Rational Design of Glycomimetic Compounds Targeting the Saccharomyces cerevisiae Transglycosylase Gas2

Article abstract of DOI:10.1111/cbdd.12650

The transglycosylase Saccharomyces cerevisiae Gas2 (ScGas2) belongs to a large family of enzymes that are key players in yeast cell wall remodeling. Despite its biologic importance, no studies on the synthesis of substrate-based compounds as potential inhibitors have been reported. We have synthesized a series of docking-guided glycomimetics that were evaluated by fluorescence spectroscopy and saturation-transfer difference (STD) NMR experiments, revealing that a minimum of three glucose units linked via a β-(1,3) linkage are required for achieving molecular recognition at the binding donor site. The binding mode of our compounds is further supported by STD-NMR experiments using the active site-mutants Y107Q and Y244Q. Our results are important for both understanding of ScGas2-substrate interactions and setting up the basis for future design of glycomimetics as new antifungal agents.

Full text of DOI:10.1111/cbdd.12650

Chem Biol Drug Des 2016; 87: 163–170
Editor’s Choice
Rational Design of Glycomimetic Compounds
Targeting the Saccharomyces cerevisiae
Transglycosylase Gas2
1
,2,
3
Ignacio Delso *, Jessika Valero-Gonz aꢀ lez ,
pharmaceutical targets (1,2). In particular, transglycosy-
lases that belong to GH72 in the CAZy database (3) are
ubiquitous enzymes present in fungal organisms. These
glycosylphosphatidylinositol (GPI)-anchored enzymes are
crucial for remodeling of fungal cell wall (4,5), which is an
essential structure in scaffolding the cytoplasmic mem-
brane and maintaining structural integrity of those micro-
organisms. While some information about the enzymes
responsible for the cell wall biosynthesis is available
1
1
Eduardo Marca , Tom aꢀ s Tejero , Ram oꢀ n
Hurtado-Guerrero * and Pedro Merino *
3
,4,
1,
1
Laboratorio de S ꢀı ntesis Asim eꢀ trica, Departamento de
S ꢀı ntesis y Estructura de Biomol eꢀ culas, Instituto de S ꢀı ntesis
Qu ꢀı mica y Cat aꢀ lisis Homog eꢀ nea (ISQCH), CSIC,
Universidad de Zaragoza, Zaragoza, Arag oꢀ n 50009, Spain
Servicio de Resonancia Magn eꢀ tica Nuclear, Centro de
Qu ꢀı mica y Materiales de Arag oꢀ n (CEQMA), CSIC,
Universidad de Zaragoza, Campus San Francisco,
Zaragoza, Arag oꢀ n 50009, Spain
Instituto de Biocomputaci oꢀ n y Fisica de Sistemas
Complejos (BIFI), BIFI-IQFR (CSIC) Joint Unit, Universidad
2
(6–10), little is known of the involvement of these transgly-
cosylases in the construction and remodeling of the fungal
cell wall (11).
3
de Zaragoza, Zaragoza, Arag oꢀ n 50009, Spain
Fundaci oꢀ n ARAID, Gobierno de Arag oꢀ n, Zaragoza,
Arag oꢀ n 50009, Spain
These transglycosylases are classified as glycosyl hydro-
lases due to their typical folding, which consists of a TIM
4
barrel domain formed by (ba)
structure resembles a glycosyl hydrolase, its activity shows
balance among hydrolysis and transglycosylation
8
(12,13). Although the overall
*
Corresponding authors: Ignacio Delso, idelso@unizar.es;
Ram oꢀ n Hurtado-Guerrero, rhurtado@bifi.es; Pedro Merino,
pmerino@unizar.es
a
depending on the length of the substrate. For example,
Saccharomyces cerevisiae Gas2 (ScGas2; Figure 1) is inac-
tive against laminaripentaose (G5) but is a pure hydrolase
against laminarihexaose and heptaose and becomes a
transglycosylase against larger sized laminarioligosaccha-
rides (13). In the case of Aspergillus fumigatus Gel1, lami-
narinonaose was the minimal tested size to have
transglycosylase activity (14). Several efforts have been
made on elucidating the mechanism of action of transgly-
cosylases (15–18), and several antibiotics that inhibit the
transglycosylation step in bacteria have been designed (19).
The transglycosylase Saccharomyces cerevisiae Gas2
(
ScGas2) belongs to a large family of enzymes that are
key players in yeast cell wall remodeling. Despite its bio-
logic importance, no studies on the synthesis of sub-
strate-based compounds as potential inhibitors have
been reported. We have synthesized a series of dock-
ing-guided glycomimetics that were evaluated by fluo-
rescence
spectroscopy
and
saturation-transfer
difference (STD) NMR experiments, revealing that a min-
imum of three glucose units linked via a b-(1,3) linkage
are required for achieving molecular recognition at the
binding donor site. The binding mode of our compounds
is further supported by STD-NMR experiments using the
active site-mutants Y107Q and Y244Q. Our results are
important for both understanding of ScGas2–substrate
interactions and setting up the basis for future design of
glycomimetics as new antifungal agents.
This family of fungal enzymes has key important biologic
roles in yeast and fungi and in some cases are even
essential for pathogens such as Aspergillus fumigatus (20).
Consequently, the targeting of these enzymes with specific
inhibitors might represent potential therapies to treat selec-
tively fungi-related pathologies such as Aspergillosis and
Candidiasis that constitute important current medical
problems (21–29).
Key words: carbohydrates, docking, glycomimetics, STD-
NMR, transglycosylases
Received 2 July 2015, revised 1 August 2015 and accepted
for publication 11 August 2015
Despite of their biologic importance, the unique enzyme
from this family whose tridimensional structure is described
is ScGas2. Several X-ray structures have been resolved
(13), including the apo-structure (E176Q mutant, PDB
entry 2W61) and structures in complex with laminaripenta-
ose (PDB entry 2W62) and with laminaritriose and
Enzymes that can transfer single or multiple activated car-
bohydrate units to a range of substrates are involved in
important biologic processes and therefore are potential
ª 2015 John Wiley & Sons A/S. doi: 10.1111/cbdd.12650
163

Delso et al.
Figure 2: O-glycosides derived from b-(1,3)-D-glucans.
molecular docking studies, STD-NMR experiments and
tryptophan fluorescence spectroscopy, and applied this
information to unambiguously determine the minimal sub-
strate length required for the design and synthesis of
glycomimetics for the ScGas2 enzyme.
Figure 1: Surface representation of ScGas2 in complex with two
G5 units at donor (positions ꢀ5 to ꢀ1) and acceptor (positions +1
to +5) sites.
laminaritetraose (PDB entry 2W63). From these crystallo-
graphic data, the architecture of this family of enzymes
was described and their catalytic mechanism was
proposed. The active site of ScGas2 shows two catalytic
residues, the nucleophile E275 and the acid–base E176,
and four well-conserved tyrosine residues (Y107, Y244,
Y307, and Y316). To explain the transglycosylation mecha-
nism, the ‘base occlusion mechanism’ was proposed, in
which the leaving hydrolyzed sugar blocks the entrance of
a key water molecule and thus avoids hydrolysis, favoring
transglycosylation (13).
Methods and Materials
NMR studies
NMR spectra were recorded in a Bruker 500 MHz spec-
trometer, equipped with a direct TBO probe. H spectra
1
were recorded with water signal suppression using excita-
tion sculpting with gradients (Pulse program zgesgp).
Water saturation pulse was adjusted to 2 ms with a
Sinc1.1000 shape. Saturation-transfer difference spectra
were acquired with water suppression using excitation
sculpting with gradients (Pulse program stddiffesgp.3). Pro-
tein saturation pulse was adjusted to 50 ms, with an Ebur-
p2.1000 shape during 2 seconds. Saturation frequency
was optimized to ꢀ0.5 ppm when on resonance and
40 ppm when off-resonance. Relaxation delay was
adjusted to 1.5 seconds and acquisition time to 2.5 sec-
onds. For both on- and off-resonance, 1024 spectra were
acquired. Sample temperature was fixed to 298 K. All mea-
sured samples were prepared on deuterated-TRIS 25 mM
in deuterium oxide at pH 8.0, dissolving ligand to 1 mM
and corresponding protein to 20 lM. STD spectra were
recorded with separated solution components to check
blank. Only protein signals between 1.25 and 0.50 ppm
appear as background, and the TRIS signal at 3.60 ppm.
Our aim was to design glycomimetics as potential inhibi-
tors targeting essential transglycosylases of the fungal cell
wall. In particular, it is well known that Gel4, a homologue
of ScGas2, is essential in Asp. fumigatus, and conse-
quently, the design of glycomimetics for this enzyme might
become future drugs to treat Aspergillosis. Due to that we
could only express and purify ScGas2, we decided to use
ScGas2 as a template for AfGel4. The study of the identity
between ScGas2 and different homologues was already
reported (13), and from these data, it was very clear that
overall all members of this family conserve all the important
residues involved in catalysis and sugar recognition. Of
note, ScGas2 might serve as a template to discover inhibi-
tors for AfGel4 given the high identity among them at the
active site level (the identity between the catalytic domains
is ~50% and reaches 100% when all the key residues rec-
ognizing the carbohydrates from ꢀ3 to +1 are considered).
This makes ScGas2 very appealing for the design of new
glycomimetics not only targeting ScGas2 but also AfGel4.
Recently, we developed a synthetic strategy for the prepa-
ration of O- and C-glycosides derived from b-(1,3)-D-
glucans (30) that demonstrated its utility for the synthesis
of di- and trisaccharide analogues.
Docking studies
Compounds 1–3 were docked into the catalytic site of
ScGas2 (PDB ID 2W62) using Glide software from
Schr o€ dinger. Protein was prepared with the Protein Prepa-
ration Wizard, and the water molecules and the acceptor
oligosaccharide were removed. Grid was prepared with a
ꢁ
box size of 45 9 45 9 45 A, centered on the donor
oligosaccharide, using OPLS-2005 force field. Ligands
were designed with Maestro, prepared with LigPrep, and
minimized using Gaussian 09 (31), with a convergence
threshold of 0.05 and 5000 maximum iterations. Afterward,
MacroModel was used to generate 1000 different confor-
mations for each ligand. Glide was run on XP mode (Extra
Precision) using as input all calculated conformations.
Herein, we have further explored the combination of differ-
ent radicals covalently bound to variable laminooligosac-
charides; in particular, we fixed our attention on
compounds 1–3 (Figure 2). The interactions of these com-
pounds with ScGas2 were analyzed by computational
1
64
Chem Biol Drug Des 2016; 87: 163–170

Glycomimetics for transglycosylases
Schr o€ dinger Suite 2013 was used for all computational
calculations: MAESTRO 9.4 for viewing and protein prepara-
tion; LIGPREP 2.6 for ligand preparation; MACROMODEL 10.0 for
ligand energy minimization; GLIDE 5.9 for molecular docking.
Inhibition assays
Activity was measured monitoring laminarihexaose (G6)
hydrolysis obtaining as major products laminaritetraose,
laminaritriose, and laminaribiose. Reactions took place at
3
0 °C on a final volume of 100 lL of ammonium formate-
buffered solution (10 mM) at pH 5.5, containing fixed con-
centrations of ScGas2 (5 lM) and G6 (0.5 mM, Megazyme,
Ireland) and variable concentrations of compound 3 (100,
200 and 500 lM, 1, 2 and 5 mM). Reaction was monitored
every 30 min, over 3 h, with two repetitions. Hydrolysis
ratio was determined by comparison of G6 concentration
(mean of two measurements). IC50 was calculated by
extrapolating the inhibitor concentration at which G6
hydrolysis is half of obtained in absence of compound 3.
Cloning, expression, and purification
The cloning was carried out as described before (13) with
the exception that the DNA sequence encoding to the
amino acids residues 26–525 of the S. cerevisiae Gas2
which was obtained by PCR from the S. cerevisiae strain
BY4741 kindly provided by Prof. Javier Arroyo (Universidad
Complutense de Madrid, Spain). The double mutant
N498D/N510D, referred to here as the wild type, and the
mutants Y244Q and Y107Q were also generated as
shown before (13).
Oligosaccharides were determined by UPLC chromatogra-
phy on a 2.1 mm i.d. 9 50 mm ACQUITY 1.7 lM BEH Amide
(Waters Corp, Milford, MA, USA) using an ACQUITY UPLC
system (Waters Corp). The column was maintained at
35 °C. Mobile phase, at a flow rate of 0.800 mL/min, con-
sisted of 68% acetonitrile and 32% water, both solvents
doped with 0.2% triethylamine. One microlitre of sample
was injected without any pretreatment. Carbohydrates
were detected with an Evaporative Light Scattering Detec-
tor (ELSD; Waters Corp) with optimized conditions as fol-
lows: temperature of drift tube 100 °C, nebulizer
temperature 45 °C, gain 1000, and nitrogen pressure
30 psi. G6 retention time, on these conditions, is 58–
62 seconds.
The expression system used was Pichia Pastoris strain
X33 (Invitrogen, Grand Island, NY, USA). Batch cultures
were performed in 1 liter of BMGY medium (1% (w/v)
yeast extract, 2% (w/v) peptone, 100 mM potassium phos-
phate (pH 6.0), 1.34% (w/v) yeast nitrogen based and 1%
(
v/v) glycerol) overnight at 30 °C, then centrifuged at
000 g for 10 min. Cells were resuspended in BMMY
medium (1% (w/v) yeast extract, 2% (w/v) peptone,
00 mM potassium phosphate pH 6.0, 1.34% (w/v) yeast
nitrogen base, and 1% (v/v) methanol) and incubated at
8 °C. Supernatant containing ScGas2 and mutants was
4
1
1
collected after 72 h of methanol induction and concen-
trated to 20–50 mL using a Pellicon XL device (Millipore,
Billerica, MA, USA), and dialyzed against 25 mM Tris and
5
mM sodium phosphate pH 7.5. Samples were loaded
Results
into 1 9 5 mL CHT Type I cartridge (Bio-Rad, Hercules,
CA, USA), previously equilibrated with 25 mM Tris and
Compounds 1–3 were prepared by glycosylation of the
corresponding glycosyl bromide, derived from peracety-
lated laminarioligosaccharide following the procedure we
described previously (30) (Scheme 1). Spectroscopic data
5
mM sodium phosphate pH 7.5. The protein was eluted
with a sodium phosphate gradient (5–500 mM). Gel filtra-
tion was carried out using SUPERDEX 75 XK26/60 column in
buffer containing 25 mM Tris pH 7.5 and 150 mM NaCl.
The protein was dialyzed in Tris–HCl 25 mM pH 7.5.
Fluorescence spectroscopy assay
Fluorescence spectroscopy assay was carried out to eval-
uate the dissociation constants of ScGas2 WT against
compounds 1–3. These experiments were conducted in a
Cary Eclipse spectrofluorometer (Varian, Palo Alto, CA,
USA) at 25 °C. Protein concentration was 1 lM in every
case, and concentration of compounds 1–3 varied from
0.1 to 4 mM in buffer 25 mM Tris pH 7.5. Fluorescence
emission spectra were registered in the 300–450 nm
range with an excitation wavelength of 280 nm.
Data analysis was performed in Prism (GraphPad Software
Inc., La Jolla, CA, USA) according to the fitting equation
below:
Scheme 1: Synthesis of compounds 1–3. (i) 30% HBr, AcOH, rt,
Fi
fk½Qꢁ
1
ꢀ
¼
15 min (76%). (ii) Ph(CH ) OH, AgCO , CH Cl , rt, 20 h (63%). (iii)
2
5
3
2
2
F0
Kd þ ½Qꢁ
NaOMe, MeOH, rt, 4 h (quant.).
Chem Biol Drug Des 2016; 87: 163–170
165

Delso et al.
for compounds 1–3, as well as for intermediates, are avail-
able in the Appendix S1.
revealed notable differences in binding energy for each
derivative with ScGas2. Furthermore, poses obtained for
each compound showed a different dispersion on their
geometry. In particular, compound 3 showed lower pose
dispersion around the minimum energy pose.
Docking studies were initially performed with G5 on the
entire enzyme ScGas2 (PDB code 2W62) using the
Schr o€ dinger Software (32–34). Analogous studies were
carried out with glycomimetics 1–3, and the best affinity
was observed for compound 3. Differences of ca. 9 and
STD-NMR experiments (35,36) were measured on the cor-
responding ligand and ScGas2 mixtures. STD-NMR spec-
tra showed depreciable signal for compound 1 (Figure 4,
traces A and B), but a slight STD signal for compound 2
(Figure 4, traces C and D). In this latter case, a STD signal
4
kcal/mol were found between the lowest energy repre-
sentative cluster for compound 3 and for compounds 1
and 2, respectively (Figure 3). Calculations with 1–3
Figure 3: Graphical representation of Glide
Docking Energy (Kcal/mol) versus RMSD
calculated from the lowest energy pose for
each product, being 1 (blue), 2 (green), and
3
(red).
1
Figure 4: H spectra and STD-NMR
spectra measured for compounds 1–3.
STD spectra are scaled up 64 times. Traces
A and B correspond to compound 1; traces
C and D correspond to compound 2;
traces E and F correspond to compound 3.
1
66
Chem Biol Drug Des 2016; 87: 163–170

Glycomimetics for transglycosylases
Figure 5: STD-NMR spectra for Y244Q
mutant (trace B), Y107Q mutant (trace C),
and the wild-type enzyme (trace D) with 3.
1
H NMR spectrum of compound 3 is given
in trace A.
corresponding to aromatic protons, with an enhancement
of 1%, could be observed; methylene groups showed an
enhancement below 1%. On the other hand, no STD
signal was observed for glucose moieties. STD spectrum
for compound 3 showed more intense signals for the
whole molecule than the other two compounds (Figure 4,
traces E and F). Under the same conditions, both the aro-
matic ring and the pentyl spacer showed an enhancement
of 3%. Laminaritriose signal showed a STD enhancement
of 1% (except those lost due to solvent suppression).
mutant, and an enhancement below 1% for aromatic and
methylene protons when Y107Q mutant is present.
Tryptophan fluorescence studies (Figure 6) provided for
compound 3 K
compounds
changes impeding to obtain K
d
of 800 ꢂ 100 lM. Unlike compound 3,
and did not show dose–response
values.
1
2
d
Inhibition data were obtained by measuring the hydrolytic
activity of the enzyme. Laminarihexose (G6) has been used
as the substrate for hydrolysis, mainly rendering lami-
naribiose, laminaritriose, and laminaritetraose without any
measurable transglycosylation activity (13). Determination
of the hydrolytic activity of ScGas2 in the presence of
compound 3 rendered an IC50 value of 1.55 mM. No
effects on laminarihexaose hydrolysis were observed when
A comparison of the STD-NMR experiments carried out with
the wild-type protein (Figure 5, trace D), and mutants Y244Q
(Figure 5, trace B) and Y107Q (Figure 5, trace C), shows an
complete disappearance of STD signals for the Y244Q
1
and 2 were added to the reaction. An additional reaction
was monitored containing 3 (2 mM) and ScGas2, in the
Figure 6: Quenching of intrinsic ScGas2 tryptophan fluorescence
measured at increasing concentrations of the compound 3. All
data points represent the means ꢂ SD for three measurements.
The K
d
for compound 3 was determined by fitting fluorescence
Figure 7: Evolution of G6 hydrolysis in the presence 3 in different
intensity data against compound 3 concentration.
concentrations.
Chem Biol Drug Des 2016; 87: 163–170
167

Delso et al.
absence of substrate (G6). No evolution was observed
over 24 h at 30 °C (Figure 7).
study allowed us to conclude that the enzyme shows higher
affinity for sugars at position ꢀ3 and ꢀ2, suggesting that
potential glycomimetics should contain at least 2 units of
glucose, specifically b-(1,3) bonded.. The presence of that
hydrophobic pocket prompted us to consider glycomimet-
ics containing aromatic radicals, such as compounds 1–3,
as good candidates for binding the active site.
Discussion
Docking results clearly match the previously described
ꢁ
binding mode of G5 (13) (RMSD of 0.35 and 0.12 A for
sugars adopting the lowest representative energy cluster
located from ꢀ5 to ꢀ1, and +5 to +1, respectively) and
reveal a higher affinity of G5 for the donor site than for the
acceptor site of the protein (Figure 8, top panel). Indeed, a
strong correlation is observed between the G5 binding
mode with the lower energy calculated poses. These
docking studies showed that the conformation of residues
in positions ꢀ2 and ꢀ3 did not change to a large extent,
remaining fixed by several H-bond interactions (Figure 8,
bottom panel).
The most representative poses obtained from docking
studies for compounds 1–3 are illustrated in Figure 9 (top
panel). In all cases, the carbohydrate units remain at the
donor site. Compound 3 is the only one occupying posi-
tions ꢀ1 to ꢀ3, as found for the natural substrate. In fact,
calculated RMSD for the carbohydrate moiety between
laminaripentaose from the PDB entry 2W62 and com-
ꢀ
ꢁ
pound 3 is 0.174 A, being the maximum deviation of
ꢀ
ꢁ
0.291 A. Furthermore, the fact of 3 occupying the ꢀ1
position with the functionalized ring allows the phenylpentyl
moiety to overpass catalytic residues E176 and E275, and
place the phenyl group in the hydrophobic pocket, formed
by Y244, Y307, and Y316. An edge-to-face p–p stacking
of the phenyl group with Y244 is clearly evidenced
(Figure 9, bottom panel).
On the other hand, residues on ꢀ4 and ꢀ5 positions,
located far from the reaction site, show higher flexibility in all
studied conformations. Residue on position ꢀ1 is involved
in CH–p interactions with the phenyl ring of amino acid resi-
due Y107 and a key H-bond with hydroxyl group on residue
Y307. In addition, the reaction site is also characterized by
the presence of a hydrophobic pocket formed by the side
chains of residues Y244, Y307, and Y316. Thus, our in silico
STD-NMR experiments were designed to confirm our
docking results. Comparing derivatives 1–3 spectra (Fig-
ure 4), the highest signal intensity for all protons from 3
Figure 8: Close-up view of the ScGas2 crystal structure in
complex with G5 (top panel), and the minima energy poses
calculated for the complex within a 1 kcal/mol energy difference
Figure 9: Most representative poses for compounds 1 (blue), 2
(green), and 3 (red) bounded to ScGas2 (top panel). Close-up
view of the compound 3 docked into the active site of ScGas2
(bottom panel).
(bottom panel).
1
68
Chem Biol Drug Des 2016; 87: 163–170

Glycomimetics for transglycosylases
clearly indicates that its affinity is higher than the other two
glycomimetics. Besides, regarding the relative intensities
between the aglycone protons and the carbohydrate sig-
nals, it can be inferred that the phenylpentyl group is the
anchoring moiety that enhances the interaction with the
protein. STD-NMR experiments with mutants confirm this
hypothesis. The complete loss of STD signal observed for
Y244Q mutant (Figure 5, trace B) demonstrates the inter-
action of phenyl group with the tyrosine-rich pocket, in
particular the proposed p–p stacking with precisely Y244.
In the case of Y107Q mutant (Figure 5, trace C), only phe-
nylpentyl group interaction remains, meaning the loss of
the carbohydrate moiety interaction. These experimental
studies support the binding mode of compound 3 that is
inferred from docking and previous STD-NMR studies.
Acknowledgments
This study was supported by the Ministerio de Econom ꢀı a y
Competitividad (MINECO) and FEDER Program (Madrid,
Spain, projects CTQ2013-44367-C2-1-P and CTQ2013-
44367-C2-2-P), and the Gobierno de Arag oꢀ n (Zaragoza,
Spain. Bioorganic Chemistry group E-10 and Protein
Targets group B-89). We acknowledge the Institute of
Biocomputation and Physics of Complex Systems (BIFI) at
the University of Zaragoza for computer time at clusters
Terminus and Memento. E. M. and J. V.G. thank MEC for
FPU and FPI predoctoral grants, respectively. The ARAID
Foundation (Gobierno de Arag oꢀ n, Spain) is also acknowl-
edged for financial support.
Tryptophan fluorescence studies confirm that compounds
Conflict of Interest
1
and 2 have a very low affinity and that compound 3 is
the best compound of our series, showing a K of
d
The authors declare no conflict of interest.
8
00 ꢂ 100 lM. This value is significant as affinity of
ScGas2 with its own substrates is so low that it cannot be
measured, so the introduction of the aromatic moiety, con-
veniently linked to laminaritriose, has been able to increase
the affinity, probably, an order of magnitude.
References
1
. Desmet T., Soetaert W., Bojarova P., Kren V., Dijkhuizen
L., Eastwick-Field V., Schiller A. (2012) Enzymatic glyco-
sylation of small molecules: challenging substrates
require tailored catalysts. Chem Eur J;18:10786–10801.
. Desmet T., Soetaert W. (2011) Enzymatic glycosyl
transfer: mechanisms and applications. Biocat Bio-
trans;29:1–18.
The inhibition assay let us confirm that glycomimetic 3 is
able to bind the active site of ScGas2 better than a
substrate as laminarihexaose. Unfortunately, the IC50
value is too high to consider 3 as a good inhibitor. It is
also remarkable that no hydrolysis of compound 3 occurs
in presence of ScGas2.
2
3
. Cantarel B.L., Coutinho P.M., Rancurel C., Bernard T.,
Lombard V., Henrissat B. (2009) The carbohydrate-
active enzymes database (CAZy): an expert resource
for glycogenomics. Nucleic Acid Res;37:D233–D238.
. Reed P., Veiga H., Jorge A.M., Terrak M., Pinho M.G.
Conclusions
4
5
These studies open a door for a rational design of com-
pounds that could inhibit ScGas2 and close homologues
such as Aspergillus fumigatus Gel4, that in turn is a drug
target for Aspergillosis. We clearly demonstrate that a mini-
mum of three monosaccharide units are required for ensur-
ing a correct recognition at the donor binding site of the
enzyme, and that the correct orientation of the hydrophobic
radical toward the hydrophobic pocket increases the affinity
of the ligand. This is supported by STD-NMR experiments
that suggested that the hydrophobic radical acts as an ‘an-
choring moiety’ for the glycomimetics. Also, from the dock-
ing studies, it appears that the representative poses
present a CH–p interaction between the phenyl group and
residue Y244. STD-NMR experiments also showed that the
affinities observed with wild-type ScGas2 are completely
loss with the corresponding mutants Y107Q and Y244Q,
thus confirming the specific interactions between these
residues and the sugar at position ꢀ1 and the hydrophobic
radical. Fluorescence studies and inhibition experiments
also confirmed the above results. Further design of other
trisaccharide analogues bearing other groups at the
anomeric center of residue ꢀ1 is currently being pursued,
and they will be reported in due course.
(2011) Monofunctional transglycosylases are not essen-
tial for Staphylococcus aureus cell wall synthesis. J
Bacteriol;193:2549–2556.
. Lovering A.L., de Castro L.H., Lim D., Strynadka N.C.
(
2007) Structural insight into the transglycosylation step
of bacterial cell-wall biosynthesis. Science;315:1402–
405.
1
6
7
8
. Farkas V. (2003) Structure and biosynthesis of fungal
cell walls: methodological approaches. Folia Micro-
biol;48:469–478.
. Bowman S.M., Free S.J. (2006) The structure and
synthesis of the fungal cell wall. BioEssays;28:799–
8
08.
. Engel J., Schmalhorst P.S., Routier F.H. (2012) Bio-
synthesis of the fungal cell wall polysaccharide galacto-
mannan requires intraluminal GDP-mannose. J Biol
Chem;287:44418–44424.
. Lipke P.N., Ovalle R. (1998) Cell wall architecture in
yeast: new structure and new challenges. J Bacte-
riol;180:3735–3740.
9
1
0. Orlean P. (1997) Biogenesis of yeast wall and surface
components. In: Pringle J., Broach J., Jones E., editors.
Chem Biol Drug Des 2016; 87: 163–170
169

Delso et al.
Molecular and Cellular Biology of the Yeast
neonates and young infants. Curr Med Chem;19:
4621–4632.
26. Scott L.J. (2012) Micafungin: a review of its use in the
prophylaxis and treatment of invasive Candida infec-
tions. Drugs;72:2141–2165.
27. Pikman R., Ami B.-R. (2012) Immune modulators as
adjuncts for the prevention and treatment of invasive
fungal infections. Immunotherapy;4:1869–1882.
28. Armstrong-James D., Harrison T.S. (2012) Immu-
notherapy for fungal infections. Curr Opin Micro-
biol;15:434–439.
29. Vandeputte P., Ferrari S., Coste A.T. (2012) Antifungal
resistance and new strategies to control fungal infec-
tions. Int J Microbiol;2012:713687.
30. Marca E., Valero-Gonzalez J., Delso I., Tejero T.,
Hurtado-Guerrero R., Merino P. (2013) Synthesis of O-
and C-glycosides derived from b-(1,3)-D-glucans. Car-
bohydr Res;382:9–18.
31. Frisch M.J., Trucks G.W., Schlegel H.B., Scuseria
G.E., Robb M.A., Cheeseman J.R., Scalmani G. et al.
(2009) Gaussian 09, Revision A.1. Wallingford CT:
Gaussian, Inc.
32. Friesner R.A., Banks J.L., Murphy R.B., Halgren T.A.,
Klicic J.J., Mainz D.T., Repasky M.P., Knoll E.H., Shaw
D.E., Shelley M., Perry J.K., Francis P., Shenkin P.S.
(2004) Glide: a new approach for rapid, accurate
docking and scoring. 1. Method and assessment of
docking accuracy. J Med Chem;47:1739–1749.
33. Halgren T.A., Murphy R.B., Friesner R.A., Beard H.S.,
Frye L.L., Pollard W.T., Banks J.L. (2004) Glide: a new
approach for rapid, accurate docking and scoring. 2.
Enrichment factors in database screening. J Med
Chem;47:1750–1759.
34. Friesner R.A., Murphy R.B., Repasky M.P., Frye L.L.,
Greenwood J.R., Halgren T.A., Sanschagrin P.C.,
Mainz D.T. (2006) Extra precision glide: docking and
scoring incorporating a model of hydrophobic enclo-
sure for protein-ligand complexes. J Med Chem;49:
6177–6196.
Saccharomyces. New York: Cold Spring Harbor Labo-
ratory Press; p. 229–362.
1
1. Lutzen A., Wittmann V. (2004) Molecular basis of pro-
tein-carbohydrate interactions. Highlights Bioorg
Chem; 119–123.
1
2. Coutinho P.M., Henrissat B. (1999) Carbohydrate-active
enzymes: an integrated database approach. In: Gilbert
H.J., Davies G., Henrissat B., Svensson B., editors.
Recent Advances in Carbohydrate Bioengineering.
Cambridge, UK: Royal Society of Chemistry; p. 3–12.
3. Hurtado-Guerrero R., Sch u€ ttelkopf A.W., Mouyna I.,
1
Ibrahim A.F.M., Shepherd S., Fontaine T., Latge J.-P.,
van Aalten D.M.F. (2009) Molecular mechanisms of
yeast cell wall glucan remodeling. J Biol Chem;
284:8461–8469.
1
4. Hartland R.P., Fontaine T., Debeaupuis J.P., Simenel
C., Delepierre M., Latge J.P. (1996) A novel beta-(1-3)-
glucanosyltransferase from the cell wall of Aspergillus
fumigatus. J Biol Chem;271:26843–26849.
1
5. Hattori T., Kato Y., Uno S., Usui T. (2013) Mode of
action of a b-(1?6)-glucanase from Penicillium multi-
color. Carbohydr Res;366:6–16.
1
6. Couturier M., Roussell A., Rosengren A., Leone P.,
Stalbrand H., Berrin J.G. (2013) Structural and bio-
chemical analyses of glycoside hydrolase families 5
and 26 b-(1,4)-mannanases from Podospora anserina
reveal differences upon manno-oligosaccharide cataly-
sis. J Biol Chem;288:14624–14635.
1
7. Reid C.W., Legaree B.A., Auzanneau F.I., Clarke A.J.
(2004) Inhibition of membrane-bound lytic transglyco-
sylase B by NAG-thiazoline. FEBS Lett;574:73–79.
8. Scheurwater E., Reid C.W., Clarke A.J. (2008) Lytic
transglycosylases: bacterial space-making autolysins.
Int J Biochem Cell Biol;40:586–591.
1
1
9. Ostash B., Walker S. (2005) Bacterial transglycosylase
inhibitors. Curr Opin Chem Biol;9:459–466.
2
0. Gastebois A., Fontaine T., Latg eꢀ J.P., Mouyna I. (2010)
The b(1-3)glucanosyltransferase Gel4p is essential for
Aspergillus fumigatus. Eukaryot Cell;9:1294–1298.
1. Fisher M.C., Henk D.A., Briggs C.J., Brownstein J.S.,
Madoff L.C., McCraw S.L., Gurr S.J. (2012) Emerging
fungal threats to animal, plant and ecosystem health.
Nature;484:186–194.
35. Meyer B., Mayer M. (1999) Characterization of ligand
binding by saturation transfer difference NMR spectro-
scopy. Angew Chem Int Ed;38:1784–1788.
36. Angulo J., Nieto P. (2011) STD-NMR: application to
transient interactions between biomolecules-a quantita-
tive approach. Eur Biophys J;40:1357–1369.
2
2
2
2
2. Goffeau A. (2008) Drug resistance: the fight against
fungi. Nature;452:541–542.
3. Paiva J.A., Pereira J.M. (2013) New antifungal antibio-
tics. Curr Opin Infect Dis;26:168–174.
4. Perfect J.R., Andes D. (2013) Antifungal therapy: cur-
rent concepts and evidence-based management. Curr
Med Res Opin;29:289–290.
Supporting Information
Additional Supporting Information may be found in the
online version of this article:
2
5. Regazzi M., Billaud E.M., Lefeuvre S., Stronati M.
(2012) Pharmacokinetics of antifungal agents in
Appendix S1. Graphical data of inhibition assays.
1
70
Chem Biol Drug Des 2016; 87: 163–170