Thioimidoyl furanosides as first inhibitors of the α-l-arabinofuranosidase AbfD3

Article abstract of DOI:10.1016/j.bmcl.2006.10.032

Two sets of five thioimidoyl α-l-arabino- and β-d-galactofuranosides were designed, synthesized and subjected to docking studies to evaluate their ability to be recognized by the active site of the α-l-arabinofuranosidase AbfD3. Further in vitro assays sh

Full text of DOI:10.1016/j.bmcl.2006.10.032

Bioorganic & Medicinal Chemistry Letters 17 (2007) 434–438
Thioimidoyl furanosides as first inhibitors of the
a-L-arabinofuranosidase AbfD3
a
a
b
a
´
`
Gerald Lopez, Richard Daniellou, Michael O’Donohue, Vincent Ferrieres
and Caroline Nugier-Chauvina,*
aEcole Nationale Supe´rieure de Chimie de Rennes, UMR CNRS 6226, Team Organic Synthesis and Organized Systems,
Avenue du Ge´ne´ral Leclerc, F-35700 Rennes, France
bLaboratoire de Technologie Enzymatique et Physico-Chimie des Agroressources, UMR URCA/INRA FARE,
8 Rue Gabriel Voisin, BP 316, F-51688 Reims, France
Received 29 August 2006; revised 12 October 2006; accepted 12 October 2006
Available online 17 October 2006
Abstract—Two sets of five thioimidoyl a-L-arabino- and b-D-galactofuranosides were designed, synthesized and subjected to dock-
ing studies to evaluate their ability to be recognized by the active site of the a-L-arabinofuranosidase AbfD3. Further in vitro assays
showed that the targeted furanosides are the first potent inhibitors of this furanosyl hydrolase and that the most efficient one, the
thiazolyl a-L-arabinofuranoside 1, is a competitive inhibitor having a KI of 1.4 lM.
Ó 2006 Elsevier Ltd. All rights reserved.
Furanosides are widely present in nature, notably in
nucleic acids. However, significant differences must be
noticed for other furanose-containing glycoconjugates:
indeed, hexofuranosides, such as D-galactofuranosides
(D-Galf), are found in microorganisms but not in mam-
malian cells.1 As a result, this observation has increased
the number of biological and chemical studies for such
potential new pharmacological targets.2–4 On the other
hand, the biodistribution of pentofuranosides is much
more important. For example, L-arabinofuranose (L-
Araf), structuraly analogous to D-Galf, is found in soil
bacteria, fungi and plants.5 In each case, galactans and
arabinans are components of the glycocalix that allows
cell survival and that also plays an important morpho-
logical role. On this basis, the development of inhibitors
of enzymes specifically involved in the biosynthesis and
catabolism of glycofuranoconjugates is of great interest
to improve knowledge connected with furanosyl trans-
ferases and/or hydrolases, and to design new com-
pounds likely to modulate their activity. Moreover, as
recently shown, the lack of activity related to the a-L-
arabinofuranosidase XYL3 resulted in the reduction of
seed size and a delayed germination without affecting
the seed viability.5 In this context, we focused our atten-
tion on an arabinofuranosidase, the AbfD3, a thermo-
stable enzyme able to degrade the biomass, and more
precisely to cleave a-L-arabinofuranosidic linkages.6
This enzyme was also used as an efficient biocatalyst
to prepare natural and non-natural disaccharides con-
taining either a-L-Araf or b-D-Galf non-reducing resi-
dues.7,8 These studies highlighted the versatility of the
AbfD3 for the recognition of substrates other than ara-
binofuranosyl derivatives.
In this article, we present a short and efficient synthesis
of the first inhibitors of the AbfD3. Based on our previ-
ous chemo-enzymatic approach, we expected the desired
inhibitors to be molecularly built up with (i) a sugar
head group recognized by the active site, that is, an a-
L-Araf group or a close structure such as b-D-Galf which
presents a hydroxymethyl added on C-5, (ii) an aglycon
part mimicking a carbohydrate entity, and also able to
create hydrogen bonds with suitable amino-acid residues
and (iii) a stable connection between these two parts.
Based on these assumptions, thioimidoyl a-L-arabino-
furanosides 1–5 and thioimidoyl b-D-galactofuranosides
6–10 could satisfy the given criteria (Table 1), that is, the
stability of the thioglycosidic linkage as well as the rec-
ognition requirements.
Keywords: Glycofuranosides; a-L-Arabinofuranosidase; Oligosaccha-
rides; Enzymes; Glycosidases; Inhibitors.
*
To test the possible potency of these arabinofurano-
sides as inhibitors, docking studies can nowadays be
Corresponding author. Tel.: +33 2 23238066; fax: +33 2 23238046;
e-mail: caroline.nugier@ensc-rennes.fr
0960-894X/$ - see front matter Ó 2006 Elsevier Ltd. All rights reserved.
doi:10.1016/j.bmcl.2006.10.032

G. Lopez et al. / Bioorg. Med. Chem. Lett. 17 (2007) 434–438
435
Table 1. Chemical structures of targeted thioimidoyl a-L-arabinofur-
anosides and b-D-galactofuranosides
OBz
O
OH
Heterocycle
(i), (ii)
O
OAc
OH
OH
Heterocycle
α- -Ara
Heterocycle
β- -Galf
O
O
OBz
OH
OBz
OH
L
f
D
OH
OH
OH
Scheme 1. Synthesis of 1-thioimidoyl a-L-arabinofuranosides 2–6.
Reagents: (i) Thione, BF3ÆOEt2, CH2Cl2; (ii) MeONa, MeOH (1, 49%;
2, 63%; 3, 55%; 4, 61%; 5, 65%).
OH
OH
N
S
Thiazolyl
S
S
S
1
2
3
4
5
6
computer model since their corresponding natural sub-
strate is known to be weakly recognized.7 Still they were
also expected to be inhibitors.
N
S
Benzothiazolyl
Pyrimidinyl
Pyridinyl
7
8
N
N
As a part of our ongoing research, a general and effi-
cient strategy has recently been developed in our
laboratory for the synthesis of 1-thioimidoyl hexofur-
anosides, readily and exclusively obtained from the
corresponding penta-O-acetyl hexofuranoses.11,7 In
the present study, we successfully extended this meth-
odology to the pentose analogue of the b-D-Galf, the
a-L-Araf, starting from the known 1-O-acetyl-2,3,5-
tri-O-benzoyl-a-L- arabinofuranose12 donor 11 using
2-mercaptothiazole, 2-mercaptobenzothiazole, 2-merca-
ptopyrimidine, 2-mercaptopyridine or 2-mercaptobenz-
imidazole, respectively, as an acceptor (Scheme 1). The
desired S-arabinofuranosides 1–5 were thus readily ob-
tained upon activation by boron trifluoride etherate
complex and subsequent deacylation under basic Zem-
plen transesterification (Fig. 1). Careful assignments of
the NMR data of these original 1,2-trans-thiofurano-
N
9
S
S
N
Benzimidazolyl
10
N
H
performed and used as a first approach. However, since
no crystallographic data for the AbfD3 protein has been
published, an in silico model was first constructed by
homology modelling to the X-ray structures of the a-
L-arabinofuranosidase from Geobacillus stearothermo-
philus T6 and from Clostridium thermocellum, sharing
26.1% and 26.0% of identity, respectively.9,10 Then, mol-
ecules 1–5 were successfully docked into the active site
(Fig. 1), according to the experimental protocol, and
compared to the ‘‘natural’’ substrate p-nitrophenyl a-
L-arabinofuranoside (p-NP-a-L-Araf). All of them
showed similar conformation inside the active site, with
free binding energy in the same order of magnitude as
for p-NP-a-L-Araf. A difference of 1–10 in their affinity
constant was nevertheless obtained when varying the
aglycon size, without any incidence due to the heteroat-
om. Therefore, these compounds were all expected to act
as competitive inhibitors of the AbfD3, with increasing
effect due to the bigger size of the aromatic substituent.
D-Galactofuranosides 6–10 were not docked into our
sides revealed
a classical chemical shift around
90 ppm for the anomeric carbon and a coupling con-
stant of JH-1,H-2 ꢀ 4 Hz.11–13 In addition to the previ-
ously prepared thioimidoyl b-D-galactofuranosides
7–10,11 thiazolyl 1-thio-b-D-galactofuranoside 6 was
prepared from penta-O-acetyl galactofuranose14 and
2-mercaptothiazole, and isolated in 55% yield.
Having thiofuranosides 1–10 in hand, their inhibitory
activity against the a-L-arabinofuranosidase AbfD3
was studied. The enzymatic assays were performed using
an aqueous solution of p-NP-a-L-Araf as a substrate,
incubated with 0.7 IU AbfD3 at 60 °C.7 The initial rates
of the hydrolytic reactions were determined by measur-
ing the release of p-nitrophenol at 400 nm for several
concentrations of inhibitor. Preliminary assays using
the unglycosylated thiones under similar conditions led
to the conclusion that none of these aromatic com-
pounds could act as an inhibitor.
Then, a first set of experiments with both series of sug-
ars, a-L-arabinofuranosides 1–5 and b-D-galactofurano-
sides 6–10, was achieved to evaluate the relative activity
of the AbfD3 during the hydrolysis reaction of p-NP-a-
L-Araf (4.9 mM) at inhibitor concentrations ranging
from 0.1 to 6 mM (Fig. 2). First glance at Figure 2
showed an expected global weaker effect of the thioga-
lactofuranosides 6–10 compared to the one caused by
the thioarabinofuranosides 1–5. This result is in com-
plete accordance with previous investigations of the
AbfD3 specificity towards p-nitrophenyl b-D-galactofur-
anoside (p-NP-b-D-Galf) revealing an extremely high Km
Figure 1. Molecular model of the AbfD3 protein, represented with the
docked inhibitor 2.

436
G. Lopez et al. / Bioorg. Med. Chem. Lett. 17 (2007) 434–438
100
1,6
1,4
80
60
40
20
0
10
9
8
1,2
1
7
6
5
4
0,8
2
1
0,6
0,4
3
0
1
2
3
4
5
6
0,2
[I] (mM)
0
-1,2
-0,2
0,8
1/[S] mM-1
1,8
2,8
3,8
Figure 2. Effects of concentration of 1-thioimidoyl a-L-arabinofur-
anosides and b-D-galactofuranosides 2–6 and 7–11, respectively, on the
relative enzymatic activity of AbfD3 (values are means of three
experiments).
Figure 3. Lineweaver–Burk plots of the AbfD3-catalysed hydrolysis of
p-NP-a-L-Araf at various concentrations [S] by the arabinofuranoside
1 at concentrations of 2 lM (Ç), 1.7 lM (j); 1.3 lM (m), 1 lM (ꢁ).
(Values are means of three experiments).
value (>50 mM) compared to the Km of 0.7 mM for the
p-NP-a-L-Araf, and suggesting a detrimental effect of
the extra hydroxymethyl group at C-5 for substrate
binding into the ꢁ1 subsite.15 Moreover, as expected
from the described docking study, it is worth mentioning
a very coherent influence of the nature of the thioimi-
doyl aglycons in both series leading. Nevertheless, the
in vitro assays showed a decreasing inhibitor effect in
the following order: 2-mercaptobenzimidazolyl (5 and
10), 2-mercaptopyridinyl (4 and 9), 2-mercaptopyrimid-
inyl (3 and 8), 2-mercaptobenzothiazolyl (2 and 7) and
2-mercaptothiazolyl (1 and 6). We could without ambi-
guity check, according to the monitoring of the reac-
tions by HPTLC, that none of these thiofuranosides
was a substrate of the enzyme under the conditions used
since no thione was released. Careful examination of the
effects of thioarabinofuranosides 1–5 indicated approxi-
mate IC50 values (molar concentration for 50% inhibi-
tion of AbfD3-catalysed hydrolysis of p-NP-a-L-Araf)
ranging from less than 0.1 mM to around 1 mM.
type of substrate analogue for AbfD3 (Fig. 3). More-
over, the intersections of the different lines with the 1/
[S] axis provided a mean KI value of 1.4 lM,17 in com-
plete agreement with the previous value estimated using
the IC50. This KI value for 1 is about 700-fold lower than
the Km (0.7 mM) of the p-NP-a-L-Araf under these con-
ditions, suggesting a much favourable binding of the
thiofuranoside 1 in the active site than the preferred syn-
thetic substrate of the AbfD3.11,7,15 We assumed that in
this compound, the thiazolyl heterocycle may accommo-
date within the +1 subsite and that the endocyclic sulfur
atom may act as a stabilising element for the inhibitor–
enzyme complex.
In conclusion, we have prepared the first thioimidoyl
furanosides able to inhibit the hydrolytic activity of
the arabinofuranosyl hydrolase AbfD3. These com-
pounds were readily obtained in only two steps from
peracyl pentose or hexose and a commercially available
thione. As it could not be predicted by preliminary in
silico study, the best inhibitor was the thiazolyl arabino-
side 1. To explain this difference and in the absence of
3D structure of the AbfD3, we assume that hydrogen
bonding may compensate the hydrophobic interactions
between a tryptophan residue present in the active site
of the enzyme and the aromatic aglycon. The present
study supplements well previous structure–activity rela-
tionship studies, but further X-ray data of the crystal-
lized or co-crystallized AbfD3 will probably give more
details about its active site and would help us in design-
ing better inhibitors.
In both series of furanosides, we could observe a signif-
icantly better inhibitor effect when an endocyclic sulfur
atom is present within the aglycon (compounds 1, 2
and 6, 7). Therefore, we presumed that the presence of
a sulfur atom in the aglycon moiety at an appropriate
distance from an electron-deficient residue within the
+1 subsite of the enzyme (H+, NH+) might be involved
in a stabilising hydrogen bond. Moreover, comparison
of the relative activity of compounds 1 and 2 suggested
a significant effect of the benzene-fused ring attached to
the thiazolyl heterocycle leading to a weaker activity for
2, presumably due to a significant steric hindrance. This
observation was corroborated by similar results ob-
tained from b-D-Galf analogues 6 and 7.
Homology modelling and docking studies. The homolo-
gy model was based on the two described structures
of L-arabinofuranosidase from the family 51.9,10 It
was constructed using the first approach mode of
the Swiss Model website—an automated comparative
protein modelling server.18 Rotatable bonds in the li-
gands were assigned with Autodock Tools—an acces-
sory program that allows the user to interact with
Autodock. Docking simulations were performed with
AutoDock 3.0.5 using a Lamarckian genetic algo-
rithm.19 The standard docking procedure was used
for a rigid protein and a flexible ligand whose torsion
In a second time, considering the greater inhibitor effect
of thioimidate 1 among all the compounds tested, we
zoomed in the appropriate concentration range (from
0 to 20 lM) the effect of the concentration of 1 on the
initial velocity of the hydrolysis reaction and thus eval-
uate the IC50 to be close to 10 lM. The calculated
KI,16 obtained from this IC50, led to the estimated
1.2 lM.
Finally, by the analysis of the Lineweaver–Burk plots,
we checked the competitive inhibitor character of this

G. Lopez et al. / Bioorg. Med. Chem. Lett. 17 (2007) 434–438
437
angles were identified (for 20 independent runs per li-
gand). A grid of 60, 60 and 60 points in x, y and z
directions was built, centred on the acid/base catalyst
(d, 1H, H-1, J1,2 = 2.4 Hz), 4.13 (dd, 1H, H-2,
J2,3 = 4.2 Hz), 4.09–3.92 (m, 2H, H-4 H-3), 3.65–3.55
(m, 2H, H-5a, H-5b); 13C NMR (CD3OD) d(ppm):
160.9 (SCS), 120.7 (2 CHN), 111.5 (CHCHCN), 92.8
(C-1), 89.2 (C-4), 85.7 (C-2), 81.2 (C-3), 65.8 (C-5);
HRMS (ESI+) for C9H12O4N2SNa: m/z [M]+: calcd
267.0415; found 267.0415.
˚
residue Glu-176. A grid spacing of 0.375 A and a dis-
tance-dependent function of the dielectric constant
were used for the calculation of the energetic map.
The default settings were used for all parameters.
At the end of docking, ligands with the most favour-
able free energy of binding and a similar orientation
as the ‘natural’ substrate were selected as the resul-
tant complex structures. All calculations were carried
out on PC based machines. The resultant structure
files were analyzed using Pymol visualization
programs.20
Pyridin-2-yl 1-thio-a-L-arabinofuranoside (4). CH2Cl2/
20
MeOH (9:1), Rf 0.6; mp: 169 °C; ½aꢂ +22 (c 1, MeOH);
D
1H NMR (CD3OD) d (ppm): 8.31 (ddd, 1H, CHN,
J = 0.9 Hz, J = 1.8 Hz, J = 4.9 Hz), 7.52 (ddd, 1H,
CHCHCHN, J = 7.5 Hz, J = 7.6 Hz), 7.21 (ddd, 1H,
CHCHCHCHN, J = 1.1 Hz), 7.00 (ddd, 1H, CHCHN),
6.05 (d, 1H, H-1, J1,2 = 3.1 Hz), 4.10 (dd, 1H, H-2,
J2,3 = 4.3 Hz), 4.00 (dd, 1H, H-3, J3,4 = 4.0 Hz), 3.96-
3.92 (m, 1H, H-4), 3.64 (dd, 1H, H-5a, J5a,5b = 11.9 Hz
J4,5a = 3.3 Hz), 3.56 (dd, 1H, H-5b, J4,5b = 4.0 Hz); 13C
NMR (CD3OD) d (ppm): 159.2 (SCN), 149.7 (CHN),
137.0 (CHCHCHN), 123.0 (CHCHCHCHN), 120.5
(CHCHN), 89.6 (C-1), 85.8 (C-4), 83.3 (C-2), 77.8 (C-
3), 62.3 (C-5); HRMS (ESI+) for C10H13O4NSNa: m/z
[M]+: calcd 243.0565; found 243.0569.
Synthesis. Optical rotations were measured on a Perkin-
Elmer 341 polarimeter. Melting points were determined
on a Reichert microscope and are uncorrected. Thin-
layer chromatography (TLC) analyses were conducted
on E. Merck 60 F254 Silica Gel non-activated plates
and compounds were revealed using a 5% solution of
H2SO4 in EtOH followed by heating. For column chro-
matography, Geduran Si 60 (40–63 lm) Silica Gel was
1
used. H, 13C, HMQC and COSY NMR spectra were
recorded on a Bruker ARX 400 spectrometer at
400 MHz for 1H and 100 MHz for 13C, respectively.
Chemical shifts are given in d-units (ppm) measured
downfield from Me4Si. The HRMS were measured at
the CRMPO with a MS/MS ZabSpec TOF Micromass
using m-nitrobenzylic alcohol as a matrix and accelerat-
ed caesium ions for ionisation.
Benzimidazol-2-yl
1-thio-a-L-arabinofuranoside
(5).
20
D
CH2Cl2/MeOH (9:1), Rf 0.61; mp: 158 °C; ½aꢂ +69
(c 1, MeOH); 1H NMR (CD3OD) d (ppm): 7.51–
7.49 (m, 2H, CHCN, CHCNH), 7.21–7.19 (m, 2H,
CHCHCN, CHCHCNH), 5.76 (d, 1H, H-1,
J1,2 = 3.8 Hz), 4.10 (dd, 1H, H-3, J2,3 = 4.4 Hz,
J3,4 = 3.9 Hz), 4.08 (ddd, 1H, H-4, J4,5a = 2.8 Hz,
J4,5b = 5.5 Hz), 3.88 (dd, 1H, H-2), 3.78 (dd, 1H, H-
5a, J5a,5b = 12.2 Hz), 3.66 (dd, 1H, H-5b); 13C NMR
(CD3OD) d (ppm): 149.6 (SCN), 134.3, 134.1 (CN,
CNH), 130.1, 128.9 (CHCHCN, CHCHCNH), 123.9,
121.3 (CHCN, CHCNH), 92.3 (C-1), 85.7 (C-4),
83.7 (C-2), 77.9 (C-3), 62.5 (C-5); HRMS (ESI+) for
C12H14O4N2SNa: m/z [M]+: calcd 305.0572; found
305.0576.
Thiazol-2-yl 1-thio-a-L-arabinofuranoside (1). CH2Cl2/
20
MeOH (9:1), Rf 0.5; mp: 157 °C; ½aꢂ +25 (c 1, MeOH);
D
1H NMR (CD3OD) d (ppm): 7.70 (d, 1H, CHS,
J = 3.3 Hz), 7.50 (d, 1H, CHN), 5.60 (d, 1H, H-1, J1,2
3.3 Hz), 4.07 (dd, 1H, H-2, J2,3 4.2 Hz), 3.99–4.03 (m,
1H, H-4), 3.94 (dd, 1H, H-3, J3,4 = 6.4 Hz), 3.72 (dd,
1H, H-5a, J4,5a = 3.3 Hz, J5a,5b = 12.4 Hz), 3.63 (dd,
1H, H-5b, J4,5b = 4.9 Hz); 13C NMR (CD3OD)
d(ppm): 160.1 (SCN), 144.1 (CHS), 123.5 (CHN), 94.3
(C-1), 88.4 (C-4), 86.5 (C-2), 79.8 (C-3), 62.7 (C-5);
HRMS (ESI+) for C8H11O4NS2Na: m/z [M]+: calcd
272.0027; found 272.0031.
Thiazol-2-yl 1-thio-b-D-galactofuranoside (6). CH2Cl2/
MeOH (9:1), Rf 0.4; H NMR (CD3OD) d (ppm): 7.72
1
(d, 1H, CHN, J = 3.3 Hz), 7.56 (d, 1H, CHS), 5.56 (d,
1H, H-1, J1,2 = 3.8 Hz), 4.12 (dd, 1H, H-3, J2,3 = 4.6 Hz,
J3,4 = 6.6 Hz), 4.07–4.04 (m, 2H, H-2, H-4), 3.72 (ddd,
1H, H-5, J4,5 = 2.8 Hz, J5,6a = 6.1 Hz, J5,6b = 6.1 Hz),
3.60–3.58 (m, 2H, H-6a, H-6b); 13C NMR (CD3OD) d
(ppm): 160.2 (SCN), 144.3 (CHS), 123.5 (CHN), 93.8
(C-1), 85.0 (C-4), 78.7 (C-2), 78.3 (C-3), 72.0 (C-5), 64.4
(C-6); HRMS (ESI+) for C9H13NO5S2Na: m/z [M]+: calcd
302.0133; found 302.00136.
Benzothiazol-2-yl 1-thio-a-L -arabinofuranoside (2).
20
D
CH2Cl2/MeOH (9:1), Rf 0.5; mp: 192 °C; ½aꢂ +74 (c
1
1, MeOH); H NMR (CD3OD) d (ppm): 7.83–7.81 (m,
2H, CHCN, CHCS), 7.49–7.47 (m, 2H, CHCHCN,
CHCHCS), 5.80 (d, 1H, H-1, J1,2 = 4.2 Hz), 3.88 (dd,
1H, H-2, J2,3 4.4 Hz), 3.85–3.75 (m, 2H, H-4, H-3),
3.60 (dd, 1H, H-5a, J4,5a = 3.4 Hz, J5a,5b = 12.1 Hz),
3.48 (dd, 1H, H-5b, J4,5b = 5.0 Hz); 13C NMR (CD3OD)
d (ppm): 150.6 (SCS), 134.9, 134.3 (CHCN, CHCS),
131.1, 129.8 (CHCHCS, CHCHCN), 124.9, 122.3
(CHCS, CHCN), 90.7 (C-1), 85.7 (C-4), 83.9 (C-2),
78.2 (C-3), 62.8 (C-5); HRMS (ESI+) for C12H13O4N-
S2Na: m/z [M]+: calcd 322.0184; found 322.0187.
Enzymatic inhibition. The hydrolytic activity of AbfD3
was quantified after incubation of the enzyme (0.7 IU)
with p-NP-a-L-Araf at 60 °C in water in the absence or
presence of inhibitor. Continuous release of p-nitrophe-
nol was measured at 400 nm. One unit of activity corre-
sponds to the amount of enzyme releasing 1 lmol of p-
nitrophenol per minute. The inhibitor and the substrate
aqueous solutions were mixed and prewarmed before
the addition of enzyme. Initial rates of the hydrolytic
reaction were used in order to determine the relative
activity using various inhibitor concentrations between
Pyrimidin-2-yl
1-thio-a-L
-arabinofuranoside (3).
20
CH2Cl2/MeOH (9:1), Rf 0.5; mp: 201 °C; ½aꢂ : +107 (c
D
1
1, MeOH); H NMR (CD3OD) d (ppm): 8.42 (d, 2H,
CHN, J = 4.9 Hz), 7.40–7.35 (m, 1H, CHCHN), 6.20

438
G. Lopez et al. / Bioorg. Med. Chem. Lett. 17 (2007) 434–438
5. Minic, Z.; Do, C.-T.; Rihouey, C.; Morin, H.; Lerouge, P.;
Jouanin, L. J. Exp. Bot. 2006, 57, 2339.
6. Debeche, T.; Bliard, C.; Debeire, P.; O’Donohue, M. J.
Protein Eng. 2002, 15, 21.
0.1 lM and 6 mM (the maximal activity corresponding
to the initial rate of the reaction without any inhibitor)
and each value is the mean of three experiments. The
inhibition constant KI was determined from Linewe-
aver–Burk plots using substrate concentrations between
0.3 and 3 mM and inhibitor concentrations between 1
and 2 lM, thanks to the following Michaelis–Menten
equation for competitive inhibition:
`
7. Euzen, R.; Lopez, G.; Nugier-Chauvin, C.; Ferrieres, V.;
Plusquellec, D.; Remond, C.; Donohue, M. O. Eur. J. Org.
´
Chem. 2005, 4860.
8. Remond, C.; Ferchichi, M.; Aubry, N.; Plantier-Royon,
R.; Portella, C.; O’Donohue, M. J. Tetrahedron Lett. 2002,
43, 9653.
9. Ho¨vel, K.; Shallom, D.; Niefind, K.; Belakhov, V.;
Shoham, G.; Baasov, T.; Shoham, Y.; Schomburg, D.
EMBO J. 2003, 22, 4922.
1=V 0 ¼ Kmð1 þ ½Iꢂ=KIÞ=V m½Sꢂ þ 1=V m
KI value was corroborated by the following equation:17
KI ¼ IC50=ð1 þ ½Sꢂ=KmÞ:
10. Taylor, E. J.; Smith, N. L.; Turkenburg, J. P.;
D’Souza, S.; Gilbert, H. J.; Davies, G. J. Biochem.
J. 2006, 395, 31.
`
11. Euzen, R.; Ferrieres, V.; Plusquellec, D. J. Org. Chem.
2005, 70, 847.
Acknowledgments
12. Wang, Z.; Prudhomme, D. R.; Buck, J. R.; Park, M.;
Rizzo, C. J. J. Org. Chem. 2000, 65, 5969.
13. Steers, E., Jr.; Cuatrecasas, P.; Pollard, H. B. J. Biol.
Chem. 1971, 246, 196.
We are grateful to the French Ministry of Education,
Research, and Technology for a grant to G.L. We thank
´
J. P. Guegan (ENSCR) for recording NMR spectra and
P. Guenot for recording HRMS spectra.
`
14. Ferrieres, V.; Gelin, M.; Boulch, R.; Toupet, L.; Plus-
´
quellec, D. Carbohydr. Res. 1998, 314, 79.
15. Remond, C.; Plantier-Royon, R.; Aubry, N.; O’Donohue,
M. J. Carbohydr. Res. 2005, 340, 637.
References and notes
16. Cheng, Y. C.; Prusoff, W. H. Biochem. Pharmacol. 1973,
22, 3099.
17. Hernandez, A.; Ruiz, M. T. Bioinformatics 1998, 14, 227.
18. Schwede, T.; Kopp, J.; Guex, N.; Peitsch, M. C. Nucleic
Acids Res. 2003, 31, 3381.
19. Morris, G. M.; Goodsell, D. S.; Halliday, R. S.; Huey, R.;
Hart, W. E.; Belew, R. K.; Olson, A. J. J. Comput. Chem.
1998, 19, 1639.
20. DeLano, W. L. In The PyMOL Molecular Graphics
System; DeLano Scientific: San Carlos, CA, USA. http://
www.pymol.org/, 2005.
1. de Lederkremer, R. M.; Colli, W. Glycobiology 1995, 5,
547.
2. Joe, M.; Sun, D.; Taha, H.; Completo, G. C.; Croudace, J.
E.; Lammas, D. A.; Besra, G. S.; Lowary, T. L. J. Am.
Chem. Soc. 2006, 128, 5059.
3. Hederos, M.; Konradsson, P. J. Am. Chem. Soc. 2006,
128, 3414.
´
4. Gandolfi-Donadıo, L.; Gallo-Rodriguez, C.; de Lederkr-
emer, R. M. J. Org. Chem. 2003, 68, 6928.