Glycobiology vol. 18 no. 8 pp. 570–586, 2008
Advance Access publication on May 22, 2008
Covalent inhibitors of glycosidases and their applications in biochemistry and biology
Brian P Rempel and Stephen G Withers1
Department of Chemistry, University of British Columbia, Vancouver, British
Columbia V6T 1Z3, Canada
Received on January 30, 2008; revised on May 09, 2008; accepted on May 12,
Glycoside hydrolases are important enzymes in a number of
essential biological processes. Irreversible inhibitors of this
class of enzyme have attracted interest as probes of both
structure and function. In this review we discuss some of the
in residue identification, structural and mechanistic investi-
gations, and finally their applications, both in vitro and in
vivo, to complex biological systems.
Keywords: affinity label/glycoside hydrolase/
Glycosidase classification and mechanism
Glycosidases are widespread enzymes that are responsible for
the hydrolytic cleavage of glycosidic bonds in contexts ranging
from primary metabolism through to glycoprotein glycan as-
enzymes, many of which are glycosidases (Davies et al. 2005).
Not surprisingly, therefore, their function or dysfunction has
been implicated in a number of different disease states, leading
to an interest in inhibitors of glycosidases as potential therapeu-
tics (Asano et al. 2000; Asano 2003a; Butters et al. 2003; de
Melo et al. 2006). There is also interest in this class of enzymes
for industrial and biotechnological applications.
Glycosidases can be classified into a number of sequence-
related families (Henrissat 1991; Henrissat and Bairoch 1993),
which can be found at http://www.cazy.org. Enzymes within a
sequence-related family catalyze the cleavage of the glycosidic
bond by the same mechanism and share a similar overall struc-
tural fold (as reviewed in Sinnott 1990; Rye and Withers 2000;
Zechel and Withers 2000; Withers 2001; Vasella et al. 2002;
Davies et al. 2005). The two most commonly employed mech-
anisms used by glycosidases to effect glycosidic bond cleavage
with overall inversion or retention of anomeric stereochemistry
are shown schematically below (Figure 1).
Inverting glycosidases (1a) effect bond cleavage through the
1To whom correspondence should be addressed: Tel: +1-604-822-3402;
Fax: +1-604-822-8869; e-mail: firstname.lastname@example.org
c ?The Author 2008. Published by Oxford University Press. All rights reserved. For permissions, please e-mail: email@example.com
and Withers 2000). Of the two carboxylic acids, only one is de-
protonated in the enzyme’s resting state and acts as a general
base, removing a proton from the incoming nucleophile (typi-
cally water under the normal glycosidase mechanism) during its
attack at the anomeric carbon. The other carboxylic acid acts as
a general acid residue, protonating the departing aglycone oxy-
gen atom and assisting in its departure from the anomeric cen-
ter. The bond-making and bond-breaking steps proceed through
a single, concerted oxocarbenium ion-like transition state in
which the developing positive charge at the anomeric carbon is
partially stabilized by electron donation from the ring oxygen.
The truncated sugar product is a hemi-acetal that initially has
the opposite configuration at the anomeric center to that of the
starting material; hence, the glycosidase is termed “inverting.”
Noteworthy features regarding this mechanism include a single
oxocarbenium ion-like transition state and the lack of any cova-
lent enzyme intermediate formed during the course of catalysis.
Most retaining glycosidases (1b), as with the inverting gly-
cosidases, also have a pair of essential carboxylic acid residues
but they are normally closer together at ∼5.5˚A apart (Zechel
and Withers 2000). One of the residues functions as a general
acid in the first mechanistic step by donating a proton during
the departure of the aglycone. In the same step, the second,
deprotonated carboxylate acts as a nucleophile, attacking the
anomeric carbon in a reaction that also proceeds through an ox-
ocarbenium ion-like transition state. This step, referred to as the
glycosylation step, leads to the formation of a covalently linked
glycosyl-enzyme intermediate that has an anomeric configura-
tion opposite to that of the starting material. The second step of
this reaction, the deglycosylation step, involves the hydrolytic
breakdown of the glycosyl-enzyme intermediate. The carboxy-
late that first acted as an acid catalyst now acts as a base by
abstracting a proton from the incoming nucleophile, normally a
ocarbenium ion-like transition state. The product thus obtained
is a hemi-acetal that initially has the same anomeric configura-
tion as the starting material. This mechanism differs from that
glycosyl-enzyme intermediate,andhenceproceeds throughtwo
oxocarbenium ion-like transition states.
that are substantially different from the inverting/retaining gly-
cosidases described above. Glycoside hydrolase family 18, 20,
56, 84, and 85 enzymes utilize a double-displacement mecha-
nism in which the catalytic nucleophile is not an enzymatic car-
boxylate, but instead is the oxygen of the substrate acetamide
group (Knapp et al. 1996; Mark et al. 2001; Macauley et al.
2005). Glycoside hydrolase family 4 and 109 enzymes utilize a
by guest on June 7, 2013
Irreversible labeling agents of glycosidases
Fig. 1. (A) Mechanism for an inverting β-glucosidase. (B) Mechanism for a retaining β-glucosidase. Note that the proton delivered during general acid catalysis is
delivered in either a syn- or anti-fashion, depending on the specific enzyme.
unique NAD+-dependent redox elimination/addition sequence
(Yip and Withers 2006). Neither of these mechanisms will be
discussed in further detail in this review.
Glycosidase inhibition: an overview
Enzyme inhibitors can be divided, broadly, into two classes:
noncovalent and covalent, with members of each class having
different applications. Noncovalent inhibitors of glycosidases
bind reversibly and have the greatest potential as therapeutics.
Asano et al. 2000; Lillelund et al. 2002; Asano 2003a,b; Butters
et al. 2003; de Melo et al. 2006), and are not the intended sub-
ject of this review. Covalent inactivators of glycosidases are
a class of molecules that ablate the enzyme activity through
the formation of a covalent bond between the enzyme and
some functionality on the inactivator (Legler 1990; Withers and
Aebersold 1995). The bond is typically formed by the attack
of an enzyme-based nucleophile onto an electrophilic portion
of the inactivator, leading to covalent attachment of the inac-
tivator. This attachment leads to loss of activity, most often
because it either physically blocks access to the enzyme ac-
tive site or modifies an active site residue that is critical for
Irreversible glycosidase inhibitors can be used for many dif-
ferent purposes. One of the earliest and still most prevalent
uses is in the identification of active site residues. Mutation
of the identified residues followed by kinetic analysis of mu-
tants modified at that position can confirm their function in
either catalytic or structural roles. Covalent inactivators have
by guest on June 7, 2013
B P Rempel and S G Withers
Fig. 2. Photoreactive probe structures.
also seen use in studying the catalytic mechanism(s) by which
glycosidases function, both through kinetic and structural ex-
amination. Highly specific probes have been used to selec-
tively inactivate a target enzyme or enzyme activity in complex
biological systems while observing the effect of this “deletion”
on the organism. This has been further extended to the design of
specific probes for the discovery and characterization of novel
Covalent glycosidase inactivators can be shown to be active
the enzyme, by incubation of the inactivator in the presence and
absence of a known active site-binding noncovalent inhibitor.
hibition should be observed if it is indeed active site-directed, as
The kinetics or extent of inactivation can usually be studied by
incubating a solution of inactivator plus enzyme and removing
enzyme aliquots at various time points. Residual enzymatic ac-
tivity in these aliquots can be assayed using a known substrate
to reveal the degree of inactivation as a function of time. Anal-
ysis of the data can reveal a time-dependent irreversible loss of
enzyme activity, and the kinetic parameters of inactivation can
the data (Kitz et al. 1965; Mosi and Withers 2002; Wicki et al.
2002). No general statements can be made about the rapidity of
inactivation with any class of compounds, as the rate constant
for inactivation is highly dependent on both the inactivator and
the specific enzyme being investigated. The time necessary to
completely ablate the activity for a given enzyme and inhibitor
pair can range from milliseconds to weeks, so interested readers
are encouraged to consult the original articles for more details
on rate constants or rates of inactivation.
Irreversible inhibitors can be divided into two general cate-
specificity for a given protein and a reactive functionality that
will irreversibly covalently modify a neighboring region of the
protein (Fersht 1999). These affinity labels can be further sub-
divided into two classes: labels that are inherently reactive as
a consequence of their chemical bonding (Fersht 1999) and la-
(Vodovozova 2007). By contrast, a mechanism-based inhibitor
it produces a species that reacts to form a covalent bond to the
enzyme (Legler 1990; Withers and Aebersold 1995). In this
review we will first discuss affinity labels and then mechanism-
based inhibitors as covalent inactivators of glycosidases, with
Photoreactive affinity labels
labeling has not received a great deal of attention. This class of
probe generally consists of compounds containing a specificity
moiety that is attached to a diazirine or aryl azide. These func-
nonspecific and difficult to predict.
One of the early examples of a PA probe applied to gly-
cosidase labeling was the use of a diazirine as the photoreac-
tive group linked through a C-glycosidic linkage to a galac-
topyranosyl residue as the specificity tag (Figure 2) (Kuhn and
Lehmann 1987; Kuhn, Lehmann, Jung 1992). This compound
(2a) was found to be a modest inhibitor and was used to label
Escherichia coli lacZ β-galactosidase upon irradiation. Only
moderate levels of inactivation were observed. A radiolabeled
version of the inactivator was prepared and used to tag the en-
tified, two short polypeptides were isolated and sequenced, and
on this basis proposed to be located close to the active site. With
the later determination of the three-dimensional structure of the
enzyme by X-ray crystallography (Jacobson et al. 1994; Juers
et al. 2000, 2001), it is possible to reexamine this labeling result
and see that, while one of the two peptides does indeed lie very
close to the active site, the other one is near the surface of the
of the probe for the enzyme active site.
A PA probe for human lysosomal hexosaminidases A
and B, which are responsible for ganglioside degradation,
was synthesized and tested (Kuhn, Lehmann, Sandhoff 1992;
Liessem et al. 1995). The thioglycoside (2b) was found
to only be a modest inhibitor of enzyme activity, and the
diazirine-derived carbene was not a very efficient protein la-
bel. Despite these drawbacks, the labeling experiment was
still successful in identifying a catalytic glutamate in hex-
osaminidase B (Liessem et al. 1995), a residue that was
subsequently shown through biochemical (Hou et al. 2001),
bioinformatic (Fernandes et al. 1997), and structural (Mark
et al. 2003) studies to act as the general acid/base during catal-
ysis. This is an example in which a covalent inactivator of
a glycosidase was used to obtain structural, mechanistic, and
by guest on June 7, 2013
Irreversible labeling agents of glycosidases
Fig. 3. Other affinity label structures.
sequence information prior to a three-dimensional structure
becoming available (Mark et al. 2003; Lemieux et al. 2006).
One feature of note is that, although enzymes in this family
(GH family 20) are retaining enzymes, they do not use an enzy-
matic carboxylate as the nucleophile. Rather, the amide oxygen
on the N-acetyl portion of the substrate acts as the nucleophile.
As a result, many covalent labeling strategies discussed below
that rely on the specific enzymatic nucleophile in the double-
displacement mechanism would not be applicable to this class
Another subclass of PA labeling involves the use of a puta-
tive transition state analog as opposed to a substrate analog to
tion. In both cases described, an aryl azide was used as the pho-
Mancini 1990) and two mammalian (Vanderhorst, Rose 1990;
Kopitz et al. 1997) sialidases, using a similar concept, where
sist in the identification of the labeled protein in complex mix-
tures or in a multisubunit enzyme complex. More recently, an
N-alkylated derivative of 1-deoxynojirimycin with an aryl azide
appended to the end of the N-alkyl chain (2d) was used to in-
hibit and label human α-glucosidase I (Romaniouk et al. 2004).
This enzyme is involved in the protein quality control machin-
ery and ensures proper folding of proteins in the endoplasmic
reticulum before they are sorted for trafficking; misfolded pro-
teins are instead targeted for degradation (Meusser et al. 2005).
Compound 2d was used to bind to and selectively tag human
Fig. 4. Mechanism of generation of reactive species from (A) glycosyl
triazene inactivators and (B) glycosyl diazomethyl ketone inactivators.
α-glucosidase I in a complex microsomal mixture of proteins.
The exact site of labeling was not determined, although the au-
thors were able to localize the label to the highly conserved
polypeptide spanning residues 582–598 that is thought to make
up part of the substrate-binding site (Romaniouk and Vijay
Other affinity labels
The other major classes of affinity labels are those wherein the
reagents possess a functional group that is inherently chemi-
cally reactive (Figure 3). The first five compounds discussed,
3a–e, despite their clever design, have not seen wide use as
glycosidase-labeling agents. This is principally due to their in-
stability toward spontaneous hydrolysis, as discussed below.
The first three molecules (3a–c), all C-glycosides, act as co-
valent glycosidase inactivators by the attack of an enzymatic
nucleophile onto the highly reactive diazomethyl group which
is activated by protonation (Marshall et al. 1981; Bemiller
et al. 1993). While 3a/3b and 3c have different modes of ac-
tivation, they all generate highly reactive carbon species that
inactivate the enzyme by undergoing rapid attack by an enzy-
matic nucleophile (Figure 4). However, all were hydrolytically
by guest on June 7, 2013
B P Rempel and S G Withers
unstable at pH <7, limiting the range of glycosidases suscep-
tible to inactivation. Glycosylmethyl-triazenes 3a and 3b were
used to inactivate a variety of retaining β-glycosidases, and also
surprisingly showed very weak inactivation of some retaining
α-glycosidases (Marshall et al. 1981). Notably, the inactivation
of E. coli lacZ β-galactosidase was studied in more detail, and
the site of labeling was determined to be Met-500 (Sinnott and
Smith 1976; Fowler et al. 1978; Sinnott and Smith 1978). Al-
though the glycosylmethyl-triazenes were tested against two in-
verting β-glycosidases, no inactivation was observed (Marshall
et al. 1981). Galactosyl-diazomethyl ketone 3c was shown to be
and the inactivation was shown to be active site directed. The
same compound failed to label E. coli lacZ β-galactosidase, the
only other enzyme against which it was tested (Bemiller et al.
1993). A family of glucosylthio-hydroquinones (generally rep-
as irreversible inhibitors of two very well studied retaining β-
glycosidases, the β-glucosidase from Agrobacterium sp. (Abg)
and the xylanase from Cellulomonas fimi (Schnabelrauch et al.
1994). These Michael-type acceptors inactivated Abg reason-
from Cellulomonas fimi. They were found to be hydrolytically
unstable in buffered solution and hence could be tested in wa-
ter alone. This hydrolytic instability has thus severely limited
the study of this class of compound and precluded identifi-
cation of the site of labeling. Another electrophilic reagent,
the glucosyl-isothiocyanate (3e) was shown to label almond β-
glucosidase (Shulman et al.1976). Although the inactivator was
shown to be active site directed and irreversible, the inactivator
efficiency was not particularly high, and it was also unstable
under the inactivation conditions. Owing to these drawbacks,
no active site residue could be identified using this inactivator.
Finally, the N-bromoacetyl glycosylamines and bromoketone
C-glycosides (represented by the general structures 3f–h) have
been used to label and inactivate several glycosidases (Naider
et al. 1972; Black et al. 1993; Keresztessy et al. 1994; Tull
et al. 1996; Howard and Withers 1998a,b; Chir et al. 2002;
Kiss et al. 2002; Vocadlo et al. 2002; Jager and Kiss 2005).
Unlike the classes of compounds 3a–e discussed above, the
N-bromoacetyl glycosylamines and bromoketone C-glycosides
have typically proven to be sufficiently stable toward spon-
taneous decomposition to be useful as labeling agents when
care is taken to select the proper inactivator, as described be-
low. The N-bromoacetyl glycosylamines of general structure
3f have proven useful as probes that often label the acid/base
catalytic residue in retaining β-glycosidases (Keresztessy
et al. 1994; Tull et al. 1996; Chir et al. 2002; Vocadlo
et al. 2002), although compounds in this class have also been
observed to label other residues, including an active site me-
thionine (Met-500) in E. coli lacZ β-galactosidase (Naider et al.
1972). Interestingly, this is the same methionine as that which
was labeled by the galactosylmethyl triazene described above
(Sinnott and Smith 1976, 1978; Fowler et al. 1978). In a second
case in which this reagent was employed, the authors propose
that the catalytic nucleophile is the site of labeling on the basis
of the pH dependence on the rate of inactivation and the pH/rate
profile of enzymatic substrate hydrolysis (Jager and Kiss 2005).
(Knowles 1976), so in the absence of a structural study and a
detailed kinetic analysis of both wild-type and site-directed mu-
tants (see Vocadlo et al. 2002 for a good example of this type of
analysis), caution should be exercised. The model enzyme Abg
also did not label cleanly and, while inactivation was shown to
two, or three N-acetyl-glucosaminyl moieties during the course
of the inactivation experiment as determined by ESI-MS (Black
et al. 1993). This result again emphasizes the danger in relying
on kinetic data alone. Thus, while the N-bromoacetyl glycosy-
lamines have proven valuable in the labeling and identification
of the acid/base catalyst in some retaining β-glucosidases, as-
signments made on this basis should be verified by other means.
sponse to the need for a covalent inactivator directed toward the
catalytic acid/base residue in retaining α-glycosidases (Howard
and Withers 1998a,b). Bromoketone C-glycosides were chosen
as labels for retaining α-glycosidases because the equivalent N-
bromoacetyl α-glycosylamines are very difficult to synthesize.
While bromoketone C-glycosides were also synthesized with
(3g), they did not meet with much success and showed no
advantages over the corresponding N-bromoacetyl glycosy-
lamines (Howard and Withers 1998a). However, bromoketone
C-glycoside 3h was synthesized and shown to be an active site-
directed inactivator of yeast α-glucosidase, a well-studied re-
covalently modified by the inactivator aligned nicely (Howard
and Withers 1998b) with the catalytic acid/base residue that
had been previously identified in other family 13 glycosidases
(Qian et al. 1994; Svensson 1994; Knegtel et al. 1995). Un-
fortunately, while bromoketone C-glycosides such as 3h are
available, a version of this analog with an equatorial hydroxyl at
C-2(gluco-configuration) rapidly underwent intramolecular cy-
clization, precluding the use of such reagents as general affinity
In contrast to affinity labels, a mechanism-based inactivator
(MBI) is a molecule that is chemically inert until activated by
the catalytic machinery of the enzyme. In the context of gly-
cosidases, this activation typically comes in one of two ways,
either the hydrolytic cleavage of a glycosidic bond releases a
enzymatic nucleophile attacks a center activated by the general
MBIs with reactive aglycones
To date, there have only been two classes of molecules that rely
on enzymatic cleavage of a glycosidic bond to release an ac-
tivated aglycone, a difluoroalkyl glucoside (Figure 5, 5a) and
a series of related compounds featuring an activated phenyl-
methyl aglycone (5b–e). After enzymatic cleavage of 5a, the
initially formed α,α-difluoro-alcohol rapidly decomposes with
the release of a molecule of HF and generation of a reactive acyl
fluoride. This acylating agent then reacts with a nucleophilic
residue in the enzyme, irreversibly inhibiting it (Figure 6).
The only example of this class of MBI to date (Halazy
et al. 1989) was found to inactivate yeast α-glucosidase rapidly
and irreversibly, although it acted as a substrate rather than an
inactivator for the sucrase-isomaltase enzyme from rat’s small
by guest on June 7, 2013
Irreversible labeling agents of glycosidases
Fig. 5. Structures of mechanism-based inactivators with reactive aglycones.
Fig. 6. Mechanism of generation of reactive aglycone from difluoroalkyl
the reactive aglycone, once released, has no inherent affinity for
the enzyme active site and is free to diffuse out where it may
Molecules of general structure 5b–e which feature an ac-
tivated phenylmethyl aglycone have been studied as glycosi-
dase inactivators for a number of enzymes (Driguez et al. 1992;
Briggs et al. 1995; Zhu et al. 1998; Ichikawa and Ichikawa
2001; Tsai et al. 2002; Kurogochi et al. 2004; Hinou et al.
2005; Lo et al. 2005, 2006; Lu et al. 2005; Shie et al. 2006).
The general mechanism by which this class of MBI generates
the reactive species is shown in Figure 7. Indeed, this class of
glycosidase inhibitor is a subset of a more general class of hy-
activated phenylmethyl leaving group has also been adopted for
the study of phosphatases (Myers and Widlanski 1993; Wang
et al. 1994) and phospho-triesterases (Lo et al. 1996). A di-
Fig. 7. Mechanism of generation of reactive aglycone from activated
fluoromethyl group is best used as the latent quinone methide
equivalent, since the monofluoromethyl was found to be hy-
drolytically unstable prior to enzymatic cleavage (Wang et al.
1994). Alternatively, in molecules such as 5d, when R = an
ester or amide, the monofluoro species is also sufficiently sta-
ble toward spontaneous hydrolysis (Lo et al. 1996, 2005; Tsai
et al. 2002; Shie et al. 2006). Both ortho (Driguez et al. 1992;
Ichikawa and Ichikawa 2001; Kurogochi et al. 2004; Hinou
et al. 2005; Lu et al. 2005; Lo et al. 2006) (e.g., 5b) and para
(Tsai et al. 2002; Lo et al. 2005; Shie et al. 2006) (5c) substitu-
tions of the fluoromethyl group on the aromatic ring have been
investigated, with a variety of different glycone portions being
investigated, including β-glucosides (Tsai et al. 2002; Lo et al.
2005; Shie et al. 2006), a β-galactoside (Kurogochi et al. 2004),
a β-xyloside (Lo et al. 2006), a β-N-acetyl-glucosaminide, and
α-sialosides (Driguez et al. 1992; Hinou et al. 2005; Lu et al.
2005). Compounds such as 5d have incorporated a reporter
group (often dansyl or biotin) attached via a linker to an amide
or ester off the phenyl ring. This class of MBI has had lim-
ited success, mostly owing to the fact that the activated quinone
methide species released has no specific affinity for the target
glycosidase. Consequently, it may well diffuse out of the active
site and label a remote residue or other nearby proteins (Bolton
et al. 1997; Lo et al. 2005). This property renders such reagents
essentially useless for proteomic analysis. Indeed, no reports
in which one specific active site residue has been selectively
labeled have appeared, although in some cases labeling outside
the active site can be quite selective (Hinou et al. 2005). In
most cases, the stoichiometry and/or site of labeling were not
It is interesting to note that a natural product, salicortin (5e),
has been isolatedwhich was shown tofragment upon enzymatic
hydrolysis to generate an activated quinone methide (Clausen
et al. 1990), and that this quinone methide was shown to be
an inactivator of the β-glucosidase Abg (Zhu et al. 1998). It
was shown to be a repeat-attack MBI, with multiple alkylations
around the enzyme active site occurring before inactivation was
covery of rationally designed and naturally isolated compounds
by guest on June 7, 2013
B P Rempel and S G Withers
Fig. 8. General mechanism of inactivation by an epoxide (X = O) or an
aziridine (X = NH).
Fig. 9. Structures of aziridine-based inactivators.
one avenue for the development of new covalent inactivators of
glycosidases may be through examination of natural products
and mimicking their functionality.
Epoxide- and aziridine-based inactivators
Aziridines and epoxides have been utilized to label the catalytic
nucleophile in a number of retaining glycosidases. When in-
corporated with a functionality that imparts specificity toward
the active site, they typically are activated by proton donation
from one catalytic carboxylic acid and are attacked by the other
to covalently label the enzyme active site (Figure 8). Of these,
the glycosyl-aziridines have received considerably less atten-
tion than the epoxides (discussed below), and examples in the
literature appear to be restricted to those shown in Figure 9.
Aziridine-based inhibitors enjoy the theoretical advantage of
having a higher initial, noncovalent affinity for a glycosidase
active site owing to the positively charged (when protonated)
nitrogen atom, which should help direct the inactivator toward
the negatively charged active site. The spiroaziridines 9a and
9b were tested against a few different enzymes and shown to
be only very weak inactivators of a single β-glucosidase and a
demands imposed at the anomeric center (Kapferer et al. 2003).
Spiroaziridine 9c was shown to be a potent inactivator of one
α-galactosidase, although it showed no activity against other
enzymes (Tong and Ganem 1988). The other glycosyl-aziridine
that has been tested is the conduritol aziridine 9d (Caron and
Withers 1989), an aza-analog of conduritol B-epoxide (CBE,
Fig. 10. Structures of epoxide-based inactivators.
discussed below). It was found to be a modestly potent irre-
versible inactivator of both the β-glucosidase Abg and the yeast
α-glucosidase, showing slightly higher activity against the α-
Considerably more work has been done using epoxides as the
active electrophile. The general structures for the three classes
of epoxides that have been studied are shown in Figure 10.
One of the earliest classes of glycosyl-epoxides to be studied
is that of the exo-alkyl epoxide glycosides, of general structure
10a. These have been employed to label a catalytic carboxy-
late in a variety of different enzymes (Legler and Bause 1973;
Shulman et al. 1976; Clarke and Strating 1989; Hoj et al. 1991,
1992; Keitel et al. 1993; Macarron et al. 1993; Havukainen
et al. 1996; Sulzenbacher et al. 1997). The labeled residue
has often been the catalytic nucleophile, and this has been
definitively proven in some cases through the solution of the
three-dimensional X-ray crystallographic structure (Keitel et al.
1993; Havukainen et al. 1996; Sulzenbacher et al. 1997). These
X-ray structures show a covalent bond between the residue
known to act as the catalytic nucleophile and the alkyl chain
of the inactivator. In one case, it was noted that the residue
labeled depended on the chain length of the inactivator: a 2,3-
epoxypropyl β-D-xyloside labeled the catalytic nucleophile as
expected, while a 3,4-epoxybutyl β-D-xyloside labeled the cat-
alytic acid/base residue in a retaining 1,4-xylanase from Tri-
choderma reesei (Havukainen et al. 1996). This unusual re-
sult highlights the flexibility of the alkyl chains and their
ability to shift the reactive epoxide into different orienta-
tions depending on the specific interactions with the enzyme
active site. Molecular dynamics simulations allowed a theo-
retical rationale for this observation to be offered (Laitinen
et al. 2000). A similar result was found in a systematic study
of exo-alkyl epoxide glycosides of different chain lengths and
different epoxide stereochemistries. It was found that the two
different enzymes studied were inactivated by epoxides of dif-
(Hoj et al. 1991).
CBE: mechanistic and structural studies
CBE (10b) is one example of a very well characterized MBI
of retaining glycosidases. There already is an excellent review
by guest on June 7, 2013
Irreversible labeling agents of glycosidases
of much of the early work on use of the conduritol epoxides
and conduritol bromo-epoxides (Legler 1990), thus only a few
examples will be discussed here. CBE was used very early on to
examine both the glycosidase active sites present in mammalian
sucrase-isomaltase. It was shown to be an effective inactiva-
tor of hydrolytic activity (Quaroni et al. 1974) and a tritiated
version of the inactivator was used to label and identify cat-
alytic carboxylates in both the sucrase and isomaltase active
sites through peptic digestion and subsequent sequencing of ra-
dioactively labeled peptides (Quaroni et al. 1974; Quaroni and
behavior was identified and the opening of the epoxide ring by
the enzyme was shown to be completely stereospecific and con-
sistent with the general mechanism shown in Figure 8 (Braun
et al. 1977).
This general strategy of using a radioactive derivative of
CBE highlights one of the most important applications for co-
valent inactivators of glycosidases: the alkylation, and subse-
quent identification of active site residues. Until bioinformat-
ics approaches permitted the prediction of active site residues
(Henrissat 1991; Henrissat and Bairoch 1993), these types of
experiments were one of the few ways of identifying can-
didate active site residues. Even with the predictive abilities
of bioinformatics, experimental verification of those predic-
tions using such reagents as CBE remains an active area of
research (see Febbraio et al. 1997; Hrmova et al. 1998; Li et al.
2001 for some examples). Once a putative residue is re-
vealed, however, a highly regarded strategy involves the cre-
ation of variant proteins in which the purported active site
residue is mutated, followed by detailed mechanistic analysis
of mutants so generated (see Ly and Withers 1999; Vocadlo
et al. 2002 for an example of this approach, reviewed in Ly and
Interestingly, owing to the element of symmetry present in
CBE, this reagent has also been used as an inactivator of some
retaining α-glucosidases (Hermans et al. 1991; Iwanami et al.
taining α-glucosidases to also label the catalytic nucleophile in
cessful trapping and identification of the nucleophile in human
lysosomal α-glucosidase (Hermans et al. 1991). Since the first
X-ray crystallographic structure of a homologous enzyme was
the structural symmetry of CBE that permits it to act as an inac-
tivator of both α-glucosidases and β-glucosidases also allows it
tobindinsomeenzyme active sitesinmorethanonemode,thus
a carboxylate using the CBE analog conduritol C cis-epoxide to
label E. coli lacZ β-galactosidase (Herrchen and Legler 1984)
and CBE itself to label human lysosomal glucocerebrosidase
(GCase) (Dinur et al. 1986), and almond β-glucosidase (Legler
and Harder 1978). In all three cases, the labeled residue was
mistakenly identified as the enzymatic nucleophile, and the nu-
cleophile was later correctly identified with a more specific
class of reagent, the 2-deoxy-2-fluoro glycosides (discussed be-
low) (Gebler et al. 1992; Miao et al. 1994; He and Withers
Fig. 11. Mechanism of inactivation by conduritol B-epoxide of (A) a retaining
α-glucosidase and (B) a retaining β-glucosidase.
CBE: biological applications
One of the most important uses of CBE has been to study mam-
malian retaining β-glucosidases, and in particular human lyso-
somal GCase. GCase is normally responsible for the hydrolytic
cleavage of the β-glucosyl residue in β-glucosyl ceramide
A deficiency in this enzyme is known to cause accumulation
of the substrate, leading to a pathological condition known as
Gaucher’s disease (Zhao and Grabowski 2002; Butters 2007).
An X-ray crystal structure of CBE covalently bound to the
GCase enzyme has been solved (Premkumar et al. 2005), which
gives some insight into the biochemical basis for some forms
of Gaucher’s disease (Dvir et al. 2003; Brumshtein et al. 2006,
2007; Lieberman et al. 2007).
CBE has been shown to be a selective inactivator of human
GCase while not affecting the activity of other known mam-
to discover and characterize novel mammalian β-glucosidases
since CBE can be added to crude enzyme mixtures to selec-
tively inactivate GCase, allowing other β-glucosidase activities
to be examined. This approach allowed the characterization of
a non-lysosomal β-GCase activity belonging to β-glucosidase
2 (Boot et al. 2007). This had not been possible previously
owing to the enzyme’s instability (Vanweely et al. 1993). CBE
has also been helpful in characterizing the activity of a broad
specificity β-glucosidase by selectively knocking out GCase
activity in cell homogenates (Hays et al. 1998). Interestingly,
another β-glucosidase activity that can cleave 6?-acylamino-4-
methylumbelliferyl β-D-glucosides was shown to be insensitive
to CBE, confirming that it was not GCase (Mikhaylova et al.
by guest on June 7, 2013
B P Rempel and S G Withers
The selectivity of CBE for GCase has also been exploited
in the selective ablation of GCase activity in both cell culture
and in animal models to answer interesting biochemical and
biological questions about the role of the enzyme and its sub-
strate. CBE allows for the selective inhibition of mammalian
GCase and permits the accumulation of unprocessed substrate
GlcCer. In one example, multi-drug resistance in cancer cells
was mimicked in normally drug-sensitive cancer cells by treat-
ment with CBE. This has lent support to the hypothesis that
increased levels of GlcCer correlate with increased drug resis-
tance in cancer cells (Morjani et al. 2001). Gaucher’s cells have
also been observed to have an altered redox state and increased
levels of reactive oxygen species, an observation that could be
directly attributed to GCase activity since healthy fibroblasts
treated with CBE demonstrated the same behavior (Deganuto
et al. 2007). Finally, the ability to selectively turn-off GCase
activity in mammalian systems using CBE has been exploited
in the characterization of the important role that glycosphingo-
lipids play in skin biochemistry (Takagi et al. 1999; Holleran
et al. 2006). CBE was used to help localize GCase activity
to the stratum corneum layer of the skin, which is known to
play a crucial role in maintaining the skin’s ability to prevent
excessive water loss and prevent absorption of foreign sub-
stances (Takagi et al. 1999). It has been suggested that proper
GCase activity in the statum corneum is important for main-
taining the lipid balance for proper function (Holleran et al.
disease, and as a result many investigators have used CBE
to mimic this aberrant “Gaucher’s cell” phenotype. CBE-
treated macrophages do in fact display the altered morphol-
ogy of the Gaucher’s cell (Yatziv et al. 1988), and this al-
teration can be exacerbated by exposure to liposomes (Das
et al. 1987) or red blood cells (Schueler et al. 2004) to ar-
tificially increase the GlcCer storage levels. The authors ar-
gued that this would better mimic the natural phagocytic role
that macrophages play in the clearance of red and white blood
cells (Schueler et al. 2004). CBE-induced storage of GlcCer in
macrophages has also been used to study altered phosphatidyl-
choline metabolism (Trajkovic-Bodennec et al. 2004), which
phate (CTP):phosphocholine cytidylyltransferase α (Kacher
cells has been investigated using CBE in order to understand
the neurological effects present in Gaucher’s disease (Prence
et al. 1996). Decreased GCase activity was found to have
a significant impact on neuronal growth and development,
with increased axon growth and branching in cultured neu-
ron cells being observed (Schwarz et al. 1995). These stud-
ies have begun to properly map out the actual cellular effects
observed in Gaucher’s neurons, with increases in intracellu-
lar calcium levels, increased response to glutamate and in-
creased calcium release from the ER in response to caffeine
observed in one study (Korkotian et al. 1999), and increased
CTP:phosphocholine cytidylyltransferase α activity in another
(Bodennec et al. 2002).
A mouse model for Gaucher’s has also been created by the
injection of CBE into otherwise healthy mice, leading to some
storage of GlcCer although not to the same degree as found
in human Gaucher’s patients (Kanfer et al. 1975). This mouse
model was used to evaluate the limit of GCase activity neces-
sary for an individual to be asymptomatic, found to be 12–16%
of control animals (Stephens et al. 1978). This sort of infor-
mation can be very important when evaluating newer potential
therapies of Gaucher’s disease, as it allows evaluation of the
degree of enzyme enhancement needed for the so-called chap-
erone therapy to be successful (Sawkar et al. 2006; Butters
2007). Interestingly, enzyme activities other than GCase were
also found to be altered in the CBE-treated mouse (Stephens
(Marshall et al. 2002). Although Gaucher’s mouse generated
by CBE treatment does not perfectly mimic the human disease
only led to nonviable mice (Tybulewicz et al. 1992; Liu et al.
1998). Much more complicated systems for generating a mouse
model have been examined with only partial success (Beutler
et al. 2002; Marshall et al. 2002; Mizukami et al. 2002; Xu et al.
2003; Sun et al. 2005; Sinclair et al. 2007), and it is only very
recently that a viable mouse model for Gaucher’s disease has
been produced (Enquist et al. 2007; Sinclair et al. 2007). Thus
molecule activity in an animal to mimic a disease state, when
the generation of a genetic model has proven problematic or
Cyclophellitol: mechanistic and biological applications
Cyclophellitol (10c) is another example of the parallel develop-
ment between the discovery of bioactive natural products and
rational design of MBIs for glycosidases. Although CBE has
proven to be a useful inactivator, its inherent symmetry allows
it to inactivate both α-glucosidases and β-glucosidases as dis-
cussed above. Thus, it had been proposed that a CBE derivative
that possessed a hydroxymethyl group analogous to the C5
hydroxymethyl group in glucose would be a more selective
and potentially also a more potent inactivator of β-glucosidases
(Caron and Withers 1989). Indeed, before any report detailing
the total synthesis of cyclophellitol was published, the com-
pound was isolated from a mushroom, Phellinus sp., and shown
to be a β-glucosidase inhibitor (Atsumi, Iinuma 1990; Atsumi,
Umezawa 1990). It was subsequently demonstrated to be a spe-
cific, active site-directed inactivator of two β-glucosidases, Abg
and almond (Withers and Umezawa 1991). Surprisingly, it was
nucleophile was solved, the enzyme being the β-glucosidase
from Thermotoga maritima (Gloster, Madsen 2007).
A diastereomer of cyclophellitol, 1,6-epi-cyclophellitol
(10d), has also been synthesized and tested as an inactivator
of α-glucosidases (Atsumi, Nosaka, Ochi 1993; Tai et al. 1995).
Compound 10d was found to be a potent irreversible inactivator
of yeast α-glucosidase and also of the α-mannosidase from jack
beans (Tai et al. 1995). 1,6-Epi-cyclophellitol was shown to be
an inhibitor of experimental metastasis, whereas cyclophellitol
itself was not shown to be active under the same conditions.
This definitively shows that one of the enzyme activities im-
portant for metastasis is that of an α-glucosidase, and not a
β-glucosidase. This distinction was suspected but unproven us-
ing noncovalent inhibitors such as castanospermine, since they
demonstrated some inhibitory activity against both retaining α-
and β-glucosidases (Hadwigerfangmeier et al. 1989). The use
by guest on June 7, 2013
Irreversible labeling agents of glycosidases
of a specific MBI allows the unequivocal ablation of a single
enzyme activity, which is often challenging in a complex en-
Cyclophellitol has also been used, in the same manner as
CBE, as a small-molecule inhibitor of mammalian GCase in
vivo to induce a Gaucher-like state in both cell culture and an
animal model (Atsumi S, Iinuma H 1990; Atsumi et al. 1992;
Atsumi, Nosaka, Iinuma 1993). Cyclophellitol was shown to be
The administration of cyclophellitol to mice also caused an
increase in the level of glucosylsphingosine detected in vari-
ous organs, leading to the suggestion that this cytotoxic sub-
stance may play a role in the pathology of Gaucher’s dis-
ease (Atsumi, Nosaka, Iinuma 1993). However, other than
these reports, and one other report on the use of cyclophel-
litol to probe the structural and dynamic aspects of a β-
glycosidase by tryptophan emission studies (Bismuto et al.
1999), cyclophellitol (10c) and 1,6-epi-cyclophellitol (10d)
have not seen use in the wide variety of applications of their
Activated fluorinated glycosides
and 2-deoxy-2,2-difluoro (12c) glycosides represent the most
specific class of MBI known for glycosidase activity. Inacti-
vation by these species invariably derives from labeling the
enzymatic nucleophile of a retaining glycosidase; there have
been no examples to date of any other residue being labeled
by this class of reagent (Mosi and Withers 2002; Wicki et al.
2002). The activated 2-deoxy-2-fluoro glycosides (12a) were
the first to be introduced and were found to be specific in-
activators of a variety of retaining β-glycosidases (Withers
et al. 1987, 1988). The 2-deoxy-2-fluoro substitution leads to
a destabilization of both the glycosylation and deglycosylation
transition states in the retaining glycosidase catalytic mech-
anism, thereby slowing both the formation of the glycosyl-
enzyme intermediate and its hydrolysis. This destabilization
of the oxocarbenium ion-like transition state arises both from
inductive effects of fluorine and from the removal of hydrogen-
bonding interactions ordinarily formed with the OH-2 on the
pyranose ring. The addition of an activated leaving group (of-
ten a dinitrophenolate, or fluoride) accelerates the glycosyla-
tion step, leading to the accumulation of the covalent glycosyl-
enzyme intermediate (Figure 13). This species formed can be
Fig. 12. Structures of activated (A) 2-deoxy-2-fluoro glycosides; (B) 5-fluoro
glycosyl fluorides and (C) trinitrophenyl 2-deoxy-2,2-difluoro glucoside. R =
F, dinitrophenol; TNP = trinitrophenyl.
Fig. 13. Mechanism of inactivation of a β-glucosidase by 2-deoxy-
moderately stable, with observed lifetimes ranging from sec-
onds to months.
The activity of enzymes inactivated in this fashion can be
recovered either through hydrolysis of the covalent glycosyl-
enzyme intermediate, or by transglycosylation onto a suit-
able acceptor substrate to restore a catalytically competent
enzyme (Mosi and Withers 2002; Wicki et al. 2002). As a con-
sequence of this recovery of activity, this class of molecule
is formally better described as being a very slow enzymatic
substrate rather than a true inactivator. However, the trapped
intermediates are usually sufficiently long lived that for all
practical purposes, these reagents function as inactivators and
will generally be referred to as such in the context of this
The activated 5-fluoro glycosides (12b) operate using a very
cosides. One structural difference lies in the replacement of a
hydrogen atom by fluorine, as opposed to the replacement of
a hydroxyl group with fluorine as seen with the 2-fluoro gly-
cosides. This change would be expected to be more strongly
destabilizing to both glycosylation and deglycosylation transi-
tion states, on the basis of the larger change in electronegativity
arising from replacement of hydrogen with fluorine. However,
the replacement of an oxygen atom with fluorine in the 2-fluoro
glycosides also strongly attenuates the hydrogen-bonding inter-
actions at that position, which have generally been shown to be
very important in both the glycosylation and deglycosylation
transition states. No such loss of hydrogen-bonding interactions
against retaining β-glycosidases, the 5-fluoro glycosides tend to
have both higher glycosylation and higher deglycosylation rates
cosylation) is rate limiting; thus, the intermediate accumulates.
Kinetically, this is revealed in very low Kmvalues (if monitored
by guest on June 7, 2013
B P Rempel and S G Withers
as a substrate) or apparent very tight binding (low Kivalues)
if monitored as a “reversible inhibitor.” For a more detailed
discussion, see Mosi and Withers (2002). Interestingly, and in
contrast to what is found with the 2-deoxy-2-fluoro glycosides,
the appropriately activated 5-fluoro glycosides are capable of
inactivating retaining α-glycosidases. Kinetic, mechanistic, and
structural studies have confirmed that this inactivation is indeed
due to the accumulation of a stable 5-fluoroglycosyl-enzyme
species (Numao et al. 2003). By contrast, the 2-deoxy-2-fluoro
α-glycosides act as slow substrates with α-glycosidases since
the deglycosylation step is faster than glycosylation. The origin
clear, although it is thought to be related to the relative distribu-
tion of partial positive charge between the anomeric carbon and
the ring oxygen in the transition state of the reaction (Zechel
and Withers 2000).
A third class of fluorinated sugars, that of the activated
2-deoxy-2,2-difluoro glycosides (12c), has also been used to
trap the covalent glycosyl-enzyme intermediate in retaining α-
on the use of this class of inactivator (Braun et al. 1995; Hart
et al. 2000; Zhang et al. 2008).
Activated fluorinated glycosides: mechanistic and
The use of activated fluorinated glycosides to identify and label
and Withers 2002; Wicki et al. 2002) and will not be covered
in detail here. In brief, the current strategy typically employed
to identify the catalytic nucleophile involves inactivation of the
target glycosidase followed by proteolysis, peptide localization,
and sequencing by HPLC/MS using collision-induced fragmen-
tation to identify the labeled residue. Important examples of
the use of activated 2-deoxy-2-fluoro glycosides in correcting
the identities of the enzymatic nucleophiles in E. coli lacZ β-
galactosidase (Gebler et al. 1992), GCase (Miao et al. 1994),
and almond β-glucosidase (He and Withers 1997) through the
use of the appropriately configured glycoside have been pub-
is in the identification, using an activated 5-fluoro-glucosyl flu-
oride, of a novel α-glucosidase from an acidophilic archaeon
in which the catalytic nucleophile was shown to be a threo-
nine rather than the more typically seen glutamate or aspartate
(Ferrer et al. 2005). This use of a different nucleophile is an
interesting example of the evolutionary adaptation presumably
necessary for the enzyme to function under the harsh condi-
tions in which the organism survives. Similarly, the catalytic
nucleophile in sialidase enzymes responsible for cleavage of
anionic sialic acid residues has been trapped and identified as
a tyrosine (Watts et al. 2003; Watts and Withers 2004). A fur-
ther example of the use of a 2-deoxy-2-fluoro glycoside was
in trapping the covalent glycosyl-enzyme intermediate of my-
rosinase (Cottaz et al. 1996). Interestingly, the aglycone in the
2-deoxy-2-fluoroglucotropaeolin is the same as the aglycone in
the natural substrate, sinigrin (McCarter et al. 1997). In this
case the natural-leaving group is relatively reactive, and indeed
the enzyme has evolved without a general acid/base catalytic
residue, and a bound ascorbate functions as the base catalyst
sugar-leaving group was shown to be sufficient to render the
2-fluoroglycoside useful in trapping of the intermediate. Nec-
essary conditions for this behavior were evaluated in that paper
(McCarter et al. 1997).
Another important application for the activated 2-deoxy-2-
fluoro and 5-fluoro glycosides is in the study of the trapped
glycosyl-enzyme intermediates by X-ray crystallography and
class of enzymes. Many examples, in particular X-ray crystallo-
only a few notable examples will be mentioned here. One of the
most significant examples was the crystallization of the cova-
lently bound glycosyl-enzyme intermediate of hen egg-white
lysozyme using 2-acetamido-2-deoxy-β-D-glucopyranosyl-
(1→4)-2-deoxy-2-fluoro-β-D-glucopyranosyl fluoride and a
removed by site-directed mutagenesis (Vocadlo et al. 2001).
This report was key in establishing a new paradigm in the un-
derstanding of glycosidase mechanisms, refuting the ion-pair
intermediate mechanism proposed by Phillips (1967) and sup-
porting the double-displacement mechanism first proposed by
Koshland (1953). Activated 2-deoxy-2-fluoro and 5-fluoro gly-
cosides have also been used in conjunction with X-ray crys-
tallography to gain mechanistic insights into the conformations
of the sugar ring during the course of enzymatic catalysis. X-
ray crystallographic structures have been solved for the cova-
lent glycosyl-enzyme intermediate in both α-retaining (Numao
et al. 2003; Lovering et al. 2005) and β-retaining (Ducros et al.
2002; Davies et al. 2003) glycosidases using activated 5-fluoro
and 2-deoxy-2-fluoro glycosides, respectively. These types of
studies of the structures of the resting enzyme, Michaelis, inter-
mediate and product complexes have allowed a mapping of the
catalysis and have led tohypotheses regarding the conformation
of the sugar ring at the enzymatic transition state. Mechanistic
studies such as these may permit the rational design of tighter
binding inhibitors that are more selective for one class of en-
zymes over another, based on knowledge of the transition state
(Gloster, Meloncelli 2007).
NMR spectroscopy has been applied to studying mechanistic
2-fluoro glycosides have proven to be useful tools in these types
of studies. One of the earliest examples was in demonstrating
the stereochemistry of the covalent glycosyl-enzyme interme-
diate formed upon reaction of Abg with an activated 2-deoxy-
2-fluoro glucoside using19F NMR spectroscopy (Withers and
Street 1988). In another case, the covalent glycosyl-enzyme in-
termediate of the xylanase from Cellulomonas fimi was trapped
and studied by NMR spectroscopy (Poon et al. 2007). It was
found that flexible portions of the protein became more ordered
upon inactivation, and the protein was also observed to be much
more stable upon inactivation by 2,4-dinitrophenyl 2-deoxy-2-
fluoro-cellobioside. The pKavalues of the acid/base and nu-
cleophilic catalytic residues of a retaining xylanase, another
important aspect of the catalytic machinery, were probed by pH
titration of both the free and inactivated enzyme that had been
site-specifically13C labeled on the two active site carboxylic
acids (McIntosh et al. 1996).13C-NMR spectra of each were
recorded at different pH values, and titration curves (chemical
shift versus pH) were constructed. The study revealed how the
pKavalue for the acid/base catalytic residue “cycles” to suit its
by guest on June 7, 2013
Irreversible labeling agents of glycosidases
role at each step through the course of the reaction in response
to the local electrostatic environment.
Activated fluorinated glycosides: biochemical applications
for their cognate glycosidase has been exploited in a similar
manner to that for which CBE was used in the characterization
cytosolic pyridoxine-β-glucosidase activity in mammalian cells
relied on both CBE and 2-deoxy-2-fluoro-β-glucopyranosyl flu-
oride to selectively ablate the GCase and broad-specificity cy-
tosolic β-glucosidase activities, respectively, during the purifi-
cation and characterization of this novel enzyme (McMahon et
of glycosidases came in the characterization of the two active
sites present on mammalian intestinal lactase phlorizin hydro-
lase enzyme (Arribas et al. 2000; Day et al. 2000; Mackey et al.
2002). In these reports, activated 2-deoxy-2-fluoro glycosides
were used to selectively inactivate either the lactase or phlorizin
hydrolysis activities. This ability to selectively knockout one
enzymatic activity over the other allows full kinetic character-
ization of only the one active site. This was previously found
to be challenging as the two active sites showed some substrate
cross-reactivity. The activated 2-deoxy-2-fluoro glycosides also
helped localize the two active sites onto different regions of the
polypeptide chain following labeling and sequencing (Arribas
et al. 2000), which helped clarify earlier conflicting reports re-
garding the location of the two active sites (Wacker et al. 1992;
Zecca et al. 1998). Activated 2-deoxy-2-fluoro glycosides may
also prove useful in live animal studies, as 2-deoxy-2-fluoro-β-
D-glucosyl fluoride has already been demonstrated to get into
all organs, including the brain, and to selectively label the β-
glucosidases in a rat model (McCarter et al. 1994).
The ability to trap a covalent glycosyl-enzyme intermedi-
ate using an activated 2-deoxy-2-fluoro glycoside has also been
enzyme intermediate formed with a fluorinated sugar can be
accelerated by the addition of another acceptor molecule (be-
sides water) which has someaffinity for the enzyme’s aglycone-
binding site and binds productively (Figure 14) (Withers et al.
1987). This behavior has been used to study the aglycone site
interactions in a number of different enzymes (Blanchard and
Withers 2001; Hommalai et al. 2005) and could be applicable
to studying the aglycone-binding site when nothing is known
of an enzyme’s specificity. Indeed the approach has been de-
veloped into a high-throughput format for the rapid screening
of aglycone specificity and represents the only realistic way of
uncovering such information.
Activated 2-deoxy-2-fluoro and 5-fluoro glycosides have
also been the basis for new activity-based proteomic profil-
ing (ABPP) probes when conjugated to a reporter group such
as biotin through a covalent linker (Vocadlo and Bertozzi
2004; Hekmat et al. 2005; Williams et al. 2006; Stubbs
et al. 2008). While this technology is still in the ear-
liest stages of development, the ability to selectively la-
bel the desired class of retaining β-glycosidase (either β-
galactosidase (Vocadlo and Bertozzi 2004), xylanase (Hekmat
et al. 2005; Williams et al. 2006), or β-glucosaminidase (Stubbs
et al. 2008)) in a complex protein mixture has already been
Fig. 14. Turnover of covalent glycosyl-enzyme intermediate by (A) hydrolysis
or (B) transglycosylation.
demonstrated. In one instance, active site peptide “fingerprint-
ing” of the labeled peptides derived from a biologically relevant
protein mixture led to the discovery of a new enzyme not previ-
et al. 2005) while the use of fluorescent tags of different colours
on inactivators of xylanases and cellulases allowed facile vi-
sual inspection of such enzymes produced in Cellulomonas fimi
grown upon different substrates (Hekmat et al. 2007). A novel
β-glucosaminidase thought to be involved in antibiotic resis-
tance has also been confirmed to exist in the pathological or-
ganism Pseudomonas aeruginosa by ABPP methods using an
activated 5-fluoro glycoside (Stubbs et al. 2008). Most recently,
by guest on June 7, 2013
B P Rempel and S G Withers
isotopically encoded versions of these reagents have been em-
ployed for quantitative proteomic analysis of glycosidase ac-
tivities on a proteome-wide basis (Hekmat and Withers 2008).
This topic has recently been elaborated on elsewhere (Stubbs
and Vocadlo 2006), so this review will not focus on the specific
logically relevant applications will arise in the near future from
such technologies, and this represents one of the many very im-
portant applications that covalent inactivators of glycosidases
may find in the future.
It can be seen that no one structural feature makes for an ideal
general covalent inactivator against all glycosidases. The ma-
jority of the compounds reviewed above are only active against
retaining glycosidases; no general reagent for the efficient co-
valent modification or inactivation of inverting glycosidases has
been described to date. Many of the MBIs described above are
restricted to, or show a preference for, inactivation of retaining
to be a more challenging class of enzyme to covalently modify,
and so better strategies for trapping this family of enzymes may
still need to be developed.
To date, covalent inactivators of glycosidases have found use
for studying a variety of processes in biochemistry. CBE has
been widely used in the study of the mammalian Gcase en-
zyme in the further understanding of Gaucher’s disease. Other
cosidases represent useful potential small molecule probes that
can be introduced to perturb biological systems in a known and
defined way, to help understand the systems under study more
fully. Also, covalent glycosidase inactivators represent poten-
tially very powerful tools in proteomics and ABPP that are only
beginning to be explored and recognized. These types of probes
maydevelop intousefultoolsforsystemsbiology andhelpusto
understand differential levels of protein expression in response
to external stimuli, questions that are only beginning to be
Conflict of interest statement
Neither author has any conflict of interests.
Abg, Agrobacterium sp. β-glucosidase; ABPP, activity-
based proteomic profiling; CBE, conduritol B-epoxide; CTP,
cytidyltriphosphate; GCase, glucocerebrosidase; GlcCer, β-
Glucosyl ceramide; MBI, mechanism-based inactivator; PA,
Arribas JCD, Herrero AG, Martin-Lomas M, Canada FJ, He SM, Withers SG.
2000. Differential mechanism-based labeling and unequivocal activity as-
J Biochem. 267:6996–7005.
Asano N. 2003a. Glycosidase inhibitors: Update and perspectives on practical
use. Glycobiology. 13:93R–104R.
Asano N. 2003b. Naturally occurring iminosugars and related compounds:
Structure, distribution, and biological activity. Curr Top Med Chem.
Asano N, Nash RJ, Molyneux RJ, Fleet GWJ. 2000. Sugar-mimic glycosidase
inhibitors: Natural occurrence, biological activity and prospects for thera-
peutic application. Tetrahedron Asymmetry. 11:1645–1680.
Atsumi S, Iinuma H, Nosaka C, Umezawa K. 1990. Biological-activities of
cyclophellitol. J Antibiot. 43:1579–1585.
Atsumi S, Nosaka C, Iinuma H, Umezawa K. 1992. Inhibition of glucocere-
brosidase and induction of neural abnormality by cyclophellitol in mice.
Arch Biochem Biophys. 297:362–367.
Atsumi S, Nosaka C, Iinuma H, Umezawa K. 1993. Accumulation of tissue glu-
inhibitor cyclophellitol. Arch Biochem Biophys. 304:302–304.
Atsumi S, Nosaka C, Ochi Y, Iinuma H, Umezawa K. 1993. Inhibition of exper-
imental metastasis by an alpha-glucosidase inhibitor, 1,6-epi-cyclophellitol.
Cancer Res. 53:4896–4899.
Atsumi S, Umezawa K, Iinuma H, Naganawa H, Nakamura H, Iitaka Y,
Takeuchi T. 1990. Production, isolation and structure determination of a
novel beta-glucosidase inhibitor, cyclophellitol, from Phellinus sp. J An-
Bemiller JN, Gilson RJ, Myers RW, Santoro MM. 1993. Suicide-substrate
inactivation of beta-galactosidase by diazomethyl beta-D-galactopyranosyl
ketone. Carbohydr Res. 250:101–112.
Beutler E, West C, Torbett BE, Deguchi H. 2002. A chimeric mouse model of
Gaucher disease. Mol Med. 8:247–250.
Bismuto E, Nucci R, Rossi M, Irace G. 1999. Structural and dynamic aspects of
beta-glycosidase from mesophilic and thermophilic bacteria by multitryp-
tophanyl emission decay studies. Proteins Struct Funct Genet. 35:163–172.
BlackTS, KissL, Tull D,Withers
glycopyranosylamines as affinity labels for a beta-glucosidase and a cel-
lulase. Carbohydr Res. 250:195–202.
Blanchard JE, Withers SG. 2001. Rapid screening of the aglycone specificity
of glycosidases: Applications to enzymatic synthesis of oligosaccharides.
Chem Biol. 8:627–633.
Bodennec J, Pelled D, Riebeling C, Trajkovic S, Futerman AH. 2002. Phos-
phatidylcholine synthesis is elevated in neuronal models of Gaucher disease
due to direct activation of CTP: Phosphocholine cytidylyltransferase by
glucosylceramide. Faseb J. 16:1814-+.
Bolton JL, Turnipseed SB, Thompson JA. 1997. Influence of quinone methide
reactivity on the alkylation of thiol and amino groups in proteins: Studies
utilizing amino acid and peptide models. Chem Biol Interact. 107:185–200.
HS, Wennekes T, Aerts J. 2007. Identification of the non-lysosomal gluco-
sylceramidase as beta-glucosidase 2. J Biol Chem. 282:1305–1312.
Braun C, Brayer GD, Withers SG. 1995. Mechanism-based inhibition of
yeast alpha-glucosidase and human pancreatic alpha-amylase by a new
class of inhibitors—2-Deoxy-2,2-difluoro-alpha-glycosides. J Biol Chem.
Braun H, Legler G, Deshusses J, Semenza G. 1977. Stereospecific ring-opening
of conduritol-B-epoxide by an active-site aspartate residue of sucrase-
isomaltase. Biochim Biophys Acta. 483:135–140.
Briggs JC, Haines AH, Taylor RJK. 1995. (Halogenomethyl)phenyl alpha-
D-glucopyranosides as enzyme-activated irreversible inhibitors of yeast
alpha-glucosidase and potential anti-HIV agents. J Chem Soc Perkin Trans.
Brumshtein B, Greenblatt HM, Butters TD, Shaaltiel Y, Aviezer D, Sil-
man I, Futerman AH, Sussman JL. 2007. Crystal structures of com-
plexes of N-butyl- and N-nonyl-deoxynojirimycin bound to acid beta-
glucosidase—Insights into the mechanism of chemical chaperone action
in Gaucher disease. J Biol Chem. 282:29052–29058.
the defective enzyme in Gaucher disease. Acta Crystallogr D Biol Crystal-
Burmeister WP, Cottaz S, Rollin P, Vasella A, Henrissat B. 2000. High reso-
lution x-ray crystallography shows that ascorbate is a cofactor for myrosi-
nase and substitutes for the function of the catalytic base. J Biol Chem.
Butters TD. 2007. Gaucher disease. Curr Opin Chem Biol. 11:412–418.
in lysosomal storage disorders. Curr Top Med Chem. 3:561–574.
Caron G, Withers SG. 1989. Conduritol aziridine—A new mechanism-based
glucosidase inactivator. Biochem Biophys Res Commun. 163:495–499.
by guest on June 7, 2013
Irreversible labeling agents of glycosidases
Chir J, Withers S, Wan CF, Li YK. 2002. Identification of the two essential
groups in the family 3 beta-glucosidase from flavobacterium meningosep-
ticum by labelling and tandem mass spectrometric analysis. Biochem J.
Clarke AJ, Strating H. 1989. Affinity labeling of schizophyllum-commune
cellulase with [1-H-3]-4,5-epoxypentyl beta-cellobioside—Synthesis of in-
hibitor and stoichiometry of interaction. Carbohydr Res. 188:245–250.
Clausen TP, Keller JW, Reichardt PB. 1990. Aglycone fragmentation ac-
companies beta-glucosidase catalyzed-hydrolysis of salicortin, a naturally-
occurring phenol glycoside. Tetrahedron Lett. 31:4537–4538.
Cottaz S, Henrissat B, Driguez H. 1996. Mechanism-based inhibition and
stereochemistry of glucosinolate hydrolysis by myrosinase. Biochemistry.
Daniels LB, Glew RH, Radin NS, Vunnam RR. 1980. A revised fluorometric
assay for Gauchers-disease using conduritol-beta-epoxide with liver as the
source of beta-glucosidase. Clin Chim Acta. 106:155–163.
Exp Cell Res. 168:463–474.
Davies GJ, Ducros VMA, Varrot A, Zechel DL. 2003. Mapping the conforma-
tional itinerary of beta-glycosidases by x-ray crystallography. Biochem Soc
Davies GJ, Gloster TM, Henrissat B. 2005. Recent structural insights into the
expanding world of carbohydrate-active enzymes. Curr Opin Struct Biol.
Day AJ, Canada FJ, Diaz JC, Kroon PA, McLauchlan R, Faulds CB, Plumb
GW, Morgan MRA, Williamson G. 2000. Dietary flavonoid and isoflavone
glycosides are hydrolysed by the lactase site of lactase phlorizin hydrolase.
Febs Lett. 468:166–170.
de Melo EB, Gomes AD, Carvalho I. 2006. Alpha- and beta-glucosidase
inhibitors: Chemical structure and biological activity. Tetrahedron.
Deganuto M, Pittis MG, Pines A, Dominissini S, Kelley MR, Garcia R, Quadri-
disease fibroblasts and impairment of adaptive response against oxidative
stress. J Cell Physiol. 212:223–235.
Dinur T, Osiecki KM, Legler G, Gatt S, Desnick RJ, Grabowski GA. 1986. Hu-
containing the catalytic site. Proc Natl Acad Sci USA. 83:1660–1664.
Synthesis of sodium-salt of ortho-(difluoromethyl)phenyl-alpha-ketoside
of N-acetylneuraminic acid—A mechanism-based inhibitor of clostridium-
perfringens neuraminidase. Bioorg Med Chem Lett. 2:1361–1366.
SG, Davies GJ. 2002. Substrate distortion by a beta-mannanase: Snapshots
of the Michaelis and covalent-intermediate complexes suggest a B-2,B-5
conformation for the transition state. Angew Chem Int Ed. 41:2824-&.
Dvir H, Harel M, McCarthy AA, Toker L, Silman I, Futerman AH, Sussman JL.
in Gaucher disease. Embo Rep. 4:704–709.
Gaucher disease. Proc Natl Acad Sci USA. 104:17483–17488.
Febbraio F, Barone R, Dauria S, Rossi M, Nucci R, Piccialli G, DeNapoli
L, Orru S, Pucci P. 1997. Identification of the active site nucleophile in
the thermostable beta-glycosidase from the archaeon sulfolobus solfataricus
expressed in Escherichia coli. Biochemistry. 36:3068–3075.
man P, Kaplan F. 1997. Identification of candidate active site residues in
lysosomal beta-hexosaminidase A. J Biol Chem. 272:814–820.
Ferrer M, Golyshina OV, Plou FJ, Timmis KN, Golyshin PN. 2005. A novel
alpha-glucosidase from the acidophilic archaeon ferroplasma acidiphilum
strain Y with high transglycosylation activity and an unusual catalytic nu-
cleophile. Biochem J. 391:269–276.
Fersht A. 1999. Structure and Mechanism in Protein Science. New York: Free-
man, pp. 273–288.
Fowler AV, Zabin I, Sinnott ML, Smith PJ. 1978. Methionine 500, site of
covalent attachment of an active site-directed reagent of beta-galactosidase.
J Biol Chem. 253:5283–5285.
Gebler JC, Aebersold R, Withers SG. 1992. Glu-537, not Glu-461, is the nu-
cleophile in the active-site of (Lac-Z) beta-galactosidase from Escherichia
coli. J Biol Chem. 267:11126–11130.
Gloster TM, Madsen R, Davies GJ. 2007. Structural basis for cyclophellitol
inhibition of a beta-glucosidase. Org Biomol Chem. 5:444–446.
cosidase inhibition: An assessment of the binding of 18 putative transition-
state mimics. J Am Chem Soc. 129:2345–2354.
Hadwigerfangmeier A, Niemann H, Tamura T. 1989. Inhibition of glycosy-
lation processing alters the growth-parameters of cells transformed by the
oncogene of simian sarcoma-virus. Arch Virol. 104:339–345.
Halazy S, Danzin C, Ehrhard A, Gerhart F. 1989. 1,1-Difluoroalkyl
glucosides—A new class of enzyme-activated irreversible inhibitors of
alpha-glucosidases. J Am Chem Soc. 111:3484–3485.
Hart DO, He SM, Chany CJ, Withers SG, Sims PFG, Sinnott ML, Brumer
H. 2000. Identification of Asp-130 as the catalytic nucleophile in the main
alpha-galactosidase from phanerochaete chrysosporium, a family 27 glyco-
syl hydrolase. Biochemistry. 39:9826–9836.
Havukainen R, Torronen A, Laitinen T, Rouvinen J. 1996. Covalent binding of
three epoxyalkyl xylosides to the active site of endo-1,4-xylanase II from
Trichoderma reesei. Biochemistry. 35:9617–9624.
Hays WS, Wheeler DE, Eghtesad B, Glew RH, Johnston DE. 1998. Expres-
sion of cytosolic beta-glucosidase in guinea pig liver cells. Hepatology.
He SM, Withers SG. 1997. Assignment of sweet almond beta-glucosidase as a
family 1 glycosidase and identification of its active site nucleophile. +98 J
Biol Chem. 272:24864–24867.
Hekmat O, Florizone C, Kim YW, Eltis LD, Warren RAJ, Withers SG. 2007.
Specificity fingerprinting of retaining beta-1,4-glycanases in the Cellu-
lomonas fimi secretome using two fluorescent mechanism-based probes.
Hekmat O, Kim YW, Williams SJ, He SM, Withers SG. 2005. Active-site
peptide “fingerprinting” of glycosidases in complex mixtures by mass
spectrometry—Discovery of a novel retaining beta-1,4-glycanase in Cel-
lulomonas fimi. J Biol Chem. 280:35126–35135.
Hekmat O, Withers S. 2008. J Proteome Res.
Henrissat B. 1991.A classification of glycosyl hydrolases based on amino-acid-
sequence similarities. Biochem J. 280:309–316.
Henrissat B,Bairoch A. 1993. New families in the classification of gly-
cosyl hydrolases based on amino-acid-sequence similarities. Biochem J.
Hermans MMP, Kroos MA, Vanbeeumen J, Oostra BA, Reuser AJJ. 1991.
J Biol Chem. 266:13507–13512.
Herrchen M, Legler G. 1984. Identification of an essential carboxylate group
at the active-site of lacZ beta-galactosidase from Escherichia coli. Eur
J Biochem. 138:527–531.
Hinou H, Kurogochi M, Shimizu H, Nishimura SI. 2005. Characterization of
vibrio cholerae neuraminidase by a novel mechanism-based fluorescent la-
beling reagent. Biochemistry. 44:11669–11675.
glutamic-acid 105 at the active-site of bacillus-amyloliquefaciens 1,3-1,4-
beta-D-glucan 4-glucanohydrolase using epoxide-based inhibitors. J Biol
Hoj PB, Rodriguez EB, Iser JR, Stick RV, Stone BA. 1991. Active site-directed
inhibition by optically pure epoxyalkyl cellobiosides reveals differences in
active-site geometry of 2 1,3-1,4-beta-D-glucan 4-glucanohydrolases—The
importance of epoxide stereochemistry for enzyme inactivation. J Biol
Holleran WM, Takagi Y, Uchida Y. 2006. Epidermal sphingolipids:
Hommalai G, Chaiyen P, Svasti J. 2005. Studies on the transglucosylation re-
actions of cassava and Thai rosewood beta-glucosidases using 2-deoxy-
2-fluoro-glycosyl-enzyme intermediates. Arch Biochem Biophys. 442:
of the Glu and Asp residues in the active site of human beta-hexosaminidase
B. Biochemistry. 40:2201–2209.
Howard S, Withers SG. 1998a. Bromoketone C-glycosides, a new class of
beta-glucanase inactivators. J Am Chem Soc. 120:10326–10331.
Howard S, Withers SG. 1998b. Labeling and identification of the postulated
acid/base catalyst in the alpha-glucosidase from Saccharomyces cerevisiae
using a novel bromoketone C-glycoside. Biochemistry. 37:3858–3864.
Hrmova M, MacGregor EA, Biely P, Stewart RJ, Fincher GB. 1998. Substrate
binding and catalytic mechanism of a barley beta-D-glucosidase/(1,4)-beta-
D-glucan exohydrolase. J Biol Chem. 273:11134–11143.
Ichikawa M, Ichikawa Y. 2001. A mechanism-based affinity-labeling agent for
possible use in isolating N-acetylglucosaminidase. Bioorg Med Chem Lett.
by guest on June 7, 2013
B P Rempel and S G Withers
Iwanami S, Matsui H, Kimura A, Ito H, Mori H, Honma M, Chiba S. 1995.
alpha-glucosidase. Biosci Biotechnol Biochem. 59:459–463.
Jacobson RH, Zhang XJ, Dubose RF, Matthews BW. 1994. Three-dimensional
Jager S, Kiss L. 2005. Investigation of the active site of the extracellular beta-
D-glucosidase from aspergillus carbonarius. World J Microbiol Biotechnol.
Juers DH, Heightman TD, Vasella A, McCarter JD, Mackenzie L, Withers SG,
Matthews BW. 2001. A structural view of the action of Escherichia coli
(lacZ) beta-galactosidase. Biochemistry. 40:14781–14794.
form reveals multiple metal-binding sites and provides a structural basis for
alpha-complementation. Protein Sci. 9:1685–1699.
Kacher Y, Golan A, Pewzner-Jung Y, Futerman AH. 2007. Changes in
macrophage morphology in a Gaucher disease model are dependent on
CTP: Phosphocholine cytidylyltransferase alpha. Blood Cells Mol Dis.
Kanfer JN, Legler G, Sullivan J, Raghavan SS, Mumford RA. 1975. Gaucher
mouse. Biochem Biophys Res Commun. 67:85–90.
Kapferer P, Birault V, Poisson JF, Vasella A. 2003. Synthesis and evaluation
as glycosidase inhibitors of carbasugar-derived spirodiaziridines, spirodi-
azirines, and spiroaziridines. Helv Chim Acta. 86:2210–2227.
Keitel T, Simon O, Borriss R, Heinemann U. 1993. Molecular and active-site
structure of a bacillus 1,3-1,4-beta-glucanase. Proc Natl Acad Sci USA.
Keresztessy Z, Kiss L, Hughes MA. 1994. Investigation of the active-site of
the cyanogenic beta-D-glucosidase (Linamarase) from manihot-esculenta
crantz (Cassava): 2. Identification of Glu-198 as an active-site carboxy-
late group with acid catalytic function. Arch Biochem Biophys. 315:323–
Kimura A, Takata M, Fukushi Y, Mori H, Matsui H, Chiba S. 1997. A catalytic
amino acid and primary structure of active site in aspergillus niger alpha-
glucosidase. Biosci Biotechnol Biochem. 61:1091–1098.
Kiss T, Erdei A, Kiss L. 2002. Investigation of the active site of the extracel-
lular ss-D-xylosidase from aspergillus carbonarius. Arch Biochem Biophys.
Kitz RJ, Ginsburg S, Wilson IB. 1965. Activity-structure relationships in reacti-
vation of diethylphosphoryl acetylcholinesterase by phenyl-1-methyl pyri-
dinium ketoximes. Biochem Pharmacol. 14:1471-&.
Knapp S, Vocadlo D, Gao ZN, Kirk B, Lou JP, Withers SG. 1996. NAG-
thiazoline, an N-acetyl-beta-hexosaminidase inhibitor that implicates ac-
etamido participation. J Am Chem Soc. 118:6804–6805.
of cyclodextrin glycosyltransferase from bacillus-circulans strain-251 with
natural substrates and products. J Biol Chem. 270:29256–29264.
KnowlesJR. 1976. Intrinsic Pka-values
enzymes—Improper deductions from pH-dependence of steady-state
parameters. CRC Crit Rev Biochem. 4:165–173.
Kopitz J, Sinz K, Brossmer R, Cantz M. 1997. Partial characterization and
enrichment of a membrane-bound sialidase specific for gangliosides from
human brain tissue. Eur J Biochem. 248:527–534.
Korkotian E, Schwarz A, Pelled D, Schwarzmann G, Segal M, Futerman AH.
1999. Elevation of intracellular glucosylceramide levels results in an in-
crease in endoplasmic reticulum density and in functional calcium stores in
cultured neurons. J Biol Chem. 274:21673–21678.
Koshland DE. 1953. Stereochemistry and mechanism of enzymatic reactions.
Biol Rev Camb Philos Soc. 28:416–436.
Kuhn CS,Lehmann J. 1987. 3,7-Anhydro-2-azi-1,2-dideoxy-D-glycero-L-
manno-octitol, a photoaffinity label for beta-D-galactosidase from Es-
cherichia coli. Carbohydr Res. 160:C6–C8.
of Escherichia coli beta-D-galactosidase by photoaffinity-labeling. Carbo-
hydr Res. 232:227–233.
Kuhn CS, Lehmann JC, Sandhoff K. 1992. Efficient photoaffinity-labeling
of human beta-hexosaminidase-a—Synthesis and application of 3-Azi-1-
[(2-acetamido-2-deoxy-1-beta-D-glucopyranosyl)Thio]butane and 3-Azi-1-
hydrolases. J Biol Chem. 279:44704–44712.
Laitinen T, Rouvinen J, Perakyla M. 2000. Inversion of the roles of the nucle-
inhibitor to the catalytic glutamates of endo-1,4-beta-xylanase (XYNII): A
molecular dynamics study. Protein Eng. 13:247–252.
Legler G. 1990. Glycoside hydrolases—mechanistic information from studies
with reversible and irreversible inhibitors. Adv Carbohydr Chem Biochem.
Legler G, Bause E. 1973. Epoxyalkyl oligo-(1–4)-beta-D-glucosides as active-
site-directed inhibitors of cellulases. Carbohydr Res. 28:45–52.
a from bitter almonds. Biochim Biophys Acta. 524:102–108.
Lemieux MJ, Mark BL, Cherney MM, Withers SG, Mahuran DJ, James MNG.
2006. Crystallographic structure of human beta-hexosaminidase A: Inter-
pretation of Tay-Sachs mutations and loss of G(M2) ganglioside hydrolysis.
J Mol Biol. 359:913–929.
Li YK, Chir J, Chen FY. 2001. Catalytic mechanism of a family 3 beta-
glucosidase and mutagenesis study on residue Asp-247. Biochem J.
Lieberman RL, Wustman BA, Huertas P, Powe AC, Pine CW, Khanna R,
Schlossmacher MG, Ringe D, Petsko GA. 2007. Structure of acid beta-
glucosidase with pharmacological chaperone provides insight into Gaucher
disease. Nat Chem Biol. 3:101–107.
Liessem B, Glombitza GJ, Knoll F, Lehmann J, Kellermann J, Lottspeich
F, Sandhoff K. 1995. Photoaffinity-labeling of human lysosomal beta-
hexosaminidase-B—Identification of Glu-355 at the substrate binding site.
J Biol Chem. 270:23693–23699.
Lillelund VH, Jensen HH, Liang XF, Bols M. 2002. Recent developments
of transition-state analogue glycosidase inhibitors of non-natural product
origin. Chem Rev. 102:515–553.
Liu YJ, Suzuki K, Reed JD, Grinberg A, Westphal H, Hoffmann A, Doring T,
Sandhoff K, Proia RL. 1998. Mice with type 2 and 3 Gaucher disease point
mutations generated by a single insertion mutagenesis procedure (SIMP).
Proc Natl Acad Sci USA. 95:2503–2508.
Lo LC, Chiang YL, Kuo CH, Liao HK, Chen YJ, Lin JJ. 2005. Study of the
preferred modification sites of the quinone methide intermediate resulting
Biophys Res Commun. 326:30–35.
Lo LC, Chu CY, Lin FA, Pan YR, Li YK, Lin JJ. 2006. Synthesis of activity
probes for beta-xylosidase. J Chin Chem Soc. 53:479–488.
Lo LC, Lo CHL, Janda KD, Kassel DB, Raushel FM. 1996. A versatile mech-
anism based reaction probe for the direct selection of biocatalysts. Bioorg
Med Chem Lett. 6:2117–2120.
and structural analysis of a family 31 alpha-glycosidase and its glycosyl-
enzyme intermediate. J Biol Chem. 280:2105–2115.
Lu CP, Ren CT, Lai YN, Wu SH, Wang WM, Chen JY, Lo LC. 2005. Design of
a mechanism-based probe for neuraminidase to capture influenza viruses.
Angew Chem Int Ed. 44:6888–6892.
Ly HD, Withers SG. 1999. Mutagenesis of glycosidases. Annu Rev Biochem.
Macarron R, Vanbeeumen J, Henrissat B, Delamata I, Claeyssens M. 1993.
Identification of an essential glutamate residue in the active-site of
endoglucanase-III from Trichoderma-Reesei. Febs Lett. 316:137–140.
Macauley MS, Whitworth GE, Debowski AW, Chin D, Vocadlo DJ. 2005.
O-GlcNAcase uses substrate-assisted catalysis—Kinetic analysis and de-
velopment of highly selective mechanism-inspired inhibitors. J Biol Chem.
Mackey AD, Henderson GN, Gregory JF. 2002. Enzymatic hydrolysis of
pyridoxine-5 ‘-beta-D-glucoside is catalyzed by intestinal lactase-phlorizin
hydrolase. J Biol Chem. 277:26858–26864.
Mark BL, Mahuran DJ, Cherney MM, Zhao DL, Knapp S, James MNG.
2003. Crystal structure of human beta-hexosaminidase B: Understand-
ing the molecular basis of Sandhoff and Tay-Sachs disease. J Mol Biol.
Mark BL, Vocadlo DJ, Knapp S, Triggs-Raine BL, Withers SG, James MNG.
beta-hexosaminidase. J Biol Chem. 276:10330–10337.
Marshall J, McEachern KA, Kyros JAC, Nietupski JB, Budzinski TL, Ziegler
RJ, Yew NS, Sullivan J, Scaria A, van Rooijen N, Barranger JA, Cheng
SH. 2002. Demonstration of feasibility of in vivo gene therapy for Gaucher
disease using a chemically induced mouse model. Mol Ther. 6:179–189.
Marshall PJ, Sinnott ML, Smith PJ, Widdows D. 1981. Active-site-directed
(para-nitrophenyl)triazenes. J Chem Soc Perkin Trans. 1:366–376.
by guest on June 7, 2013