Synthesis and evaluation of malonate-based inhibitors of phosphosugar-metabolizing enzymes: Class II fructose-1,6-bis-phosphate aldolases, type i phosphomannose isomerase, and phosphoglucose isomerase

Article abstract of DOI:10.1016/j.bmc.2011.12.050

In the design of inhibitors of phosphosugar metabolizing enzymes and receptors with therapeutic interest, malonate has been reported in a number of cases as a good and hydrolytically-stable surrogate of the phosphate group, since both functions are dianio

Full text of DOI:10.1016/j.bmc.2011.12.050

Bioorganic & Medicinal Chemistry 20 (2012) 1511–1520
Contents lists available at SciVerse ScienceDirect
Bioorganic & Medicinal Chemistry
journal homepage: www.elsevier.com/locate/bmc
Synthesis and evaluation of malonate-based inhibitors of
phosphosugar-metabolizing enzymes: Class II fructose-1,6-bis-phosphate
aldolases, type I phosphomannose isomerase, and phosphoglucose isomerase
Stéphanie Desvergnes , Stéphanie Courtiol-Legourd , Racha Daher, Maciej Dabrowski,
⇑
⇑
Laurent Salmon , Michel Therisod
Univ. Paris-Sud, Laboratoire de Chimie Bioorganique et Bioinorganique, ICMMO, UMR8182, LabEx LERMIT, Orsay F-91405, France
CNRS, Orsay F-91405, France
a r t i c l e i n f o
a b s t r a c t
Article history:
In the design of inhibitors of phosphosugar metabolizing enzymes and receptors with therapeutic inter-
est, malonate has been reported in a number of cases as a good and hydrolytically-stable surrogate of the
phosphate group, since both functions are dianionic at physiological pH and of comparable size. We have
investigated a series of malonate-based mimics of the best known phosphate inhibitors of class II (zinc)
fructose-1,6-bis-phosphate aldolases (FBAs) (e.g., from Mycobacterium tuberculosis), type I (zinc) phos-
phomannose isomerase (PMI) from Escherichia coli, and phosphoglucose isomerase (PGI) from yeast. In
the case of FBAs, replacement of one phosphate by one malonate on a bis-phosphorylated inhibitor (1)
led to a new compound (4) still showing a strong inhibition (Ki in the nM range) and class II versus class
I selectivity (up to 8 ꢀ 104). Replacement of the other phosphate however strongly affected binding effi-
Received 27 September 2011
Revised 20 December 2011
Accepted 22 December 2011
Available online 2 January 2012
Keywords:
Enzyme inhibitor
Hydroxamate
Malonate
Phosphate
Monosaccharide
ciency and selectivity. In the case of PGI and PMI, 5-deoxy-5-malonate-
yielded a strong decrease in binding affinities when compared to its phosphorylated parent compound
5-phospho- -arabinonohydroxamic acid (2). Analysis of the deposited 3D structures of the kinetically
D-arabinonohydroxamic acid (8)
D
evaluated enzymes complexed to the phosphate-based inhibitors indicate that malonate could be a good
phosphate surrogate only if phosphate is not tightly bound at the enzyme active site, such as in position 7
of compound 1 for FBAs. These observations are of importance for further design of inhibitors of phos-
phorylated-compounds metabolizing enzymes with therapeutic interest.
Ó 2012 Elsevier Ltd. All rights reserved.
1. Introduction
recognize the phosphoester group have been considered as attrac-
tive therapeutic targets. Particularly, because of their implication
Number of enzymes act on phosphorylated substrates or cata-
lyze the transfer of phosphate groups to/from various metabolic
intermediates. Abnormal levels of phosphorylated biological enti-
ties in mammals have been linked to a wide range of diseases such
as cancer, diabetes, atherosclerosis, immunodeficiency, cystic
fibrosis (or mucoviscidosis) and inflammation. In addition, viru-
lence and development of pathogenic microorganisms (bacteria,
protozoae) have been shown to partly or entirely rely on levels
of phospho-entities. Consequently, proteins and enzymes that
in key metabolic pathways, phosphosugar-metabolizing enzymes
such as 6-phosphofructo-2-kinase,1 phosphoglucose isomerase,2
phosphomannose isomerase,3 3-phosphoglycerate kinase,4 glycer-
aldehyde-3-phosphate dehydrogenase,5 fructose-1,6-bisphosphate
aldolase (FBA),5,6 and glutamine:fructose-6-phosphate amido-
transferase7 have been considered for the clinical development of
chemotherapeutic agents, under the form of enzyme inhibitors.
Indeed, selective inhibitors of most of the fore-mentioned en-
zymes have been prepared, which have been proved to be effi-
ciently active in vitro. Very disappointing results however were
most often obtained when these compounds were tested for an
in vivo activity. For example, we previously reported the synthesis
of potent and selective inhibitors of class II (zinc dependent)
FBA.8–10 These compounds have a potential therapeutic interest
as antibiotics, since class II FBA is present in many pathogenic
bacteria (Yersinia pestis, Mycobacterium tuberculosis, Clostridium dif-
ficile, Pseudomonas aeruginosa...), yeasts (Candida albicans...), proto-
zoae (Giardia lamblia...) and absent from human where class I FBA
Abbreviations: 5PAH, 5-phospho-
6-phosphate; FBA, fructose-1,6-bis-phosphate aldolase; FBP,
phosphate; G6P, -glucose 6-phosphate; HEI, high energy intermediate; M6P,
mannose-6-phosphate; b-M6P, b- -mannopyranose 6-phosphate; PDB, Protein
D
-arabinonohydroxamic acid; F6P,
D-fructose
D-fructose 1,6-bis-
D
D-
D
Data Bank; PGI, phosphoglucose isomerase; PMI, phosphomannose isomerase.
⇑
Corresponding authors. Tel.: +33 1 69 15 63 11; fax: +33 1 69 15 72 81.
E-mail addresses: laurent.salmon@u-psud.fr (L. Salmon), michel.therisod@u
-psud.fr (M. Therisod).
These authors contributed equally to this work.
0968-0896/$ - see front matter Ó 2012 Elsevier Ltd. All rights reserved.
doi:10.1016/j.bmc.2011.12.050

1512
S. Desvergnes et al. / Bioorg. Med. Chem. 20 (2012) 1511–1520
operates. They were rationally designed as transition-state ana-
logues of the catalyzed reaction of cleavage of -fructose 1,6-bis-
D
phosphate (FBP), and bear a zinc-chelating group to induce a selec-
tivity for class II FBA (Fig. 1).
Compound 1 (and others reported in Refs. 8–10) presents inter-
esting inhibitory power with Ki in the nanomolar range and selec-
tivity up to 105 for microbial enzymes. Crystallographic studies
also confirmed that 1 is tightly bound into the active site of class
II FBA.10 However, when 1 and its congeners were assayed
in vivo against cultivated pathogens (M. tuberculosis, Y. pestis, C.
albicans), no growth inhibition could be observed at concentrations
up to 1 mM (there are strong evidence that at least in M. tubercu-
losis, class II fba is an essential gene). This negative result is not
so surprising in view of the polarity of these phosphorylated prod-
ucts, unlikely to spontaneously cross (microbial) cytoplasmic
membranes. Moreover, phosphatases are ubiquitously present,
which can dephosphorylate 1 before it can reach its target. Simi-
Figure 2. D-Glucose 6-phosphate (G6P) and D-mannose 6-phosphate (M6P) as
substrates of phosphoglucose isomerase (PGI) and phosphomannose isomerase
(PMI), respectively; HEI: high energy intermediate of the reactions catalyzed by the
larly, 5-phospho-D-arabinonohydroxamic acid 5PAH (2, Fig. 2) de-
signed as a high energy intermediate (HEI) analogue was first
reported as the best inhibitor of phosphoglucose isomerase (PGI),
two enzymes; 5-phospho-
inhibitor of both PGI and PMI; 6-deoxy-6-malonate-b-
malonylated analogue inhibitor of the PMI cyclic substrate b -
D
-arabinonohydroxamic acid (2) as
-mannopyranose (3) as the
-mannopyranose 6-
a HEI analogue
D
which interconverts D-glucose 6-phosphate (G6P) and D-fructose
D
6-phosphate (F6P), with Ki in the low micromolar range.11–13 The
same compound turned out to behave also as the strongest known
inhibitor of phosphomannose isomerase (PMI), which intercon-
phosphate (b -M6P); malonylated analogue (8) of 2, and its ‘naked’ counterpart (9).
this function is also easy to derivatize into fatty-acid esters.
Inhibitors bearing such a malonic ester could be considered as
lipophilic prodrugs, able to cross biological membranes, and
readily hydrolyzed in the cell by various non-specific lipases or
esterases. Malonate has been used more or less successfully as
surrogate of phosphate in synthesis of inhibitors of various
enzymes.19–28 We recently reported the preparation and successful
verts D-mannose 6-phosphate (M6P, epimer of G6P at carbon 2)
and F6P.14 Kinetic and molecular modeling studies on PMI from
C. albicans allowed us to propose a detailed mechanism of the
isomerization reaction catalyzed by the enzyme.15,16 Besides its
use as probe in elucidating enzyme mechanisms, compound 2 is
of therapeutic interest since the multifunctional glycolytic enzyme
PGI is also the cytokine autocrine motility factor involved in tumor
cell migration and evolution of metastatic lesions.17,18 PMI was re-
ported to be essential for the survival or pathogenesis of several
bacteria, parasites, and yeasts, including the fungal pathogen C.
albicans involved in candidosis.3 Similarly to 1, compound 2 is un-
likely to have any in vivo biological activity in view of its high
polarity and the presence of a hydrolysable phosphate group.
For these reasons, we decided to build up analogues of these
strong inhibitors bearing a non-hydrolyzable surrogate of the
phosphate group(s). We chose to introduce a malonate group, since
evaluation of the malonylated analogue of b-
6-phosphate (b-M6P), namely 6-deoxy-6-malonate-
D
-mannopyranose
-mannopyra-
D
nose (3, Fig. 2) as a hydrolytically resistant inhibitor of PMI.29 This
product was characterized as a purely competitive substrate ana-
logue inhibitor of PMI, with inhibition constants Ki in the micromo-
lar range, and Km/Ki ratio values of 10 to 30, depending on the
enzyme source. We therefore undertook the synthesis and evalua-
tion on class II FBA of compounds 4 and 5 as mono-malonylated
analogues of inhibitor 1. For a useful comparison, we also prepared
and tested the ‘naked’ compounds 6 and 7 (Fig. 1). Similarly, com-
pound 8, the malonylated analogue of 5PAH 2, as well as its ‘naked’
analogue 9, were synthesized and evaluated on PMI and PGI
(Fig. 2).
2. Results and discussion
2.1. Chemistry
The chemical syntheses of new class II FBA inhibitors 4, 5, and 7
are depicted in Scheme 1 (the syntheses of 1 and 6 were reported
previously10). Indeed, compound 109 was acylated in methanol /
triethylamine with the para-methoxybenzyl (PMB) ether of glyco-
lyl chloride,30,31 chosen for having a protection orthogonal to the
benzyl group, to afford 11, which was subsequently mesylated to
give compound 12. The dibenzylmalonate group was introduced
through displacement of the mesylate to afford 13, then 14 after
removal of the PMB ether with DDQ. 4 was finally obtained after
a classical phosphorylation with dibenzyl-N-diisopropylphosph-
oramidite and oxidation with tbutylhydroperoxide which gave
15, followed by one-step removal of the five benzyl groups (40%
overall yield from 10). Very similarly, the dibenzylmalonyl glycolyl
chloride 17 was condensed in methanol/triethylamine with the
N-alkylhydroxylamine 1810 to afford 19, which was then phos-
phorylated as described above to give compound 20. The five ben-
zyl groups were finally removed by hydrogenolysis to lead to the
Figure 1. D-Fructose 1,6-bis-phosphate (FBP) as substrate of class II D-fructose-1,6-
bis-phosphate aldolase (FBA); TS: transition-state of the reaction catalyzed by FBA;
N-(4-phosphobutyl)-phosphoglycolohydroxamic acid (1) as a selective inhibitor of
FBA; malonylated analogues (4 and 5) of 1, and their ‘naked’ counterparts (6 and 7,
respectively).

S. Desvergnes et al. / Bioorg. Med. Chem. 20 (2012) 1511–1520
1513
Scheme 1. Synthesis of class II D-fructose-1,6-bis-phosphate aldolase inhibitors 4, 5, and 7. Reagents and conditions: Synthesis of 4—(a) Cl-CO-CH2-O-PMB, MeOH, 0 °C; (b)
t
MsCl, CH2Cl2, DIPEA, 0 °C; (c) CH2(COOBn)2, NaH, DMF, rt; (d) DDQ, CH2Cl2/H2O, rt; (e) (iPr)2NP(OBn)2, triazole/imidazole, CH2Cl2, rt then BuOOH; (f) H2, Pd-C 10%, 2 equiv
NaHCO3, EtOH/H2O. Synthesis of 5—(a) NaH, BrCH2COOtBu, DMF, 0 °C to rt; (b) TFA, CH2Cl2, rt then (COCl)2, DMF cat., CH2Cl2, rt; (c) MeOH, NEt3, 0 °C; (d) (iPr)2NP(OBn)2,
triazole/imidazole, CH2Cl2, rt then tBuOOH; (e) H2, Pd-C 10%, 2 equiv NaHCO3, THF/H2O. Synthesis of 7—(a) H2, Pd-C 10%, 1 equiv NaHCO3, THF/H2O, on a previously reported
synthetic intermediate.10
t
free compound 5 (44% overall yield from butylbromoacetate and
dibenzylmalonate). Compound 7 was simply obtained after depro-
tection of 2110 whose synthesis was previously described. As pro-
ven by 13C NMR, throughout the syntheses as well as on the final
deprotected compounds, one single conformer of the hydroxamate
function was formed. We cannot however decide if it is the s-Z or s-
E conformer. Moreover, there is no evidence that the conformer
present in solution is the same as the one (s-Z) present in the active
site of the enzymes.10
in the next reaction. Addition to a dry THF solution of ditertiobutyl-
malonate and sodium hydride gave compound 25 ( –anomer: d(C-
1) = 105 ppm) in 85% yield (two steps). Following deprotection of
the hydroxyl groups by hydrogenolysis on Pd(OH)2/C to give com-
pound 26 quantitatively, oxidation of the anomeric hydroxyl group
a
led to 5-deoxy-5-ditertiobutylmalonate-D-arabinono-1,4-lactone
27 in 60% yield. Deprotection of the malonate group was success-
fully achieved with trifluoroacetic acid in dry dichloromethane to
afford the lactone 28, which, upon reaction with 50% aqueous
The chemical synthesis of the new PGI/PMI inhibitor 5-deoxy-5-
hydroxylamine gave the final product 5-deoxy-5-malonate-D-ara-
dicarboxymethyl-
Scheme 2 (the syntheses of 2,11,12 3,29 and 932,33 were reported pre-
viously). Benzylation of 5-O-trityl-
-arabinofuranose 2234,35 was
achieved using the conditions reported by Cabaret et al. for the
synthesis of the
enantiomers.36 The two anomers were separated
by flash-chromatography to afford 23 (d(C-1) = 105 ppm) and 23
b (d(C-1) = 99 ppm) in 39% and 16%, respectively (overall yield:
55%; lit.35 <40%). Following removal of the trityl group of 23
in
D
-arabinonohydroxamic acid 8 is depicted in
binonohydroxamic acid (hydroxylammonium salt) 8 in 82% yield
(two steps).
D
2.2. Inhibition studies on FBAs
L
a
The inhibitory properties of compounds 1, 4–7 were next eval-
uated on FBAs from various sources following standard proce-
dures.8 The quantitative results are given in Table 1.
All the tested compounds gave purely competitive-inhibition
kinetics patterns on the C. albicans enzyme, considered as repre-
sentative of class II FBAs. As reported previously, compound 1, in
which the two phosphate groups at positions 1 (p1) and 7 (p7,
Fig. 1) mimic the phosphates present at p1 and p6 of the substrate
a
AcOH/H2O as reported, 24 (
a
-anomer: d(C-1) = 105 ppm; lit.35
105 ppm) was obtained in 86% yield (lit.35 72%). Reaction with tri-
fluoromethanesulfonic anhydride and diisopropylethylamine
afforded benzyl 2,3-di-O-benzyl-5-O-trifluoromethanesulfonyl-
a-
D
-arabinofuranoside which was not purified and used immediately

1514
S. Desvergnes et al. / Bioorg. Med. Chem. 20 (2012) 1511–1520
Scheme 2. Synthesis of PGI/PMI inhibitor 8. Reagents and conditions: (a) KOH, BnBr, Bu4NBr; (b) AcOH/H2O, 65 °C; (c) Tf2O, DIPEA, CH2Cl2, ꢁ40 °C; (d) (tBuOOC)2CHNa, THF
reflux; (e) H2, Pd(OH)2/C, rt; (f) Br2, BaCO3, H2O/dioxane, rt; (g) TFA, CH2Cl2, rt; (h) 50% aq NH2OH.
p1 is further highlighted by the fact that its non-phosphorylated
analogue (unpublished result) proved to be inactive against FBAs.
Table 1
In vitro biochemical evaluation of inhibitor 1, 4–7 on D-fructose-1,6-bis-phosphate
The above kinetic results confirm that the two phosphate groups
at p1 and p7 are not equally important for an analogue of FBP to
be recognized by FBA. Indeed, analysis of crystallographic data ob-
tained with 1 in the active site of H. pylori10 (Fig. 3, PDB ID co-
de = 3N9S) and M. tuberculosis37 (PDB ID code = 4A22) FBAs
indicates that the phosphate group at p7 is bound to only two ac-
tive site residues (Arg280 and Ser49) through three hydrogen
bonds, and to seven water molecules. The case of the phosphate
group at p1 is rather different: the binding network implies only
one water molecule, but six hydrogen bonds to five active site res-
idues (Gly181, Lys184, Ser213, Asp255, and Thr256), with four over
aldolases (FBA) from various sourcesa
FBA source (class)
KM
M)
Ki (lM)
(l
FBP
1b
4
5
6b
7
Rabbit muscle (I)c
C. albicans (II)
H. pylori (II)c
17
57.5
0.003
0.07
257
0.05
0.04
500
2300 0.3
250 5.5
400
>1000d
0.76
1.2
350
37
M. tuberculosis (II)c
Y. pestis (II)c
21b
55b
0.0016 0.004 500
0.018
0.19 0.175
0.08
0.008 1500 0.2
a
b
c
Estimated standard error value = 5%.
Values from Ref. 10.
Ki estimated from IC50, measured at [S] = KM
No inhibition at the given concentration.
six H-bonds coming from NH groups of the C
a chain. In addition,
.
d
two of these residues are consecutive in the sequence (Asp255 and
Thr256). Consequently, the phosphate group at p1 appears much
more tightly bound to the protein than the one at p7. Although
the structure of a malonate is chemically different from its phos-
phate bio-isoster, both functions are most likely dianionic at pH
>7 used for the kinetics.29 However, molecular dynamics revealed
that the malonate group occupies about 13% more volume than the
phosphate.20 This small difference is of little importance in
replacement of a loosely bound phosphate (p7 of 1) by a malonate
(case of 4). In the case of a tight-binding interaction, like for the
phosphate at p1 of 1, the malonate substitution could induce steric
clashes (case of 5). Moreover, a malonate function has three rotat-
able bonds, while a phosphate ester group has only one. This im-
plies a more negative binding entropy for a malonylated than a
phosphorylated analogue to the enzyme active site, and still a
more negative binding entropy of a malonate at p1, where the
phosphate group is very ordered, than at p6, where it is less. Our
results clearly show that in this series of FBA inhibitors, malonate
can effectively replace phosphate only if this group is not essential
for binding of the FBP analogue to the enzyme active site, that is,
the phosphate group at p6 of FBP (or p7 in analogues 1, 4–7).
The phosphate group in p1 position being crucial to enzyme bind-
ing of substrate and analogues is likely to be difficult to mimic. Our
hypothesis is also in accordance with dihydroxyacetone phosphate
being the unique entity that binds to FBA with the phosphate
group in p1, while a large spectrum of aldehydes are accepted by
the enzyme for binding at the p6 domain.
D
-fructose 1,6-bis-phosphate (FBP), respectively, strongly and
selectively inhibits class II versus class I FBAs.10 Compound 4, in
which a malonate mimics the p7 phosphate group of compound
1, is still a strong and selective inhibitor of class II FBAs, with,
depending on the FBA source, a limited loss (2.5 to 16), or an even
better inhibitory power (case of Helicobacter pylori) and selectivity
as compared to 1. The comparison between 4 and 6 (on which a
simple hydroxyl group is present at p7 while maintaining the
phosphate group at p1 also confirms the efficiency of a malonate
as a surrogate of phosphate at this position. On the other hand,
the replacement of the phosphate group at p1 of 1 by a malonate
in 5 gave Ki values in the millimolar range and a complete loss of
Class II versus Class I selectivity. The comparison between 5 and
7 (on which a simple hydroxyl group is present at p1 while main-
taining the phosphate group at p6) also indicates that a malonate
at this position not only does not mimic the phosphate at p1 of
FBP, but has a negative effect on binding of the inhibitor. Interest-
ingly, compound 7 lacking the phosphate group at p1 is ignored by
class I FBA and efficiently recognized by class II FBAs, with low Ki
values and high selectivities. The fact that compounds 6 (phos-
phorylated at p1) and 7 (phosphorylated at p7) gave similar inhibi-
tion constants may suggest that either phosphate group in
compound 1 is important for binding at the active site of FBA.
However, 6 and 7 are ‘naked’ compounds, which lack a phosphate
or a malonate group at p7 or p1, respectively, so that these two
compounds are only good but not strong inhibitors. Our discussion
about the relative importance of either phosphate in compound 1
is based on the effect of its replacement by a malonate group.
Clearly, comparison of the inhibition constants of compounds 4
and 5 shows that such a replacement of the phosphate group at
p1 or p7 by a malonate leads to very different conclusions. In the
case of compound 4, the importance of the phosphate group at
2.3. Inhibition studies on PGI/PMI
The inhibitory properties of compounds 2, 8, and 9 were evalu-
ated on
-mannose-6-phosphate isomerase (PMI) from Escherichia coli
following standard procedures.12,16 The quantitative results are
D-glucose-6-phosphate isomerase (PGI) from yeast and
D

S. Desvergnes et al. / Bioorg. Med. Chem. 20 (2012) 1511–1520
1515
Figure 3. Schematic representation of the active site of FBA from H. pylori complexed to N-(4-phosphobutyl)-phosphoglycolohydroxamic acid (1) (PDB ID code = 3N9S10
)
showing the bonding patterns of the two phosphate groups in positions 1 (p1) and 7 (p7). Crystallographic water molecules (W) are highlighted as spheres and hydrogen
bonds by dashed lines (H-bonds distances are given in angstrom). Distances involving water molecules are given between oxygen atoms.
Table 2
In vitro biochemical evaluation of inhibitors 2, 8, and 9 on D-glucose-6-phosphate
isomerase (PGI) and D-mannose-6-phosphate isomerase (PMI)a
Enzyme
KM
(l
M)
M6P
Ki (lM)
F6P
2
8
9
Yeast PGI
E. coli PMI
84
—
—
354
0.23b
0.08d
2290c
162
5540c
10000c,e
a
b
c
Estimated standard error value = 5%.
Ref. 11.
IC50 value.
Ref. 16.
Ref. 38.
d
e
given in Table 2. The phosphate analogue inhibitor 2 is known as
the most powerful competitive inhibitor of all PGI and PMI evalu-
ated so far,11,12,14,16 with Ki values in the sub-micro-molar range
and KM/Ki ratios from 400 to 4500 for yeast PGI and E. coli PMI,
respectively. Considering now compound 8, the malonylated ana-
logue of 2, a much higher Ki value was determined for E. coli PMI
with a 2000-fold decrease in binding affinity. A similar tendency
is observed for yeast PGI with an IC50 value above 2 mM for 8 as
compared to a Ki value of 0.23 lM for 2. In the case of these
transition state analogue inhibitors, malonate does definitely not
appear as a good mimic of phosphate. Interestingly, compound 3
(Fig. 2), the weak malonylated analogue inhibitor of the PMI sub-
strate b-M6P, gave a Ki value in the same range of magnitude
(115 l lM for 8). This comparison confirms that phos-
M29 vs 162
Figure 4. Schematic representations of the active site of (A) rabbit muscle PGI and
(B) Candida albicans PMI complexed to 5-phosphate- -arabinonohydroxamic acid 2,
phate replacement of compound 2 by malonate in compound 8 is
likely responsible of the dramatic loss in enzyme binding affinity.
However, the situation is worse in the case of compound 9 (the
non-phosphorylated/-non-malonylated transition state analogue
of 2/8), which does not inhibit neither yeast PGI nor E. coli PMI.
These results confirm that the phosphate group in substrates and
analogues accounts for most of the binding affinity to either en-
zyme, is thus definitely essential, but can be very partly replaced
by a malonate group. Examination of the previously reported X-
ray crystal structure (rabbit muscle PGI-2 complex,13 Fig. 4A) and
model structure (Candida albicans PMI-2 complex,16 Fig. 4B) gives
interesting information that could explain the above kinetic re-
sults. Active site residues of PGI from yeast and rabbit muscle are
highly conserved, so that kinetic results determined with yeast
PGI can be corroborated by conclusions drawn from the rabbit
muscle PGI-2 complex structure. A similar comment can be formu-
lated for both E. coli and C. albicans PMI, for which no X-ray crystal
structures of the enzyme–inhibitor complexes are available. In the
D
depicted from the deposited crystal structure (PDB ID code = 1KOJ)13 and from the
reported polarizable molecular mechanics study,16 respectively, showing the
bonding pattern of the phosphate group in both complexes. Crystallographic (A)
and modeled (B) water molecules (W) are highlighted as spheres and hydrogen
bonds by dashed lines (distances are given in angstrom). In Figure 4A, all distances
are given between heteroatoms, while in Figure 4B, H-bond distances are provided.
case of the PGI-2 complex (Fig. 4A), binding pattern of the phos-
phate group to the enzyme involves three water molecules (four
H-bonds) and five residues (six H-bonds). Interestingly, three of
these residues (Ser209, Lys210, and Thr211) are consecutive (al-
most four with Thr214), which likely gives a rather rigid structure
of the phosphate subsite. In addition, two of these residues (Lys210
and Thr211) interact with the phosphate through the NH group of
the C
a chain, enhancing binding and ordering of the phosphate
group. Consequently, and as described above for FBA, binding of
a malonate analogue to the PGI active site is likely to strongly

1516
S. Desvergnes et al. / Bioorg. Med. Chem. 20 (2012) 1511–1520
affect the binding energy cost both in terms of enthalpy and entro-
py (more negative for a malonate than for a phosphate), and thus
appears unfavorable. In PMI-2 complex (Fig. 4B), two water mole-
cules and three active site residues are involved in the binding of
the phosphate group of inhibitor 2, five over seven H-bonds com-
ing from the three residues. Arg304 and Lys310 appear to strongly
interact with the phosphate group through two H-bonds each, with
H-bond distances even shorter than most of those observed in the
p1 phosphate subsite of FBA which has been described as quite
well ordered (Fig. 3). The strong bonding pattern at the PMI phos-
phate subsite may well explain why phosphate replacement by a
malonate is not energetically favorable. However, why binding of
the malonylated substrate analogue 3 and transition state ana-
logue 8 are in the same order of magnitude cannot be clearly ex-
plained from the model depicted in Figure 4B. Polarizable
molecular mechanics studies of the PMI-2 and PMI-8 complexes
involving discrete structural water molecules will be performed
and published in due course.
filtered. Concentration of solutions was performed under dimin-
ished pressure at temperature <30 °C using rotary evaporator. All
air- and moisture-sensitive reactions were performed under an
atmosphere of argon. Analytical TLC was performed using Silica
Gel 60 F254 pre-coated aluminum plates (E. Merck). Spots were
visualized by treatment with 5% ethanolic H2SO4 followed by heat-
ing and/or by absorbance of UV light at 254 nm. Optical rotations
were measured at 25 °C on a Jasco DIP-370 digital polarimeter at
589 nm. NMR spectra were recorded at 297 K in CDCl3, CD3OD
and D2O with Bruker DRX 400 (1H at 400.13 MHz and 13C at
100.62 MHz), Avance 360 (1H at 360.13 MHz and 13C at
90.56 MHz), DRX 300 (1H at 300.13 MHz and 13C at 75.47 MHz)
or DPX 250 (1H at 250.13 MHz, 13C at 62.90 MHz, and 31P at
101.26 MHz) spectrometer using NMR Notebook 2.0 WinNMR soft-
ware. Chemical shift are reported in ppm (d) and coupling con-
stants in Hz (Jij). 1H NMR spectra were referenced to internal
residual chloroform (d 7.26), CD2HOD (d 3.31) and HOD (d 4.78)
for solutions in CDCl3, CD3OD and D2O, respectively. 13C NMR spec-
tra were referenced to solvent for solutions in CDCl3 (d 77.0) and
CD3OD (d 49.1), and to dioxane (d 67.4) for solutions in D2O. In
most cases, COSY, HSQC, and/or DEPT135 NMR spectra were re-
corded for assigning resonances. Infrared spectra were recorded
with a FTIR Bruker IFS-66 spectrometer. Low-resolution mass spec-
trometry (MS) and high-resolution mass spectrometry (HRMS)
analyses were performed by electrospray with positive (ESI+) or
negative (ESIꢁ) ionization mode.
2.4. Biological studies
Compounds 1, 4, and 5 were assayed for growth inhibition of
cultivated M. tuberculosis and Y. pestis. No effect was observed at
concentrations up to 1 mM.
3. Conclusion
With a dianionic ionization state at physiological pH as for phos-
phate and a steric hindrance only slightly larger, malonate has been
widely considered as a good phosphate surrogate in several exam-
ples reported in the literature. Actually, in our design of an hydrolyt-
ically stable phosphosugar analogue as potential inhibitor of class II
FBA, phosphate replacement by a malonate group gave aneffectively
strong inhibitor (compound 4). The same success was obtained in
4.2. Synthesis of compound 4
4.2.1. (E)-N-(Benzyloxy)-N-(4-hydroxybut-2-enyl)-2-(4-methoxy
benzyloxy)acetamide (11)
(E)-4-(Benzyloxyamino)but-2-en-1-ol (10)9 (1.21 g, 6.23 mmol)
and triethylamine (1.05 mL, 7.5 mmol) were dissolved in anhy-
drous methanol (15 mL) at 0 °C. 4-Methoxy-benzyloxyacetyl chlo-
ride30,31 dissolved in dichloromethane was then added dropwise.
The mixture was further stirred during 2 h, diluted with 20 mL of
dichloromethane and washed three times with water. The organic
phase was dried over sodium sulfate and evaporated. The product
was purified by flash-chromatography (pentane/ethyl acetate 6:4)
to afford the title compound (1.78 g, 77%). Rf = 0.34 (pentane/AcOEt
2:8); 1H NMR (CDCl3, 250 MHz) d 3.79 (s, 3H, CH3), 4.10 (d, 2H,
NCH2, J 6.8 Hz), 4.16 (s, 2H, OCH2CO), 4.24 (d, 2H, J 8.8 Hz, CH2OH),
4.50 (s, 2H, OCH2PhOMe), 4.76 (s, 2H, CH2Ph), 5.67–5.89 (m, 2H,
CH@CH), 6.88 (d, 2H, J 12.4 Hz, o-Har-PhOMe), 7.28–7.37 (m, 7H,
H-Ph, m-Har-PhOMe); 13C NMR (CDCl3, 62.9 MHz): d 48.2 (NCH2),
55.3 (CH3), 62.6 (CH2OH), 67.2 (OCH2CO), 72.9 (OCH2PhOMe),
77.0 (CH2Ph), 113.8 (o-Car-PhOMe), 124.2 (CH@CH–CH2OH),
128.7, 129.1, 129.4, 129.8, 134.2 (C-Ph, CH@CH-CH2OH, m-Car-
PhOMe), 159.4 (C@O).
inhibition of PMI by 6-deoxy-6-malonate-D-mannopyranose 3, the
malonylated analogue of the substrate b-M6P. In other cases, such
as for PMI or PGI inhibition by 8, the malonylated analogue of the
strong inhibitor 2, or in inhibition of FBA by 5, such a replacement
yielded inhibitors with poor or no significant binding properties.
Analysis of our previously reported tridimensional crystal (FBA,
PGI) and theoretical model (PMI) structures of enzyme–inhibitor
complexes gives informations in accordance with the kinetic results.
Somehow logically, malonate appears as a good phosphate surro-
gate only if phosphate binding is not crucial for the whole binding
of the phosphorylated entity at the enzyme active site. We propose
that binding enthalpy and entropy could be determining parameters
in the overall binding energy of the malonylated analogue at the
enzyme active site. In conclusion, malonate is not a versatile phos-
phate surrogate: such a replacement appears rather dependent on
the enzyme studied, and in the case of bis-phosphorylated sub-
strates or inhibitors, further dependent on the substitution position.
These observations are of importance for further design of enzyme
inhibitors with therapeutic interest.
4.2.2. (E)-4-(N-(benzyloxy)-2-(4-methoxybenzyloxy)acetamido)
but-2-enyl methanesulfonate (12)
Compound 11 (1.88 g, 5.1 mmol), a catalytic amount of DMAP
and DIPEA (1.3 mL, 7.5 mmol) were dissolved in anhydrous dichlo-
romethane (20 mL) at 0 °C. To this stirred solution was added drop-
wise mesyl chloride (0.47 mL, 6.12 mmol), and the mixture was
further stirred during 1 h. The mixture was diluted with more
dichloromethane, washed with water, 1 M HCl, and saturated NaH-
CO3. The organic phase was dried and evaporated to afford the title
compound (2.125 g, 94%). 1H NMR (CDCl3, 250 MHz) d 2.98 (s, 3H,
SCH3), 3.80 (s, 3H, OCH3), 4.20 (s, 2H, OCH2CO), 4.28 (d, 2H, J 6.8 Hz,
NCH2), 4.54 (s, 2H, CH2PhOMe), 4.70 (d, 2H, J 8.8 Hz, CH2OMs), 4.80
(s, 2H, NOCH2Ph), 5.80–5.90 (m, 2H, CH@CH), 6.90 (d, 2H, J 13.6 Hz,
o-Har-PhOMe), 7.29–7.40 (m, 7H, H-Ph, m-Har-PhOMe); 13C NMR
(CDCl3, 62.9 MHz) d 38.2 (SCH3), 47.8 (NCH2), 55.3 (OCH3), 67.2
(CH2OMs), 69.4 (OCH2CO), 72.9 (CH2PhOMe), 77.1 (NOCH2Ph),
4. Experimental protocols
4.1. General materials and methods
Unless otherwise stated, all chemical reagents were of analyti-
cal grade, obtained from Acros, Alfa Aesar, or Aldrich, and used
without further purification. Solvents were obtained from SDS or
VWR-Prolabo. MeOH, CH2Cl2, and THF were dried by refluxing
with, respectively, Mg/I2, CaH2, and Na/benzophenone, then dis-
tilled and used immediately. Flash chromatography was performed
using silica gel (35–70
lm, E. Merck) under N2 pressure. Unless
otherwise stated, all organic extracts were dried over Na2SO4 and

S. Desvergnes et al. / Bioorg. Med. Chem. 20 (2012) 1511–1520
1517
113.9 (o-Car-PhOMe), 126.3 (NCH2CH@), 128.8, 129.1, 129.4, 129.8,
134.2 (C-Ph, NCH2CH@CH, m-Car-PhOMe), 159.4 (C@O).
126.7 (NCH2CH@), 128.8, 129.1, 129.4, 129.8, 135.3 (C-Ph,
NCH2CH@CH), 168.4 (CO2Bn), 169.0 (NC@O).
4.2.3. (E)-Dibenzyl 2-(4-(N-(benzyloxy)-2-(4-methoxybenzyloxy)
acetamido)but-2-enyl)malonate (13)
4.2.6. Sodium 2-carboxy-6-(2-(hydrogenphosphonatooxy)-N-hy
droxyacetamido)hexanoate (4)
A solution of dibenzylmalonate (2.15 mL, 8.8 mmol) in anhy-
drous DMF (2 mL) was added dropwise to a suspension of 60%
NaH (0.126 g, 5.26 mmol) in the same solvent (6 mL) at 0 °C. After
30 min, compound 12 (2.12 g, 4.38 mmol) dissolved in DMF (2 mL)
was added dropwise, and the mixture was stirred 2 h at rt. Water
(20 mL) was then added and the mixture was extracted with
diethyl ether (3 ꢀ 50 mL). The organic phase was washed with
water, brine, then dried and evaporated. Flash-chromatography
(pentane/ethyl acetate 7:3) afforded the title product (2.17 g,
78%). Rf = 0.44 (pentane/AcOEt 7:3); 1H NMR (CDCl3, 250 MHz) d
2.72 (t, 2H, J 7.0 Hz, CH2CH(CO2Bn)2), 3.70 (t, 1H, J 7.0 Hz,
CH(CO2Bn)2), 3.84 (s, 2H, OCH3), 4.18 (m, 4H, NCH2, OCH2CO),
4.56 (s, 2H, CH2PhOMe), 4.74 (s, 2H, NOCH2Ph), 5.15, 5.16 (2s,
4H, CO2CH2Ph), 5.55–5.70 (m, 2H, CH@CH), 6.90 (d, 2H, J 9.3 Hz,
o-Har-PhOMe), 7.29–7.40 (m, 17H, H-Ph, m-Har-PhOMe); 13C NMR
(CDCl3, 62.9 MHz) d 31.4 (CH2CH(CO2Bn)2), 41.6 (NCH2), 51.6
(CH(CO2Bn)2), 55.3 (OCH3), 67.2 (OCH2CO), 67.3 (CO2CH2Ph), 72.9
(CH2PhOMe), 77.0 (NOCH2Ph), 113.8 (o-Car-PhOMe), 126.3
(NCH2CH@), 128.8, 129.1, 129.4, 129.8, 134.2 (C-Ph, NCH2CH@CH,
m-Car-PhOMe), 159.4 (Cq-Ph), 166.3 (NC@O), 168.5 (CO2Bn).
Compound 15 (0.31 g, 0.4 mmol) was dissolved in a 15:1 mix-
ture of ethanol and water, together with sodium hydrogenocarbon-
ate (0.068 g, 0.8 mmol). Pd/C (35 mg) was added and the mixture
was vigorously stirred overnight under hydrogen (4 bars). After fil-
tration and evaporation, compound 4 was obtained as a white
powder (0.15 g, 100%). 1H NMR (D2O, 360 MHz) d 1.25–1.26 (m,
2H, CH2CH2CH), 1.60 (m, 2H, NCH2CH2), 1.78 (m, 2H, CH2CH),
3.18 (t, 1H, J 5.0 Hz, CH), 3.58 (t, 2H, J 5.5 Hz, NCH2), 4.63 (d, 2H,
J 5.3 Hz, COCH2O); 13C NMR (D2O, 90.5 MHz) d 23.8 (CH2CH2CH),
25.4 (NCH2CH2), 29.2 (CH2CH), 48.1 (NCH2), 53.7 (CH), 62.2 (d, JPC
3.2 Hz, COCH2O), 170.4 (NC@O), 177.0, 177.1 (COO); HRMS (ESIꢁ)
calcd for C9H15NO10P: 328.0434 [M–2Na+H]–, found m/z 328.0420.
4.3. Synthesis of compound 5
4.3.1. Dibenzyl 2-(2-tert-butyloxy-2-oxoethyl)malonate (16)
A solution of dibenzylmalonate (4 g, 14.08 mmol) in anhydrous
DMF (17.5 mL) was added dropwise to a suspension of 60% NaH
(338 mg, 8.45 mmol) in anhydrous DMF (18.7 mL) at 0 °C. After
30 min, a solution of tert-butyl-bromoacetate (1 mL, 6.77 mmol)
in DMF (6.6 mL) was added dropwise at 0 °C, and the mixture
was stirred for 2 h at rt. The reaction was quenched by addition
of water (50 mL) and the mixture was extracted with diethyl ether
(3 ꢀ 100 mL). The organic layers were washed with water, brine,
then dried over MgSO4, filtered and concentrated under vacuum
to give compound 16 used in the next step without further purifi-
cation. 1H NMR (CDCl3, 250 MHz) d 1.37 (s, 9H, t-Bu), 2.85 (d, 2H, J
7.5 Hz, CH2CH), 3.88 (t, 1H, J 7.5 Hz, CH), 5.13 (s, 4H, 2ꢀCH2Ph),
7.18–7.46 (m, 10H, CHar).
4.2.4. (E)-Dibenzyl 2-(4-(N-(benzyloxy)-2-hydroxyacetamido)but-
2-enyl)malonate (14)
DDQ (0.93 g, 4.0 mmol) was added at once to an heterogeneous
mixture of 13 (2.16 g, 3.4 mmol) in water/dichloromethane 7:1
(25 mL). The mixture was stirred overnight at rt, then washed with
sodium thiosulfate until no color persisted, and brine. After drying
and evaporation, the product was purified by flash-chromatogra-
phy (pentane/ethyl acetate 7:3) to afford the title compound
(1.4 g, 81%). Rf 0.56 (pentane/AcOEt 7:3); 1H NMR (CDCl3,
360 MHz) d 2.71 (t, 2H, J 5.5 Hz, CH2CH(CO2Bn)2), 3.56 (t, 1H, J
5.6 Hz, CH(CO2Bn)2), 4.22–4.41 (m, 4H, NCH2, CH2OH), 4.77 (s,
2H, NOCH2Ph), 5.15, 5.16 (2s, 4H, CO2CH2Ph), 5.57–5.70 (m, 2H,
CH@CH), 7.28–7.41 (m, 15H, H-Ph); 13C NMR (CDCl3, 90.5 MHz) d
31.3 (CH2CH(CO2Bn)2), 48.9 (NCH2), 51.6 (CH(CO2Bn)2), 60.3
(CH2OH), 67.2 (CO2CH2Ph), 76.8 (NOCH2Ph), 126.6 (NCH2CH@),
128.8, 129.1, 129.4, 129.8, 135.3 (C-Ph, NCH2CH@CH), 166.3
(NC@O), 168.3 (CO2Bn).
4.3.2. Dibenzyl 2-(2-chloro-2-oxoethyl)malonate (17)
A solution of compound 16 (6.77 mmol) in CH2Cl2/TFA 1:1
(5.4 mL) was stirred at room temperature for 2 h. The mixture
was concentrated under vacuum. Following flash chromatography
on silica gel (petroleum ether/AcOEt 10:0 to 4:6), the deprotected
carboxylic acid was obtained (1.62 g, 70% from tert-butyl bromo-
acetate). 1H NMR (CDCl3, 360 MHz) d 3.00 (d, 2H, J 7.3 Hz, CH2CH),
3.91 (t, 1H, J 7.3 Hz, CH), 5.15 (s, 4H, 2ꢀCH2Ph), 7.17–7.44 (m, 10H,
CHar). Oxalyl chloride (755 lL, 8.79 mmol) was added dropwise to
4.2.5. (E)-Dibenzyl 2-(4-(N-(benzyloxy)-2-dibenzylphosphoaceta
mido)but-2-enyl)malonate (15)
a solution of this product (1.00 g, 2.92 mmol) in CH2Cl2 (2 mL) con-
taining 2 drops of anhydrous DMF. The mixture was stirred for 2 h
at room temperature and then was concentrated under vacuum to
give compound 17 used in the next step without further
purification.
A mixture of compound 14 (1.42 g, 2.75 mmol) and dibenzyl-
diisopropylphosphoramidite (1.9 g, 5.51 mmol) was kept under
vacuum until no more bubbles formed. Triazole (0.57 g,
8.25 mmol), imidazole (0.375 g, 5.51 mmol) and anhydrous dichlo-
romethane were added and the mixture was stirred overnight at rt
under argon. tButylhydroperoxide (aqueous solution, 0.65 mL,
5.51 mmol) was then added, and the mixture was stirred for addi-
tional 3 h. Dichloromethane (25 mL) and aqueous thiosulfate
(25 mL) were added. The organic phase was washed with saturated
sodium hydrogenocarbonate, dried, and evaporated. The crude
product was purified by flash-chromatography (pentane/ethyl ace-
tate 6:4) to afford the title compound (1.86 g, 87%). Rf 0.38 (pen-
tane/AcOEt 4:6); 1H NMR (CDCl3, 360 MHz) d 2.70 (t, 2H, J 5.7 Hz,
CH2CH(CO2Bn)2), 3.55 (t, 1H, J 5.5 Hz, CH(CO2Bn)2), 4.15 (d, 2H, J
4.5 Hz, NCH2), 4.40 (d, 2H, J 9.2 Hz, NCOCH2O), 4.74 (s, 2H,
NOCH2Ph), 5.14 (d, 4H, J 7.2 Hz, P(OCH2Ph)2), 5.15, 5.16 (2s, 4H,
CO2CH2Ph), 5.57–5.70 (m, 2H, CH@CH), 7.28–7.38 (m, 25H, H-
Ph); 13C NMR (CDCl3, 90.5 MHz) d 31.4 (CH2CH(CO2Bn)2), 48.6
(NCH2), 51.6 (CH(CO2Bn)2), 64.4 (d, JPC 5.4 Hz, NCOCH2O), 67.2
(CO2CH2Ph), 69.5 (d, JPC 6.3 Hz P(OCH2Ph)2), 76.7 (NOCH2Ph),
4.3.3. Dibenzyl 2-(2-(benzyloxy(4-hydroxybutyl)amino)-2-oxoet
hyl)malonate (19)
A solution of compound 17 (2.92 mmol) in CH2Cl2 (6.3 mL) was
added dropwise to a solution of N-(4-hydroxybutyl)-O-benzyl-
hydroxylamine 1810 (3.65 mmol) and triethylamine (611
lL,
4.39 mmol) in anhydrous methanol (7.9 mL) at 0 °C. The reaction
mixture was stirred for 30 min at rt, and then diluted with CH2Cl2
(20 mL) and washed with water (3 ꢀ 10 mL). The organic layer was
dried over MgSO4, filtered and concentrated under vacuum. After
purification by flash chromatography on silica gel (petroleum
ether/AcOEt 10:0 to 4:6), compound 19 was obtained (1.10 g,
72% from compound 17). 1H NMR (CDCl3, 250 MHz) d 1.43–1.58
(m, 2H, HOCH2CH2), 1.60–1.77 (m, 2H, CH2CH2N), 3.06 (d, 2H, J
7.5 Hz, CH2CH), 3.52–3.73 (m, 4H, HOCH2 and CH2N), 4.02 (t, 1H,
J 7.5 Hz, CH), 4.81 (s, 2H, NOCH2Ph), 5.10 (d, 2H, J 12.5 Hz,
2ꢀCH2Ph), 5.17 (d, 2H, J 12.5 Hz, 2ꢀCH2Ph), 7.17–7.44 (m, 15H,

1518
S. Desvergnes et al. / Bioorg. Med. Chem. 20 (2012) 1511–1520
CHar); 13C NMR (CDCl3, 62.9 MHz)
d
23.3 (CH2CH2N), 29.5
59.3 (CH2OH), 64.5 (d, JPC 4.5 Hz, POCH2CH2), 171.8 (CO); 31P
(HOCH2CH2CH2), 32.0 (CH2CH), 45.5 (CH2N), 47.7 (CH2CH), 62.1
(HOCH2CH2), 67.4 (2ꢀCO2CH2Ph), 76.5 (NOCH2Ph), 128.2–129.2
(15ꢀCHar), 134.4 (Car, NOBn), 135.4 (2ꢀCar, CO2Bn), 168.7
(2ꢀCO2Bn), 171.5 (CO); MS (ESI+) m/z 542 (100%) [M+Na]+.
NMR (D2O, 101.2 MHz) d 3.65; HRMS (ESIꢁ) calcd for C6H13NO7P:
242.0430 [MꢁH]ꢁ, found m/z 242.0439.
4.5. Synthesis of compound 8
4.3.4. Dibenzyl2-(2-(benzyloxy(4-dibenzylphosphobutyl)amino)
-2-oxoethyl)malonate (20)
4.5.1. Preparation of benzyl 2,3-di-O-benzyl-5-O-trityl-
D-arabinof
uranoside (23
a
/23b35
1,2,4-Triazole (239 mg, 3.46 mmol) and imidazole (353 mg,
5.19 mmol) were added to a solution of compound 19 (900 mg,
1.73 mmol) and dibenzyl-diisopropylphosphoramidite (1.20 g,
3.46 mmol) in anhydrous CH2Cl2 (24 mL). The reaction mixture
was stirred overnight at rt under argon. A 70% aqueous solution
To a solution of trityl arabinose 2234,35 (5 g, 12.74 mmol) in ben-
zyl bromide (15 mL) was added crushed KOH (2.5 g). After addition
of tetrabutylammonium bromide (250 mg) as catalyst, the mixture
was stirred overnight at rt. Diethyl ether was added (50 mL) and
the solution was decanted, washed with water (2 ꢀ 20 mL), dried
over Na2SO4, and evaporated. The residue was purified by chroma-
tography with petroleum ether/ethyl acetate (9:1) as eluent to give
of tert-butylhydroperoxide (494 lL, 3.46 mmol) was then added
at 0 °C, and the mixture was stirred another 2 h at rt. A 1 M solu-
tion of sodium thiosulfate (25 mL) was added and the aqueous
layer was washed with CH2Cl2 (3 ꢀ 25 mL). The collected organic
23
all yield = 55%). 23
+32.47° (c 0.85, dichloromethane); 13C NMR (CDCl3,
a
as a colorless oil (3.29 g) and 23b as a colorless oil (1.35 g) (over-
a: Rf 0.53 (petroleum ether/ethyl acetate: 9:1);
layers were washed with
a
saturated solution of NaHCO3
[a]
589
(25 mL), dried over MgSO4, filtered and concentrated under vac-
uum. After purification by flash chromatography on silica gel
(petroleum ether/AcOEt 10:0 to 6:4), compound 20 was obtained
(1.19 g, 88%). 1H NMR (CDCl3, 250 MHz) d 1.54–1.80 (m, 4H,
POCH2CH2CH2), 3.09 (d, 2H, J 7.5 Hz, CH2CH), 3.55–3.70 (m, 2H,
CH2N), 3.98 (q, 2H, J 6.5 Hz, POCH2), 4.05 (t, 1H, J 7.5 Hz, CH),
4.81 (s, 2H, NOCH2Ph), 5.03 (d, 2H, J 7.5 Hz, POCH2Ph), 5.06 (d,
2H, J 7.5 Hz, POCH2Ph), 5.13 (s, 2H, COOCH2Ph), 5.21 (s, 2H,
COOCH2Ph), 7.19–7.53 (m, 25H, CHar); 13C NMR (CDCl3,
90.5 MHz) d 23.0 (CH2CH2N), 27.4 (d, JPC 6.5 Hz, POCH2CH2CH2),
32.0 (CH2CH), 45.2 (CH2N), 47.7 (CH2CH), 67.4 (d, JPC 6.0 Hz,
POCH2CH2), 67.4 (2ꢀCO2CH2Ph), 69.3 (d, J 5.5 Hz, 2ꢀPOCH2Ph),
76.6 (NOCH2Ph), 128.0–129.4 (25ꢀCHar), 134.4 (Car, NOBn), 135.4
(2ꢀCar, CO2Bn), 136.0 (2ꢀCar, POBn), 171.5 (CO), 179.2 (2ꢀCO2Bn);
31P NMR (CDCl3, 101.2 MHz) d ꢁ0.91; MS (ESI+) m/z 802 (22%)
[M+Na]+.
90.5 MHz) d 64.0 (C-5
OCH2, C-3 –OCH2), 83.3 (C-2
(CPh3), 105.3 (C-1
a
), 69.0 (C-1
a
–OCH2), 72.1, 72.2 (C-2
a –
a
a), 84.1 (C-3
a
), 86.9 (C-4 ), 88.5
a
a
), 129.0–127.1 (CHar Ph), 137.8, 138.2 (Cqar Bn),
144.2 (Cqar CPh3); MS (ESI+) m/z 685.52 (100%) [M+Na]+, 701.47
(10%) [M+K]+; 23b: Rf 0.36 (petroleum ether/ethyl acetate: 9:1);
[
a
]
589 ꢁ37.40° (c 1, dichloromethane); 13C NMR (CDCl3, 90.5 MHz)
d 66.2 (C-5b), 68.9 (C-1 b –OCH2), 72.5, 72.6 (C-2 b –OCH2, C-3 b –
OCH2), 77.4 (C-2b), 80.7 (C-3b), 83.1 (C-4b), 84.4 (CPh3), 99.0 (C-
1b), 127.2–129.4 (CHar Ph), 137.8, 138.1 (Cqar Bn), 144.1 (Cqar
CPh3); MS (ESI+) m/z 685.52 (100%) [M+Na]+, 701.47 (10%) [M+K]+.
4.5.2. Preparation of benzyl 2,3-di-O-benzyl-a-D-arabinofuranosi
de (24)35
Compound 23a(3.35 g, 5.05 mmol) was dissolved in a mixture of
acetic acid (44 mL) and water (11 mL). The reaction was heated to
65 °C. After 3 h, TLC (petroleum ether/ethyl acetate: 1:1) indicated
completed consumption of starting material. The solvent was re-
moved in vacuo (azeotrope 3ꢀ with toluene) and the crude product
was purified by wet-flash chromatography (the column was initially
eluted with petroleum ether/ethyl acetate (9:1) to wash out TrOH
and the product was obtained on elution with petroleum ether/ethyl
acetate (1:1) to afford 24 as a colorless oil (2.1 g, 99%). Rf 0.27 (petro-
4.3.5. Sodium 2-(2-(hydroxy(4-(hydrogenphosphonatooxy)butyl)
amino)-2-oxoethyl)malonate (5)
To a solution of compound 20 (156 mg, 0.20 mmol) and sodium
hydrogenocarbonate (34 mg, 0.40 mmol) in a 5:1 mixture of THF
and H2O (15 mL) 10% Pd/C (35 mg) was added. After to be purged
first with nitrogen and then with hydrogen, the reaction mixture
was stirred vigorously at room temperature for 5 h under hydro-
gen. The mixture was filtered through Celite, and the filtrate was
concentrated under vacuum to give compound 5 (75 mg, 100%).
1H NMR (D2O, 360 MHz): d 1.56–1.67 (m, 2H, POCH2CH2CH2),
1.67–1.79 (m, 2H, CH2CH2CH2N), 2.97 (d, 2H, J 7.0 Hz, CH2CH),
3.51 (t, 1H, J 7.5 Hz, CH), 3.68 (t, 2H, J 6.5 Hz, CH2N), 3.80 (m,
POCH2); 13C NMR (D2O, 90.5 MHz): d 22.6 (CH2CH2N), 27.5 (d, JPC
7.0 Hz, POCH2CH2CH2), 33.0 (CH2CH), 47.9 (CH2N), 54.3 (CH2CH),
64.0 (d, JPC 4.5 Hz, POCH2CH2), 173.6 (CO), 178.5 (2ꢀCO2H); 31P
NMR (D2O, 101.2 MHz): d 3.76; HRMS (ESIꢁ) calcd for C9H15NO10P:
328.0434 [Mꢁ2Na+H]ꢁ, found m/z 328.0306.
leum ether/ethyl acetate: 1:1); [a]589 +41.6° (c 1, dichloromethane);
1H NMR (CDCl3, 360 MHz) d 3.65 (dd, 1H, H-3, J5 5 12.3, J5 4 4.1 Hz),
0
0
0
3.83 (dd, 1H, H-5, J55 12.3, J54 2.7 Hz), 4.00 (dd, 1H, H-3, J34 6.8, J32
2.7 Hz), 4.10 (dd, 1H, H-2, J23 2.7, J21 0.9 Hz), 4.38–4.40 (m, 1H, H-4,
0
J43 6.8, J45 4.1, J45 2.7 Hz), 4.45 (d, 1H, C-3,2-OCHAHB, JBA 12.3 Hz),
4.49 (d, 1H, C-3,2-OCHAHB, JAB 12.3 Hz), 4.50 (d, 1H, C-2,3-OCHAHB,
JBA 12.3 Hz), 4.51 (d, 1H, C-2,3-OCHAHB, JAB 12.3 Hz), 4.59 (d, 1H, C-
1-OCHAHB, JBA 12.3 Hz), 4.77 (d, 1H, C-1-OCHAHB, JAB 12.3 Hz), 5.10
(s, 1H, H-1), 7.26–7.36 (m, 15H, Ph); 13C NMR (CDCl3, 90.5 MHz) d
65.2 (C-5), 69.0 (C-1-CH2), 72.1, 72.4 (C-2-CH2, C-3-CH2), 82.4 (C-
4), 83.0 (C-3), 88.2 (C-2), 105.5 (C-1), 127.1, 127.8, 127.9, 128.1,
128.6, 128.7 (CHar), 137.5, 137.9, 138.0 (Cqar); MS (ESI+) m/z 443.33
(100%) [M+Na]+.
4.4. Synthesis of compound 7
4.5.3. Benzyl 2,3-di-O-benzyl-5-deoxy-5-ditertiobutylmalonate-
4.4.1. 4-(N,2-Dihydroxyacetamido)butyl dihydrogen phosphate
a-D
-arabinofuranoside (25)
(7)
Compound 24 (600 mg, 1.43 mmol) was dissolved in dry CH2Cl2
A
reaction mixture containing compound 2110 (103 mg,
(10 mL) under argon. Diisopropylethylamine (0.61 mL, 3.71 mmol)
wasaddedandthesolutionwas stirredandcooledto ꢁ40 °C. Follow-
ing dropwise addition of trifluoromethane-sulfonic anhydride
(0.36 mL, 2.15 mmol), the reaction mixture was stirred for 3 h.
Thereafter, the solution was allowed to warm to rt and ethyl acetate
(5 mL) was added. The resulting solution was washed with satd
NH4Cl (5 mL), then neutralized with satd NaHCO3 (5 mL). The organ-
ic phase was dried and concentrated in vacuo to afford benzyl 2,3-di-
0.20 mmol), sodium hydrogenocarbonate (34 mg, 0.40 mmol) in
THF/H2O 5:1 (15 mL), and 10% Pd/C (35 mg) was stirred vigorously
at rt for 2 h under hydrogen. The mixture was filtered through Cel-
ite, and the filtrate was concentrated under vacuum to give com-
pound 7 (49 mg, 100%). 1H NMR (D2O, 250 MHz) d 1.45–1.76 (m,
4H, CH2CH2CH2N), 3.58 (t, 2H, J 6.5 Hz, CH2N), 3.69 (q, 2H, J
6.5 Hz, POCH2), 4.32 (s, 2H, CH2OH); 13C NMR (D2O, 90.5 MHz) d
22.4 (CH2CH2N), 27.5 (d, JPC 7.5 Hz, POCH2CH2CH2), 48.5 (CH2N),
O-benzyl-5-trifluoromethane-sulfonyl-a-D-arabinofuranosideas an

S. Desvergnes et al. / Bioorg. Med. Chem. 20 (2012) 1511–1520
1519
intense red oil which was not further purified and used immediately
for the next reaction: Rf 0.89 (petroleum ether/ethyl acetate: 8:2). In
a two-neck round bottom flask placed under argon and containing
dry THF (100 mL) was dissolved ditertiobutyl malonate (0.64 mg,
2.86 mmol). NaH (60% in oil, 74 mg) was then slowly added to the
solution under vigorous stirring. The reaction mixture was further
stirred at rt for 30 min. A solution of benzyl 2,3-di-O-benzyl-5-O-tri-
(C-2), 78.1 (C-3), 78.7 (C-4), 82.6, 82.7 (CMe3), 168.5, 168.6 (COOtBu),
174.2 (C-1); mmax 3364 (br s, OH), 2980, 2934 (Csp3-H), 1799 (C@O
lactone), 1725 (C@O ester) cmꢁ1; MS (ESI+) m/z 369.31 (100%)
[M+Na]+; HRMS (ESI+) calcd for C16H26O8Na: 369.1525 [M+Na]+,
found m/z 369.1513.
4.5.6. 5-Deoxy-5-dihydrogenomalonate-D-arabinono-1,4-lactone
fluoromethanesulfonyl-a-D-arabinofuranoside previously synthe-
(28)
sized in dry THF (25 mL) was then added dropwise to the sodium
ditertiobutyl malonate THF solution. The reaction mixture was stir-
red underrefluxfor1 h, thencooledtort, andneutralizedbyaddition
of methanol (0.5 mL) and satd NH4Cl (10 mL). The aq phase was ex-
tractedwithdiethylether(3 ꢀ 10 mL). Thecombinedorganicphases
were washedwith satdNaCl (10 mL), driedover Na2SO4, filtered, and
concentrated in vacuo. The residue was purified by wet-flash chro-
matography (petroleum ether/ethyl acetate: 95:5) to give 25 as col-
orless oil (500 mg, 85% for two steps). Rf 0.81 (petroleum ether/ethyl
Compound 27 (90 mg, 0.26 mmol) was dissolved in dry dichloro-
methane (1.5 mL) under argon. The solution was cooled to 0 °C and
TFA was added (1.5 mL). The reaction mixture was stirred at 0 °C
for 35 min then 2.5 h at rt. The solution was concentrated in vacuo
(azeotrope 3 ꢀ 20 mL with toluene) at rt and lyophilized to afford
the title compound 28 as a colorless oil (50 mg, 82%). 1H NMR
(D2O, 360 MHz) d 2.30 (dd, 1H, H-50, J5 5 14.5, J5 4 9.1 Hz), 2.52 (dd,
0
0
0
1H, H-5, J55 14.5, J54 2.7 Hz), 3.61–3.80 (m, 1H, H-6), 4.16 (dd, 1H,
0
H-3, J32 9.1, J34 9.1 Hz), 4.37 (ddd, 1H, H-4, J43 9.1, J45 9.1, J45
acetate: 8:2); [
a]
589 +51.6° (c 0.9, dichloromethane); 1H NMR (CDCl3,
2.7 Hz), 4.58 (d, 1H, H-2, J23 9.1 Hz); 13C NMR (D2O, 90.5 MHz) d
31.0 (C-5), 73.7 (C-2), 77.0 (C-4), 78.9 (C-3), 172.8, 173.1 (COOH),
175.9(C-1);MS(ESIꢁ)m/z233.05(100%)[MꢁH]ꢁ; HRMS(ESIꢁ)calcd
for C8H9O8: 233.0297 [MꢁH]ꢁ, found m/z 233.0293.
0
360 MHz) d 1.38 (s, 18H, CH3), 3.38 (dd, 1H, H-6, J65 8.6, J65 6.4 Hz),
3.48 (ddd, 1H, H-50, J5 5 12.3, J5 4 8.6, J5 6 6.4 Hz), 3.68 (dd, 1H, H-3,
0
0
0
0
J32 3.6, J34 0.9 Hz), 3.78 (ddd, 1H, H-5, J55 12.3, J56 8.6, J54 4.1 Hz),
4.13 (dd, 1H, H-2, J23 3.6, J21 0.9 Hz), 4.15–4.23 (m, 1H, H-4), 4.51
(d, 1H, C-3,2-OCHAHB, JBA 12.3 Hz), 4.52 (d, 1H, C-3,2-OCHAHB, JAB
12.3 Hz), 4.58 (d, 1H, C-2,3-OCHAHB, JBA 12.3 Hz), 4.60 (d, 1H, C-
2,3-OCHAHB, JAB 12.3 Hz), 4.65 (d, 1H, C-1-OCHAHB, JAB 12.3 Hz),
4.82 (d, 1H, C-1-OCHAHB, JAB 12.3 Hz), 5.10 (s, 1H, H-1), 7.35–7.43
(m, 15H, Ph); 13C NMR (CDCl3, 90.5 MHz) d 28.1 (CH3), 32.5 (C-5),
51.4 (C-6), 68.7 (C-1-OCH2), 72.3 (C-2-OCH2, C-3-OCH2), 78.7 (C-4),
81.5, 81.6 (CMe3), 87.8 (C-3), 89.0 (C-2), 104.7 (C-1), 127.8, 127.9,
128.0, 128.1, 128.2, 128.5, 128.6 (CHar), 137.7, 137.9, 138.1 (Cqar),
168.6, 168.7 (C@O); MS (ESI+) m/z 641.42 (100%) [M+Na]+; HRMS
(ESI+) calcd for C37H46O8: 641.3090 [M+Na]+, found m/z 641.3095.
4.5.7. 5-Deoxy-5-hydrogenomalonate-
acid, hydroxylammonium salt (8)
D-arabinonohydroxamic
Compound 28 (50 mg, 0.21 mmol) was dissolved in a 50% aq
NH2OH solution (1.5 mL, 20.3 mmol). The reaction mixture was
stirred at rt for 25 min, concentrated under high vacuum at rt
and lyophilized to afford the title compound 8 as a white solid
(70 mg, 100%);
[
a
]
+17.6° (c 0.9, water); 1H NMR (D2O,
589
400 MHz) d 1.51 (dd, 1H, H-50, J5 5 14.1, J5 4 9.8 Hz), 1.98 (d, 1H,
0
0
0
0
H-5, J55 14.1 Hz), 3.37 (dd, 1H, H-4, J45 9.8, J43 8.2 Hz), 3.45 (d,
1H, H-3, J34 8.2 Hz), 4.23 (s, 1H, H-2); 13C NMR (D2O, 100.6 MHz)
d 33.7 (C-5), 54.9 (C-6), 68.9 (C-3), 70.2 (C-2), 74.3 (C-4), 171.1
(C-1), 179.1, 179.4 (COOꢁ); mmax 3087 (br s, OH), 2290 (Csp3-H),
4.5.4. 5-Deoxy-5-ditertiobutylmalonate-D-arabinofuranose (26)
Compound 25 (500 mg, 0.8 mmol) dissolved in ethyl acetate
and Pd(OH)2/C (92 mg) were stirred under H2 for 3 days. The mix-
1636 (C@O), 1553 (COOꢁ) cmꢁ1 HRMS (ESIꢁ) calcd for
;
C8H12NO9: 266.0512 [MꢁNH3OH]ꢁ, found m/z 266.0515.
ture was then filtered and concentrated to give 26 (a/b = 6:4) as a
colorless oil (360 mg, 100%). 1H NMR (CD3OD, 360 MHz) d 1.49 (s,
18H, CH3), 2.10–2.20 (m, 1H, H-50), 2.22–2.29 (m, 1H, H-5), 3.46–
Acknowledgments
We are very grateful to Professor Jurgen Sygusch (Université de
Montréal, Canada) for the kind supply of H. pylori FBA, to Dr. Mary
Jackson (Colorado State University, U.S.A.) for the kind supply of M.
tuberculosis and Y. pestis FBAs, and to Dr. Jean-Michel Bruneau
(Novexel, Romainville, France) for the kind supply of C. albicans
FBA. The ‘Région Ile-de-France’ (SD and RD) and the ‘Ministère de
l’Enseignement Supérieur et de la Recherche’ (SCL) are gratefully
acknowledged for financial support, as well as Delphine Arquier
(ICMMO) for mass spectrometry data. Our laboratory is a member
of the Laboratory of Excellence LERMIT supported by a grant from
ANR (ANR-10-LABX-33).
3.52 (m, 0.6H, H-6
2b, J2b3b 2.3 Hz), 3.84 (dd, 0.6H, H-2
3.94–4.07 (m, 2H, H-4, H-3), 5.19 (d, 0.6H, H-1
a), 3.67–3.74 (m, 0.4H, H-6b), 3.79 (d, 0.4H, H-
a, J2
4.8, J2
4.5 Hz),
a
3a
a1a
a, J1
2.7 Hz),
a2a
5.25 (d, 0.4H, H-1b, J1b2b 4.5 Hz); 13C NMR (CD3OD, 90.5 MHz) d
33.8, 35.6 (C-5
81.6, 82.4, 84.4 (C-2, C-3, C-4CMe3), 97.3 (C-1b), 103.4 (C-1
170.1, 170.4 (C@O); mmax 3442 (br s, OH), 2979, 2934 (Csp3-H),
1727 (C@O ester) cmꢁ1
a, C-5b), 45.1 (CH3), 52.2 (C-6), 78.9, 80.5, 80.7,
a
),
.
4.5.5. 5-Deoxy-5-ditertiobutylmalonate-D-arabinono-1,4-lactone
(27)
Compound 26 (100 mg, 0.3 mmol) was dissolved in water (1 mL)
and 1,4-dioxane (2 mL). BaCO3 (178 mg, 0.9 mmol) was added and
the solution was stirred and cooled to 0 °C. Following dropwise addi-
tion of bromine (0.05 mL, 0.9 mmol), the reaction mixture was stir-
red for 2 days at rt. The solution was neutralized with 1 M sodium
thiosulfate (1 mL). The mixture was extracted with ethyl acetate
(3 ꢀ 20 mL). The combined organic phases were washed with satd
aq NaCl (10 mL), dried over MgSO4, filtered, and concentrated in va-
cuo. The residue was purified on preparative plate (petroleum ether/
ethyl acetate: 95:5) and lyophilized to afford the lactone 27 as a col-
Supplementary data
Supplementary data (NMR spectra of new compounds evalu-
ated on enzymes) associated with this article can be found, in
the online version, at doi:10.1016/j.bmc.2011.12.050.
References and notes
1. Clem, B.; Telang, S.; Clem, A.; Yalcin, A.; Meier, J.; Simmons, A.; Rasku, M. A.;
Arumugam, S.; Dean, W. L.; Eaton, J.; Lane, A.; Trent, J. O.; Chesney, J. Mol.
Cancer Ther. 2008, 7, 110.
2. Haga, A.; Tanaka, N.; Funasaka, T.; Hashimoto, K.; Nakamura, K. T.; Watanabe,
H.; Raz, A.; Nagase, H. J. Mol. Biol. 2006, 358, 741.
3. Smith, D. J.; Proudfoot, A. E. I.; De Tiani, M.; Wells, T. N. C.; Payton, M. A. Yeast
1995, 11, 301.
4. Kotsikorou, E.; Sahota, G.; Oldfield, E. J. Med. Chem. 2006, 49, 6692.
5. Cáceres, A. J.; Michels, P. A. M.; Hannaert, V. Mol. Biochem. Parasitol. 2010, 169, 50.
orless oil (62 mg, 60%). [
a
]
+28.5° (c 1, water); 1H NMR (CDCl3,
589
300 MHz) d 1.43 (s, 18H, CH3), 2.15 (ddd, 1H, H-50, J5 5 12.3, J5 6 8.5,
0
0
0
0
J5 4 6.0 Hz), 2.44 (ddd, 1H, H-5, J55 12.3, J52 8.5, J54 3.8 Hz), 3.45 (dd,
0
1H, H-6, J65 8.5, J65 6.0 Hz), 4.11 (dd, 1H, H-3, J34 8.5, J32 8.5 Hz),
4.22 (td, 1H, H-4, J43 8.5, J45 3.8 Hz), 4.51 (d, 1H, H-2, J23 8.5 Hz);
13C NMR (CDCl3, 75.5 MHz) d 28.1 (CH3), 31.7 (C-5), 50.5 (C-6), 74.8

1520
S. Desvergnes et al. / Bioorg. Med. Chem. 20 (2012) 1511–1520
6. Azéma, L.; Lherbet, C.; Baudoin, C.; Blonski, C. Bioorg. Med. Chem. Lett. 2006, 16,
3440.
23. Miller, M. J.; Cleary, D. G.; Ream, J. E.; Snyder, K. R.; Sikorski, J. A. Bioorg. Med.
Chem. 1995, 3, 1685.
7. Buse, M. G. Am. J. Physiol. Endocrinol. Metab. 2006, 290, E1.
8. Fonvielle, M.; Coinçon, M.; Daher, R.; Desbenoit, N.; Kosieradzka, K.; Barilone,
N.; Gicquel, B.; Sygusch, J.; Jackson, M.; Therisod, M. Chem. Eur. J. 2008, 14,
8521.
24. Ye, B.; Akamatsu, M.; Shoelson, S. E.; Wolf, G.; Giorgetti-Peraldi, S.; Yan, X.;
Roller, P. P.; Burke, T. R. J. Med. Chem. 1995, 38, 4270.
25. Clavel, C.; Barragan-Montero, V.; Montero, J.-L. Tetrahedron Lett. 2004, 45,
7465.
9. Daher, R.; Therisod, M. ACS Med. Chem. Lett. 2010, 1, 101.
10. Daher, R.; Coinçon, M.; Fonvielle, M.; Gest, P. M.; Guerin, M. E.; Jackson, M.;
Sygusch, J.; Therisod, M. J. Med. Chem. 2010, 53, 7836.
11. Hardré, R.; Bonnette, C.; Salmon, L.; Gaudemer, A. Bioorg. Med. Chem. Lett. 1998,
8, 3435.
12. Hardré, R.; Salmon, L. Carbohydr. Res. 1999, 318, 110.
13. Arsenieva, D.; Hardré, R.; Salmon, L.; Jeffery, C. J. Proc. Natl. Acad. Sci. USA 2002,
99, 5872.
14. Roux, C.; Lee, J. H.; Jeffery, C. J.; Salmon, L. Biochemistry 2004, 43, 2926.
15. Roux, C.; Gresh, N.; Perera, L. E.; Piquemal, J.-P.; Salmon, L. J. Comput. Chem.
2007, 28, 938.
16. Roux, C.; Bhatt, F.; Foret, J.; de Courcy, B.; Gresh, N.; Piquemal, J.-P.; Jeffery, C. J.;
Salmon, L. Proteins 2011, 79, 203.
17. Liotta, L. A.; Mandler, R.; Murano, G.; Katz, D. A.; Gordon, R. K.; Chiang, P. K.;
Schiffmann, E. Proc. Natl. Acad. Sci. USA 1986, 83, 3302.
18. Watanabe, H.; Takehana, K.; Date, M.; Shinozaki, T.; Raz, A. Cancer Res. 1996, 56,
2960.
19. Berkowitz, D. B.; Maiti, G.; Charette, B. D.; Dreis, C. D.; MacDonald, R. G. Org.
Lett. 2004, 6, 4921.
20. Chen, H.; Luzy, J.-P.; Gresh, N.; Garbay, C. Eur. J. Org. Chem. 2006, 2329.
21. Gao, Y.; Luo, J.; Yao, Z.-J.; Guo, R.; Zou, H.; Kelley, J.; Voigt, J. H.; Yang, D.; Burke,
T. R. J. Med. Chem. 2000, 43, 911.
26. Gurgui-Ionescu, C.; Toupet, L.; Uttaro, L.; Fruchier, A.; Barragan-Montero, V.
Tetrahedron 2007, 63, 9345.
27. Jeanjean, A.; Garcia, M.; Leydet, A.; Montero, J.-L.; Morère, A. Bioorg. Med. Chem.
2006, 14, 3575.
28. Miller, M. J.; Anderson, K. S.; Braccolino, D. S.; Cleary, D. G.; Gruys, K. J.; Han, C.
Y.; Lin, K.-C.; Pansegrau, P. D.; Ream, J. E.; Douglas Sammons, R.; Sikorski, J. A.
Bioorg. Med. Chem. Lett. 1993, 3, 1435.
29. Foret, J.; de Courcy, B.; Gresh, N.; Piquemal, J.-P.; Salmon, L. Bioorg. Med. Chem.
2009, 17, 7100.
30. Piers, E.; Lemieux, R. Organometallics 1995, 14, 5011.
31. Lal, B.; Gangopadhyay, A. K.; Rajagopalan, R.; Vasantrao Ghate, A. Bioorg. Med.
Chem. 1998, 6, 2061.
32. Kobashi, K.; Kumashi, K.; Hase, J. Biochim. Biophys. Acta 1971, 227, 429.
33. Salmon, L.; Prost, E.; Merienne, C.; Hardré, R.; Morgant, G. Carbohydr. Res. 2001,
335, 195.
34. Izumi, M.; Tsuruta, O.; Hashimoto, H. Carbohydr. Res. 1996, 280, 287.
35. Winzar, R.; Philips, J.; Kiefel, M. Synlett 2010, 583.
36. Cabaret, D.; Wakselman, M. Can. J. Chem. 1990, 68, 2253.
37. de la Paz Santangelo, M.; Gest, P. M.; Guerin, M. E.; Coinçon, M.; Pham, H.;
Ryan, G.; Puckett, S. E.; Spencer, J. S.; Gonzalez-Juarrero, M.; Daher, R.; Lenaerts,
A. J.; Schnappinger, D.; Therisod, M.; Ehrt, S.; Sygusch, J.; Jackson, M. J. Biol.
Chem. 2011, 286, 40219.
22. Dourlat, J.; Valentin, B.; Liu, W.-Q.; Garbay, C. Bioorg. Med. Chem. Lett. 2007, 17,
3943.
38. Foret, J. Ph.D Thesis, Université Paris-Sud (Orsay, France), 2009.