Streptococcus pneumoniae endohexosaminidase D; Feasibility of using N-glycan oxazoline donors for synthetic glycosylation of a GlcNAc-asparagine acceptor

Article abstract of DOI:10.1039/b926078a

Endohexosaminidase D, a family 85 glycoside hydrolase from S. pneumoniae and the first endohexosaminidase to be discovered, is found to be capable of catalysing the glycosylation of a glycosyl amino acid bearing a GlcNAc residue using a variety of N-glycan oxazoline donors. Although enzyme-catalysed oxazoline hydrolysis is a significant competing reaction that is not countered by the addition of organic co-solvents or variation of reaction pH, a high yielding synthetic process can be achieved by the sequential addition of multiple equivalents of oxazoline donor, demonstrating the synthetic potential of this enzyme as a biocatalyst for the synthesis of defined glycoconjugates. Notably Endo-D does not appear to hydrolyse the resulting products under the conditions used. The synthetic activity displayed by Endo D implies that other, as yet untested, family GH85 enzymes may display similar synthetic potential. Furthermore since Endo D is capable of cleaving N-glycans attached to monoclonal antibodies (mAbs), and also of cleaving glycans that are core-fucosylated, the development of Endo D as a useful biocatalyst for the synthesis of important defined homogeneous complex glycoconjugates may have significant future potential, provided that the limitation of direct oxazoline hydrolysis can be surmounted.

Full text of DOI:10.1039/b926078a

View Article Online / Journal Homepage / Table of Contents for this issue
PAPER
www.rsc.org/obc | Organic & Biomolecular Chemistry
Streptococcus pneumoniae endohexosaminidase D; feasibility of using
N-glycan oxazoline donors for synthetic glycosylation of a
GlcNAc-asparagine acceptor†‡
Thomas B. Parsons,a Mitul K. Patel,a Alisdair B. Boraston,b David J. Vocadloc and Antony J. Fairbanks*d
Received 10th December 2009, Accepted 4th February 2010
First published as an Advance Article on the web 25th February 2010
DOI: 10.1039/b926078a
Endohexosaminidase D, a family 85 glycoside hydrolase from S. pneumoniae and the first
endohexosaminidase to be discovered, is found to be capable of catalysing the glycosylation of a
glycosyl amino acid bearing a GlcNAc residue using a variety of N-glycan oxazoline donors. Although
enzyme-catalysed oxazoline hydrolysis is a significant competing reaction that is not countered by the
addition of organic co-solvents or variation of reaction pH, a high yielding synthetic process can be
achieved by the sequential addition of multiple equivalents of oxazoline donor, demonstrating the
synthetic potential of this enzyme as a biocatalyst for the synthesis of defined glycoconjugates. Notably
Endo-D does not appear to hydrolyse the resulting products under the conditions used. The synthetic
activity displayed by Endo D implies that other, as yet untested, family GH85 enzymes may display
similar synthetic potential. Furthermore since Endo D is capable of cleaving N-glycans attached to
monoclonal antibodies (mAbs), and also of cleaving glycans that are core-fucosylated, the development
of Endo D as a useful biocatalyst for the synthesis of important defined homogeneous complex
glycoconjugates may have significant future potential, provided that the limitation of direct oxazoline
hydrolysis can be surmounted.
hydrolytic activity of endohexosaminidase D (Endo D), observed
in the culture fluid of Streptococcus pneumoniae and reported by
Muramatsu in 1971, was in fact the first endohexosaminidase
Introduction
Endo-b-N-acetylglucosaminidases1 (endohexosaminidases, or
activity to be disclosed.2 Subsequently, more than 15 other endo-
hexosaminidases have been identified, and several of these enzymes
have now been cloned and expressed. According to the CAZy3
enzyme classification system all of the endohexosaminidases
identified to date belong to either family 18 (GH18) or family
85 (GH85) of the superfamily of glycoside hydrolases.
ENGases) are a class of enzyme that specifically cleave the
chitobiose core [GlcNAcb(1–4)GlcNAc] of N-linked glycans
between the two N-acetyl glucosamine residues (Fig. 1). The
Endohexosaminidases, such as endohexosaminidase H (Endo
H,4 a member of family GH18) from Streptomyces plicatus
(griseus)5 have been long-standing tools used to remove the mix-
ture of heterogeneous N-glycans from glycoproteins, either to fa-
cilitate protein handling and analysis, or to allow separation of the
released oligosaccharides for glycan profiling. However, recently
there has been resurgence in interest in endohexosaminidases as
potent biocatalytic tools for the chemoenzymatic synthesis of
defined homogeneous glycoconjugates, including glycoproteins.
The natural heterogeneity of the glycan portions of glycopro-
teins produced intracellularly, compounded with the inseparability
of these mixtures of protein ‘glycoforms’, has meant that access
to pure homogeneous single glycoforms of glycoproteins has
become a major scientific objective.6 Such access would not only
facilitate more precise biological investigation into the effects
that different glycan structures have on protein properties, but is
also an important commercial goal in the field of glycoprotein
therapeutics, since these glycoproteins are currently approved
and marketed as heterogeneous mixtures that can be difficult to
reproducibly generate. Furthermore the emergence during the past
decade of monoclonal antibodies (mAbs) as probably the most im-
portant class of new biotherapeutic agent has added considerable
Fig. 1 Structure of the core region of N-glycans showing the sites
of endohexosaminidase cleavage, core fucosylation and decoration with
additional sugars that leads to the typical high mannose, hybrid and
complex structures.
aDepartment of Chemistry, Chemistry Research Laboratory, University of
Oxford, Mansfield Road, Oxford, UK OX1 3TA
bDepartment of Biochemistry and Microbiology, University of Victoria,
Victoria, British Columbia, Canada V8W 3P6
cDepartment of Chemistry, Simon Fraser University, Burnaby, British
Columbia, Canada V5A 1S6
dDepartment of Chemistry, University of Canterbury, Private Bag 4800,
Christchurch, 8140, New Zealand
† This paper is part of an Organic & Biomolecular Chemistry web theme
issue on biocatalysis.
‡ Electronic supplementary information (ESI) available: Characterisation
data for glycosylation products and details of enzyme kinetics. See DOI:
10.1039/b926078a
This journal is
The Royal Society of Chemistry 2010
Org. Biomol. Chem., 2010, 8, 1861–1869 | 1861
©

View Article Online
financial impetus to the development of technologies that permit
the production of defined glycoproteins in homogenous form.
For example, human IgG antibodies contain N-linked glycans
at the two Asn-297 residues of the dimeric Fc (constant fragment)
region, and extensive work has revealed that the precise identity
of the glycans at these sites is key for modulating antibody-
mediated responses such as antibody dependent cell-mediated
cytotoxicity (ADCC) and complement-dependent cytotoxicity
(CDC).7 Moreover, the efficacy of mAb-based therapies has also
been demonstrated to be dependent on polymorphism of the FcR
receptor of natural killer (NK) cells. Optimal tuning of mAb
glycosylation patterns to take account of such polymorphisms
represents a significant opportunity to increase the efficacy of
mAb therapies for individuals for whom current therapies fail.
Therefore it is not surprising that considerable effort has recently
been expended in the attempted optimisation of the glycosylation
patterns of mAbs. For example, the Biotech companies GlycoFi8
and GlyCart9 have focussed on cell engineering approaches
to produce expression systems that can be used to produce
mAbs with more ‘humanised’ glycosylation patterns, although
these approaches give no guarantee as to complete glycoprotein
homogeneity since they rely on cell biosynthesis.
tion of a Man3GlcNAc tetrasaccharide using the corresponding
oxazoline donor, though almost 100 equivalents of this material
were required in order to drive this reaction to completion. In a
similar vein, the potential use of Endo D as a selective synthetic
biocatalyst, using oxazolines as donors, would appear to be a
potentially fruitful line of study for several reasons. Firstly, Endo
D, in consort with a panel of other glycoside hydrolases produced
by S. pneumoniae, naturally hydrolyses complex glycans attached
to human IgGs; it may therefore prove to be a particularly suitable
biocatalyst for the remodelling of mAbs. Secondly, Endo D is
a family GH85 enzyme that is capable of hydrolysing glycans
that contain a core fucose, i.e. which are fucosylated at the
6-position of the innermost GlcNAc residue. Notably Endo A11
and Endo M10b are not capable of hydrolysing glycans that are
fucosylated in this fashion. Therefore the development of Endo
D as a synthetic biocatalyst may also allow access to defined
glycoconjugates that contain a core fucose. This report therefore
details preliminary investigations into the potential use of Endo D
to catalyse the glycosylation of acceptors bearing GlcNAc residues
using N-glycan oxazolines as activated glycosyl donors.
Results and discussion
An alternative and attractive approach that may allow access
to a wide range of glycoproteins bearing homogenous defined
glycan structures, including mAbs, is that of in vitro glycoprotein
remodelling. The remodelling approach involves the initial trim-
ming back of the oligosaccharide chains of naturally produced
heterogeneous mixtures of glycoforms to single GlcNAc residues
using endohexosaminidases; a process that is routinely performed
with enzymes such as Endo H. In a second, and considerably
more challenging step, defined oligosaccharides can then be
added to these ‘GlcNAc handles’, using enzymatic catalysis.
Two endohexosaminidases have so far shown useful synthetic
glycosylation activity in this latter respect; Endo M10 from Mucor
Hiemalis and Endo A11 from Arthrobacter protophormiae, both
of which are members of the family GH85. Early demonstrations
of the synthetic potential of these two enzymes relied upon the
use of Asn-linked oligosaccharides to act as the glycosyl donors.
Since the enzymes naturally operate hydrolytically, competitive
product hydrolysis was, not surprisingly, found to compromise
achievable product yield; transglycosylations using these ‘un-
activated donors’ were therefore quite inefficient. However the
use of carbohydrate oxazolines as donors for family GH85
endohexosaminidase-catalysed glycosylation of acceptors bearing
GlcNAc residues, first demonstrated by Shoda and co-workers,12
has since led to substantial increases in synthetic efficiency. Indeed,
subsequently both ourselves13 and Wang14 and co-workers have
published extensively on the application and further development
of both Endo A and Endo M as biocatalysts for the synthesis of a
variety of defined glycoconjugates bearing N-glycans. Recently
acquired crystallographic data for Endo A15 and its E173Q
mutant16 will greatly facilitate future enzyme engineering in the
pursuit of refining these enzymes as efficient biocatalysts.
Following the initial reports of hydrolytic activity, Endo D
was subsequently isolated from S. pneumoniae by Muramatsu,
purified,17 and its specific hydrolytic activity was also investigated
with respect to substrate structure. Much later it was cloned
and expressed,18 and truncation and mutational studies revealed
residues in the enzyme active site critical for catalytic activity.19
Sequence analysis placed Endo D in family GH85, alongside Endo
A and Endo M, the two enzymes that have already demonstrated
useful synthetic activity with oxazoline donors. Furthermore
very recently the crystal structure of the catalytic domain of
Endo D (SpGH85)20 was solved providing key structural insight
into the catalytic site and some insight into the oligosaccharide
binding motifs of the enzyme. It is envisaged that this structural
information will prove extremely valuable for fine-tuning and
optimisation of any synthetic activity displayed by this enzyme
by future protein engineering.
Endo D naturally cleaves complex N-glycans, acting in
collaboration with neuraminidase, b-galactosidase and exo-
b-N-acetylglucosaminidase enzymes produced by S. pneumo-
niae. Following the sequential action of these other enzymes,
Endo D itself cleaves either the core N-glycan pentasaccharide
(Man3GlcNAc2Asn, Fig. 1 structure shown within dotted line) or
the more extended Man5GlcNAc2Asn heptasaccharide structure,
either of which may be core fucosylated at the terminal GlcNAc.
In particular Endo D requires17b,17e,21 an exposed 2-hydroxyl group
in the a-mannose residue which is linked by an a(1–3) linkage
to the inner mannose of the core structure and so, for example,
Endo D will not cleave high mannose glycans. It may be expected
therefore that Endo D would have a strict tolerance for which
donor oxazolines may be processed. Studies to clarify the scope of
which donor oxazolines are processed by Endo D were therefore
undertaken as described below.
When considering the future potential applications of
endohexosaminidase-mediated glycosylation reactions the ther-
apeutic importance of mAbs highlights these proteins as key sub-
strates for methodological development and application. Indeed
such remodelling of a human IgG1-Fc fragment has already been
demonstrated by Wang14i who used Endo A to catalyze the addi-
Variation of reaction parameters
The tetrasaccharide oxazoline donor 1 was chosen as a model
substrate for initial studies since it corresponds to the core
1862 | Org. Biomol. Chem., 2010, 8, 1861–1869
This journal is
The Royal Society of Chemistry 2010
©

View Article Online
mannose structure pentasaccharide that is a known substrate for
the hydrolytic action of Endo D. The glycosyl amino acid 213a
was used as the model glycosyl acceptor. A truncated version of
Endo D, termed SpGH85§ comprised of amino acids 159-807 of
the complete protein and which contains the catalytic domain,
was used as the catalyst. SpGH85, which has previously been
demonstrated to posses all of the catalytic activity of the full-
length Endo D, was cloned, expressed and purified as previously
described.20 A series of glycosylation experiments were undertaken
in which reaction parameters were systematically varied. The
synthetic efficiency of each reaction catalysed by Endo D was
assessed by HPLC monitoring of the conversion of acceptor 2
into the corresponding glycosylated product 3 (Scheme 1). Time
course studies of these reactions are shown in Fig. 2.
Firstly, although a definition of unit activity for full-length Endo
D already existed,22 this assay was not readily amenable since it re-
quired access to 14C labelled acetyl Man5GlcNAc2Asn. Therefore,
an initial study focussed on determining the approximate quantity
of enzyme required to efficiently promote any glycosylation
reaction. Reactions were conducted in phosphate buffer (0.1 M)
at pH 6.5, and the total reaction volume was kept constant at
50 mL. The amount of Endo D used to catalyse the reaction was
varied between 0.0116 mg mL-1 and 1.16 mg mL-1 (corresponding
crudely to 0.155–15.5 mU mL-1)¶, as shown in Fig. 2A. The use
of 1.16 mg mL-1 of Endo D produced pentasaccharide 3 in 30%
Scheme 1 Endo D catalysed glycosylation of GlcNAc-Asn acceptor 2
with tetrasaccharide oxazoline donor 1.
yield after approximately 30 min. This yield remained constant
even after extended reaction times, implying either that, whilst the
enzyme was capable of processing the tetrasaccharide oxazoline 1
it was incapable of hydrolysing the product 3 at an appreciable
§ The terms Endo D and SpGH85 are used interchangeably after this point.
¶ This crude estimation is made on the basis of the assumptions of a similar
level of enzyme purity and Muramatsu’s 13.4 units mg-1 quantification. See
ref. 17d and 22.
Fig. 2 Time-course studies of glycosylation of acceptor 2 with oxazoline 1 catalysed by truncated Endo D (SpGH85) to produce pentasaccharide
glycosyl amino acid 3. (A) Variation of Endo D concentration; (B) effect of the addition of various organic co-solvents; (C) effect of reaction pH;
(D) increased conversion to product by multiple additions of oxazoline donor.
This journal is
The Royal Society of Chemistry 2010
Org. Biomol. Chem., 2010, 8, 1861–1869 | 1863
©

View Article Online
rate, or alternatively, that the enzyme was losing activity over
time. Decreasing the concentration of enzyme by a factor of
two (to 0.58 mg mL-1) only led to a slight decrease in both the
product yield (29%) and also the rate by which it was formed;
approximately 1 h was required to achieve this maximum yield.
A subsequent decrease in the enzyme concentration by a further
factor of five to 0.116 mg mL-1 led to a significant decrease in
the rate of glycosylation; this reaction took approximately 8 h to
produce 3 in 29% yield. Decreasing the amount of enzyme by a
further factor of ten to 0.0116 mg mL-1 led to a slight reduction in
the maximum observable yield of 3 (26%), and to a further increase
in the time required to achieve this (approximately 20 h). Since the
initial two-fold decrease in the quantity of enzyme from 1.16 to
0.58 mg mL-1 had led to minimal differences in both the rate and
yield of reaction, it was deemed most appropriate and convenient
to use a concentration of 0.58 mg mL-1 of Endo D as the standard
quantity of enzyme employed in subsequent investigations.
As mentioned above, the maximum yield of the glycosylation
process using a single addition of 3 equivalents of oxazoline donor
was a rather modest 30%. This compared somewhat unfavourably
with results obtained using either Endo A or Endo M under similar
conditions; Endo A afforded 3 in a maximum yield of 88% after
30 min, after which point product hydrolysis led to a progressive
decrease in the efficiency of the reaction, whilst Endo M afforded 3
in a maximum yield of 41% after 3 h, though again after this time,
product hydrolysis led to a progressive decrease in the efficiency
of the reaction. The observation that the enzyme was still active at
20 h, coupled with the absence of any time-dependent decrease in
product formation, strongly suggests that incomplete conversion
to product was due to competitive enzyme-catalysed hydrolysis of
the oxazoline donor 1, rather than to product hydrolysis. These
data reveal a significant difference between the activities of these
different family GH85 enzymes.
Recent investigations into the synthetic utility of Endo M and
Endo A have shown that both a variation in reaction pH and the
addition of an organic co-solvent to the reaction medium may have
a beneficial effect on the efficiency of the synthetic glycosylation
reaction.23 A brief survey of these two reaction parameters was
therefore carried out to determine whether the modest yield of
the Endo D catalysed reaction could be improved by effecting a
decrease in the rate of direct oxazoline hydrolysis relative to that
of the desired synthetic reaction.
Firstly the use of three different organic co-solvents was
investigated: DMSO, DMF and acetone (Fig. 2B). The addition of
10% DMSO led to a decrease in the rate of the reaction, but did not
alter the overall yield of 3 (28% yield after around 5 h). In contrast,
the addition of 10% DMF had a more detrimental effect on the
reaction; the reaction rate was reduced considerably such that the
maximum product yield was not achieved until after 12 h, and
the maximum observable yield was also reduced to 23%. Finally,
the addition of 10% acetone to the reaction mixture again led to
a decrease in the rate of the reaction, but actually led to a very
slight increase in the observed yield of 3 (31%, reached after 5 h),
although this minor increase is within the bounds of experimental
error. Once again, in all three of these instances no product
hydrolysis was observed during extended reaction times. However,
since no significant increase in product yield was achieved, it
appears that the addition of organic co-solvent has no particularly
beneficial effect on the yield of glycosylations catalysed by Endo D.
Variation of reaction pH was also investigated (Fig. 2C). Endo
D has been shown to have a pH optimum of 6.5 for hydrolysis,
with half maximal values of 5.3 and 8.2.17a A series of reactions
were conducted at pH 6.5, pH 7.0 and pH 8.0 using a phosphate
buffer (0.1 M). However as can be seen from Figure 3C, increasing
the reaction pH had no beneficial effect on the yield of the
glycosylation reaction; rather an increase in pH merely caused
a slight decrease in the rate of the reaction.
It was concluded that unfortunately oxazoline hydrolysis cat-
alyzed by Endo D (SpGH85) competes significantly with glyco-
sylation and is unavoidable under the conditions tested here. To
investigate if this limit could be overcome and push the conversion
of acceptor 2 to product 3 towards completion the addition of
further equivalents of the oxazoline donor 1 after particular time
intervals was attempted. This multiple addition procedure worked
satisfactorily as shown in Fig. 2D. Herein, the addition of 3 further
equivalents of oxazoline donor 1 every hour for 4 h (i.e. a total
of 15 equivalents of donor) allowed production of 3 in 87% yield.
Although such a procedure was rather inefficient with respect to
the quantity of oxazoline donor required, it did demonstrate that
reactions catalysed by Endo D may be driven to near completion,
ruling out “product” inhibition and also supporting the view
that competitive oxazoline hydrolysis was the major factor which
limited synthetic efficiency. Studies with donor 1 were concluded
by investigating if de-glycosylated ribonuclease B (dRNase B),
which has previously been used as a model protein acceptor for
enzymatic glycosylations using both Endo A and Endo M, could
be glycosylated with Endo D. However, no appreciable amount
of glycosylation of dRNase B was observed using the previously
applied conditions, even when a large excess of donor 1 was used.
The precise reasons for the lack of glycosylation activity in this
instance are as yet unclear, particularly as both Endo A and
Endo M are capable of glycosylating dRNase B with oxazoline
donor 1. We have established that Endo D itself does not catalyse
hydrolysis of commercially available ribonuclease B (RNase B),
but presumably this lack of hydrolytic activity is primarily due to
the fact that RNase B is itself decorated with high mannose glycans
that Endo D cannot cleave, rather than to any innate inability of
Endo D to hydrolyse glycans that are attached to RNase B. It is
therefore most probable that the lack of synthetic activity so far
observed with dRNase B is because direct competitive oxazoline
hydrolysis is a much more rapid process than glycosylation of
the GlcNAc residue attached to the protein under the conditions
investigated to date. Further kinetic investigations will be required
in order to confirm this fact, and more importantly to develop a
reaction process in which useful synthetic activity can be observed.
Variation of oxazoline donors
As mentioned previously, Endo D has been reported to possess
quite strict substrate requirements for hydrolytic activity towards
Asn-linked N-glycans. It may therefore have been expected that
the synthetic activity of Endo D could also be quite limited with
respect to the oxazoline structures it was able to process. The caveat
to this presumption is that the ability of an endohexosaminidase
to catalyse hydrolysis of, or glycosylation of an acceptor with,
an N-glycan oxazoline donor does not necessarily correlate
with its corresponding hydrolytic activity towards the product
of that synthetic reaction. Such variation may stem from the
1864 | Org. Biomol. Chem., 2010, 8, 1861–1869
This journal is
The Royal Society of Chemistry 2010
©

View Article Online
consideration that the normal glycoside bond cleavage reaction
catalyzed by Endo D is a two-step process involving the transient
formation of a relatively high-energy oxazoline intermediate
and its subsequent hydrolysis. The hydrolysis of a high-energy
oxazoline intermediate, however, is only the second half of the
overall process for glycoside bond cleavage. Therefore, it is quite
possible that Endo D would be capable of hydrolysing an oxazoline
intermediate, since this could be a fast step in the normal glycoside
bond cleavage reaction, and yet be unable to form the same
oxazoline from the glycoside substrate, since formation of the
oxazoline could be the slow rate determining step. Given this
uncertainty, the flexibility of the Endo D with respect to oxazoline
substrates it is able to process was investigated by assaying a
variety of alternative oxazoline donors as synthetic substrates
using the standard reaction conditions detailed previously (pH 6.5,
no co-solvent, single addition of 3 equivalents of oxazoline donor,
Scheme 2).
Firstly, use of the disaccharide oxazoline donor 4 produced
the trisaccharide 5 in a maximum 43% yield after around 2.5 h.
Interestingly, if this reaction was left for an extended period of
time the yield of product 5 was found to decrease slightly to 38%.
However, rather than being due to product hydrolysis, it was found
that this reduction in yield was in fact due to a second glycosylation
of 5. Close inspection of the HPLC trace revealed the formation
of a new minor peak, and when this product was collected and
subjected to mass spectrometric analysis it was identified as a
pentasaccharide, presumably arising from glycosylation of the
4-hydroxyl of the terminal mannose residue of 5 by a second
equivalent of the disaccharide oxazoline 4. Previous studies on
both Endo A11b and Endo M24 have revealed the ability of sugars
other than GlcNAc to act as acceptors, including mannose.
Moreover Wang has recently demonstrated14k the potential for
Endo A to effect multiple glycosylations and so yield oligomers,
albeit for an acceptor/donor having a terminal glucose unit. It
appears that a similar situation operates here, and that Endo D can
also glycosylate an acceptor other than GlcNAc using an oxazoline
donor. The addition of a further 3 equivalents of oxazoline donor
4, in an attempt to push the reaction towards completion did
result in an increase in the yield of product 5 obtained (63%),
but there was also a corresponding increase in the yield of the
pentasaccharide product arising from double glycosylation.
Endo D was also found to be capable of effecting glycosylation
of the acceptor 2 with both the (1–3) linked trisaccharide donor
6, which gave the tetrasaccharide product 7 in a maximum yield
of 14%, and the (1–6) linked trisaccharide donor 8, which gave
tetrasaccharide 9 in a maximum yield of 24%. Although the
efficiency of both of these processes appeared to be significantly
lower than when the tetrasaccharide donor 1 had been employed,
both reactions were complete within one hour. In both cases the
product yield did not decrease after prolonged reaction times,
indicating that Endo D was unable to hydrolyse the products 7 or
9 at an appreciable rate, and again implying that the modest yields
were due to competitive oxazoline hydrolysis.
analogue of 10, was also processed by the enzyme and afforded
heptasaccharide 13 in 9% yield after approximately 9 h. It has been
shown previously that both Endo M13b and Endo A13c are able to
tolerate a manno- or glucopyranose at this internal position, using
oxazoline donors to synthesise products that these enzymes are
then unable to hydrolyse. The fact that 13 was produced in this
study indicated that Endo D can also process oxazoline donors
containing a central glucopyranose unit, though in this instance
the rate of reaction was significantly decreased. The fact that
product 13 was not hydrolysed by Endo D was in this instance
perhaps less surprising than the analogous observations made
using either Endo A or Endo M, since in these experiments Endo
D did not display hydrolytic activity against any of the reaction
products. Finally, hexasaccharide 14, an oxazoline structure that
corresponds to a truncated complex-type hexasaccharide, was also
processed by Endo D and gave heptasaccharide 15, albeit in a very
modest 5% yield. In this case the reaction was complete within
30 min, and again no product hydrolysis was observed. This final
result is of particular interest since Endo D has been reported
to be unable to hydrolyse complex N-glycans before removal of
all the external residues, including all of the GlcNAc units. In this
instance the successful, yet inefficient, synthesis of 15 indicates that
the previously identified requirement17b,18e,21 for Endo D activity for
an exposed 2-hydroxyl group on the a(1–3)-linked mannose is not
strictly adhered to when using an oxazoline donor. However, the
very low product yield means that this process is not currently
synthetically useful.
Investigation of enzyme kinetics
In order to confirm the hypothesis that direct oxazoline hydrolysis
was the main competing side reaction, a brief investigation of
the kinetics of Endo D catalysed oxazoline and product hydrolysis
was undertaken using one representative oxazoline. Thus, the rates
of enzyme-catalysed hydrolysis of the (1–6)-linked trisaccharide
oxazoline 8 and the corresponding synthetic product of the
enzymatic glycosylation reaction, tetrasaccharide amino acid 9
(which could be a substrate for the enzyme) were measured.
Kinetic studies were performed by the use of HPLC analysis of
aqueous reaction mixtures in which either material (8 or 9) was
treated with Endo D. The rate of hydrolysis was determined by the
use of an evaporative light scattering detector (ELSD) in the case
of 8, or a UV detector in the case of 9, to analyze reaction mixtures
that were quenched after 5 min. For the hydrolysis of oxazoline
8, Michaelis–Menten analysis (see the ESI‡) revealed Km to be
9 mM ( 3). The hydrolytic rate constant kcat was measured as
220 min-1 ( 20), and the apparent second order rate constant of the
hydrolytic process represented by kcat/Km was therefore estimated
to be 24 min-1 mM-1. Kinetic studies on the hydrolysis of the
corresponding synthetic product 9 (the product obtained from
glycosylation of acceptor 2 with oxazoline 8) were also attempted.
However the rate of hydrolysis of 9 was found to be so slow that
measurement of kinetic parameters proved to be impracticable.
For example, no hydrolysis of 9 was observed after 60 min under
equivalent reaction conditions and concentrations that had led
to significant hydrolysis of oxazoline 8 after 5 min. Subsequent
incubation of tetrasaccharide 9 with Endo D for extremely long
periods of time revealed that only ~10% hydrolysis was achieved
after 7000 min of reaction time; a result that contrasted with the
Glycosylations were also undertaken using three different hexas-
accharide oxazolines 10, 12 and 14 (Scheme 2). Under the standard
reaction conditions the high-mannose type hexasaccharide 10 gave
heptasaccharide 11 in 12% yield; the reaction was complete after
approximately 3.5 h, and no product hydrolysis was observed
after this time. The donor 12, a non-natural gluco-modified
This journal is
The Royal Society of Chemistry 2010
Org. Biomol. Chem., 2010, 8, 1861–1869 | 1865
©

View Article Online
Scheme 2 Reagents and conditions: (i) Endo D (SpGH85), Na2HPO4 (100 mM), pH 6.5, 23 ◦C.
1866 | Org. Biomol. Chem., 2010, 8, 1861–1869
This journal is
The Royal Society of Chemistry 2010
©

View Article Online
observation of ~50% hydrolysis of 8 under identical conditions
after merely 5 min. Thus kcat/Km for the hydrolysis of synthetic
product 9 by Endo D was crudely estimated to be at least three
orders of magnitude lower than that for hydrolysis of oxazoline 8.
Based on previous structure–activity relationships reported for
the hydrolytic activity of Endo D it is unsurprising that Endo
D is unable to effect hydrolysis of 9; tetrasaccharide 9 does
not contain a mannose unit at the 3-position of the central
branching mannose residue. However the rapid hydrolysis of the
corresponding oxazoline structure 8, which also does not possess a
mannose unit at the 3-position of the central branching mannose,
confirms the considerable discrepancy that may exist between the
rates of oxazoline and product hydrolysis. In this instance the
high energy oxazoline intermediate 8 is hydrolysed by Endo D
rapidly, whereas the corresponding synthetic reaction product 9 is
effectively inert to hydrolysis.
related to Endo D should also display the ability to glycosylate
acceptors using oxazolines as donors, and thus themselves may
be useful catalysts for the synthesis of defined glycoconjugates.
Furthermore, the recently reported crystal structure of Endo D20
will undoubtedly provide key structural insight that will allow
the future fine-tuning and optimisation of the synthetic activity
of Endo D by protein engineering; for example by effecting a
reduction in the rate of competing oxazoline hydrolysis. Such
mutational studies on Endo D, and investigations into the effect of
core fucosylation on the transfer reaction are currently in progress,
and the results will be reported in due course.
Experimental
General
◦
Enzymatic experiments were performed at 23 C and monitored
This discrepancy in the ability of Endo D to process the
oxazoline intermediate by transfer to either water or a glycosyl
acceptor, as compared to its inability to hydrolyse the correspond-
ing glycosylation product, suggests that the Endo D-catalyzed
glycosylation reaction has potential scope for improvement.
Currently, the problem presented here is that Endo D does not
discriminate effectively between water or acceptor 2 during the
processing of the oxazolines studied here. This lack of selectivity
in the group transfer reaction may stem from the fact that acceptor
2 is itself not recognized efficiently by Endo D, which is consistent
with the fact that glycoside 9 is a very poor substrate for this
enzyme. Two interesting lines of study that we are now pursuing,
following this initial feasibility study, is that inclusion of fucose
on the acceptor may favour partitioning of the oxazoline toward
transfer to the glycan acceptor rather than to water. The second
idea is that mutation of Endo D to favour partitioning to a glycan
acceptor over water should also be feasible.
by reverse phase HPLC by one of two methods. Method A:
using a Hewlett-Packard 1050 HPLC instrument using Clarity
software (version 2.4.1.43) connected to an Agilent 1100 variable
wavelength detector at 257 nm wavelength. An analytical HPLC
(Phenomenex Gemini 5 m C-18 column, 250 ¥ 4.6 mm) was used
to monitor the synthetic reactions, with 2 mL aliquots taken
at appropriate time intervals. The column was eluted with 22%
MeCN–H2O for 20 min at a flow rate of 1 mL min-1. The
yield was determined by integration of the product and acceptor
peaks. Method B: using a Dionex Ultimate 3000 instrument using
Chromeleon software (version 6.80, build 2212) connected to
a Dionex Ultimate 3000 variable wavelength detector at 207,
220, 254 and 280 nm wavelengths. An analytical HPLC (Grace
Platinum 3 m C-18 column, 100 ¥ 4.6 mm) was used to monitor the
reactions, with 1 mL aliquots taken at appropriate time intervals
and diluted into 10 mL of water. The column was eluted by a
linear gradient of 25-35% MeCN containing 0.3% TFA for 15 min,
followed by a 95% MeCN wash for 2 min and reequilibration at
25% MeCN for 3 min, all at a flow rate of 1 mL min-1.
Conclusions
The above studies demonstrate for the first time that Endo D, like
Endo A and Endo M, two other family 85 glycoside hydrolase
enzymes, is capable of catalysing the glycosylation of an acceptor
bearing a GlcNAc residue using N-glycan oxazolines as activated
donors. Methodological studies revealed that synthetic efficiency
is somewhat lower than for Endo A or Endo M, and that the major
competing process is direct enzyme-catalysed oxazoline hydrolysis.
The significant rate of direct oxazoline hydrolysis was confirmed
by kinetic studies in one instance. These studies revealed that direct
hydrolysis of an oxazoline occurred rapidly even though Endo D
was not capable of hydrolysing the corresponding product that
was produced by the synthetic reaction employing this oxazoline
together with a GlcNAc acceptor. Unfortunately attempted Endo
D-catalyzed glycosylation of de-glycosylated RNase B (dRNase
B) was unsuccessful, presumably due to the high rate of direct
competing oxazoline hydrolysis. However, it is pertinent to note
that Endo D is capable of employing a variety of structurally
different oxazolines as donors and, as alluded to above, tolerated
structures are not limited to known hydrolytic substrates for
the enzyme. Indeed in the present study none of the glycosyl
amino acid products were found to be hydrolysed by Endo D
at an appreciable rate. The synthetic capabilities demonstrated
by Endo D indicate therefore that other family GH85 enzymes
Materials
The catalytic domain (SpGH85) of Endo D was produced in E. coli
(BL21) and purified as previously described.20 N-glycan oxazoline
donors 1, 4, 6, 8, 10, 12 and 14 and the Asn-GlcNAc glycosyl amino
acid acceptor 2 were all synthesised as previously described.13a,13c,13e
Glycosyl amino acid products 3, 5, 7, 9, 11, 13 and 15 exhibited
spectral data identical to those reported previously,13c,13e and which
are provided in the ESI.§
Typical glycosylation procedures
Method 1: variation of the quantity of Endo D employed, and the
use of various oxazoline donors. Glycosyl amino acid acceptor
2
13a (100 mg, 621 nmol) and oxazoline donor (3 eq.) were dissolved
in sodium phosphate buffer (0.1 M solution, pH 6.5, total reaction
volume: 50 mL). Endo D (5 mL of a 5.8 mg mL-1 solution;
390 mU or overall concentration of 7.8 mU mL-1)17d,22 was added
and the temperature maintained at 23 ◦C. Reaction progress was
monitored by HPLC analysis (method A) and the yield determined
by UV integration of the product and acceptor peaks. To drive the
reaction between 1 and 2 towards completion the reaction was
conducted as described above except that 3 further equivalents
This journal is
The Royal Society of Chemistry 2010
Org. Biomol. Chem., 2010, 8, 1861–1869 | 1867
©

View Article Online
of donor 1 were added after every hour for 4 h (i.e. to a total
of 15 equivalents of donor). The progress of the reaction was
monitored by HPLC analysis of aliquots taken immediately before
each subsequent addition.
volume: 8 mL) in the presence of Endo D (5.8 ng, 1 mL of a
5.8 mg ml-1 solution of enzyme; molecular weight, 73 828.48; final
enzyme concentration: 9.82 ¥ 10-6 M). The extent of oxazoline
hydrolysis was measured after 5 min by HPLC analysis (method
A). Samples were spiked with a known quantity of methionine to
act as internal standard, and quantities of oxazoline and hydrolytic
product were obtained by using the calibration curve data for each
component. The initial rate of reaction (mol min-1) was obtained
by division by 5. The kinetic parameters, kcat, Km, and Vmax, were
then obtained by fitting this experimental data to the Michaelis–
Menten model using the non-linear regression program Graphpad
PrismTM: kcat = 219.5 min-1 ( 18.3); Km = 0.0089 M ( 0.003); and
Vmax = 0.002156 mol min-1.
Method 2: variation of co-solvent. Oxazoline donor 1 (214 mg,
310 nmol 3 eq.) and glycosyl amino acid acceptor 2 (50 mg,
103 nmol) were dissolved in a mixture of the appropriate organic
solvent (HPLC grade), water and sodium phosphate buffer (0.4 M
solution, pH 6.5, total reaction volume: 25 mL; overall buffer
concentration 0.1 M). Endo D (2.5 mL of a 5.8 mg mL-1 solution;
195 mU or overall concentration of 7.8 mU mL-1) was added
and the temperature maintained at 23 ◦C. Reaction progress was
monitored by HPLC analysis (method A) and the yield determined
by UV integration of the product and acceptor peaks.
Hydrolysis of tetrasaccharide 9
Method 3: variation of reaction pH. Similarly 2 (50 mg,
103 nmol) and oxazoline donor 1 (214 mg, 310 nmol, 3 eq.) were
dissolved in sodium phosphate buffer (0.1 M solution, pH 6.5,
total reaction volume: 25 mL). Endo D (2.5 mL of a 5.8 mg mL-1
solution; 195 mU or overall concentration of 7.8 mU mL-1) was
added and the temperature maintained at 23 ◦C. Reaction progress
was monitored by HPLC analysis (method A) and the yield
determined by UV integration of the product and acceptor peaks.
Hydrolytic reaction reactions were investigated using tetrasac-
charide 9 (2 mL of an 11 mg ml-1 solution; overall substrate
concentration 3.67 mg ml-1) at 23 ◦C in sodium phosphate buffer
(3 mL of a 0.1 M solution, pH 6.5; total reaction volume: 6 mL) in
the presence of Endo D at two enzyme concentrations: (a) 0.58 ng,
1 mL of a 0.58 mg ml-1 solution of enzyme; molecular weight,
73 828.48; final enzyme concentration: 1.3 ¥ 10-6 M; (b) 5.8 ng, 1 mL
of a 5.8 mg ml-1 solution of enzyme; final enzyme concentration:
13.1 ¥ 10-6 M). In both cases no appreciable amount of hydrolysis
of tetrasaccharide 9 was observed after 60 min by HPLC analysis
(method B). After 4 d, 17 h and 15 min (6795 min) of reaction time
analysis of (b) indicated that 10% of tetrasaccharide 9 had been
hydrolysed.
Method 4: attempted glycosylation of dRNase B. Deglycosy-
lated RNase B (prepared previously) (200 mg, 14.4 nmol) and
oxazoline donor 1 (50 mg, 72 nmol, 5 eq.) were dissolved in sodium
phosphate buffer (0.1 M solution, pH 6.5, total reaction volume:
12 mL). Endo D (1.0 mL of a 5.8 mg mL-1 solution; 78 mU or overall
concentration of 6.5 mU mL-1) was added and the temperature
maintained at 23 ◦C. Reaction progress was monitored by HPLC
analysis (method B).
Acknowledgements
We thank Dr Christoph Heidecke for the synthesis of oxazoline
donors, and Mr Robin Proctor for assistance with mass spectro-
metric analysis. We also thank the EPSRC (DTA studentship to
TBP) for funding.
Kinetic studies
HPLC analysis for enzyme kinetics was carried out by one of
two methods. Method A: using a Dionex Ultimate 3000 HPLC
system using Chromeleon software (version 6.80 SP2 Build 2284)
and a Phenomenex Gemini 5 m C-18 column, 250 ¥ 4.6 mm. The
column was eluted with 5% MeCN–H2O over a 5 min period for
each analysis. This was followed by a gradient to 95% MeCN over
1 min, 95% MeCN for 2 min, gradient to 5% MeCN over 1 min and
re-equilibrate for 5 min. Detection was carried out using a Polymer
Laboratories PL-ELS 2100 Ice system. The non-linear response
of the evaporative light scattering detector (ELSD) was accounted
for by separate calibration for each compound involved.25 Method
B: using a Hewlett-Packard 1050 HPLC instrument using Clarity
software (version 2.4.1.43) connected to an Agilent 1100 variable
wavelength detector at 257 nm wavelength. An analytical HPLC
(Phenomenex Gemini 5 m C-18 column, 250 ¥ 4.6 mm) was used
to monitor the synthetic reactions by analysing 1 mL aliquots. The
column was eluted with 22% MeCN–H2O for 20 min at a flow
rate of 1 mL min-1. The yield was determined by integration of the
product and acceptor peaks.
References
1 (a) F. Maley, R. B. Trimble, A. L. Tarentino and T. H. Plummer, Anal.
Biochem., 1989, 180, 195–204; (b) Y. Karamanos, S. Bourgerie, J. P.
Barreaud and R. Julien, Res. Microbiol., 1995, 146, 437–443.
2 T. Muramatsu, J. Biol. Chem., 1971, 246, 5535–5537. Note that
Diplococcus pneumoniae was renamed Streptococcus pneumoniae in
1974; the D of Endo D originating from Diplococcus.
3 http://www.cazy.org/fam/acc_GH.html.
4 (a) A. L. Tarentino and F. Maley, J. Biol. Chem., 1974, 249, 811–817;
(b) A. L. Tarentino, T. H. Plummer and F. Maley, J. Biol. Chem., 1974,
249, 818; (c) R. J Trumbly, P. W. Robbins, M. Belfort, F. D. Ziegler, F.
Maley and R. B. Trimble, J. Biol. Chem., 1985, 260, 5683–5690; (d) V.
Rao, C. Guan and P. Van Roey, Structure, 1995, 3, 449–457.
5 Note that Streptomyces plicatus (ATCC 27800) was previously classified
as Streptomyces griseus.
6 (a) J. R. Rich and S. G. Withers, Nat. Chem. Biol., 2009, 5, 206–215;
(b) R. J Sola´ and K. Griebenow, J. Pharm. Sci., 2009, 98, 1223–1245;
(c) C. S. Bennett and C.-H. Wong, Chem. Soc. Rev., 2007, 36, 1227–
1238; (d) N. Sethuraman and T. A. Stadheim, Curr. Opin. Biotechnol.,
2006, 17, 341–346; (e) A. Brik, S. Ficht and C.-H. Wong, Curr. Opin.
Chem. Biol., 2006, 10, 638–644.
7 T. Shinkawa, K. Nakamura, N. Yamane, E. Shoji-Hosaka, Y. Kanda,
M. Sakurada, K. Uchida, H. Anazawa, M. Satoh, M. Yamasaki, N.
Hanai and K. Shitara, J. Biol. Chem., 2003, 278, 3466–3473.
8 S. Wildt and T. U. Gerngross, Nat. Rev. Microbiol., 2005, 3, 119–128.
9 P. Umana, E. Moessner, R. Grau, C. Gerdes, A. Nopora, C. Schmidt,
P. Strein, S. Bauer, T. Friess, K. Dabbagh, J. DalPorto and C. Klein,
Ann. Oncol., 2008, 19, 115–115.
Hydrolysis of oxazoline 8
Hydrolytic reactions were performed at various concentrations of
oxazoline 8 (4, 8, 12, 16, 20, 25, 32, 40 and 50 mg ml-1) at 23 ◦C in
sodium phosphate buffer (0.1 M solution, pH 6.5, total reaction
1868 | Org. Biomol. Chem., 2010, 8, 1861–1869
This journal is
The Royal Society of Chemistry 2010
©

View Article Online
10 (a) K. Yamamoto, S. Kadowaki, J. Watanabe and H. Kumagai,
Biochem. Biophys. Res. Commun., 1994, 203, 244–252; (b) K.
Yamamoto, S. Kadowaki, M. Fujisaki, H. Kumagai and T. Tochikura,
Biosci., Biotechnol., Biochem., 1994, 58, 72–77; (c) K. Haneda, T. Inazu,
K. Yamamoto, H. Kumagai, Y. Nakahara and A. Kobata, Carbohydr.
Res., 1996, 292, 61–70; (d) K. Yamamoto, K. Fujimori, K. Haneda, M.
Mizuno, T. Inazu and H. Kumagai, Carbohydr. Res., 1997, 305, 415–
422; (e) M. Mizuno, K. Haneda, R. Iguchi, I. Muramoto, T. Kawakami,
S. Aimoto, K. Yamamoto and T. Inazu, J. Am. Chem. Soc., 1999, 121,
284–290.
11 (a) K. Takegawa, M. Nakoshi, S. Iwahara, K. Yamamoto and T.
Tochikura, Appl. Environ. Microbiol., 1989, 55, 3107–3112; (b) K.
Takegawa, M. Tabuchi, S. Yamaguchi, A. Kondo, I. Kato and S.
Iwahara, J. Biol. Chem., 1995, 270, 3094–3099; (c) J.-Q. Fan, K.
Takegawa, S. Iwahara, A. Kondo, I. Kato, C. Abeygunawardana and
Y. C. Lee, J. Biol. Chem., 1995, 270, 17723–17729; (d) J.-Q. Fan, M. S.
Quesenberry, K. Takegawa, S. Iwahara, A. Kondo, I. Kato and Y. C.
Lee, J. Biol. Chem., 1995, 270, 17730–177359; (e) J.-Q. Fan, L. H.
Huynh, B. B. Reinhold, V. N. Reinhold, K. Takegawa, S. Iwahara, A.
Kondo, I. Kato and Y. C. Lee, Glycoconjugate J., 1996, 13, 643–652.
12 M. Fujita, S.-i. Shoda, K. Haneda, T. Inazu, K. Takegawa and
K. Yamamoto, Biochim. Biophys. Acta, Gen. Subj., 2001, 1528,
9–14.
13 (a) T. W. D. F. Rising, T. D. W. Claridge, N. Davies, D. P. Gamblin,
J. W. B. Moir and A. J. Fairbanks, Carbohydr. Res., 2006, 341, 1574–
1596; (b) T. W. D. F. Rising, T. D. W. Claridge, J. W. B. Moir and A. J.
Fairbanks, ChemBioChem, 2006, 7, 1177–1180; (c) T. W. D. F. Rising,
C. D. Heidecke, J. W. B. Moir, Z. Ling and A. J. Fairbanks, Chem.–
Eur. J., 2008, 14, 6444–6464; (d) C. D. Heidecke, Z. Ling, N. C. Bruce,
J. W. B. Moir, T. B. Parsons and A. J. Fairbanks, ChemBioChem, 2008,
9, 2045–2051; (e) T. B. Parsons, J. W. B. Moir and A. J. Fairbanks, Org.
Biomol. Chem., 2009, 7, 3128–3140.
14 (a) B. Li, Y. Zeng, S. Hauser, H. J. Song and L.-X. Wang, J. Am.
Chem. Soc., 2005, 127, 9692–9693; (b) H. Li, B. Li, H. Song, L. Breydo,
I. V. Baskakov and L.-X. Wang, J. Org. Chem., 2005, 70, 9990–9996;
(c) L.-X. Wang, H. J. Song, S. W. Liu, H. Lu, S. B. Jiang, J. H. Ni and
H. G. Li, ChemBioChem, 2005, 6, 1068–1074; (d) Y. Zeng, J. S. Wang,
B. Li, S. Hauser, H. G. Li and L.-X. Wang, Chem.–Eur. J., 2006, 12,
3355–3364; (e) B. Li, H. Song, S. Hauser and L.-X. Wang, Org. Lett.,
2006, 8, 3081–3084; (f) M. Umekawa, W. Huang, B. Li, K. Fujita, H.
Ashida, L.-X. Wang and K. Yamamoto, J. Biol. Chem., 2008, 283,
4469–4479; (g) L.-X. Wang, Carbohydr. Res., 2008, 343, 1509–1522;
(h) W. Huang, H. Ochiai, X. Zhang and L.-X. Wang, Carbohydr. Res.,
2008, 343, 2903–2913; (i) Y. Wei, C. Li, W. Huang, B. Li, S. Strome and
L.-X. Wang, Biochemistry, 2008, 47, 10294–10304; (j) H. Ochiai, W.
Huang and L.-X. Wang, J. Am. Chem. Soc., 2008, 130, 13790–13803;
(k) H. Ochiai, W. Huang and L.-X. Wang, Carbohydr. Res., 2009, 344,
592–598; (l) W. Huang, C. Li, B. Li, M. Umekawa, K. Yamamoto,
X. Zhang and L.-X. Wang, J. Am. Chem. Soc., 2009, 131, 2214–2223.
15 J. Yin, L. Li, N. Shaw, Y. Li, J. K. Song, W. Zhang, C. Xia, R. Zhang, A.
Joachimiak, H.-C. Zhang, L.-X. Wang, Z.-J. Liu and P. Wang, PLoS
One, 2009, 4, e4658.
16 M. D. L. Suits, Z. Ling, R. J. Bingham, N. C. Bruce, G. J Davies, A. J.
Fairbanks, J. W. B. Moir and E. J. Taylor, J. Mol. Biol., 2009, 389, 1–9.
17 (a) N. Koide and T. Muramatsu, J. Biol. Chem., 1974, 249, 4897–4904;
(b) T. Tai, K. Yamashita, M. Ogata-Arakawa, N. Koide, T. Muramatsu,
S. Iwashita, Y. Inoue and A. Kobata, J. Biol. Chem., 1975, 250, 8569–
8575; (c) N. Koide, M. Nose and T. Muramatsu, Biochem. Biophys.
Res. Commun., 1977, 75, 838–844; (d) T. Muramatsu, N. Koide and
K.-i. Maeyama, J. Biochem., 1978, 83, 363–370; (e) T. Muramatsu,
Methods Enzymol., 1978, 50, 555–559.
18 H. Muramatsu, H. Tachikui, H. Ushida, X.-j. Song, Y. Qiu,
S. Yamamoto and T. Muramatsu, J. Biochem., 2001, 129, 923–928.
19 S. Yamamoto, H. Muramatsu and T. Muramatsu, Glycoconjugate J.,
2005, 22, 35–42.
20 D. W. Abbott, M. S. Macauley, D. J. Vocadlo and A. B. Boraston,
J. Biol. Chem., 2009, 284, 11676–11689.
21 T. Mizuochi, J. Amano and A. Kobata, J. Biochem., 1984, 95, 1209–
1213.
22 Previously 1 unit of enzyme activity has been defined by Mura-
matsu as the amount of enzyme required to hydrolyse 1 mmol
of (Man)5(GlcNAc)2-Asn-[14C]acetyl to give (Man)5GlcNAc and
GlcNAc-Asn-[14C]acetyl per minute, at 37 ◦C. According to ref. 18d
this corresponds approximately to 13.4 units mg-1 of purified enzyme.
23 C. D. Heidecke, T. B. Parsons and A. J. Fairbanks, Carbohydr. Res.,
2009, 344, 2433–2438.
24 (a) T. Yamanoi, M. Tsutsumida, Y. Oda, E. Akaike, K. Osumi, K.
Yamamoto and K. Fujita, Carbohydr. Res., 2004, 339, 1403–1406; (b) K.
Fujita, K. Kobayashi, A. Iwamatsu, M. Takeuchi, H. Kumagai and K.
Yamamoto, Arch. Biochem. Biophys., 2004, 432, 41–49.
25 N. Vervoot, D. Daemen and G. To¨ro¨k, J. Chromatogr., A, 2008, 1189,
92–100.
This journal is
The Royal Society of Chemistry 2010
Org. Biomol. Chem., 2010, 8, 1861–1869 | 1869
©