Chemical Synthesis of CD52 Glycopeptides Containing the Acid-Labile Fucosyl Linkage

Article abstract of DOI:10.1021/jo034773s

Glycopeptide 1 with the fucosylated trisaccharide, β-D-GlcNAc(1→ 4)[α-L-Fuc(1→6)]-β-D-GlcNAc, linked to the Asn of CD52 peptide was prepared by two methods, both of which used the free glycosyl Asn 12 and glycotripeptide 21 as key intermediates. Thus, after the trisaccharide was prepared and linked to Asn, the carbohydrate moiety was deprotected to give 12. From 12, 21 was constructed in homogeneous NMP solutions by elongating the peptide chain alone the N-terminus. Though the glycopeptides were easily soluble in NMP, they were barely soluble in diethyl ether, because of the free trisaccharide. Consequently, addition of diethyl ether to the reaction mixtures could precipitate the glycopeptides, and the products were conveniently isolated and purified in the solid form. The coupling of 21 with a free nonapeptide 24 in NMP afforded 1. 1 was also prepared by solid-phase synthesis, using the acid-sensitive 2-chlorotrityl resin. In this case, 21 was attached to the nonapeptide on the resin, and the resulting glycopeptide was then released with dilute acetic acid. Deprotection of the peptide under moderate acidic conditions gave 1. The acid-labile α-fucose was not affected in these syntheses.

Full text of DOI:10.1021/jo034773s

Ch em ica l Syn th esis of CD52 Glycop ep tid es Con ta in in g th e
Acid -La bile F u cosyl Lin k a ge
Ning Shao, J ie Xue, and Zhongwu Guo*
Department of Chemistry, Case Western Reserve University, Cleveland, Ohio 44106
zxg5@cwru.edu
Received J une 5, 2003
Glycopeptide 1 with the fucosylated trisaccharide, â-D-GlcNAc(1f4)[R-L-Fuc(1f6)]-â-D-GlcNAc,
linked to the Asn of CD52 peptide was prepared by two methods, both of which used the free glycosyl
Asn 12 and glycotripeptide 21 as key intermediates. Thus, after the trisaccharide was prepared
and linked to Asn, the carbohydrate moiety was deprotected to give 12. From 12, 21 was constructed
in homogeneous NMP solutions by elongating the peptide chain alone the N-terminus. Though the
glycopeptides were easily soluble in NMP, they were barely soluble in diethyl ether, because of the
free trisaccharide. Consequently, addition of diethyl ether to the reaction mixtures could precipitate
the glycopeptides, and the products were conveniently isolated and purified in the solid form. The
coupling of 21 with a free nonapeptide 24 in NMP afforded 1. 1 was also prepared by solid-phase
synthesis, using the acid-sensitive 2-chlorotrityl resin. In this case, 21 was attached to the
nonapeptide on the resin, and the resulting glycopeptide was then released with dilute acetic acid.
Deprotection of the peptide under moderate acidic conditions gave 1. The acid-labile R-fucose was
not affected in these syntheses.
In tr od u ction
useful models for structure-activity relationship studies
of glycopeptides. To this end, pure CD52 glycopeptides
are essential. The problem is that it is difficult to obtain
pure CD52 glycopeptides by biological methods because
of biological microheterogeneity, a common problem in
glycoscience. A promising solution to the problem is
chemical synthesis.
However, chemical synthesis of glycopeptides is a
significant challenge.8,9 What makes chemical synthesis
of CD52 glycopeptides more complex is the presence of a
fucose residue in its N-glycan. It is well established that
the R-fucosidic linkage is particularly labile to acids. For
example, the glycosidic bond of benzylated fucose is easily
broken by trifluoroacetic acid (TFA) used to retrieve
glycopeptides from resins in the traditional solid-phase
synthesis.10 The glycosidic bond of acylated fucose is more
stable, so it is usually the acyl-protected fucose that is
used to prepare fucosylated glycopeptides.11-15 A draw-
back of utilizing acyl protective groups is that, to form
the R-fucosidic linkage and avoid neighboring group
participation in the glycosylation, the fucosyl donor must
be ether-protected and later on it has to be transformed
to the acylated form in several steps.14-16 Though a one-
CD52 is a glycosylphosphatidylinositol (GPI)-anchored
glycopeptide expressed by virtually all human lympho-
cyte and sperm cells.1,2 Lymphocyte CD52 and sperm
CD52 share an identical peptide, but they have distinct
functions. The former plays an important role in the
human immune system,3 while the latter is involved in
the reproduction process of human beings.4,5 Sperm
CD52-specific antibodies that did not cross-react with
lymphocyte CD52 were also identified.6,7
The peptide of CD52 antigens is extremely short,
consisting of only 12 amino acids, and it has one N-
glycosylation site, to which complex glycans having the
fucosylated core are linked.1,2 The glycans have a decisive
influence on the biological activities of CD52. For in-
stance, it is the N-glycans of sperm CD52 that the sperm-
specific antibodies bind to.6,7
Because of their short peptide, simple glycosylation
pattern, and interesting bioactivities, CD52 antigens are
(1) Schroter, S.; Derr, P.; Conradt, H. S.; Nimtz, M.; Hale, G.;
Kirchhoff, C. J . Biol. Chem. 1999, 274, 29862-29873.
(2) Treumann, A.; Lifely, M. R.; Schneider, P.; Ferguson, M. A. J .
J . Biol. Chem. 1995, 270, 6088-6099.
(3) Hale, G.; Xia, M.-Q.; Tighe, H. P.; Dyer, J . S.; Waldmann, H.
Tissue Antigens 1990, 35, 118-127.
(8) Kunz, H. Pure Appl. Chem. 1993, 65, 1223-1232.
(9) Seitz, O. ChemBioChem 2000, 1, 214-246 and references therein.
(10) Kunz, H.; Unverzagt, C. Angew. Chem., Int. Ed. Engl. 1988,
27, 1697-1699.
(11) von dem Bruch, K.; Kunz, H. Angew. Chem., Int. Ed. Engl. 1994,
33, 102-103.
(12) Kunz, H.; Maerz, J . Synlett 1992, 591-593.
(13) Unverzagt, C.; Kunz, H. Bioorg. Med. Chem. 1994, 2, 1189-
1201.
(4) Diekman, A.; Norton, E. J .; Klotz, K. L.; Westbrook, V. A.;
Shibahara, H.; Naaby-Hansen, S.; Flickinger, C. J .; Herr, J . C. FESEB
J . 1999, 13, 1303-1313.
(5) Eccleston, E. D.; White, T. W.; Howard, J . B.; Hamilton, D. W.
Mol. Reprod. Dev. 1994, 37, 110-119.
(6) Tsuji, Y. In Immunology of Human Reproduction; Kurpisz, M.,
Fernandez, N., Eds.; BIOS Science Pub.: Oxford,UK, 1995; pp 23-
32.
(14) Hietter, H.; Schultz, M.; Kunz, H. Synlett 1995, 1219-1230.
(15) Peters, S.; Lowary, T. L.; Hindsgaul, O.; Medal, M.; Bock, K. J .
Chem. Soc., Perkin Trans. 1 1995, 3017-3022.
(7) Tsuji, Y.; Clausen, H.; Nudelman, E.; Kaizu, T.; Hakomori, S.-
I.; Isojima, S. J . Exp. Med. 1988, 168, 343-356.
10.1021/jo034773s CCC: $25.00 © 2003 American Chemical Society
Published on Web 10/21/2003
J . Org. Chem. 2003, 68, 9003-9011
9003

Shao et al.
glycopeptide was previously prepared by solid-phase
synthesis,23 its N-glycan did not contain the problematic
fucosyl residue. Bertozzi and co-workers24 recently
achieved a more complex glycopeptide of CD52 by chemi-
cal ligation of glycans, which was not fucosylated either.
As mentioned above, the glycosidic linkage of fucose,
especially if it is protected by benzyl groups, is extremely
labile to acids.10 Our studies indicated that the half-life
of the core trisaccharide having a benzylated fucose was
less than an hour in 10-20% TFA/dichloromethane
(DCM) commonly used to deprotect peptide side chains.
Therefore, conventional solid-phase synthesis with ben-
zylated glycosyl amino acids is not applicable even if an
acid-sensitive resin is utilized to facilitate the retrieval
of glycopeptides. In contrast, the free trisaccharide is
significantly more stable to acids. For example, we did
not notice any obvious change of the free trisaccharide
after it was treated with 20% TFA/DCM at room tem-
perature for 3 h. Thus, using unprotected oligosaccha-
rides as building blocks should be of general benefit to
fucosylated glycopeptides.
To deal with the stability problem of carbohydrates,
as well as the potential problems related to their final
deprotection, we recently explored a new strategy for
glycopeptide synthesis, namely “solution-phase synthesis
with solid-phase workup”, using unprotected glycosyl
amino acids as the key building blocks and phase tags.25,26
Its underlining concept is that the free oligosaccharides
are used to facilitate the isolation of reaction products
by the precipitation method, while neither detachment
of the phase tags nor final deprotection of carbohydrates
is required. This strategy should be especially suitable
for synthetic targets such as 1.
An alternative solution to the problem is solid-phase
synthesis with unprotected glycosyl amino acids as
building blocks,27-29 because the free oligosaccharides are
more stable to acidic conditions. Nevertheless, to avoid
treating fucosylated glycopeptides with strong acids, it
is beneficial to employ an acid-sensitive resin as the solid-
phase support. The glycosidic linkage of free fucose can
then withstand the mild acidic conditions involved in the
retrieval of glycopeptides from the resin and the depro-
tection of peptide side chains. Moreover, the new solution-
phase synthesis can be combined with solid-phase syn-
thesis to formulate a novel and highly convergent method.
Syn th esis of th e Un p r otected Tr isa cch a r id e-Asn
Con ju ga te (Sch em e 1). The monosaccharide building
blocks 3, 4, and 8 were prepared from D-glucosamine (2)
and L-fucose (7) according to the reported procedures.30-32
F IGURE 1. The synthetic target.
step procedure to prepare R-linked fucosyl amino acids
from peracetylated fucose was reported,17 to the best of
our knowledge no similar method has been developed for
fucosylated oligosaccharides. More importantly, the final
deprotection of acyl-protected glycopeptides is tricky.
While strict control of pH is critical to prevent â-elimina-
tion and other side reactions in the deprotection,9 more
basic conditions may be necessary for the complete
removal of acyl groups in complex oligosaccharides.
Another potential problem caused by the intimate core
fucosyl group is the steric hindrance that may hinder the
coupling reactions especially in solid-phase glycopeptide
synthesis. Enzymatic and chemoenzymatic synthesis is
a useful method for fucosylated glycopeptides,18-22 while
its downside is the limited availability and substrate
specificity of the enzymes.
Fucosylation is very common in nature and it has a
significant influence on the conformations and biological
functions of natural glycoproteins. Therefore, more robust
and practical synthetic methods for fucose-containing
glycopeptides are highly desirable not only for the study
of CD52 but also for the study of other glycoproteins.
In this paper, two related methods were exploited in
the preparation of fucose-containing CD52 glycopeptides.
One is solution-phase synthesis with solid-phase workup
of the glycopeptide products, and the other is combined
solution-phase and solid-phase synthesis with use of the
acid-sensitive 2-chlorotrityl resin. Both utilized a glycosyl
asparagine with its carbohydrate chain deprotected as
the key building block, and both should be applicable to
other glycopeptides.
Resu lts a n d Discu ssion
One of our synthetic targets was glycopeptide 1 that
has a fucosylated core trisaccharide linked to the Asn of
an intact CD52 peptide (Figure 1). Though a CD52
(23) Guo, Z.; Nakahara, Y.; Nakahara, Y.; Ogawa, T. Angew. Chem.,
Int. Ed. Engl. 1997, 36, 1464-1467 and references therein.
(24) Pratt, M. R.; Bertozzi, C. R. J . Am. Chem. Soc. 2003, 125, 6149-
6159.
(16) Maerz, J .; Kunz, H. Synlett 1992, 589-590.
(17) Elofsson, M.; Roy, S.; Salvador, L. A.; Kihlberg, J . Tetrahedron
Lett. 1996, 37, 7645-7648.
(25) Xue, J .; Guo, Z. J . Org. Chem. 2003, 68, 2713-2720.
(26) Wen, S.; Guo, Z. Org. Lett. 2001, 3, 3773-3776.
(27) Otvos, L. J .; Urge, L.; Hollosi, M.; Wroblewski, K.; Graczyk,
G.; Fasman, G. D.; Thurin, J . Tetrahedron Lett. 1990, 41, 5889-5892.
(28) Reimer, K. B.; Meldal, M.; Kusumoto, S.; Fukase, K.; Bock, K.
J . Chem. Soc., Perkin Trans. 1 1993, 925-932.
(29) Laczko, I.; Hollosi, M.; Urge, L.; Ugen, K. E.; Weiner, D. B.;
Mantsch, H. H.; Thurin, J .; Otvos, L., J r. Biochemistry 1992, 31, 4282-
4288.
(18) Schuster, M.; Wang, P.; Paulson, J . C.; Wong, C.-H. J . Am.
Chem. Soc. 1994, 116, 1135-1136.
(19) Witte, K.; Sears, P.; Martin, R.; Wong, C.-H. J . Am. Chem. Soc.
1997, 119, 2114-2118.
(20) Seitz, O.; Wong, C.-H. J . Am. Chem. Soc. 1997, 119, 8766-
8776.
(21) Leppanen, A.; Mehta, P.; Ouyang, Y.-B.; J u, T.; Helin, J .; Moore,
K. L.; van Die, I.; Canfield, W. M.; McEver, R. P.; Cummings, R. D. J .
Biol. Chem. 1999, 274, 24838-24848.
(22) Hokke, C. H.; van den Eijnden, D. H. Carbohydr. Res. 1997,
305, 463-468.
(30) Gurjar, M. K.; Rajendran, V.; Rao, V. B. Tetrahedron Lett. 1998,
39, 3803-3806.
(31) Lemieux, R. U.; Takeda, T.; Chung, B. Y. ACS Symp. Ser. 1976,
39, 90-115.
(32) Peters, T.; Weimar, T. Liebigs Ann. Chem. 1991, 237-242.
9004 J . Org. Chem., Vol. 68, No. 23, 2003

CD52 Glycopeptides Containing the Acid-Labile Fucosyl Linkage
SCHEME 1a
to construct the target glycopeptide 1 by coupling the free
N-terminal glycopeptide segment Gly1-Asn3[glycan] to the
free nonapeptide segment Asp4-Ser12. This convergent
design can minimize the number of C-terminal elonga-
tions that require acidic treatment.25 The key elements
of this synthesis are the preparation of the glycotripeptide
segment 16 and the coupling between the free peptide
and glycopeptide segments.
Glycotripeptide 16 was prepared from 12 by solution-
phase elongation of the peptide chain alone the N-
terminus. N-Methylpyrrolidinone (NMP), which could
dissolve the unprotected glycopeptides easily, was used
as the solvent to achieve homogeneous reactions,25 and
the commercially available pentafluorophenyl (Pfp) esters
of amino acids were employed for the peptide coupling
reactions.
The synthetic procedure is shown in Scheme 2. After
12 was dissolved in NMP, to the solution was added the
Pfp ester of Gln at room temperature. The progress of
the coupling was monitored by TLC, which indicated a
very clean reaction completed in 2 h. Then, diethyl ether
(10 times the volume of NMP) was added to the mixture
to precipitate the expected glycopeptide 13, while the
excessive Gln-Pfp and the side products, such as pen-
tafluorophenol, remained in solution. After the solution
was removed, the precipitate was washed with diethyl
ether to afford 13 (96%) that was practically pure. The
less than perfect yield of this reaction might be due to
the mass loss during the sample transfer. N-Terminal
deprotection of 13 was achieved with 20% piperidine in
NMP, and the product 14 was also isolated by the
precipitation method and purified by washing with
diethyl ether. The coupling protocol described for 13 was
also applied to introduce Gly and give the glycotripeptide
15. Finally, the C-terminus of 15 was deprotected with
15% TFA in DCM at room temperature to yield 16. It
was proved that this treatment did not affect the unpro-
tected carbohydrate moiety. Glycopeptide segment 16
with a free C-terminus was ready to be activated and
coupled to a peptide fragment.
To investigate the coupling reaction between the gly-
copeptide and a peptide fragment, 16 was first linked to
a protected tripeptide 17. The reaction was achieved in
DMF with use of 3 equiv of 17 with DCC/HOBt as the
condensation reagent. TLC indicated a clean and com-
plete reaction. The product was again precipitated out
and washed with diethyl ether, and then deprotected in
two steps to give 19, a partial structure of CD52 glyco-
peptide, that was finally purified by HPLC (71% yield
based on 18). DCU formed during the activation of 16
was soluble in DMF and in the mixture of DMF/ether,
so it was removed from the product during the precipita-
tion and washing process, as proved by the NMR spectra
of crude 18 and 19. The HPLC results and NMR spectra
of 19 did not show any racemization of amino acids or
decomposition of the glycan.
a
Reagents and conditions: (a) AgOTf, MS 4 Å, DCM, -40 °C
to rt, overnight, 75%; (b) TBAF, HOAc, THF, rt, 72 h, 87%; (c)
CuBr2, Bu4NBr, MS 4 Å, DCM/DMF, rt, 2 h, 85%; (d) NH2CH2-
CH2NH2, BuOH, 90 °C, overnight; Ac2O, pyr., rt, overnight;
NaOMe, MeOH, rt, overnight, 86%; (e) Lindlar catalyst, H2; Fmoc-
Asp(OBt)-OtBu, NMP, rt, 2 h, 87%; (f) 20% Pd(OH)2/C, H2, rt, 20
h, 96%.
Stereoselective glycosylation of 3 by 4 was achieved by
using silver triflate (AgOTf) as the promoter.33 After the
removal of tert-butyldiphenylsilyl (TBDPS) in 5, the
R-fucose was introduced to the exposed 6-O-position.
Several promoters were evaluated for this glycosylation,
and CuBr2/Bu4NBr produced the highest stereoselectivity
and yield (85%).34,35 Then, the phthalimido groups in 9
were substituted for acetamido groups by a one-pot
procedure to give 10 (86%).
The azido group at the reducing end of trisaccharide
10 was selectively reduced by hydrogenation with Lindlar
catalyst to give the glycosylamine that was immediately
treated with a freshly prepared active ester, Fmoc-Asp-
(OBt)-OtBu, to form the trisaccharide-Asn conjugate 11
(H-1: δ 4.93, J ) 3.0 Hz; δ 4.76, J ) 8.4 Hz; and δ 4.67,
J ) 8.4 Hz). No R-anomer was observed. The product was
easily purified by silica gel column chromatography. The
benzyl and Fmoc groups were finally removed under a
H2 atmosphere by using Pd(OH)2 as the catalyst to afford
12 (96%), which had a free R-amino group and was ready
for the peptide elongation.
The synthesis of 1 is shown in Scheme 3. After the
N-terminus of 15 was deprotected and capped by an
acetyl group, its tert-butyl group, as well as the trityl
group, was removed to give 21. Meanwhile, the free
nonapeptide segment 24 was prepared by the conven-
tional solid-phase method with use of Wang resin and
Fmoc chemistry. Because 24 had two free carboxyl groups
that may interfere with the reaction, instead of utilizing
Solu tion -P h a se Syn th esis of CD52 Glycop ep tid es
w ith Solid -P h a se Wor k u p s. Our synthetic plan was
(33) Suzuki, K.; Maeta, H.; Matsumoto, T. Tetrahedron Lett. 1989,
30, 4853-4856.
(34) Lemieux, R. U.; Hendriks, K. B.; Stick, R. V.; J ames, K. J . Am.
Chem. Soc. 1975, 97, 4056-4062.
(35) Dan, A.; Lergenmu¨ller, M.; Amano, M.; Nakahara, Y.; Ogawa,
T.; Ito, Y. Chem. Eur. J . 1998, 4, 2182-2190.
J . Org. Chem, Vol. 68, No. 23, 2003 9005

Shao et al.
SCHEME 2a
a
Reagents and conditions: (a) Fmoc-Gln(Trt)-OPfp, NMP, rt, 2 h, 96%; (b) 20% piperidine/NMP, rt, 1 h, 98%; (c) Fmoc-Gly-OPfp,
NMP, rt, 2 h, 98%; (d) TFA/DCM/Et3SiH (3/17/2), rt, 2 h, >98%; (e) DCC, HOBt, NMP, 0 °C to rt, overnight, 93%; (f) 20% TFA/DCM, rt,
2.5 h; (g) 20% piperidine/NMP, rt, 1 h, 71% (two steps overall, after HPLC).
SCHEME 3a
a
Reagents and conditions: (a) 20% piperidine/NMP, rt, 1 h; (b) Ac2O/MeOH, 91% (2 steps); (c) 20% TFA/DCM, rt, 2 h; (d) DCC, PfpOH,
NMP, rt, 5 h; (e) NMP, rt, overnight, 42% (3 steps after HPLC, 30% of 21 recovered).
the one-pot coupling protocol described for 19, we had to
use the active ester of 21 free of DCC. Thus, 21 was
treated with pentafluorophenol and DCC in NMP to form
22. The reaction was monitored with TLC. After precipi-
tation and washing with diethyl ether, the obtained
product 22 was essentially DCC and DCU-free. As such,
the new strategy enabled the partial purification of the
active ester without passing through a column. 22 was
then coupled with 24 (3 equiv) in NMP to give 1, which
was precipitated out and finally purified by HPLC. 1 was
obtained in a 42% yield, and its structure was character-
ized by 1D, 2D NMR and MS. As shown by the HPLC
diagram (Figure 2a), 1 (Rt ) 25.3 min) was easily
separated from the unreacted glycotripeptide 21 (30%
recovered, Rt ) 8.1 min), the nonapeptide 24 (Rt ) 12.2
min), and a side product (Rt ) 29.3 min, ca. 12%). The
NMR data for 1 and 19, including the chemical shifts and
coupling constants of all reporter signals, listed in Table
1 further support the structural assignment.
On the other hand, the side product (HPLC: Rt ) 29.3
min) was identified as a diastereoisomer of 1, probably
having the asparagine racemized. The racemization could
happen during the activation of 21, but it could also
happen during the reaction between the active ester 22
9006 J . Org. Chem., Vol. 68, No. 23, 2003

CD52 Glycopeptides Containing the Acid-Labile Fucosyl Linkage
F IGURE 2. HPLC diagrams of the final product (1): (a) solution-phase synthesis with PfpOH/DCC as the coupling reagent; (b)
solution-phase synthesis with HOBt/DCC as the coupling reagent; (c) solid-phase synthesis. HPLC conditions: semipreparative
C18 column (1.5 cm × 25 cm); eluent, 0.1% TFA and 2% i-PrOH in H2O; flow rate, 2 mL/min; dual wavelength detection at 200
and 220 nm.
and the free peptide 24. Trying to answer this question,
the coupling reaction between 21 and 24 with HOBt/DCC
as the condensation reagent was examined.
not give the racemization product suggests that the
racemization was not caused by the activation procedure.
Instead, it might result from the slower reaction of the
free peptide and/or the prolonged workup procedure,
which can also explain why the Bt ester, which is less
stable than Pfp ester, gave more of the side product.
Con ver gen t Solid -P h a se Syn th esis of CD52 Gly-
cop ep tid e. The target glycopeptide 1 was also prepared
by coupling the unprotected glycotripeptide segment
obtained via solution-phase synthesis to the nonapeptide
segment on the 2-chlorotrityl resin36 (Scheme 4). The
ester bond that attaches the glycopeptide to this resin
After the reaction of 21 and HOBt/DCC was finished,
the active ester was isolated by precipitation and washed
with diethyl ether. It was then dissolved in NMP and
reacted with 24. HPLC analysis (Figure 2b) of the product
indicated a higher conversion of 21, while the synthetic
target 1 was formed in a yield similar to that with Pfp
ester. The side product was obtained in obviously higher
yield (ca. 25%). The fact that the one-pot reaction between
16 and 17 with HOBt/DCC as the coupling reagent did
J . Org. Chem, Vol. 68, No. 23, 2003 9007

Shao et al.
TABLE 1. Selected NMR Da ta of 1, 19 a n d th e Sid e
P r od u ct
HOBt in NMP by the one-pot protocol to afford the resin-
bound glycopeptide 27. The coupling reaction yield was
no less than 76%, based on comparing the weight gain
of the resin, which reflected the quantity of 21 attached
to the resin, to the actual peptide loading on the resin.
27 was then treated with HOAc/TFE/DCM (1/1/8) to
release the glycopeptide 28 (92%). This condition was
mild enough to leave the tBu and Trt groups intact, so
the product could be purified by silica gel column chro-
matography. Finally, treatment of 28 with TFA/DCM/
Et3SiH afforded 1 in a good yield (70% after HPLC). Its
spectroscopic and HPLC properties were identical with
those of the product prepared by solution-phase synthe-
sis. HPLC (Figure 2c) showed the same racemization
product obtained above with solution-phase synthesis. Its
proportion was between that shown in panels a and b of
Figure 2, which may suggest that the slower reaction
between the active ester and resin-bound peptide is the
main cause of active ester reacemization.
In summary, CD52 glycopeptides with a fucosylated
trisaccharide were prepared by convergent solution-phase
and solid-phase syntheses. Homogeneous CD52 glyco-
peptide samples were obtained in sufficient amounts for
their structural and structure-activity relationship stud-
ies. The acid-labile fucose linkage proved to be stable to
all reactions involved. The results of coupling between
the glycopeptide and peptide segments varied. The reac-
tion of the glycotripeptide and a protected peptide 17 in
solution with HOBt/DCC as the coupling reagent gave
excellent yield of the expected product, while a similar
reaction in solid-phase synthesis resulted in some race-
mization products. Glycopeptide activation with PfpOH/
DCC seemed to cause fewer problems than that with
DCC/HOBt. Nonetheless, the final products were easily
purified. The advantage of the solution-phase synthetic
chemical shift (J in Hz)
proton signal
1
19
side product
Glc-1
Glc-1
5.03 (9.6)
4.66 (8.4)
4.88 (3.6)
4.12 (6.0)
1.21
5.04 (9.6)
4.66 (8.4)
4.88 (4.2)
4.13 (6.6)
1.20 (6.6)
4.33 (9.6)
2.10
5.04
4.66
Fuc-1
Fuc-5
Fuc-6
Glu2-R
Glu2-â
4.11
1.21
4.31
2.19-1.96
2.19-1.96
2.40-2.29
4.73 (6.0, 7.2)
2.87 (6.0, 13)
2.78 (7.2, 13)
4.79 (6.0, 7.8)
2.95 (6.0, 17)
2.84 (7.8, 17)
4.38 (4.2)
4.30
1.21
4.82 (6.6, 6.6)
4.26
4.41
4.52, 4.47, 4.45
4.50
2.19
2.13
2.40
2.00
Glu2-γ
Asn3-R
Asn3-â
2.38 (7.8)
4.76 (6.0, 7.2)
2.85 (7.2, 17)
2.79 (7.2, 17)
4.80 (6.0, 7.2)
2.93 (6.0, 16)
2.88 (6.0, 16)
4.37 (4.2)
4.24 (6.6, 4.2)
1.20 (6.6)
4.96 (6.0, 6.0)
-
2.90
2.73
Asp4-R
Asp4-â
3.00
2.87
Thr5-R
Thr5-â
Thr5-γ
Ser6-R
Glu7-R
Thr8-R
Ser9,10,12-R
Pro11-R
Ac
-
-
-
2.07, 2.07, 2.01
2.08, 2.02
can be easily cleaved under very mild acidic conditions
that would not affect oligosaccharide structures.
The synthesis started from the commercially available
Ser-loaded resin 25. Conventional Fmoc chemistry was
utilized to construct the nonapeptide on an automatic
peptide synthesizer with 2-(1H-9-azabenzotriazole-1-yl)-
1,1,3,3-tetramethyluronium hexafluorophosphate (HATU)
as the condensation reagent. Thereafter, the glycotri-
peptide 21 was coupled to 26 under the influence of DCC/
SCHEME 4a
a
Reagents and conditions: (a) 21, DCC, HOBt, NMP, rt, overnight, 76%; (b) HOAc/TFE/DCM (1/1/8), rt, 2 h, 92%; (c) TFA/DCM/
Et3SiH (4/16/2), rt, 3 h, 70% (after HPLC).
9008 J . Org. Chem., Vol. 68, No. 23, 2003

CD52 Glycopeptides Containing the Acid-Labile Fucosyl Linkage
added and the reaction mixture was stirred at room temper-
ature overnight. The solution was filtered off, washed with
brine, and dried and then concentrated in a vacuum. Column
chromatography of the residue gave 9 as a white solid (1.06 g,
method is that it can avoid operations such as the
retrieval of glycopeptides and the deprotection of peptide
side chains, and it is applicable to the preparation of large
glycoproteins. The advantage of solid-phase glycopep-
tide-peptide coupling is that the unreacted glycotripep-
tide can be readily recovered and reused. The best result
was obtained with a one-pot protocol of solution-phase
coupling between the glycopeptide and a protected pep-
tide. Nevertheless, the optimization of the coupling
protocols and the application of solution-phase glycopep-
tide-peptide coupling to the preparation of complex
glycoproteins is under further study.
1
85%). [R]D -58.1 (c 1.0, CHCl3). H NMR (600 MHz, CDCl3) δ
7.91-7.90 (m, 2 H), 7.78-7.76 (m, 3 H), 7.68-7.64 (m, 2 H),
7.56-7.54 (m, 3 H), 7.48-7.40 (m, 4 H), 7.36-7.23 (m, 9 H),
6.98-6.96 (m, 2 H), 6.79-6.78 (m, 3 H), 5.69 (dd, J ) 9.6, 10.8
Hz, 1 H), 5.60 (d, J ) 8.4 Hz, 1 H), 5.15-5.12 (m, 3 H), 5.10
(d, J ) 3.6 Hz, 1 H), 5.05 (dd, J ) 9.6, 10.2 Hz, 1 H), 4.98 (d,
J ) 11.4 Hz, 1 H), 4.95 (d, J ) 11.4 Hz, 1 H), 4.87 (d, J ) 11.4
Hz, 1 H), 4.82 (dd, J ) 11.4 Hz, 12.0, 2 H), 4.63 (d, J ) 11.4
Hz, 1 H), 4.44 (d, J ) 13.2 Hz, 1 H), 4.30-4.26 (m, 2 H), 4.16-
4.08 (m, 3 H), 4.02-3.95 (m, 3 H), 3.90 (dd, J ) 2.4, 12.0 Hz,
1 H), 3.78 (d, J ) 10.2 Hz, 1 H), 3.70 (d, J ) 2.4 Hz, 1 H),
3.41-3.37 (m, 2 H), 3.30 (dd, J ) 2.4, 10.8 Hz, 1 H), 1.90 (s, 3
H), 1.85 (s, 3 H), 1.82 (s, 3 H), 1.07 (d, J 6.6 Hz, 3 H). FAB-MS
calcd for C68H67N5O19 1257.4, found 1256.7 (M + H+).
Exp er im en ta l Section
O-(2,3,4-Tr i-O-a cet yl-2-d eoxy-2-p h t h a lim id o-â-D-glu -
copyr an osyl)-(1f4)-6-O-(ter t-bu tyldiph en ylsilyl)-3-O-ben -
zyl-2-d eoxy-2-p h th a lim id o-â-D-glu cop yr a n osyl Azid e (5).
To a solution of 2,3,4-tri-O-acetyl-2-deoxy-2-phthalimido-â-D-
glucopyranosyl acetate (3.74 g, 7.8 mmol) in DCM (20 mL) was
added 33% HBr in HOAc (3.0 mL) at 0 °C. After being stirred
at room temperature for 4 h, the reaction mixture was washed
with saturated NaHCO3 solution and brine, dried over Na2-
SO4, and concentrated in a vacuum to give the glycosyl bromide
4 as a syrup that was directly used without further purifica-
tion. After a mixture of 4, 3 (2.5 g, 3.8 mmol), and MS 4 Å (4
g) in dry DCM (20 mL) was stirred at room temperature for 2
h, it was cooled to -40 °C. To this mixture was added AgOTf
(2.33 g, 9.0 mmol). The mixture was then warmed to room
temperature and stirred overnight. The solution was filtered
off, washed with brine, dried over Na2SO4, and then concen-
trated in a vacuum. Column chromatography of the residue
afforded 5 as a white foam (3.0 g, 75%). [R]D +13.6 (c 1.0,
CHCl3). 1H NMR (200 MHz, CDCl3) δ 7.88-7.65 (m, 13 H),
7.60-7.40 (m, 5 H), 7.07-7.02 (m, 2 H), 6.88-6.82 (m, 3 H),
5.90 (dd, J ) 9.0, 10.8 Hz, 1 H), 5.72 (d, J ) 8.4 Hz, 1 H),
5.24-5.14 (m, 2 H), 4.88 (d, J ) 12.4 Hz, 1 H), 4.57-4.34 (m,
4 H), 4.20 (dd, J ) 8.4, 10.6 Hz, 1 H), 4.07 (dd, J ) 2.0, 12.2
Hz, 1 H), 4.01 (dd, J ) 9.4, 10.5 Hz, 1 H), 3.90-3.70 (m, 3 H),
3.28 (d, J ) 9.5 Hz, 1 H), 2.02 (s, 3 H), 1.87 (s, 3 H), 1.15 (s, 9
H).
O-(2,3,4-Tr i-O-a cet yl-2-d eoxy-2-p h t h a lim id o-â-D-glu -
cop yr a n osyl)-(1f4)-3-O-ben zyl-2-d eoxy-2-p h th a lim id o-â-
D-glu cop yr a n osyl Azid e (6). After 5 (3.0 g, 2.8 mmol) was
dissolved in THF (15 mL), to the solution were added TBAF
(1 M solution in THF, 5.0 mL, 5.0 mmol) and acetic acid (300
µL, 5.0 mmol) at 0 °C. The solution was then warmed to room
temperature and stirred for 72 h. The reaction mixture was
diluted with EtOAc (200 mL), and the organic layer was
washed with saturated NaHCO3 (150 mL) and water (150 mL),
dried over Na2SO4, and finally concentrated under vacuum to
dryness. The residue was purified on a silica gel column to
afford 6 as white foam (2.05 g, 87%). [R]D +10.8 (c 1.0, CHCl3).
1H NMR (300 MHz, CDCl3) δ 7.94-7.65 (m, 8 H), 7.14-6.98
(m, 2 H), 6.91-6.82 (m, 3 H), 5.81 (dd, J ) 9.1, 10.6 Hz, 1 H),
5.67 (d, J ) 8.4 Hz, 1 H), 5.17 (dd, J ) 9.2, 9.2 Hz, 1 H), 4.87
(d, J ) 12.6 Hz, 1 H), 4.44 (d, J ) 12.7 Hz, 1 H), 4.39-4.26
(m, 3 H), 4.14-3.96 (m, 3 H), 3.87-3.81 (m, 1 H), 3.74-3.69
(m, 1 H), 3.50-3.40 (m, 2 H), 2.01 (s, 3 H), 1.99 (s, 3 H), 1.85
(s, 3 H). HRFAB-MS calcd for C41H39N5O15Na (M + Na+)
864.2340, found 864.2332.
O-(2-Deoxy-2-a ceta m id o-â-D-glu cop yr a n osyl)-(1f4)-[O-
(2,3,4-tr i-O-ben zyl-r-L-fu cop yr a n osyl)-(1f6)]-3-O-ben zyl-
2-d eoxy-2-a ceta m id o-â-D-glu cop yr a n osyl Azid e (10). After
the mixture of 9 (1.06 g, 0.78 mmol), ethylenediamine (2 mL,
29 mmol), and n-butanol (10 mL) was stirred at 90 °C
overnight, it was concentrated to dryness under vacuum. The
residue was dissolved in Ac2O/pyridine (16 mL, 1/2) and stirred
overnight at room temperature. After the reaction mixture was
diluted with EtOAc (150 mL), the solution was washed with
saturated NaHCO3 (80 mL) and water (50 mL), dried over Na2-
SO4, and concentrated in a vacuum to dryness. The residue
was purified on a silica gel column. After the product was
treated with MeOH (10 mL) containing NaOMe (0.05 M) at
room temperature for 3 h, it was neutralized to pH 6-7 with
Amberlyst 15 resin. The resin was filtered off and washed with
MeOH, and the washings were combined and concentrated.
The crude product was purified on a silica gel column to give
10 as a white foam (640 mg, 86%). [R]D -64.5 (c 1.0, CH3OH).
1H NMR (600 MHz, CD3OD) δ 7.40-7.21 (m, 20 H), 5.04 (d, J
) 10.8 Hz, 1 H), 4.89 (d, J ) 3.6 Hz, 1 H), 4.86-4.76 (m, 6 H),
4.62 (d, J ) 11.4 Hz, 1 H), 4.54 (d, J ) 10.8 Hz, 1 H), 4.48 (d,
J ) 9.0 Hz, 1 H), 4.10 (q, J ) 6.6 Hz, 1 H), 4.04 (dd, J ) 9.6,
9.6 Hz, 1 H), 4.01 (dd, J ) 3.0, 10.8 Hz, 1 H), 3.96 (dd, J )
3.0, 10.2 Hz, 1 H), 3.88-3.72 (m, 6 H), 3.60-3.40 (m, 5 H),
3.25 (dd, J ) 9.6, 9.6 Hz, 1 H), 1.97 (s, 3 H), 1.91 (s, 3 H), 1.10
(d, J ) 6.6 Hz, 3 H). HRFAB-MS calcd for C50H62O5N14 (M +
H+) 956.4293, found 956.4307.
Nr-(9-F lu r oen ylm et h oxyca r b on yl)-Nγ-[O-(2-d eoxy-2-
a ceta m id o-â-D-glu cop yr a n osyl)-(1f4)-[O-(2,3,4-tr i-O-ben -
zyl-r-L-fu cop yr a n osyl)-(1f6)]-3-O-ben zyl-2-d eoxy-2-a ce-
tam ido-â-D-glu copyr an osyl]-L-aspar agin e ter t-Bu tyl Ester
(11). To a solution of Fmoc-Asp-OtBu in CH2Cl2 (10 mL) and
NMP (1 mL) was subsequently added HOBt and DCC at room
temperature. The mixture was stirred at room temperature
for 1 h to give the active ester solution after filtration to remove
the solid DCU. A mixture of 10 (110 mg, 0.115 mmol), Lindlar
catalyst (110 mg), and MeOH (5 mL) was stirred at room
temperature for 2.5 h under a H2 atmosphere. After filtration
through Celite, the solution was concentrated in a vacuum,
and to the residue were added CH2Cl2 (2 mL) and the freshly
prepared active ester. The mixture was stirred at room
temperature overnight, concentrated, and purified on a silica
gel column to give 11 as a white foam (132 mg, 87%). [R]D
1
-26.2 (c 1.0, CH2Cl2/MeOH 1/1). H NMR (600 MHz, CD3OD)
O-(2,3,4-Tr i-O-a cet yl-2-d eoxy-2-p h t h a lim id o-â-D-glu -
cop yr a n osyl)-(1f4)-[O-(2,3,4-tr i-O-ben zyl-r-L-fu cop yr a -
n osyl)-(1f6)]-3-O-ben zyl-2-d eoxy-2-p h th a lim id o-â-D-glu -
cop yr a n osyl Azid e (9). After the mixture of 8 (957 mg, 2.0
mmol), 6 (670 mg, 0.8 mmol), Bu4NBr (650 mg, 2.0 mmol), and
MS 4 Å (4 g) in dry CH2Cl2/DMF (18 mL, 5/1) was stirred at
room temperature for 2 h, CuBr2 (450 mg, 2.0 mmol) was
δ 7.77 (d, J ) 7.2 Hz, 2 H), 7.61 (d, J ) 7.8 Hz, 2 H), 7.38-
7.22 (m, 24 H), 5.04 (d, J ) 10.8 Hz, 1 H), 4.97 (d, J ) 10.2
Hz, 1 H), 4.93 (d, J ) 3.0 Hz, 1 H), 4.88-4.74 (m, 4 H), 4.76
(d, J ) 8.4 Hz, 1 H), 4.67 (d, J ) 8.4 Hz, 1 H), 4.62 (d, J )
11.4 Hz, 1 H), 4.54 (d, J ) 11.4 Hz, 1 H), 4.42 (dd, J ) 5.4, 6.0
Hz, 1 H), 4.33 (dd, J ) 7.2, 10.8 Hz, 1 H), 4.24 (dd, J ) 7.2,
10.8 Hz, 1 H), 4.16 (dd, J ) 7.2, 6.6 Hz, 1 H), 4.10 (q, J ) 6.6
Hz, 1 H), 3.98-3.90 (m, 4 H), 3.86-3.74 (m, 5H), 3.58 (dd, J
) 9.6, 9.6 Hz, 1 H), 3.53-3.17 (m, 5 H), 2.63 (d, J ) 6.6 Hz, 2
(36) Barlos, K.; Gatos, D.; Kapolos, S.; Papaphotiu, G.; Schaefer, W.;
Yao, W. Tetrahedron Lett. 1989, 30, 3947-3950.
J . Org. Chem, Vol. 68, No. 23, 2003 9009

Shao et al.
H), 1.93 (s, 3 H), 1.85 (s, 3 H), 1.41 (s, 9 H), 1.12 (d, J ) 6.6
Hz, 3 H). FAB-MS calcd for C73H86O4N19 1322.6, found 1323.3
(M + H+).
freeze-drying. HPLC: Rt ) 16.5 min. 1H NMR (600 MHz, D2O,
DHO at δ 4.79 as reference, 25 °C) δ 5.04 (d, J ) 9.6 Hz, 1 H,
GlcNAc-1), 4.96 (dd, J ) 6.0, 6.0 Hz, 1 H, Ser-R), 4.88 (d, J )
4.2 Hz, 1 H, Fuc-1), 4.80 (dd, J ) 6.0, 7.2 Hz, 1 H, Asp-R),
4.76 (dd, J ) 6.0, 7.2 Hz, 1 H, Asn-R), 4.66 (d, J ) 8.4 Hz, 1 H,
GlcNAc-1), 4.37 (d, J ) 4.2 Hz, 1 H, Thr-R), 4.33 (dd, J ) 6.0,
8.4 Hz, 1 H, Glu-R), 4.24 (dt, J ) 6.0, 11.4 Hz, 1 H, Thr-â),
4.13 (t, J ) 6.6 Hz, 1 H, Fuc-5), 3.94-3.73 (m, 13 H), 3.70-
3.65 (m, 2 H), 3.56 (dd, J ) 4.8, 6.6 Hz, 1 H), 3.51-3.45 (m, 2
H), 3.16 (s, 3 H, NMe), 2.96 (s, 3 H, NMe), 2.93 (dd, J ) 6.0,
16.2 Hz, 1 H, Asp-â), 2.85 (dd, J ) 7.2, 16.8 Hz, 1 H, Asn-â),
2.88 (dd, J ) 6.0, 16.2 Hz, 1 H, Asp-â), 2.79 (dd, J ) 7.2, 16.8
Hz, 1 H, Asn-â), 2.38 (t, J ) 7.8 Hz, 2 H, Glu-γ), 2.10 (m, 1 H,
Glu-â), 2.08 (s, 3H, Ac), 2.02 (s, 3 H, Ac), 2.00 (m, 1 H, Glu-â),
1.20 (2 d, J ) 6.6 Hz, 6 H, Thr-γ and Fuc-6). MALDI-TOF-MS
calcd for C46H77O11N26 1200.5, found 1223.8 (M + Na+), 1239.6
(M + K+).
Glycop ep tid e 20. Glycopeptide 15 (16 mg, 0.0115 mmol)
was treated with piperidine/DMF (1 mL) at room temperature
for 1 h, and the deprotected product was precipitated out by
Et2O (10 mL). To the crude product were added MeOH (2 mL)
and Ac2O (0.1 mL), and the reaction mixture was stirred at
room temperature for 2 h. Addition of Et2O (20 mL) to the
reaction mixture gave 20 as a white precipitate (12.6 mg, 91%).
Nγ-[O-(2-Deoxy-2-acetam ido-â-D-glu copyr an osyl)-(1f4)-
[O-(r-L-fu cop yr a n osyl)-(1f6)]-2-d eoxy-2-a ceta m id o-â-D-
glu cop yr a n osyl]-L-a sp a r a gin e ter t-Bu tyl Ester (12). After
11 (132 mg, 0.1 mmol) in MeOH (6 mL) and H2O (2 mL) was
stirred with Pd(OH)2 (50 mg) under a H2 atmosphere at room
temperature for 20 h, the mixture was filtered off through a
Celite pad, and the catalyst was washed with water. The
filtrates were combined and concentrated in a vacuum to
remove MeOH. The resulting water solution (5 mL) was
washed with Et2O (10 mL) twice and lyophilized to afford 12
(71 mg, 96%) as a white solid. 1H NMR (300 MHz, D2O) δ 5.04
(d, J ) 9.6 Hz, 1 H), 4.89 (d, J ) 3.9 Hz, 1 H), 4.67 (d, J ) 8.5
Hz, 1 H), 4.29 (dd, J ) 4.4, 4.7 Hz, 1 H), 4.12 (q, J ) 6.4 Hz,
1 H), 4.95-3.46 (m, 15 H), 3.10 (dd, J ) 4.9, 17.5 Hz, 1 H),
2.92 (dd, J ) 4.5, 17.5 Hz, 1 H), 2.08 (s, 3 H), 2.03 (s, 3 H),
1.48 (s, 9 H), 1.21 (d, J ) 6.6 Hz, 3 H). HRFAB-MS calcd for
C
30H53O4N17 (M + H+) 741.3406, found 741.3411.
Glycop ep tid e 16. To the solution of 12 (92 mg, 0.125 mmol)
in NMP (2 mL) was added Fmoc-Gln(Trt)-OPfp (284 mg, 0.375
mmol), and the solution was stirred at room temperature for
2 h. Et2O (20 mL) was added to the solution with gentle
stirring to get a white precipitate. The mixture was then
centrifuged and the solvent removed by pipet. The precipitate
was washed with Et2O (5 mL) twice. After being dried under
a flash of nitrogen, the crude product 13 (160 mg, 96%) was
dissolved in NMP (1.6 mL) and piperidine (0.4 mL). The
reaction mixture was stirred at room temperature for 1 h. The
deprotection product was again precipitated and washed
following the same procedure described above to give the crude
product as a white solid (130 mg, 97.5%). This product was
again dissolved in NMP (2 mL) and Fmoc-Gly-OPfp (216 mg,
0.466 mmol) was added. Two hours later precipitation of the
product gave glycotripeptide 15 as a white solid (160 mg, 98%).
After the crude 15 (10 mg) was dissolved in TFA/CH2Cl2/Et3-
SiH (0.3/1.7/0.2 mL) and stirred at room temperature for 2 h,
toluene (4 mL) was added. The mixture was concentrated
under vacuum, and the residue was dissolved in H2O (5 mL)
and washed with Et2O (5 mL) twice. The water solution was
finally lyophilized to give 16 as a white solid, which was used
in the next step of reaction without further purification.
Tr ip ep tid e 17. Tripeptide 17 was prepared by conventional
solution-phase synthesis using the DCC/HOBt protocol and
Fmoc chemistry. [R]D +38.5 (c 1.0, CHCl3). 1H NMR (600 MHz,
CDCl3, 25 °C) δ 7.75 (d, J ) 7.2 Hz, 2 H), 7.68 (d, J ) 7.2 Hz,
1 H), 7.59 (dd, J ) 6.6, 6.6 Hz, 2 H), 7.48 (d, J ) 5.4 Hz, 1 H),
7.39 (dd, J ) 7.2, 7.8 Hz, 2 H), 7.32-7.30 (m, 2 H), 5.98 (d, J
) 8.4 Hz, 1 H), 5.01 (dd, J ) 7.8, 13.2 Hz, 1 H), 4.58 (br s, 1
H), 4.39 (br s, 2 H), 4.31 (dd, J ) 3.6, 6.0 Hz, 1 H), 4.24 (dd,
J ) 6.6, 7.2 Hz, 1 H), 4.16 (br s, 1 H), 3.54 (dd, J ) 5.4, 8.4
Hz, 1 H), 3.39 (dd, J ) 7.8, 8.4 Hz, 1 H), 3.09 (s, 3 H), 2.95 (s,
3 H), 2.95-2.92 (m, 2 H), 2.64 (dd, J ) 5.4, 16.8 Hz, 1 H), 1.45
(s, 9 H), 1.27 (s, 9H), 1.11 (s, 9 H), 1.03 (d, J ) 6.0 Hz, 3 H).
HRFAB-MS calcd for C40H59O9N4 (M + H+) 739.4282, found
739.4286.
Glycop ep tid e 19. To a mixture of glycopeptide 16 (5 mg,
0.0046 mmol), tripeptide 17 (15 mg, 0.029 mmol), and HOBt
(7 mg, 0.052 mmol) in DMF (1 mL) was added DCC (10 mg,
0.048 mmol) at 0 °C under N2 atmosphere. After the solution
was stirred at room temperature overnight, the product was
precipitated out by Et2O to give 18 as a white solid (6.77 mg,
93%). It was then treated with TFA/CH2Cl2 (0.4/1.6 mL) at
room temperature for 2.5 h, whereupon toluene (4 mL) was
added and the solution was concentrated in a vacuum. The
residue was treated with piperidine/DMF (0.2/0.8 mL) at room
temperature for 1 h, and addition of Et2O (10 mL) to the
reaction mixture afforded the crude product that was further
purified by HPLC with use of a C18 column (1.5 cm × 25 cm)
with 1.5% i-PrOH in water containing 0.1% TFA as the eluent
(2 mL/min) to give 19 as a white solid (3.6 mg, 71%) after
Non a p ep tid e 24. An Fmoc-protected nonapeptide linked
to resin (648 mg) was synthesized after eight cycles of the
standard synthesizer program of condensation with HATU-
DIEA activated Fmoc amino acid (1 mmol each) starting from
Fmoc-Ser-loaded HMP resin (325 mg, 0.25 mmol). A part of
the resin (34 mg) was treated with 20% piperidine in NMP at
room temperature for 2 h, and then the resin was washed with
NMP, DCM, and Et2O. It was treated with a mixture of TFA
(1.9 mL), HSCH2CH2SH (0.05 mL), and water (0.05 mL) at
room temperature for 2 h. The mixture was filtered off and
the filtrate was concentrated under vacuum. The residue was
suspended in water (5 mL) and extracted with Et2O (3 × 5
mL). The water solution was concentrated and dissolved in
methanol/AcOH (0.5 mL/0.1 mL). On addition of Et2O (15 mL),
24 was obtained as a white solid (14 mg) after washing with
1
Et2O and drying. H NMR (600 MHz, D2O, DHO at δ 4.79 as
reference, 25 °C) δ 4.83 (m, 2 H), 4.54-4.40 (m, 9 H), 4.29-
4.25 (m, 2 H), 3.96-3.73 (m, 10 H), 3.01 (dd, J ) 5.4, 18.0 Hz,
1 H), 2.93 (dd, J ) 7.2, 18.0 Hz, 1 H), 2.39 (dd, J ) 7.2, 7.8
Hz, 2 H), 2.32 (m, 1 H), 2.17 (m, 1 H), 2.07-1.99 (m, 4 H),
1.25 (d, J ) 6.0 Hz, 3 H), 1.21 (d, J ) 6.0 Hz, 3 H). MALDI-
TOF-MS calcd for C34H56O19N10 908.9, found 932.3 (M + Na+).
Glycop ep tid e 1. The tert-butyl and trityl groups of 20 (5
mg, 0.005 mmol) were removed with 20% TFA/DCM as
described for 16 to afford 21, which was directly used in the
next step of reaction without further purification. To the
mixture of 21 obtained above and pentafluorophenol (9 mg,
0.05 mmol) in NMP (1 mL) was added DCC (21 mg, 0.1 mmol),
and the solution was stirred at room temperature for 5 h. The
reaction product was precipitated out by Et2O followed by
washing. After the crude product was dissolved in NMP, the
free nonapeptide 24 (10 mg, 0.011 mmol) was added, and the
reaction mixture was stirred at room temperature for 12 h.
Et2O was added to the reaction mixture to give the solid
product, which was finally purified by HPLC with use of a C18
column (1.5 cm × 25 cm) with 2% i-PrOH in water containing
0.1% TFA as the eluent (2 mL/min). A part of the nonapeptide
24 (5 mg, 0.005 mmol, Rt ) 12.2 min) and the glycotripeptide
21 (1.9 mg, 0.0015 mmol, Rt ) 8.1 min) were recovered. In
addition to the synthetic target 1 (Rt ) 25.3 min) obtained as
a white solid after freeze-drying (3.8 mg, 0.0021 mmol, 42%),
HPLC also gave another product (Rt ) 29.3 min, 1.0 mg, 12%)
1
identified as a stereoisomer of 1. 1: H NMR (600 MHz, D2O,
DHO at δ 4.79 as reference, 25 °C): δ 5.03 (d, J ) 9.6 Hz, 1
H, GlcNAc-1), 4.88 (d, J ) 3.6 Hz, 1 H, Fuc-1), δ 4.82 (dd, J )
6.6, 6.6 Hz, 1 H, Ser-R), 4.79 (dd, J ) 7.8, 7.8 Hz, 1 H, Asp-R),
4.73 (dd, J ) 6.0, 7.2 Hz, 1 H, Asn-R), 4.66 (d, J ) 8.4 Hz, 1 H,
GlcNAc-1), 4.52 (dd, J ) 5.4, 10.8 Hz, 1 H, Ser-R), 4.50 (dd, J
9010 J . Org. Chem., Vol. 68, No. 23, 2003

CD52 Glycopeptides Containing the Acid-Labile Fucosyl Linkage
) 4.2, 3.6 Hz, 1 H, Pro-R), 4.47 (dd, J ) 5.4, 5.4 Hz, 1 H, Ser-
R), 4.45-4.43 (m, 2 H, Thr-â and Ser-R), 4.41 (d, J ) 4.2 Hz,
1 H, Thr-R), 4.38 (d, J ) 4.2 Hz, 1 H, Thr-R), 4.31-4.29 (m, 2
H, Thr-â and Glu-R), 4.26 (dd, J ) 4.8, 6.0 Hz, 1 H, Glu-R),
4.12 (q, J ) 6.0 Hz, 1 H, Fuc-5), 3.97-3.45 (m, 29H), 2.95 (dd,
J ) 6.0, 16.8 Hz, 1 H, Asp-â), 2.89-2.82 (m, 2 H, Asn-â and
Asp-â), 2.78 (dd, J ) 7.2, 12.8 Hz, 1 H, Asn-â), 2.40-2.29 (m,
5 H), 2.19-1.96 (m, 7 H), 2.07 (s, 3 H, Ac), 2.07 (s, 3H, Ac),
2.01 (s, 3 H, Ac), 1.22-1.19 (m, 9 H, 2 Thr-γ and Fuc-6). HMQC
ture for 2 h. After the mixture was washed and dried, the resin
was mixed with 21 (15 mg, 0.016 mmol) and HOBt (15 mg,
0.11 mmol) in NMP (1 mL). The mixture was cooled to 0 °C
under N2, and then DCC (30 mg, 0.14 mmol) was added. The
reaction mixture was allowed to warm to room temperature
and stirred overnight. The resin was washed with NMP and
Et2O thoroughly and dried under vacuum to give a yellowish
resin 27 (58 mg, 76%).
Glycop ep tid e 28. After the resin-bonded glycopeptide 27
was treated with HOAc/TFE/DCM (0.1 mL/0.1 mL/0.8 mL) at
room temperature for 2 h, it was filtered off and the filtrate
was concentrated. The residue was purified on a silica gel
column (eluent MeOH/DCM/EtOAc 1/1/1) to give glycopeptide
28 (28 mg, 92%). MALDI-TOF-MS calcd for C116H181O39N17
2436.3, found 2459.4 (M + Na+).
Glycop ep tid e 1. After glycopeptide 28 (4 mg, 1.64 µmol)
was treated with TFA/DCM/Et3SiH (0.4/1.6/0.2 mL) at room
temperature for 3 h, to the mixture was added toluene (5 mL).
The mixture was concentrated in a vacuum, and the residue
was dissolved in water (5 mL) and extracted with Et2O (3 × 3
mL). After the water solution was lyophilized, the product was
purified by HPLC as described above to give glycopeptide 1 (2
mg, 70%) as a white solid.
1
(13C 150 MHz, H 600 MHz, D2O, 25 °C) δ 80.6/5.03 (GlcNAc-
1), 101.5/4.88 (Fuc-1), 55.8/4.82 (Ser-R), 53.0/4.79 (Asp-R), 52.2/
4.73 (Asn-R), 103.1/4.66 (GlcNAc-1), 57.6/4.52 (Ser-R), 62.9/
4.50 (Pro-R), 58.2/4.47 (Ser-R), 55.2/4.45 (Thr-â), 61.3/4.43 (Ser-
R), 61.5/4.41 (Thr-R), 58.7/4.38 (Thr-R), 69.1/4.32 (Gln-R), 55.5/
4.31 (Thr-â), 69.2/4.26 (Gln-R), 69.0/4.12 (Fuc-5), 38.6/2.87
(Asp-â), 38.5/2.79 (Asn-â), 33.2/2.37 (Gln-γ), 31.4/2.31 (Pro),
28.9/2.18 (Gln-γ), 28.8/2.09 (Gln-â), 24.3/2.07 (Ac), 23.8/2.07
(Ac), 26.6/2.05 (Pro), 31.4/2.03 (Pro), 24.2/2.01 (Ac), 28.8/1.99
(Gln-â), 20.8/1.23 (Thr-γ), 17.5/1.22 (Thr-γ), 28.4/1.19 (Fuc-6).
MALDI-TOF-MS calcd for C69H111O39N17 1825.6, found 1825.4
(M + Na+). The side product: 1H NMR (600 MHz, D2O, DHO
at δ 4.79 as reference, 25 °C) δ 5.04 (d, J ) 9.0 Hz, 1 H), 4.88
(br, 1 H), 4.70 (m, 1 H), 4.66 (d, J ) 8.4 Hz, 1 H), 4.54-4.51
(m, 2 H), 4.48 (t, J ) 5.4 Hz, 1 H), 4.45 (dd, J ) 5.4, 7.2 Hz, 1
H), 4.41 (br d, J ) 4.2 Hz, 1 H), 4.38 (br d, J ) 4.2 Hz, 1 H),
4.31-4.25 (m, 3 H), 4.12 (q, J ) 6.0 Hz, 1 H), 3.98 (dd, J )
4.8, 3.0 Hz, 1 H), 3.92 (br d, J ) 3.6 Hz, 2 H), 3.91-3.68 (m,
28 H), 3.66 (t, J ) 9.0 Hz, 1 H), 3.49 (m, 1 H), 3.00 (dd, J )
6.0, 17.9 Hz, 1 H), 2.90-2.86 (m, 2 H), 2.73 (dd, J ) 8.4, 16.2
Hz, 1 H), 2.41-2.31 (m, 6 H), 2.18 (m, 1 H), 2.11-1.96 (m, 6
H), 2.08 (s, 3 H), 2.07 (s, 3H), 2.02 (s, 3 H), 1.22-1.19 (m, 9
H).
Ack n ow led gm en t. This work was supported in part
by the Research Corporation (Research Innovation
Award RI 0663) and the Petroleum Research Fund,
administered by the American Chemical Society (ACS
PRF grant 37743-G). The authors are also grateful to
Dr. S. Chen and Mr. H. Faulk for the MS measure-
ments.
Resin -Bon d ed Glycop ep tid e (27). Starting from Fmoc-
Ser-loaded 2-chlorotrityl resin (417 mg, 0.25 mmol), a non-
apeptide linked to the resin (661 mg) was prepared on an
automatic peptide synthesizer after eight cycles of peptide
elongation following the standard Fmoc chemistry with HATU-
DIEA as the condensation regent. A part of the resin (54.3
mg) was treated with 20% piperidine/NMP at room tempera-
Su p p or tin g In for m a tion Ava ila ble: The general meth-
ods of the experiments and selected 1H, COSY and HMQC
NMR and MS spectra of the intermediates and the final
product. This material is available free of charge via the
Internet at http://pubs.acs.org.
J O034773S
J . Org. Chem, Vol. 68, No. 23, 2003 9011