Efficient substitution reaction from cysteine to the serine residue of glycosylated polypeptide: Repetitive peptide segment ligation strategy and the synthesis of glycosylated tetracontapeptide having acid labile sialyl-T N antigens

Article abstract of DOI:10.1021/jo8026164

This paper reports the synthesis of a 40-residue glycopeptide having two antigenic sialyl-T_N (NeuAc-α-(2,6)-GalNAcThr) residues in the MUC1 sequence. This target glycopeptide is a tandem repeat form of 20-residue glycopeptides. For the synthesi

Full text of DOI:10.1021/jo8026164

Efficient Substitution Reaction from Cysteine to the Serine Residue of
Glycosylated Polypeptide: Repetitive Peptide Segment Ligation Strategy
and the Synthesis of Glycosylated Tetracontapeptide Having Acid Labile
Sialyl-TN Antigens
Ryo Okamoto, Shingo Souma, and Yasuhiro Kajihara*
International Graduate School of Arts and Sciences, Yokohama City UniVersity, 22-2, Seto, Kanazawa-ku,
Yokohama, 236-0027, Japan
kajihara@yokohama-cu.ac.jp
ReceiVed NoVember 26, 2008
This paper reports the synthesis of a 40-residue glycopeptide having two antigenic sialyl-TN (NeuAc-R-(2,6)-GalNAc-
Thr) residues in the MUC1 sequence. This target glycopeptide is a tandem repeat form of 20-residue glycopeptides.
For the synthesis of this large molecule, native chemical ligation (NCL) at the serine site was used (CysNCLSer). The
concept of CysNCLSer relies on the following: (1) conventional NCL between peptide-R-thioester and the cysteine
residue of another peptide segment; (2) methylation of the thiol that was used for NCL; (3) acidic CNBr conversion
of the cysteine residue to the serine residue forming an O-ester linkage; and (4) an O- to N-acyl shift to couple the
two glycopeptides through a native amide bond. To synthesize glycopeptide having an acid-labile sugar moiety, a
20-residue glycopeptide-R-thioester and 20-residue glycopeptide having a cysteine residue at the N-terminal were
synthesized by solid phase glycopeptide synthesis, and then coupled by CysNCLSer. As the result of extensive
investigation, CNBr activation with an additional acid (trifluoroacetic acid) was found to be essential to obtain good
reactivity and yield, and this condition afforded a tandem repeat form of 40-residue sialylglycopeptide having two
sialyl-TN residues. In addition to this, it was demonstrated that the cysteine thiol protected by the acetoamidomethyl
(Acm) group did not react with the CNBr reagent, and therefore CysNCLSer can be used for repetitive native chemical
ligation in the presence of a protecting N-terminal cysteine residue with an Acm group.
Introduction
copeptide chemical syntheses have been extensively examined.
Glycoproteins have been prepared by cell expression methods,
but not in a homogeneous form due to the heterogeneous
structure of oligosaccharides on the expressed protein.2 Recently,
chemical synthesis has come to be recognized as a new strategy
Glycoproteins are functional on the cell surface and in body
fluids.1 To study glycoprotein function, glycoprotein and gly-
* Corresponding author.
2494 J. Org. Chem. 2009, 74, 2494–2501
10.1021/jo8026164 CCC: $40.75 2009 American Chemical Society
Published on Web 02/23/2009

Synthesis of Glycosylated Tetracontapeptide
FIGURE 1. Conversion of Cys to Ser residue.
for the preparation of homogeneous glycoproteins.3 A chemical
strategy is expected to solve the role of the posttranslational
modification patterns made by the puzzlingly diverse oligosac-
charide structures. In this strategy, preparation of the target
glycosylated polypeptide chain relies on a peptide-segment
coupling, including the glycopeptide segments.3,4 Therefore this
peptide segment coupling is a key reaction step to the production
of a long polypeptide chain, which is subsequently folded into
the desired three-dimensional protein structure. There are several
methods which have been developed for peptide coupling. Of
these, native chemical ligation (NCL)5 has been widely used
for the synthesis of proteins and it has also been applied to the
synthesis of glycoproteins. However, NCL normally requires
the use of a cysteine residue at the junction between two peptide
segments, and this means that proteins without cysteine residues
cannot be synthesized by means of NCL. Therefore, attempts
at the improvement and further development of NCL have been
conducted.6 These can be roughly divided into two categories.
The first strategy uses an auxiliary mediated type of reaction.7
Although this method theoretically can be used for all of the
amino acid residues, it is usually applied for the less sterically
hindered amino acid residues such as alanine and glycine at
the junction points. The most recently reported strategy is a
conversion of cysteine after NCL.8 This type uses the reduction
reaction of a thiol group after conventional NCL and enables
the use of alanine, phenylalanine, and valine as the ligation sites.
This method overcame the site limitation problem in the NCL
reaction.
However, to synthesize glycopeptides and glycoproteins,
further development of this conversion reaction is essential.
Ideally, NCL would be applicable to any positions in a
polypeptide chain. In addition to these reactions, we also
demonstrated an efficient glycopeptide ligation at serine
(
CysNCLSer) based on conversion strategy after conventional
NCL.9 This strategy includes three reaction steps, as shown in
Figure 1: (a) cysteine methylation,10 (b) CNBr conversion
reaction,11 and (c) O- to N-acyl shift.12
Of these reactions, CNBr conversion is one of the critical
reaction steps. CNBr has conventionally been used for the
C-terminal peptide cleavage of methionine or chemically
methylated cysteine residues under an acidic condition such as
with formic acid.11 However, this reaction has also been shown
to cause formylation toward alcohols of serines or threonines
in peptides.11,13 In terms of glycopeptide synthesis, this reaction
condition has come to be considered unsuitable, because long-
term reactions under acidic conditions may lead to the cleavage
of labile glycosylated linkages (e.g., sialyl-linkage and fucosyl
linkage) as well as the undesired formylation of sugar alcohols.
To synthesize glycoproteins having acid labile sialylglycopep-
tides or glycoproteins by means of CysNCLSer, we have exten-
sively studied CysNCLSer through a synthesis of the MUC1
peptide with sialyl-TN residues. This class of sialylglycopeptide
has potentially important value as a cancer vaccine, but the
synthesis has been thought to be an extremely difficult task.3,14
Here we report a practical synthesis of MUC1 tandem repeat
40-residue sialylglycopeptide having two sialyl-TN residues by
means of CysNCLSer. In addition, we also report that this
CysNCLSer method can be used for powerfully repetitive NCL
reactions, indicating that CysNCLSer can be performed in the
presence of methionine9 and other cysteine residues in the
peptide chain.
(1) Dwek, R. A. Chem. ReV. 1996, 96, 683.
(2) Walsh, G.; Jefferis, R. Nat. Biotechnol. 2006, 24, 1241.
(3) (a) Nakahara, Y. Biopolymers 2007, 88, 308. (b) Liu, L.; Bennett, C. S.;
Wong, C.-H. Chem. Commun. 2006, 21. (c) Gambin, D. P.; Scanlan, E. M.;
Davis, B. G. Chem. ReV. 2009, 109, 131.
(4) Nilsson, B. L.; Soellner, M. B.; Raines, R. T. Annu. ReV. Biophys. Biomol.
Struct. 2005, 34, 91.
(5) Dawson, P. E.; Muir, T. W.; Clark-Lewis, I.; Kent, S. B. H. Science
1994, 266, 776.
(6) (a) Macmillan, D. Angew. Chem. Int. Ed 2006, 45, 7668. (b) Haase, C.;
Seitz, O. Angew. Chem., Int. Ed 2008, 47, 1553. (c) Hackenberger, C. P. R.;
Schwarzer, D. Angew. Chem., Int. Ed 2008, 47, 10030.
(7) (a) Canne, L. E.; Bark, S. J.; Kent, S. B. H. J. Am. Chem. Soc. 1996,
118, 5891. (b) Low, D. W.; Hill, M. G.; Carrasco, M. R.; Kent, S. B. H.; Botti,
P. Proc. Natl. Acad. Sci. U.S.A. 2001, 98, 6554. (c) Kawakami, T.; Akaji, K.;
Aimoto, S. Org. Lett. 2001, 3, 1403. (d) Offer, J.; Boddy, C. N. C.; Dawson,
P. E. J. Am. Chem. Soc. 2002, 124, 4642. (e) Macmillan, D.; Anderson, D. W.
Org. Lett. 2004, 6, 4659. (f) Chen, G. D.; Warren, J. D.; Chen, J.; Wu, B.; Wan,
Q.; Danishefsky, S. J. J. Am. Chem. Soc. 2006, 128, 7460. (g) Brik, A.; Ficht,
S.; Yang, Y. Y.; Bennett, C. S.; Wong, C. H. J. Am. Chem. Soc. 2006, 128,
15026.
(8) (a) Yan, L.; Dawson, P. E. J. Am. Chem. Soc. 2001, 123, 526. (b)
Pentelute, B. L.; Kent, S. B. H. Org. Lett. 2007, 9, 687. (c) Crich, D.; Benerjee,
A. J. Am. Chem. Soc. 2007, 129, 10064. (d) Botti, P.; Tchertchian, S. WO/
2006/133962. (e) Wan, Q.; Danishefsky, S. J. Angew. Chem., Int. Ed. 2007, 46,
9248. (f) Haase, C.; Rohde, H.; Seitz, O. Angew. Chem., Int. Ed. 2008, 47, 6807–
6810. (g) Chen, J.; Wan, Q.; Tuan, Y.; Zhu, J.; Danishefsky, S. J. Angew. Chem.,
Int. Ed. 2008, 47, 8521–8524.
(9) Okamoto, R.; Kajihara, Y. Angew. Chem., Int. Ed. 2008, 46, 9248.
(10) Heinrikson, R. L. J. Biol. Chem. 1971, 246, 4090.
(11) (a) Kaiser, R.; Metzka, L. Anal. Biochem. 1999, 266, 1. (b) Gross, E.;
Morell, J. L. Biochem. Biophys. Res. Commun. 1974, 59, 1145.
(12) (a) Vila-Perello, M.; Hori, Y.; Ribo´, M.; Muir, T. W. Angew. Chem.,
Int. Ed. 2008, 47, 7764. (b) Souma, Y.; Sasaki, M.; Hayashi, Y.; Kimura, T.;
Kiso, Y. Chem. Commun. 2004, 124.
(13) Richardson, J. P.; Macmillan, D. Org. Biomol. Chem. 2008, 6, 3977.
(14) (a) Buskus, T.; Ingale, S.; Boons, G.-J. Glycobiology 2006, 16, 113R–
136R. (b) Seitz, O. ChemBioChem 2000, 1, 214.
J. Org. Chem. Vol. 74, No. 6, 2009 2495

Okamoto et al.
SCHEME 1. CNBr Conversion Reaction with 11-Residue Peptide 1 (LFRVYC(Me)NFLRG)
TABLE 1. Formic Acid Concentration Dependence of the CNBr
Conversion Reactiona
TABLE 2. Varying the Acidic Condition for the CNBr Conversion
Reactiona
yield (%)
yield (%)
entry
acid condition
100% HCOOH
70% HCOOH/30% H2O
50% HCOOH/50% H2O
20% HCOOH/80% H2O
1
2
3
entry
acid condition
80% HCOOH/20% H2O
1
2
3
1
2
3
4
86
60
48
28
13
37
21
5
1
1
2
3
4
5
6
7
8
9
25 58 17
30 33 35
9
6
8
11
11 63 25
10 63 26
44 34
3
31
67
0.1 M HCl
60% H2O/40% CH3CN containing MsOH (1% v/v)
60% H2O/40% CH3CN containing TsOH (3% v/v)
60% H2O/40% CH3CN containing H3PO4 (3% v/v)
60% H2O/40% CH3CN containing TBA (3% v/v)
60% H2O/40% CH3CN containing TFA (3% v/v)
60% H2O/40% CH3CN containing TFA (2% v/v)
50% TFA/50% H2O
66 24
68 22
5
2
83
61
a Reaction condition: peptide 1 (1equiv, concentration was adjusted to
1 mM), CNBr (100 equiv). Yields were estimated by HPLC area
intensity.
7
Results and Discussion
a Reaction condition: peptide 1 (1 equiv, concentration was adjusted
to 2 mM), CNBr (200 equiv). Yields were estimated by HPLC area
intensity.
Protein cleavage at the methionine or methylated cysteine
with CNBr has been examined with formic acid or HCl.11 In
the case of methionine, the cleavage reaction takes place at the
C-terminal amide bond of methionine, while the N-terminal
amide bond of methylated cysteine is cleaved by the CNBr
reaction. Although this CNBr reaction under acidic conditions
toward methionine has been well investigated, there are few
reports which discuss the reaction conditions for the cleavage
reaction at the methylated cysteine site. Therefore, we first
investigated the relationship of the formic acid concentration
and the reactivity of the CNBr conversion reaction toward the
peptide having methylated cysteine residues. The model 11-
residue peptide (LFRVYC(Me)NFLRG) was prepared by the
reported method (Scheme 1).9 We performed the CNBr conver-
sion reaction at several HCOOH concentrations (Table 1).
As shown in Figure 1, this reaction requires H2O, therefore
the reaction shown in entry 1 (100% HCOOH) did not proceed
well. In the case of entries 2-4, decrease of the formic acid
concentration led to a decrease in the product amount and an
increase of the sulfoxide formed by oxidation of the starting
material. In addition to the formic acid condition, we performed
the CNBr conversion reaction by using several different kinds
of acid (Table 2). Entries 1 and 2 are the conventional reaction
conditions in the case of peptide cleavage at methionine sites.11
In contrast to these conditions, applying a low concentration of
an acid such as methanesulfonic acid (MsOH), p-toluenesulfonic
acid (TsOH), or trifluoroacetic acid (TFA) exhibited good yield,
as shown in entries 3, 4, 7, and 8. On the other hand, phosphoric
acid (entry 5) or tribromoacetic acid (TBA, entry 6) generated
the largest amount of sulfoxide. As shown in Table 2, we found
that the conditions of entry 7 or 8 are suitable for CNBr
activation. In addition, because TFA can be removed from
reaction mixture by evaporation, we used this condition for
glycopeptides synthesis.
We have also performed the conversion reaction by another
activation system, as shown in Scheme 2. In the case of the
CNBr condition, it generated the sulfonium cation through the
activating SMe group of the CN cation, as shown in Figure 1.
We expected this generation of the sulfonium cation could be
obtained directly from the sulfuhydryl group. Therefore, we used
the commercially available reagent diethyl meso-2,5-dibromoa-
dipate. This type of reagent afforded a five-membered-ring
intermediate, as shown in Scheme 2.15 However, as a result,
this reaction afforded only a ꢀ-elimination product and did not
afford any desired O-ester product (Scheme 2, data shown in
the Supporting Information).16 We speculate that this is because
of the steric hindrance of the five-membered ring. This ring
cannot apparently be placed in a position where the amide
carbonyl oxygen could attack the ꢀ carbon of cysteine. On the
other hand, in the case of the CNBr activation, the intermediate
of the sulfonium cation may be placed in a suitable position
for oxazolone formation rather than elimination, because the
sulfonium cation formed by CNBr appears to be a sterically
(15) Holmes, T. J.; Lawton, R. G. J. Am. Chem. Soc. 1977, 99, 1986.
(16) Several ꢀ elimination reactions of cysteine are reported: Bernardes,
G. J. L.; Chalker, J. M.; Errey, J. C.; Davis, B. G. J. Am. Chem. Soc. 2008, 130,
5052.
2496 J. Org. Chem. Vol. 74, No. 6, 2009

Synthesis of Glycosylated Tetracontapeptide
SCHEME 2. Activation of Sulfuhydryl Group Directly through a Five-Membered-Ring-Type Intermediatea
a Reagents and conditions: (a) peptide 4, diethyl meso-2,5-dibromoadipate, CH3CN, 0.2 M phosphate buffer containing 6 M guanidine (pH 6.3).
SCHEME 3. Cys to Ser Conversion Reaction with 8-Residue Peptide 6a
a Reagents and conditions: (a) methyl 4-nitrobenzenesulfonate, CH3CN, 0.25 M tris buffer (pH 8.6) containing 6 M guanidine and 3.3 mM EDTA2Na;
(b) CNBr, 80% HCOOH solution or CNBr, 40% CH3CN solution, TFA (2% v/v); (c) 5% hydrazine solution, then TFA, NH4I, SMe2; (d) 50% CH3CN
solution containing 0.1% TFA (v/v), AgOAc.
smaller reaction species than that activated by meso-2,5-
dibromoadipate. Therefore, we concluded that CNBr would be
a suitable reagent for the conversion reaction.
react with CNBr. To determine whether the TFA/CNBr condi-
tion would avoid both esterification of the hydroxyl groups and
activation of the protected cysteine at the N-terminal, we applied
this condition for the model 8-residue peptide 6 having three
hydroxyl groups, a guanidine group, an N-terminal amino group,
a free cysteine group, and a Acm-protected cysteine group at
the N-terminal (Scheme 3). If the conversion reaction of Cys
at the 6 position to a serine residue acts selectively, the
N-terminal cysteine can be used in the second NCL.
The 8-residue peptide 7 having a methyl cysteine residue
(C(Acm)YRNTC(Me)SL) was synthesized by the selective
methylation of 6, which was synthesized by SPPS. We per-
formed the CNBr conversion reaction using both the conven-
tional and new TFA condition. As shown in Figure 2a, a
conventional reaction condition with HCOOH generated mul-
tiple products. Mass analysis of these products suggested that
byproducts seemed to be the result of formylation toward
alcohols. In contrast, in the case of the new reaction condition
(Figure 2b), there was no side reaction except for slight oxidation
of the starting material. A one-pot reaction of the O- to N-acyl
We also expected that a new acidic condition with use of
TFA for the CNBr conversion reaction would not cause any
acylation of the hydroxyl groups. In terms of formylation during
the CNBr activation, we speculated that a simple formylation
would be caused under the acidic condition (HCOOH). In the
case of the TFA/CNBr condition, we suppose that the same
acidic trifluoroacetylation may be caused. However, it is known
that a trifluoroacetyl group is easily hydrolyzed in a solution,
meaning that the TFA/CNBr condition may avoid unexpected
esterification of the hydroxyl groups. In addition to overcoming
the formylation problem, we have had an interest in repetitive
NCL by the use of CysNCLSer. In this case, a cysteine at the
N-terminal should not react with CNBr during the CNBr
conversion reaction. In our experiments, since the free thiol
group of cysteine potentially reacts with CNBr, we introduced
the N-terminal cysteine in an acetoamidomethyl (Acm)-protected
form. We expected that the Acm-protected cysteine would not
J. Org. Chem. Vol. 74, No. 6, 2009 2497

Okamoto et al.
FIGURE 2. CNBr conversion reaction with 8-residue peptide 7. The HPLC chart of 80% HCOOH condition reaction shows a multiple peak
pattern at 31 h ((i) peptide 8, (ii) monoformylated peptide of 8, (iii) diformylated peptide of 8, (iv) peptide 9, (v) monoformylated peptide of 9, and
(vi) diformylated peptide of 9). Each inset of purified product was observed from ESI-MS spectra (calcd for [M + H]+ 1014.5, [M + 2H]2+ 507.6).
The asterisk indicates the diastereomer of sulfoxides.
shift afforded the desired peptide 9. To recover the starting
material, a reduction process of sulfoxide was included in this
one-pot O- to N-acyl shift process. This result proved that the
reaction yield under the new condition was better than that under
the conventional condition (54% yield in HCOOH condition
(Figure 2a) and 75% yield in the 2% TFA condition (Figure
2b); these yields were estimated by HPLC area intensities), and
the recovered starting material could be used again to accumulate
target peptide 9. In addition, as expected, the N-terminal
protected cysteine did not react with CNBr. Deprotection of
the Acm group by the conventional condition17 afforded the
peptide having a cysteine residue at the N-terminal 10 (data
shown in the Supporting Information). This means that CysNCLSer
can be applied to repetitive NCL.
Since the suitable condition had been determined, we then
examined the synthesis of glycopeptides having acid labile
sialyloligosaccharides. We selected 40-residue glycopeptide in
a MUC1 variable number tandem repeat (VNTR) region having
two sialyloligosaccharides.18 This region consists of a 20-residue
glycopeptide unit, in which the polypeptide chain possesses
abundant serines, threonine, and prolines, in tandem. It is known
that the sialyloligosaccharides are expressed in the MUC1
VNTR region on endothelial cancer cells.18,19 Such special
glycopeptides having multiple sialyloligosaccharides are ex-
pected to have potential as a cancer vaccine. However, the
repeating sialylglycopeptides of the VNTR region had not been
synthesized. Therefore we targeted the synthesis of repeated
MUC1 glycopeptides having two sialyl-TN antigens.
To synthesize such a sialylglycopeptide, we prepared 20-
residue sialylglycopeptide-R-thioester 11 and 20-residue sia-
lylglycopeptide 12. The sialyl-TN methyl ester threonine unit
for sialylglycopeptide segments was synthesized by the reported
method (Scheme 4, data shown in the Supporting Information).20
To couple these glycopeptides, NCL was performed under
the reported condition,20 and then selective methylation of the
cysteine thiol10 afforded the desired 40-residue glycopeptide
having the two sialyl-TN 14 in good yield (81% and 90% yield,
respectively). Following this step, the CNBr conversion reaction
was performed in acetonitrile-H2O-TFA solution (Table 2,
entry 8 condition). As expected, there was no acylation in the
reaction. Hence, slight oxidation of the starting material was
observed. To recover the starting material, reduction of sulfoxide
was performed by the use of TFA/NH4I/SMe2 at -10 °C, and
this afforded O-ester linked 40-residue sialylglycopeptide 15
in 38% yield. In addition, the recovered substrate was found to
be repeatedly useful for this CNBr conversion reaction. Sub-
sequently, the O- to N-acyl shift reaction and deprotection of
two methyl ester groups of 15 under a basic condition was
undertaken. Although this treatment afforded a small amount
of cleaved peptide (ca. 30%, data shown in the Supporting
Information), a subsequent deprotection step of the methyl ester
afforded sialylglycopeptide in moderate yield (ca. 50% isolated
(17) Durek, T.; Torbeev, V. Y.; Kent, S. B. H. Proc. Natl. Acad. Sci. U.S.A.
2007, 104, 4846.
(18) Singh, R.; Bandyopadhyay, D. Cancer Biol. Ther. 2008, 6, 481.
(19) (a) Becker, T.; Dziadek, S.; Kunz, H. Curr. Cancer Drug Targets 2006,
6, 491. (b) Dube, D. H.; Bertozzi, C. R. Nat. ReV. Drug DicoVery 2005, 4, 477.
(c) Danishefsky, S. J.; Allen, J. R. Angew. Chem., Int. Ed 2000, 39, 836. (d)
Hang, C. H.; Bertozzi, C. R. Bioorg. Med. Chem. 2005, 13, 5021.
(20) Okamoto, R.; Souma, S.; Kajihara, Y. J. Org. Chem. 2008, 73, 3460.
(21) Johnson, E. C. B.; Kent, S. B. H. J. Am. Chem. Soc. 2006, 128, 6640.
2498 J. Org. Chem. Vol. 74, No. 6, 2009

Synthesis of Glycosylated Tetracontapeptide
a
SCHEME 4. Synthesis of MUC1 Repeated Glycopeptide Having Two Sialyl-TN
a Reagents and conditions: (a) 0.2 M phosphate buffer (pH 7.2) containing 6 M guanidine, 20 mM tris(2-carboxylethyl)phosphine and 4-mercaptophenylacetic
acid (1% v/v); (b) methyl 4-nitrobenzenesulfonate, CH3CN, 0.25 M tris buffer (pH 8.6) containing 6 M guanidine and 3.3 mM EDTA2Na; (c) CNBr, 40%
CH3CN solution, TFA (2% v/v), then TFA, NH4I, SMe2;. (d) 5 mM NaOH, then 50 mM NaOH.
yield, 70% conversion yield estimated by HPLC area intensity:
see the Supporting Information). In these reaction steps, we did
not observe the cleavage of the NeuAc residue in sialyl-TN
moiety under acidic CNBr reaction or ꢀ-elimination of the sialyl-
TN moiety under the basic condition. The 1H NMR of the MUC1
repeating glycopeptide having two sialyl-TN 16 is shown in
Figure 3. As shown in this NMR spectrum, CysNCLSer can afford
glycopeptides in moderate yield.
J. Org. Chem. Vol. 74, No. 6, 2009 2499

Okamoto et al.
FIGURE 3. 1H NMR of synthesized 40-residue sialylglycopeptide having two sialyl-TN 16.
To the solution was then added NH4I (11 mg) and dimethyl sulfide
(5.6 µL). The mixture was stirred for 50 min at room temperature
and ice cooled H2O (2.0 mL) was added. This mixture was washed
with CCl4. The water phase was concentrated in vacuo. Purification
of the residue by reverse-phase HPLC (condition: C-18 column, 5
µm, 250 × 4.5 mm, Isocratic of 0.1% TFA solution for 5 min,
followed by the linear gradient of 0f54% CH3CN containing 0.1%
TFA in 0.1% TFA solution over 30 min at a flow rate 1.0 mL/
min) afforded peptide 9. ESIMS: calcd for C41H68N13O15S [M +
H]+ 1014.5; found [M + H]+ 1014.8.
In conclusion, we report the synthesis of a 40-residue MUC1
sialylglycopeptide having two sialyl-TN. To synthesize such a
large sialylglycopeptide, we used CysNCLSer. We have demon-
strated that this condition has afforded acid labile sialylglyco-
peptides and avoided activation of the protected cysteine at the
N-terminal of the peptide. This means that CysNCLSer can be
used not only for the synthesis of acid labile sialylglycopeptides,
but also for repetitive NCL. Investigation of the synthesis of
several glycoproteins by means of both conventional NCL and
CysNCLSer is currently in progress.
Synthesis of 13 (Native Chemical Ligation of Sialylglycopep-
tide Thioester 11 and Sialylglycopeptide 12). Sialylglycopeptide
R-thioester 11 (5.0 mg, 2.0 µmol) and sialyl glycopeptide 12 (4.8
mg, 2.0 µmol) were dissolved in 0.2 M sodium phosphate buffer
(pH 7.2, 1.0 mL) containing 6 M guanidine, 20 mM tris(2-
carboxylethyl)phosphine, and 1.0% v/v 4-mercaptophenylacetic
acid. This mixture was stirred for 12 h. Purification of the mixture
by RP-HPLC (condition: C-8 column, 5 µm, 250 × 10 mm linear
gradient of 9f36% CH3CN containing 0.1% TFA in 0.1% TFA
aqueous over 30 min at a flow rate of 4.0 mL/min) afforded MUC1
glycopeptide having two sialyl-TN 13 (7.6 mg, 81%). ESIMS: calcd
for C200H317N54O80S [M + 3H]3+ ) 1597.3, [M + 4H]4+ 1198.3,
[M + 5H]5+ 958.8; found [M + 3H]3+ 1597.3, [M + 4H]4+ 1198.1,
[M + 5H]5+ 958.7.
Synthesis of 14 (S-Methylation of MUC1 Repeated Glycopep-
tide 13). Compound 13 (6 mg, 1.25 µmol) was dissolved in 0.25
M tris buffer (pH 8.6, 1.25 mL) containing 6 M guanidine and 3.3
mM EDTA-2Na. To the mixture was added a solution of methyl
4-nitrobenzenesulfonate (8.1 mg, 67.5 µmol) in acetonitrile (416
µL) and the mixture was stirred for 40 min. Then the mixture was
neutralized by 10% TFA solution (0.10 mL) and the solution was
concentrated in vacuo to remove organic solvent. Purification of
the residue to remove excess salt by RP-HPLC (condition: C18
column, 5 µm, 250 × 4.5 mm Isocratic of 0.1% TFA aqueous for
10 min, then linear gradient of 9f36% CH3CN containing 0.1%
TFA in 0.1% TFA aqueous over 30 min at a flow rate of 1.0 mL/
min) afforded product 14 (5.5 mg, 90% isolated yield). In this step,
this material was used for the next step without further purification.
ESIMS: calcd for C201H319N54O80S [M + 3H]3+ 1602.0, [M + 4H]4+
Experimental Section
Synthesis of 7 (S-Methylation of Peptide 6). Compound 6 (3.3
mg, 3.21 µmol) was dissolved in 0.25 M tris buffer (pH 8.6, 3.2
mL) containing 6 M guanidine and 3.3 mM EDTA-2Na. To this
mixture was added 2-mercaptoethanol (2.3 µL), then the solution
was stirred for 15 min. Next a solution of methyl 4-nitrobenzene-
sulfonate (21.0 mg, 66.3 µmol) in acetonitrile (1.0 mL) was added
to the mixture, which was then stirred for 40 min. The mixture
was neutralized by 10% TFA solution (0.4 mL) and washed with
Et2O (3 mL, 3 times). The solution was then concentrated in vacuo
to remove organic solvent. Purification of the mixture by RP-HPLC
(HPLC condition: C-18 column, 5 µm, 250 × 4.5 mm, linear
gradient of 13.5f31.5% CH3CN containing 0.1% TFA in 0.1%
TFA aqueous over 60 min at a flow rate of 1.0 mL/min) afforded
the desired product 7 (2.7 mg, 83% isolated yield). ESIMS: calcd
for C42H70N13O14S2 [M + H]+ 1044.5; found [M + H]+ 1044.9.
Typical Procedure of the Synthesis of 9 (CNBr Conversion
Reaction of 7). Peptide 7 (4 mg, 3.83 µmol) was dissolved in acid
solution (80% formic acid or 40% CH3CN solution containing 2%
v/v TFA, 1.9 mL). To the solution was added CNBr (81 mg) and
the mixture was stirred for 31 h in the dark at room temperature.
Then the mixture was lyophilized. Subsequently, the residue was
treated as in the following condition to reduce oxidized starting
material and to perform O- to N-acyl shift reaction under slight
basic condition (in the case of formic acid condition, this basic
condition removed the formyl group). The residue was dissolved
in 5% hydrazine hydrate solution (380 µL) and this solution was
stirred for 25 min. Then, to this solution was added TFA (1.5 mL).
2500 J. Org. Chem. Vol. 74, No. 6, 2009

Synthesis of Glycosylated Tetracontapeptide
1201.8, [M + 5H]5+ 961.6; found [M + 3H]3+ 1601.5, [M + 4H]4+
1201.6, [M + 5H]5+ 961.6.
added 50 mM NaOH solution (1.3 mL) and the reaction mixture
was allowed to gradually warm to room temperature and stirred
for 28 h. Then, the solution was neutralized by AcOH (6 µL).
Purification of the mixture by RP-HPLC (condition: C18 column,
5 µm, 250 × 4.5 mm, linear gradient of 9f27% CH3CN in 50
mM NH4OAc solution over 60 min at a flow rate of 1.0 mL/min)
afforded MUC1 repeated segment 16 (ca. 1.0 mg, ca. 50% isolated
yield, 70% yield estimated by HPLC peak area intensity). ESIMS:
calcd for C198H313N54O81 [M + 3H]3+ 1582.6, [M + 4H]4+ 1187.2,
[M + 5H]5+ 950.0; found [M + 3H]3+ 1582.8, [M + 4H]4+ 1187.5,
[M + 5H]5+ 950.2.
Synthesis of 15 (CNBr Conversion Reaction of Sialylglycopep-
tide 14). Glycopeptide 14 (5.8 mg) was dissolved in 40% CH3CN
solution (600 µL) containing 2% v/v TFA (12 µL). To this solution
was added CNBr (64 mg) and the mixtuer was stirred for 30.5 h in
the dark at room temperature. Then H2O (1.0 mL) was added and
the mixture was lyophilized. Subsequently, the residue was treated
as in the following condition to reduce oxidized starting material.
The residue was dissolved in cooled (at -10 °C) TFA (2.9 mL).
To this solution was added NH4I (7.8 mg) and dimethyl sulfide
(2.6 µL) at -10 °C. The mixture was stirred for 30 min at room
temperature and then ice cooled H2O (1.2 mL) was added. This
mixture was washed by ice cooled CCl4. The water phase was
concentrated in vacuo at 10 °C. Purification of the residue by
reverse-phase HPLC (condition: C18 column, linear gradient of
4.5f36% CH3CN containing 0.1% TFA in 0.1% TFA solution over
30 min at flow rate 1.2 mL/min) afforded MUC1 repeated
sialylglycopeptide 15 (2.2 mg, 38% isolated yield). ESIMS: calcd
for C200H317N54O81 [M + 3H]3+ 1592.0, [M + 4H]4+ 1194.2, [M
+ 5H]5+ 955.6; found [M + 3H]3+ 1592.0, [M + 4H]4+ 1194.4,
[M + 5H]5+ 955.8.
Acknowledgment. We would like to thank Dr. Yukishige
Ito (RIKEN, Saitama) and Otsuka Chemical Co. for their support
and encouragement. Financial support from the Japan Society
for the promotion of Science (Grant-in-Aid for Creative
Scientific Research No. 17GS0420) is acknowledged.
Supporting Information Available: Full experimental details
for the CNBr conversion of 1 and the synthesis of 5, 9, 10, and
16, and NMR spectra of all new compounds and coupling
compounds. This material is available free of charge via the
Internet at http://pubs.acs.org.
Synthesis of 16 (O to N Intramolecular Acyl Shift and Remov-
ing Methylester of Sialyl-TN Moiety). Sialylglycopeptide 15 (2.0
mg, 0.42 µmol) was dissolved in 5 mM NaOH solution (1.67 mL)
at 0 °C and the mixture was stirred for 25 min. To the mixture was
JO8026164
J. Org. Chem. Vol. 74, No. 6, 2009 2501