close

Вход

Забыли?

вход по аккаунту

?

Enzymes in Organic Synthesis Application to the Problems of Carbohydrate Recognition (Part 2).

код для вставкиСкачать
REVIEWS
Enzymes in Organic Synthesis :
Application to the Problems of Carbohydrate Recognition (Part 2)**
Chi-Huey Wong,* Randall L. Halcomb, Yoshitaka Ichikawa, and Tetsuya Kajimoto
Recognition of carbohydrates by
proteins and nucleic acids is highly
specific, but the dissociation constants
are relatively high (generally in the mM
to high p~ range) because of the lack of
hydrophobic groups in the carbohydrates. The high specificity of this weak
binding often comes from many hydrogen bonds and the coordination of metal
ions as bridge between sugars and receptors. Though weak hydrophobic interactions between sugars and proteins
have also been identified, the unique
shape of a complex carbohydrate under
the influence of anomeric and ex0
anomeric effects (the glycosidic torsion
angles are therefore often not flexible
but are typically somewhat restricted)
and the topographic onentation of the
hydroxyl and charged groups contribute
most significantly to the recognition process. Studies on the structure-function
relationship of a complex carbohydrate
therefore require deliberate manipulation of its shape and functional groups,
and synthesis of oligosaccharide analogs
from modified monosaccharides is often
useful to address the problem. The
availability of various monosaccharides
and their analogs for the synthesis of
complex carbohydrates together with
the information resulting from structural studies (such a N M R or X-ray
studies on sugar-protein complexes)
will certainly provide a basic understanding of complex carbohydrate
recognition. An ultimate goal is to develop simple and easy-to-make non-carbohydrate molecules that resemble the
active structure involved in carbohy-
5. Enzymatic Glycosidic Bond Formation
Nature employs two groups of enzymes in the biosynthesis of
oligosaccharides: the enzymes of the Leloir pathway['361 and
those of non-Leloir pathways. The Leloir pathway enzymes are
responsible for the synthesis of most N- and 0-linked glycoproteins and other glycoconjugates in mammalian systems. Nlinked oligosaccharides are characterized by a b-glycosidic linkage between a GlcNAc residue and the 6-amide nitrogen atom
[*I
[**I
PI-of. Dr. C'.-H. Wong. Dr. R. L. Halcotnb
Dcpartinent of Chemistry. The Scripps Research Institute
10666 North Torrej Pines Road, La Jolla, CA 92037 (USA)
Telet'ax' l n t . code + (619)554-6731
ProC Dr.Y lchikawa
Department or Phai-macology and Molecular Sciences
Thc l o h n s Hopkins University. Baltimore (USA)
Dr. T. K q m o t o
Fronticr Rescarch Program on Glycotechnology
l l i e InsLitute ofPhysica1 and Chemical Research ( R I K E N ) , Wdko City (Japan)
Part 1 : i I ! i , y i w C'hc~ni. 1995, 107, 453; A n g w . Chivn. I n f . Ed. EnjiI. 1995. 34,
417. The numbering ofthe sections. references, schemes. and Tdbles follows on
t'rom Part 1 .
drate-receptor interaction or the transition-state of an enzyme-catalyzed transformation (for example. glycosidase or
glycosyltransferase reactions) and have
the approprite bioavailability to be used
to control the carbohydrate function in a
specific manner. In part one of this review we described various enzymatic approaches to the synthesis of monosaccharides, analogs, and related structures.
We describe in this part enzymatic and
chemoenzymatic approaches to the synthesis of oligosaccarides and analogs, including those involved in E-selectin recognition, and strategies to inhibit glycosidases and glycosyltransferases.
Keywords: carbohydrates. enzymes . organic synthesis
J
of an asparagine residue. The less common 0-linked glycoproteins contain an 3-glycosidic linkage between a GalNAc (or
xylose) and the hydroxyl group of a serine or threonine residue.
Both 0-and N-linked glycoproteins are cotranslationally glycosylated in the endoplasmic reticulum and the Golgi apparatus.'"' 371 The biosynthesis of the N-linked type involves an
initial synthesis of a dolichyl pyrophosphoryl oligosaccharide
intermediate in the endoplasmic reticulum by the action of GlcNAc-transferases and mannosyltransferases. This structure is
further glucosylated, and then the entire oligosaccharide moiety
is transferred to an ASn residue of the growing peptide chain by
the enzyme oligosaccharyltransferase.['371The Asn is typically
part of the amino acid sequence Asn-X-Ser(Thr). where X is not
Pro or Asp.[' 3 7 - l4I1 Before transport into the Golgi apparatus,
the glucose residues and some mannose residues are removed in
a process called trimming by the action of glucosidase I and I1
and a mannosidase to reveal a core pentasaccharide (peptideAsn-(GlcNAc),-(Man),). The resulting core structure is further
processed by mannosidases and glycosyltransferases present in
the Golgi apparatus to produce either the high-mannose type,
the complex type, o r the hybrid type oligosaccharides. Mono-
C.-H. Wong et al.
REVIEWS
saccharides are then added sequentially to this core structure to
provide the fully elaborated oligosaccharide chain.
In contrast to the dolichyl pyrophosphate mediated synthesis
of N-linked oligosaccharides, the glycosyltransferases necessary
for the synthesis of 0-linked oligosaccharides are located in the
Golgi apparatus.[137JThe biosynthetic route to the 0-linked
type is also different. Monosaccharide residues are added sequentially to the growing oligosaccharide chain.
All mammalian cells, with the exception of erythrocytes, contain the necessary elements for glycosylation. In certain secretory
cells, however, the preponderance of transferases is greater."421
As glycosyl donors, the glycosyltransferases of the Leloir
pathway in mammalian systems utilize monosaccharides which
are activated as glycosyl esters of nucleoside mono- or diphosphates.['371 Non-Leloir transferases typically utilize glycosyl
phosphates as activated donors. The Leloir glycosyltransferases
employ primarily eight nucleoside mono- or diphosphate sugars
as monosaccharide donors for the synthesis of most oligosaccharides : UDP-Glc, UDP-GlcNAc, UDP-Gal, UDP-GalNAc,
GDP-Man, GDP-Fuc, UDP-GlcUA, and CMP-NeuAc. Many
other monosaccharides, such as the anionic or sulfated sugars of
heparin and chondroitin sulfate, are also found in mammalian
systems, but they usually are a result of modification of a particular sugar after it is incorporated into an oligosaccharide structure. A very diverse array of monosaccharides (for example,
xylose, arabinose, 3-deoxy-rnanno-octulosonate (KDO), deoxysugars) and oligisaccharides is also resent in microorganisms,
plants, and invertebrate^."^^. 1441 The enzymes responsible for
their biosynthesis, however, have not been extensively exploited
for synthesis, though they follow the same principles as those in
mammalian systems.
The glycosyltransferases from the Leloir and non-Leloir
pathways as well as glycosidase~['~]
have been exploited for the
synthesis of oligosaccharides and glyco~onjugates.['~~
145, 1461
The function of glycosidases in vivo is to cleave glycosidic
bonds; however, under appropriate conditions they can be useful synthetic catalysts. Each group of enzymes has certain advantages and disadvantages for synthesis. Glycosyltransferases
are highly specific in the formation of glycosides; however, the
availability of many of the necessary transferases is limited.
Fortunately, the recent advances in genetic engineering and recombinant techniques are rapidly alleviating these drawbacks.
Glycosidases have the advantage of wider availability and lower
cost, but they are not as specific or high-yielding in synthetic
reactions. Several other enzymatic methods have also been used
to synthesize N-glycosides, such as nucleosides.
5.1. Donor Substrates for Glycosyltransferases
Most of the sugar nucleoside phosphates used as substrates
for gly~osyltransferases~'~~~
are biosynthesized in vivo from the
corresponding monosaccharides. The initial step is a kinasemediated phosphorylation to produce a glycosyl phosphate.
This glycosyl phosphate then reacts with a nucleoside triphosphate (NTP), catalyzed by sugar nucleoside diphosphate pyrophosphorylase, to afford an activated sugar nucleoside
diphosphate [Eq. (a)]. Other sugar nucleoside phosphates, such
as GDP-Fuc and UDP-GlcUA, are biosynthesized by further
522
enzymatic modification of these existing key sugar nucleoside
phosphates. Another exception is CMP-NeuAc, which is
formed by the direct reaction of NeuAc with CTP [Eq. (b)].
Some of the enzymes involved in the biosynthesis of sugar nucleotides also accept nonnatural sugars as substrates. In general,
however, the rates are quite slow, thus limiting the usefulness of
this approach.
-
Sugar-1-P + NTP
NeuAc
+ CTP
NDP-Sugar
+ PP,
CMPNeuAc + PP,
(4
(b)
5.1.1. Preparation of Nucleoside Triphosphates
Chemical syntheses of some sugar nucleoside phosphates have
been
Most of these methods involve the reaction of
an activated nucleoside monophosphate (NMP)['48-'521with a
glycosyl phosphate to produce a sugar nucleoside diphosphate.
Of the commonly used activated NMP derivatives, phosphoramidates such as phoshorimidazolidates[153
- ' 5 5 1 and phosphorornorpholidate~['~~1 5 2 J are considered the most effective.
These activated NMPs may also be used to prepare NTPs by
reaction with pyrophosphate.[' 5 5 1 A number of chemical methods
are available for the synthesis of glycosyl phosphates. Reactions
of phosphates with activated glycosyl donors" 5 6 J or chemical
phosphorylation of anomeric hydroxyl groups['51 - 53. ' 571 h ave
proven to be convenient. Additionally, routes via glycosyl phosphites are useful.[' Enzymatic procedures are also available.
Glycogen phosphorylase[' 591 and sucrose phosphorylase['601
were used to produce cc-glucose-1-phosphate.Phosphoglucomutase can also be used to prepare glucose-I -phosphate from glu~ose-6-phosphate,['~'~
which is in turn synthesized from glucose
by hexokinase.
The appropriate nucleoside triphosphates are utilized as substrates for the biosynthesis of sugar nucleoside phosphates.
Biosynthesis-based enzymatic preparation of these donors for
use in glycosylations therefore requires a synthesis of NTPs
suitable for scale-up.
Most preparative-scale enzymatic syntheses of NTPs use
commercially available NMPs as starting materials. Alternatively, all of the NMPs can be obtained from yeast RNA digests
at low cost,[1621
or can be easily prepared chemically. In general,
these methods involve the sequential use of two kinases to transform NMPs to NTPs, via the corresponding nucleoside diphosphates (NDPs). Either of three kinases may be used to synthesize NTPs from the corresponding NDPs, each of which uses a
different phosphoryl donor: pyruvate kinase [EC 2.7.1.401 uses
phosphoenolpyruvate (PEP)['63, 164] as a phosphoryl donor,
acetate kinase [EC 2.7.2.11 uses acetyl phosphate, and nucleoside-diphosphate kinase [EC 2.7.4.61 uses ATP. Pyruvate kinase
is generally the enzyme of choice, because it is less expensive
than nucleoside-diphosphate kina~e,[''~%
1651 and because PEP is
more stable and provides a more thermodynamically favorable
driving force for phosphorylation than does acetyl phosphate
(Scheme 34).
The preparation of NDPs from NMPs is more complicated
and requires different enzymes for each NMP. Adenylate kinase
[EC 2.7.4.31 phosphorylates AMP['s51 and CMP,['661and also
slowly phosphorylates UMP. Guanylate kinase [EC 2.7.4.83 catalyzes the phosphorylation of GMP. Nucleoside-monophosphate
Angew. Chem. Int. Ed. EngI. 1995,34, 521-546
Enzymes in Organic Synthesis (Part 2)
a' NMPW
-
NDP
1
NTP
NDP
**
REVIEWS
NTP
OP
analogous fashion with UDP-Gal pyrophosphorylase!
Additionally, UDP-Gal can be generated from UDP-Glc by
epimerization of C-4 with UDP-glucose epimerase[I6'l
(Scheme 36). Though the equilibrium for this reaction favors
UD-Glc, it can be shifted by coupling the reaction to an in situ
Scheme 36. Synthesis of UDP-Gal and its use in giycosylations.
t
1. P(OCH&
2. HpOiUOH
Scheme 34. a ) Synthesis of nucleoside triphoshates (NTPs): 1 adenylate kinase
([EC 2.7.4.31, N = A, C. U), guanylate kinase ((EC 2.7.4.81. N = C), nucleoside
monophosphate kinase ([EC 2.7.4.41, N = U); 2. pyruvate kinase [EC 2.7.1.401.
b) Synthesis of phosphoenol pyruvate (PEP)
kinase [EC 2.7.4.41 uses ATP to phosphorylate AMP, CMP,
GMP. and UMP; however, the enzyme is relatively expensive
and unstable.['"] Both CMP and UMP kinases exist but are not
commercially available. If ATP is required as a phosphorylating
agent. it is usually used in a catalytic amount and recycled from
ADP with pyruvate kinase/PEP or acetate kinase/acetyl phosphate." 671 Phosphoenolpyruvate may be prepared chemically
from p y r ~ v a t e [ ' or
~ ~generated
]
enzymatically from ~-3-phosphoglyceric acid['"I (Scheme 34).
Comparisons of chemical and enzymatic methods for the synthesis of NTPs[' 551 lead to the conclusion that enzymatic methods
provide the most convenient route to CTP and GTP. Chemical
dedmination of CTP is the best method for preparing UTP."
ATP is relatively inexpensive from commerical sources; nevertheless, it has also been synhesized enzymatically from AMP on
a 50 mmol scale. Mixtures of NTPs can be prepared fromRNA
by sequential reactions catalyzed by nuclease PI, polynucleotide
phosphorylase, and pyruvate kinase.[168]This mixture can be
selectively converted into a sugar nucleotide with a particular
sugar nucleoside diphosphate pyrophosphorylase.['
glycosylation with galactosyltransferase. This concept has been
applied to a large-scale synthesis of N-acetyllactosarnine.[' 6 1 1
For gram amounts, UDP-Gal has been prepared from Gal-lphosphate and UDP-Glc with UDP-Gal uridyltransferase,['721
followed in situ by glycosylation (see Section 5.2.8). Furthermore, UDP-Gal has been synthesized from UMP and Gal by
means of dried cells of Torulopsis randida.'' 6 y 1 In this system,
Gal-I-phosphate and UTP were generated in Jitu as substrates
for UDP-Gal pyrophosphorylase. A recent chemical synthesis
of UDP-2-deoxygalactose and its use in glycosylations has been
recently reported." 7 3 1
5.1.3. UDP-N-Acetylglucosamine (UDP-GlcNAc)
Two enzymatic methods have been developed for the synthesis
of UDP-GlcNAc. The first involves a reaction between GlcNAcI-phosphate and UTP, catalyzed by UDP-GlcNAc pyrophosphorylase.[' 691 UDP-GlcNAc pyrophosphorylase is currently not
commercially available; to produce it, however, a whole-cell process using Baker's yeast can be employed. The enzyme from calf
liver"701 can also be used in the synthesis. The second procedure
exploits UDP-Glc pyrophosphorylase to catalyze a condensation between UTP and glucosamine-I -phosphate (GlcN-1 -P)to
afford U D P - g l u c ~ s a m i n e [ ' (Scheme
~ ~ " ~ 37). The product UDPGlcN can then be selectively N-acetylated to provide UDP-
0.po:
"&O
HO
H
NHz
5.1.2. UDP-Glucose (UDP-Glc) and
UDP-Galactose ( U DP-Gal)
UDP-Glucose has been prepared from UTP and glucose-lphosphate under catalysis by UDP-glucose pyrophosphorylase
(Scheme 35).[155.1 6 1 , 1 6 9 1 UDP-G al can be synthesized in an
UTP
UDP
PYr
PEP
6
H2NO-POt-
&$,
UTP
HO
HO
UDP-Glc
H O o . ~ ~pyrophosptwryiase
-
Scheme 35 Synthesis of UDP-Glc.
An,ni,w.
< hiw
Inr. Ed. Engl. 1995, 34. 521 -546
H~o&?$OH
b
*
HO
HO
-
-**
UTP
PPi
H2N0-UDP
IE5
2 pi
Scheme 37. Synthesis of UDP-GlcNAc. E l = hexokinase from yeast, E' =
pyruvate kinase, E' = phosphoglucomutase. E4 = UDP-Glc pyrophosphorylase.
Es = inorganic pyrophosphatase. Pyr = pyruvate.
523
C.-H. Wong et al.
REVIEWS
GlcNAc. GlcN-1-P was synthesized from GlcN by phosphorylation of the 6-position with hexokinase to give GlcN-6-P, followed by a phosphoglucomutase-catalyzed isomerization to
provide GlcN-1-P. The second sequence can be performed in a
hollow fiber rea~tor."'~'
5.1.4. UDP- N-A cetylgalactosamine (UDP-GalNA c )
UDP-GalNAc can be prepared from GalNAc-1-P and UTP
with UDP-GalNAc pyrophosphorylase, or from UDP-GlcNAc
with an epimerase ; however, the necessary enzymes are not
readily available. An alternative procedure has been reported
that is based on a UMP exchange reaction between UDP-Glc
and GalN-1-P,under catalysis by UDP-glucose: galactosylphosphate uridyltransferase [EC 2.7.7.121, which is commercially available (Scheme 38) .1171, 1751 Galactose-I-phosphate is
5.1.5. GDP-Mannose (CDP-Man) and CDP-Fucose (GDP-Fuc)
GDP-Man has been prepared from Glc and GMP with dried
Baker's yeast cells.[821The procedure involves the biocatalytic
conversion of glucose to Man-1 -P,and a subsequent conversion
to GDP-Man with GDP-Man pyrophosphorylase. A cell-free
extract from Baker's yeast has also been used to synthesize GDPMan from m a r ~ n o s e . A
~ ' direct
~ ~ ~ synthesis from chemically
prepared Man-I-P and GTP, catalyzed by GDP-Man pyrophosphorylase [EC 2.7.7.131 is useful for a large scale
(Scheme 39) .[I 5 5 1
Man-1-P
GDP-Man pyrophosphorylase
baker's
yeast
Glc
H
O
4
+
H
Z
H2NO.pO:
fUC0se
1
r-
fucokinase
pfucose-1-P
GDP-FUC
pyrophosphorylase CH;@Z-GDp
GTP
HO
t
HO
lE*
HO
OH
Scheme 38. Synthesis of UDP-Ga1NAc.E' = UDP-GIc
uridyltransferase. E': phosphoglucomutase.
galactosylphosphate
the natural substrate for the enzyme, but 2-deoxygalactose-lphosphate, 2-deoxyglucose-l-phosphate,
and galactosamine-I phosphate are also accepted. The equilibrium constant for the
exchange reaction is about 1 ; phosphoglucomutase was therefore added to remove the product Glc-1-P and shift the equilibrium toward UDP-GalN. The UDP-GalN thus produced
was acetylated with acetic anhydride in a subsequent step to give
UDP-GalNAc.
A modification of the latter procedure has been adapted for
a large-scale synthesis of UDP-GalNAc.[' 72b1 In this procedure,
galactosamine-1-phosphate was prepared with galactokinase
coupled to an ATP regeneration system, and used as a substrate
for the uridyltransferase. UDP-Glc was regenerated in situ from
UTP and the product Glc-1-P, under catalysis by UDP-Glc
pyrophosphorylase. This also shifts the equilibrium toward the
formation of UDP-GalN. Finally, a chemical N-acetylation of
the UDP-GalN thus produced with N-acetoxysuccinimide provides UDP-GalNAc. This procedure has also been adapted to
an in situ synthesis of UDP-GalN for glycosylation, with cofactor regeneration, to provide 8-glycosides of GalN (see Section 5.2.8).
524
CH*%Dp]
0
0-GDP
HO
HoO-UDP
q
GMP-
E'
+
lGTP [
NADPH
PP,
Scheme 39. Synthesis of GDP-Fuc.
GDP-fucose is biosynthesized in vivo from GDP-Man by an
NADPH-dependent oxidoreductase enzyme system. Such systems have also been utilized for in vitro synthesis of GDP-Fuc.
For example, the conversion of GDP-Man to GDP-Fuc was
accomplished with a crude enzyme preparation from Agrobacteriurn radiobacter.[' 7 7 1 NADPH was regenerated in situ from
NADP with glucose-6-phosphate dehydrogenase and Glc-6P.11781
In a similar procedure, GDP-Fuc has been generated in
situ for use in a glycosylation reaction with ctl - 3fucosyltransferase (see Section 5.2.3).['791Enzymes from a minor biosynthetic pathway which synthesize GDP-Fuc from L-fucose-Iphosphate"791or L-fucose"
have also been exploited for synthesis." 7 9 , lSo1 Fucose was phosphorylated by fucokinase
[EC 2.7.1.521 to produce Fuc-I-P, which subsequently underwent a GDP-fucose pyrophosphorylase-catalyzed reaction with
GTP to provide GDP-Fuc. Several practical chemical syntheses
of GDP-Fuc have also been reported." 'I
5.1.6. UDP-Glucuvonic Acid (UDP-GlcUA)
UDP-Glucuronic acid is biosynthesized by oxidation of
UDP-Glc with UDP-Glc dehydrogenase, an NAD-dependent
enzyme. Enzyme preparations from bovine liver have been employed for gram-scale syntheses of UDP-GlcUA (Scheme
40).['55.
The NAD cofactor was regenerated with lactate
dehydrogenase in the presence of pyruvate. Additionally, extracts from guinea pig liver have been used to generate UDPGlcUA in situ for use in enzymatic glycosylations with glucuronykransferases." 381
Angew. Chem. Int. Ed. Engl. 1995, 34, 521-546
Enzymes in Organic Synthesis (Part 2)
HO
PNAD
2NADH
x
L-LDH
REVIEWS
several NeuAc derivatives as substrates. For example, 9-deoxy7,9-dideoxy-, and 4,7,9-trideoxy-NeuAc are all converted into
the corresponding CMP-NeuAc derivative.11y51On the other
hand, neither the 4-0x0, 7-0x0, or &ox0 derivatives of NeuAc,
nor their dimethylacetals, are substrates for CMP-NeuAc synt h e t a ~ e . ~The
' ~ ~ enzyme
]
accepts a variety of modifications at
the 9-position, and the hydroxyl group can be replaced with
several different groups with little effect on the Michaelis constant Km,1186,197-1991
Scheme 40. Synthesis of UDP-GlcUA
5.1.7. CMP-N-Acetylneuraminic Acid (CMP-NeuAc)
CMP-N-acetylneuraminic acid has been prepared enzymatically on small scales (less than 0.5 mmol) from CTP and NeuAc,
under catalysis by CMP-NeuAc synthetase [EC 2.7.7.431
An improvement in this procedure, which is suitable for multigram-scale synthesis,['851has been developed: CTP is itself synthesized in situ from C M P with adenylate kinase and pyruvate
kinase. Adenylate kinase catalyzes the equilibration of CTP and
C M P to CDP, which is subsequently phosphorylated by pyruvate kinase to provide CTP. Another procedure has been employed in which the NeuAc used in the synthesis of CMP-NeuAc was prepared in a NeuAc aldolase-catalyzed reaction of
pyruvate with N-acetylmannosamine, which was itself generated from N-acetylglucosamine by a base-catalyzed epimerization.['"" In a one-pot synthesis of CMP-NeuAc based on the
latter procedure, NeuAc is prepared in situ from N-acetylmannosamine and pyruvate, catalyzed by sialic acid aldolase
(Scheme 41) .[' 861 Chemical syntheses of CMP-NeuAc have also
been reported.['*'- l S 9 ]
The gene encoding E. coli CMP-NeuAc synthetase has been
cloned[lyO,"'I and overexpressed in E. coli using L Z A P vector
and the LacZ p r ~ m o t o r [ ' ~ 1931
' , o r PKK223 vector-tac promotor."941 The enzyme from calf brain has also been cloned and
overexpressed. CMP-NeuAc synthetase was shown to accept
&
HO
HO
OH
NHAc
CMP
PEP
CTP
CDP
Pyr
PEP
Pyr
Scheme 41. Multienzymatic synthesis of CMP-NeuAc. El = NeuAc aldolase, EZ:
CMP-NeuAc synthetase. E 3 : pyruvate kinase, E4: adenylate kinase, Pyr =
pyruvate.
Anfiiw C'hetn In,. Ed. E~zfil.1995. 34. 521-546
5.2. Substrate Specificity and Synthetic Applications
of Glycosyltransferases
In general, for each sugar nucleotide glycosyl donor, many
glycosyltransferases exist, each of which transfers the particular
donor to different acceptors. These enzymes are generally considered to be specific for a given glycosyl donor and acceptor, as
well as for the position and the configuration of the newly
formed glycosidic bond. This specificity has led to the "one
enzyme- one linkage" concept." In other words, the specificity of the glycosyltransferases ensures fidelity in oligosaccharide
sequences in vivo without the use of a template scheme. Though
systematic investigations of the in vitro substrate specificity of
most glycosyltransferases have not been carried out, some deviations from this picture of absolute specificity have been observed, both in the glycosyl donors and acceptors. Moreover,
studies toward the design of inhibitors of glycoprotein biosynthesis[2001have also shown that the specificities of glycosyltransferases are not absolute.
5.2.1. Galactosyltvansferase (CalT)
Because of its availability, PI -4-galactosyltransferase
(UDP-Gal : N-acetylglucosamine /31 -4-galactosyltransferase,
[EC 2.4.1 .22])[20'.2 0 2 1 is one of the most extensively studied
mammalian glycosyltransferases with regard to synthesis and
substrate specificity. This enzyme catalyzes the transfer of galactose from UDP-Gal to the 4-position of ,+linked GlcNAc
residues to produce the GalBl -4GlcNAc substructure. In the
presence of lactalbumin, however, glucose is the preferred acceptor, resulting in the formation of lactose, Gal/jl-4Glc. The enzyme has been employed in the in vitro synthesis of N-acetyllactosamine and glycosides thereof, as well as other galactosides.
Galactosyltransferase utilizes as acceptor substrates N-acetylglucosamine and glucose and B-glycosides thereof, 2-deoxyglucose, D-XylOSe, 5-thioglucose, N-acetylmuramic acid, and myoinositol.[2011Modifications at the 3- or 6-position of the acceptor
GlcNAc are also tolerated. For example, Fucal -6GlcNAc and
NeuAca2 -6GlcNAc are substrates.[203]Acceptor substrates that
are derivatized at the 3-position include 3 - 5 - m e t h y l - G l c N A ~ , [ ~ ~ ~ ~
3-deoxy-GlcNAc, 3-O-aIlyI-GlcNAc/3OBu, and 3-0x0-GlcN A C . [ ~All
~ ~glycosides
]
of GlcNAc that are substrates for the
galactosyltransferase have 8-glycosidic linkages. Both a- and
P-glycosides of glucose are acceptable; however, the presence of
lactalbumin is required for galactosyl transfer onto cc-glcosides.
Neither D-mannose, D-allose, D-galactose, D-ribose, nor D-xylose are substrates. Monosaccharides that have a negative
charge, such as glucuronic acid and cc-glucose-1-phosphate, are
525
REVIEWS
C.-H. Wong et al.
also not accepted as substrates. Scheme 42 depicts several disaccharides that have been synthesized with galactosyltransferase.[1733
204-2061 A particularly interesting example is the
j$-1 ,I -linked disaccharide, in which the anomeric hydroxyl of
3-acetamido-3-deoxyglucose serves as the acceptor
The acetamido function apparently controls the position of glycosylation.
Table 5. Relative rates of fi1-4galactosyltransferase-catalyzed transfer of donor
substrates.
UDP-Gal
Ref.
Hoq
HO
100
(211 a]
0.3
[210, 211 a]
HoO-UDP
UDP-Glc
HO
k,,
Donor substrates
OH
6
HO
HO
HoO-UDP
!&$,
HO
+
H HOO
G
HoO-UDP
o R
~ H A C
GalT
Mn2+
UDP-4-desoxy-Glc
Hog
5.5
[211a]
HoOUJP
OR
NHAc
OH
OH
UDP-GalNAc
4.0
no
H
CH,
= CH,F
=
=
[211 a]
[211 b]
1211 b]
[2101
AcHN&UDp
~2041
.."Go&
OH
4.0, R
1.3, R
0.2, R
UDP-Ara
Ho
UDP-GlcNAc
HO
~2041
6
HO
HO
OR
NHAc
0.00
AcHNO-UDP
[173,205]
UDP-GlcN
Hao*
0.09
H2NO-UDP
OR
NHAc
[2121
UDP-5-thio-Gal
R' = H, OH, OCHS, 0 (3-OXO),
OCHpCH=CHp
12041
Scheme 42. Some disaccharides synthesized with /I1 -4galactosyltransferase.
UDP-2-desoxy-Gal
"q
HO
90
0-UDP
fil -4Galactosyltransferase has also been employed in solidphase oligosaccharide synthesis and has been used to galactosylate gluco or cellobio subunits of polymer-supported oligosaccharides and p o l y s a c c h a r i d e ~ . The
~ ~ ~resulting
~l
oligosaccharides can
then be removed from the support by either a photochemical
cleavage or a chymotrypsin-mediated hydrolysis. The types of
polymer supports employed include polyacrylamide and a
water-soluble poly(viny1 alcohol). N-Acetylglucosaminyl amino
acids and peptides have also been used as substrates for galactosyltransferase to afford galactosylated g l y c o p e p t i d e ~ . ' 2081
~~~.
The carbohydrate chain can then be further extended with other
transferases, such as s i a l y l t r a i i ~ f e r a s e .2081
' ~ ~ ~S'imilarly,
~
the
synthesis of a ceramide glycoside, which was subsequently enzymatically sialylated, provided a GM3analog.[2091
As donor substrates, the B-galactosyltransferase also transfers glucose, 4- and 6-deoxygalactose, arabinose, glucosamine,
gahctosamine, N-acetylgalactosamine, 2-deoxygaiactose, and
2-deoxyglucose from their UDP-derivatives to provide an enzymatic route to oligosaccharides that terminate in j-1,4-linked
residues other than galactose (Table 5).[210-2131
An example
worthy of note IS the transfer of 5-thiogalactose (entry 8).12121
Although the rate of the enzyme-catalyzed transfer of many of
526
[a] Generated In situ from 2-deoxygalactose with galactokinase/uridyltransferase
[213]. [b] Generated in situ from 2-deoxyglucose with hexokinase/mutase/UDPG k pyrophosphatase/epimerase.
these nonnatural donor substrates is quite slow, this method is
useful for milligram-scale synthesis. The rl -3-galactosyltransferase and rwl -3GlcNAc transferase involved in the B and A
blood-group antigens, respectively, have also been studied.'" Ob1
5.2.2. Sialyltransferase (Sia T )
Several a2- 6- and ct2 - 3 sialyltransferases have been used for
oligosaccharide synthesis.121 4 - 2 1 6 1 These sialyltransferases generally transfer N-acetylneuraminic acid to either the 6- or 3position of terminal Gal or GalNAc residues. The r2-8-sialyltransferase is involved in the synthesis of a-2,8-linked polysialic
acids.[" '1 Some sialyltransferases have been shown to accept
CMP-NeuAc analogs that are derivatized at the 9-position of
the sialic acid side chain,[197-199.2181
such as those in which the
hydroxyl group at C-9 is replaced with an amino, fluoro, azido,
acetamido, or benzamido group. Analogs of the acceptors
Angew. Chew. Int. Ed. Engl. 1995, 34, 521-546
Eni-ymc\ ir Organic Synthesis (Part 2)
REVIEWS
Gal/ll 4G!cNAc and GaI,91-3GalNAc in which the acetamido
function is replaced by an azide, phthalimide, carbamate, or
pivaloyl functionality are also substrates for the enzymes.[21g1
A
recent synthesis of a G,, analog started with a disaccharyl ceraniide derikative in which the fatty acid amide group was reTo incorporate sialic acid
placed with an azide
analogs into sialosides is, however, problematic because sialyltransferases are very specific for CMP-sialic acid.
fucose and L-arabinose are transferred to Galfi-1 -4GlcNAcflO(CH,),CO,CH, at a rate of 2.3 % and 5.9 Yo.respectively, relative to ~ - f u c o s e . [ Furthermore,
~*~~
this enzyme will transfer a
fucose residue that is substituted at C-6 by a very large sterically
demanding group. In particular, a synthetic blood-group antigen was attached, and the resulting ,,ohgosacharide" was transferred to an acceptor from its GDP derivative by fucosyltransf e r a ~ e . [ ~ This
' ~ I approach has been used to alter the antigenic
properties of cell-surface glycoproteins.
5.2.3. Fucosyltransfevase (FucT)
5.2.4. N-Acetylghcosaminyittvansfevase
Fucosyltransferases are involved in the biosynthesis of many
oligosaccharide structures such as blood-group substances and
In vivo, the N-acetylglucosaminyl transferases control the
antigens associated with the cell surface and tumors. Fucosylabranching pattern of N-linked glycoproteins.[226,22'1 Each of
the enzymes transfers a BGlcNAc residue from the donor UDPtion is one of the last modifications of oligosaccharides in vivo.
GlcNAc to a mannose or other acceptor. The GlcNAc transferasSeveral fucosyltransferases have been isolated and used for in
vitro synthesis.[221-2241 F or example, xl - 3fucosyltransferase
es I-VI, which catalyze the addition of the GlcNAc residues
has been used to L-fucosylate the 3-position of the GlcNAc of
to the core pentasaccharide of asparagine glycoproteins as
N-acetyllactosamine and of sialyl x2-3N-acetyllactosamine to
outlined in Scheme 44, have been identified and characterprovide the Lewis" and sialyl Lewis" structural motifs, respectively." '. 2 2 ' 1 Several other acceptor substrates with modifications
in the GlcNAc residue can also be fucosylated (Scheme 43).L17g1
GaIfll-4Glc. Galp1-4Ghca1, and Galpl-4(5-SGlc) are all
HOOH
substrates. A similar enzyme, ctl-3(4fucosyltransferase, has
also been used for synthesis. This enzyme fucosylates either the
GlcNAc 3-position of Gal/31-4GlcNAc or the GlcNAc 4-position of Galbl -3GlcNAc to afford Lewis" or Lewis", respective0
Iy.r22'.2 2 2 1 The corresponding sialylated substrates have also VI ----H;o
v/
HO
OH11
been employed as
The Lewis" X I -4fucosyltransferase has been shown to transfer
Scheme 44. Specificity of GlcNAc transferases I-VI.
nonnatural fucose derivatives from their GDP esters. 3-Deoxy-
/
-H
Galpl-4GlcNAc
K, = 35 mM. V = 100
K,=
Galpl-4Glc
500 mM, V = 160
Galpl-4(5-S-Glc)
12 my, V = 51
K,=
J
HO
NHAc
Galpl-3GlcNAc
K, =600mM,V = 130
HO
OH
OH
Galpl-4Glucal
Km=34mM,V=l0
OH
HO
AcHN
K, = 100 mM, V = 620
OH
HO
HO
AcHN
NeuAca2-6Gal@l-4GlcNAc
K, = 70 mM, V = 13
OH
NeuAca2-3Ga1@1-4Glul
K, = 64 mM, V = 330
Galpl-4-deoxynojirimycin
IC5,, = 8 mM
Scheme 43. Substrates and inhibitors for fucosyltransferase. The reason that CaI/?l-4deoxynojirimycin is an inhibitor
may be due to a hydrogen bond between the NH group of the inhibitor and the base in the enzyme that abstracts the 3-OH
proton of the acceptor (indicated by an arrow)
Angen.. Chrni.h f .Ed. Engf. 19Y.5, 34, 521 ~ 5 4 6
ized.1226-22S1
These, as well as other GlcNAc transferases, have been
exploited for purposes of oligosaccharide synthesis.[229.2301
GIcNAc transferases have also
been utilized to transfer nonnatural
residues to oligosaccharides. In addition to transferring GlcNAc, Nacetylglucosaminyl transferase I
from human milk catalyzes the
transfer of 3-, 4-. or 6-deoxyGlcNAc from its UDP derivative
to Manxl-3(Mannl-6)ManflO(CH,),CO,CH, .Izz9]The 4- and 6deoxy-GlcNAc analogs can also
be transferred by GlcNAc transferase 11; however UDP-3-deoxyGlcNAc is not a substrate for this
enzyme.[z291 One of the many
synthetic applications of GlcNAc
transferases is the attachment
of the terminal GlcNAc of
GlcNAc/ll-4GlcNAcx
dolichyl
pyrophosphate, a substance employed in the study of oligosaccharyl t ~ - a n s f e r a s e . [ ~A~murine
'~]
kidney GlcNAc transferase was used
527
C.-H. Wong et al.
REVIEWS
in the synthesis of a hexasaccharide containing sialyl Lewis".
This enzyme catalyzes the transfer of GlcNAc to 6-OH of
Galat - ~ G I c N A c . ' ~ ~ ~ ~ ]
5.2.5. Mannosyltransfevase
Various mannosyltransferases have been shown to transfer
mannose and 4-deoxymannose from their GDP adducts to acc e p t o r ~ .The
~ ~ stl
~ ~-2mannosyltransferase
]
from yeast has been
cloned and overexpressed in E. coli (about I U L - ' ) and was
employed to transfer mannose to the 2-position of various derivatized cx-mannosides and x-mannosyl peptides to produce the
Manx1 -2Man structural
A recent report indicates
that mannosyltransferases from porcine liver accept GlcNAcpl4GlcNAc phytanyl pyrophosphate, an analog of
the natural substrate in which the phytanyl moiety replaces doli-
5.2.8. In Situ Regeneration of the Cofactor
Though analytical- and small-scale synthesis with glycosyltransferases is extremely powerful, the high cost of sugar nucleotides and product inhibition caused by the released nucleoside mono- or diphosphates present major obstacles to largescale synthesis. A simple solution to both of these problems is
the regeneration of the sugar nucleotide in situ from the released
nucleoside diphosphate. The first example of the use of such a
strategy is the galactosyltransferase-catalyzed synthesis of N-acetyllactosamine.['6 'I (Scheme 46). A catalytic amount of UDP-Gal
&
HO
HO
OH
NHAc
I
HO
5.2.6. Sucrose Synthetase
The fructose derivatives 1-azido-1 -deoxy-, 1 -deoxy-I -fluoro-,
6-deoxy-, 6-deoxy-6-fluoro-, and 4-deoxy-4-fluorofructose have
been used as glycosyl acceptors in the sucrose synthetase cata- 2Pi
lyzed synthesis of sucrose analogs (Scheme 45) .r2351 6-Deoxyand 6-deoxy-6-fluorofructose were generated in situ from the
corresponding glucose derivatives under catalysis by glucose
i s ~ m e r a s e . Because
~ ~ ~ ~ ] the reaction is reversible, the sucrose
synthetase from rice was used for the preparation and regeneration of UDP-Glc. The enzyme also accepts TDP, ADP, and
GDP.12361
-c
E'
NHAc
\
UDP
PPj
E3, E4
UTP
0
It
0.m:-
Ho&&OH
HO
-OH'
'
R
Scheme 45. Synthesis of sucrose analogs with sucrose synthetase. The system can
also be used in the regeneration of UDP-Glc from sucrose and UDP. a:
R' = R2 = O H ; b: R' = F, R2 = O H ; C : R' = O H , R 2 = E
5.2.7. Oligosaccharyltransferase
As mentioned earlier, oligosaccharyltransferase catalyzes the
transfer of an oligosaccharide consisting of two GlcNAc, nine
mannose, and three glucose units from a dolichyl pyrophosphate intermediate to an Asn residue of a nascent peptide or
protein.11371The enzyme also transfers the minimal unit GlcNAcpl -4GlcNAc from the corresponding dolichyl pyrophosphate donor or from a derivative in which the lipid component
is truncated or simplified."411 The minimal peptide that will
serve as an acceptor is the tripeptide Asn-X-Ser/Thr. Oligosaccharyltransferase has been utilized for the in vitro synthesis of
several peptides containing glycosylated Asn residues. The glycopeptides Bz-Asn(oligosacchary1)-Leu-Thr-NH,
Bz-Asn(GlcNAc,)-Leu-Thr-NH,,[2381
and Ac-Asn(GlcNAc,)-LeuThr-OCH3/NHCH3['401were synthesized as well as glycosylated cyclic peptides such as Cys-Als-Asn(GlcNAc,)-Cis-ThrSer-Ala.1'401
528
OH
Scheme 46. Galactosyltransferase-catalyzed glycosylation involving UDP-glucose
epimerase with in situ regeneration of UDP-Gal and UDP-Glc. E' = pl-4galactosyltransferase, E2 = pyruvate kinase, E3 = UDP-Glc pyrophosphorylase, E4 =
UDP-Glc epimerase. E5 = inorganic pyrophosphorylase. E6 = phosphoglucomutase.
is initially used to glycosylate GlcNAc; UDP-Gal is regenerated
from the product UDP and galactose with an enzyme-catalyzed
reaction sequence that requires stoichiometric amounts of a
phosphorylating agent. Several oligosaccharides have been prepared with routes based on this
A second regeneration system for UDP-Gal, which is based on the use of galactose1-phosphate uridyltransferase, has also been developed[2131and
has been used in the preparation of analogs such as 2'-deoxyLacNAc and 2-amino-2'-deoxy-LacNAc (Scheme 47). A third
regeneration method for UDP-Gal is based on sucrose synthetase catalyzed formation of UDP-Glc from sucrose and
UDP.[2361
In situ cofactor regeneration offers several advantages. First,
a catalytic amount of nucleoside diphosphate and a stoichiometric amount of monosaccharide can be used as starting materials rather than a stoichiometric quantity of sugar nucleotide,
thus reducing costs tremendously. Second, product inhibition
by the released NDP is minimized due to its low concentration
Angrw. Chew. Int. Ed. Engl. 1995. 34, 521 -546
Enzymes in Organic Synthesis (Part 2)
&
REVIEWS
Pgalactosidase
from Bacillus sp.
OR
HOHO
\
NHAc
I
OR
NHAc
I
HO
Hs
HO
Pyr
NHAc
UTP
I
PEP
Scheme 47. Cialnctocyltransferase-catalyzed glycosylation involving galactose-lphosphate uridyhmsferdse with in situ regeneration of UDP-Gal and UDP-Glc.
E ' = /I1 4g;ilactosyltransferase. E' = pyruvate kinase. E3 = UDP-Glc pyrophosphorylase. E3 = &ctose-l -phosphate uridyltransferase, ES = galactokiuase. Pyr
= pyruvate.
in solution. And third, isolation of the product is greatly
facilitated.
A regeneration system for CMP-NeuAc (Scheme 48) has also
been developed.['92.2 3 9 1 The UDP-Gal and CMP-NeuAc regeneration schemes can also be combined in a one-pot reaction
(Scheme 49) and applied to the synthesis of sialyl oligosaccharides.[' 791 The development of these regeneration systems, as
well as the more recent development of regeneration schemes
or U D P - G I C N A C [ ~G~ D
~P
~ -, M ~ I I , [ ' ~ ~and
] GDP-FUC,['~~]
as well as U D P - G I C U A [ " ~ ~
should facilitate the more widespread use of glycosyltransferases for oligosaccharide synthesis. One example is the recent synthesis of sialyl Lewis"
(Scheme 50).['791
2Pi
'3
NeuAc
CTP
b
Scheme 48. Enzymatic sialylation
nucleoside monophosphate kinase
pyrophosphate.
At7ge~s.Chcni.In!. Ed.
4
lactose
OH
fi-
x
UDP-Gal
UDP
Pyr
PEP+Gal
OH
NHAc
LacNAc
x
pyvp
+
NeuAc
ManNAc
Scheme 49. Synthesis of a trisaccharide with a galactosidast':dykransferase or
aalactosyltransferase/sialyltransferaseenzyme system. Enzymatic sialylation of
LacNAc in situ gave the trisaccharide that is no longer sensitive to galactosidase.
Pyr = pyruvate.
5.3. Cloning and Expression of Glycosyltransferases
and Sugar Nucleotide Synthetases
Although many glycosyltransferases catalyze similar reactions
and in many cases use the same donor substrate, there appears
to be little sequence homology among the different transferases.
There is, however, a significant homology between the same
enzymes from different species. For instance, the protein sequence of PI ~4galactosyltransferasefrom humans has an 86 %
identity with that of the enzyme from rat. The different glycosyltransferases d o exhibit some similarity in that all the cDNA sequences determined to date
encode regions consistent with a
short N-terminal tail, a hydrophobic transmem brane sequence,
a short stem sequence, and a large
C-terminal catalytic domain.[2441
\=
OH
In
addition to the membraneOH OH
bound form of the glycosyltransHO OH
ferases, soluble forms of the en%$&OH
NHAc
zymes have also been identified in
CMP
various body fluids such as the
blood, milk, and colostrum. Indeed, some of these enzymes have
been isolated and purified from
these
2 4 3 1 A comparison of the cDNA sequences of
these soluble enzymes with the Nterminal protein sequence of the
with in situ regeneration of CMP-NeuAc. El = a2-3sialyltransferase. E' =
glycosyltransferases that have
or adenylate kinase. E' = pyruvate kinase, E4 = CMP-NeuAc synthetase. ES =
been sequenced suggests that the
Engl. 1995, 34, 521 -546
pZHS0
529
REVIEWS
OH
'iGD
%
C.-H. Wong et al.
coated with the lectin. The adherent
cells were isolated and repanned for
further purification. Each of these
techniques makes use of libraries in
kinase
which
the desired
there gene.
are very
A greater
few copies
chance
of
of success may be possible if the
al-3-fucosyl
number of copies of the genes could
transferase
be amplified. The introduction in
GDP-FUC
1985 of an in vitro amplification
pymphosC H , v - p o : method
based on the polymerase
GDP-FUC
AcHN
HOoH
chain
reaction
(PCR) fulfilled this
H O S Y & : &
HO OH
OH
NHAcO
\/\\
phosphatase
need.[254.2551
Of course, PCR (and
PPI -2pi
Expression
Cassette-PCR) ,[25
Scheme 50. Enzymatic fucosylation with in situregeneration of GDP-fucose
like the hybridization screening, requires a specific knowledge of the
sequence.
Once identified, the genes are sequenced with standard procestem region has been cleaved to release the large catadures. An enlightened approach to the recloning of the gene into
lytic domain from the membrane. Presumably, this theme of
an expression vector is then used to develop an expression syssignal sequence cleavage is typical for all the glycosyltranstem. This recloning into expression systems has been performed
ferase~.[~~~]
with only a few of the glycosyltransferases. Toghrol et al. have
The amount of the glycosyltransferase that can be isolated
inserted the mouse liver glucuronyltransferase gene into the
from a natural source is often limited by the low concentrations
yeast vector pEVPll and expressed the enzyme in Succhuof these enzymes present in most tissues and body fluids. The
romyces c e r e ~ i s i u eThe
. ~ ~rat
~ ~ ~liver glucuronyltransferase, on
purification of glycosyltransferases is further complicated by the
the other hand, has been expressed in COS cells with the SV40
relative instability of this group of enzymes.[' For this reason,
Bovine PI -4galactosyltransferase was cloned
a great deal of interest has been directed toward the cloning of
analogously.[2471
A noteworthy approach toward the expression
the glycosyltransferase genes into convenient expression sysof the glycosyltransferase in E. coli has been developed to obtain
tems. The general strategy involved in this procedure is outlined
human /I1 - 4 g a I a c t o ~ y l t r a n s f e r a s e . ~
A~unique
~ ~ ~ RsrII restricin Scheme 51 a. The glycosyltransferase gene must first be idention site in the galactosyltransferase gene allowed the dissection
tified and isolated from the mRNA pool by cloning of the
of the sequence at the location of signal peptidase cleavage. The
cDNA to make a cDNA library. This library is then screened
cohesive terminus was digested with Klenow fragment (the large
to identify the glycosyltransferase gene of interest among apsubunit of the D N A polymerase I, which has a 3' -+ 5' exonucleproximately lo6 different sequences present in the library. Once
ase activity and cleaves the remaining bases at the restriction
identified, the gene is sequenced and a more complete cloning
site), and the blunt end ligated to ~ I N - I I I - o ~ ~ at
A EcoRI
,[~~~~
strategy is developed in order to incorporate the gene into
site that had also been treated with a Klenow fragment. Thus
an expression vector. This laborious path has successfully been
the code was generated for a fusion protein of the soluble
practiced by several groups.[5.2 4 4 - 2 5 3 , 2 5 6 - 2 6 5 1 From various
form of galactosyltransferase with the ompA signal sequence
tissue sources cDNA is recovered, and, for example, the double
(ompA = outer membrane protein A, an E. coli protein that
stranded cDNA is ligated in h phage through a convenient linkcontrols secretion through its cleavable N-terminal signal seer and packed into bacteriophages. The bacteriophages are then
quence). Transcription and translation of this sequence in E. coli
plated onto a lawn of E. coli and screened for the desired gene
produced an active enzyme that was released into the periplasor gene product. The identification of the glycosyltransferase
mic space. Purification and N-terminal sequencing of the engene has most frequently been achieved by the hybridization of
zyme verified the expresson of the soluble form of galactosylthe gene to specific radiolabeled D N A probes.r245- 2481 Screentransferase with an added N-terminal tail comprising three
ing in this manner requires a previous knowledge of the gene
amino acids. The kinetic parameters of this enzyme appear to be
sequence, information that in some cases may be obtained by
identical to the isolated native enzyme. This system, however,
extrapolation from a partial protein sequence or from the D N A
only produced a very small amount of enzyme ( <0.014 U L-').
sequence of the glycosyltransferase from a related source. Two
Though a 30-fold increase of productivity has been achieved
other approaches have been used to screen glycosyltransferase
with a modified E. coli
the level of productivity is still
cDNA libraries, both requiring successful transcription and
too low to be of practical synthetic value. More practical exprestranslation of the gene product. In the cloning of the a2-6sion systems are perhaps those based on the baculo virus['791
sialyltransferase from rat liver, Weinstein et al. used polyclonal
and yeast.[2581
antibodies raised against the purified enzyme to screen the h
To date, very few glycosyltransferases have been cloned,
phages.[2531The other approach used by Larsen et al. alleviated
expressed, and produced in quantities sufficient for enzymatic
the need for a previous knowledge of the sequence.[2511This
synthesis.[2661 However, given the advantages of enzymatic
method made use of the specificity of a lectin that recognizes the
synthesis of oligosaccharides over traditional schemes, research
surface-expressed glycoconjugate product of the a1 - 3galactosinto the overexpression of glycosyltransferases will continue
yltransferase. The transfected cells were then panned in dishes
Pyr
sialyl LeX
530
Angew. Chem. Int. Ed. Engl. 1995. 34, 521 -546
Enzymes in Organic Synthesis (Part 2)
REVIEWS
target sequence
5'
3'
primer CMP5
b)
5-ATATTGAAlTCTAAACTAGTCGCCAAGGAGACAGTCATM>A&A
-Err
5'
3'
1. denaturation
"C
at
. 96
~.
start
primer CMP3
5 ' 6 C G TCTAGACTAlTAAGAACCGTAGTCCGGAACGTCGTACGGG
2. annealing of primers
at 55 "C
%3--
decapep*
TAmAACAATCTCGGCTAlT(;
gene c-lsrmiwa
3
5'
\
/
-
f
ShinsDabamosasnce
s
CMP-NeuAcsynIhetase
structureaene f1.3kb)
restlictlon sites,
start and stop,
a 3-x
primer extension
with Taq polymerase
SlgMlS
at 72 OC
3'
5'
I
activationof
resfridon site
3
-5'
*
4
'
3
c)..
expression cassette
antibody
\5
C'
s..
0-P-0
d)
5' ATAlTGmCTAAACTAGTCGCmAGGAG ACAGT
CATAcGAGAACAAACG3'
mmme primer
SGCGCTCTAGACTATTAllTAACAATCTCCGCTAm3
Xba I
alkaline
phosphatase
C-terminus
CMPSIL-1 plasmid
(digestedwim EcoRl and Xbal)
PCR amplification and digestion
by restriclion enzyme
ECOR I
*
1
5
tagged CMP-NeuAc
synmetase gene
blue
- _
49
CMP-NeuAc Synthetase
Scheme 51. a) General strategy for the overexpression of CMP-NeuAc synthetdse. b) Primer sequences
containing an additional decapeptide tag sequence and restriction sites for the amplification of the
CMP-NeuAc synthetase gene are prepared for the construction of a phdgemid containing the tagget
CMP-NeuAc synthetdse gene. c) Selection of a clone using an antibody against the decapeptide conjugdted with an alkaline phosphatase. The blue positive clones were picked for overexpression. d) RemOVdi of
the tag of new designed primers by PCR, exchange of the tagged gene with the one without the tag, and
overexpression of the gene under Lac2 promotor (50 mgL-' from E. coli).
A n p w . C'liern.
Itii.
Ed. En$ 1995, 34. 521 -546
531
REVIEWS
C.-H. Wong et al.
to flourish. Schemes 51 b-51 d describe the procedures used
in the overexpression of CMP-sialic acid synthetase in E. coli
(-5OmgL-').
now
OH
HO
potato
phosphorylase
6H
sucrose
phosphorylase
5.4. Non-Leloir Glycosyltransferases: Transfer of
Glycosyl Donors from Glycosyl Phosphates and Glycosides
+
OH
no
OH
Oligosaccharides can also be prepared using non-Leloir glycosyltransferases. Phosphorolysis of polysaccharides is catalyzed by a group of these enzymes called glucan phosphorylases.
The reaction is reversible and can be used in the synthesis of
oligo- or polysaccharides. Two particularly important examples
are the syntheses of sucrose and trehalose, catalyzed by
sucrose p h o ~ p h o r y l a s e [ ~ and
~ ' ] trehalose p h o ~ p h o r y l a s e [ ~ ~ ' ]
(Scheme 52). Examples of other enzymes of this class are those
involved in synthesis of dextrans and levans.[2681
b
primer
HO
-Primer
n
r
I
1
/OH
I
sucrose, Pi
SUCrOse
phosphotyiase
-
R'-
/OH
potato
phosphorylase
(OH
OH
". .-.,.--
H a
phosphorylase
OH
HO
6H
OOH
a' *
HOO-PO,'
-+
HO
OH
trehalose
O
*"H
L
~
HO
phosphorylase
n>6
Scheme 53. Phosphorylase-catalyzed polysaccharide synthesis.
no
Cyclodextrin ~1l-4glucosyltransferase [EC 2.4.1.191 from
Bacillus macerans catalyzes the cyclization of oligomaltose to
form CI-,p-, and 6-cyclodextrin, as well as the transfer of sugars
from cyclodextrin to an acceptor to form oligosaccharides.[271,2 7 2 1 The enzyme can also transform a-glucosyl fluoride into a mixture of CI-and 8-cyclodextrins and maltooligomers in almost equal amounts.[2731When immobilized on
a silica gel support that was functionalized with glutaraldehyde,
the enzyme was very stable, and no loss of activity was observed
after four weeks at 4 "C. With an appropriate choice of acceptor,
this type of enzymatic catalysis may provide a new route to
cyclodextrin analogs and novel oligosaccharides.
Nonnatural sugar acceptors that are structurally similar to
glucose are also substrates for cyclodextrin glucosyl transferase.
With deoxynojirimycin as acceptor and cyclodextrin as donor,
an oligoglucosyl deoxynojirimycin was produced, which was
subsequently hydrolyzed by glucoamylase to give 4-O-CI-D-glUcopyranosyl deoxynojirimycin in approximately 60 % yield
(Scheme 54).[2741A number of N-substituted derivatives of deoxynojirimycin were also good substrates for the transferase,
OH
Scheme 52. Phosphorylase-catalyzed synthesis of sucrose and trehalose.
Non-Leloir transferases have also been used to synthesize a
variety of polysaccharides. The synthesis of modified polysaccharides may provide materials with more desirable physical
and biological properties than their natural counterparts. Approaches to influencing the characteristics of polymers include
the control of genes encoding the enzymes responsible for their
production, regulation of the activity of these enzymes, or the
influence of their in vivo synthesis.[2691Potato phosphorylase
[EC 2.4.1.11 has been used in vitro to prepare maltose oligomer,[159]and a family of linear and star- and comb-shaped polymer~.[~'']
Improvement of this system has recently been accomplished
with the use of a coupled enzyme system in which glucose-l-phosphate is generated in situ from sucrose and inorganic phosphate
catalyzed by sucrose phosphorylase.[1601The inorganic phosphate liberated by potato phosphorylase is used by sucrose phosphorylase to drive the formation of polymer, thereby increasing
the yield. This coupled enzyme system also allows for regulation
of the molecular weight of the polysaccharide product by control of the concentration of the primer. Nonnatural primers
bearing functional groups can also be used to prepare tailormade polysaccharides for further manipulation, for example,
for attachment to proteins or other compounds (Scheme 5 3 ) .
532
t
OH
Scheme 54. Synthesis of glycosyl N-alkyldeoxynojirimycin derivatives.
Angew. Chem. I n f . Ed. Engl. 1995,34, 521 -546
Enzymes in Organic Synthesis (Part 2)
REVIEWS
and the products can be degraded with glucoamylase to glucosyl
aza sugars. One of these glucosyl aza sugars, 4-0-a-D-glucopyranosyl-N-methyldeoxynojirimycin, was reported to be a potent inhibitor of glucosidase.
A key step in the biosynthesis of lipid A is the glycosylation
of the 6-position of a 2,3-diacylglucosamine-l -phosphate by the
donor UDP-2.3-diacylglucosamine to produce a lipid A precursor. This transforination is catalyzed by the enzyme lipid A
synthetase. which has been cloned and overe~pressed!~'~~
It has
been used for the in vitro synthesis of the lipid A precursor and
analogs thereof. Some examples include C - g l y c o ~ i d e [ and
~~~'
phosphate["'1 analogs (Scheme 55).
/OH
+
6
:jX
IipidA
HO0
0';
synthetase
HO
CllH23
C1,Hn
/OH
X = OPOi- (lipid A precursor)
X = CHZCO;
x = PoiScheme 55. Synthesis of lipid A analogs with lipid A synthetase
5.5. Glycosidases and Transglycosidases
Glycosides may be synthesized with glycosidase catalysts under either equilibrium or kinetically controlled conditions.['31
Though quite easy to perform, the equilibrium approach provides poor yields generally not exceeding 15%. Also, the low
yield of the desired product and the formation of side products
generally make purification difficult.
Kinetically controlled synthesis relies on the trapping of a reactive intermediate generated from an activated glycosyl donor with
exogenous nucleophiles to form a new glycoside bond.[131Suitable glycosyl donors for this transglycosylation reaction include
di- or oligosaccharides, aryl glycosides, and glycosyl fluorides.
This reaction must be carefully monitored, and stopped when the
glycosyl donor is consumed in order to minimize glycoside hydrolysis.
Yields in kinetically controlled synthesis generally range from
20 to 40%. Although addition of organic solvent might be expected to increase product yields, as in the equilibrium controlled approach, this effect generally has not been observed.
The kinetically controlled approach has primarily been applied to the glycosidases that give products with retention of
configuration. However, an inverting glycosidase has been used
with glycosyl fluorides as glycosyl donors to afford products
having the configuration at the anomeric position opposite to
that of the donor.["'] For example, the r,a-linkage of X-D-ghcopyranosyl-a-D-xylopyranoside has been prepared utilizing pglucosylfluoride and a - t r e h a I a ~ e . [ ~ " ~
In general, the primary hydroxyl group of the acceptor reacts
preferentially over secondary hydroxyl groups to give a 1,6-glycosidic linkage. Some control of selectivity has been demonstrated
by the selection of an appropriate dono/acceptor combination.[2801For example, the reaction catalyzed by a-galactosidase
of r-Gal-OPh-p-NO, with cc-Gal-OMe (or ally1 r-galactoside)
and 8-Gal-OMe forms predominantly x-1,3 and n-1.6 linkages,
respectively.[z80".b1 The configuration of the anomeric center of
the acceptor controls, to some extent, the position of glycosylation. This situation was also observed in the /l-N-acetylgalactosaminidase reactions.[280c1cc-gal-OPh-p-NO, acting both as
donor and acceptor, forms referentially the %-I,3 linkage whereas the artho nitrophenyl glycoside reacts in a similar fashion to
form predominantly the cc-I ,2-linkage.[2s0",b1 With 8-galactosidase, mainly the P-1,3-linked disaccharides were formed when
benzyl or ally1 8-galactoside was used as
2 8 0 a , b1 The
use of glycals as acceptors has also been employed as a means of
controlling selectivity,[281]giving the 1,3-linked glycoside as the
major product when the 6-position was free or blocked.
One can also use glycosidases from different species to control
the regioselectivity. For example, the 8-galactosidase from
testes catalyzes the formation of Gal/ll-3GlcNAc from lactose
or Galbl -3GlcNAc/BEt[2821 from either GlcNAc or GlcNAcPSEt. Minor products were then hydrolyzed by E. coli Pgalactosidase, which preferentially hydrolyzes /l-1,6-linked
galactosyl residue. The overall yield of the /l-1.3-linked disaccharides was around 10-20 YO.Syntheses of polysaccharides
based on kinetically controlled glycosidase reactions have been
accomplished, as examplified by the cellulase-catalyzed polymerization of P-cellobiosyl fluoride to form cellulose. with a degree
of polymerization of greater than 22.[283a1The 11-galactosidase
from Bacillus circulans was used in large-scale synthesis of N a c e t y l l a c t ~ s a m i n e [from
~ ~ ~ lactose
~~
and GlcNAc. When this
reaction was coupled with the sialyltransferase reaction in situ,
a sialyl lactosamine was prepared[2R3c1
(see Scheme 49), thus
preventing the secondary hydrolysis of the galactosidase
product.
Glycosyl transfer to non-sugar acceptors has also been
demonstrated. These reactions are especially interesting with
chiral, racemic, or meso alcohols, as some degree of diastereoselectivity due to the asymmetric microenvironment of an enzyme
active site might be expected. Indeed moderate to exceptional
diastereoseIectivitie~[~~]
have been observed.
Transglycosidases are related to glycosidases in that they
cleave glycoside bonds, but they differ in that they usually transfer the glycosyl moiety to another acceptor with a minimal
amount of hydrolysis. Transglycosidases are thus useful catalysts for glycosylation. For example, a 8-fructofuranosidase
from Anlhrrobacter sp. K-1 transfers fructose from sucrose to
the 6-position of the glucose residues of stevioside and rubus o ~ i d e . Also,
[ ~ ~ a~ sucrase
~
from Bacillus subtilis catalyzes the
reversible transfer of fructose from sucrose to the 6-hydroxyl of
a fructose unit at the nonreducing end of a levan ~hain.['*~1
Several nonnatural sucrose derivatives have been prepared by
taking advantage of this process.[2861
A transsialidase from Trypanosoma cruzi has been shown to
transfer sialic acid reversibly to and from the 3-position of terminal @-Galresidues.[2871
Chains terminating in cc-linked galactose
are not substrates. A number of oligosaccharides containing the
533
C.-H. Wong et al.
REVIEWS
NeuAca2-3Gal/" substructure have been synthesized with this
t r a n ~ s i a l i d a s e . [Moreover,
~~~~
this enzyme has been shown to
resialylate the terminal galactose units of the cell-surface glycoproteins and glycolipids of sialidase-treated erythrocytes.[28y1
and a g a l a c t o ~ i d e . [Thus,
~ ~ ~ l the 7:cruzi transsialidase potentially provides a useful alternative to a2- 3sialytransferase.
5.6. Synthesis of Nucleosides
Nucleosides and their derivatives are ubiquitous in nature
and are involved in a myriad of biochemical phenomena, most
notably the storage and transfer of genetic information. Interest
in this class of compounds has been stimulated by the efficacy of
certain nucleosides as antipara~itic['~~]
(for example, 1-/"-D-nboallopurinol ribofuranosyl-I H-pyrazolo[3,4-d]pyrimidin-4-one,
side) and antiviral (for example, 3'-azido-3'-deoxythymidine,
azidovudine, AZT) agents.[2y2
- 2 9 3 1 Nucleosides have traditionally been prepared by various chemical methods.[2941These are
multistep syntheses requiring protection and deprotection steps
and glycosyl activation. Other problems include control of
anomeric configuration, especially when preparing B-D-arabinofuranosyl- and 2'-deoxy-0-D-ribofuranosylnucleosides,
and
regiospecific C- N glycoside formation when there are several
possible nucleophilic groups in the purine or pyrimidine base.
Two enzymatic methods have been used in the synthesis of
nucleosides (Scheme 56a). Most enzymatic preparations of
both natural and nonnatural nucleosides have been reported
with nucleoside phosphorylases as
These enzymes
catalyze the reversible formation of a purine or pyrimidine nucleoside and inorganic phosphate from ribose-I-phosphate
a)
HO
HO
X
E'
HO
HO
X
E2
c--
HO
X
HO
X=H,OH
+.
X
0GN
~1
N".
\
HO
OH
H
H3N' )+4t4
HO
OH
Scheme56. a)Two strategies for enzyme-catalyzed nucleoside synthesis. E' =
nucleoside phosphorylase, EZ = transribosylase. Both E' and EZcatalyze reversible
reactions. b) When 7-N-methylguanosine
is the donor, the reaction catalyzed by E'
. is irreversible.
534
(R-1-P) and a purine or pyrimidine base. The equilibrium
strongly favors nucleoside formation. Nucleoside synthesis has
relied on the transfer of the ribose moiety of a readily available
nucleoside to a different purine or pyrimidine base or analogs
through the intermediacy of R-1-P. This work has been done
primarily with isolated
but whole cells have also
been employed in a few cases.[2971The effects of deleterious
hydrolases present in whole cells could be largely suppressed by
conducting the reactions at 60 "C, a temperature at which the
nucleoside phosphorylases maintain more than 70 YOof their
original activity for three to five days.[2971
Two basic strategies have generally been employed. The first
involves isolation of R-1-P, which is prepared in good yield from
a nucleoside in the presence of a high concentration of phosphate.[298]The isolated R-1-P is then used as the glycosyl donor
in an enzymatic coupling reaction with added purine or pyrimidine bases or analogs. The second strategy involves a one-pot
exchange of one nucleoside base for another in the presence of
a catalytic amount of inorganic phosphate without isolation of
R-1 -P. The first strategy (isolation of R-I -P) is the most general,
and almost any heterocycle that is a substrate for a nucleoside
phosphorylase can be glycosylated by this method. The second
strategy (in situ generation of R-1-P) is more limited. At best, it
results in formation of an equilibium mixture of the substrate
and the product nucleosides, from which the product must be
isolated. In less favorable cases, the natural purine or pyrimidine
base released from the glycosyl donor may be a potent competitive inhibitor with respect to the purine or pyrimidine analog,
for which the enzyme has lower affinity. For example, because
of competitive inhibition by hypoxanthine ( K , = 5.6 mM),
1,2,4-triazole-3-carboxamide
(TCA, the aglycon component of
virazole, K , = 167 mM) cannot be glycosylated if inosine is the
ribosyl donor and purine nucleoside phosphorylase (PNPase)
the catalyst.[29y1
It was, however, possible to synthesize virazole by
isolating of R-1-P and subsequently using it as the ribosyl
donor.[29y1An alternative way to circumvent the inhibition
problems is to employ a pyrimidine nucleoside as the glycosyl
donor and a purine (or purine analog) as the acceptor, since the
released pyrimidine base does not inhibit the purine nucleoside
p h o s p h o r y l a ~ e . By
[ ~ ~this
~ ~method, both pyrimidine nucleoside
phosphorylase and purine nucleoside phosphorylase are required.
Recently, direct purine-to-purine exchange reactions have been
conducted without isolation of R-l -P with activated purine
derivatives as the ribosyl donors.[301]The activated purine derivatives were prepared by 7-N-methylation of inosine and guanosine
to provide derivatives that are excellent substrates for phosphorolytic cleavage by PNPase. The cleaved 7-N-methylpurines do
not show any measurable product inhibition, and the equilibrium of the reaction greatly favors the product. The effectiveness
of this approach was demonstrated by the one-pot synthesis of
virazole from 3,2,4-triazole-3-carboxamideand 7-N-methyl
guanosine (Scheme 56 b).
The nucleoside phosphorylases accept a wide range of
nucleoside analogs modified in both the base and glycosyl
components, as substrates. Most of these reactions have
been carried out in one step without isolation of the sugar
phosphate, which has, however* been demonstrated to be an
intermediate.
Angew. Chem. In[. Ed. Engl. 1995, 34, 521 -546
Enzymes in Organic Synthesis (Part 2)
The use of nonnatural bases has met with much success with
both natural and nonnatural glycosyl donors. However, a few
limitations have been observed. For example, in the purine nucleoside phosphorylase catalyzed synthesis of certain imidazole[4,5-c]pyridine nucleosides (3-deazapurine nucleosides) , the
normally observed regiospecificity was lost, and a mixture of
N-I and N-3 glycosylated products was isolated. This problem
was not encountered with purines, which retain the nitrogen at
the 3-position: 2'-deoxyribosylation of unsubstituted purine
Apparently either the
gave only the N-9 glycosyl
N-3 nitrogen atom of the purine base or an appropriate substituent at C-6 is necessary for proper orientation of the base for
regiospecific glycosylation.
Sugar-modified nucleosides can be synthesized from
glycosyl donors prepared by chemical modification of readily
available nucleosides such as uridine and cytidine. Good yields
of ara bi no- '1 and 2'-amino-2-deoxyribonucleosides
have
also been obtained enzymatically, although that of the
enzymatic synthesis of 3'-amino-2,3'-dideoxyribonucleosides
is
This low yield may be due to competition of the two
pyrimidine heterocycles for enzyme binding or to an overall
decreased reaction rate caused by alteration of the 3'-position,
which may be important in substrate binding by the en,yme~l"o"l
N-Transribosylases have been employed in the synthesis of
nucleoside analogs.[2y5.3041 Two classes of transribosylases have
been identified: type 1enzymes catalyze the transfer of the sugar
moiety between two purine bases, and type I1 catalyze the transfer between any two bases. Like nucleoside phosphorylases,
transribosylases are stereospecific for the p-anomer of the nucleoside product. Thymidine and 2'-deoxycytidine are the best
glycosyl donors, and a reasonable amount of variation in the
acceptor bases is tolerated. The transribosylase from L. leichmanii has been used to prepare 2-chloroadenosine, an antileukemic and immunosuppressive n u c l e ~ s i d e . [ ~ ~ ~ " ~
5.7. Silyl Lewis" and Related Structures
Carbohydrate-mediated cell adhesion is a n important event,
which can be initiated by tissue injury or infection and is involved in metastasis.['01 One such adhesion process that was
recently discovered is the interaction between the glycoprotein
E-selectin (formerly called endothelial leukocyte adhesion molecule or ELAM-I), which is expressed on the surface of endothelial cells during inflammation and an oligosaccharide structure
displayed on the surface of neutrophils. The ligand recognized
by E-selectin has been identified to be the tetrasaccharide sialyl
LewisX(SLe"), which is present a t the terminus of glycolipids
displayed on the surface of n e u t r o p h i l ~ . [The
~ ~ ~adhesion
I
process[loI is stimulated by signaling molecules (cytokines) o r other
inflammatory factors (for example, toxins, lipopolysaccharides,
leukotrienes) that induce the production of E-selectins. After
binding to the endothelial cells, the white blood cells roll along
the surface of the endothelium. Further adhesion occurs by
protein-protein interactions, mediated by integrins on white
blood cells and a protein ligand containing the R G D (Arg-GlyAsp) sequence (called intercellular adhesion molecule-1, ICAM1 ) on endothelial cells. The white blood cells are then able to
Anpii
CIwin I n i Ed Enpl 1995, 34, 521 -546
RRIIEWS
squeeze through gaps between endothelial cells and enter the
adjacent tissue to help repair injury (Scheme 57).
If too many white blood cells are recruited to the site of injury,
normal cells can also be destroyed. This can occur in the condition of septic shock, in chronic inflammatory diseases such as
psoriasis and rheumatoid arthritis, and in the reperfusion tissue
A.
B.
injury
E-selectins
C.
D.
E.
F.
w'o
CAM-1
'.. integrin
Scheme 57. Inflainmation of tissue involves E-selectin-mediated cell adhesion.
When tissue injury occurs (A), cytokines are released to stimulate synthesis of
E-selection (B). Adhesion then occurs through interaction of E-selectins and SLe"
on the surface of leukocytes (C, D). As leukocytes roll along the surface ofendothelial cells, further adhesion occurs through interaction between integrins on leukocytes and the ligand o n the endothelial cells that contain the sequence Arg-Gly-Asp
(RGD) (called [CAM-I) (E). The leukocytes reach the injury site through small
openings between the endothelial cells (F).
injury that occurs following heart attack, stroke, and organ
transplant. High levels of sialyl Lewis" have also been found on
the surfaces of certain tumor and cancer cells (for instance,
colon and lung cancer cells), which suggests that cancer cells
may exploit this phenomenon of adhesion to metastasize after
entering the blood stream."'] In addition to E-selectin, two
other carbohydrate-binding proteins, P- and L-selectin, also recognize sialylated ligands in cell-adhesion processes.
The discovery that the ligand for E-selectin is sialyl Lewis"[3051
provides new impetus for the development of therapeutic agents
in the treatment of inflammation-related diseases and cancers.
Since sialyl Lewis" in solution competes with the white blood
cells for binding to E-selectin, thus inhibiting the adhesion process, it may be useful as an anti-inflammatory to anticancer
agent.['01 On the other hand, understanding the active conformation of sialyl LewisXrecognized by E-selectin may also lead to
the development of simple and easy-to-make inhibitors that are
analogs or non-carbohydrate mimetics of sialyl Lewis" with better bioavialability. Another approach is to inhibit the enzymes
(such as the a1 -3fucosyltransferase) associated with the biosynthesis of sialyl Lewis".
Considerable research has been directed toward the chemical
synthesis of sialyl Lewisx;r306-3091
these syntheses tyically require multiple protection and deprotection steps. making largescale production difficult. Enzymatic syntheses of sialyl Lewis"
based o n glycosyltransferases proceed regio- and stereoselectively in aqueous solution, without the need for protecting
groups.['7y~3'0-3"1With the advances in recombinant DNA
technology for the cloning and overproduction of enzymes and
535
C.-H. Wong et al.
REVIEWS
known, further experiments with defined multivalent ligands or
the development of sugar nucleotide regeneration, sialyl Lewis"
mimetics with appropriate spacers should clarify the nature of
can now be produced in large quantities. (Cytel Co., San Diego,
E-selectin-mediated cell adhesion and may suggest new approachCalifornia, is producing SLe" in kg quantities as a drug candies to the discovery and development of antiadhesion molecules.
date for the treatment of reperfusion tissue injury.)
The enzymatic method has also been used for the synthesis of
Of particular interest is the development of non-carbohydrate
molecules to mimic the topography of the sialyl Lewis" binding
['3C]Gal-1 -labeled sialyl
which was used for conforregion composed of carboxylate, galactose. and fucose.
mational analysis with N M R spectroscopy. In addition, sialyl
Lewis" (SLe"),[L79*
3121 SLe" glyca1,[1791and L e " - 3 ' - ~ u l f a t e [ ~ ' ~ ~ The glycosidic bonds in SLe", SLe" glycal, and Le" can be
formed enzymatically with glycosyltransferases. Introduction of
are also c ~ n f o r m a t i o n a l l y [ similar
' ~ ~ ~ to SLe" and biologically
a sulfate group to the 3'-position of Le" with a sulfate-transfer
as active (Scheme 58). The results of these studies indicate that
enzyme coupled with regeneration of PAPS for large-scale prothe active binding domain of sialyl Lewis" consists primarily of
cess, however, has not been developed. The synthesis of the
the galactose, fucose, and the carboxylate group of the sialic
bivalent oligosaccharide ligand containing two SLe" groups
acid. The methyl group of the fucose is not essential, as the
was very straightforward, starting with a chemically prepared
fucose residue can be replaced with a r a b i n ~ s e . [ ~ The
' ~ ' three
trisaccharide followed by three glycosyltransferase reactions
hydroxyl groups of the fucose, however, are required.[3141
(Scheme 59) .[3 14] Since the glycosyltransferase required for the
GkNAcp-0
Galf3-1-4GlcNA@-O
0
UDP-Gal (2equiv)
b
OH
"A
OH
HO
AcHN
bivalent sialyl LeX
ligand
sialyl Lex
HO
I OH
Scheme 59. Synthesis of a bivalent ligdnd containing two SLex groups. Two sugar
residues are are incorporated in each enzymatic step.
HO
sialyi LeXglycal
active binding site
Scheme 58. Ligands for E-selectin and the minimum active binding site. Sialyl
Lewis", sialyl Lewis'. Lex-3'-O-sulfate, and SLe" glycal exhibit almost the same
binding activity to E-selectin in vitro (lCs0 = 1 - 2 mM).The three OH groups of Fuc
are essential, while the CH, group i s not required. Ion-spray mass analysis using the
collision-induced decomposition technique indicated that the CaZt ions may be
located in the Lewisx moiety.
Further studies on the nature of ligand recognition by Eselectin revealed that a bivalent ligand consisting of two sialyl
Lewis" groups anchored onto ethyl [I-galactoside by 8-1,3- and
8-1.6-linkages was five times better at blocking adhesion
(IC5,, = 0.4 mM) than monovalent SLe". This observation suggests the possibility of a multivalent ligand-receptor interact i ~ n . [ ~All
' ~these
'
activities were measured in vitro based on an
enzyme-linked immunosorbant assay (ELISA). The level of in
vivo activity of sialyl Lewis" is, however, not very clear. The
IC,,for protecting lung injury in rats appears to be, for example,
about 1 pM.[3151While the structure of the E-selectin-ligand
complex and the role of Ca2 in the ligand binding are still not
+
536
synthesis of the GalPl-3GlcNAc core unit of SLe" is not readily
available, this disaccharide was prepared chemoenzymatically
from glucal by subtilisin-catalyzed acetylation followed by
/I-galactosidase-catalyzed galactosylation and azidonitration
(Scheme 60) .11501 The remaining glycosidic bonds were then
formed with appropriate glycosyltransferases. Another SLe"
analog derived from SLe" glycal was prepared by chloroperoxidase-catalyzed halohydration (Scheme 61) .I3 61 This halogenated SLe" may have the same activity as SLe", since the active
conformation should not change. Although the halohydration
is, in general, not stereoselective, the enzymatic halohydration
of sialic acid glycal is regio- and stereoselective, giving 3-bromosialic acid with the bromide in the axial position. Presumably
a bound hypobromite intermediate is involved.
It is noteworthy that though enzymatic synthesis of oligosaccharides is potentially useful for large-scale processes, the method
is quite limited, as many oligosaccharide analogs cannot be
prepared enzymatically. Synthesis of modified oligosaccharides
thus requires the use of chemical g l y c ~ s y l a t i o n .'1[ ~The
~ scope
and mechanism of recently developed glycosyl phosphites in
g l y c ~ s y l a t i o nhas
~ ~ ~been
~ ~ examined in detail,r3191and the
Angrn. C k m . Inr. Ed. Engl. 1995, 34, 521 -546
REVIEWS
Enzymes in Organic Synthesis (Part 2)
sublilisin
HO
*
P-galactosidase
*
HO
@OAc
Hoqoo
90 %
DMF
~d
HO
AcHN
H
HO
I
o
chloroperoxidase
OH
HO
2. NaN3. CAN*
&
50 %
OH
HO
3. [HI
4. Ac20
OH
H202,KBr
pH=3
OH
NHAc
OH
5. saponification
OH
a2-3SiaT
al-3l4FU~T
OH
CMP-N~~A~)
regeneration
P
GDP-FUC
NHAC
HO
chloroperoxidase
H202,KBr
AcHN
*O*C02CH3
*
H o M C O z C H s
AcHN
pH=3
OH OH
OHoH
B~
90 %
H
O
W
AcHN
I
OHoH
HO
B~
2:3 74%
sialyl Lea
HO
HO
OH
OH
NHAc
Scheme 60. Sgnrhesis of sialyl Lewisa with galactosidase/sialyltransferase:fucosyltransferase. Glucal was regioselectively acylated in DMF in the presence of 3 %
aqueoua phosphate buffer. CAN = cerium ammonium nitrate.
Scheme 61. Chloroperoxidase-catalyzedbromination of glycals.
method appears to be particularly effective for sialylation
(Scheme 62) and fucosylation. Mechanistic study of the glycosylation reactions indicates that triflic acid is involved in
the activation when trimethylsilyl trifluoromethanesulfonate
(TMSOTf) is used in catalytic amounts (Scheme 63).[3191
TMSOTf
RO-TMS
\
A
AcHN
c o
S
AcO OAC
C
O
~
C
(BnO)2PNEtp
H
~
tetrazole
THF
Ad3 OAC
-0Tf
It
H
O=P(OBn),
ROH
TMSOTf
CH3CN
4 0 OC
A
c
O
AnRcii
Chwn. Inr Ed. Enyl. 1995. 34, 521 -546
Scheme 6 3 . Proposed mechanism of glycosylation with phosphites.
AcO OAC
R = BnO
Scheme 6 2 . Chemical sialylation with a sialyl phosphite.
:C
AcHN
40-80 %
HO
s
NHAc
6. Glycopeptide Synthesis and
Glycoprotein Remodeling
A number of the proteins of interest as human pharmaceuticals
(tissue plasminogen activator, juvenile human growth hormone,
CD4) are glycoproteins. There is substantial interest in developing methods that will permit modification of oligosaccharide
537
C.-H. Wong et al.
REVIEWS
structures on these glycoproteins by removing and adding sugar
units (“remodeling”) and in making new types of proteinoligosaccharide conjugates.[3203
3 2 1 1 The motivation for these
efforts is the hope and that modification of the sugar components of naturally occurring or nonnatural glycoproteins might
increase serum lifetime and the solubility of the drug, decrease
antigenicity, and promote uptake by target cells and tissues.
Enzymes are feasible catalysts for manipulating the oligosaccharide content and structure of glycoproteins. The delicacy and
polyfunctional character of proteins and the requirement for
high selectivity in their modification indicate that classical synthetic methods will be of limited use. The major obstacle to the
widespread use of enzymes for glycoprotein remodeling and
Cbz-Ala-Ser-OCH3
I
‘z-0
-H
Cbz-Ala-Sar-Gly-Ala-NH2
z wI ~
thlosubtillsin
+ H2N-Gly-Ala-Nb
mutant
pH = 9.0.50 OC
OH
A useful method for glycopeptide synthesis incorporates glycosylamino acids into oligopeptides chemically[3z21
or enzymati ~ a l l y and
[ ~ subsequently
~~~
introduces additional sugars with
glycosyltransferases. Enzymatic formation of peptide and glycoside bonds is quite effective, because both procedures can be
carried out in aqueous solution, thus minimizing protection/deprotection steps in peptide synthesis. Glycosyl amino acids can
be used as the P,, P3, Pi, and Pi residue in subtilisin-catalyzedglycopeptide segment condensation. Using a thermostable variant developed by site-directed mutagenesis in which the activesite Ser is converted to Cys, the enzymatic coupling of glycopeptide segments can be carried out efficiently at 60°C in
aqueous solution (Scheme 64) .[3231 The enzyme prefers aminol-
OH
60 %
~,Asp-Ala-Ser-9CH3
no
HO
NH
NHAc
Ho&
HO
??
OH
J
cofactor
regeneration
Pyr,PI
Phe-Leu-NHp
subtillsin BPN’ 8397
Ho
generation is the lack of availability of many of the glycosyl
transferases that are plausible candidates and the uncertainty
whether glycosyltransferases that probably act on unfolded or
partially folded proteins in vivo will be active at the surface of
a completely folded protein.
a1
aminolysis
OH
I-
PEP-ip H o H ~ o ~ T * * ” , - ~ - “ “
Gal-,
HO
HO
NH
I.,
UDP
1. H2. PdlC
2. UDP-Gal
HO
p-4-GalTase
cofactor
regeneration
&~;Ay-Ah-SrocU~
HO
x
UDP-Gal
-
NH
NHAc
Scheme 64. Synthesis of a glycopeptide with
engineered subtilisins and galactosyltransferase. The thiosubtilisin mutant contains the
following changes: MetSO -+ Phe, Am76 +
Asp, Gly169 +Ala, Am218 + Ser, Ser221 +
cys.
ysis over hydrolysis by a factor about 10000, and kinetics studies indicate that the selectivity arises because the acyl-enzyme
intermediate reacts more selectively with the amine nucleophile
than the wild-type enzyme (Scheme 65).[3231Chemical synthesis
of glycopeptides followed by enzymatic glycosylation, however,
hydrolysis
hydrolysis
aminolysis
E+P2
1
Reaction coordinate
Scheme 65. a) Energy diagram for aminolysis/hydrolysis catalyzed by subtilisin (-) and thiosubilisin (....). AGrc,is in units of kcalmol-I. E = enzyme, ES’ = acylated
enzyme, S = peptide, P, = amine nucleophile. P2= acid. b) Mechanism of thiosubtilisin-catalyzed hydrolysis and aminolysis. (For clarity the oxygen atom of serine and the
sulfur atom of cysteine have been explicitly depicted.)
538
Angew. Chem. In!. Ed. Engl. 1995, 34, 521 -546
Enzymes in Organic Synthesis (Part 2)
REVIEWS
works very well on an aminopropylsilica
A
key element in this strategy is the attachment of a proper
acceptor-spacer group with a selectively cleavable bond.
The method allows rapid iterative formation of peptide and
glycosidic bonds in organic and aqueous solvents, respectively ;
the glycopeptides are released from the support enzymatically
(Scheme 66).
charged transition-state structure is generated by general-acidand general-base-catalysis, presumably from carboxylic acid
and carboxylate residues in the active site of the enzyme. Compounds which resemble the transition-state structure of the glycosidic cleavage or glycosyltransfer reactions are therefore potent
inhibitors of the enzyme. The aza sugars described previously
seem to be inhibitors resembling the transition state, as they
n
,NHAc
I
HO
1. ~1-4-galactos~ransfeferase
UDP-gal
2. a2-3-dalflranSfemW
CMP-NeuAC
NHAc
I
1. chymotrypsln
2. al-3-fumsyltransferam
GDP-FW
8
NHAc
7. Inhibition of Glycosidases and Glycosyltransferases
Glycosidases and glycosyltransferases are important enzymes
for the processing of various oligosaccharide-containing glycoproteins and glycolipids. The profound impact of these enzymes
on life processes has made them desirable targets for inhibition." 5 , l 6 ] Glycoprocessing inhibitors have been used to treat
diabetes and other metabolic disorders, and have been implicated
in the blocking of infection, inflammation, and metastasis. Some
representative examples of naturally occurring and synthetic
inhibitors of glycosidases and glycosyltransferases are presented
in Schemes 67 and 68.[3251
The mechanisms of the reactions with
glycosidases and to some extent with glycosyltransferases have
been
It is generally postulated that the reactions
proceed through a half-chair (or twist-boat) transition state with
substanial sp2 character at the anomeric carbon. This positively
Angew. Chem. h t . Ed. Engl. 1995, 34, 521-546
Scheme 66. Solid-supported chemoenzymatic synthesis of a glycopeptide containing SLe". The
hexapeptide spacer (Gly,) , chymotrypsin-sensitive
ester linkage, and oligopeptide-containing GlcNAc were attached to the support by conventional
solid-phase synthesis (chemical coupling). Additional sugar residues were then added with glycos yltransferases.
show potent inhibition activities. Both five- and six-membered
aza sugars can be easily prepared by the aldolase reactions, and
they can be considered as building blocks for the development
of sequence-specific glycosidase or glycosyltransferase inhibitors. The synergistic inhibition of ctl - 3fucosyltransferase
with the fucosyl-like five-membered aza-sugars and GDP
(Scheme 69, see p. 542)['791is of particular interest, as it provides a new direction in the design of glycosyltransferase inhibitors. Another interesting observation is that most naturally
occurring glycosyltransferase inhibitors do not have he pyrophosphate moiety. Perhaps sugars (for instance, in tunicamycin)
or peptides can be used to mimic the pyrophosphate-Mn"
component of the sugar nucleotide in glycosyltransferase reactions. Other mimics of pyrophosphate-Mn2' complexes may
be useful as a linker between aza sugars and nucleosides for the
development of aza sugar nucleosides as synthetic glycosyl539
REVIEWS
-'
C.-H. Wong et al.
I castanosaermine
I
I
deoxynojkmycin
Kmethvldeomoiirimvcin
Glc-
I
deoxymannonqirimycin
SGlc- A
Glc-Man-
A
Man-
Man-
Man
Man
Man
Man
.
\
Man
-GlcNAc -GlcNAc - Asn
'
mannosidase I1
mannosidase I
nu
deoxynojirimycin
deoxymannonqirimycin
mannostatin
castanospennine
siastatin B
swainsonine
klfunensine
N(CH3)2
allosamidin
chitinase inhibitor
IC, = 3.5 x lo-' M
trestatin A-C
a-amylase inhibitor
icW
to
H
O
k OHO
?
OH
trahalostatin
trehalase InhlMtor
Icso = lo-' M
transferase inhibitors. The issue of stability and bioavailability
associated with sugar-based inhibitors will have to be examined
further if carbohydrates are to be drug candidates. The discovery that C-linked glycosides possess the same conformation as
the parent 0-linked g l y ~ o s i d e s3~2 8~1 ~is~ important,
.
because
C-linked glycosides may be metabolically stable. Similarly, Clinked aza glycosides, guanido-sugar-containing glyc~sides,[~
and 0-linked carbocycles may have a desirable conformation and
may be useful as sequence-specificendoglycosidase and glycosyltransferase inhibitors (Scheme 70, see p. 542). With regard to the
preparation of inhibitors to block the multivalent interaction of
receptors or enzymes with their ligands or inhibitors, the phospholipase D-catalyzed tran~phosphatidylation[~~~]
may offer a
540
Scheme 67. a) Naturally occurring glycosidase inhibitors and their site of inhibition in the biosynthesis of ylycoproteins [325a].
new method for the preparation of multivalent inhibitors, as
phospholipids tend to form a liposome by self-assembly
(Scheme 71, see p. 543).
8. Future Opportunities
The pace of development of carbohydrate-derived pharmaceutical agents has, in general, been slower than that of more
convenient classes of materials, undoubtedly partly because of
the difficulties in the synthesis and analysis of carbohydrates. At
least three areas of biology and medicinal chemistry have rediAngrw. Chem. hi.Ed. Engl. 1995.34, 521 -546
REVIEWS
Enzymes in Organic Synthesis (Part 2)
HoQ7
AcHN
8-
OH
Hok-cy/
AcHN
AcHN
'OR
p03H-
NI1
OH
J
H2NW N H 2
r
6+
Ki = 5.3 x 1O'5 M
sialidase (V. choke)
1325bl
_ _ 1*
[325c]
6-
vn
dH
L
Ki = 2 X lO-''M
influenza virus sialidase
J
Ki=lX10-*M
a-fucosidase
(human liver) [325d]
Ki = 1.4X lO-'u
a-fucosidese (bovine kidney)
[541
*
HOHO
&2
NHAc
NHAc
Ki = 1.9X lo4 M
~-hl-acetylglucosaminidase
(Jack beans) 1581
Ki = 3.8 X lo-' M
pN-acetylglucosaminidase
(bovine Wdney) [51]
Hs Hs2
HO
HO
OH
Ki = 5.0 x lo-' M
agalactosidase
(green coffee beans) [325e]
OH
Ki = 1.6 x lo-' M
agalactosidase
(green coffee beans) [325fl
K~= 5 x 1 0 - ~M
a-hamnosidase
-$&
HO
HO
bH
[32591
HO
NH2
Ki = lo-' M [325g]
~glucosldase
a-mannosidase
a-galactosklase
HO
OH
Ki = ~ x ~ O [325h]
- ~ M
flglucosldase
6-
HO
HO
OH
inhibitor of rat sucrase
K~= 2 x lo-' '[325i]
rected attention to carbohydrates. First, interfering with the
assembly of bacterial cell wallsr3311remains one of the most
successful strategies for the development of antimicrobials. As
bacterial resistance to the classical p-lactam antibiotics, the penams and cephams, becomes more widespread, there is increasing interest in disturbing with the biosynthesis of the characteristic carbohydrate components of the cell wall, especially KDO,
AnEcw. Climi. Int. Ed. Ennl. 1995. 343 521 -546
OH
anti-HW giucosidase
[325i1
*
Scheme 67. b) Synthetic glycosidase inhibitors containing
five- or six-membered sugar units.
heptulose, lipid A, peptidoglycans and related materials. Interest in cell-wall constituents is also heigtened by their relevance
to vaccines and to starting point for the development of nonprotein immunomodulating compounds. Second, cell-surface
carbohydrates are central to cell -cell communication, cell adhesion, infection differentiation, and development. They may be
relevant to abnormal states of differentiation, such as those
541
REVIEWS
C.-H. Wong et al.
HO
gauche
anti
''
U
HO
OH
nikkomycln2
chitin synthetase Inhibitor
Ki = 1.4 X
M
papulacandin 6
glucan synthase inhibitor
Icm = 1.2 X lo8 M
gauche
HO
OH
polyoxin D
chitin synthetase inhibitor
tunicamycin
lnhlbitor of transfer of UDP-GlcNAc
Ki = 9.0 x
M
OH
GlcNAc phosphotransferase reactlon
Scheme 68. Naturally occurring glycosyltransferase inhibitors. Tunicamycin appears to mimic the
transition state of the GlcNAc phosphotransferase reaction.
t
6:- - I _
fE:-
H\
/ O -Acceptor
anti
-- I':
Scheme 70. E m anomeric effect and the conformation of
synthetic glycosidase inhibitors. a-Glycosides prefer the
gauche conformation as a result of the anomeric and exo
anomeric effects. The C-linked glycosides exhibit a similar
conformation to the parent 0-linked glycosides. The conformations of C-linked aza glycosides, N-linked guanidino
sugars, and 0-linked carbocycles (potential sequencespecific glycosidase and glycosyltransferase inhibitors)
must still be established.
-
0 Accaptor
0-
- I
*-.
o-y=o
HO
OH
Scheme 69. Mechanism of fucosyltransferase-catalyzed reaction and synergistic inhibition
of the enzyme by GDP and an aza sugar (a fucosidase inhibitor).
characterizing some malignancies. Synthesis of these cellsurface ligands on large scales by enzymatic or chemoenzymatic methods for therapeutic evaluation are beginning
to be realized. The enzymes involved in the biosynthesis
of these ligands also represent interesting targets for
therapeutic development. Third, the broad interest in diagnostics has finally begun to generate interest in carbohydrates
542
as markers of human health. In addition, there are a
number of other possible applications of carbohydrates
(for example, as dietary constituents, as antivirals, or
as components of liposomes) that warrant attention. Enzymatic methods of synthesis will contribute to further research in all of these areas by rendering carbohydrates more
accessible.
Angew. Chem. Inl.
Ed. Engl. 1995,34, 521 -546
Enzymes in Organic Synthesis (Part 2)
REVIEWS
@ inhibitor phosphate
Hfl
HA d H
HO
0
Scheme 71. Synthesis of phospholipid-inhibitor conjugates catalyzed by phospholipase D (PLD). They self-assembleinto a liposome bilayer with numerous inhibitors
on the surfaces (a simple system for the synthesis of polyvalent ligands and inhibitors). Examples shown are an azasugar, a nucleoside, RGD peptide (the ligand
of integrin). and a sialic acid bound to a phospholipid (influenza hemagglutinin
receptor)
Work at The Scripps Research Institute was supported by the
National Institutes of Health, the National Science Foundation,
and Cytel Corporation, San Diego, and at RIKEN by the Frontier
Research Program sponsored by the Japanese Science and Technology Agency. R.L.H. acknowledges support from the American
Cancer Society in the,form of a postdoctoral fellowship.
Received: January 13, 1994 [A 47bl
German version: Angeu. Chem. 1995, 107, 559
L. F. Leloir, Science 1971, 172, 1299.
R. Kornfeld, S. Kornfeld, Annu. Rev. Biochem. 1985,54, 631.
W. J. Lennarz. Biochemistry 1987, 26, 7205.
S. C. Hubbard, Annu. Rev. Biochem. 1981,50, 555.
a) B. Imperiali, K. L. Shannon, K. W. Rickert, J. A m . Chem. SOC.1992, 114,
7942: b) B. Imperiali, K. L. Shannon, M. Unno, K. W. Rickert, ibid. 1992,
114. 7944.
a) H. A. Kaplan, J. K. Welply, W. J. Lennarz, Biochim. Biophys. Acta 1987,
906.161 ; b) A. Abbadi, M. Mcharfi, A. Aubry, S. Promilat, G. Boussard, M.
Marrand. J. Am. Chem. SOC.1991,113,2729; c) B. Imperiali, K. L. Shannon,
BiochemiArrv 1991. 30, 4374; d) C. B. Sharmd, L. Lehle. W. Tanner, Eur. J.
Bioihem. 1981. 116, 101.
J. Roth, Biochim. Biophys. Acta 1987, 906, 405.
H. Grisebach. Adv. Curbohydr. Chem. Biochem. 1978, 35, 80.
J. H. Hash, Methods Enzymol. 1975, 43; R. N. Russell, H.-W. Liu, J. Am.
Chem. Sot.. 1991,113,7777;J. A. Vara, C. R. Hutchinson, J. Biol. Chem. 1988,
263, 14992: P. J. Oths, R. M. Hayer, H. G. Floss, Carbohydr. Res. 1990, 198,
91.
0. Hindsgaul, Semin. Cell Biol. 1991, 2, 319.
Angew. Chem. Int. Ed. Engl. 1995.34, 521-546
11461 D. G. Drueckhdmmer, W. J. Hennen, R. L. Pederson, C. F. Barbas 111, C. M.
Gautheron, T. Krach, C.-H. Wong, Synthesis 1991, 499.
[I471 a) J. E. Heidlas, K. W. Williams, G. M. Whitesides, Acc. Chem. Res. 1992,25,
307; b) N. K. Kochetkov, V. N. Shihaev, Adv. Carbohydr. Chem. Biochem.
1973, 28, 307.
H. G. Khorana, Some Recent Developments in the Cherni.7tr.v of Phosphate
Esters of Biological Interests, Wiley, New York, 1961
A. M. Michelson, The Chemistry of Nucleosides and Nucleotides, Academic
Press, New York, 1963.
V. M. Clark, D. W. Hutchinson, A. J. Kirby, S. G. Warren, Angew. Chem.
1964, 76,704; Angew. Chem. Int. Ed. Engl. 1964, 3,678.
L. A. Slotin, Synthesis 1977, 737.
K. H. Scheit, Nucleotides Analogs, Synthesis and Biological Function, Wiley,
New York, 1980.
F. Cramer, H. Neunhoeffer, Chem. Ber. 1962, 95, 1664.
D. E. Hoard, D. G. Ott, % Am. Chem. SOC.1965.87, 1785.
E. S. Simon, S. Grabowski, G. M. Whitesides, J Org. Chem. 1990, 55,
1834.
a) R. R. Schmidt, Angew. Chem. 1986, 98, 213; Angew. Chem. Int. Ed.
Engl. 1986, 25, 212; b) P. Pale, G. M. Whitesides. J. Org. Chem. 1991,
56, 4547.
U. B. Gokhale, 0. Hindsgaul, M. M. Palcic, Can. J. Chem. 1990, 68,
1063.
M. M. Sim. H. Kondo, C.-H. Wong, J. Am. Chem. Soc. 1993, 115, 2260.
a) B. Pfannemuller, Staerke 1968, 11, 341; b) W. Praznik, R. Ebermann,
StarchlStaerke 1979, 31, 288.
H. Wdldmann, D. Gygax, M. D. Bednarski, W. R. Shangraw. G. M. Whitesides, Carbohydr. Res. 1986, 157, C4.
C.-H. Wong, S. L. Haynie, G. M. Whitesides, J. Org. Chem. 1982, 47,
5416.
H . J. Leucks, J. M. Lewis, V. M. Rios-Mercadillo, G. M. Whitesides, J. Am.
Chem. SOC.1979, 101, 5829.
B. L. Hirschbein, F. P. Mazenod, G. M. Whitesides, J Org. Chem. 1982, 47,
3765.
E. S. Simon, S. Grabowski, G. M. Whitesides, J. Am. Chem. SOC.1989, l l f ,
8920.
M.-J. Kim, G. M. Whitesides, Appl. Biochem. Biotechnol. 1987, 16, 95.
E. S. Simon, M. D. Bednarski, G. M . Whitesides, J. Am. Chem. SOC.1988,
110, 7159.
H. K. Chenault, E. S. Simon, G. M. Whitesides in Biotechnology and Genetic
Engineering Reviews, Vol. 6 (Ed.: G. E. Russell). Intercept, Wimborne, Dorset, 1988, Chapter 6.
C.-H. Wong, S. L. Haynie, G. M. Whitesides, J. Am. Chem. SOC.1983, 105,
115.
a) K. Kawaguchi, H. Kawai, T. Tochikura, Methods Carhohydr. Chem. 1980.
8,261; b) T. Tochikura, K. Kawaguchi, H. Kawai, Y Mugibayashi, K. Ogata,
Hakko Kogaku Zasshi 1968,46,970; c) T. Tochikura, H. Kawai, S. Tobe, K.
Kawaguchi, M. Osugi, K. Ogata, ibid. 1968, 46, 957.
U. Korf, J. Thimm, J. Thiem, Synletr 1991, 313.
V. Ginsburg, Adv. Enzymol. 1964, 26, 35.
a) J. E. Heidlas, W. J. Lees, P. Pale, G. M. Whitesides, J Org. Chem. 1992, 57,
146; b) J. E. Heidlas, W. J. Lees, G. M. Whitesides, ibid. 1992, 57, 152.
G. Srivastava, 0. Hindsgaul, M. M. Palcic, Corbohydr. Res. 1993, 245, 137.
P. A. Ropp, P.-W Cheng, Anal. Biochem. 1990, 187. 104.
F. Maley, Merhods Enzymol. 1972, 28, 271.
T. J. Grier, J. R. Rasmussen, Anal. Biochem. 1982, 127, 100.
V. Ginsberg, J. Biol. Chem. 1960, 235, 2196.
K. Yamamoto, T. Maruyama, H. Kumagai, T. Tochikura, T. Seno, H. Yamaguchi, Agric. Biol. Chem. 1984, 48, 823.
Y. Ichikawa, Y.-C. Lin, D. P. Dumas, G.-J. Shen, E. Garcia-Junceda, M. A.
Williams, R. Bayer, C. Ketcham, L. E. Walker, J. C. Paulson, C.-H. Wong, J.
Am. Chem. Soc. 1992, 114, 9283.
R. Stiller. J. Thiem, Liebigs Ann. Chem. 1992,467.
a) H. A. Nunez, J. V. OConnor, P. R. Rosevear, R. Barker, Can. J. Chem.
1981, 59, 2086; b) V. B. Gokhale, 0. Hindsgaul, M. M. Palcic, ibid. 1990.68,
1063; c) R. R. Schmidt, B. Wegmann, K.-H. Jung, Liebigs Ann. Chem. 1991,
191,121;d) G. H. Veeneman. H. J. G. Broxterman, G. H. van der Marel, J. H.
van Boom, Tetrahedron Lett. 1991,32,6175;e ) Y. Ichikawa, M. M. Sim, C:H.
Wong, J Org. Chem. 1992,57, 2943.
E. J. Toone, E. S. Simon, G. M. Whitesides, J. Org. Chem. 1991, 56,
5603.
[183] D. Gygax, P. Spies, T. Winkler, U. Pfarr, Tetrahedron 1991, 28, 5119.
[184] a) H. H. Higa. J. C . Paulson, J. Biol. Chem. 1985.260, 8838; b) J. Thiem, W.
Treder, Angew. Chem. 1986, 98, 1100; Angew. Chem. hi.Ed. Engl. 1986, 25,
1096; c) D. H. van den Eijnden, W van Dijk, Hoppe-Seyler's Z . Physiol.
Chem. 1972,353, 1817; d) J. Haverkdmp, J. M. Beau, R. Schauer, ibid. 1972,
360, 159; e) E. L. Kean, J B i d . Chem. 1970, 9,2391; f ) C. Auge, C. Gautheron, Tetrahedron Lett. 1988, 29, 789.
[185] J. Thiem, P. Stangier, Liebigs Ann. Chem. 1990. 1101.
I1861 J. L.-C. Liu, G.-J. Shen, Y. Ichikawa, J. F. Rutan, G. Zapata, W. F. Vann,
C.-H. Wong, J. A m . Chem. SOC.1992, 114, 3901.
543
REVIEWS
[187] T. J. Martin, R. R. Schmidt, Tetrahedron Leu. 1993, 34. 1765.
[188] H. Kondo. Y. Ichikawa. C:H. Wong, J. Am. Chem. Soc. 1992, 114, 8748.
[I891 S. Makino. Y Ueno, M . Ishikawa, Y Hayakawa, T. Hata, Tetrahedron Lett.
1993, 34, 2775.
[I901 L. Warren, R. Blacklow, J. Biol. Chem. 1962, 237, 3527: W. F. Vann, R. P.
Silver, C. Abeiyon. K . Chang, W. Aaronson, A. Sutton, C. W. Finn, W. Lindner, M. Kotsatos, 1 Biol. Chem. 1987, 262. 17562.
[191] G. Zapata, W. F. Vann. W. Aaronson, M. S. Lewis, M. Moos, J. B i d . Chem.
1989.264. 14769.
11921 Y. Ichikawa. G.-J. Shen, C:H. Wong. J. Am. Chem. Soc. 1991. 113, 4698.
[I931 G.-J. Shen, J. L.-C. Liu, C.-H. Wong. Biocata/ysis 1992. 6, 31.
(1941 S. L. Shames, E. S. Simon. C . W. Christopher, W. Schmid. G. M. Whitesides,
L.-L. Yang, Glycobiologv 1991, I , 187.
11951 E. Schreiner, R. Christian, E. Zhiral, Liebigs Ann. Chem. 1990. 93.
11961 M. Hartmann. R. Christian, E. Zbiral, Liebigs Ann. Chem. 1990, 83.
[197] H. J. Gross, A. Buensch, J. C. Paulson, R. Brossmer, Eur. J. Biochem. 1987,
168, 595.
[198] H. J. Gross. R. Brossmer, Eur. J Biochem. 1988, 177, 583.
[199] R. Schduer. M. Wember, C. F. d o Amaral, Hoppe-Seyler’s 2. Physiol. Chem.
1972, 353, 883.
[200] R:T.
Schwarz, R. Datema. Adv. Curbohydr. Chrm. Biochem. 1982, 40,
287.
[201] a) F. L. Schanbacher, K. E. Ebner, J B i d . Chem. 1970, 24s. 5057; b) L. J.
Berliner, M. E. Davis, K. E. Ebner, T. A. Beyer, J. E. Bell, Mol. CeN. Biochem.
1984, 62, 37; c ) H . A. Nunez. R. Barker, Biochemi5lry 1980, 19. 489.
[202] a) I P. Trdyer, R. L. Hill, J. Biol.Chem. 1971.246.6666; b) P. Andrews, FEBS
Lett. 1970, 9.297; c ) R. Barker. K. W. Olsen, J. H . Shaper. R L. Hill. J. Biol.
Chem. 1972, 247, 7135; d) A. K. Rao, F. Garver, J. Mendicino, Biochemistrj
1976. IS, 5001.
[203] a) M. M. Pdlcic, 0. P. Srivastdva, 0. Hindsgad, Curhohydr. Rrs. 1987, 159,
315; b) C. Auge, S. David, C. Mathieu, C. Gautheron, Tetrahedron Lett. 1984,
2.7, 1467.
[204] C.-H. Wong, Y. Ichikawd, T. Krach, C. Gautheron-Le Narvor, D. P. Dumas,
G. C. Look, J. Am. Chem. Soc. 1991. 113, 8137.
[205] J. Thiem, T Wiemann. Angenv. Chem. 1990, 102, 78; Angew. Chem. Int. Ed.
Engl. 1990, 29, 80.
[206] a) Y Nishida, T. Wiemann. J. Thiem, Tetrahedron Lett. 1992, 33, 8043; b) Y
Nishida, T. Wiemann, V. Sinwell, J. Thiem, 1 Am. Chem. Soc. 1993,115,2536;
c ) Y Nishida, T. Wiemann, J. Thiem, Tetrahedron Lett. 1993, 34, 2905.
[207] a) U. Zehavi, M. Herchmdn, Curhohydr. Res. 1984, 133, 339: b) U . Zehavi. S.
Sadeh, M. Herchman, ibid. 1983, 124, 23.
[208] C. Unverzagt. H. Kunz, J. C. Padson. J. Am. Chem. Soc. 1990, 112,9308; C.
Auge, C . Gautheron, H. Pora, Carbohydr. Res. 1989. 193, 288.
(2091 8 . Guilbert, T. H. Khan. S. L. Flitsch, J. Chem. Soc. Cheni. Commun. 1992.
1526.
[210] a) M. M. Palcic, 0.Hindsgaul, Glycobiology 1991, 1 , 205; b) T. L. Lowary, 0.
Hindsgaul. Carbohydr. Res. 1993. 249, 163.
[ a l l L. J. Berliner, R. D. Robinson, Biochemistry 1982, 21. 6340.
[212] a) H. Yuasa, 0. Hindsgdul, M. M. Palcic, J. Am. Chem. Soc. 1992.114, 5891;
b) H. Kodama, Y. Kajihara, T. Endo, H. Hashimoto, Tetrahedron Lett. 1993,
34, 6419.
12131 C.-H. Wong, R. Wang, Y. Ichikawd, J. Org. Chem. 1992, 57, 4343.
[214] S. Sabesan, J. C. Paulson, J Am. Chem. Soc. 1986, 108, 2068.
[215] J. Thiem, W. Treder. Angew. Chem. 1986, 98, 1100; Angew Chem. I n [ . Ed.
Ennl. 1986,2S, 1096.
[216] C. Ange. C. Gautheron, Tetrahedron Lett. 1988, 29, 789.
[217] R. D . McCoy, E. R. Vimr, F. A. Troy, J. Biol. Chem. 1985,260, 12695; F. A.
Troy, M. A. McCloskey, J. Biol. Chem. 1979, 2, 7377; J. Finne, Trends Biochem. Sci. 1985, 129.
[218] H. S. Condrddt, A. Bunsch, R. Browwmer, FEES Lett. 1984, 170,295; C. R.
Petrie, M. Sharma, 0. D. Simmons. W. Korytnyk, Carbohydr. Res. 1989, J86,
326.
[219] Y. Ito, J. J. Gaudino, J. C. Paulson, Pure Appl. Chem. 1993, 65, 753.
12201 K. K.-C. Liu, S. J. Danishefsky, J. Am. Chem. Sac. 1993. 115,4933.
[221] M. M . Palcic, A. P. Venot, R. M. Ratcliffe, 0. Hindsgaul, Carbohvdr. Res.
1989, 190, 1.
[222] D . P. Dumas, Y. Ichikawa, C.-H. Wong. J. B. Lowe. R. P. Nair. Bioorx. Med.
Chem. Lett. 1991, I , 425.
[223] P. R. Rosevear, H . A. Nnnez, R. Barker, Biochemistry 1982.21, 1421.
[224] U. B. Gokhale, 0. Hindsgdul, M. M. Palcic, Can. J Cliem. 1990, 68.
1063.
[225] G. Srivastava, K. I. Kaur, 0. Hindsgdul, M. M . Palcic, J. Biol. Chem. 1992,
267, 22356.
[226] H. Schachter, Biochem. Cell B i d . 1986, 64, 163.
I2271 I. Brockhausen, E. Hull, 0. Hindsgaul, H. Schachter, R. N . Shah, S. W. Michnick, J. P. Carver, J. B i d . Chem. 1989, 264, 11 21 1.
[2281 I. Brockhausen, J. Carver, H . Schachter, Biochem. Cell Biol. 1988, 66, 1134.
12291 G. Srivastava, G. Alton, 0. Hindsgaul, Carbohydr. Res. 1990, 207, 259.
[230] a) K. J. Kaur, G. Alton, 0. Hindsgaul, Carhohydr. Res. 1990,210,145; b) T.
Linker, S. C. Crdwley, 0. Hindsgaul, ibid. 1993, 245, 323; c) 1. Lindh. 0.
Hindsgaul, J. Am. Cl7em. Sor. 1991, 113. 216.
544
C.-H. Wong et al.
[231] a) B. Imperiali, J. W. Zimmerman, Tetrahedron Lett. 1990, 31. 6485; b) R.
Oehrlein, 0. Hindsgaul, M. M. Palcic, Carbohydr. Res. 1993, 244. 149.
[232] W. McDowell, T. J. Crier, J. R. Rasmussen, R. T. Schwarz. Biochem. J. 1987,
248. 523.
[233j P. Wang, G.-J. Shen, Y.-F. Wang, Y. Ichikawa, C.-H. Wong, J. Org. Chem.
1993, 58, 3985.
[234] a) S. L. Flitsch, J. P. Taylor, N. J. Turner, J. Chem. Soc. Clirm. Commun. 1991,
380, 382; b) S. L. Flitsch. H . L. Pinches. J. P. Taylor. N. J. Turner, J. Chem.
Soc. Perkin Trans. 1 1992, 2087.
[235] a) P. 1. Card, W D. Hitz, J. Am. Chem. Soc. 1984, 106, 5348; b) P. J. Card,
W. D . Hitz, K. G. Ripp. J. Am. Chem. Soc. 1986, 108, 158.
[236] L. Elling, M. Grothus, M.-R. Kula, Gl.ycobiology 1993, 3, 349.
[237] R. S. Clark, S. Banerjee, J. K . Coward, J. Org. Chem. 1990, 55, 6275.
[238] J. Lee. J. K . Coward, J Org. Chem. 1992, 57, 4126.
[239] Y. Ichikdwa, J. J.-C. Liu, G.-J. Shen, C.-H. W0ng.J. Am. Chem. Soc. 1991,113.
6300.
[240] G. C. Look, Y. Ichikawa, G.-J. Shen, G.-J. Cheng, C.-H. Wong. J. Org. Chem.
1993. 58,4326; J. C. Pdulson. J. Weinstein. E. L. Ujita, K . J. Riggs, P.-H. Ldi,
Biochem. So<. Trans. 1987. 1s. 618.
[241] K. Brew, F. J. Castellino, J. C. Vanaman, R. L. Hill, J. B i d . Chem. 1970, 245,
4570.
[242] B. A Bartholomew. G. W. Jourdian, S. Roseman. J. Bid. Chem. 1973, 248,
5751.
12431 M. Nagai, V. Dave. B. E. Kaplan, A. Yoshida, J. Bid. Chem. 1978, 253,
377.
[244] J. C. Pdukon, K. J. Colley. J Biol. Chern. 1989, 264, 17615.
[245] F. Toghrol, T. Kimura, I. S. Owens, 5iochemi.rtry 1990, 29, 2349.
[246] P. I. MacKenzie. J. Biol. Chem. 1986. 261. 6119.
[247] A. S. Masibay, P. K. Qasba, Proc. Nut/. Acad. Sci. U S A 1989, 86, 5733.
12481 H. Narimdtsu. S. Sinha, K. Brew, H. Okaydma, P. K. Qasba, Proc. Narl.
Acad. Sci. U S A 1986,83,4720.
[249] N. L. Shaper. W. W. Wright, 1. H. Shaper, Proc. Nut/. Acad. Sci. USA 1990,
87, 791.
[250] D. H. Joziasse, J. H. Shaper. D . H. van den Eijnden, A. J. Van Tunen. N. L.
Shaper, J. Biol. Chem. 1989, 264, 14290.
[251] R. D. Larsen, V. P. Rajan, M. M. Ruff. J. Kukowska-Latallo, R. D. Cummings, J. B. Lowe, Proc. Nut/. Acad. Sci. USA 1989, 86, 8227.
[252] R. D . Larsen. L. K. Ernst, R. P. Nair, J. B. Lowe, Proc. Natl. Acad. Scr. U S A
1990, 87, 6674.
[253] J. Weinstein, E. U. Lee, K. McEntee, P.-H. Lai, J. C. Paulson, J. Biol. Chem.
1987,262, 17 735.
[254] a) R. K . Saiki, S. Scharf, F. Faloona. K . B. Mullis, G. T. Horn, H. A. Erlich,
N. Arnheim. Science 1985,230,1350: b) N . Amheim, C. H. Levenson, Chem.
Eng. News 1990. 68(40), 36.
[255] K . D . MacFerrin, M. P. Terranova, S. L. Schreiber, G . L. Verdine, Proc.
Nutl. Acud. Sci. U S A 1990,87,1937; K. Nakazawa, K . Furukawa, H . Narimdtsu, A. Kobdta, J. Biochem. 1993. 113, 747.
[256] D. Aoki, H. E. Appert, D . Johnson, S. S. Wong, M . N. Fukuda, EMBO J.
1990, 9, 3171.
[257] J. Ghrayab, H. Kimura, M. Takdhara, H. Hsiung, Y Masui. M. Inouye,
EMBO J. 1984, 3, 2437.
[258] C. H. Krezdorn, G. Watsele, R. B. Kleene. S. X. Ivanov, E. G . Berger, Eur. J.
Biochem. 1993, 212, 113.
[259] a) W. Gillespie, S. Kelm, J. C. Paulson, J. Biol. Chem. 1992, 267, 21 004; b)
D. X. Wen. B. D. Livingston, K. F. Medzihradszky, A. L. Burlingame, J. C .
Pdulson, J. Biol. Chem. 1992, 267, 21 011.
[260] N. L. Shaper, G. F. Hollis, J. G. Douglas, I. R. Kirsch, J. H. Shaper, J. Biol.
Chem. 1988, 263, 10420.
[261] G. D’Agostaro, B. Bendiak. M. Tropak, Eur. J Biochem. 1989, 183,
211.
[262] K. Nakazawa, T. Ando, T. Kimura, H. Narimatsu. J1 Biochem. 1988, 104.
165.
[263] N. L. Shaper, J. H. Shaper. J. L. Meuth, J. L. Fox, H. Chang, I. R. Kirsch,
G. F. Hollis. Proc. Nail. Acad. Sci. USA 1986,83, 1573.
[264] K. A. Masri, H. E. Appert, M. N . Fukuda, Biochem. Biophys. Res. Commun.
1988, 157, 657.
[265] a) FucT 111: J. F. Kukowska-Latallo, R. D. Larsen, R. P. Nair, J. B. Lowe,
Genes Dev. 1990,4,1288; b) FucT IV: R. Kumar, B. Potvin, W. A. Muller, P.
Stanley, J. Bio/. Chem. 1991,266,21777; c ) FucT V: B. W Weston, R. P. Nair,
R. D. Larsen, J. B. Lowe, ibid. 1992. 267, 4152; d) FucT VI: B. W. Weston,
P. L. Smith, R. J. Kelly, J. B. Lowe, ibid. 1992, 267, 24575.
12661 Review: J. B. Lowe. Seminars in Cell Biology 1991, 2, 289.
[267] S. L. Haynie, G. M. Whitesides, Appl. Biochem. Biotechnol. 1990, 23, 205.
[268] R. Dedonder, A4ethod.s Enzymol. 1966. 8 , 500.
[269] a) D . Botstein, R. W. Davis in Molecular Biology ofthe Yeast Saccharomyces;
Metabolism and Gene Expression (Eds.: J. N. Strathern, E. W. Jones, J. R.
Broach). Cold Spring Harbor Laboratory, Plainview, NY, 1981, p. 607; b)
D. M. Carlson, Pure Appl. Chem. 1987. 59, 1489.
[270] G . Ziegast. B. Pfannemuller, Curbohydr. Res. 1987, 160, 185.
[271] a) D. French, Adv. Carhohydr. Chem. Bfoehem. 1957,lZ. 189; b) W. Saenger,
Angew. Chem. 1980. 92, 343; Angew. Chem. I n t . Ed. Engl. 1980, 19, 344.
Angew. Chem. In[. Ed. Engl. 1995, 34, 521-546
EnLyiiies in Organic Synthesis (Part 2)
[272] a ) H. Bender. Curhohi.ilr. Rrs. 1980. 78, 133; h) K. Wallenfels. B. Foldi. H.
Niermann. H. Bender, D. Linder. ihrd. 1978, 61, 359.
[273] W. Treder. J. Thiem, M . Schlingmann, Terruhi&on Let/. 1986. 27. 5605.
12741 ii) Y. E7ure. Agric. Bid. Chem. 1985. 49. 2159: h) Y. Ezure, S. Maruo, N.
Ojima. K. Konno. H. Yamashita. K . Miyazaki. T. Seto, N. Yamada, M.
Sugiyama, ihrd. 1989, 53. 61.
[275] D. N . Crowcll. M. S. Anderson, C. R. H. Raetz, J. Buercriol. 1986. 168.
52
(2763 H Vypel. D. Scholz, 1. Macher. K . Schindlmdier, E. Schutze. J. m i d . Chon.
1991, 34. 2759.
(2771 D. ScholL. K. Bednarik. G. Ehn. W. Neruda, E. Janzek. H . Loibner. K. Briner.
A Va~ella.J. M r d . C h e m 1992. 35, 2070.
(2781 a ) E. S. Hchrc. T. Sawai. C. F. Brewer, M . Nakano. T. Kanda, Biixheinlrfr.r:
1982, .?I.
3090; h) T. Kasumi. Y. Tsumuraya, C. F. Brewer, H . Kersters-Hilderson. M. Claeyssens. E. S. Hehre. ihirl. 1987. 26. 3010.
[279] I: Kasumi. C. F. Brcwer. E. T. Reesc. E. S. Hehre. Carhohydr. Re.\. 1986, 146.
7').
[280] a ) K G . I . Nilsson, Curhokydr. Rcs. 1987. 167. 95; b) rhid. 1988. 180, 53. c)
D. H. G . ('rout. D . A. MacManus. J.-M. Ricca, S. Singh, P. Crithley. W. T.
Gibson. Piire Appl. Chmii. 1992, 64. 1079; d) J. Lehmann. E. Schroter. Curholiwlr. Rc.5 1979. 71, 65.
(281) G. C . Look. C.-H. Wong, fiwuhrriron Lett. 1992. 33, 4253.
[282] L Hcdbya. E Johansson. K. Mosbach, P. 0 . Larsson. A. Gunnarsson. S.
Svensson. H. Lonn. G/woumjugntr~J 1989, 6. 161.
[283] a) S. Kohayashi. K. Koshiwa. T. Kawasaki, S.-I. Shoda. J. Am. Chem. Soc.
1991. 113. 1079; b) G. F. Herrmann. U. Kragl, C. Wandrey, Angew. Chem.
1993. 105. 139Y; Anjiew. Chem. fnf. Ed. EnRI. 1993, 32. 1342; c) G . F. Herrmann. Y Ichikawa. C. Wandrey. F. C. A. Gaeta, J. C. Paulson, C.-H. Wong,
G/ruhidrotr L ~ t t 1993.
.
34. 3091.
12841 H. Iahikawa. S. Kitahata. K. Ohtani. C . Ikuhara, 0 . Tanaka. Agrrc. Bid.
C/ic,m 1990. 54. 3137.
[285] a1 R. Dcdonder, Methoib Enrymol. 1964. 8 , 500: h) F. Kunst, M. Pascal, J.-A.
Lepesant. J. Wdlle, R. Dedonder. Eur. J. Biuchem. 1974.42.611.
[286] E. B. Rathhone. A. J. Hacking, P. S. J. Cheetham, US-A 4617269. 1986.
12871 S. Schenkmun. J. Man-Shiow. G. W Hart, V. Nussenzweig, CeN 1991, 65,
1117.
(2881 F. Vanderkerckhove. S. Schenkman, L. P. de Carvalho, S. Tomlinson, M .
Kiso, M. Yoshida, A. Hasegawa. V. Nussenzweig, Glycobiology 1992,3. 541.
[289] S. Tomlinson. L. P. de Carvalho. F. Vanderkerckhove, V. Nussenzweig, Glyrohiidqqi, 1991. 2. 549.
(2901 Y Ho. J. C. Paulson. J. Am. Cheni. Soc. 1993. 115, 7862.
[291] D. J. Hupe. . h i u Rry. Merl. Cliem. 1986, 21, Chapter 23.
[292] M. M . Mansuri. J C. Martin, Annu. Rep. Med. Chem. 1987, 22, Chapter 15.
12931 M. M. Manaurl. J. C. Martin, Annu. Rep. Mrd. Chem. 1988, 23, Chapter 17.
[294] a) . Y r d m A m / C h w n r ~ f rPurr
~ , , 3 (Eds.: L. B. Townsend. R. S. Tipson), Wiley. New York. 1986: b) Nu~leosidt.Analogs: Chemistry, Biofoyy and Medicrr i d Applicu/roir.\ (Eds.: R. T. Walker, E. Declerez, F. Eckstein), Plenum, New
York. 1979.
[295] D. W. Hutchinson. Trmd.r Biorechnui. 1990, 8. 348.
[296] T. A. Krenltsky. G. W. Kosrdlka, J. V. Tuttle, J. L. Rideout, G. B. Ehon, Carho/r!.dr. Re\. 1981, 97. 139.
[297] a) T. Utagawa. H. Morisawa. F. Yoshinaga, A. Yamazaki, K. Mitsugi. Y
Hirose. Agri(c Eiol. Chem. 1985. 49, 1053; b) T. Utagawa. H. Morisawa, S.
Yamanaka. A. Yamazaki. F. Yoshinaga, Y. Hirose, ibid. 1985, 49. 2167.
(2981 a) G . M. Tencr. H. G. Khorand, J. Chcm. Sue. 1957, 79. 437; h) Y. Inoue, F.
Ling. A. Kimura. Agric B i d . Chem. 1991, 55, 629.
[299] T. Utagawa. H. Morisasa. S. Yamanaka, A. Yamazaki. F. Yoshinaga, Y Hirose. A,qri(c Brol. C%em 1986, 50. 121.
[300] T. A. Krenitaky. J. 1. Rideout, E. Y. Chao, G. W. Koszalka, F. Gurney,
R C. Crouch. h.K. Cohn, G. Wolherg. R. Vinegar. J. Med. Chem. 1986.
29. 118.
13011 W. J. Hennen. C -H. Wong. J. Org. Chcm. 1989, 54, 4692.
[302] T. A. Krenit5ky. G. A. Freeman. S. R. Shaver, L. M. Beacham, S. Hurlbert,
N. K . Cohn. L. P. Elwell, J. W. T. Selway, J. M e d . Chem. 1983, 26, 891.
[303] J. D. Stoeckcr. S. E. Ealick, C. E. Bugg, R. E. Parks. Jr.. Pror. Fed. Am. Soc.
E.vp. Brill. 1986. 45. 2773.
13041 a) J. Holguin. R. Cardinaud, EUF.J Biuchem. 1975,54, 505; h) ihid. 1975,54,
575: c ) D. A. Carson. D. B. Wasson. E. Beutler. Pruc. Natl. Acad. Sci. USA
1984. $1. 2232: d ) D. A. Carson. D. B. Wasson, Biorhcm. Biophy. Res. Comniun 1988. 15.7. X29; e) D. Betbeder. D. W. Hutchinson. A. 0 . L. Richards,
,Vu</r,r~~
A i i d i K c s 1989. 17. 4217.
[305] M 1.Phillipa. E. Nudelman, F. C . A. Gaeta, M. Perez. A. K. Singhal, S.
Hakomori. J. C. Paulson. Sciiwce 1990. 250, 1130; G. Waltz, A. Arufto, W.
Kalanus. M. Revilacqua. B. Seed, ihrd. 1990. 250, 1132; J. B. Lowe, L. M.
Stoolman. R. P. Nair, R. D. Larson, T. L. Berhend, R. M. Marks, Cell 1990.
63, 475. X-ray crystal structure analysis of E-selectin: B. J. Graves, R. L.
Crowrher. C. Chandran. J. M. Rumberger, S . Li, K.-S. Huang, D . M. Presky,
P. C , Familletti, R. A. Wolitzky. D . K . Burns, Narurr (London) 1994, 367,
532.
[306] A. Kameyiirna. H. Ishida. M. Kiso, A. Hasegawa, Carhohydr. Rcs. 1991.
20Y. "1.
Angen. Chwrr. h i t .
€11.
EngI. 1995. 34. 521 -546
REVIEWS
(3071 K. C Nicobaou, C . W. Hummel. N. J. Bockovich, C.-H Wong, J. Chrw. Soc.
Chem. Commun. 1991, 870.
[308] H. Kondo, Y. Ichikawa. C.-H. Wong, J Am. Chim Soc. 1992, 114,
8748.
[309) S. J. Danishefsky. K. Koseki, D. A. Griffith, 3. Gervay. J. M . Peterson, F. E.
McDonald, T. Oriyama. J. Am. Chrm. Soc. 1992, /14. X331.
[310] M. M. Palcic, A. Venot. R. M. Ratcliffe, 0.Hindsgaul. C'urhohydr. Re.?. 1989.
190. 1.
131 11 G. E. Ball. R. A. O'Neill, J. E. Schultz. J. B. Lowe, B. W. Weston, J. 0. Nagy.
E. G. Brown. C. .I. Hohbs. M . D. Bednarski. J Ant. C h e n l . Sot.. 1992. 114,
5449.
(3121 D. Tyrrell, P. James. N. Rao, C. Foxall. S. Ahbas. F. Dasgupta, M. Nashed,
A. Hasegawa. M. Kiso. D. Asa. J. Kidd, B. K. Brandley, Pror. Nut/. Acud. Scr.
U S A 1991.88, 10372.
(3131 C.-T. Yuen, A.M. Lawson, W. Chai. M. Ldrkin, M S Stoll, A. C. Stuart.
F. X . Sullivan. T. J. Ahern. T, Feizi, Biochenir.r/ry 1992. 31, 9126.
13141 S. A. De Frees, F. C. A. Gaeta. Y-C. Lin. Y. Ichikawa. C.-H. Wong. J. Am.
Chem. Soc. 1993. 115, 7549.
[315] M. S. Mulligdn, J. C . Pdulson, S. A. DeFrees. Z:L. Zheng, J. B. Lowe. P. A.
Ward. Nature (London) 1993, 364, 149.
(3161 H . Fu. H. Kondo, Y Ichikawa. G. C. Look. C.-H. Wong. J Org. Chem. 1992,
57, 7265.
[317] Representative chemical glycosylations: a) R. U . Lemieux, Chmi. Soc. Rev.
1978. 7.423; b) H. Paulsen. Angew. Chem. 1982. 94. 184: Angew. Chem. Int.
Ed. Engl. 1982. 21. 155; c) Ref. [156a]; d) H. Kunz. Angrw. Chem. 1987. 99.
297; Anpew. Chenr. In[. Ed. Engl. 1987, 26. 294; e) K . Okamoto, T. Goto.
Ptruhedron 1989, 45, 5835; f) K. C. Nicolaou, T. Caulfield. H . Kataoka, P.
Kumazawa, J. Am. Chem. Soc. 1988,110,7910; g) D . R. Mootoo. P. Konradsson, U. Udodong, B. Fraser-Reid, ibid. 1988. I i U . 5583; R . L. Halcomb, S. J.
Danishefsky, ihid. 1989, 1 11. 6661; i) R. W. Friesen. S. J. Danishefsky. ihrd.
1989, f l t . 6656;J)A. G. M. Barrett, B. C. B. Bezuidenhoudt. A. F. Gasiecki,
A. R. Howell, M. A. Russell, ibid. 1989, 111. 1391; k) K Briner, A. Vesella,
Heir. Chim. Acra 1989, 72. 1371; I) D . Kahne, S. Wdiker. Y. Cheng. D. Van
Engen, J. Am. Chem. Sue. 1989. Ill, 6881; m) K. Suzuki, H. Maeta, T.
Matsumoto, Tefrahedron Let/ 1989, 30, 4853; n ) Y. Ito. T, Ogawa, Wtrahedron 1990, 46, 89; 0)A. Hasegawa, T. Nagahama. H. Ohki. H. Hotta, M.
Kiso, J. Carbohydr. Chem. 1991, 10, 493; p) A. Marra, P.Sinay, Curhohydr.
Rex 1990, 19S, 303;q) H. Lonn. K. Stenvall, Tetruhedroit Lt,rt. 1992, 33, 115;
r) E. Kirchner, F. Thiem, R. Dernick, J. Heukeshoven, J. Thiem. J Carbohydr.
Chem. 1988, 7. 453; s) F. Barres. 0. Hindsgaul, J. Am. Chem. Soc. 1991, 113,
9376: t) G. Stork, G . Kim, rbid. 1992, 114, 1087; u) P. J. Garegg, ACT.Chem.
Res. 1992, 25, 575; v) K . Toshima, K. Tatsuta, C h ~ mR. w . 1993, 93. 1503.
[318] Sialylations with TMSOTfas catalyst: a) T. J. Martin. R. R. Schmidt. Terruhedron L e t f . 1992. 33, 6123; h) H. Kondo, Y Ichikawa, C.-H. Wong, J. Am.
Chem. Soc. 1992, t14, 8748; M. M. S n n , H. Kondo, C.-H. Wong, rhid. 1993,
115.2260. With ZnC1, as Lewis acid: c) Y. Watanahe. C. Nakamoto, S. Ozaki,
Synletf 1993, 115; E. J. Corey, Y.-J. Wu, J. A m . Chem. Sot.. 1993, ti5. 8871;
d ) Synthesis of CMP-NeuAc: T. J. Martin, R. R. Schmidt. fifrahedron Lcft.
1993, 34. 1765.
[319] H. Kondo. S. Aoki, Y. Ichikawa. R. L. Halcomh. H. Ritren, C.-H. Wong, J.
Org. Cheni. 1994, 59, 864.
[320] a) H. S. Conradt, H. Egge, J. Peter-Katalinic. W. Reiser, T. Siklosi, K. Schaper, J Bid. Chem. 1987,262, 14600; b) S. P. Little, N . U. Bang, C. S. Harms,
C. A. Marks, L. E. Mattler, Bi0chemi.sfr.r. 1984, 23, 6191.
[321] B. D. Livingston. E. M. D . Robertis, J. C. Paulson. G/y~ohio/ogy1990,
1, 39.
[322] H. Kunz. Pare Appl. Chem. 1993, 65, 1223; H. Garg. R. W. Jeanloz, Adv.
Curhohydr. Chem. Biochem. 1985, 43, 135; T. Bielfeldt. S. Peters. M. Meldal.
K. Bock. H. Paulsen, Angew. Chem. 1992, 104, 881 ; Angidw. Ch[)m.Int. Ed.
Enyf. 1992. 3 1 , 857: F. E. McDonald, S. J. Danishefsky, J Org. Chem. 1992,
57. 7001; S. T. Cohen-Anisfeld. P. T. Lansburg Jr., J. An?. Chem. Sot. 1993,
115, 10531: A. L. Handlon, B. Fraser-Reid, ihrd. 1993. /IS, 3796.
[323] C.-H. Wong, M. Schuster, P. Wang, P. Sears, J. Am. C h m . Soc.. 1993. 115.
5893.
13241 M . Schuster. P. Wdng, J. C. Paulson, C.-H. Wong, J. A m . Chcvn. Soc. 1994,
116. 1135.
132.51 a) For a review see Ref. [15]: b) L. Czollnev, J. Kuszmann, A. Vasella, H d v .
Chim. Actu 1990, 73, 1338: c) M. von Itzstein, W.-Y Wu. C. B. Kok, M. S.
Pegg, J. C. Dyason, D. Jin, T. van Phan, M. L. Smythe, H. E White, S. W.
Oliver, P. M. Colman, J. N. Varghese, D . M . Ryan, J. M. Woods, R. C. Bethell, V. J. Hotham, J. M. Cameron, C. R. Penn. Nuture (London) 1993,363.
418: d) B. Winchester, C. Barker. S. Baines. G. S. Jacob, S. K. Namgoong, G.
Fleet, Biochrw. J. 1990, 265, 277: e) Y.-F. Wang. Y Takaoka, C.-H. Wong,
Angew Chem., 1994, 106, 1343: Angen.. Chem. Int. Ed. Engl. 1994, 33, 1242:
f) R. C. Bernotas. M. A. Pezzone, B. Ganem, Curhuhydr. R i s . 1987,167,305:
g ) M . K.Tong,G. Papandreou, B.Ganem,J.Anz. Chem.Soc. 1990,112,6237.
h) P. Ermert, A. Vasella, Helv. Chim. Actu 1991, 74, 2043; i) K. M. Robinson.
M . E. Begovic, B. L. Rhinehart. E. W. Heineke. J.-B Ducep, P. R. Kastner.
F. N. Marshall, C . Danzin, Diuhetes 1991, 40, 825; j) G. B. Karlsson. T. D.
Butters. R. A. Dwek. F. M. Platt. J. Bio/. Chem. 1993,268.570: k) L. Provencher, D. H . Steensma, C.-H. Wong, Bioorg. Mcd. Chrm., in press.
545
REVIEWS
13261 M.L. Sinnott, Chem. Rev. 1990,90, 1171; M.Ashwell, X.Guo, M. L. Sinnott, J. Am. Chem. Sor. 1992,ff4,10158;P.Deslongchamps, Pure Appl. Chem.
1993,65,1161;
R.Kuroki, L.H. Weaver, B. W. Matthews, Science 1993,262,
2030;N. C.J. Strynadka, M. N. G. James, J. Mol. Biol. 1991,220,401;S. G.
Withers, R. A. Warren, I. P. Street, K . Rupitz, J. B. Kempton, R. J. Aebersold, J. Am. Chem. SOC.1990, fi2, 5887; A.N. Singh, L. S. Hester, F. M.
Raushel, J. B i d . Chem. 1987.262,2554.
(3271 T.-C. WU, P. G. Goekjian, Y. Kishi, J. Org. Chem. 1987,52,4819.
546
C.-H. Wong et al.
(3281 K.N.Houk, J. E. Eksterowicz, Y.-D. Wu, C. D. Fuglesang, D. B. Mitchell, J.
Am. Chem. SOC.1993,ff5,4170.
[3291 C. Fotsch, C.-H. Wong, Tetrahedron Left.,in press.
[330] P. Wang, M. Schuster, Y.-F. Wang, C.-H. Wong, J. Am. Chem. SOC.1993,ff5,
10487.
13311 C.T Walsh,J. Bid. Chem. 1989,264,2393;
E.T.Rietschel, H. Brade, Sci. Am.
1992,267, 54.
Angew. Chem. Inr. Ed. Engl. 1995, 34,521 -546
Документ
Категория
Без категории
Просмотров
3
Размер файла
2 369 Кб
Теги
synthesis, part, application, problems, enzymes, organiz, recognition, carbohydrate
1/--страниц
Пожаловаться на содержимое документа