mL) was added dropwise at À78 ꢀC to a
stirred solution of 11 (125 mg, 0.49 mmol) and pyridine
(0.12 mL, 1.5 mmol) in CH2Cl2 (2 mL). The reaction was
stirred 2 h and another 0.2 mL of the dibenzylphos-
phoiodinate solution was added. After 1h at the same
temperature, the reaction was filtered and partitioned
between CH2Cl2 and dilute HCl solution. The organic
phase was dried (MgSO4) and concentrated to give a
crude yellow oil. Chromatography (Isolute 5 g SI) with
hexane/EtOAc: 0!100% afforded 13 as a colourless oil
1
methyl ester 23 as a colourless oil (24 mg); H NMR
(CD3OD) d 4.2–3.8 (m, 4H); 3.7 (s, 3H); 2.52 (quint,
1H); 1.55 (m, 2H); 1.18 (d, 3H, J=6.8 Hz); 13C NMR
(CD3OD) d 177.4 (Cq); 72.2 (CH2); 70.2 (CH); 68.6 (d,
CH, J=7.5 Hz); 52.6 (CH3); 47.8 (CH); 39.4 (CH2);
12.7 (CH3); 31P NMR (CD3OD)
d
À1.89 (br);
18
½aꢁD =À18.7 (c 0.1, MeOH); MS (ESÀ) m/z 271[M ÀH];
HRMS (ESÀ) m/z 271.0579 (C8H17O8P requires:
271.0583).
1
(125 mg, 50%). H NMR (CDCl3) d 7.4 (br s, 15H); 5.2
(s, 2H); 5.1(m, 2H); 5.06 (m, 2H); 4.44 (m, 1H); 4.2–3.9
(m, 3H); 3.8–3.5 (br, 2H); 2.57 (m, 2H); 1.6 (t, 2H); 13C
NMR (CDCl3) d 172.8 (Cq); 136.1 (Cq); 136.0 (Cq);
129.1 (2CH); 128.8 (CH); 128.7 (CH); 128.5 (CH); 72.2
(d, CH2, J=7.5 Hz); 70.1(d, CH 2, J=7.5 Hz); 67.8
(CH); 67.0 (CH2); 65.4 (CH); 42.0 (CH2); 38.7 (CH2);
(3S)-3,6-Bis-(bis-benzyloxy - phosphoryloxy) - (5S) - 5 -
hydroxy-(2R)-2-methyl-hexanoic acid benzyl ester (25).
Iodine (908 mg/3.6 mmol) was added at 0 ꢀC to a solu-
tion of tribenzylphosphite (1.3 g, 3.7 mmol) in CH2Cl2
(10 mL). After 10 min at 0 ꢀC and 10 min at room
temperature, the red solution was added drop wise at
À78 ꢀC to a solution of 10 (400 mg, 1.49 mmol) and
pyridine (0.6 mL, 7.5 mmol) in CH2Cl2 (6 mL). The
reaction was stirred 2 h 15 min at À78 ꢀC, diluted
with CH2Cl2, filtered, washed with water, dried
(MgSO4) and concentrated in vacuo. The crude oil
was chromatographed (Isolute 20 g SI) with hexane/
EtOAc: 25!100% to give the diphosphate as a light
yellow oil. The product dissolved in CH2Cl2 was treated
with activated charcoal and filtered. Removal of the
23
31P NMR (CDCl3) d+0.92; ½aꢁD =À3.7 (c 0.54,
MeOH); MS (ES+) m/z 532 [M+NH4], 515 [M+H];
ESHRMS m/z 515.1839 [M+H] (C27H32O8P requires:
515.1835). Anal. calcd for C27H31O8P/0.9H2O: C, 61.10;
H, 6.23; P, 5.84; found: C, 61.06; H, 5.69; P, 5.66.
6-(Bis-benzyloxy-phosphoryloxy)-(3S,5S)-3,5-bis-(tetra-
hydro-pyran-2-yloxy)-hexanoic acid benzyl ester (19). A
solution of 13 (65 mg, 0.13 mmol), 3,4-dihydro-2H-
pyran (0.2 mL, 2.5 mmol) and PPTS (12 mg, 0.05 mmol)
in CH2Cl2 (2 mL) was stirred at room temperature for
48 h. The volatiles were removed in vacuo and the resi-
due was chromatographed (Isolute 1g SI) with hexane/
solvent in vacuo yielded 25 as
a colourless oil
(715 mg, 61%); 1H NMR (CDCl3) d 7.35 (br s,
25H); 5.18 (s, 2H); 5.12–4.98 (m, 8H); 4.75 (m,
1H); 4.1 (m, 2H); 3.95 (m, 2H); 2.55 (m, 1H);
1.75–1.55 (m, 2H); 1.25 (d, 3H, J=7.2 Hz); 13C NMR
(CDCl3) d 175.2 (Cq); 136.3 (Cq); 136.1 (Cq); 136.0
(Cq); 129.0 (CH); 128.7 (CH); 128.6 (CH); 128.4
(3CH); 74.9 (CH); 70.2 (d, CH2); 69.85 (d, CH2); 69.6
(CH2); 67.4 (CH); 66.8 (CH2); 45.6 (CH); 36.8 (CH2);
12.6 (CH3); 31P NMR (CDCl3) d+1.13;+0.24;
ESHRMS m/z 806.2860 [M+NH4] (C42H50NO11P2
requires 806.2859). Anal. calcd for C42H46O11P2: C,
63.96; H, 5.88; P, 7.85; found: C, 63.93; H, 6.00; P:
7.81.
1
EtOAc: 0!70% to yield 19 as an oil (72 mg, 81%). H
NMR (acetone-d6) d 7.5–7.3 (m, 15H); 5.2–5.0 (m, 6H);
4.8 (m, 1H); 4.67 (m, 1H); 4.4–3.75 (m, 6.5H); 3.4 (m,
2H); 2.7–2.35 (m, 1.5H); 1.9–1.3 (m, 14H); 31P NMR
d+0.574;+0.501;+0.379;+0.342; MS (ES+) m/z 705
[M+Na];
ESHRMS
m/z
705.2789
[M+Na]
(C37H47O10NaP requires: 705.2805).
(3S,5S)-3,5 - Dihydroxy - 6 - phosphonooxy-hexanoic acid
(22). A suspension of 19 (55 mg) and 5% Pd/C28

C. Dardonville et al. / Bioorg. Med. Chem. 11 (2003) 3205–3214
3213
Biological tests
excitation wavelength of 530 nm and emission wavelength
of 590 nm. Fluorescence development was expressed as
percentage of the control, and IC50 values determined.
6PGDH enzyme assay. The recombinant T. brucei
6PGDH, overexpressed in Escherichia coli, was purified
by a technique which was slightly modified compared to
the original of Barrett.9 Cells were lysed by freeze–thaw
cycling using liquid nitrogen and then resuspended in
50 mM triethanolamine/HCl containing 1mM EDTA,
pH 7.5 (TEA buffer). Cells were then centrifuged at
39,000 rpm for 30 min in a Beckman XL-70 ultra-
centrifuge using a Ti70 rotor. The supernatant was
applied to a 15-mL DEAE-Sepharose column equili-
brated with TEA buffer, then washed with the same
buffer, and the flowthrough material absorbing at
280 nm was loaded directly onto a 5 mL 20,50-ADP-
Sepharose column, equilibrated with TEA buffer. After
washing, the enzyme was eluted by a TEA buffer con-
taining 0.5 M NaCl and the specific activity assayed in
the presence of 1.5 mM 6PG and 0.25 mM NADP+.
The whole purification lasted less than one day and was
monitored both by SDS-PAGE and activity assays.
Enzyme was stored in the presence of 50% glycerol
at À20 ꢀC. The sheep liver 6PGDH was purified as
reported.18 Activity of the enzymes at 20 ꢀC is followed
spectrophotometrically (Kontron Uvikon 930 spectro-
photometer) at 340 nm, measuring the production of
NADPH in absence or presence of inhibitor.
In vitro pharmacology, T. cruzi. Rat skeletal myoblasts
(L-6 cells) were seeded in 96-well microtiter plates at
2000 cells/well/100 mL in RPMI 1640 medium with 10%
FBS and 2 mM l-glutamine. After 24 h, 5000 trypo-
mastigotes of T. cruzi (Tulahuen strain C2C4 containing
the galactosidase (Lac Z gene) were added in 100 mL per
well with 2Â of a serial drug dilution. The plates were
incubated at 37 ꢀC in 5% CO2 for 4 days. After 96 h, the
minimum inhibitory concentration (MIC) was deter-
mined microscopically. For measurement of the IC50
the substrate CPRG/Nonidet was added to the wells.
The colour reaction which developed during the fol-
lowing 2–4 h was read photometrically at 540 nm. From
the sigmoidal inhibition curve IC50 values were calcu-
lated. Cytotoxicity was assessed in the same assay
using non-infected L-6 cells and the same serial drug
dilution. The MIC was determined microscopically
after 4 days.
In vitro pharmacology, L. donovani. Mouse peritoneal
macrophages were seeded in RPMI 1640 medium with
10% heat-inactivated FBS into Lab-tek 16-chamber
slides. After 24 h L. donovani amastigote were added at
a ratio of 3:1(amastigotes to macrophages). The med-
ium containing free amastigotes was replaced by fresh
medium 4 h later. Next day, the medium was replaced
by fresh medium containing different drug concentra-
tions. The slides were incubated at 37 ꢀC under a 5%
CO2 atmosphere for 96 h. Then the medium was
removed, the slides fixed with methanol and stained
with Giemsa. The ratio of infected to non-infected
macrophages was determined microscopically, expres-
sed as percentage of the control and the IC50 value cal-
culated by linear regression.
Compounds to assay as inhibitors were dissolved in
either methanol, DMSO or water. Inhibition type and
degree was evaluated by performing the assays in TEA
buffer, pH 7.5 at variable 6PG concentration (8–40 mM),
whilst keeping NADP+ concentration fixed to 0.26 mM.
When lack of solubility of the compound did not allow
to assay a high inhibitor concentration in the conditions
described above, enzyme activity was assayed in TEA
buffer in the presence of either 20% methanol or 10 or
20% DMSO. The concentration of inhibitors was fixed
and the reaction was initiated by the addition of either
0.2 mg T. brucei 6PGDH or 0.5 mg sheep liver 6PGDH
per mL reaction mixture. The mean Ki values were cal-
culated by Lineweaver–Burk plots where the standard
errors of linear regressions were below 0.05 and after
experiments were repeated several times. The selectivity
degree is valued as the ratio between the mammal and
parasite enzymes Ki values.
Acknowledgements
We gratefully acknowledge the Wellcome Trust (CD)
and the European Union (E.R., S.H. and M.P.B.; con-
tract No. IC18CT98-0357) for financial support and the
EPSRC National Mass Spectrometry Service Centre,
Swansea, for accurate mass spectra.
Compounds 20 and 22 were assayed as substrates in the
presence of 0.25 mM NADP+ in TEA buffer, pH 7.5, at
concentrations of 2.6 and 3.4 mM, respectively, against
both sheep and T. brucei enzymes.
References and Notes
1. WHO/CDS/CSR/ISR/2000.1, http://www.who.int/emc.
2. http://www.who.int/tdr/diseases/tryp/diseaseinfo.htm.
3. World Health Organ. Tech. Rep. Ser. 1998, 881, 1.
4. (a) De Koning, H. P. Int. J. Parasitol. 2001, 31, 512. (b)
Denise, H.; Barrett, M. P. Biochem. Pharmacol. 2001, 61, 1 .
5. Keiser, J.; Stich, A.; Burri, C. TRENDS Parasitol. 2001, 17,
42.
6. (a) Opperdoes, F. R. Biochem. Soc. Trans. 1990, 18, 729.
(b) Opperdoes, F. R.; Michels, P. A. M. Int. J. Parasitol. 2001,
31, 482.
In vitro pharmacology, T.b. rhodesiense. Minimum
Essential Medium (50 mL) supplemented with 2-mer-
captoethanol and 15% heat-inactivated horse serum
was added to each well of a 96-well microtiter plate.
Serial drug dilutions were added to the wells. Then
50 mL of trypanosome suspension (T.b. rhodesiense
STIB 900) was added to each well and the plate incu-
bated at 37 ꢀC under a 5% CO2 atmosphere for 72 h.
Alamar Blue (10 mL) was then added to each well and
incubation continued for a further 2–4 h. The plate was
then read with a Millipore Cytofluor 2300 using an
7. Barrett, M. P. Parasitol. Today 1997, 13, 1 1 .
8. Barrett, M. P.; Le Page, R. W. F. Mol. Biochem. Parasitol.
1993, 57, 89.

3214
C. Dardonville et al. / Bioorg. Med. Chem. 11 (2003) 3205–3214
9. Barrett, M. P.; Phillips, C.; Adams, M. J.; LePage, R. W. F.
Protein Expres. Purif. 1994, 5, 44.
10. Krepinsky, K.; Plaumann, M.; Martin, W.; Schnarren-
berger, C. Eur. J. Biochem. 2001, 268, 2678.
11. Phillips, C.; Dohnalek, J.; Gover, S.; Barrett, M. P.;
Adams, M. J. J. Mol. Biol. 1998, 282, 667.
21. Details of the development of our synthetic approach for
the preparation of the (2R)-2-methyl-4-deoxy analogues
(compounds 2, 3, 6, 8, 10, 12, 14–18, 20, 21) and (2R)-2-
methyl-4,5-dideoxy analogues (compounds 26–32) have been
described elsewhere: Dardonville, C.; Gilbert, I. H. Org. Bio-
mol. Chem. 2003, 1, 552.
12. Barrett, M. P.; Gilbert, I. H. Curr. Topics Med. Chem.
2002, 2, 471.
13. Marchand, M.; Kooystra, U.; Wierenga, R. K.; Lambeir,
A.-M.; Beeumen, J. V.; Opperdoes, F. R.; Michels, P. A. Eur.
J. Biochem. 1989, 184, 455.
14. Lobo, Z.; Maitra, P. K. Mol. Gen. Genet. 1982, 185, 367.
15. Gvozdev, V. A.; Gerasimova, T. I.; Kogan, G. L.; Bra-
slavskaya, O. Yu. FEBS Lett. 1976, 64, 85.
22. Evans, D. A. Aldrichim. Acta 1982, 15, 23.
23. (a) Meyers, A. I.; Lawson, J. P.; Walker, D. G.; Linder-
man, R. J. J. Org. Chem. 1986, 51, 5111. (b) Hanessian, S.;
Ugolini, A.; Therien, M. J. Org. Chem. 1983, 48, 4427. (c)
Boger, D. L.; Curran, T. T. J. Org. Chem. 1992, 57, 2235.
24. (a) Ueno, Y.; Miyamo, T.; Okawara, M. Tetrahedron Lett.
1982, 23, 443. (b) Vedejs, E.; Powell, D. W. J. Am. Chem. Soc.
1982, 104, 2046.
16. (a) Phillips, C.; Gover, S.; Adams, M. J. Acta Cryst. 1995,
D51, 290. (b) Adams, M. J.; Ellis, G. H.; Gover, S.; Naylor,
C. E.; Phillips, C. Structure 1994, 2, 651.
17. Pasti, C.; Rinaldi, E.; Cervellati, C.; Dallocchio, F.; Hardre,
R.; Salmon, L.; Hanau, S. Bioorg. Med. Chem. 2003, 11, 1207.
18. (a) Hanau, S.; Rippa, M.; Bertelli, M.; Dallocchio, F.;
Barrett, M. P. Eur. J. Biochem. 1996, 240, 592. (b) Rippa, M.;
Giovannini, P. P.; Barrett, M. P.; Dallocchio, F.; Hanau, S.
Biochim. Biophys. Acta 1998, 1429, 83.
25. Evans, D. A.; Britton, T. C.; Ellman, J. A. Tetrahedron
Lett. 1987, 28, 6141.
26. (a) Gefflaut, T.; Lemaire, M. Valentin, M-L., Bolte, J. J.
Org. Chem. 1997, 62, 5920. (b) Kis, K.; Wungsintaweekul, J.;
Eisenreich, W.; Zenk, M. H.; Bacher, A. J. Org. Chem. 2000,
65, 587.
27. It should be noted that compounds 20 and 22 could not be
obtained in a pure form directly by hydrogenolysis of 12 and
13, respectively. Phosphate migration ability of this kind of
molecules has been discussed in another article. See ref 21.
28. We used 5% Pd/C from Fluka Chemical company to
avoid early deprotection of the THP. Kaisalo, L. H.; Hase,
T. A. Tetrahedron Lett. 2001, 42, 7699.
19. Bertelli, M.; El-Bastawissy, E.; Knaggs, M. H.; Barrett,
M. P.; Hanau, S.; Gilbert, I. H. J. Comput. Aid. Mol. Des.
2001, 15, 465.
20. Berdis, A. J.; Cook, P. F. Biochemistry 1993, 32, 2041.