Synthesis of Unprecedented Sulfonylated Phosphono-exo-Glycals Designed as Inhibitors of the Three Mycobacterial Galactofuranose Processing Enzymes

Article abstract of DOI:10.1002/chem.201603161

This study reports a new methodology to synthesize exo-glycals bearing both a sulfone and a phosphonate. This synthetic strategy provides a way to generate exo-glycals displaying two electron-withdrawing groups and was applied to eight different carbohydrates from the furanose and pyranose series. The Z/E configurations of these tetrasubstituted enol ethers could be ascertained using NMR spectroscopic techniques. Deprotection of an exo-glycal followed by an UMP (uridine monophosphate) coupling generated two new UDP (uridine diphosphate)-galactofuranose analogues. These two Z/E isomers were evaluated as inhibitors of UGM, GlfT1, and GlfT2, the three mycobacterial galactofuranose processing enzymes. Molecule 46-(E) is the first characterized inhibitor of GlfT1 reported to date and was also found to efficiently inhibit UGM in a reversible manner. Interestingly, GlfT2 showed a better affinity for the (Z) isomer. The three enzymes studied in the present work are not only interesting because, mechanistically, they are still the topic of intense investigations, but also because they constitute very important targets for the development of novel antimycobacterial agents.

Full text of DOI:10.1002/chem.201603161

DOI: 10.1002/chem.201603161
Full Paper
&
Carbohydrates |Hot Paper|
Synthesis of Unprecedented Sulfonylated Phosphono-exo-Glycals
Designed as Inhibitors of the Three Mycobacterial
Galactofuranose Processing Enzymes
Christophe J.-M. Frꢀdꢀric,[a] Abdellatif Tikad,[a] Jian Fu,[a] Weidong Pan,[b] Ruixiang B. Zheng,[c]
Akihiko Koizumi,[c] Xiaochao Xue,[c] Todd L. Lowary,[c] and Stꢀphane P. Vincent*[a]
Abstract: This study reports a new methodology to synthe-
size exo-glycals bearing both a sulfone and a phosphonate.
This synthetic strategy provides a way to generate exo-gly-
cals displaying two electron-withdrawing groups and was
applied to eight different carbohydrates from the furanose
and pyranose series. The Z/E configurations of these tetra-
substituted enol ethers could be ascertained using NMR
spectroscopic techniques. Deprotection of an exo-glycal fol-
lowed by an UMP (uridine monophosphate) coupling gener-
ated two new UDP (uridine diphosphate)-galactofuranose
analogues. These two Z/E isomers were evaluated as inhibi-
tors of UGM, GlfT1, and GlfT2, the three mycobacterial galac-
tofuranose processing enzymes. Molecule 46-(E) is the first
characterized inhibitor of GlfT1 reported to date and was
also found to efficiently inhibit UGM in a reversible manner.
Interestingly, GlfT2 showed a better affinity for the (Z)
isomer. The three enzymes studied in the present work are
not only interesting because, mechanistically, they are still
the topic of intense investigations, but also because they
constitute very important targets for the development of
novel antimycobacterial agents.
Introduction
quence, this mAPG constitutes a true Achilles heel of the bac-
terium as inhibition of its biosynthesis prevents M. tuberculosis
survival in the infected host. Therefore, the AG is still a major
target for the development of antitubercular drugs because of
its essential role for the viability of mycobacteria.[3] The best-
characterized enzymes involved in the AG biosynthesis are
UDP (uridine diphosphate)-galactopyranose mutase (UGM),
two galactofuranosyltransferases (GlfTs), and several arabinosyl-
transferases (AraTs).[2a,4] UGM is a flavoenzyme with a flavine
adenine dinucleotide (FAD) cofactor that plays an unprece-
dented catalytic role as a nucleophile in the isomerization of
UDP-galactopyranose (1, UDP-Galp) into its furanose isomer
UDP-galactofuranose (2, UDP-Galf, Figure 1).[5] The polymeric
galactan is constructed by two GlfTs that catalyze the glycosy-
lation reactions.[6] The first transferase (GlfT1)[7] adds the two in-
itial galactofuranose (Galf) units to the disaccharide linker,
while the second transferase (GlfT2)[3a,8] catalyzes a polymeri-
zation[9] that yields a final galactan composed of 30–35 galac-
tofuranose moieties (Figure 1A).[3d,10] The inhibition of UGM
and the GlfTs would prevent the biosynthesis of AG and, thus,
weaken the M. tuberculosis cell wall. Several attempts have
been made to inhibit UGM,[2a,11] but only very few inhibition
studies of GlfT1 and GlfT2 have been described so far.[2a,12]
In previous studies, we discovered that exo-glycals such as 3
and 4-(Z,E) could inhibit UGM and/or GlfT2 (Figure 1B).[13]
The principle used in the design of these compounds is that
the sp2 hybridization of the anomeric carbon of the phospho-
nylated exo-glycals 3, 4, or 6 may lock the molecule in a confor-
Despite the availability of potent drugs and vaccines, tubercu-
losis (TB) remains a major threat for public health worldwide,
mainly because of the emergence of multi-drug resistant
strains.[1] In this context, the inhibition of the mycobacterial
cell wall biosynthesis of Mycobacterium tuberculosis, the causa-
tive agent of tuberculosis, has received significant attention.[2]
The main mycobacterial cell wall component is the mycolyl–
arabinogalactan–peptidoglycan (mAPG) complex, which is
composed of an arabinogalactan (AG) polysaccharide linked to
the peptidoglycan by a disaccharide linker. About two thirds of
the arabinan chains are covalently linked to mycolic acids that
enhance the cellular resistance against antibiotics. As a conse-
[a] C. J.-M. Frꢀdꢀric, Dr. A. Tikad, J. Fu, Prof. S. P. Vincent
University of Namur (UNamur)
Dꢀpartement de Chimie, Laboratoire de Chimie Bio-Organique
rue de Bruxelles 61, 5000 Namur (Belgium)
Fax: (+32)81-72-45-17
E-mail: stephane.vincent@unamur.be
[b] Dr. W. Pan
The Key Laboratory of Chemistry for Natural Products of Guizhou Province
Chinese Academy of Sciences
202, Sha-chong South Road, Guiyang 550002 (P. R. China)
[c] R. B. Zheng, Dr. A. Koizumi, Dr. X. Xue, Prof. T. L. Lowary
Department of Chemistry and Alberta Glycomics Centre
University of Alberta, Gunning-Lemieux Chemistry Centre
11227 Saskatchewan Drive, Edmonton, AB T6G 2G2 (Canada)
Supporting information and the ORCID(s) for the author(s) of this article
can be found under http://dx.doi.org/10.1002/chem.201603161.
Chem. Eur. J. 2016, 22, 1 – 9
1
ꢁ 2016 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
&
&
These are not the final page numbers! ÞÞ

Full Paper
Figure 1. A) The mycobacterial galactan biosynthesis. B) Structures of previous GlfT2 (GlfT=galactofuranosyltransferase) inhibitors 3 and 4. C) Design principle
of the present work.
mation mimicking the putative oxacarbenium-ion-like transi-
tion state 5 (Figure 1C).
glycals. The scope of this transformation with regards to sub-
strate and protecting groups will be discussed. Then, the de-
tailed synthesis of the UDP-Galf analogues will be shown, fol-
lowed by their biochemical characterization using inhibition
assays.
It is generally believed that the transition states of glycosyl-
transferase-catalyzed reactions have a substantial cationic char-
acter at the anomeric carbon.[14] We, thus, consider that the pu-
tative structure 5 may mimic the transition states of both GlfT1
and GlfT2. Moreover, several mechanistic studies have suggest-
ed that the transition state of the UGM-catalysed isomerization
also exhibits a partial positive charge at the anomeric posi-
tion.[5c] Therefore, these three Galf-processing enzymes could
share some common conformational and electronic features in
their respective transition states. Based on this analysis, we
became interested in exo-glycals of the general structure 6
(Figure 1C); the presence of an electron-withdrawing group
(EWG) would likely enhance the polarization of the conjugated
exocyclic enol ether, which might improve the interactions be-
tween the substrate analogues and the enzyme and/or the in-
coming nucleophile (the cofactor for UGM or the growing gal-
actan for the two GlfTs). Our preliminary results with fluorinat-
ed exo-glycals 4 support this hypothesis.[13b]
Results and Discussion
The main methods to synthesize exo-glycals are Wittig-like ole-
finations,[16] the Ramberg–Bꢂcklund rearrangement,[17] the
modified Julia olefination,[18] and the stepwise addition–elimi-
nation of sugar lactones.[19]
Recently, different functionalizations,[20] such as the Sonoga-
shira,[21] Suzuki,[22] and Stille cross-coupling reactions,[23] have al-
lowed the preparation of tetrasubtituted exo-glycals. However,
to our knowledge, there is no general method to synthesize
tetrasubstituted exo-glycals bearing two electron-withdrawing
groups. In fact, the methodology developed by Lin,[19] which
consists of a nucleophilic addition of a lithiated methyl-phos-
phonate to a sugar lactone, does not yield the desired addition
product when a lithiated phosphonate bearing an electron-
withdrawing group is used (Figure 2, Path A). This lack of reac-
tivity is likely due to the poor nucleophilicity of the phospho-
nate.
To explore this concept, we describe the synthesis of sulfo-
nylated exo-glycals and their inhibition properties not only
against UGM and GlfT2, but also against the poorly studied
GlfT1 enzyme. The sulfonyl group was selected because of its
remarkable electron-withdrawing properties, as evidenced by
the unique Michael acceptor ability of vinyl sulfones.[15]
Given the problems encountered in the direct addition of
lithiated phosphonates bearing an electron-withdrawing
group, we explored another strategy (Figure 2, Path B) that
consists of an addition of a lithiated phosphonate bearing an
This study will begin with a description of a new general
methodology to generate sulfonylated tetrasubstituted exo-
&
&
Chem. Eur. J. 2016, 22, 1 – 9
www.chemeurj.org
2
ꢁ 2016 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
ÝÝ These are not the final page numbers!

Full Paper
Scheme 1. Reagents and conditions: a) NaH, THF, 408C, 16–24 h.
As an initial experiment, the phosphonate 13 was lithtiated
and added to the perbenzylated lactone 14 to afford the lactol
21 (see scheme in Table 1). The following elimination was ach-
ieved by treating 21 with pyridine and trifluoroacetic anhy-
dride (TFAA) to give the exo-glycal 29-(Z,E) in 80% yield over
two steps in a 70:30 Z/E ratio (Table 1, Entry 1).
Figure 2. Addition of a lithiated phosphonate bearing an electron-withdraw-
ing group. [E]=elimination, [O]=oxidation.
Having established the optimal conditions for this stepwise
addition–elimination sequence, the scope of this transforma-
tion was examined with various lactones. All starting lactones
15–20 were prepared following procedures described in the
literature.[24] The results gathered in Table 1 show that, in all
cases, satisfactory yields were obtained. Moreover, we noticed
that the reaction times of the addition and elimination steps
depend on the nature of the lactone as well as the phospho-
nate reagents used. Using the phosphonate 12 instead of 11
gave the lactol 24 in a modest 33% yield, indicating that the
electron-donating group SR followed by an oxidation to ulti-
mately give the desired sulfonyl moiety. To develop this new
approach, we selected the commercially available tribenzylated
arabinolactone 14 as a model substrate (Table 1).
The phosphonylated thioethers 11 and 12 required for this
study were obtained through a classical Michaelis–Becker reac-
tion as depicted in Scheme 1, while the diethyl (thiomethyl)-
methyl phosphonate 13 is commercially available.
Table 1. Synthesis of thiomethylphosphono-exo-glycals 29–36.
Entry
Lactone
Addition
T [h]
Elimination
Yield [%][a]
Lactol
Yield [%][a]
exo-Glycal
T [h]
Ratio (Z/E)[b]
1
2
3
4
5
6
7
8
14
15
16
15
17
18
19
20
21
22
23
24
25
26
27
28
1
3
1
22
2
3
2.5
5
99
56
88
33
77
42
71
51
29
30
31
32
33
34
35
36
2
5
2
4
2
3.5
5
80
87
55
60
99
75
48
51
70:30
60:40
92:8
70:30
96:4
40:60
65:35
70:30
5
[a] Isolated yield. [b] Determined by 31P NMR spectroscopic analysis of the crude reaction mixture.
Chem. Eur. J. 2016, 22, 1 – 9
www.chemeurj.org
3
ꢁ 2016 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
&
&
These are not the final page numbers! ÞÞ

Full Paper
kinetics of the addition reaction of lithiated 12 to lactones is
significantly slower than the methyl derivative 11. This is prob-
ably because of the significant steric hindrance arising from
the phenyl groups (Table 1, Entries 2–4). For all of the em-
ployed furanosides 14–17, the formation of the (Z) isomer pre-
dominates and the (Z/E) ratios ranged from 60:40 to 96:4
(Table 1, Entries 1–5).
tween 17.7–16.4 ppm for the (E) isomers. In addition, the cou-
pling constants between C-1 and the phosphorous atom were
always between 38–42 Hz when these two elements are trans,
while this coupling constant was always close to 10 Hz when
the 13C and the 31P are cis (Table 2).
The Z/E diastereomers were very easily separated by stan-
dard silica gel chromatography except for compound 33,
which required a more tedious purification. We also observed
that the (Z)-exo-glycals were always more polar than their (E)
isomers (based on silica gel migratory ability). The latter trend
is in agreement with all cis/trans phosphono-exo-glycals al-
ready described in our previous studies.[13a,25]
However, a moderate (Z) selectivity was observed for pyra-
nosidic lactones in the reaction between 11 and the glucopyra-
noside 19 (Table 1, Entry 7). Surprisingly, the reaction with the
galactopyranoside 18 gave preferentially the (E) isomer
(Table 1, Entry 6). This difference in diastereoselectivity can be
explained by the fact that the two pyranosides adopt different
conformations in the transition state, because of the benzyl
group at the C-4 position that is axial for 34 and equatorial for
35. The result reported in Entry 5 (Table 1) show that protect-
ing groups influence the course of the reaction. Indeed,
a good yield of the elimination step (99%) with an excellent
selectivity (Z/E 96:4) was observed when acetonides were used
as protecting groups instead of benzyl ethers.
After optimizations, we found that the oxidation of the thio-
methylphosphono-exo-glycal 29-(Z) could be carried out using
two equivalents of m-chloroperoxybenzoic acid (m-CPBA) in di-
chloromethane at 08C. After one hour, the methyl sulfide was
chemoselectively oxidized leading to the desired product 37-
(Z) in quantitative yield (Table 3, Entry 1). Under these oxidizing
conditions, the epoxidation of the enol ether was not ob-
served. The second isomer 29-(E) underwent the oxidation
under the same conditions to afford 37-(E) in 66% yield
(Table 3, Entry 2). Changing the phosphonate alkyl group did
not affect the course of the reaction, as the oxidation of both
isomers 30-(Z) and 30-(E) provided the desired compounds
38-(Z) and 38-(E) in 68% and 69% yield, respectively (Table 3,
Entries 3 and 4). In addition, a Me or Ph substituent on the sul-
phur atom had no significant effect in this oxidation (Table 3,
Entries 3-6). The oxidation of 33-(Z) was performed in 90%
yield after only one hour, which indicates that an acetonide
could be also used as a protecting group of the glycoside
(Table 1, Entry 8). In the pyranose series, all the reactions start-
ing from d-Gal, d-Glu, and l-Fuc (34–36) efficiently produced
the corresponding sulfones 42–44 in good to excellent yields
in a few hours (Table 2, Entries 9–12).
The Z/E assignment of tetrasubstituted exo-glycals is chal-
lenging, because the exocyclic double bond of 29–36 is substi-
tuted by four distinct elements (O, C, S, and P). However, their
structures have been fully ascertained by NMR spectroscopy,
using 1H, 13C, 31P, 1H/1H correlations, and 1H/13C correlations
(HSQC, HMBC). To demonstrate the configurations of all exo-
glycals reported in this study, 1D homonuclear NOE spectra
(1D NOE) were also acquired and compared for each E/Z dia-
stereomeric pair.
For all (E)-configured enol ethers, 1D NOE experiments
showed a NOE effect between the protons of the SCH3 group
and the H-2 proton of the carbohydrate ring, while the ab-
sence of this effect in their corresponding (Z) isomers unam-
biguously confirms the Z/E assignment (Table 2). Moreover, for
all compounds 29–36, the H-2 signal of the (Z) isomers are
always shifted to lower field compared to the (E) isomers. The
H-2 chemical shifts were recorded at 6.3–5.4 ppm for the (Z)
isomers and 5.5–4.9 ppm for the (E) isomers. A similar trend
was observed for the 31P-signal chemical shifts, which were re-
corded between 19.7–17.8 ppm for the (Z) isomers and be-
Globally, we were able to optimize a general methodology
starting from a variety of lactones from the furanose and pyra-
nose series. After the lithiated-phosphonate addition, the sub-
sequent dehydration generally leads to the generation of the
(Z) isomer.
The oxidation of all the resulting thioether exo-glycals che-
moselectively gave the corresponding sulfones in good yields.
Importantly, to the best of our knowledge, there are no other
general synthetic methods yielding phosphonylated exo-gly-
cals bearing two electron-withdrawing groups.
2
Table 2. Comparison between JC1ÀP of both exo-glycals (E) and (Z).
Given the importance of galactofuranose (Galf) in the cell
walls of major pathogens, we naturally explored the synthesis
of modified phosphonylated analogs of this carbohydrate
starting from 38-(E,Z) (Scheme 2). To remove the benzyl
groups, the usual hydrogenation conditions cannot be used
because of the possibility of an enol ether saturation.[26] The
deprotections were, thus, carried out with boron trichloride
(BCl3) in dry dichloromethane.[27] After two hours at tempera-
tures between À788C and room temperature, the deprotected
exo-glycals 45-(Z) and 45-(E) were quantitatively isolated for
both isomers 38-(Z) and 38-(E), respectively.
Entry
exo-Glycal
2JC1ÀP (Z)
2JC1ÀP (E)
1
2
3
4
5
6
7
8
29
30
31
32
33
34
35
36
38.3
38.3
39.3
39.3
37.4
41.2
41.2
40.3
9.6
8.6
9.6
8.6
–
6.7
8.6
7.7
The coupling between 45-(Z) and uridine 5’-phosphorimida-
zolide in DMF in the presence of MgCl2 as the Lewis acid cata-
&
&
Chem. Eur. J. 2016, 22, 1 – 9
www.chemeurj.org
4
ꢁ 2016 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
ÝÝ These are not the final page numbers!

Full Paper
Table 3. Chemoselective oxidation of phosphono-exo-glycals 29–36.
Entry
1
SM[a]
Structure
T
[h]
Product Yield[b]
[%]
29-(Z)
1
2
37-(Z)
37-(E)
99
66
2
29-(E)
3
4
5
6
30-(Z)
30-(E)
32-(Z)
32-(E)
1
3
6
3
38-(Z)
38-(E)
39-(Z)
39-(E)
68
69
54
80
Scheme 2. Reagents and conditions: a) BCl3, DCM, À788C–rt, 2 h; b) Uridine
5’-phosphorimidazolide, MgCl2, DMF, rt, 16 h.
HPLC purifications (Scheme 2). This result indicates that the (E)
isomer is prone to an E-to-Z isomerization during the coupling
step, whereas the Z isomer was configurationally stable.
Biochemical investigation
7
8
31-(Z)
33-(Z)
3
1
40-(Z)
41-(Z)
78
90
The two final molecules, 46-(E) and 46-(Z), were first assayed
against UDP-galactopyranose mutase (UGMMt from M. tubercu-
losis) following two different procedures. First, the Kd values
were determined using a well-established fluorescence-polari-
zation assay.[29] This assay measures the competition between
the inhibitors and a fluorescent probe for the nonreduced
UGM (UDP-fluorescein) at a constant probe concentration and
at varying concentrations of the inhibitors. We found that the
(E) isomer 46-(E) was a significantly better inhibitor (Kd =
77 mm) than its (Z) isomer (Kd =470 mm). In our previous
studies[13a,30] we had shown that exo-glycals could behave as
time-dependent inactivators of E. coli UGM. In these experi-
ments the (E) fluorinated isomer was found to be kinetically
superior to its (Z) isomer; however, the relative affinity of both
isomers towards UGM had not been measured. The Kd meas-
urements of the sulfonylated exoglycals 46 clearly prove that
UGM has a clear preference for the (E) isomer. It should also
be emphasized that such an inhibition level against UGM
(<100 mm) has rarely been reached for UDP-galactose ana-
logues in the literature.[2a,6c]
9
34-(E)
35-(Z)
2
4
42-(E)
43-(Z)
70
99
10
11
12
35-(E)
36-(Z)
3
2
43-(E)
44-(Z)
71
88
We then evaluated the ability of 46-(Z) and 46-(E) to behave
as covalent irreversible inhibitors. Indeed, vinyl sulfones are
generally excellent Michael acceptors, which often act as a co-
valent trap of enzyme nucleophilic residues.[15b,c] Moreover, the
UGM reaction proceeds by a unique mechanism involving the
cofactor as covalent nucleophile. The X-ray structure of UGM
in complex with our fluorinated UDP-Galf analogues confirms
the spatial proximity of the FAD nucleophilic center and the
Galf anomeric position.[11d] Therefore, we measured the per-
[a] SM=starting material. [b] Isolated yield of tetrasubstituted exo-glycals.
lyst afforded the desired unprotected nucleotide sugar 46-(Z)
in modest 16% yield.[28] To our surprise, the same coupling
procedure starting from the other isomer 45-(E) gave an E/Z
mixture of the nucleotide sugar, which could be easily separat-
ed by size-exclusion chromatography and semi-preparative
Chem. Eur. J. 2016, 22, 1 – 9
www.chemeurj.org
5
ꢁ 2016 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
&
&
These are not the final page numbers! ÞÞ

Full Paper
centage of inhibition of the exo-glycals 46 by a standard HPLC
activity assay (at UDP-Galf 25 mm and inhibitor 500 mm). In
these experiments the formation of UDP-Galp is followed by
HPLC at different reaction times. However, for each experi-
ment, the enzyme and the inhibitor are pre-incubated before
the addition of the substrate (UDP-Galf), which starts the reac-
tion. The pre-incubation time varies from one experiment to
the other; if a covalent irreversible inhibition occurs, the per-
centage of inhibition should increase with pre-incubation time.
Interestingly, independent of the pre-incubation time (0–
30 min), the level of inhibition was always the same (23% and
46% for the (Z) and the (E) isomers, respectively, Table 4). Thus,
sulfonylated exo-glycals are not time-dependent inactivators of
UGM despite their Michael acceptor character. When compared
to our preliminary inactivation studies using nonfluorinated
and fluorinated exo-glycals,[13a,30] these results suggest that the
enol ethers need to be protonated first to act as inactivators.
This would generate an extremely electrophilic oxacarbenium
ion that could react with the nucleophilic FAD cofactor. The
strong electron-withdrawing character of the sulfone group on
46 likely decreases the propensity of the enol ether to be pro-
tonated. Therefore, a standard Michael addition (nucleophilic
addition followed by a protonation) of the FAD cofactor on the
electrophilic anomeric carbon of exo-glycals 46 does not occur.
In addition, these experiments also gave us an estimation of
the nucleophilicity of the FAD cofactor; if an electrophilic, non-
reversible inhibitor were to be designed, it should possess an
even more electrophilic functional group than the sulfonylated
enol ether present in molecules 46.
We then pursued the biochemical investigation with the
two subsequent mycobacterial Galf processing enzymes, the
galactofuranosyltransferases GlfT1 and GlfT2. The GlfT1 and
GlfT2 inhibition assays are based on the same principle; in
both cases the inhibitors compete with the donor substrate for
the transferase UDP-Galf (2) in presence of a specific acceptor
substrate (glycosyl phospholipid 47 and trisaccharide 48 for
GlfT1 and GlfT2, respectively; see the Supporting Information
for detailed procedures and Figure 3 for the structures). In
both cases, the production of UDP during the glycosyl transfer
reaction is coupled with a NADH oxidation, which can be
easily monitored spectrophotometrically.[12f]
Interestingly, the (E) isomer 46-(E) showed a better affinity
for GlfT1 than its (Z) isomer (Table 4), while 46-(Z) proved to
be superior for GlfT2. The level of inhibition is in the range of
the affinity of the donor substrate 2 for the two transferases
Table 4. Inhibition of UGM, GlfT1, and GlfT2 by 46-(Z) and 46-(E).
UGMMt
GlfT1
GlfT2
[a]
[c]
[e]
Kd
[mM]
Inhib[b] IC50
Inhib[d] IC50
[%]
Inhib[f]
[%]
[%]
[mM]
[mM]
[g]
46-(Z) 470Æ2 23Æ1
46-(E)
77Æ1 46Æ1
–
33Æ6
3.85Æ0.06 55Æ4
[g]
1.09Æ0.02 62Æ2
–
26Æ5
[a] FP assay: fluorescent probe (15 nm), UGM (580 nm). [b] HPLC assay: I
(Inhibitor, 500 mm), reduced UGM (60 nm), UDP-Galf (25 mm). [c] I (range,
see the Supporting Information for details), UDP-Galf (3 mm), acceptor
(2 mm). [d] I (4 mm), UDP-Galf (3 mm), acceptor (2 mm). [e] I (range, see
the Supporting Information for details), UDP-Galf (3 mm), acceptor
(2 mm). [f] I (4 mm), UDP-Galf (3 mm), acceptor (2 mm). [g] IC50 values
were determined only for compounds displaying >50% inhibition at I=
4 mm.
(Km(GlfT1)=566 mm;[31] Km(GlfT2)=380 mm[12f]). These results are
consistent with the inhibition data measured for glycosyltrans-
ferases in general. Indeed, the very best nucleotide-sugar ana-
logues reported as glycosyltransferase inhibitors usually display
inhibition levels at concentrations around the Km of the
enzyme-donor substrate.[14,32] This study is the first example in
which the relative affinities of the same donor analogues are
measured against both M. tuberculosis galactofuranosyltransfer-
ases. Perhaps most interesting, these experiments show that
the two enzymes do not accommodate UDP-Galf the same
way, because they show a distinct binding preference for the
two Z/E diastereomers.
While GlfT1 has no well characterized inhibitor to date, sev-
eral attempts have been made to inhibit GlfT2; Galf,[12a,33]
UDP,[12e] UDP-Galf,[12b,c] and acceptor analogues[12d] have been
produced and assayed against this transferase. The inhibition
level usually observed is in the same range like that seen for
46, which leaves open the question of how to design tight-
binding GlfT inhibitors.
The 3D structures of GlfT2, free and in complex with UDP,
have been recently obtained by X-ray crystallography.[34] These
structures revealed a unique tetrameric topology that shed
light on the mechanism of the galactan polymerization, a cen-
tral and debated question.[9c,35] In absence of X-ray structures
of both GlfT1 and GlfT2 in complex with UDP-Galf (or at least
UDP), the experiments detailed in this study provide useful in-
formation about the relative binding modes of the nucleotide
sugars within the catalytic pockets of both transferases.
Conclusion
This study reports a novel methodology to synthesize exo-gly-
cals bearing two electron-withdrawing groups, a sulfone and
a phosphonate. This methodology was generalized to eight
different carbohydrates from the furanose and pyranose series.
The Z/E configurations of these tetrasubstituted enol ethers
Figure 3. Structures of acceptor substrates 47 and 48 employed for the
GlfT1 and GlfT2 assays, respectively.
&
&
Chem. Eur. J. 2016, 22, 1 – 9
www.chemeurj.org
6
ꢁ 2016 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
ÝÝ These are not the final page numbers!

Full Paper
vin, V. Ferrieres, T. L. Lowary, R. Daniellou, K. Mikusova, Chem. Biol. 2010,
17, 1356–1366.
[4] S. K. Angala, J. M. Belardinelli, E. Huc-Claustre, W. H. Wheat, M. Jackson,
Crit. Rev. Biochem. Mol. Biol. 2014, 49, 361–399.
[5] a) P. M. Nassau, S. L. Martin, R. E. Brown, A. Weston, D. Monsey, M. R.
McNeil, K. Duncan, J. Bacteriol. 1996, 178, 1047–1052; b) J. N. Barlow,
M. E. Girvin, J. S. Blanchard, J. Am. Chem. Soc. 1999, 121, 6968–6969;
c) Q. Zhang, H.-w. Liu, J. Am. Chem. Soc. 2000, 122, 9065–9070; d) S.
Bornemann, Nat. Prod. Rep. 2002, 19, 761–772; e) S. O. Mansoorabadi,
C. J. Thibodeaux, H.-w. Liu, J. Org. Chem. 2007, 72, 6329–6342; f) M. Sol-
tero-Higgin, E. E. Carlson, T. D. Gruber, L. L. Kiessling, Nat. Struct. Mol.
Biol. 2004, 11, 539–543.
could be ascertained using NMR spectroscopic techniques. De-
protection of the exo-glycals followed by an UMP coupling
generated two original UDP-Galf analogues 46-(E) and 46-(Z).
The Z/E isomers were independently evaluated as inhibitors of
UGM, GlfT1, and GlfT2, the three mycobacterial galactofura-
nose processing enzymes. Molecule 46-(E) was, thus, the first
characterized inhibitor of GlfT1 reported to date and was also
found to efficiently inhibit UGM in a reversible manner. Inter-
estingly, GlfT2 showed a better affinity for 46-(Z).
This study nicely complements the pioneering work of Fer-
riꢃres, Mikusova, and Daniellou who compared a series of syn-
thetic UDP-furanoses on cell-free assays, in which both GlfT1
and GlfT2 were present.[3d] Although this earlier study did not
report the relative affinities of these nucleotide sugars towards
the enzymes, it showed that some of the molecules were sub-
strates of GlfT1. These compounds were found to inhibit the
galactan polymerization by producing short “dead-end” inter-
mediates, resulting from the incorporation of the modified car-
bohydrate residues into the growing carbohydrate chain.
The three enzymes studied in the present work are not only
interesting because, mechanistically, they are still the topic of
intense investigations, but also because they constitute very
important targets for the development of novel antimycobac-
terial agents. A very recent study showed that GlfT2 plays
a key role in the spatiotemporal cell growth and cell-envelope
elongation in mycobacteria,[36] thus reinforcing the interest in
the fundamental understanding of this enzyme.
[6] a) S. Berg, D. Kaur, M. Jackson, P. J. Brennan, Glycobiology 2007, 17,
35R–56R; b) P.-H. Tam, T. L. Lowary, Curr. Opin. Chem. Biol. 2009, 13,
618–625; c) M. R. Richards, T. L. Lowary, ChemBioChem 2009, 10, 1920–
1938.
[7] a) L. J. Alderwick, L. G. Dover, N. Veerapen, S. S. Gurcha, L. Kremer, D. L.
Roper, A. K. Pathak, R. C. Reynolds, G. S. Besra, Protein Expression Purif.
2008, 58, 332–341; b) M. A. Martinez Farias, V. A. Kincaid, V. R. Annama-
lai, L. L. Kiessling, J. Am. Chem. Soc. 2014, 136, 8492–8495.
[8] N. L. Rose, G. C. Completo, S.-J. Lin, M. McNeil, M. M. Palcic, T. L. Lowary,
J. Am. Chem. Soc. 2006, 128, 6721–6729.
[9] a) J. F. May, R. A. Splain, C. Brotschi, L. L. Kiessling, Proc. Natl. Acad. Sci.
USA 2009, 106, 11851–11856; b) A. Bernardi, J. Jimꢀnez-Barbero, A. Cas-
nati, C. De Castro, T. Darbre, F. Fieschi, J. Finne, H. Funken, K.-E. Jaeger,
M. Lahmann, T. K. Lindhorst, M. Marradi, P. Messner, A. Molinaro, P. V.
Murphy, C. Nativi, S. Oscarson, S. Penades, F. Peri, R. J. Pieters, O. Renau-
det, J.-L. Reymond, B. Richichi, J. Rojo, F. Sansone, C. Schaeffer, W. B.
Turnbull, T. Velasco-Torrijos, S. Vidal, S. Vincent, T. Wennekes, H. Zuilhof,
A. Imberty, Chem. Soc. Rev. 2013, 42, 4709–4727; c) M. R. Levengood,
R. A. Splain, L. L. Kiessling, J. Am. Chem. Soc. 2011, 133, 12758–12766.
[10] a) G. C. Completo, T. L. Lowary, J. Org. Chem. 2008, 73, 4513–4525;
b) M. G. Szczepina, R. B. Zheng, G. C. Completo, T. L. Lowary, B. M. Pinto,
ChemBioChem 2009, 10, 2052–2059; c) M. G. Szczepina, R. B. Zheng,
G. C. Completo, T. L. Lowary, B. M. Pinto, Bioorg. Med. Chem. 2010, 18,
5123–5128.
Experimental Section
[11] a) C. Ansiaux, I. N’go, S. P. Vincent, Chem. Eur. J. 2012, 18, 14860–14866;
b) S. El Bkassiny, I. N’Go, C. M. Sevrain, A. Tikad, S. P. Vincent, Org. Lett.
2014, 16, 2462–2465; c) I. N’Go, S. Golten, A. Ardꢄ, J. CaÇada, J. Jimꢀ-
nez-Barbero, B. Linclau, S. P. Vincent, Chem. Eur. J. 2014, 20, 106–112;
d) K. E. van Straaten, C. M. Sevrain, S. A. Villaume, J. Jimꢀnez-Barbero, B.
Linclau, S. P. Vincent, D. A. R. Sanders, J. Am. Chem. Soc. 2015, 137,
1230–1244; e) A. Caravano, S. P. Vincent, Eur. J. Org. Chem. 2009, 1771–
1780; f) G. Eppe, P. Peltier, R. Daniellou, C. Nugier-Chauvin, V. Ferriꢃres,
S. P. Vincent, Bioorg. Med. Chem. Lett. 2009, 19, 814–816; g) A. Caravano,
P. Sinay¨, S. P. Vincent, Bioorg. Med. Chem. Lett. 2006, 16, 1123–1125.
[12] a) J. Li, T. L. Lowary, Med. Chem. Commun. 2014, 5, 1130–1137; b) M. B.
Poulin, R. Zhou, T. L. Lowary, Org. Biomol. Chem. 2012, 10, 4074–4087;
c) K. Vembaiyan, J. A. Pearcey, M. Bhasin, T. L. Lowary, W. Zou, Bioorg.
Med. Chem. 2011, 19, 58–66; d) J. Frigell, J. A. Pearcey, T. L. Lowary, I.
Cumpstey, Eur. J. Org. Chem. 2011, 1367–1375; e) A. E. Trunkfield, S. S.
Gurcha, G. S. Besra, T. D. Bugg, Bioorg. Med. Chem. 2010, 18, 2651–
2663; f) N. L. Rose, R. B. Zheng, J. Pearcey, R. Zhou, G. C. Completo, T. L.
Lowary, Carbohydr. Res. 2008, 343, 2130–2139.
All the experimental details are gathered in the Supporting Infor-
mation.
Acknowledgements
This work was funded by the FNRS (Mandat de Chargꢀ de re-
cherche for A.T., F.R.I.A. fellowship to C.J-M.F.), the China Schol-
arship Council (CSC, PhD grant NO.201308520011to J.F.), and
the Alberta Glycomics Centre. A.K. thanks the Naito Foundation
for a fellowship. We thank Prof. Laura Kiessling and Prof. Pablo
Sobrado for the plasmid constructions of M. tuberculosis UGM.
Keywords: glycals
· glycosyltransferase · phosphonates ·
[13] a) A. Caravano, H. Dohi, P. Sinay¨, S. P. Vincent, Chem. Eur. J. 2006, 12,
3114–3123; b) L. Dumitrescu, G. Eppe, A. Tikad, W. Pan, S. E. Bkassiny,
S. S. Gurcha, A. Ardꢄ, J. Jimꢀnez-Barbero, G. S. Besra, S. P. Vincent, Chem.
Eur. J. 2014, 20, 15208–15215.
[14] a) M. D. Burkart, S. P. Vincent, A. Dꢅffels, B. W. Murray, S. V. Ley, C.-H.
Wong, Bioorg. Med. Chem. 2000, 8, 1937–1946; b) L. Tedaldi, G. K.
Wagner, Med. Chem. Commun. 2014, 5, 1106–1125.
[15] a) J. J. Reddick, J. Cheng, W. R. Roush, Org. Lett. 2003, 5, 1967–1970;
b) L. S. Brinen, E. Hansell, J. Cheng, W. R. Roush, J. H. McKerrow, R. J.
Fletterick, Structure 2000, 8, 831–840; c) E. Dunny, P. Evans, Curr. Bioact.
Compd 2011, 7, 218–236.
[16] a) M. Lakhrissi, Y. Chapleur, Angew. Chem. Int. Ed. Engl. 1996, 35, 750–
752; Angew. Chem. 1996, 108, 833–834; b) Y. Lakhrissi, C. Taillefumier,
M. Lakhrissi, Y. Chapleur, Tetrahedron: Asymmetry 2000, 11, 417–421;
c) Y. Lakhrissi, C. Taillefumier, F. Chretien, Y. Chapleur, Tetrahedron Lett.
2001, 42, 7265–7268.
sulfones · tuberculosis
[1] a) S. M. Blower, T. Chou, Nat. Med. 2004, 10, 1111 – 1116 ; b) A. Koul, E. Ar-
noult, N. Lounis, J. Guillemont, K. Andries, Nature 2011, 469, 483–490.
[2] a) G. Eppe, S. E. Bkassiny, S. P. Vincent, Galactofuranose Biosynthesis: Dis-
covery, Mechanisms and Therapeutic Relevance, The Royal Society of
Chemistry, Cambridge, 2015, pp. 209–241; b) I. Chlubnova, L. Legentil,
R. Dureau, A. Pennec, M. Almendros, R. Daniellou, C. Nugier-Chauvin, V.
Ferriꢃres, Carbohydr. Res. 2012, 356, 44–61.
[3] a) K. Mikusova, T. Yagi, R. Stern, M. R. McNeil, G. S. Besra, D. C. Crick, P. J.
Brennan, J. Biol. Chem. 2000, 275, 33890–33897; b) L. Kremer, L. G.
Dover, C. Morehouse, P. Hitchin, M. Everett, H. R. Morris, A. Dell, P. J.
Brennan, M. R. McNeil, C. Flaherty, K. Duncan, G. S. Besra, J. Biol. Chem.
2001, 276, 26430–26440; c) D. C. Crick, S. Mahapatra, P. J. Brennan, Gly-
cobiology 2001, 11, 107R–118R; d) P. Peltier, M. Belanova, P. Dianiskova,
R. Zhou, R. B. Zheng, J. A. Pearcey, M. Joe, P. J. Brennan, C. Nugier-Chau-
Chem. Eur. J. 2016, 22, 1 – 9
www.chemeurj.org
7
ꢁ 2016 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
&
&
These are not the final page numbers! ÞÞ

Full Paper
[17] a) F. K. Griffin, D. E. Paterson, P. V. Murphy, R. J. K. Taylor, Eur. J. Org.
Chem. 2002, 1305–1322; b) Y. Ohnishi, Y. Ichikawa, Bioorg. Med. Chem.
Lett. 2002, 12, 997–999; c) G. D. McAllister, D. E. Paterson, R. J. K. Taylor,
Angew. Chem. Int. Ed. 2003, 42, 1387–1391; Angew. Chem. 2003, 115,
1425–1429; d) R. J. K. Taylor, G. D. McAllister, R. W. Franck, Carbohydr.
Res. 2006, 341, 1298–1311; e) P. Pasetto, R. W. Franck, J. Org. Chem.
2003, 68, 8042–8060.
Chem. 1968, 33, 1780–1783; c) K. Bennis, P. Calinaud, J. Gelas, M.
Ghobsi, Carbohydr. Res. 1994, 264, 33–44.
[25] G. Eppe, L. Dumitrescu, O. Pierrot, T. Li, W. Pan, S. P. Vincent, Chem. Eur.
J. 2013, 19, 11547–11552.
[26] a) A. Caravano, W. Pan, S. P. Vincent, Arkivoc 2007, x, 348–364; b) A. Car-
avano, D. Mengin-Lecreulx, J.-M. Brondello, S. P. Vincent, P. Sinay¨, Chem.
Eur. J. 2003, 9, 5888–5898.
[27] a) D. R. Williams, D. L. Brown, J. W. Benbow, J. Am. Chem. Soc. 1989, 111,
1923–1925; b) V. Liautard, V. Desvergnes, K. Itoh, H.-w. Liu, O. R. Martin,
J. Org. Chem. 2008, 73, 3103–3115.
[28] a) T. Li, A. Tikad, W. Pan, S. P. Vincent, Org. Lett. 2014, 16, 5628–5631;
b) P. Dabrowski-Tumanski, J. Kowalska, J. Jemielity, Eur. J. Org. Chem.
2013, 2147–2154.
[18] a) M. Benltifa, S. Vidal, D. Gueyrard, P. G. Goekjian, M. Msaddek, J.-P.
Praly, Tetrahedron Lett. 2006, 47, 6143–6147; b) B. Bourdon, M. Corbet,
P. Fontaine, P. G. Goekjian, D. Gueyrard, Tetrahedron Lett. 2008, 49, 747–
749; c) D. Gueyrard, R. Haddoub, A. Salem, N. S. Bacar, P. G. Goekjian,
Synlett 2005, 520–522; d) D. Gueyrard, P. Fontaine, P. G. Goekjian, Syn-
thesis 2006, 1499–1503.
[29] M. Soltero-Higgin, E. E. Carlson, J. H. Phillips, L. L. Kiessling, J. Am. Chem.
Soc. 2004, 126, 10532–10533.
[19] a) W. B. Yang, C. F. Chang, S. H. Wang, C. F. Teo, C. H. Lin, Tetrahedron
Lett. 2001, 42, 4657–4660; b) W. B. Yang, C. Y. Wu, C. C. Chang, S. H.
Wang, C. F. Teo, C. H. Lin, Tetrahedron Lett. 2001, 42, 6907–6910; c) W.-
B. Yang, Y.-Y. Yang, Y.-F. Gu, S.-H. Wang, C.-C. Chang, C.-H. Lin, J. Org.
Chem. 2002, 67, 3773–3782.
[20] R. N. Monrad, R. Madsen, Tetrahedron 2011, 67, 8825–8850.
[21] a) A. Gꢆmez, A. Pedregosa, A. Barrio, S. Valverde, C. Lopez, Tetrahedron
Lett. 2004, 45, 6307–6310; b) A. M. Gꢆmez, A. Barrio, A. Pedregosa, C.
Uriel, S. Valverde, J. C. Lopez, Eur. J. Org. Chem. 2010, 2910–2920.
[22] a) A. Novoa, N. Pellegrini-Moise, S. Lamande-Langle, Y. Chapleur, Tetrahe-
dron Lett. 2009, 50, 6484–6487; b) A. M. Gꢆmez, G. O. Danelon, A. Pe-
dregosa, S. Valverde, J. C. Lopez, Chem. Commun. 2002, 2024–2025;
c) A. Novoa, N. Pellegrini-Moise, S. Bourg, S. Thoret, J. Dubois, G.
Aubert, T. Cresteil, Y. Chapleur, Eur. J. Med. Chem. 2011, 46, 3570–3580;
d) H.-T. T. Thien, A. Novoa, N. Pellegrini-Mo¨ıse, F. Chrꢀtien, C. Didierjean,
Y. Chapleur, Eur. J. Org. Chem. 2011, 6939–6951.
[30] A. Caravano, S. P. Vincent, P. Sinay¨, Chem. Commun. 2004, 1216–1217.
[31] A. Koizumi, R. B. Zhang, T. L. Lowary, unpublished results.
[32] a) S. Wang, S. Vidal, Carbohydr. Chem. 2013, 39, 78–101; b) H. Dohi, R.
Pꢀrion, M. Durka, M. Bosco, Y. Rouꢀ, F. Moreau, S. Grizot, A. Ducruix, S.
Escaich, S. P. Vincent, Chem. Eur. J. 2008, 14, 9530–9539; c) M. D. Bur-
kart, S. P. Vincent, C.-H. Wong, Chem. Commun. 1999, 1525–1526.
[33] S. Cren, A. J. Blake, G. S. Besra, N. R. Thomas, S. S. Gurcha, Org. Biomol.
Chem. 2004, 2, 2418–2420.
[34] R. W. Wheatley, R. B. Zheng, M. R. Richards, T. L. Lowary, K. K. S. Ng, J.
Biol. Chem. 2012, 287, 28132–28143.
[35] J. F. May, M. R. Levengood, R. A. Splain, C. D. Brown, L. L. Kiessling, Bio-
chemistry 2012, 51, 1148–1159.
[36] J. M. Hayashi, C.-Y. Luo, J. A. Mayfield, T. Hsu, T. Fukuda, A. L. Walfield,
S. R. Giffen, J. D. Leszyk, C. E. Baer, O. T. Bennion, A. Madduri, S. A. Shaff-
er, B. B. Aldridg, C. M. Sassetti, S. J. Sandler, T. Kinoshit, D. B. Moody, Y. S.
Morita, Proc. Natl. Acad. Sci. USA 2016, 113, 5400–5405.
[23] A. M. Gꢆmez, A. Barrio, I. Amurrio, S. Valverde, S. Jarosz, J. C. Lopez, Tet-
rahedron Lett. 2006, 47, 6243–6246.
[24] a) J.-S. Thomann, F. Monneaux, G. Creusat, M. V. Spanedda, B. Heurtault,
C. Habermacher, F. Schuber, L. Bourel-Bonnet, B. Frisch, Eur. J. Med.
Chem. 2012, 51, 174–183; b) L. M. Lerner, B. D. Kohn, P. Koh, J. Org.
Received: July 1, 2016
Published online on && &&, 0000
&
&
Chem. Eur. J. 2016, 22, 1 – 9
www.chemeurj.org
8
ꢁ 2016 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
ÝÝ These are not the final page numbers!

Full Paper
FULL PAPER
New antimycobacterial agents: This
study reports a new methodology to
synthesize exo-glycals bearing both
a sulfone and a phosphonate, which
was applied to prepare two original
UDP-galactofuranose analogues. The
latter were evaluated as inhbitors of
UGM, GlfT1, and GlfT2, three enzymes
involved in the Mycobacterium tubercu-
losis cell wall biosynthesis.
&
Carbohydrates
C. J.-M. Frꢀdꢀric, A. Tikad, J. Fu, W. Pan,
R. B. Zheng, A. Koizumi, X. Xue,
T. L. Lowary, S. P. Vincent*
&& – &&
Synthesis of Unprecedented
Sulfonylated Phosphono-exo-Glycals
Designed as Inhibitors of the Three
Mycobacterial Galactofuranose
Processing Enzymes
Chem. Eur. J. 2016, 22, 1 – 9
www.chemeurj.org
9
ꢁ 2016 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
&
&
These are not the final page numbers! ÞÞ