A Native Ternary Complex Trapped in a Crystal Reveals the Catalytic Mechanism of a Retaining Glycosyltransferase

Article abstract of DOI:10.1002/anie.201504617

Glycosyltransferases (GTs) comprise a prominent family of enzymes that play critical roles in a variety of cellular processes, including cell signaling, cell development, and host-pathogen interactions. Glycosyl transfer can proceed with either inversion

Full text of DOI:10.1002/anie.201504617

.
Angewandte
Communications
International Edition: DOI: 10.1002/anie.201504617
German Edition: DOI: 10.1002/ange.201504617
Enzyme Catalysis
A Native Ternary Complex Trapped in a Crystal Reveals the Catalytic
Mechanism of a Retaining Glycosyltransferase**
David Albesa-JovØ, Fernanda Mendoza, Ane Rodrigo-Unzueta, Fernando Gomollón-Bel,
Javier O. Cifuente, Saioa Urresti, Natalia Comino, Hansel Gómez, Javier Romero-García,
JosØ M. Lluch, Enea Sancho-Vaello, Xevi BiarnØs, Antoni Planas, Pedro Merino,
Laura Masgrau, and Marcelo E. Guerin*
Abstract: Glycosyltransferases (GTs) comprise a prominent
family of enzymes that play critical roles in a variety of cellular
processes, including cell signaling, cell development, and host–
pathogen interactions. Glycosyl transfer can proceed with
either inversion or retention of the anomeric configuration with
respect to the reaction substrates and products. The elucidation
of the catalytic mechanism of retaining GTs remains a major
challenge. A native ternary complex of a GT in a productive
mode for catalysis is reported, that of the retaining glucosyl-3-
phosphoglycerate synthase GpgS from M. tuberculosis in the
presence of the sugar donor UDP-Glc, the acceptor substrate
phosphoglycerate, and the divalent cation cofactor. Through
a combination of structural, chemical, enzymatic, molecular
dynamics, and quantum-mechanics/molecular-mechanics
(QM/MM) calculations, the catalytic mechanism was unrav-
eled, thereby providing a strong experimental support for
a front–side substrate-assisted SNi-type reaction.
particularly apparent in the maintenance of the structural
integrity of the cell in the modulation of molecular recog-
nition events, including cell signaling, cell–cell communica-
tion, and cell–pathogen interactions.[3] GTs can be classified
as either ꢀinvertingꢁ or ꢀretainingꢁ enzymes according to the
anomeric configuration of the reaction substrates and prod-
ucts (Figure S1 in the Supporting Information).[1] The reac-
tion mechanism of inverting GTs seems to follow a single-
displacement mechanism with an oxocarbenium ion like
transition state and an asynchronous SN2 mechanism, analo-
gous to that observed for inverting glycosyl hydrolases.[1] By
contrast, the catalytic mechanism for retaining GTs is
currently a matter of strong debate. By analogy with glycosyl
hydrolases, a double displacement mechanism via the for-
mation of a covalent glycosyl-enzyme intermediate was first
suggested. Such a mechanism would involve an enzymatic
nucleophile positioned within the active site on the b-face of
the donor substrate in close proximity to the anomeric
reaction center (Figure 1A). Supporting this notion, chemical
rescue of the mammalian a-(1!3)-galactosyltransferase
Glu317Ala mutant by sodium azide has been reported.[4]
More recently, molecular dynamics simulations and density
functional theory (DFT) Quantum Mechanics/Molecular
Mechanics (QM/MM) calculations have also confirmed the
formation of a covalent glycosyl-enzyme intermediate in the
G
lycosyltransferases (GTs) play a central role in nature.
GTs catalyze the transfer of a sugar moiety from nucleotide-
sugar or lipid-phospho-sugar donors to a wide range of
acceptor substrates, including mono-, oligo-, and polysaccha-
rides, lipids, proteins, small organic molecules, and nucleic
acids.[1,2] As a consequence, GTs generate a significant
amount of structural diversity in biological systems, which is
[*] Dr. D. Albesa-JovØ, A. Rodrigo-Unzueta, Dr. J. O. Cifuente,
Dr. S. Urresti, N. Comino, Dr. E. Sancho-Vaello, Prof. M. E. Guerin
Unidad de Biofísica, Consejo Superior de Investigaciones Científ-
icas - Universidad del País Vasco/Euskal Herriko Unibertsitatea
(CSIC-UPV/EHU) and Departamento de Bioquímica, Universidad
del País Vasco
J. Romero-García, Dr. X. BiarnØs, Prof. A. Planas
Laboratory of Biochemistry, Institut Químic de Sarrià
Universitat Ramon Llull, 08017 Barcelona (Spain)
Dr. D. Albesa-JovØ, Prof. M. E. Guerin
IKERBASQUE, 48013 Bilbao (Spain)
[**] This work was supported by the EU Contract HEALTH-F3-2011-
260872, MINECO Contract BIO2013-49022-C2-2-R, and the Basque
Government (to M.E.G.); MINECO Contracts CTQ2011-24292 and
CTQ2014-53144-P (to J.M.LL.) and “UAB - Banco Santander
Program” (to L.M.); CTQ2013-44367-C2-1-P (to P.M.) and MINECO
Contract BIO2013-49022-C2-1-R (to A.P). We acknowledge Dia-
mond Light Source (proposals mx8302/10130), Soleil (proposal
20140843), and BioStruct-X (proposals 2460/5594/7687) for pro-
viding access to synchrotron radiation facilities. We gratefully
acknowledge S. López-Fernµndez and P. Arrasate (Unit of Bio-
physics, CSIC, UPV/EHU, Spain), E. Ogando and T. Mercero
(Scientific Computing Service UPV/EHU, Spain) for technical
assistance. F.G.-B. and F.M. acknowledge support from the JAE
Predoc Program (CSIC) and “Becas de Doctorado en el Extranjero -
Becas Chile - CONICYT” Program, respectively.
48940 Leioa, Bizkaia (Spain)
E-mail: mrcguerin@gmail.com
Dr. L. Masgrau
Institut de Biotecnologia i de Biomedicina (IBB) (Spain)
F. Mendoza, Prof. J. M. Lluch
IBB and Departament de Química
Universitat Autònoma de Barcelona, 08193 Bellaterra (Spain)
Dr. H. Gómez
IBB and Joint BSC-CRG-IRB Program in Computational Biology
IRB Barcelona, 08028 Barcelona (Spain)
F. Gomollón-Bel, Prof. P. Merino
Laboratorio de Síntesis AsimØtrica, Departamento de Síntesis
y Estructura de BiomolØculas, Instituto de Síntesis Química
y Catµlisis HomogØnea (ISQCH), Universidad de Zaragoza, CSIC
50009 Zaragoza, Aragón (Spain)
Supporting information for this article is available on the WWW
under http://dx.doi.org/10.1002/anie.201504617.
9898
ꢀ 2015 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Angew. Chem. Int. Ed. 2015, 54, 9898 –9902

Angewandte
Chemie
in the reaction mixture.[18] The strategy for capturing a native
ternary complex GpgS·Mn2+·UDP-Glc·PGA was to carry out
quick-soak experiments of unliganded GpgS crystals with the
sugar donor UDP-Glc and acceptor substrate PGA in the
presence of Mn2+ at different time points. We thus obtained
three snapshots of the reaction center at resolutions of 2.3 Š,
2.3 Š, and 2.6 Š, thereby providing for the first time the
atomic coordinates of a native Michaelis complex for a GT in
the absence of any substrate derivative or protein mutant
(Figure 2 and Table S1 and Figure S5 in the Supporting
Information).
Figure 1. A) Proposed catalytic mechanisms for enzymatic glycosyl
transfer with retention of the anomeric configuration: a double dis-
placement mechanism (1) and a front–face mechanism (2) B. Overall
structure of GpgS in complex with UDP-Glc, PGA, and the Mn2+
cofactor.
enzyme.[5–7] Covalent intermediates have also been detected
for the human blood group synthesizing a-(1!3)-N-acetyl-
galactosaminyltransferase (GTA) and a-(1!3)-galactosyl-
transferase (GTB) mutants by mass spectrometry.[8,9] How-
ever, in the absence of a residue near the reaction center that
can act as a nucleophile to form the glycosyl-enzyme
intermediate, an alternative mechanism known as the SNi
ꢀinternal returnꢁ, also called the SNi-like mechanism, has been
suggested (Figure 1A).[1,10–12] In this mechanism, leaving
group departure and nucleophilic attack occur on the same
face of the sugar[13] and involve either a short-lived oxocar-
benium ion intermediate (SNi-like)[6,9,14–16] or an oxocarbe-
nium ion transition state (SNi).[17]
To further advance the understanding of the catalytic
mechanism of retaining glycosyl transfer reactions, we inves-
tigated the glucosyl-3-phosphoglycerate synthase (GpgS)
from Mycobacterium tuberculosis. GpgS is a retaining gluco-
syltransferase that initiates the biosynthetic pathway of 6-O-
methylglucose lipopolysaccarides (MGLPs) in mycobacteria,
by transferring a glucose (Glc) moiety from uridine diphos-
phate (UDP)-Glc to the 2 position of the phosphoglycerate
(PGA) to form glucosyl-3-phosphoglycerate (Figure 1B and
Figure S2).[18,19] As with most of the members of the GT-A
superfamily of GTs, GpgS uses a divalent cation as an
essential cofactor for enzymatic activity. Kinetic studies have
demonstrated that the enzyme prefers Mg2+ for maximal
activity in vitro. However, GpgS was less but still enzymati-
cally active when another group II metal ion (Ca2+) or
transition metal ions (Mn2+, Co2+, or Fe2+) were introduced
Figure 2. Three snapshots of the reaction center as visualized in the
crystal structures of the ternary complexes GpgS·Mn2+·UDP-Glc·PGA-3
(A; PDB code 4Y9X), GpgS·Mn2+·UDP-Glc·PGA-2 (B; pre-Michaelis
complex, PDB code 4Y6U), and GpgS·Mn2+·UDP-Glc·PGA-1 (C;
Michaelis complex, PDB code 4Y6N). D) Structural comparison of
a selected region of the active site in the GpgS·Mn2+·UDP-Glc·PGA-1,
GpgS·Mn2+·UDP-Glc·PGA-2, and GpgS·Mn2+·UDP-Glc·PGA-3 com-
plexes.
GpgS is a homodimer, with each monomer displaying
a well-defined and positively charged tunnel, which is
compatible with the binding of phosphate-containing sub-
strates such as UDP-Glc and PGA (Figure 1B and Fig-
ure S4).[18] This tunnel is shaped in the form of two opposing
funnels, separated by a flexible loop that seems to modulate
substrate binding and playing a critical role during the
catalytic cycle.[18] Both the UDP-Glc and PGA substrates
and the cofactor are clearly visible in the electron density
maps, and are located in the center of the tunnel where the
Glc transfer reaction takes place (Figure S4). Importantly,
close inspection of the active site of GpgS revealed the
absence of a putative nucleophile residue that could result in
the formation of a glycosyl-enzyme covalent intermediate.
Therefore, it is expected that the reaction catalyzed by GpgS
would proceed through a front–side substrate-assisted SNi-
type mechanism. The first step in such a reaction would be the
À
breaking of the glucosidic OP C1’ bond, which accounts for
Angew. Chem. Int. Ed. 2015, 54, 9898 –9902
ꢀ 2015 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.angewandte.org 9899

.
Angewandte
Communications
most of the activation energy and is facilitated by a critical
stabilizing interaction of the acceptor hydrogen atom O3 of
the acceptor PGA with the b-phosphate of the nucleotide
sugar. Interestingly, in the first crystal structure
(GpgS·Mn2+·UDP-Glc·PGA-3, PDB code 4Y9X), the
acceptor oxygen atom O3 of PGA is located at a large
distance of 3.7 Š from the anomeric carbon C1’ of the sugar
and 3.2 Š from the O1B atom of the b-phosphate (Fig-
ure 2A). In a second crystal structure (GpgS·Mn2+·UDP-
Glc·PGA-2, PDB code 4Y6U), a pre-Michaelis complex
shows the anomeric carbon C1’ of the sugar at a position
3.4 Š from the acceptor oxygen atom O3 of PGA. The O1B
atom of the b-phosphate is 3.0 Š from the O3 atom of PGA
(Figure 2B). In the Michaelis complex (GpgS·Mn2+·UDP-
Glc·PGA-1, PDB code 4Y6N), the anomeric carbon C1’ of
the sugar approaches only 2.6 Š from the acceptor oxygen
atom O3 of PGA, which in turn hydrogen bonds with the O1B
atom of the b-phosphate (Figure 2C,D). The configuration of
the active site in the native Michaelis complex of GpgS for the
wild type enzyme and with the natural substrates thus
provides strong experimental evidence in support of the
mechanism described above.
cation (GpgS·Mn2+·UDP-Glc·G3P PDB code 4Y7G; Fig-
ure 3B). The phosphate group of PGA forms hydrogen
bonds with the lateral chains of Arg185 and Asn260, as
previously visualized in the GpgS·Mn2·UDP-Glc·PGA-1 and
GpgS·Mn2·UDP-Glc·PPA complexes. However, the rest of
the G3P molecule displays a different structural arrangement
(r.m.s.d. of 2.5 Š). Specifically, the oxygen atom O2 of G3P,
which is equivalent to the acceptor oxygen atom O3 of PGA,
moves away from the Glc moiety to form new electrostatic
interactions with the guanidinium group of Arg256 and the
lateral chain of His258. The oxygen atom O1 forms a hydro-
gen bond with the side chain OG1 atom of Thr187 residue
(Figure 3B). Altogether the experimental data strongly sup-
port the idea of the carboxyl moiety of PGA playing a key
role in the generation of a competent reaction center for
GpgS.
To further investigate the importance of the carboxyl
moiety of PGA, we synthesized a PGA derivative in which
the carboxyl group was replaced by amide (PGD; see the
Supporting Information). Interestingly, PGD, as with G3P,
could not serve as an acceptor for Glc, although it contains the
oxygen atom O3 of PGA (the Supporting Information). In
addition, despite much effort, we were unable to crystallize
GpgS in complex with the PGD derivative, even through
soaking or co-crystallization experiments. A plausible explan-
ation is that the carboxyl O2 of PGA forms a strong hydrogen
bond with the main-chain amino group of Thr187 in
GpgS·Mn2+·UDP-Glc·PGA-1. The presence of an amide
group in PGD might thus lead to electrostatic repulsion
with Thr187, thereby preventing its binding to GpgS.
Two additional ternary complexes provide significant
insight not only into the binding mode of the sugar donor
and acceptor substrates in the active site, but also into the
catalytic mechanism of GpgS (Figure 3, Table S1, and Fig-
ure S6). The first complex was solved with UDP-Glc, 3-
(phosphonooxy)propanoic acid (PPA, an analogue of PGA
The experimental native Michaelis complex of GpgS is in
good agreement with the predicted Michaelis complexes of
lipopolysaccharyl-a-1,4-galactosyltransferase C (LgtC),[11,17]
trehalose-6-phosphate synthase (OtsA),[15,20] and the recently
obtained polypeptide N-acetylgalactosaminyltransferase 2,
which contains the UDP-GalNAc derivative UDP-5SGalNAc
and the truncated incompetent mEA2 peptide STCPA
(GalNAc-T2,
GalNAc = N-acetylgalactosamine;
Fig-
ure S7).[16] The LgtC Michaelis complex was modeled based
on a crystal structure containing the two substrate analogues
UDP 2-deoxy-2’-F-Gal and 4-deoxylactose. The attacking
hydroxy group of lactose has the oxygen atom O4 at a distance
of 3.1 Š from the anomeric carbon C1’ of the donor Gal
moiety and 2.7 Š from the glycosidic oxygen atom. Moreover,
the O3 of the acceptor lactose also forms a hydrogen bond
with the b-phosphate of UDP, thus stabilizing leaving group
departure.[17] The OtsA Michaelis complex was constructed
based on its complex with UDP and validoxylamine-6-
phosphate (VA6P), a compound that structurally resembles
to one of the reaction products, trehalose-6-phosphate. The
anomeric carbon C1’ of the Glc moiety is 3.0 Š from the O1’
of the Glc-6-phosphate acceptor.[15] In GalNAc-T2, the
hydroxy oxygen atom OG1 of the acceptor Thr is 2.5 Š
from the anomeric carbon C1’ of the GalNAc moiety, and 2.7
and 3.6 Š from the two b-phosphate oxygen atoms of UDP
moiety.[16] In addition, the backbone amide of the acceptor
Thr is also hydrogen bonded to the b-phosphate.[21] Similarly,
in GpgS the acceptor oxygen atom O3 of the acceptor PGA is
placed 2.6 Š from the anomeric carbon C1’ in the Glc residue,
Figure 3. The catalytic site as visualized in the crystal structures of the
ternary complexes GpgS·Mn2+·UDP-Glc·PPA (A; PBD code 4Y7F) and
GpgS·Mn2·UDP-Glc·G3P (B; PDB code 4Y7G).
that lacks the glucose-accepting hydroxy group), and Mn2+ as
a divalent cation (GpgS·Mn2+·UDP-Glc·PPA, PDB code
4Y7F; Figure 3A). We confirmed that the enzyme was
unable to transfer a Glc residue to PPA (see the Supporting
Information). The carboxyl group of PPA superimposes well
with the corresponding moiety of PGA, as observed in the
ternary complex GpgS·Mn2+·UDP-Glc·PGA-1. However, C2,
C3, and the phosphate moiety adopt a different conformation
(root-mean-square deviation (r.m.s.d.) of 0.9 Š; Figure 2). As
expected, there is no electron density that could indicate the
formation of a covalent adduct between the GT and the Glc
moiety in the GpgS·Mn2+·UDP-Glc·PPA complex. The
second complex was solved with UDP-Glc, glycerol 3-
phosphate (G3P, an analogue of PGA in which the carboxyl
group is replaced by a hydroxy group), and Mn2+ as a divalent
9900 www.angewandte.org
ꢀ 2015 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Angew. Chem. Int. Ed. 2015, 54, 9898 –9902

Angewandte
Chemie
and 2.4 and 3.0 Š from the corresponding b-phosphate
oxygen atoms of the UDP moiety. No other interactions are
observed between the donor and acceptor substrates in GpgS.
Extensive theoretical studies using QM/MM calculation
simulations supported the occurrence of the SNi ꢀinternal
returnꢁ mechanism in OtsA, LgtC, and GalNAc-T2.[15–17,21]
We employed molecular dynamics simulations combined
with QM(DFT)/MM calculations to further understand the
reaction mechanism of GpgS (Figure 4; see Figure S9 and
discussion in the Supporting Information for further details).
The sugar transfer was modeled by using the reaction
coordinate RC = [d(OP-C1’)-d(O3PGA-C1’)-d(HO3PGA-OP)]
to drive the system from reactants to products at a QM-
(DFT)/MM level of theory that we have successfully applied
in other retaining GTs.[6,17,21] Two transition states (TS1 and
TS2) and a short-lived ion-pair intermediate (IP) connecting
them were found (Figure 4 and Table S2), so that an
asynchronous mechanism is obtained. As depicted in
À
Figure 4, the reaction starts with the OP HO3PGA hydrogen
bond getting shorter (by ca. 0.4 Š, RC = À3.6 Š). This
interaction between the donor and acceptor substrates
occurs in all retaining GTs studied to date and seems to be
essential in facilitating nucleotide–sugar bond breakage
À
(substrate-assisted catalysis). In TS1, the OP C1’ distance
increases to 2.68 Š, HO3PGA is 1.65 Š from OP, and the
À
O3PGA C1’ distance shortens to 2.62 Š. Breakage of the
UDP–Glc bond results in a positive charge increase of
Dq(C1’ + H1’ + O5’) = 0.36 a.u. at the anomeric center
(Table S2). Importantly, the ring conformation changes from
a 4C1 chair in the reactants to a 4E half chair in TS1, with the
À
C1’ O5’ distance indicating the development of double-bond
character (Table S2). The energy cost to reach TS1 is
21.4 kcalmolÀ1. After TS1, the energy along the reaction
coordinate decreases by approximately 2 kcalmolÀ1, and
a short-lived oxocarbenium intermediate is formed (sugar
ring conformation between 4E and 4H5). In this IP, the
À
glucosidic bond is definitely broken (the OP C1’ distance is
À
À
now 3.20 Š), whereas the HO3PGA O3PGA and C1’ O5’
distances remain practically invariant with respect to TS1.
The attacking O3PGA atom gets 0.3 Š closer to C1’. The energy
of the optimized IP is 0.6 kcalmolÀ1 lower than that of TS1.
From the IP, acceptor attack takes place through TS2, in
À
À
which the C1’ O3PGA and HO3PGA OP distances are 2.26 Š
and 1.56 Š, respectively. TS2 lies 21.8 kcalmolÀ1 above the
Michaelis complex and 0.5 kcalmolÀ1 above the IP. Therefore,
most of the overall energy barrier for the transfer reaction to
À
take place is due to glucosidic OP C1’ bond dissociation.
These energies indicate that the IP could be a very short-lived
species with no time to thermally equilibrate. The calculated
rate-limiting energy barrier (21.8 kcalmolÀ1) is in reasonable
agreement with the experimentally derived phenomenolog-
ical free energy of activation of 17.5 kcalmolÀ1 (derived from
a kcat value of 800 minÀ1; see the Supporting Information),
thus supporting the idea that GpgS follows a front–side attack
mechanism.
Intriguingly, the active sites of a few GTs display
a carboxylate residue, which allows these enzymes to follow
a double-displacement reaction,[6,7,9] whereas most of retain-
ing GTs, in the absence of such nucleophile, seem to follow
a front–side mechanism. We propose that the two mecha-
nisms might represent different manners of stabilizing the
oxocarbenium ion like species that forms upon cleavage of the
donor sugar–phosphate bond. In GpgS, the electrostatic
potential at the reaction center can stabilize the oxocarbe-
nium ion like intermediate for a very short period of time,
thereby allowing the active site to reorganize and the
oxocarbenium ion species and acceptor to move towards
each other. By contrast, in the a-1,3-galactosyltransferase
GTB[22] or the N-acetyllactosaminide a-1,3-galactosyl trans-
ferase a3GalT,[23] the oxocarbenium ion like transition state is
stabilized by the formation of a covalent bond with the
nucleophile residue present in these enzymes. Thus, as we
Figure 4. A) Atomic rearrangement along the reaction pathway in
GpgS: optimized reactants (R), structure at RC=À3.6 Š, transition
state 1 (TS1), ion-pair intermediate (IP), transition state2 (TS2), and
products (P). B. Structural and energetic changes along the reaction
coordinate.
Angew. Chem. Int. Ed. 2015, 54, 9898 –9902
ꢀ 2015 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.angewandte.org 9901

.
Angewandte
Communications
initially proposed,[7] the two modes of operation could be
considered as variations of a common mechanism: a two-step
reaction via oxocarbenium ion like transition states that flank
an intermediate, either an oxocarbenium ion or a covalent
glycosyl-enzyme intermediate, depending of the active-site
configuration.
[11] K. Persson, H. D. Ly, M. Dieckelmann, W. W. Wakarchuk, S. G.
Withers, N. C. Strynadka, Nat. Struct. Biol. 2001, 8, 166 – 175.
[12] R. P. Gibson, J. P. Turkenburg, S. J. Charnock, R. Lloyd, G. J.
Davies, Chem. Biol. 2002, 9, 1337 – 1346.
[13] S. S. Lee, S. Y. Hong, J. C. Errey, A. Izumi, G. J. Davies, B. G.
Davis, Nat. Chem. Biol. 2011, 7, 631 – 638.
ˇ
ˇ
[14] A. Bobovskµ, I. TvaroSka, J. Kóna, Org. Biomol. Chem. 2014,
12, 4201 – 4210.
[15] A. Ard›vol, C. Rovira, Angew. Chem. Int. Ed. 2011, 50, 10897 –
10901; Angew. Chem. 2011, 123, 11089 – 11093.
Keywords: enzyme catalysis · enzymes · glycosyltransferases ·
reaction mechanisms · structure elucidation
[16] E. Lira-Navarrete, J. Iglesias-Fernµndez, W. F. Zandberg, I.
CompaÇón, Y. Kong, F. Corzana, B. M. Pinto, H. Clausen, J. M.
Peregrina, D. J. Vocadlo, C. Rovira, R. Hurtado-Guerrero,
Angew. Chem. Int. Ed. 2014, 53, 8206 – 8210; Angew. Chem.
2014, 126, 8345 – 8349.
How to cite: Angew. Chem. Int. Ed. 2015, 54, 9898–9902
Angew. Chem. 2015, 127, 10036–10040
[17] H. Gómez, I. Polyak, W. Thiel, J. M. Lluch, L. Masgrau, J. Am.
Chem. Soc. 2012, 134, 4743 – 4752.
[1] L. L. Lairson, B. Henrissat, G. J. Davies, S. G. Withers, Annu.
Rev. Biochem. 2008, 77, 521 – 555.
[18] S. Urresti, D. Albesa-JovØ, F. Schaeffer, H. T. Pham, D. Kaur, P.
Gest, M. J. van der Woerd, A. Carreras-Gonzµlez, S. López-
Fernµndez, P. M. Alzari, P. J. Brennan, M. Jackson, M. E.
Guerin, J. Biol. Chem. 2012, 287, 24649 – 24661.
[19] M. Jackson, P. J. Brennan, J. Biol. Chem. 2009, 284, 1949 – 1953.
[20] J. C. Errey, S. S. Lee, R. P. Gibson, C. Martinez Fleites, C. S.
Barry, P. M. Jung, A. C. O’Sullivan, B. G. Davis, G. J. Davies,
Angew. Chem. Int. Ed. 2010, 49, 1234 – 1237; Angew. Chem. 2010,
122, 1256 – 1259.
[21] H. Gómez, R. Rojas, D. Patel, L. A. Tabak, J. M. Lluch, L.
Masgrau, Org. Biomol. Chem. 2014, 12, 2645 – 2655.
[22] J. A. Alfaro, R. B. Zheng, M. Persson, J. A. Letts, R. Polakowski,
Y. Bai, S. N. Borisova, N. O. Seto, T. L. Lowary, M. M. Palcic,
S. V. Evans, J. Biol. Chem. 2008, 283, 10097 – 10108.
[23] E. Boix, Y. Zhang, G. J. Swaminathan, K. Brew, K. R. Acharya,
J. Biol. Chem. 2002, 277, 28310 – 28318.
[2] D. Albesa-JovØ, D. Giganti, M. Jackson, P. M. Alzari, M. E.
Guerin, Glycobiology 2014, 24, 108 – 124.
[3] A. Varki, Essentials of Glycobiology, 2nd ed. (Eds.: A. Varki, R.
Cummings, J. Esko, H. Freeze, P. Stanley, C. R. Bertozzi, G. Hart,
M. E. Etzler) Cold Spring Harbor Laboratory Press, Cold Spring
Harbor, NY, 2009.
[4] A. Monegal, A. Planas, J. Am. Chem. Soc. 2006, 128, 16030 –
16031.
[5] H. Gómez, J. M. Lluch, L. Masgrau, Carbohydr. Res. 2012, 356,
204 – 208.
[6] H. Gómez, J. M. Lluch, L. Masgrau, J. Am. Chem. Soc. 2013, 135,
7053 – 7063.
[7] V. Rojas-Cervellera, A. Ard›vol, M. Boero, A. Planas, C.
Rovira, Chem. Eur. J. 2013, 19, 14018 – 14023.
[8] N. Soya, Y. Fang, M. M. Palcic, J. S. Klassen, Glycobiology 2011,
21, 547 – 552.
ˇ
ˇ
[9] A. Bobovskµ, I. TvaroSka, J. Kóna, Glycobiology 2015, 25, 3 – 7.
[10] M. L. Sinnott, W. P. Jencks, J. Am. Chem. Soc. 1980, 102, 2026 –
2032.
Received: May 21, 2015
Published online: July 1, 2015
9902 www.angewandte.org
ꢀ 2015 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Angew. Chem. Int. Ed. 2015, 54, 9898 –9902