Chemoenzymatic synthesis of a PSGL-1 N-terminal glycopeptide containing tyrosine sulfate and α-O-linked sialyl Lewis X [15]

Full text of DOI:10.1021/ja0004938

J. Am. Chem. Soc. 2000, 122, 4241-4242
4241
Scheme 1. Retro-synthetic Analysis of the PSGL-1
N-terminal Sulfated Glycopeptide
Chemoenzymatic Synthesis of a PSGL-1 N-Terminal
Glycopeptide Containing Tyrosine Sulfate and
r-O-Linked Sialyl Lewis X
Kathryn M. Koeller, Mark E. B. Smith,
Rong-Fong Huang, and Chi-Huey Wong*
Department of Chemistry
The Scripps Research Institute and
Skaggs Institute for Chemical Biology
10550 North Torrey Pines Road
La Jolla, California 92037
ReceiVed February 9, 2000
Tyrosine sulfation and O-linked glycosylation are post-
translational modifications of P-selectin glycoprotein ligand-1
(PSGL-1) that are required for high-affinity binding interactions
with P-selectin.1 Interestingly, N-terminal fragments of PSGL-1
containing a core 2 O-linked glycan and tyrosine sulfate bind to
P-selectin with affinity similar to that of the full-length homo-
dimeric PSGL-1 protein.2 This suggests that simple sulfated
glycopeptide structures may be suitable for disruption of PSGL-
1/P-selectin binding events during excessive leukocyte infiltration
in the inflammatory response.3 The PSGL-1 core 2 O-glycan
attached to Thr16 displays sialyl Lewis X (sLex), the minimum
selectin binding determinant, on the â-1,6-branch of an R-linked
GalNAc.4 Of greater interest is the role of the tyrosine sulfate.
The isolated sLex tetrasaccharide binds to P-selectin only weakly
(Kd ≈ 3-4 mM), yet the N-terminal fragment of PSGL-1
containing sLex and tyrosine sulfate adheres with nanomolar
affinity.2
The chemoenzymatic synthesis of sulfated glycopeptides has
not been previously reported on a preparative scale. To develop
methodology toward the synthesis of sulfated glycopeptides, as
well as to investigate PSGL-1 biosynthesis and P-selectin interac-
tions, a portion of the PSGL-1 N-terminus was selected as a
synthetic target. Traditional chemical synthesis of sulfated gly-
copeptides has been prohibitive, due to lability of sulfate esters
to the acid catalysts used in chemical glycosylation reactions.
However, the utility of enzymes as catalysts provides another
synthetic avenue. Enzymatic glycosylations with glycosyltrans-
ferases proceed at essentially neutral conditions and are therefore
compatible with the presence of a sulfate ester. By combining
chemical and enzymatic methodology, the synthesis of a PSGL-1
glycopeptide containing tyrosine sulfate and an R-linked sLex
O-glycan was accomplished. The synthetic sequence outlined in
Scheme 1 should be applicable to the chemoenzymatic synthesis
of sulfated glycopeptides in general.
readily available glycosyltransferases â-1,4-galactosyltransferase
(â-1,4-GalT), R-2,3-sialyltransferase (R-2,3-SiaT), and R-1,3-
fucosyltransferase-V (R-1,3-FucT V).
Initially, the chemical portion of the synthesis required a
glycosylated threonine residue for incorporation into solid-phase
peptide synthesis (SPPS). The disaccharide-linked threonine
structure 15 represents the (â-1,6)-branch of the core 2 O-glycan
on which sLex is constructed. Utilizing glycoconjugate 1 in SPPS
allowed the subsequent enzymatic synthesis to proceed with the
Glycosyl-threonine 1 was incorporated into an Fmoc-based
SPPS strategy on Rink amide functionalized resin (Scheme 2).
The iterative synthesis provided glycosylated octapeptide 2
attached to the solid support. Cleavage from the resin was then
performed with TFA/H2O, conditions that also resulted in
deprotection of the t-Bu esters and ether, yielding 3. With
protected glycoconjugate 3 in hand, access to sulfated and
unsulfated versions of the glycopeptide was possible. In the first
case, tyrosine sulfation was achieved through reaction of 3 with
SO3-pyr complex. Subsequently, basic hydrolysis of the acetate
protecting groups gave sulfated 4a. Simple deprotection of 3 under
identical conditions afforded unsulfated 4b.
As previously reported, sulfation on tyrosine affected the ability
of â-1,4-GalT to catalyze the addition of galactose to glycopeptide
4a.5 Under conditions successfully utilized for reaction with
unsulfated 4b (Scheme 3), reaction of â-1,4-GalT with sulfated
4a proceeded slowly. As a further hindrance, proteolytic products
(1) (a) Wilkins, P. P.; Moore, K. L.; McEver, R. P.; Cummings, R. D. J.
Biol. Chem. 1995, 270, 22677. (b) DeLuca, M.; Dunlop, 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.
(2) Leppanen, A.; Mehta, P.; Ouyang, Y.-B.; Ju, T.; Helin, J.; Moore, K.
L.; van Die, I.; Canfield, W. M.; McEver, R. P.; Cummings, R. D. J. Biol.
Chem. 1999, 274, 24838.
(3) (a) Kansas, G. S. Blood 1996, 88, 3259. (b) Rosen, S. D.; Bertozzi, C.
R. Curr. Biol. 1996, 6, 261.
(4) Wilkins, P. P.; McEver, R. P.; Cummings, R. D. J. Biol. Chem. 1996,
271, 18732.
(5) Koeller, K. M.; Smith, M. E. B.; Wong, C.-H. J. Am. Chem. Soc. 2000,
122, 742.
10.1021/ja0004938 CCC: $19.00 © 2000 American Chemical Society
Published on Web 04/15/2000

4242 J. Am. Chem. Soc., Vol. 122, No. 17, 2000
Communications to the Editor
Scheme 2. Solid-phase Synthesis of Sulfated and Unsulfated Glycopeptidesa
a
(a) Fmoc-AA-OH, HBTU, HOBt, NMM, DMF; (b) Ac2O, Pyr; (c) DMF/morpholine; (d)repeat a-c; (e) TFA/H2O/EDT (68%, based on initial
loading); (f) SO3-pyr, pyr; (g) NaOH/MeOH (81%, 2 steps); or (g alone) NaOH/MeOH (89%, 1 step).
Scheme 3. Enzymatic Synthesis of Sulfated and Unsulfated Glycopeptidesa
a
(a) UDP-Gal, â-1,4-GalT (bovine), alkaline phosphatase, 130 mM HEPES, pH 7.4, 0.25% Triton X-100, MnCl2, protease inhibitor cocktail (“PIC”,
Sigma) (5a: 75%, 5b: 72%); (b) CMP-NeuAc, R-2,3-SiaT (6a: bacterial, 6b: rat liver), alkaline phosphatase, 130 mM HEPES, pH 7.4, 0.25% Triton
X-100, MnCl2, (6a: +PIC, 69%, 6b: -PIC, 48%); (c) GDP-Fuc, R-1,3-FucT V (human), alkaline phosphatase, 100 mM MES, pH 6.0, 0.25% Triton
X-100, MnCl2, protease inhibitor cocktail (Sigma) (7a: +PIC, 87%, 7b: -PIC, 65%).
were observed to form at long reaction times. Fortunately, this
problem was solved by the addition of a protease inhibitor cocktail
into the reaction. The addition of this cocktail allowed the reaction
to proceed for long duration without destruction of the reactant
and product glycopeptides. Under the new conditions, it was
possible to isolate and characterize sulfated glycopeptide 5a.
Moreover, addition of protease inhibitor cocktail into all glycos-
yltransferase reactions investigated was found to ease isolation
and raise product yield. As such, the isolation of 5b was also
accomplished, a task which was not possible on prior occasions.
The commercially available rat liver R-2,3-SiaT was not
reactive toward sulfated glycopeptide 5a but did catalyze reaction
with unsulfated 5b, giving sialylated product 6b.5 The activity
of rat liver derived R-2,3-SiaT appears to be dramatically affected
by sulfation of its potential substrates, as it was also observed to
be inactive toward certain sulfated chito-oligomers terminating
in galactose.6 Fortunately, examination of sialyltransferases from
other sources proved beneficial. It was found that reaction of
sulfated glycopeptide 5a with CMP-NeuAc in the presence of
R-2,3-SiaT from Escherichia coli7 afforded product 6a. Again,
the addition of protease inhibitor cocktail was necessary to avoid
the formation of cleavage products.
afforded 7a, containing the complete sLex structure. As such, the
synthesis of the PSGL-1 N-terminal sulfated glycopeptide con-
taining both sLex and tyrosine sulfate was accomplished.
This is the first report of the chemoenzymatic synthesis of a
sulfated glycopeptide on semipreparative scale and is the first in
which the sulfate ester was introduced prior to the peripheral
carbohydrates. This sulfated glycopeptide structure would be
difficult to synthesize by chemical methods alone. The use of
glycosyltransferases allowed the addition of peripheral sugars
under neutral conditions, which did not result in destruction of
the tyrosine sulfate ester. In the past, the combination of SPPS
and enzymatic glycosylation has been a successful synthetic route
to glycopeptides.8 This methodology can now be extended to
include sulfated glycopeptides as well.
Acknowledgment. This research was supported by the NIH (GM
44154). K.M.K. thanks the American Chemical Society Organic Division
for a graduate fellowship.
Supporting Information Available: Characterization data for 5a,
5b, 6a, 7a, as well as conditions for the cloning and overexpression of
bacterial R2,3-SiaT (PDF). This information is available free of charge
via the Internet at http://pubs.acs.org.
Finally, as described for the synthesis of unsulfated 7b,5
reaction of 6a with GDP-Fuc in the presence of R-1,3-FucT V
JA0004938
(6) Burkart, M. D.; Izumi, M.; Chapman, E.; Wong, C.-H., submitted for
publication.
(8) (a) Seitz, O.; Wong, C.-H. J. Am. Chem. Soc. 1997, 119, 8766. (b)
Schuster, M.; Wang, P.; Paulson, J. C.; Wong, C.-H. J. Am. Chem. Soc. 1994,
116, 1135.
(7) R2,3-SiaT from Neisseria meningitidis was cloned and overexpressed
in E. coli. See the Supporting Information for details.