Chemoenzymatic synthesis of PSGL-1 glycopeptides: Sulfation on tyrosine affects glycosyltransferase-catalyzed synthesis of the O-glycan

Article abstract of DOI:10.1016/S0968-0896(00)00041-9

PSGL-1 is the primary glycoprotein ligand for P-selectin during the inflammatory response. Interestingly, the N-terminal sequence, containing both a site of tyrosine sulfation and an O-glycan, has been shown to bind to P-selectin with an affinity similar to full-length PSGL-1. To further characterize this system, the synthesis of glycopeptides from PSGL-1 was undertaken. The synthesis involved both solution- and solid-phase synthesis, as well as enzymatic transformations. During the synthesis, notable reactivity differences of the glycosyltransferases toward sulfated and unsulfated versions of the same glycopeptides were observed. (C) 2000 Elsevier Science Ltd.

Full text of DOI:10.1016/S0968-0896(00)00041-9

Bioorganic & Medicinal Chemistry 8 (2000) 1017±1025
Chemoenzymatic Synthesis of PSGL-1 Glycopeptides:
Sulfation on Tyrosine A€ects Glycosyltransferase-Catalyzed
Synthesis of the O-Glycan
Kathryn M. Koeller, Mark E. B. Smith and Chi-Huey Wong*
Department of Chemistry, The Scripps Research Institute and Skaggs Institute for Chemical Biology, 10550 N. Torrey Pines Road,
La Jolla, CA, 92037, USA
Received 27 October 1999; accepted 20 December 1999
AbstractÐPSGL-1 is the primary glycoprotein ligand for P-selectin during the in¯ammatory response. Interestingly, the N-terminal
sequence, containing both a site of tyrosine sulfation and an O-glycan, has been shown to bind to P-selectin with an anity similar
to full-length PSGL-1. To further characterize this system, the synthesis of glycopeptides from PSGL-1 was undertaken. The
synthesis involved both solution- and solid-phase synthesis, as well as enzymatic transformations. During the synthesis, notable
reactivity di€erences of the glycosyltransferases toward sulfated and unsulfated versions of the same glycopeptides were observed.
#
2000 Elsevier Science Ltd. All rights reserved.
Introduction
activity does not rely on the presence of an O-linked
glycan, no previous studies have investigated the e€ect
of sulfation on glycosyltransferase activity. Unlike
phosphorylation, sulfation is most often accepted to be
an irreversible protein modi®cation in mammalian sys-
tems, as an appropriate protein sulfatase activity has not
P-selectin glycoprotein ligand-1 (PSGL-1) is the primary
counter-receptor for P-selectin during leukocyte extra-
vasation in the in¯ammatory response. Previously, it
has been determined that the O-glycan attached to
threonine 16 and at least one site of tyrosine sulfation
within the N-terminal 19 amino acids are required for
1
4
been identi®ed. It is, therefore, not clear whether the rela-
tionship between sulfation and glycosylation on PSGL-1 is
of an on/o€ nature (analogous to phosphorylation), or
whether it is associated with the construction of a high
versus low anity P-selectin ligand.
2
optimal recognition of PSGL-1 by P-selectin. In fact,
the N-terminal glycosulfopeptide binds to P-selectin
nearly as eciently as the full-length dimeric PSGL-1,
and ꢀ5 orders of magnitude tighter than the unsulfated
3
sequence.
To further examine PSGL-1/P-selectin recognition, as
well as to investigate PSGL-1 biosynthesis, a glyco-
peptide from the PSGL-1 N-terminus was selected as
a synthetic target (Fig. 1b). Structures of the natural
O-linked glycans from PSGL-1 have recently been
elucidated, and contain a terminal sialyl Lewis x
During protein biosynthesis, modi®cations such as O-
linked glycosylation and sulfation occur within the
Golgi apparatus. Although the order in which a given
protein is decorated remains largely unknown, it has
generally been reported that sulfation occurs as the ®nal
x
(sLe ) tetrasaccharide extended o€ a core 2 glycan (Fig.
6
4
modi®cation before exit from the Golgi. In support of
1a). The target structure represents a minimal sequence
this idea, Cummings et al. have shown that tyrosyl-
protein sulfotransferase (TPST) can catalyze sulfation
of the native glycosylated PSGL-1 N-terminal peptide,
which contains a complete core 2 branched hepta-
containing tyrosine and the glycosylated threonine resi-
due, corresponding to amino acid residues 10±17 of the
mature PSGL-1 protein. The synthetic strategy com-
bined chemical methods and solid-phase peptide syn-
thesis to arrive at a disaccharide-linked octapeptide.
Utilizing this strategy, access to both the sulfated and
unsulfated disaccharide-linked peptides was possible.
Glycosyltransferase-catalyzed elaboration of the glycan
portion was then studied with respect to either the sul-
3
saccharide. However, it had been shown previously
that TPST also accepts the non-glycosylated PSGL-1
5
sequence as a substrate. Although it appears that TPST
*
2
Corresponding author. Tel.: +1-619-784-2487; fax: +1-619-784-
409; e-mail: wong@scripps.edu
7
fated or unsulfated forms of the glycopeptides. Results
0968-0896/00/$ - see front matter # 2000 Elsevier Science Ltd. All rights reserved.
PII: S0968-0896(00)00041-9

1
018
K. M. Koeller et al. / Bioorg. Med. Chem. 8 (2000) 1017±1025
indicate that sulfation on tyrosine at its semi-remote
position within the peptide in¯uences the reactivity of
the glycosyltransferases responsible for the synthesis of
Results and Discussion
Initially, a glycosylated threonine residue was required
for use in solid-phase peptide synthesis. The strategy
employed to obtain this construct is outlined in Scheme
x
sLe on the attached O-glycan.
8
1
version to trichloroacetimidate 2, which was isolated as
. Anomeric deacetylation of 1 was followed by con-
9
a mixture of anomers (a:b, 2:3). Treatment of 2 (a:b,
2
:3) and 310 with TMSOTf as activator in a dichloro-
ꢁ
methane:diethyl ether mixture (1:1) at � 30 C gave
conjugate 411 in excellent yield as predominantly the
a-anomer (a:b, 6:1). Complete reaction at low tempera-
ture prevented any loss of the tert-butyl ester upon
addition of TMSOTf. The reaction was quenched with
diethylisopropylamine to avoid loss of the Fmoc group
which was observed when triethylamine was utilized as
the quenching reagent.
Deacetylation of 4 was then achieved with sodium
methoxide in methanol. The pH was maintained at less
than 8.5 to prevent Fmoc cleavage. Conversion to 3,4-
1
2
isopropylidene 5 has been reported by Paulsen et al.
Attempts to repeat this procedure failed, due to pre-
cipitation of the 4,6-isopropylidene from solution dur-
ing the course of the reaction. However, it was observed
that using a mixed solvent of dichloromethane and 2,2-
dimethoxypropane (1:1) was more bene®cial. In this
case, the 4,6-isopropylidene was again the predominant
product early in the reaction, but did not precipitate.
Figure 1. (a) The N-terminal structure of PSGL-1. (b) The synthetic
target structure.
Ê
CCN, 4A MS, Cs
Scheme 1. Synthesis of the glyco-threonine building block. Reagents and conditions: (a) H
2
NNH
2
-HOAc, DMF; (b) Cl
3
2
CO
3
2
,
,
ꢁ
CH
CH
2
Cl
Cl
2
, 76% in 2 steps; (c) 3, TMSOTf, CH
ꢁ
2
Cl
Cl
2
/Et
2
O, � 30 C, 91%; (d) NaOMe, MeOH; (e) 2,2-DMP, TsOH, 48% in 2 steps; (f) 6a, BF
3
-OEt
Ê
, 4A MS, 0 C, 75%; (h) AcOH/H
ꢁ
2
2
, � 30 C, 100%; (g) 6b, DMTST, CH
2
2
2
O; (i) Ac
2
O/pyr; (j) Zn, Ac
2
O, 80% in 3 steps; (k) TFA/H O,
2
1
00%.

K. M. Koeller et al. / Bioorg. Med. Chem. 8 (2000) 1017±1025
1019
Over time, equilibration to 3,4-isopropylidene 5 occur-
red, and this intermediate was isolated in good yield.
removal of the tBu ester and ether protecting groups.
The crude peptide obtained was initially puri®ed by
ether precipitation and small portions then further
puri®ed using RP-HPLC to give sequence 11.
Formation of disaccharide-threonine conjugate 7 was
then successfully achieved by using two separate glyco-
sylation protocols. Combination of acceptor 5 and
In order to sulfate tyrosine within the glycopeptide
sequence, 11 was treated with sulfur trioxide-pyridine.
An ecient work-up method was then sought, as
attempts to repeat published procedures largely resul-
ted in the isolation of the unsulfated starting mate-
1
3
trichloroacetimidate donor 6a in dichloromethane at
�
ꢁ
jugate 7. Alternatively, 7 was formed by reaction of 5
30 C with BF -OEt activation gave desired con-
2
3
1
0
with thioglycoside donor 6b in dichloromethane at
0
ꢁ
14
of 7 with AcOH:H O (4:1) at 45 C then removed the
16
C using DMTST as the activating agent. Treatment
ꢁ
rial. Fortunately, it was found that quenching the
sulfating reagent with methanol, followed immediately
by silica gel chromatography was an optimal work-up
procedure.17 This allowed the sulfated sequence to be
isolated in high yield with minimal observed cleavage.
Saponi®cation of the acetate esters then gave depro-
tected glycopeptide 12a for the subsequent glycosyl-
transferase-catalyzed glycosylations. Similarly, 11 was
treated with sodium hydroxide in methanol to provide
unsulfated 12b. Thus, the procedure by which the gly-
copeptide was sulfated post-SPPS was optimal for
synthesis of the desired sequence.
2
isopropylidene group. Subsequent acetylation and con-
version of the amino functionalities yielded intermediate
8. Transformation of 8 to building block 9, required for
solid-phase peptide synthesis, was quantitatively per-
formed by cleavage of the tert-butyl ester in a mixture of
TFA:H O (95:5).
2
Building block 9 was initially incorporated into glyco-
peptide sequences on two di€erent solid supports. The
sequences were prepared to ascertain whether tyrosine
sulfation could be best achieved during or post-SPPS.
The sequence incorporating commercially available
�
Glycosyltransferase-catalyzed glycosylations of the gly-
copeptides were then investigated (Scheme 3). Galactose
was appended in a b1,4 linkage to GlcNAc in unsul-
fated sequence 12b by treatment with b1,4-galactosyl-
transferase (b1,4-GalT) in the presence of donor
substrate UDP-Gal to give 13b. Similarly, addition of
sialic acid was achieved by incubation of 13b with a2,3-
sialyltransferase (a2,3-SiaT) in the presence of CMP-
Fmoc-Tyr(OSO )-OH on solid phase was synthesized
3
using a hyper acid labile Sieber Amide modi®ed resin.
Synthesis and cleavage from the resin proceeded with-
out diculty. However, all attempts to remove the tBu
protecting groups also resulted in cleavage of the sulfate
1
5
ester. As such, the sulfated sequence was never iso-
lated in higher than trace yields.
x
NeuAc to obtain 14b. Finally, the sLe structure was
In contrast, non-sulfated sequence 10 was generated
using a Rink Amide modi®ed resin (Scheme 2). Fol-
lowing N-terminal acetylation, 10 was treated with 95%
TFA, H O, and ethane dithiol as a scavenger. These
2
conditions caused simultaneous liberation of the
sequence from the resin as the C-terminal amide and
completed by the addition of branch point carbohydrate
fucose in an a1,3 linkage to GlcNAc. This was accom-
plished by incubation of 14b and GDP-Fuc with a1,3-
fucosyltransferase V (a1,3-FucT V) to arrive at 15.
Alkaline phosphatase was added to all transferase reac-
tions to reduce product inhibition.
Scheme 2. Solid-phase synthesis of glycopeptides. Reagents and conditions: (a) Fmoc-AA-OH, HBTU, HOBt, NMM, DMF; (b) Ac
DMF/morpholine; (d) repeat a±c; (e) TFA:H
or (g) alone NaOH/MeOH 89%.
2
O/pyr; (c)
2 3
O:EDT (95:2.5:2.5) 68%, based on initial loading; (f) SO -pyr, pyr; (g) NaOH/MeOH, 81% in 2 steps

1
020
K. M. Koeller et al. / Bioorg. Med. Chem. 8 (2000) 1017±1025
Scheme 3. Di€erences in glycosyltransferase reactivity with sulfated versus unsulfated glycopeptides.
In contrast, for sulfated glycopeptide sequence 12a, the
enzymatic reactions did not proceed as smoothly as for
the unsulfated substrate. The b1,4-GalT reaction
proceeded slowly, resulting in the dicult isolation of
product 13a. Proteolytic cleavage was also encountered
at long reaction times. However, if the reaction was
allowed to proceed until all starting material had been
consumed, isolation of 13a was possible. An even more
pronounced variation of glycosyltransferase activity was
observed with a2,3-SiaT. In this case, no formation of
product 14a was observed. It appears, therefore, that
sulfation on tyrosine signi®cantly interferes with the
sialyltransferase reaction. This was unexpected, as the
sulfate is not in close proximity to the glycan portion of
the molecule.
In order to further investigate the cause of reactivity
di€erences between the substrates, a structural study
was undertaken.18 Results indicate that there are no
signi®cant structural changes between glycopeptide
sequences 12a and 12b. The 2-dimensional ROESY
spectra show nearly identical overlap. It is assumed,
then, that the di€erences in reactivity towards b1,4-
GalT may be caused by unfavorable electrostatic inter-
actions of the sulfate with the enzyme. However, the
most pronounced variation between the activity of sul-
fated and unsulfated peptides was with a2,3-SiaT. The
sulfate appears to interfere with the transfer of sialic
acid from CMP-NeuAc (Fig. 3).
Conclusion
To further characterize these observed di€erences, a
kinetic study of the enzymatic glycosylations by b1,4-
GalT was carried out. An assay that allowed the
determination of kinetic parameters for the b1,4-GalT
reaction was developed (Fig. 2, Table 1). Notably, the
kinetic parameters determined from initial reaction rates
A combination of chemical and solid-phase synthesis
provided access to sulfated and unsulfated peptides
from the N-terminus of PSGL-1. Glycosyltransferase-
catalyzed additions of galactose, sialic acid, and fucose
to the glycopeptides were then studied. Current results
indicate that sulfation on tyrosine has the potential to
regulate the activity of the glycosyltransferases required
di€er for 12a and 12b. By examination of K , it appears
m
that sulfation on tyrosine interferes with the binding of
the substrate to the enzyme in some manner. However,
this is compensated for by Vmax, resulting in an overall
value of kcat/Km that di€ers by ꢀ3 fold for the sulfated
and non-sulfated substrates.
x
for the synthesis of sLe attached to PSGL-1. This was
unexpected, as the tyrosine residue does not appear to
make contacts to the glycan portion of the molecule. Work
is in progress to investigate any possible conformational

K. M. Koeller et al. / Bioorg. Med. Chem. 8 (2000) 1017±1025
1021
Experimental
General
All non-aqueous reactions were run in oven-dried
glassware under an inert Ar atmosphere. Reactions were
monitored by thin-layer chromatography (TLC) utiliz-
ing p-anisaldehyde or cerium molybdate stain as the
developing reagent. Unless otherwise noted, reagents
and materials were obtained from commercial sources
and used as provided. All non-aqueous solvents were
1
13
distilled prior to use. 1D H and C NMR spectra were
recorded on Bruker AMX-400, AMX-500, DRX-500,
or DRX-600 MHz spectrometers, and were referenced
1
13
to residual solvent peaks (CDCl : H d 7.24, C d 77.0;
3
1
13
1
CD OD: H d 3.30, C d 49.0; D O: H d 4.76). 2D
3
2
COSY, ROESY, and TOCSY spectra were recorded on
a Bruker DRX-600 or DRX-500 spectrometer equipped
with either a broadband or inverse probe for greater
sensitivity. Solid-phase resins and Fmoc-amino acids
were purchased from Novabiochem. UDP-Gal, CMP-
NeuAc, GDP-Fucose, b1,4-galactosyltransferase from
bovine, a2,3-sialyltransferase from rat liver, and a1,3-
fucosyltransferase V from human were purchased from
Calbiochem.
Figure 2. Lineweaver±Burk plot of b1,4-galactosyltransferase reac-
�
tions: *, for 12a (Tyr-OSO
3
); *, for 12b (Tyr-OH).
1
,3,4,6-Tetra-O-acetyl-2-azido-2-deoxy-D-galactopyrano-
Table 1. Kinetic parameters for the b1,4-GalT reaction
side (1). A solution of NaN (8.94 g, 137 mmol) in H O
3
2
�
ꢁ
(22 mL) was cooled to 0 C and CH Cl (36 mL) added.
2 2
Substrate
K
m
(mM)
V
max
k
cat (s 1)
k
(mM
cat/K
m
1 � 1
�
(mM/min)
s )
The mixture was stirred vigorously and Tf O (4.68 mL,
2
�
�
5
4
27.90 mmol) added over a period of 5 min, and the
ꢁ
Sulfated 12a
Unsulfated 12b
4073
223
2.190
0.339
0.2740
0.0425
6.72Â10
1.90Â10
reaction stirred at 0 C for 2 h. The organic phase was
then separated and the aqueous phase washed twice
with CH Cl . The combined organic phases were
washed once with saturated Na CO (sat) (75 mL). The
2
2
2
3
total volume of TfN in CH Cl was 75 mL, and this
2
3
2
reagent solution was used without further puri®cation.
GalNH .HCl (3.0 g, 13.95 mmol) was dissolved in H O
2
2
(
45 mL) and treated with K CO (2.88 g, 20.93 mmol)
2 3
and CuSO4 (21 mg, 132 mmol). MeOH (90 mL) was
added followed by careful addition of the freshly pre-
pared TfN3 solution. More MeOH was added to
achieve a homogeneous solution and the reaction was
then stirred for 6 h at rt. The solvent was then removed
under vacuum and the residue redissolved in pyridine
ꢁ
(
75 mL). After cooling to 0 C, Ac O (45 mL) was added
2
and the mixture stirred at rt for 2 h. Excess reagents
were removed under vacuum and the residue dissolved
Figure 3. E€ect of sulfation on enzymatic glycosylation.
in
EtOAc
(300 mL)
before
washing
with
saturated CuSO (aq) (3Â200 mL) and NaHCO (aq)
4
3
(
200 mL). After drying over MgSO , the solvent was
4
removed under vacuum to give a residue which was
puri®ed by ¯ash chromatography (EtOAc:hexanes (3:7))
to give the title compound as a colorless syrup (5.20 g,
97%). Spectral analysis in agreement with ref 6.
change of the sulfated PSGL-1 glycopeptide upon
galactosylation, and to study the glycosyltransferases
from humans and other species. In any case, this
study may have relevant implications in carbohydrate-
mediated receptor recognition, as the sulfated PSGL-1
glycopeptide binds to P-selectin approximately ®ve
orders of magnitude more strongly than does the non-
O-(3,4,6-Tri-O-acetyl-2-azido-2-deoxy-D-galactopyrano-
side) trichloroacetimidate (2). Azide
1
(8.97 g,
24.05 mmol) was dissolved in DMF (50 mL) and hydra-
zine acetate (2.21 g, 24.05 mmol) added. After 2 h, the
reaction was diluted with EtOAc (300 mL) and washed
once with H O (300 mL). After drying over MgSO , the
3
sulfated ligand. Along with previous studies, the results
discussed herein suggest that sulfation and glycosylation
have the potential to act as a regulatory mechanism for
the construction of PSGL-1 as either a low or high a-
nity P-selectin ligand.
2
4
solvent was removed and the residue puri®ed by ¯ash
chromatography (EtOAc:hexanes (35:65)) to yield the

1
022
K. M. Koeller et al. / Bioorg. Med. Chem. 8 (2000) 1017±1025
hemiacetal product as a white foam (7.90 g, 99%).
Spectral analysis in agreement with ref. 7). Then freshly
powdered K CO (1.67 g, 12.01 mmol), that had been
added dropwise. Reaction was complete within 5 min
and quenched by the addition of Hunig's base (42 mL,
0.24 mmol). The mixture was diluted with dichloro-
methane (100 mL) and washed once with 0.1 M HCl
(
2
3
¯
ame-dried under vacuum immediately prior to the
experiment, was added to a vigorously stirred solution
of hemiacetal (5.00 g, 15.11 mmol) and fresh trichloro-
acetonitrile (3.33 mL, 33.23 mmol) in CH Cl (70 mL).
After 6 h (reaction not complete), the mixture was ®l-
tered through celite, concentrated under vacuum and
the residue puri®ed by ¯ash chromatography (EtOAc:
hexanes (3:7) to give the title compound as a white foam
(aq) (100 mL). After drying over MgSO , the solution
4
was concentrated and the crude product puri®ed by
¯ash chromatography (2:3 EtOAc:hexane) to give the
title compound as a white foam in quantitative yield.
2
2
Ê
Method B. 4A molecular sieves (1 g) was added to a
stirred solution of donor 6b (940 mg, 1.76 mmol) and
acceptor 1 (880 mg, 1.41 mmol) in CH Cl (5 mL). The
(5.55 g, 77%) (a:b, 1:2). Spectral analysis in agreement
with ref 7.
2
2
ꢁ
suspension was cooled to 0 C and freshly prepared
DMTST (1.76 ml, 1 M stock solution in CH Cl ) was
2
2
ꢀ
N -(Fluoren-9-ylmethoxycarbonyl)-O-(3,4,6-tri-O-acetyl-
2-azido-2-deoxy-ꢀ-D-galactopyranosyl)-L-threonine tert-
added. The reaction was complete after 15 min and the
reaction was quenched on addition of diisopropylethyl-
amine (613 mL, 3.52 mmol), diluted with CH Cl
butyl ester (4). Trichloroacetimidate 2 (3.42 g, 7.20mmol)
and threonine derivative 3 (1.91 g, 4.80 mmol) were
combined and dried overnight under vacuum, then dis-
2
2
(100 mL) and washed once with 0.1 M HCl (100 mL).
The acid wash dissolved the molecular sieves used in the
reaction. After drying over MgSO , the solution was
solved in a mixture of dichloromethane:diethyl ether
(
4
ꢁ
.72 mmol) was added dropwise. The mixture was
stirred at � 30 C for 30 min after which the reaction was
quenched with Hunig's base (125 mL, 0.72 mmol). The
solution was diluted with dichloromethane (100 mL)
and washed once with 0.1 M HCl (100 mL). After dry-
1:1) (50mL). After cooling to � 30 C, TMSOTf (144 mL,
concentrated under vacuum to give a residue which was
puri®ed by ¯ash chromatography (2:3 EtOAc:hexane)
to give the title compound as a white foam (1.15 g,
0
ꢁ
75%). 1H NMR (500 MHz, CDCl ) d 7.77 (2H, d,
3
J=7.5 Hz), 7.63 (2H, d, J=7.5 Hz), 7.41 (2H, app t, J
7.5), 7.32 (2H, app t, J=7.5 Hz), 5.58 (1H, d, J=9 Hz),
5.50 (1H, d, J=9 Hz), 5.31 (1H, t, J=9.5 Hz), 5.07 (1H,
t, J=9.5 Hz), 4.95 (1H, d, J=3.5 Hz), 4.88 (1H, d,
J=12.5 Hz), 4.76 (1H, d, J=8.5 Hz), 4.52 (1H, d,
J=12 Hz.), 4.31 (1H, dd, J=10, 7 Hz), 4.36 (2H, dd,
J=8, 5.5 Hz), 4.26 (4H, m), 4.15 (2H, m) 4.11 (1H, dd,
J=3.5 Hz), 4.08 (1H, dd, J=11, 2.5 Hz), 3.80 (1H, dd,
J=11, 8.5 Hz), 3.69 (1H, m), 3.59 ( 1H, m), 3.36 1H, dd,
J=8, 3.5 Hz), 2.08, 2.02, 1.96 (9H, 3s), 1.50 (12H, bs),
1.35 (3H, s), 1.32 (3H, d, J=6.5 Hz); ESIMS calcd for
C H N O Cl 1086/1088/1090, found 1086/1088/
ing over MgSO , the solution was concentrated and
4
the crude product puri®ed by ¯ash chromatography
(
EtOAc:hexanes (3:7), to give the title compound as a
white foam (3.10 g, 91%) (a:b, 6:1). Spectral analysis in
agreement with ref 9.
ꢀ
N -(Fluoren-9-ylmethoxycarbonyl)-O-(2-azido-2-deoxy-
3,4-O-isopropylidene-ꢀ-D-galactopyranosyl)-L-threonine
tert-butyl ester (5). NaOMe (1% solution in MeOH)
was added dropwise to a stirred solution of conjugate 4
47
58
5
18
3
(
3.10 g, 4.37 mmol) at a rate such that pH was always
1090.
between 8 and 8.5. After 6 h, deacetylation was com-
plete and the reaction was quenched by the addition of
AcOH until the pH was 6. The solvent was removed
under vacuum to yield a residue that was puri®ed using
ꢀ
0
N -(Fluoren-9-ylmethoxycarbonyl)-O-{O-[2 -acetamido-
0
0
0
0
0
3 ,4 ,6 -tri-O-acetyl-2 -deoxy-ꢁ-D-glucopyranosyl]-(1 ,6)-
2-acetamido-3,4-di-O-acetyl-2-deoxy-ꢀ-D-galactopyrano-
syl}-L-threonine tert-butyl ester (8). Acetonide 7 (4.33 g,
3.99 mmol) was dissolved in acetic acid:water (4:1,
100 mL) and stirred at 45 C for 8 h. Addition of toluene
and concentration under vacuum gave a residue which
was dissolved in pyridine (50 mL). The solution was
¯
ash chromatography (EtOAc:hexanes, (9:1), to yield
the triol as a white foam (1.71 g, 67%). (Spectral analy-
sis in agreement with ref 9.) Triol (2.41 g, 4.13 g) was
then dissolved in a mixture of dimethoxypropane:
CH Cl (1:1, 60 mL) and TsOH (cat.) added. The
ꢁ
2
2
ꢁ
mixture was stirred for 12 h, diluted with CH Cl2
2
cooled to 0 C and acetic anhydride (25 mL) added.
(
(
150 mL) and washed once with saturated NaHCO (aq)
3
100 mL). After drying over MgSO , the solvent was
4
After 5 min the ice bath was removed and the reaction
stirred at room temperature for 3 h. Addition of toluene
and concentration under vacuum gave a further residue
removed under vacuum and the residue puri®ed using
ash chromatography (ether:hexanes (7:3), to give the
¯
which was dissolved in THF:AcOH:Ac O (3:2:1). Zinc
2
title compound as a white foam (1.85 g, 72%). Spectral
analysis in agreement with ref 10.
dust (3 g), activated by stirring in a 2% solution of
CuSO (aq) (100 mL), was added to the THF mixture
4
while still moist. After 30 min reaction was complete
and the mixture was ®ltered through celite. Concentra-
tion of the solution gave a white solid which was redis-
solved in CH Cl (100 mL) before washing with 0.1 M
ꢀ
0
0
0
acetyl-2 -deoxy-2 -(2,2,2-trichloroethoxycarbonylamino)-
N -(Fluoren-9-ylmethoxycarbonyl)-O-{O-[3 ,4 ,6 -tri-O-
0
-D-glucopyranosyl]-(1 ,6)-2-azido-2-deoxy-3,4-O-isopro-
0
0
ꢁ
2
2
pylidene-ꢀ-D-galactopyranosyl}-L-threonine tert-butyl ester
7). Method A. Trichloroacetimidate 6a (1.39 g,
.36 mmol) and alcohol 5 (0.98 g, 1.57 mmol) were
HCl (aq). After drying over MgSO , the solution was
4
(
2
concentrated and the crude product puri®ed by ¯ash
chromatography (neat EtOAc) to give the title com-
1
combined and dried overnight under vacuum, then dis-
solved in anhydrous dichloromethane (20 mL). After
pound as a white foam (3.10 g, 80%). H NMR (400
MHz, CDCl ) d 7.78 (2H, d, J=7.5 Hz), 7.65 (2H, t,
J=7Hz), 7.41 (2H, t, J=7.5Hz), 7.33 (2H, t, J=7.5 Hz),
3
ꢁ
cooling to � 30 C, BF .OEt (29 mL, 0.24 mmol) was
3
2

K. M. Koeller et al. / Bioorg. Med. Chem. 8 (2000) 1017±1025
1023
6
.03 (1H, d, J=9.5 Hz), 5.73 (1H, d, J=9 Hz), 5.57 (1H,
d, J=8 Hz), 5.9±5.33 (2H, m), 5.07 (1H, dd, J=3 Hz),
.01 (1H, t, J=9.5 Hz), 4.88 (1H, d, J=3 Hz), 4.80 (1H,
d, J=8 Hz), 4.58 (1H, dd, J=10, 2.5 Hz), 4.47±4.44
initial loading of resin with amino acid). 1NMR
(500 MHz, CD OD) d 7.27±7.15 (5H, m), 7.02 (1H, d,
3
5
J=9 Hz), 6.69 (2H, d, J=9 Hz), 5.39 (1H, d, J=3 Hz),
4.64 (3H, m), 4.58 (2H, dd, J=9, 5 Hz), 4.52 (1H, d,
J=2 Hz), 4.50 (2H, t, J=6 Hz), 4.45±4.34 (8H, m),
4.32±4.25 (3H, m), 4.20 (1H, dd, J=6, 2 Hz), 4.08 (1H,
dd, J=12, 2 Hz), 3.84 (2H, dd, J=10, 9 Hz), 3.78±3.76
(4H, m), 3.61±3.57 (3H, m), 3.22 (1H, dd, J=14, 4 Hz),
3.00±2.94 (3H, m), 2.74 (1H, dd, J=14, 9 Hz), 2.59 (1H,
dd, J=16, 6 Hz), 2.50 (1H, dd, J=16, 7 Hz), 2.41±2.32
(6H, m), 2.10 (3H, s), 2.04 (3H, s), 2.03 (3H, s), 1.99
(3H, s), 1.96 (3H, s), 1.92 (3H, s), 1.91 (3H, s), 1.90 (3H,
s), 1.71±1.66 (2H, m), 1.57±1.54 (1H, m), 1.28 (3H, d,
(
(
2H, m), 4.32±4.25 (3H, m), 4.23±4.18 (2H, m), 4.06
1H, d, J=12 Hz), 3.79 (1H, dd, J=10.5, 5 Hz), 3.69±
3
2
.66 (1H, m), 3.61 (2H, ABQ, J=7.5 Hz), 2.16 (3H, s),
.08 (3H, s), 2.00±1.98 (12H, m), 1.94 (3H, s), 1.47 (9H,
1
3
s), 1.33 (3H, d, J=6.5 Hz). C NMR (100 MHz,
CDCl ) d 170.8, 170.64, 170.58, 170.35, 170.3, 170.0,
3
1
1
7
4
69.4, 156.5, 143.9, 143.7, 141.3, 127.11, 127.08, 125.1,
20.01, 119.99, 100.8, 100.2, 99.4, 83.1, 76.4, 71.94,
1.89, 68.9, 68.5, 68.3, 67.7, 67.4, 67.3, 61.9, 58.8, 55.1,
7.4, 47.1, 28.1, 23.3, 23.2, 20.74, 20.68, 20.6, 20.5, 18.4.
J=6 Hz), 0.95 (3H, d, J=4 Hz), 0.94 (3H, d, J=4 Hz);
�
HRMS: calcd for C H N O Cs 1146; found 1146.
12
ES-MS (neg) calcd for C H N O [M-H] 1669.7,
75 103 11 32
71
84
2
found 1670.
ꢀ
0
N -(Fluoren-9-ylmethoxycarbonyl)-O-{O-[2 -acetamido-
0
0
0
0
0
�
3
,4 ,6 -tri-O-acetyl-2 -deoxy-ꢁ-D-glucopyranosyl]-(1 ,6)-
-acetamido-3,4-di-O-acetyl-2-deoxy-ꢀ-D-galactopyrano-
Ac-Tyr(SO )-Asp-Phe-Leu-Pro-Glu-Thr(GlcNAc(ꢁ1,6)
3
2
syl}-L-threonine (9). Ester 8 (269 mg, 0.27 mmol) was
GalNAcꢀ)-Glu-NH2 (12a). Compound 11 (29.0 mg,
17.4 mmol) was dissolved in dry pyridine (870 mL). Sul-
fur trioxide±pyridine complex was added, and the reac-
tion stirred at ambient temperature for 5 h. MeOH
(ꢀ1 mL) was added to quench the sulfating reagent.
After stirring 5 min, the solution became clear and sol-
vent was evaporated. The residue was immediately pur-
dissolved in TFA:H O (95:5, 4 mL) and stirred at room
2
temperature for 2 h. Addition of toluene and con-
centration under vacuum gave the title compound as a
white solid in quantitative yield.
Ac-Tyr-Asp-Phe-Leu-Pro-Glu-Thr(Ac GluNAcꢁ(1,6)Ac2
3
i®ed by silica gel chromatography (5:4:1 CHCl :MeOH:
3
GalNAcꢀ)-Glu-NH2 (11). Pretreatment of rink amide
resin. Rink amide resin (loading 0.573 mmol/g, 440 mg)
was shaken in DMF:morpholine (1:1, 10 mL) for
H O, CHCl neutralized by passing through a plug of
2
3
basic Al O ). Fractions were collected, concentrated
2
3
and used without further puri®cation in the next reaction.
50 min, ®ltered and washed with DMF. Fmoc removalsÐ
resin bound peptide intermediates were shaken in
DMF:morpholine (1:1, 10 mL) for 50 min, ®ltered and
washed with DMF. CouplingsÐthe resin was treated
with a 5-fold excess of the Fmoc-amino acid (except for
building block 9 which was employed in a 1.1-fold
excess) in a 0.08 M solution in DMF, which contained
The solid residue was dissolved in MeOH (174 mL), and
1 M NaOH added until the solution had been adjusted
to pH 8. The reaction was stirred at ambient tempera-
ture for 1.5 h, and solvent was evaporated. The resulting
residue was puri®ed by passing through a Sephadex
G-25 size exclusion column, eluting with H O. Frac-
2
1
.5 equiv of HOBt, 2.0 equiv of NMM and 1.0 equiv of
tions containing product were lyophilized to yield 12a
1
HBTU per equivalent of Fmoc-amino acid. The mixture
was shaken and coupling times were 5 h except for that
of the building block 9 and the subsequent 3 couplings
each of which were of an 18 h duration. The excess
reactants were then removed by ®ltration and the sup-
port washed with DMF. Capping and subsequent
acetylation of the N-terminusÐthe resin was suspended
(21.8 mg, 81%, 2 steps). H NMR (600 MHz, D O) d
2
7.35±7.34 (2H, m), 7.29±7.21 (7H, m), 4.86 (1H, d,
J=4 Hz, GalNAc H-1), 4.62±4.52 (6H, m, Thr H-a, Leu
H-a, Asp H-a, Tyr H-a, Phe H-a, GlcNAc H-1), 4.42±
4.37 (2H, m, Glu H-a, Pro H-a), 4.26±4.22 (2H, m, Thr
H-b, Glu H-a), 4.10±4.04 (3H, m, GalNAc H-2, H-4,
GlcNAc H-5), 3.94±3.88 (3H, m, GlcNAc H-6 , Gal-
a
in Py/Ac O (3:1, 8 mL) and shaken for 10 min. The
2
NAc H-6 ,H-6 ), 3.74±3.67 (4H, m, GlcNAc H-2, H-6 ,
b
a
b
mixture was then ®ltered and the resin washed with
DMF. Peptide cleavageÐfollowing acetylation of the
N-terminus, the resin was washed with CH Cl and then
GalNAc H-3, Pro H-d), 3.63±3.59 (1H, m, Pro H-d),
3.52±3.49 (1H, m, GlcNAc H-3), 3.46±3.39 (2H, m,
GlcNAc H-4, H-5), 3.11±3.06 (2H, m, Phe H-b), 3.01
(1H, dd, J=14, 8 Hz, Tyr H-b), 2.84 (1H, dd, J=14,
9 Hz, Tyr H-b), 2.56 (1H, dd, J=16, 5 Hz, Asp H-b),
2.48 (1H, dd, J=16, 8 Hz, Asp H-b), 2.42±2.37 (1H, m,
Pro H-b), 2.31±2.21 (5H, m, Pro H-b, Glu H-g), 2.04
(3H, s, AcO), 2.03 (3H, s, AcO), 1.91 (3H, s, AcO),
2.08±1.91 (6H, m, Pro H-g, Glu H-b), 1.58±1.51 (3H, m,
Leu H-b, H-g), 1.23 (3H, d, J=6 Hz, Thr H-g), 0.91
(3H, d, J=6Hz, Leu H-d), 0.88 (3H, d, J=6Hz, Leu H-
2
2
methanol, before drying under vacuum. The resin was
then suspended in TFA:H O:EDT (95:2.5:2.5, 10 mL)
2
and shaken for 4 h. The mixture was ®ltered and the
resin washed with AcOH and DMF. The solution
obtained was concentrated to leave a solution in DMF,
and crude peptide obtained on ether precipitation. The
precipitate was redissolved in MeOH and reprecipitated
with ether to give a ¯u€y, white precipitate which was
isolated by gravity ®ltration. The white solid obtained
13
d); C NMR (100MHz, D O/CD OD) d 151.06, 137.07,
2
3
was dissolved in H O and lyophilized to give a white
2
135.31, 131.37, 130.20, 129.69, 128.16, 122.51, 102.17,
100.06, 77.89, 76.92, 75.09, 70.91, 70.70, 70.43, 69.71,
68.80, 61.76, 61.10, 58.09, 56.44, 55.76, 55.52, 54.38, 50.82,
50.61, 40.27, 37.79, 37.32, 30.39, 29.03, 25.58, 25.12, 23.44,
23.34, 23.27, 22.59, 21.78, 19.48; ES±MS (neg) calcd for
powder (346 mg). The partially puri®ed glycopeptide
was further puri®ed in 30 mg portions using preparative
RP-HPLC (C18; 0.1 M NH OAc/MeCN; 20±40%
4
MeCN over 30 min; R 19.47 min) to give 25 mg (per
t
portion) of the title sequence. Yield: 68% (based on
�
C H N O S [M� H] 1538.6, found 1539.
65
92 11 30

1
024
K. M. Koeller et al. / Bioorg. Med. Chem. 8 (2000) 1017±1025
Ac-Tyr-Asp-Phe-Leu-Pro-Glu-Thr(GlcNAc(ꢁ1,6)GalNAc
)-Glu-NH (12b). Compound 11 (45.1 mg, 27.0 mmol)
300 mM NH HCO ), followed by an additional pass
4 3
ꢀ
through Biogel P-2 for ®nal desalting. This procedure
gave the product 14b (3.3 mg, 48%, 2 steps) as well as
2
was dissolved in MeOH (270 mL). The pH was adjusted
to ꢀ8 with 1 M NaOH, and the reaction allowed to stir
1
unreacted starting material 13b (1.7 mg). H NMR
(500 MHz, D O) identi®able resonances d 7.30±7.14
4.5 h at ambient temperature. Solvent was evaporated,
and the residue loaded onto a Sephadex G-25 size
2
(5H, m, Phe aromatic), 7.00 (2H, d, J=8 Hz, Tyr aro-
matic), 6.74 (2H, d, J=8 Hz, Tyr aromatic), 2.65 (1H,
exclusion column, eluting with H O. Fractions contain-
2
ing product were lyophilized to give 12b (35.1 mg, 89%).
1
dd, J=12, 8 Hz, NeuAc H-3 ), 1.96 (3H, s), 1.95 (3H,
eq
H NMR (500 MHz, D O) d 7.31±7.16 (5H, m, Phe
s), 1.94 (3H, s), 1.83 (3H, s), 1.72 (1H, t, J=12 Hz,
NeuAc H-3 ), 1.15 (3H, d, J=6 Hz, Thr -CH ), 0.84
(3H, d, J=6 Hz, Leu±CH ), 0.81 (3H, d, J=6 Hz, Leu±
2
aromatic H-2, H-3, H-4, H-5, H-6), 7.03 (2H, d,
J=8 Hz, Tyr aromatic H-2, H-6), 6.76 (2H, d, J=8 Hz,
Tyr aromatic H-3, H-5), 4.90 (1H, d, J=4 Hz, GalNAc
H-1), 4.53±4.47 (6H, m, Thr H-a, Leu H-a, Asp H-a,
Tyr H-a, Phe H-a, GlcNAc H-1), 4.38±4.32 (2H, m, Glu
H-a, Pro H-a), 4.23±4.17 (2H, m, Thr H-b, Glu H-a),
ax
3
3
�
CH ); ES-MS (neg) calcd for C H N O [M� H]
3
82 120 12 40
1911.8, found 1913.
Ac-Tyr-Asp-Phe-Leu-Pro-Glu-Thr-(NeuAc(ꢀ2,3)Gal(ꢁ1,4)
[Fucꢀ1,3]GlcNAc(ꢁ1,6)GalNAcꢀ)-Glu-NH (15). Com-
4
3
3
.05±4.00 (3H, m, GalNAc H-2, H-4, GlcNAc H-5),
2
.88±3.82 (3H, m, GlcNAc H-6 , GalNAc H-6 , H-6 ),
a
pound 14b (2.3 mg, 1.2 mmol) and GDP-Fuc (0.8 mg,
1.3 mmol) were dissolved in bu€er (100 mM MES, pH
6.0,+0.25% Triton X-100, 240 mL) containing a freshly
a
b
.68±3.64 (4H, m, GlcNAc H-2, H-6 , GalNAc H-3,
b
Pro H-d), 3.58±3.54 (1H, m, Pro H-d), 3.45 (1H, t, J=9
Hz, GlcNAc H-3), 3.38±3.35 (2H, m, GlcNAc H-4, H-
prepared MnCl solution (1.3 mmol). a1,3-FucT V (24 mL,
2
5
), 3.06±2.91 (3H, m, Phe H-b, Tyr H-b), 2.79±2.71 (1H,
m, Tyr H-b), 2.52±2.40 (2H, m, Asp H-b), 2.30±2.21
6H, m, Pro H-b, Glu H-g), 1.99 (6H, s, AcO), 1.88 (3H,
s, AcO), 2.08±1.88 (6H, m, Pro H-g, Glu H-b), 1.54±
.45 (3H, m, Leu H-b, H-g), 1.18 (3H, d, J=6 Hz, Thr
H-g), 0.85 (3H, d, J=5 Hz, Leu H-d), 0.83 (3H, d,
12 mU) and alkaline phosphatase (0.1 mL, 100 mU) were
added and the reaction shaken gently at ambient tem-
perature for 2 days. The reaction was loaded onto a
Biogel P-2 size exclusion column, eluting with water.
Further puri®cation was accomplished on a DEAE
Sephadex A-25 anion exchange column (gradient 0±
500 mM NH HCO ), followed by an additional pass
(
1
1
3
J=5 Hz, Leu H-d); C NMR (100 MHz, D O) d
2
4
3
1
1
6
4
2
55.36, 137.07, 131.48, 130.18, 129.69, 129.15, 128.14,
16.40, 102.17, 100.05, 76.91, 75.03, 70.90, 70.73, 70.46,
9.69, 61.75, 56.43, 56.16, 55.52, 54.42, 52.32, 48.79,
0.28, 39.13, 37.79, 37.11, 34.42, 30.38, 25.58, 25.10,
through Biogel P-2 for ®nal desalting. The product was
then isolated by lyophilization to give 14b (1.6 mg,
65%). H NMR (500 MHz, D O) identi®able reso-
2
nances d 7.38±7.27 (5H, m, Phe aromatic), 7.11 (2H, d,
J=8 Hz, Tyr aromatic), 6.84 (2H, d, J=8 Hz, Tyr aro-
matic), 5.10 (1H, d, J=3 Hz, Fuc H-1), 2.78 (1H, dd,
1
3.44 23.31, 22.66, 21.80, 19.50; ES±MS (neg) calcd for
,
�
C H N O [M� H] 1459, found 1459.
6
5
93 11 27
J=12, 9 Hz, NeuAc H-3 ), 2.05 (3H, s), 2.04 (6H, s),
eq
Ac-Tyr-Asp-Phe-Leu-Pro-Glu-Thr(Gal(ꢁ1,4)GlcNAc
(ꢁ1,6)GalNAcꢀ)-Glu-NH2 (13b). Compound 12b
1.95 (3H, s), 1.81 (1H, t, J=12 Hz, NeuAc H-3 ), 1.25
ax
(3H, d, J=5 Hz, Fuc H-6), 1.17 (3H, d, J=6 Hz, Thr
-CH ), 0.93 (3H, d, J=6 Hz, Leu -CH ), 0.90 (3H, d,
(6.0 mg, 4.1 mmol) and UDP-Gal (2.8 mg, 4.5 mmol) were
dissolved in bu€er (130 mM HEPES, pH 7.4,+0.25%
3
3
J=6 Hz, Leu -CH ); ES±MS (neg) calcd for C H
3 88 130
�
Triton X-100, 800 mL) containing a freshly prepared
N O [M� H] 2058, found 2059.
12
44
MnCl solution (4.5 mmol). b1,4-GalT (60 mL, 41 mU)
2
�
and alkaline phosphatase (0.4 mL, 400 mU) were added
and the reaction shaken gently at ambient temperature
for 2 days. The mixture was loaded onto a Biogel P-2
size exclusion column, eluting with water. The isolated
product (5.4 mg, 81%) contained ꢀ30% unreacted
starting material. The mixture was carried on directly to
the next transferase reaction without further puri®ca-
tion. ES±MS (neg) calcd for C H N O [M� H]�
Ac-Tyr(OSO )-Asp-Phe-Leu-Pro-Glu-Thr(Gal(ꢁ1,4)Glc-
3
NAc (ꢁ1,6)GalNAcꢀ)-Glu-NH (13a). Compound 12a
2
(3.5 mg, 2.3 mmol) and UDP-Gal (1.4 mg, 2.3 mmol) were
dissolved in bu€er (130 mM HEPES, pH 7.4,+0.25%
Triton X-100, 460 mL) containing a freshly prepared
MnCl solution (2.3 mmol). b1,4-GalT (33 mL, 23 mU)
2
and alkaline phosphatase (0.2 mL, 200 mU) were added
and the reaction shaken gently at ambient temperature
for 2 days. The mixture was loaded onto a Biogel P-2
size exclusion column, eluting with water. Further pur-
i®cation was accomplished on DEAE Sephadex A-25
(gradient 0±500 mM NH HCO ) to give 13a (2.8 mg,
71
103 11 32
1620.7, found 1621.
Ac-Tyr-Asp-Phe-Leu-Pro-Glu-Thr(NeuAc(ꢀ2,3)Gal(ꢁ1,4)
GlcNAc(ꢁ1,6)GalNAcꢀ)-Glu-NH2 (14b). Compound
4
3
1
13b (4.6 mg, 2.8 mmol), UDP-Gal (0.9 mg, 1.4 mmol),
71%). H NMR (500 MHz, D O) identi®able reso-
2
and CMP-NeuAc (2.0 mg, 3.1 mmol) were dissolved in
bu€er (130 mM HEPES, pH 7.4,+0.25% Triton X-100,
nances d 7.37±7.25 (9H, m), 2.06 (3H, s), 1.96 (3H, s),
1.94 (3H, s), 1.59±1.54 (3H, m), 1.25 (3H, d), 0.93 (3H,
5
60 mL) containing a freshly prepared MnCl solution
d), 0.90 (3H, d); ES±MS (neg) calcd for C H N O S
71 102 11 35
[M� H] 1700.6, found 1701.
2
�
(
5.7 mmol). b1,4-GalT (41 mL, 28 mU), a2,3-SiaT (27 mL,
8 mU), and alkaline phosphatase (0.3 mL, 300 mU)
2
were added and the reaction shaken gently at ambient
temperature for 2 days. The reaction was loaded onto a
Biogel P-2 size exclusion column, eluting with water.
Further puri®cation was accomplished on a DEAE
Sephadex A-25 anion exchange column (gradient 0±
HPLC assay
b1,4-galactosyltransferase reactions were set up for the
sulfated and unsulfated glycopeptides at concentrations
of 1000, 500, 100, 50 and 10 mM in 100 mM HEPES

K. M. Koeller et al. / Bioorg. Med. Chem. 8 (2000) 1017±1025
1025
bu€er, pH 7.4 containing 100 mM MnCl , using condi-
3. Leppanen, A.; Mehta, P.; Ouyang, Y.-B.; Ju, T.; Helin, J.;
Moore, K. L.; van Die, I.; Can®eld, W. M.; McEver, R. P.;
Cummings, R. D. J. Biol. Chem. 1999, 274, 24838.
2
tions as described for 13a. b1,4-GalT concentration for
each reaction was 0.13 mM. Alkaline phosphatase was
added in excess to each reaction tube. A 5 mL aliquot
was removed from each reaction, diluted to 50 mL, and
immediately frozen at the following timepoints: 0, 1, 3,
4
5
. Huttner, W. B. Ann. Rev. Physiol. 1988, 50, 363.
. Ouyang, Y.-B.; Lane, W. S.; Moore, K. L. Proc. Natl. Acad.
Sci. USA 1998, 95, 2896.
. Wilkins, P. P.; McEver, R. P.; Cummings, R. D. J. Biol.
Chem. 1996, 271, 18732.
6
6, 23 h. Reaction samples were thawed immediately
prior to injection on a C18 analytical column (A:H O,
2
7. For a previous chemoenzymatic approach to glycopeptide
synthesis, see Seitz, O.; Wong C.-H. J. Am. Chem. Soc. 1997,
119, 8766.
8. Alper, P. B.; Hung, S.-C.; Wong, C.-H. Tetrahedron Lett.
1996, 37, 6029.
B: CH CN; gradient 15±20% B, 10 min). Product for-
3
mation over the initial 6 h of reaction time was mea-
sured as the increase of uridine concentration over time.
A calibration curve for uridine concentration was pre-
pared with authentic sample, and measured peak areas
calculated directly from the equation of the calibration
plot. A Lineweaver±Burk plot was constructed to
determine the kinetic parameters.
9
1
1
1
1
. Grundler, G.; Schmidt, R. R. Liebigs Ann. Chem. 1984,
826.
0. Schulz, M.; Kunz, H. Tetrahedron Asymm. 1993, 4, 1205.
1. Paulsen, H.; Adermann, K. Liebigs Ann. Chem. 1989, 751.
2. Mathieux, N.; Paulsen, H.; Meldal, M.; Bock, K. J. Chem.
Soc., Perkin Trans. 1 1997, 2359.
3. Dullenkopf, W.; Castro-Palomino; J. C.; Manzoni, L.;
Schmidt, R. R. Carbohydr. Res. 1996, 296, 135.
4. Zhang, Z.; Ollmann, I. R.; Ye, X.-S.; Wischnat, R.; Baa-
sov, T.; Wong, C.-H. J. Am. Chem. Soc. 1999, 121, 734.
5. (a) Yagami, T.; Shiwa, S.; Futaki, S.; Kitagawa, K. Chem.
Pharm. Bull. 1993, 41, 376. (b) Kitagawa, K.; Futaki, S.;
Yagami, T.; Sumi, S.; Inoue, K. Int. J. Peptide Protein Res.
1994, 43, 190.
1
Acknowledgements
1
This research was supported by the N. I. H. (GM44154)
K. M. K. thanks the American Chemical Society
Organic Division for a fellowship and also Dr. Dee-Hua
Huang and Dr. Laura Pasternack for aid in NMR
experiments and analysis.
1
16. Marseigne, I.; Roy, P.; Dor, A.; Durieuz, C.; Pelaprat, D.;
Reibaud, M.; Blanchard, J. C.; Roques, B. P. J. Med. Chem.
1
1
989, 32, 445.
7. The work up procedure employed was previously used for
References and Notes
1
. (a) Kansas, G. S. Blood 1996, 88, 3259. (b) Rosen, S. D.;
Bertozzi, C. R. Curr. Biol. 1996, 6, 261.
. (a) Wilkins, P. P.; Moore, K. L.; McEver, R. P.; Cummings,
the sulfation of carbohydrate hydroxyls: Sanders, W. J.;
Manning, D. D.; Koeller, K. M.; Kiessling, L. L. Tetrahedron
1997, 53, 16391.
2
R. D. J. Biol. Chem. 1995, 270, 22677. (b) DeLuca, M.; Dun-
lop, L. C.; Andrews, R K.; Flannery, J. V. Jr.; Ettling, R.;
Cumming, D. A.; Veldman, G. M.; Berndt, M. C. J. Biol.
Chem. 1995, 270, 26734. (c) Sako, D.; Comess, K. M.; Barone,
K. M.; Camphausen, R. T.; Cumming, D. A.; Shaw, G. D.
Cell 1995, 83, 323. (d) Pouyani, T.; Seed, B. Cell 1995, 83, 333.
18. For previous studies of the e€ect of tyrosine sulfation on
peptide conformation see: (a) Durieux, C.; Belleney, J.; Lalle-
mand, J.-Y.; Roques, B. P.; Fournie-Zaluski, M.-C. Biochem.
Biophys. Res. Comm. 1983, 114, 705; (b) Fournie-Zaluski,
M.-C.; Belleney, J.; Lux, B.; Durieux, C.; Gerard, D.; Gacel,
G.; Maigret, B.; Roques, B. P. Biochemistry 1986, 25, 3778.