Synthesis and evaluation of 2-deoxy-2-amino-β-cellobiosides as cellulase inhibitors

Article abstract of DOI:10.1080/07328303.2010.508142

The cellulase mixture of Hypocrea jecorina (formerly Trichoderma reesei) contains a variety of exo- and endoglucanases that belong to different structural families. As such, these enzymes form an interesting model system to study the enzyme-ligand interac

Full text of DOI:10.1080/07328303.2010.508142

This article was downloaded by: [New York University]
On: 07 December 2014, At: 22:43
Publisher: Taylor & Francis
Informa Ltd Registered in England and Wales Registered Number: 1072954 Registered
office: Mortimer House, 37-41 Mortimer Street, London W1T 3JH, UK
Journal of Carbohydrate Chemistry
Publication details, including instructions for authors and
subscription information:
http://www.tandfonline.com/loi/lcar20
Synthesis and Evaluation of 2-Deoxy-2-
amino-β-cellobiosides as Cellulase
Inhibitors
Tom Desmet a , Marc Claeyssens b , Kathleen Piens c & Wim Nerinckx
b
a Department of Biochemical and Microbial Technology , Ghent
University , Coupure links 653, B-9000, Ghent, Belgium
b Department of Biochemistry and Microbiology , Ghent University ,
K.L. Ledeganckstraat 35, B-9000, Ghent, Belgium
c Department of Molecular Biology , Uppsala University , Biomedical
Center, Box 596, SE-75124, Uppsala, Sweden
Published online: 26 Aug 2010.
To cite this article: Tom Desmet , Marc Claeyssens , Kathleen Piens & Wim Nerinckx (2010) Synthesis
and Evaluation of 2-Deoxy-2-amino-β-cellobiosides as Cellulase Inhibitors, Journal of Carbohydrate
Chemistry, 29:4, 164-180, DOI: 10.1080/07328303.2010.508142
To link to this article: http://dx.doi.org/10.1080/07328303.2010.508142
PLEASE SCROLL DOWN FOR ARTICLE
Taylor & Francis makes every effort to ensure the accuracy of all the information (the
“Content”) contained in the publications on our platform. However, Taylor & Francis,
our agents, and our licensors make no representations or warranties whatsoever as to
the accuracy, completeness, or suitability for any purpose of the Content. Any opinions
and views expressed in this publication are the opinions and views of the authors,
and are not the views of or endorsed by Taylor & Francis. The accuracy of the Content
should not be relied upon and should be independently verified with primary sources
of information. Taylor and Francis shall not be liable for any losses, actions, claims,
proceedings, demands, costs, expenses, damages, and other liabilities whatsoever or
howsoever caused arising directly or indirectly in connection with, in relation to or arising
out of the use of the Content.
This article may be used for research, teaching, and private study purposes. Any
substantial or systematic reproduction, redistribution, reselling, loan, sub-licensing,
systematic supply, or distribution in any form to anyone is expressly forbidden. Terms &

Conditions of access and use can be found at http://www.tandfonline.com/page/terms-
and-conditions

Journal of Carbohydrate Chemistry, 29:164–180, 2010
ꢀ
C
Copyright Taylor & Francis Group, LLC
ISSN: 0732-8303 print / 1532-2327 online
DOI: 10.1080/07328303.2010.508142
Synthesis and Evaluation
of 2-Deoxy-2-amino-β-
cellobiosides as Cellulase
Inhibitors
Tom Desmet,1 Marc Claeyssens,2 Kathleen Piens,3
and Wim Nerinckx2
1Department of Biochemical and Microbial Technology, Ghent University, Coupure
links 653, B-9000, Ghent, Belgium
2Department of Biochemistry and Microbiology, Ghent University, K.L.
Ledeganckstraat 35, B-9000, Ghent, Belgium
3Department of Molecular Biology, Uppsala University, Biomedical Center, Box 596,
SE-75124, Uppsala, Sweden
The cellulase mixture of Hypocrea jecorina (formerly Trichoderma reesei) contains a va-
riety of exo- and endoglucanases that belong to different structural families. As such,
these enzymes form an interesting model system to study the enzyme-ligand interac-
tions in glycoside hydrolases. The nucleophilic carboxylate of retaining β-glycosidases is
believed to form a hydrogen bond with the 2-hydroxyl group of their substrate. Conse-
quently, replacing this hydroxyl group with an amino group should result in a stronger
electrostatic interaction and thus an increased affinity for the ligand. In this study, sev-
eral modified cellobiosides were synthesized and evaluated as cellulase inhibitors. The
introduction of an amino group was found to have an unpredictable effect on the in-
hibitory power of the ligands. However, the enzymes display a very high affinity for the
corresponding 2-azido compounds, precursors in the synthetic route. The new ligand
m-iodobenzyl 2-deoxy-2-azido-β-cellobioside even is the strongest inhibitor of cellobio-
hydrolase I known to date (KI = 1 µM).
Keywords Glycosidase; Cellulase; Inhibition; 2-Amino-2-deoxy-cellobioside
INTRODUCTION
Glycoside hydrolases (GH, EC 3.2.1) are a diverse and ubiquitous group of
enzymes that are responsible for the degradation of carbohydrate chains,
Received May 12, 2010; accepted July 10, 2010.
Address correspondence to Tom Desmet, Department of Biochemical and Microbial
Technology, Ghent University, Coupure links 653, B-9000, Ghent, Belgium. E-mail:
t.desmet@ugent.be
164

Cellulase Inhibitors
165
Figure 1: General reaction mechanisms of β-glycoside hydrolases. A: The double
displacement mechanism of retaining glycosidases. B: The single displacement mechanism
of inverting glycosidases. Conformational changes during hydrolysis are not shown.[28]
glycosides, and glycoconjugates. They are classified in more than 100 different
families, reflecting their wide variation in sequence and structure.[1] Neverthe-
less, nearly all glycosidases follow one of two general reaction mechanisms that
involve two carboxylic amino acids (Asp/Glu) as catalytic residues (Fig. 1).[2]
A
single displacement reaction results in inversion of the anomeric configuration,
while a double displacement reaction results in net retention.
With a half-life in the order of millions of years, glycosidic bonds are known
to be extremely stable.[3] Consequently, glycoside hydrolases rank among the
most proficient biocatalysts, achieving rate enhancements up to 1017. This
means that glycosidases have a very high affinity for the transition state (TS)
and hence, are very sensitive to inhibition by TS mimics.[4] Such compounds
can have numerous applications (e.g., in the treatment of diabetes or viral in-
fections).[5, 6] Examples of commercially available therapeutic glycosidase in-
hibitors are acarbose (Glucobay) and oseltamivir (Tamiflu), which target α-
glucosidase and neuraminidase, respectively.
Inhibitor design requires a thorough knowledge of enzyme-ligand interac-
tions, one of the most important being that between the catalytic nucleophile of
retaining β-glycosidases and the 2-hydroxyl group of their substrate.[7] Indeed,

T. Desmet et al.
166
Figure 2: The interaction between the catalytic nucleophile of retaining β-glycosidases and
the 2-hydroxyl group of their substrate. The interaction is believed to be optimized, both
geometrically and electronically, at the level of the transition state.
this interaction is believed to be optimized, both electronically and geometri-
cally, at the level of the transition state (Fig. 2) and contributes strongly (30 to
40 kJ/mol) to its stabilization.[8–10] Replacing the 2-hydroxyl group of a transi-
tion state mimic by an amino group should, therefore, strengthen the interac-
tion with the negatively charged nucleophile, resulting in an increased affinity
for the inhibitor. This has indeed been observed with glucosamine-based in-
hibitors of β-glucosidases.[11] On average, a 10-fold increase in affinity could
be achieved, although the effect was dependent on the nature of the glycosidic
heteroatom.
To determine whether this strategy can be expanded to other β-
glycosidases, aromatic 2-deoxy-2-amino-β-cellobiosides have presently been
synthesized and tested as inhibitors of cellulases from Hypocrea jecorina
(formerly Trichoderma reesei). The coupling of a nonactivated aromatic
aglycon to the (modified) cellobiosides should allow the ligands to span the
active site in a distorted, transition state-like conformation without being
hydrolyzed.[2] Cellulases consist of a complex mixture of different enzymes
that comprise both endoglucanases (EG, EC 3.2.1.4) and cellobiohydrolases
(CBH, EC 3.2.1.91).[12] These enzymes belong to different GH-families and,
therefore, form an interesting model system to study the general features of
enzyme-ligand interactions.

Cellulase Inhibitors
167
The four major cellulases of H. jecorina are CBH I (Cel7A), CBH II (Cel6A),
EG I (Cel7B), and EG II (Cel5A), which together make up almost 99% of the
cellulase content (nomenclature according to [13]). They have been extensively
characterized and are well suited for the evaluation of novel ligands, as they
display a variety of structures and mechanisms. All these enzymes hydrolyze
their substrate with retention of the anomeric configuration, except for CBH
II, which follows an inverting mechanism. Both cellobiohydrolases have an
active site located in a tunnel, with CBH I and CBH II degrading cellulose
from the reducing and nonreducing end, respectively. The endoglucanases, in
contrast, have an active site located in a cleft and hydrolyze their substrate
somewhere in the middle. Different binding modes of inhibitors are, therefore,
often observed with the different enzymes.
RESULTS
Synthesis of Inhibitors
All compounds were synthesized starting from per-O-acetyl cellobiose
(1) (Fig. 3). A nitrogen functionality was introduced at the 2-position by
azidonitration[14, 15] of per-O-acetyl cellobial (2), which was prepared by con-
verting 1 to acetobromo cellobiose followed by reductive elimination with
Zn/H2PtCl6.[16] Treatment of 2 with ceric ammonium nitrate and sodium azide
at –15◦C yielded the 2-azido-2-deoxy-cellobiose derivatives 3 and 4, both show-
ing their 2-azido function in equatorial position. Although expected as a minor
side product, the D-glucosyl-β(1ꢁ,4)-2-deoxy-2-azido-D-mannose derivate 5 with
its 2-azido function in axial position appeared not to be present (TLC analysis),
in contrast to what has been reported for the azidonitrations of D-galactal and
D-glucal, where the corresponding axial 2-azido carbohydrate derivatives were
formed in substantial amounts.[14, 15] Selective hydrolysis of the 1-nitrate ester
of 3 yielded 2-deoxy-2-azido-3,6,2ꢁ,3ꢁ,4ꢁ,6ꢁ-hexa-O-acetyl cellobiose (6). In prin-
ciple, the yield (36%) of the desired hemiacetal 6 may be further increased by
converting the amide 4 (21% yield) to 2-deoxy-2-azido cellobiose (7) by means
of acid hydrolysis, followed by peracetylation and selective anomeric deprotec-
tion.
Coupling of benzyl alcohol and m-iodobenzyl alcohol to the hemiacetal 6
and to 1-deprotected per-O-acetyl cellobiose[17] was with the trichloroacetimi-
date method.[18] Formation of the trichloroacetimidates 8 and 9[19] was at rt to
generate the thermodynamically more stable α-anomer. The coupling reactions
were at –15◦C to avoid anomeric interconversion, respectively yielding the per-
acetates 10, 11, 12, and 13. Zemplen deacetylations to the final products 14,
15, 16, and 17 were quantitative.
Attempts at catalytic reduction with H2/Pd-C of the 2-azides 14 and
15 to the corresponding amines failed in our hands: the former underwent

T. Desmet et al.
168
Figure 3: Synthetic routes for the different cellobiosides and their C2 analogues. a:
hydrazine, DMF, 50◦C (96%)[17]; b: CCl3CN, DBU, CH2Cl2, rt, 1 h (90%)[19]; c: benzyl alcohol or
m-iodobenzyl alcohol, BF3.OEt2, CH2Cl2, –15◦C, 1 h (56% for 12, 57% for 13); d: NaOMe,
MeOH, rt, 2 h (quant.); e: HBr/HOAc, CHCl3, rt, 1 h (87%); f: Zn, H2PtCl6, HOAc, rt, 4 h (57%)[16]
g: (NH+4 )2CeIV(NO3−)6, NaN3, MeCN, 12 h, rt (21% for 4); h: NaNO2, H2O, dioxane, 8 h, 80◦C
(36% for 6 from 2); i: CCl3CN, DBU, CH2Cl2, rt, 2 h (58%), j: benzyl alcohol or m-iodobenzyl
alcohol, BF3.OEt2, CH2Cl2, –15◦C, 1h (79% for 10; 72% for 11); k: NaOMe, MeOH, rt, 2 h
(quant.); l: triphenyl phosphine, MeOH/H2O (9/1), rt, 12 h (95% for 18, 94% for 19).
;
aglycon cleavage, whereas the latter did not react, presumably due to cat-
alyst poisoning by the iodide function. They were successfully converted to
the desired amines 18 and 19 by a Staudinger reduction[20] with triphenyl
phosphine.

Cellulase Inhibitors
169
Inhibition of Cellulases
Different aromatic cellobiosides (14–19) have been synthesized as poten-
tial inhibitors of cellulases (Fig. 3). Assuming that these compounds bind in
subsites –2 to +1, distortion of the carbohydrate ring in subsite –1 to a tran-
sition state-like conformation should be possible. In that way, the replacement
of the 2-hydroxyl group by an amino group could result in a stronger interac-
tion with the catalytic nucleophile and hence, tighter binding of the inhibitor.
Preliminary results have indicated that the corresponding 2-azido compounds,
precursors in the synthetic route, incidentally display a high inhibitory power
and they have therefore also been included in this study.
The use of a nonactivated aromatic aglycon is required to preclude enzy-
matic hydrolysis of the ligands. A benzyl group is known to accommodate ef-
ficiently in subsite +1 of most cellulases, because it resembles a carbohydrate
ring in size and shape.[21] An iodobenzyl group was also evaluated as aglycon,
as its increased polarity might result in improved interactions with the en-
zyme. Inhibition studies have been performed with four different cellulases,
consisting of endo- and exo-acting enzymes from three different GH-families
(Table 1). All inhibitions were found to be of the competitive type.
For CBH I (Cel7A), replacing the 2-OH by an amino group did not result
in tighter binding but instead increased the KI of the ligands. However, intro-
ducing an azido group at this position had the opposite effect, resulting in KI
values in the low µM range. When the aglycons are compared, a slight prefer-
ence for an iodobenzyl over a benzyl is observed. Surprisingly, the cellobiosides
16 and 17 are hydrolyzed rather efficiently by EG I (Cel7B), excluding their
use as inhibitors. All other ligands were only marginally hydrolyzed after 24 h
incubation with the enzyme (not shown). The endoglucanase was also found to
have a higher affinity for an azido than for an amino group at C2.
Unexpectedly, none of the synthesized ligands are able to inhibit EG II
(Cel5A) to a significant extent. In contrast, inhibition was observed with the
Table 1: Inhibition constants for the various ligands synthesized in this study
Inhibitor
aglycon
KI (µM)
Nr.
C2
CBH I
EG I
CenA
16
17
14
15
18
19
benzyl
OH
N3
14
7
—
417
n.d.
394
n.d.
661
n.d.
iodobenzyl
benzyl
—
2
409
248
623
667
iodobenzyl
benzyl
iodobenzyl
1
72
59
NH2
Spectroscopic assays were performed continuously at pH 5.7 and 37◦C. All measurements were
performed in triplicate and had a CV of less than 10%. n.d., not determined; —, hydrolysis.

T. Desmet et al.
170
endoglucanase CenA from family GH-6. This enzyme from Cellulomonas fimi
was chosen for practical reasons because it is active on chromogenic substrates,
while no such substrates have yet been identified for CBH II (Cel6A) from H.
jecorina.[22] Only inhibitors with a benzyl group as aglycon have been tested,
but this still allows an evaluation of the influence of the C2 substituent. An
amino group has been found to decrease the ligands’ affinity, while an azido
group does not have a significant effect.
DISCUSSION
As shown in this study, replacing the 2-OH group of cellobiosides with an
amino functionality has a rather unpredictable effect on their inhibitory power
toward cellulases. In fact, the proposed interaction with the catalytic nucle-
ophile can only be realized if these ligands bind in subsites –2 to +1, which
might not be the case for all cellulases. CBH I and EG I, for example, interact
differently with cellobiose, although both cellulases belong to the same family
GH-7. The former enzyme binds cellobiose very tightly (KI = 28 µM) in sub-
sites +1/+2, while the latter binds it rather loosely (KI = 6 mM) in subsites
–2/–1.[23]
Considering its positioning of cellobiose, EG I might be expected to bind
aromatic cellobiosides in subsites –2/+1. This is corroborated by the hydrol-
ysis of ligands 16 and 17 at the heterosidic linkage. The strongest binding
was achieved by the introduction of a 2-N3 group, as compared to a 2-NH2
group. Inspection of the crystal structure of the enzyme in complex with cel-
lobiose (PDB 2OVW) reveals that subsite –1 of the active-site cleft is spacious
enough to accommodate an azido group at C2 (Fig. 4). An azido functional-
ity could potentially also interact with the catalytic nucleophile, as its central
Figure 4: Binding of cellobiose in GH-7 cellulases. The ligand is positioned in subsites –2/–1
of EG I (left) and in subsites +1/+2 of CBH I (right). The figure was drawn using PyMol, with
cellobiose and the catalytic amino acids in yellow and other important residues in
pink. (Figure available in color online.)

Cellulase Inhibitors
171
nitrogen atom carries a partial positive charge. Surprisingly, different affini-
ties for the aglycons are observed for the different types of inhibitors. The ac-
commodation of an azido versus an amino group might, however, require a
different conformation of the carbohydrate ring in subsite –1, resulting in dif-
ferent interactions with the aglycon in subsite +1. As the active site of EG
I is composed of only four subsites (–2/+2), the only alternative positioning
of cellobiosides would be in subsites –1/+2. In that case, the most likely in-
teraction of the C2 substituent would be with a histidine residue (His209) in
subsite +1, but it is uncertain how this would affect the binding strength of the
inhibitors.
Of all H. jecorina cellulases, CBH I has the largest active site, contain-
ing at least 10 subsites (–7/+3). Consequently, many different binding modes
could be envisaged for the ligands described here. Crystallographic and kinetic
studies, however, have revealed that CBH I positions cellobiosides preferen-
tially in subsites +1/+3.[24] This could explain why a 2-NH2 group does not
induce tight binding. Remarkably, the 2-N3 compounds interact very strongly
with the enzyme, resulting in KI values as low as 1 to 2 µM. The ligand m-
iodobenzyl 2-deoxy-2-azido-β-cellobioside (15) is even the strongest CBH I in-
hibitor known to date. Its high affinity for CBH I has also been confirmed
by mass spectrometry.[25] Inspection of the crystal structure of the enzyme
in complex with cellobiose (PDB 3CEL) reveals that the C2 substituent in
subsite +2 is pointing away from the active site, with relatively few inter-
actions (Fig. 4). Only two residues (Tyr252 and Asp259) are in close vicinity
˚
(∼4 A) and could potentially form an interaction, especially with the longer
azido functionality. For all inhibitor types, a similar preference for the agly-
con groups is observed, implying that these are always positioned in the same
subsite.
The cellulase CenA from family GH-6 has also been included in this study,
although it operates by an inverting mechanism in which no catalytic nu-
cleophile is present close to the 2-group of the substrate. It is, therefore, not
surprising that modification of this group does not result in a significant im-
provement in binding strength. For EG II, the lack of inhibition by any of the
reported ligands is probably due to their short chain length. Indeed, this en-
zyme is known to be active on chromogenic cellotriosides but not on the corre-
sponding cellobiosides.[21]
In conclusion, the replacement of the 2-OH group of cellobiosides cannot
be considered to be a generic strategy for increasing their inhibitory power
toward cellulases. Nevertheless, a novel ligand for CBH I has been identified
that ranks among its most efficient inhibitors known to date. As this enzyme
is the major constituent of cellulase mixtures (up to 60% in H. jecorina), its in-
hibition has a dramatic effect on the total cellulolytic activity.[12] Consequently,
compound 15 might be considered for practical applications, such as the treat-
ment of termite infestations.[26]

T. Desmet et al.
172
EXPERIMENTAL
General
1H NMR spectra (500 MHz) were recorded in CDCl3 or D2O on a Bruker
WP-500 or AM-500 spectrometer, with residual protons of the solvent as inter-
nal standard; chemical shifts are expressed in ppm relative to tetramethylsi-
lane. IR spectra were recorded on a Beckman Acculab 4 spectrometer. MS data
were recorded by LCMS with an Agilent 1100 series HPLC equipped with an
Agilent G1946C ES-single quadrupole detector (as electrospray in positive or
negative mode from 1:1 0.5 mM aqueous ammonium acetate-CH3CN). Melt-
ing points were determined with a Reichert 269156 microscope and are uncor-
rected. Optical rotations were determined with a Perkin-Elmer 141 polarime-
ter. Thin-layer chromatography (TLC) was on Merck 60F254 silica gel plates,
with detection by observation under UV (254 nm) and carbonization on a hot
plate after dipping in a 10% soln of H2SO4 in isopropanol. Column chromatog-
raphy was with Merck silica gel 100 (0.063–0.200 mm), which was dried in
vacuo at 200◦C when indicated. Toluene and EtOAc were distilled prior to use,
CH2Cl2 was freshly distilled over P2O5, and Et2O a◦nd MeCN were kept over
˚
4 A molecular sieves that were dried in vacuo at 200 C. All reactions were run
with magnetic stirring under nitrogen atmosphere.
Azidonitration and Nitrate Hydrolysis
N-acetyl-3,6,2ꢁ,3ꢁ,4ꢁ,6ꢁ-hexa-O-acetyl-2-deoxy-2-azido-cellobiosylamine (4) and
2-deoxy-2-azido-3,6,2ꢁ,3ꢁ,4ꢁ,6ꢁ-hexa-O-acetyl cellobiose (6)
A soln of ceric ammonium nitrate (39.1 g, 71.4 mmol) in MeCN (150 mL)
◦
˚
was dried over 4 A molecular sieves (10 g) for 3 h at rt, and then cooled to –30 C
(cooling bath heptanol/solid CO2). Sodium azide (2.32 g, 35.7 mmol) and 2 (10 g,
17.86 mmol) were added under vigorous stirring. The mixture was allowed to
reach rt and was further stirred overnight. After pouring in ice-cold water (100
mL) and stirring for 15 min, celite (20 g) was added and the mixture was fil-
tered; the solids were washed with CH2Cl2 (3 × 100 mL). The combined organic
layers were washed with water (2 × 100 mL) and concentrated in vacuo. The
residue was dissolved in dioxane (100 mL) and a soln of sodium nitrite (10
g) in water (20 mL) was added. After stirring for 8 h at 80◦C, the soln was
cooled to rt and extracted with CH2Cl2 (3 × 100 mL). The combined organic
layers were washed with water (100 mL) and dried over anhydrous MgSO4.
After concentration in vacuo, the residue was preliminary purified by column
chromatography (eluent: toluene/EtOAc: 1/1). The first eluted compound was
recrystallized from isopropanol (100 mL; the mother liquor was concentrated
to 20 mL for a second crop), yielding 6 as white crystals (4.0 g, 36%). The

Cellulase Inhibitors
173
second eluted compound was further purified by column chromatography us-
ing the same eluent, yielding 4 as a white powder (2.48 g, 21%).
N-acetyl-3,6,2ꢁ,3ꢁ,4ꢁ,6ꢁ-hexa-O-acetyl-2-deoxy-2-azido-cellobiosylamine (4)
Brutoformula: C24H36O17N4 (MW = 660); mp 87◦C; [α]D23: +25 (c = 0.1;
CHCl3); Rf (toluene/EtOAc: 4/6): 0.28; IR (KBr): 2100, 1750, 1700, 1530, 1370,
1
1230, 1040 cm−1; H NMR (500 MHz; CDCl3): 6.55 (m, 1H), 5.72 (dd, 1H, J =
5.6 and 6 Hz), 5.27 (dd, 1H, J = 9 and 10 Hz), 5.14 (dd, 1H, J = 9.4 and 9.4 Hz),
5.08 (dd, 1H, J = 9.7 and 9.7 Hz), 4.93 (dd, 1H, J = 8.1 and 9.1 Hz), 4.47 (d, 1H,
J = 7.8 Hz), 4.39 (dd, 1H, J = 4.3 and 12.5 Hz), 4.36 (dd, 1H, J = 1 and 12 Hz),
4.12 (dd, 1H, J = 4.4 and 12 Hz), 4.09 (dd, 1H, J = 1 and 12 Hz), 3.87 (dd, 1H,
J = 5.6 and 10.2 Hz), 3.75 (m, 1H), 3.70 (dd, 1H, J = 8.6 and 8.6 Hz), 3.65 (m,
1H), 2.13 (s, 3H), 2.10 (s, 3H), 2.09 (s + s, 6H), 2.03 (s, 3H), 2.01 (s, 3H), 1.97 (s,
3H) ppm; MS: m/z (pos) 169 (8%), 331 [M/2+H+] (32%), 661 [M+H+] (100%),
683 [M+Na+] (20%); m/z (neg) 659 [M−H+] (62%), 719 [M+OAc−] (100%).
2-Deoxy-2-azido-3,6,2ꢁ,3ꢁ,4ꢁ,6ꢁ-hexa-O-acetyl cellobiose (6)
Brutoformula: C24H33O16N3 (MW = 619); mp 172◦C; [α]D25: +47 (c = 2.0;
CHCl3); Rf (toluene/EtOAc: 4/6): 0.48; IR (KBr): 3600–3300, 2100, 1750, 1430,
1370, 1230, 1030, 900 cm−1; 1H NMR (500 MHz; CDCl3): 5.49 (dd, 1H, J = 9.3
and 10.4 Hz), 5.34 (m, 1H), 5.17–5.07 (m, 2H), 4.93 (dd, 1H, J = 8 and 8 Hz),
4.54 (d, 1H, J = 8.0 Hz), 4.50 (m, 1H), 4.39 (dd, 1H, J = 4.1 and 8.4 Hz), 4.18 (m,
1H), 4.13 (dd, 1H, J = 4.5 and 12.0 Hz), 4.07 (dd, 1H, J = 2.0 and 12.2 Hz), 3.72
(dd, 1H, J = 9.7 and 9.7 Hz), 3.67 (m, 1H), 3.38 (dd, 1H, J = 8.3 and 10.2 Hz),
3.28 (dd, 1H, J = 2.7 and 10.5 (Hz), 2.13 (s, 3H), 2.11 (s, 3H), 2.09 (s, 3H), 2.03
(s, 3H), 2.01 (s, 3H), 1.99 (s, 3H) ppm; MS: m/z (pos) 331 (14%), 637 [M+NH+4 ]
(100%); m/z (neg) 410 (37%), 470 (84%), 530 (100%), 618 [M−H+] (95%).
Coupling of the Aglycons
2-Deoxy-2-azido-1a-trichloroacetamido-hexa-O-acetyl cellobioside (8)
To a soln of 6 (2 g, 3.23 mmol) and trichloroacetonitrile (1.3 mL, 12.9 mmol)
in CH2Cl2 (40 mL) at rt, 1,8-diazabicyclo-[5,4,0]-undec-7-ene (0.323 mmol,
48 µL) was added. After stirring for 2 h, the solvent was removed in vacuo
and the residue was purified by column chromatography with oven-dried silica
gel (eluent: toluene/EtOAc: 7/3), yielding 8 as a slightly yellowish foam (1.44 g,
58%). This compound is unstable and must be used within hours.
Brutoformula: C26H33O16N4Cl3 (MW = 763.5); mp 100◦C (dec); Rf
(toluene/EtOAc: 4/6): 0.58; IR (KBr): 2100, 1750, 1680, 1370, 1220, 1030 cm−1
.
2-Deoxy-2-azido-1β-benzyl-hexa-O-acetyl cellobioside (10)
A soln of 8 (610 mg, 0.799 mmol) and benzyl alcohol (0.163 mL, 1.60 mmol)
˚
in CH2Cl2 (10 mL) with 4 A molecular sieves (0.5 g) was stirred for 2 h at rt.

T. Desmet et al.
174
After cooling to –15◦C (bath with benzyl alcohol/solid CO2), a soln of BF3.OEt2
(9.8 µL, 0.080 mmol) in CH2Cl2 (100 µL) was added dropwise during 5 min; the
reaction was monitored by TLC and was complete after 1 h. Solid NaHCO3 (100
mg) was added and the mixture was brought to rt. After filtration, the solvent
was removed in vacuo and the residue was purified by column chromatography
(eluent: toluene/EtOAc: 7/3) yielding 10 as a white powder (450 mg, 79%).
Brutoformula: C31H39O16N3 (MW = 709); mp 133◦C; [α]D23: –20 (c = 0.1;
CHCl3); Rf (toluene/EtOAc: 6/4): 0.63; IR (KBr): 2100, 1750, 1430, 1370, 1230,
1160, 1050, 900 cm−1; 1H NMR (500 MHz; CDCl3): 7.37–7.30 (m, 5H), 5.13 (dd,
1H, J = 9.4 and 9.4 Hz), 5.06 (dd, 1H, J = 9.7 and 9.7 Hz), 4.93 (dd, 1H, J =
9.4 and 10.4 Hz), 4.90 (d, 1H, J = 11.8 Hz), 4.89 (dd, 1H, J = 8.0 and 9.4 Hz),
4.67 (d, 1H, J = 11.8 Hz), 4.53 (dd, 1H, J = 1.8 and 12.0 Hz), 4.48 (d, 1H, J =
8.0 Hz), 4.40 (d, 1H, J = 8.1 Hz), 4.38 (dd, 1H, J = 4.3 and 12.5 Hz), 4.11 (dd,
1H, J = 5.1 and 12.0 Hz), 5.05 (dd, 1H, J = 2.1 and 12.5 Hz), 3.69 (dd, 1H, J =
9.5 and 9.5 Hz), 3.66 (ddd, 1H, J = 2.2, 4.2 and 9.8 Hz), 3.52 (ddd, 1H, J =
1.8, 4.9 and 9.9 Hz), 3.46 (dd, 1H, J = 8.1 and 10.3 Hz), 2.15 (s, 3H), 2.09 (s,
3H), 2.08 (s, 3H), 2.02 (s, 3H), 2.01 (s, 3H), 1.97 (s, 3H) ppm; MS: m/z (pos) 727
[M+NH+4 ] (100%); m/z (neg) 769 [M+OAc−] (100%).
2-Deoxy-2-azido-1β-[m-iodo]benzyl-hexa-O-acetyl cellobioside (11)
Recipe as for 10. Starting with 8 (610 mg, 0.799 mmol) and m-iodobenzyl
alcohol (0.209 mL, 1.60 mmol), 11 was obtained as a white powder (497 mg,
72%).
Brutoformula: C31H38O16N3I (MW = 835); mp 134◦C; [α]D23: –18 (c = 0.1;
CHCl3); Rf (toluene/EtOAc: 6/4): 0.66; IR (KBr): 2100, 1750, 1430, 1380, 1230,
1
1160, 1040, 900 cm−1; H NMR (500 MHz; CDCl3): 7.71 (s, 1H), 7.65 (d, 1H,
J = 8 Hz), 7.32 (d, 1H, J = 8 Hz), 7.09 (dd, 1H, J = 8.0 and 8.0 Hz), 5.13 (dd,
1H, J = 9.3 and 9.3 Hz), 5.06 (dd, 1H, J = 9.8 and 9.8 Hz), 4.95 (dd, 1H, J =
9.3 and 10.3 Hz), 4.90 (dd, 1H, J = 8.0 and 9.3 Hz), 4.82 (d, 1H, J = 12.1 Hz),
4.60 (d, 1H, J = 12.1 Hz), 4.52 (dd, 1H, J = 2.0 and 7.9 Hz), 4.48 (d, 1H, J =
7.9 Hz), 4.38 (d, 1H, J = 8.1 Hz), 4.37 (dd, 1H, J = 4.4 and 12.5 Hz), 4.09 (dd,
1H, J = 5.1 and 12.1 Hz), 4.04 (dd, 1H, J = 2.2 and 12.4 Hz), 3.69 (dd, 1H, J =
9.5 and 9.5 Hz), 3.65 (ddd, 1H, J = 2.2, 4.4 and 9.9 Hz), 3.53 (ddd, 1H, J = 2.0,
5.1 and 9.9 Hz), 3.46 (dd, 1H, J = 8.1 and 10.3 Hz), 2.145 (s, 3H), 2.085 (s, 3H),
2.075 (s, 3H), 2.015 (s, 3H), 2.00 (s, 3H), 1.97 (s, 3H) ppm; MS: m/z (pos) 853
[M+NH+4 ] (100%); m/z (neg) 894 [M+OAc−] (100%).
1β-Benzyl-hepta-O-acetyl cellobioside (12)
Recipe as for 10. Starting with 9 (1.42 g, 2.133 mmol) and benzyl alcohol
˚
(0.441 mL, 4.266 mmol) in CH2Cl2 (30 mL), with 4 A molecular sieves (1 g), a
soln of BF3.OEt2 (27 µL, 0.22 mmol) in CH2Cl2 (0.5 mL), workup with NaHCO3
(200 mg), and column chromatography with the same eluent, 12 was obtained
as a white powder (870 mg, 56%).

Cellulase Inhibitors
175
Brutoformula: C33H42O18 (MW = 726); mp 170◦C; [α]D23: +44 (c = 0.1;
CHCl3); Rf (toluene/EtOAc: 4/6): 0.46; IR (KBr): 2940, 2860, 1750, 1430, 1370,
1230, 1040, 900 cm−1; 1H NMR (500 MHz; CDCl3): 7.36–7.25 (m, 5H), 5.13 (2 ×
dd, 2H, J obs. = 9.5 and 9.5 Hz), 5.06 (dd, 1H, J = 9.8 and 9.8 Hz), 4.97 (dd, 1H,
J = 8.0 and 9.5 Hz), 4.92 (dd, 1H, J = 8.6 and 8.7 Hz), 4.86 (d, 1H, J = 12.2 Hz),
4.59 (d, 1H, J = 12.2 Hz), 4.54 (dd, 1H, J = 2 and 11.9 Hz), 4.51 (d, 1H, J =
7.9 Hz), 4.49 (d, 1H, J = 7.9 Hz), 4.37 (dd, 1H, J = 4.1 and 12.5 Hz), 4.10 (dd,
1H, J = 4.8 and 11.9 Hz), 4.03 (dd, 1H, J = 2 and 12.5 Hz), 3.79 (dd, 1H, J = 9.5
and 9.5 Hz), 3.65 (ddd, 1H, J = 2, 4 and 9 Hz), 3.57 (ddd, 1H, J = 2, 4 and 9 Hz),
2.15 (s, 3H), 2.07 (s, 3H), 2.03 (s, 3H), 2.01 (s, 3H), 2.00 (s, 3H), 1.98 (s, 3H) ppm;
MS: m/z (pos) 744 [M+NH+4 ] (100%); m/z (neg) 785 [M+OAc−] (100%).
1β-[m-Iodo]benzyl-hepta-O-acetyl cellobioside (13)
Recipe as for 10. Starting with 9 (500 mg, 0.641 mmol) and m-iodobenzyl
˚
alcohol (0.163 mL, 1.28 mmol) in CH2Cl2 (10 mL), with 4 A molecular sieves
(0.5 g), a soln of BF3.OEt2 (8.0 µL, 0.065 mmol) in CH2Cl2 (100 µL), workup
with NaHCO3 (100 mg), and column chromatography with the same eluent, 13
was obtained as a white powder (266 mg, 57%).
Brutoformula: C33H41O17I (MW = 852); mp 202◦C; [α]D23: +40 (c = 0.1;
CHCl3); Rf (toluene/EtOAc: 1/1): 0.59; IR (KBr): 1750, 1430, 1240–1210, 1160,
1130, 1050–1030, 900, 780, 730 cm−1; 1H NMR (500 MHz; CDCl3): 7.64 (s, 1H),
7.63 (d, 1H, J = 7 Hz), 7.21 (d, 1H, J = 7.7 Hz), 7.07 (dd, 1H, J = 7 and 7.7 Hz),
5.15 (dd, 1H, J = 9.4 and 9.4 Hz), 5.13 (dd, 1H, J = 9.5 and 9.3 Hz), 5.06 (dd,
1H, J = 9.8 and 9.8 Hz), 4.97 (dd, 1H, J = 8.0 and 9.6 Hz), 4.92 (dd, 1H, J = 8.0
and 9.2 Hz), 4.79 (d, 1H, J = 12.7 Hz), 4.54 (dd, 1H, J = 2.0 and 12.1 Hz), 4.53
(d, 1H, J = 12.7 Hz), 4.51 (d, 1H, H, J = 8.0 Hz), 4.48 (d, 1H, J = 8.0 Hz), 4.37
(dd, 1H, J = 4.4 and 12.5 Hz), 4.10 (dd, 1H, J = 4.9 and 12.1 Hz), 4.04 (dd, 1H,
J = 2.1 and 12.5 Hz), 3.80 (dd, 1H, J = 9.6 and 9.6 Hz), 3.65 (ddd, 1H, J = 2.2,
4.3 and 9.9 Hz), 3.37 (ddd, 1H, J = 1.9, 4.8, and 9.9 Hz), 2.149 (s, 3H), 2.078 (s,
3H), 2.058 (s, 3H), 2.029 (s,+3H), 2.015 (s, 3H), 2.008 (s, 3H), 1.980 (s, 3H) ppm;
MS: m/z (pos) 870 [M+NH4 ] (100%); m/z (neg) 911 [M+OAc−] (100%).
Deprotection
2-Deoxy-2-azido-1β-benzyl cellobioside (14)
Compound 10 (200 mg, 282 µmol) was taken in a soln of sodium
methanolate (0.1 eq., 28.2 µmol) in MeOH (3.2 mL of an 8.7 µM soln). Workup
as for 16 gave 14 as a white powder (quant).
Brutoformula: C19H26O10N3 (MW = 457); mp 175◦C (isopropanol); [α]D23:
1
–10 (c = 0.1; MeOH); H NMR (500 MHz; D2O): 7.540–7.450 (m, 5H), 5.000
(d, 1H, J = 11.7 Hz), 4.835 (d, 1H, J = 11.7 Hz), 4.675 (d, 1H, J = 8.2 Hz),
4.545 (d, 1H, J = 7.8 Hz), 4.025 (dd, 1H, J = 1.9 and 12.3 Hz), 3.940 (dd, 1H,

T. Desmet et al.
176
J = 1.9 and 12.3 Hz), 3.865 (dd, 1H, J = 5.3 and 12.3 Hz), 3.751 (dd, 1H, J =
5.7 and 12.3 Hz), 3.735 (dd, 1H, J = 9.5 and 9.5 Hz), 3.620 (dd, 1H, J = 9.5
and 10.1 Hz), 3.610 (ddd, 1H, J = 1.9, 5.3, and 9.5 Hz), 3.535 (dd, 1H, J =
9.3 and 9.4 Hz), 3.515 (ddd, 1H, J = 1.9, 7.7, and 9.3 Hz), 3.440 (dd, 1H, J =
8.9 and 9.4 Hz), 3.400 (dd, 1H, J = 8.5 and 10.1 Hz), 3.335 (dd, 1H, J = 8.2 and
8.9 Hz) ppm; MS: m/z (pos) 250 (80%), 296 (26%), 350 (27%), 475 [M+NH+4 ]
(100%), 480 (85%); m/z (neg) 456 [M−H+] (100%), 457 (24%), 516 [M+OAc−]
(14%), 570 (17%), 913 [2M−H+] (29%).
2-Deoxy-2-azido-1β-[m-iodo]benzyl cellobioside (15)
Compound 11 (200 mg, 239 µmol) was taken in a soln of sodium
methanolate (0.1 eq., 23.9 µmol) in MeOH (2.7 mL of an 8.7 µM soln). Workup
as for 16 gave 15 as a white powder (quant).
Brutoformula: C19H25O10N3I (MW = 583); mp 197◦C (isopropanol); [α]D23:
1
–12 (c = 0.1; MeOH); H NMR (500 MHz; D2O): 7.94 (s, 1H), 7.83 (d, 1H, J =
8.2 Hz), 7.51 (d, 1H, J = 7.6 Hz), 7.25 (dd, 1H, J = 7.6 and 8.2 Hz), 4.94 (d,
1H, J = 11.7 Hz), 4.77 (d, 1H, J = 11.7 Hz), 4.65 (d, 1H, J = 8.2 Hz), 4.54 (d,
1H, J = 7.8 Hz), 4.02 (dd, 1H, J = 1.9 and 12.3 Hz), 3.94 (dd, 1H, J = 2.0 and
12.3 Hz), 3.86 (dd, 1H, J = 5.0 and 12.3 Hz), 3.76 (dd, 1H, J = 5.0 and 12.3 Hz),
3.73 (dd, 1H, J = 9.1 and 9.7 Hz), 3.62 (dd, 1H, J = 9.1 and 9.7 Hz), 3.60 (ddd,
1H, J = 1.9, 5.0, and 9.1 Hz), 3.53 (dd, 1H, J = 9.1 and 9.5 Hz), 3.51 (ddd, 1H,
J = 2.0, 5.0, and 9.5 Hz), 3.44 (dd, 1H, J = 9.1 and 9.1 Hz), 3.39 (dd, 1H, J =
7.8 and 9.1 Hz), 3.33 (dd, 1H, J = 8.2 and 9.1 Hz) ppm; MS: m/z (pos) 150
(68%), 172 (43%), 350 (32%), 388 (52%), 432 (55%), 476 (48%), 601 [M+NH+4 ]
(100%), 606 (73%), 964 (75%); m/z (neg) 582 [M−H+] (100%), 617 (13%), 642
[M+OAc−] (12%), 696 (29%).
1β-Benzyl cellobioside (16)
Compound 12 (200 mg, 275 µmol) was taken in a soln of sodium
methanolate (0.1 eq., 27.5 µmol) in MeOH (3.2 mL of an 8.7 µM soln, freshly
prepared from 10 mg Na in 50 mL MeOH). The reaction was complete after 2 h
at rt. A spoontip of silica gel was added, and after 5 min the soln was filtered
and concentrated in vacuo, yielding 16 as a white powder (quant).
Brutoformula: C19H28O11 (MW = 432); mp 162◦C (isopropanol); [α]D23: –38
1
(c = 0.1; MeOH); H NMR (500 MHz; D2O): 7.52–7.43 (m, 5H), 4.970 (d, 1H,
J = 11.7 Hz), 4.785 (d, 1H, J = 11.7 Hz), 4.585 (d, 1H, J = 8.2 Hz), 4.540
(d, 1H, J = 8.2 Hz), 4.020 (dd, 1H, J = 1.6 and 12.3 Hz), 3.940 (dd, 1H, J =
1.9 and 12.3 Hz), 3.850 (dd, 1H, J = 5.0 and 12.3 Hz), 3.760 (dd, 1H, J =
5.7 and 13.3 Hz), 3.685 (dd, 1H, J = 9.2 and 9.2 Hz), 3.630 (dd, 1H, J = 8.8
and 9.2 Hz), 3.610 (ddd, 1H, J = 1.6, 5.0, and 9.2 Hz), 3.535 (dd, 1H, J =
9.2 and 9.3 Hz), 3.515 (ddd, 1H, J = 1.9, 5.7, and 9.2 Hz), 3.440 (dd, 1H, J =
9.0 and 9.3 Hz), 3.380 (dd, 1H, J = 8.2 and 8.7 Hz), 3.335 (dd, 1H, J = 8.2
and 8.7 Hz) ppm; MS: m/z (pos) 163 (32%), 253 (22%), 271 (40%), 325 (100%),

Cellulase Inhibitors
177
450 [M+NH+4 ] (79%), 455 [M+Na+] (53%), 887 [2M+Na+] (79%); m/z (neg) 431
[M−H+] (100%), 863 [2M−H+] (29%).
1β-[m-Iodo]benzyl cellobioside (17)
Compound 13 (500 mg, 587 µmol) was taken in a soln of sodium
methanolate (0.1 eq., 58.7 µmol) in MeOH (6.7 mL of an 8.7 µM soln). Workup
as for 16 gave 17 as a white powder (quant).
Brutoformula: C19H27O11I (MW = 558); mp 183◦C (isopropanol); [α]D23: –35
1
(c = 0.1; MeOH); H NMR (500 MHz; D2O): 7.92 (s, 1H), 7.81 (d, 1H, J =
8.2 Hz), 7.49 (d, 1H, J = 7.9 Hz), 7.24 (dd, 1H, J = 7.9 and 8.2 Hz), 4.91 (d,
1H, J = 11.9 Hz), 4.75 (d, 1H, J = 11.9 Hz), 4.56 (d, 1H, J = 8.2 Hz), 4.54 (d,
1H, J = 8.2 Hz), 4.01 (dd, 1H, J = 1.9 and 12.3 Hz), 3.94 (dd, 1H, J = 1.9 and
12.3 Hz), 3.85 (dd, 1H, J = 5.1 and 12.3 Hz), 3.76 (dd, 1H, J = 5.6 and 12.3 Hz),
3.68 (dd, 1H, J = 8.6 and 8.6 Hz), 3.63 (dd, 1H, J = 8.6 and 8.8 Hz), 3.60 (ddd,
1H, J = 1.9, 5.1, and 8.6 Hz), 3.53 (dd, 1H, J = 9.3 and 9.3 Hz), 3.52 (ddd, 1H,
J = 1.9, 5.6, and 9.2 Hz), 3.44 (dd, 1H, J = 9.1 and 9.3 Hz), 3.38 (dd, 1H, J =
8.2 and 6.8 Hz), 3.34 (dd, 1H, J = 8.2 and 9.1 Hz) ppm; ESIMS: m/z (pos) 271
(26%), 325 (100%), 576 [M+NH4+] (73%), 581 [M+Na+], (49%) 1042 [2M+Na+]
(35%); m/z (neg) 557 [M−H+] (100%), 1015 [2M−H+] (15%).
Reduction of the Azido Functionality
2-Deoxy-2-amino-1β-benzyl cellobioside (18)
To a soln of 14 (63 mg, 138 µmol) in MeOH/water: 9/1 (1 mL), triphenyl
phosphine (72 mg, 276 µmol) was added. After stirring for 2 d at rt, the mix-
ture was concentrated in vacuo and purified by column chromatography (elu-
ent: isopropanol/water/saturated aq. ammonia: 18/2/1), yielding 18 as a white
powder (56 mg, 95%).
Brutoformula: C19H29O10N (MW = 431); mp 105◦C; [α]D23: –36 (c = 0.05;
1
MeOH); Rf (isopropanol/water/saturated aq. ammonia: 18/2/1): 0.35; H NMR
(500 MHz; D2O): 7.52–7.44 (m, 5H), 4.98 (d, 1H, J = 11.7 Hz), 4.77 (d, 1H, J =
11.7 Hz), 4.57 (d, 1H, J = 8.2 Hz), 4.54 (d, 1H, J = 8.2 Hz), 4.03 (dd, 1H, J =
1.9 and 12.3 Hz), 3.94 (dd, 1H, J = 2.2 and 12.3 Hz), 3.87 (dd, 1H, J = 5.0 and
12.3 Hz), 3.75 (dd, 1H, J = 6.0 and 12.3 Hz), 3.69 (dd, 1H, J = 9.5 and 9.8 Hz),
3.61 (ddd, 1H, J = 1.9, 5.0, and 9.8 Hz), 3.57 (dd, 1H, J = 9.5 and 10.0 Hz),
3.54 (dd, 1H, J = 9.5 and 9.5 Hz), 3.51 (ddd, 1H, J = 2.2, 6.0, and 9.5 Hz), 3.44
(dd, 1H, J = 9.5 and 9.5 Hz), 3.34 (dd, 1H, J = 8.2 and 9.5 Hz), 2.80 (dd, 1H,
J = 8.2 and 10.0 Hz) ppm; MS: m/z (pos) 324 (29%), 432 [M+H+] (100%), 454
[M+Na+] (15%), 734 (26%), 863 [2M+H+] (72%), 864 (33%), 885 [2M+Na+]
(18%); m/z (neg) 430 [M−H+] (100%), 466 (27%), 490 [M+OAc−] (37%), 439
(14%), 544 (14%), 861 [2M−H+] (37%).

T. Desmet et al.
178
2-Deoxy-2-amino-1β-[m-iodo]benzyl cellobioside (19)
To a soln of 15 (100 mg, 171 µmol) in MeOH/water: 9/1 (2 mL), triphenyl
phosphine (89 mg, 342 µmol) was added. After stirring for 2 d at rt, the mix-
ture was concentrated in vacuo and purified by column chromatography (elu-
ent: isopropanol/water/saturated aq. ammonia: 18/2/1), yielding 19 as a white
powder (89 mg, 94%).
Brutoformula: C19H28O10NI (MW = 557); mp 130◦C; [α]D23: –32 (c = 0.1;
1
MeOH); Rf (isopropanol/water/saturated aq. ammonia: 18/2/1): 0.40; H NMR
(500 MHz; D2O): 7.93 (s, 1H), 7.82 (d, 1H, J = 8.2 Hz), 7.50 (d, 1H, J = 7.7 Hz),
7.24 (dd, 1H, J = 7.7 and 8.2 Hz), 4.96 (d, 1H, J = 11.7 Hz), 4.77 (d, 1H, J =
11.7 Hz), 4.60 (d, 1H, J = 8.2 Hz), 4.54 (d, 1H, J = 8.0 Hz), 4.03 (dd, 1H, J =
1.9 and 9.5 Hz), 3.94 (dd, 1H, J = 2.0 and 9.5 Hz), 3.86 (dd, 1H, J = 5.0 and
9.5 Hz), 3.75 (dd, 1H, J = 5.0 and 9.5 Hz), 3.70 (dd, 1H, J = 9.5 and 9.5 Hz),
3.60 (ddd, 1H, J = 1.9, 5.0, and 9.5 Hz), 3.58 (dd, 1H, J = 9.5 and 10.0 Hz), 3.54
(dd, 1H, J = 9.5 and 9.5 Hz), 3.51 (ddd, 1H, J = 2.0, 5.0, and 9.5 Hz), 3.44 (dd,
1H, J = 9.5 and 9.5 Hz), 3.33 (dd, 1H, J = 8.2 and 9.5 Hz), 2.82 (dd, 1H, J = 8.0
and 10.0 Hz) ppm; MS: m/z (pos) 324 (15%), 558 [M+H+] (100%), 580 [M+Na+]
(23%), 1015 [2M+H+] (46%), 1016 (13%), 1037 [2M+Na+] (12%); m/z (neg) 556
[M−H+] (100%), 592 (17%), 616 [M+OAc−] (31%), 770 (12%), 1113 [2M−H+]
(28%).
Enzyme Assays
The cellulases CBH I, EG I, EG II, and CenA were kindly supplied
by J. Sta˚hlberg (Uppsala University, Zweden), M. Siika-aho (VTT, Finland),
R. Fagerstro¨m (Ro¨hm Enzyme, Finland), and J.-P. Belaich (Universite´ Mar-
seille I, France), respectively. Their activity was determined using 2-chloro-
4-nitrophenyl β-lactoside (CBH I and EG I), β-cellobioside (CenA), or β-
cellotrioside (EG II) as substrate.[21] The release of the chromophore was mea-
sured continuously in a microplate reader (EL 808, Bio-Tek Instruments) at
405 nm. All experiments were performed in a 0.1 M sodium phosphate buffer
at pH 5.7 and 37◦C. The substrate and inhibitor concentrations were varied
between 1/5 and 5× the value of KM and KI, respectively. The inhibition con-
stants were calculated from Dixon plots.[27] The measurements were performed
in triplicate and had a CV of less than 10%. Hydrolysis of the inhibitors was
checked by means of HPAEC-PAD, using a CarboPac PA-100 column, and the
settings supplied by the manufacturer (Dionex, California).
ACKNOWLEDGMENTS
The authors wish to thank Dr. Jerry Sta˚hlberg, Dr. Matti Siika-aho, Dr.
Richard Fagerstro¨m, and Dr. Jean-Pierre Belaich for providing enzyme

Cellulase Inhibitors
179
samples. This work was supported by a grant of the ‘Instituut voor de aan-
moediging van Innovatie door Wetenschap en Technologie in Vlaanderen’
(IWT-Flanders) to T.D.
REFERENCES
1. Henrissat, B.; Davies, G. Structural and sequence-based classification of glycoside
hydrolases. Curr. Opin. Struct. Biol. 1997, 7, 637–644.
2. Vocadlo, D.J.; Davies, G.J. Mechanistic insights into glycosidase chemistry. Curr.
Opin. Chem. Biol. 2008, 12, 539–555.
3. Wolfenden, R.; Lu, X.; Young, G. Spontaneous hydrolysis of glycosides. J. Am.
Chem. Soc. 1998, 120, 6814–6815.
4. Gloster, T.M.; Davies, G.J. Glycosidase inhibition: assessing mimicry of the tran-
sition state. Org. Biomol. Chem. 2010, 8, 305–320.
5. Asano, N. Sugar-mimicking glycosidase inhibitors: bioactivity and application.
Cell. Mol. Life Sci. 2009, 66, 1479–1492.
6. von Itzstein, M. The war against influenza: discovery and development of sialidase
inhibitors. Nat. Rev. Drug Disc. 2007, 6, 967–974.
7. Vasella, A.; Davies, G.J.; Bohm, M. Glycosidase mechanisms. Curr. Opin. Chem.
Biol. 2002, 6, 619–629.
8. McCarter, J.D.; Adam, M.J.; Withers, S.G. Binding energy and catalysis. Biochem.
J. 1992, 286, 721–727.
9. Namchuk, M.N.; Withers, S.G. Mechanism of Agrobacterium beta-glucosidase: ki-
netic analysis of the role of noncovalent enzyme/substrate interactions. Biochemistry
1995, 34, 16194–16202.
10. Roeser, K.-R.; Legler, G. Role of sugar hydroxyl groups in glycoside hydrolysis:
cleavage mechanism of deoxyglucosides and related substrates by beta-glucosidase A3
from Aspergillus wentii. Biochim. Biophys. Acta 1981, 657, 321–333.
11. Panday, N.; Meyyappan, M.; Vasella, A. A comparison of glucose- and glucosamine-
related inhibitors: probing the interaction of the 2-hydroxy group with retaining beta-
glucosidases. Helv. Chim. Acta 2000, 83, 513–538.
12. Foreman, P.K.; Brown, D.; Dankmeyer, L.; Dean, R.; Diener, S.; Dunn-Coleman,
N.S.; Goedegebuur, F.; Houfek, T.D.; England, G.J.; Kelley, A.S.; Meerman, H.J.;
Mitchell, T.; Mitchinson, C.; Olivares, H.A.; Teunissen, P.J.M.; Yao, J.; Ward, M. Tran-
scriptional regulation of biomass-degrading enzymes in the filamentous fungus Tricho-
derma reesei. J. Biol. Chem. 2003, 278, 31988–31997.
13. Henrissat, B.; Teeri, T.T.; Warren, R.A.J. A scheme for designating enzymes
that hydrolyse the polysaccharides in the cell walls of plants. FEBS Lett. 1998, 425,
352–354.
14. Lemieux, R.U.; Ratcliffe, R.M. Azidonitration of tri-O-acetyl-D-galactal. Can. J.
Chem. Rev. 1979, 57, 1244–1251.
15. Seeberger, P.H.; Roehrig, S.; Schell, P.; Wang, Y.; Christ, W.J. Selective formation of
C-2 azidodeoxy-D-glucose derivatives from D-glucal precursors using the azidonitration
reaction. Carbohydr. Res. 2000, 328, 61–69.

T. Desmet et al.
180
16. Kanda, T.; Brewer, C.F.; Okada, G.; Hehre, E.J. Hydration of cellobial by exo-
and endo-type cellulases: evidence for catalytic flexibility of glycosidases. Biochemistry
1986, 25, 1159–1165.
17. Excoffier, G.; Gagnaire, D.; Utille, J.P. Coupure se´lective par l’hydrazine des
groupements ace´tyles anome`res de re´sidus glycosyles ace´tyle´s. Carbohydr. Res. 1975,
39, 368–373.
18. Schmidt, R.R.J. Oligosaccharide synthesis via trichloroacetimidates. In Prepara-
tive Carbohydrate Chemistry; Hanessian, S., Eds.; Marcel Dekker: New York, 1997;
283–308.
19. Urban, F.J.; Moore, B.S.; Breitenbach, R. Synthesis of tigogenyl β-O-cellobioside
heptaacetate and glycoside tetraacetate via Schmidt’s trichloroacetimidate method;
some new observatons. Tetrahedron Lett. 1990, 31, 4421–4424.
20. Tian, W.Q.; Wang, Y.A. Mechanisms of Staudinger reactions within density func-
tional theory. J. Org. Chem. 2004, 69, 4299–4308.
21. Van Tilbeurgh, H.; Loontiens, F.G.; De Bruyne, C.K.; Claeyssens, M. Fluorogenic
and chromogenic glycosides as substrates and ligands of carbohydrases. Methods Enzy-
mol. 1988, 160, 45–59.
22. Damude, H.G.; Ferro, V.; Withers, S.G.; Warren, R.A.J. Substrate specificity of
endoglucanase A from Cellulomonas fimi: fundamental differences between endoglu-
canases and exoglucanases from family 6. Biochem. J. 1996, 315, 467–472.
23. Sulzenbacher, G.; Schulein, M.; Davies, G.J. Structure of the endoglucanase I
from Fusarium oxysporum: native, cellobiose, and 3,4-epoxybutyl beta-D-cellobioside-
˚
inhibited forms, at 2.3 A resolution. Biochemistry 1997, 36, 5902–5911.
24. Stahlberg, J.; Divne, C.; Koivula, A.; Piens, K.; Claessens, M.; Teeri, T.T.; Jones,
T.A. Activity studies and crystal structures of catalytically deficient mutants of cel-
lobiohydrolase I from Trichoderma reesei. J. Mol. Biol. 1996, 264, 337–349.
25. De Vriendt, K.; Sandra, K.; Desmet, T.; Nerinckx, W.; Van Beeumen, J.; Devreese,
B. Evaluation of automated nano-electrospray mass spectrometry in the determina-
tion of non-covalent protein-ligand complexes. Rapid Comm. Mass Spectr. 2004, 18,
3061–3067.
26. Zhoua, X.; Wheelera, M.M.; Oia, F.M.; Scharf, M.E. Inhibition of termite cellulases
by carbohydrate-based cellulase inhibitors: evidence from in vitro biochemistry and in
vivo feeding studies. Pest. Biochem. Physiol. 2008, 90, 31–41.
27. Cornish-Bowden, A. Fundamentals of Enzyme Kinetics; Portland Press; London,
1995.
28. Nerinckx, W.; Desmet, T.; Claeyssens, M. Itineraries of enzymatically and non-
enzymatically catalyzed substitutions at O-glycopyranosidic bonds. Arkivoc 2006,
90–116.