Solid-phase synthesis of peptide and glycopeptide thioesters through side-chain-anchoring strategies

Article abstract of DOI:10.1002/chem.200701978

An efficient new strategy for the synthesis of peptide and glycopeptide thioesters is described. The method relies on the side-chain immobilization of a variety of Fmoc-amino acids, protected at their C-termini, on solid supports. Once anchored, peptides were constructed using solid-phase peptide synthesis according to the Fmoc protocol. After unmasking the C-terminal carboxylate, either thiols or amino acid thioesters were coupled to afford, after cleavage, peptide and glycopeptide thioesters in high yields. Using this method a significant proportion of the proteinogenic amino acids could be incorporated as C-terminal amino acid residues, therefore providing access to a large number of potential targets that can serve as acyl donors in subsequent ligation reactions. The utility of this methodology was exemplified in the synthesis of a 28 amino acid glycopeptide thioester, which was further elaborated to an N-terminal fragment of the glycoprotein erythropoietin (EPO) by native chemical ligation.

Full text of DOI:10.1002/chem.200701978

DOI: 10.1002/chem.200701978
Solid-Phase Synthesis of Peptide and Glycopeptide Thioesters through
Side-Chain-Anchoring Strategies
[a]
Simon Ficht,[a] Richard J. Payne,[a] Richard T. Guy, and Chi-Huey Wong*[a, b]
Abstract: An efficient new strategy for
the synthesis of peptide and glycopep-
tide thioesters is described. The
method relies on the side-chain immo-
bilization of a variety of Fmoc-amino
acids, protected at their C-termini, on
solid supports. Once anchored, pep-
tides were constructed using solid-
phase peptide synthesis according to
the Fmoc protocol. After unmasking
the C-terminal carboxylate, either
thiols or amino acid thioesters were
coupled to afford, after cleavage, pep-
tide and glycopeptide thioesters in high
yields. Using this method a significant
proportion of the proteinogenic amino
acids could be incorporated as C-termi-
nal amino acid residues, therefore pro-
viding access to a large number of po-
tential targets that can serve as acyl
donors in subsequent ligation reactions.
The utility of this methodology was ex-
emplified in the synthesis of a 28
amino acid glycopeptide thioester,
which was further elaborated to an N-
terminal fragment of the glycoprotein
erythropoietin (EPO) by native chemi-
cal ligation.
Keywords: glycopeptides
proteins · ligation · peptide thio-
esters · solid-phase synthesis
· glyco-
AHCTREUNG
Introduction
from a drug discovery perspective since many of these bio-
molecules represent potential therapeutic leads.[6,7] General-
ly, the isolation of single glycoforms from natural sources is
not possible, and the current belief is that the high demand
for homogeneous glycopeptides and glycoproteins can only
be met by chemical intervention.[8–11] Of the methods avail-
able for the synthesis of glycoproteins, chemical ligation is
amongst the most promising. Current technologies include
native chemical ligation (NCL), sugar-assisted ligation
(SAL) and the traceless Staudinger ligation.[12–20] All of
these techniques rely on peptide or glycopeptide thioesters
as acyl donors, and therefore expedient access to these enti-
ties is necessary if the ligation-based assembly of glycopepti-
des and glycoproteins is to be practical. Unglycosylated pep-
tide thioesters are traditionally synthesized on solid support
using the tert-butyloxycarbonyl (Boc)-strategy.[21,22] The
major drawbacks of this method include the requirement for
TFA during repetitive Boc-deprotection steps and the use of
highly acidic conditions for example, triflic acid or HF for
release of the fully assembled peptide from the resin. Un-
fortunately, glycopeptide thioesters cannot be synthesized in
this way due to the inherent instability of glycosidic linkages
to strongly acidic conditions.[23] These difficulties have led to
the search for efficient and mild Fmoc-based strategies for
their construction.[24] Hence, many research groups chose to
synthesize protected glycopeptide thioesters containing a
free C-terminus which can be derivatized to the thioester in
solution.[25–27] Variants of this solution phase strategy include
More than 50% of all human proteins are glycosylated, a
modification known to influence protein folding, solubility
and biological half-life.[1,2] Glycosylation is implicated in a
variety of biological recognition events such as cell adhe-
sion, cell differentiation, and cell growth.[3–5] Aberrant glyco-
sylation of proteins is known to perturb intracellular recog-
nition, and is associated with several serious illnesses includ-
ing autoimmune diseases, infectious diseases and cancer.[5]
Study of these effects requires pure samples of the relevant
glycoproteins, but in nature they are typically expressed as
heterogeneous mixtures of glycoforms in which identical
protein backbones bear a variety of different glycans. This
has greatly impeded the field of glycobiology due to the dif-
ficulty of elucidating the exact role of a specific glycan.
Access to homogeneous glycoproteins is also important
[a]Dr. S. Ficht, Dr. R. J. Payne,# Dr. R. T. Guy, Prof. Dr. C.-H. Wong
#
Department of Chemistry, The Scripps Research Institute
10550 N. Torrey Pines Road, La Jolla, CA 92037 (USA)
Fax : (+1)858-748-2409
E-mail: wong@scripps.edu
[b]Prof. Dr. C.-H. Wong
The Genomics Research Center, Academia Sinica 128
section 2, Academia Road, Nankang, Taipei (Taiwan)
[#]S. Ficht and R. J. Payne contributed equally.
Supporting information for this article is available on the WWW
under http://www.chemeurj.org/ or from the author.
3620
ꢀ 2008 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2008, 14, 3620 – 3629

FULL PAPER
the synthesis of masked glycopeptide esters which can rear-
range via O!S acyl shifts to generate the desired glycopep-
tide thioester.[28] However, the necessity to perform reac-
tions in solution, in conjunction with the poor solubility
characteristics of protected peptides, has hampered applica-
tions of such a strategy to longer sequences. Alternative ap-
proaches which may be amenable to the synthesis of glyco-
peptide thioesters include Fmoc removal using non-nucleo-
philic base cocktails, Lewis acid-catalyzed release, racemiza-
tion-free photochemical approaches, and oxidation-based
safety catch strategies.[29–36] The most common and widely
used approach for obtaining glycopeptide thioesters involves
synthesis on sulfamylbutyryl resin using an activation and
thiol release strategy.[37–40] Based on this technique, a variant
of the O-linked glycoprotein diptericin e and a fragment of
the N-linked glycoprotein RNase B have been successfully
synthesized.[41,42] However, over-acylation and epimerization
during resin loading are major shortcomings of this method,
and the use of capping steps throughout the synthesis has
been reported to reduce yields considerably.[42] In our hands,
alkylation of the sulfonamide prior to thiol release also re-
sulted in derivatization of the peptide backbone. The recent
development of a self-purification feature of this method
has allowed for the production of peptide thioesters in
higher purity and with fewer truncated by-products, howev-
er, this variant eliminates the potential for sequential liga-
tions since the N-terminus is tied up in the self-purification
process.[43] Alternate strategies for the synthesis of peptide
thioesters rely on attachment of the growing peptide to an
appropriate handle through a backbone amide.[44,45] In one
of these, the backbone amide linker (BAL) strategy, a pro-
tected C-terminal carboxyl group can be elongated with an
amino acid thioester after the peptide has been fully assem-
bled. However, currently only a limited set of amino acids
have been shown to be amenable to backbone anchoring,
and to the best of our knowledge this method has yet to be
implemented in the synthesis of glycopeptide thioesters.
An efficient method for the Fmoc-based synthesis of pep-
tide thioesters involves immobilization of the C-terminal
amino acid residue through its side chain.[25,46–48] In this
manner, Fmoc-Glu-OAllyl, Fmoc-Asp-OAllyl and Fmoc-
Lys-OAllyl have been immobilized onto Wang- or Rink
amide resin.[49] After the peptide is assembled on the resin,
removal of the C-terminal allyl group, on-resin thioester for-
mation and subsequent cleavage yields the corresponding
glutamine, asparagine and lysine thioesters.[50] This method
has recently been applied to the synthesis of a 36-amino
acid glycopeptide thioester by our laboratory, which was
subsequently ligated to a glycopeptide fragment using SAL
to afford the antibacterial glycoprotein diptericin e.[17] To
date this approach has been restricted to the construction of
peptide thioesters bearing a limited set of amino acids on
the C-terminus. Herein we demonstrate that a much larger
number of amino acids can serve as an anchoring point for
the solid-phase peptide synthesis (SPPS) of peptide and gly-
copeptide thioesters on a range of resins. The scope of the
method is significantly broadened by elongation of the im-
mobilized peptides with a variety of amino acid thioesters in
the N- to C-terminal direction (Scheme 1). As a demonstra-
tion of the applicability of the side chain anchoring strategy
to the construction of more complex targets, a 28-amino
acid glycopeptide thioester was successfully synthesized and
ligated to form a fragment of the therapeutic glycoprotein
human erythropoietin (EPO).
Scheme 1. General strategy for the solid-phase synthesis of peptide and
glycopeptide thioesters using a side chain anchoring strategy (AA1 =an-
chorable amino acid; AA2 =any proteinogenic amino acid).
Results and Discussion
Our initial goal was to synthesize peptide thioesters by im-
mobilizing the side chains of aspartic acid and glutamic acid
to a solid support. To this end, commercially available
Fmoc-Asp-OAllyl and Fmoc-Glu-OAllyl were loaded onto
Rink amide resin using PyBOP and N-methylmorpholine in
DMF for two hours.[49] Pre-loadings proved to be high yield-
ing in all cases as determined by deprotection of the Fmoc
group with 10% piperidine/DMF and measurement of the
Fmoc-piperidine adduct at 302 nm. Peptide elongation ac-
cording to the established Fmoc SPPS protocol, followed by
the selective removal of the allyl ester using [Pd(PPh3)4]and
G
phenylsilane in dichloromethane, gave a free carboxylic acid
at the C-terminus to which the thiol ethyl 3-mercaptopropio-
nate could be directly coupled (Scheme 2A). Conditions for
on-resin thioesterification have been described by Miranda
and co-workers, who showed that DMF/CH2Cl2 solvent mix-
tures in combination with high DIC and HOBt concentra-
tions (>0.5m) and the use of excess thiol derivative (0.5–
0.6m) reduced C-terminal epimerization to ꢀ1%.[49] This
protocol was therefore used to generate thioesters in our
studies. Cleavage of the peptide thioester from the solid sup-
port by treatment with TFA/thioanisole/triisopropylsilane/
water (85:5:5:5 by volume) proceeded with concomitant
amino acid side chain deprotection and gave the desired C-
terminal glutamine and asparagine peptide thioesters in 96
Chem. Eur. J. 2008, 14, 3620 – 3629
ꢀ 2008 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
3621

C.-H. Wong et al.
Ac-LYRANG-SR after depro-
tection and cleavage from the
solid support. The trace is pre-
dominantly a single peak which
was confirmed to be the desired
product by MALDI-TOF mass
spectrometry (Figure 1B). After
purification by preparative
HPLC, the peptide thioester
was obtained in quantitative
yield. We were delighted to find
that the efficiency of generating
these C-terminally extended
peptide thioesters using the
HATU-mediated
coupling
method was not limited to ex-
tension with glycine thioester 5,
as similar yields were also ob-
tained for the other three thio-
esters 6–8 (Table 1, entries 8–
Scheme 2. Synthesis of peptide thioesters by immobilizing the side chains of A) aspartic acid (n=1) and gluta-
mic acid (n=2) onto Rink amide resin and B) aspartic acid (n=1) and glutamic acid (n=2) onto Wang resin,
followed by C-terminal deprotection and thiol or amino acid thioester coupling (conditions a: 1) [Pd
ACHTREUNG
PhSiH, CH2Cl2; 2) R1SH, DIC, HOBt, DIEA, CH2Cl2, DMF; 3) TFA/TIS/TA/H2O (85:5:5:5); conditions b: 1)
[Pd
(PPh3)4], PhSiH, CH2Cl2, 2) H2N-CHR2-COSR1·HCl, HATU, DIEA, CH2Cl2, 3) TFA/TIS/TA/H2O
A
2
(85:5:5:5); R1 =ꢁ
C
(Scheme 2A) where glutamic
acid served as anchor (Table 1,
A
fluoroacetic acid; TIS=triisopropylsilane; TA=thioanisole).
entries 3–6).
At this stage, we were inter-
ested in whether other resins
and 95% yield, respectively, after HPLC purification
(Table 1, entries 1 and 2).
could be used for immobilization of amino acid side chains.
To this end, Fmoc-Asp-OAllyl and Fmoc-Glu-OAllyl were
loaded onto Wang resin using the symmetrical anhydride
method.[51] Subsequent peptide synthesis and allyl deprotec-
tion were conducted in a similar fashion to that described
for Rink amide resin (Scheme 2B). We chose not to attempt
direct thioesterification of the C-terminus due to the report-
ed formation of five- and six-membered cyclic anhydrides
for peptide thioesters containing C-terminal aspartic and
glutamic acid residues.[52] Anhydride formation would not be
expected for thioesters extended at the C-terminus. Indeed,
HATU-mediated C-terminal coupling of glycine, alanine,
methionine and phenylalanine thioesters 5–8 gave, after
cleavage from the resin and purification by preparative
HPLC, the corresponding peptide thioesters in excellent
yields (88%–quant.) (Table 1, entries 11–18).
In order to further broaden the scope of this approach,
our attention now turned to the immobilization of other
amino acid side chains using an alternative resin. There are
several commercially available resins commonly employed
in solid-phase organic synthesis, which can theoretically be
used for the immobilization of amino acids. The resin
chosen must allow for the construction of amino acid-resin
linkages which are stable to the conditions of standard
Fmoc SPPS, but cleave when treated with acid. Bromo-(4-
methoxyphenyl)methyl polystyrene fulfilled these require-
ments and has previously been used for the immobilization
of alcohols, amines and thiols.[53] We therefore explored the
possibility of using this resin for the side chain anchoring of
We next examined the possibility of extending the scope
of the side chain anchoring strategy, such that it could serve
as a general strategy for the production of a larger range of
thioester targets. Barany and co-workers have previously
used amino acid thioesters as coupling partners to the C-ter-
minus of solid-phase bound peptides to access thioesters
without inducing epimerization.[44] Encouraged by this
report, we sought to carry out a similar strategy using side
chain anchored peptide 1. To this end, glycine, alanine, phe-
nylalanine and methionine thioesters were first synthesized
in solution. This entailed DIC- and HOBt-mediated cou-
pling of the appropriate Boc-protected amino acids with
ethyl 3-mercaptopropionate, followed by acid-catalyzed Boc
removal to afford the desired amino acid thioesters 5–8 in
good yields (Scheme 3). Glycine was chosen as its thioesters
are known to be very efficient acyl donors in ligation reac-
tions, alanine is an example of an amino acid carrying an
alkyl chain, phenylalanine is a representative aromatic
amino acid and methionine contains a heteroatom in the
side chain but is unable to serve as an anchor. These four
amino acids cannot be easily immobilized onto solid sup-
ports and therefore their corresponding C-terminal thioest-
ers are inaccessible by the method previously described.
Palladium-catalyzed allyl deprotection of solid-supported
peptide 1 was followed by HATU-mediated coupling of
amino acid thioesters 5–8.[44] Figure 1A shows a representa-
tive analytical HPLC trace of the crude peptide thioester
3622
www.chemeurj.org
ꢀ 2008 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2008, 14, 3620 – 3629

Peptide and Glycopeptide Thioesters
FULL PAPER
Table 1. Scope of the side chain anchoring strategy for the synthesis of
peptide thioesters using glutamic acid and aspartic acid bound to Rink
amide and Wang resin (R=S-(CH2)2COOEt).
the amino acids serine, threonine, tyrosine and cysteine.
These constructs could then be used in the solid-phase syn-
thesis of peptide thioesters using our method. In order to
achieve this goal, we first had to generate the desired amino
acid coupling partners containing free side chains and an
allyl-protected C-terminus. Their synthesis was carried out
in solution by treating Fmoc-Ser-OH, Fmoc-Thr-OH, and
Fmoc-Tyr-OH with allyl bromide in the presence of diiso-
propylethylamine, to give the desired allyl esters in good
yields (Scheme 4A). Fmoc-Cys-OAllyl was synthesized from
Entry Resin
Amino acid anchor
AA1
C-terminal
extension
AA2
Isolated yield
[%]
1
Rink
amide
Rink
amide
Rink
amide
Rink
amide
Rink
amide
Rink
amide
Rink
amide
Rink
Gln
Asn
Gln
Gln
Gln
Gln
Asn
Asn
Asn
Asn
–
96
Fmoc-Cys(Trt)-OH in two steps, via allyl ester formation
G
2
–
95
followed by trityl deprotection to liberate the thiol side
chain (Scheme 4B).
3
Gly
Ala
Phe
Met
Gly
Ala
Phe
Met
quant.
quant.
quant.
quant.
quant.
quant.
96
4
5
6
7
8
amide
Rink
amide
Rink
9
10
90
amide
Wang
Wang
Wang
Wang
Wang
Wang
Wang
Wang
Scheme 4. Synthesis of A) Fmoc-Ser-OAllyl, Fmoc-Thr-OAllyl and
Fmoc-Tyr-OAllyl and B) Fmoc-Cys-OAllyl.
11
12
13
14
15
16
17
18
Glu
Glu
Glu
Glu
Asp
Asp
Asp
Asp
Gly
Ala
Phe
Met
Gly
Ala
Phe
Met
95
quant.
quant.
99
quant.
97
Pre-loading of bromo-(4-methoxyphenyl)methyl resin was
conducted by reacting derivatized amino acids 9–12 in the
presence of N,N-diisopropylethylamine for 18–48 h to give
final resin loadings of 0.79–1.28 mmolgꢁ1 (depending on the
steric bulk of the amino acid side chain). Reactions were
carried out in the absence of light to prevent any potential
photochemical side reactions on the resin.[53] Elongation of
peptides was achieved following the standard Fmoc protocol
to give fully assembled constructs 13–16 (Scheme 5), fol-
90
88
lowed by allyl deprotection by treatment with [Pd(PPh3)4]
G
and phenylsilane in dichloromethane to give the free car-
boxylic acid. At this stage, peptide thioesters could be gener-
ated by direct DIC/HOBt-mediated coupling of ethyl 3-mer-
captopropionate as previously described for Rink amide
resin (Scheme 2A). Acid-catalyzed side chain deprotection
and cleavage from the resin gave the desired thioesters in
good yields in all cases. As a representative example,
Figure 2 shows the crude analytical HPLC trace of Ac-
LYRAC-SR. In contrast to the crude trace of Ac-
LYRANG-SR (Figure 1A), there are two products produced
in the synthesis. The first (peak a, Figure 2A) accounts for
ꢀ25% and corresponds to the peptide acid Ac-LYRAC-
OH, whereas the second accounts for ꢀ75% (peak b, Fig-
ure 2A) and corresponds to the desired peptide thioester
Scheme 3. Synthesis of amino acid thioesters.
Ac-LYRAC-S(CH2)2CO2Et. Unfortunately, extending reac-
A
Figure 1. A) Crude analytical HPLC trace of Ac-LYRANG-SR after
cleavage from Rink amide resin and B) MALDI-TOF mass spectrum of
the desired product (R=S-(CH2)2COOEt; [M+H]+calcd =851.99;
matrix=a-cyano-4-hydroxycinnamic acid).
tion times in an attempt to convert all of the free acid to the
desired thioester did not increase the yield substantially.
However, under the conditions described here, the C-termi-
Chem. Eur. J. 2008, 14, 3620 – 3629
ꢀ 2008 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
3623

C.-H. Wong et al.
The application of the side
chain anchoring approach to
the synthesis of glycopeptide
thioesters was next examined.
These are currently difficult to
obtain, and a general strategy
for their synthesis would be ex-
tremely useful. Accordingly,
bromo-(4-methoxyphenyl)meth-
yl resin was first derivatized via
the side chain of Fmoc-Ser-
OAllyl (9). Standard Fmoc-
strategy SPPS was used to con-
struct peptide 17, incorporating
the glycosylated serine residue
as the N-terminal residue. Allyl
deprotection of 17 was followed
by coupling of methionine thio-
ester 7 or phenylalanine thio-
ester 8. Subsequent deprotec-
tion and cleavage from the
resin gave the desired C-termi-
nal methionine- and phenylala-
nine-containing
glycopeptide
thioesters in 66 and 73% isolat-
ed yield, respectively, after
HPLC purification (Table 3, en-
tries 1 and 2).
Having demonstrated that
the side chain anchoring strat-
Scheme 5. Synthesis of peptide thioesters by side chain anchoring of A) serine and threonine B) tyrosine and
C) cysteine onto bromo-(4-methoxyphenyl)methyl resin, C-terminal deprotection and thiol or amino acid thio-
ester coupling (conditions a: 1) [Pd
(PPh3)4], PhSiH, CH2Cl2; 2) R1SH, DIC, HOBt, DIEA, CH2Cl2, DMF; 3)
N
TFA/TIS/TA/H2O (85:5:5:5); conditions b: 1) [Pd
(PPh3)4], PhSiH, CH2Cl2, 2) H2N-CHR2-COSR1·HCl, HATU,
N
DIEA, CH2Cl2, 3) TFA/TIS/TA/H2O (85:5:5:5); R1 = -(CH2)2COOEt; R2 =amino acid side chain; R3 =H or
Me.
Table 2. Scope of the side chain anchoring strategy using serine, threo-
nine, tyrosine and cysteine bound to bromo-(4-methoxyphenyl)methyl
resin (R=S-(CH2)2COOEt).
Entry
Amino acid anchor
AA1
Added amino acid
AA2
Isolated yield
[%]
1
2
3
4
5
6
7
8
Ser
Thr
Tyr
Cys
Ser
Ser
Ser
Ser
Thr
Thr
Thr
Thr
Tyr
Tyr
Tyr
Tyr
Cys
Cys
Cys
Cys
–
–
–
–
70
51
74
70
69
68
75
55
81
80
90
67
58
71
65
80
56
64
55
54
Figure 2. A) Crude analytical HPLC trace of direct thioester formation
from 16. Peak a=Ac-LYRAC-OH, peak b=Ac-LYRAC-SR and B)
MALDI-TOF mass spectrum of peak b (R=S-(CH2)2COOEt; [M+H]+
calcd =783.98; matrix=2’,4’,6’-trihydroxyacetophenone).
Gly
Ala
Phe
Met
Gly
Ala
Phe
Met
Gly
Ala
Phe
Met
Gly
Ala
Phe
Met
nal cysteine thioester was isolated in 70% yield after purifi-
cation by preparative HPLC (Table 2, entry 4).
9
10
11
12
13
14
15
16
17
18
19
20
The corresponding C-terminal serine, threonine and tyro-
sine peptide thioesters were also obtained in good yields
ranging from 51–74% (Table 2, entries 1–3). In a similar
manner to Wang and Rink amide immobilized peptides 1–4,
amino acid thioesters 5–8 could be coupled to the free acid
of 13–16 using the HATU-mediated coupling reaction.
Using this method, C-terminal glycine, alanine, methionine
and phenylalanine thioesters were obtained in good yields
(55–91%; Table 2, entries 5–20).
3624
www.chemeurj.org
ꢀ 2008 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2008, 14, 3620 – 3629

Peptide and Glycopeptide Thioesters
FULL PAPER
Table 3. Synthesis of glycopeptide thioesters using the side chain anchor-
ing strategy.
synthesized on Wang resin using standard Fmoc SPPS. Cou-
pling of the N-linked glycosyl amino acid building block
Fmoc-Asn(Ac3AcNH-b-Glc)-OH was performed in a similar
fashion to that described for glycopeptide thioester 19 and a
similar drop in the loading was observed. The remainder of
the synthesis proceeded smoothly without a further drop in
loading, and Boc-Cys(Trt)-OH was coupled as the final
A
amino acid residue. Treatment with TFA/thioanisole/triiso-
propylsilane/water (85:5:5:5 by volume) removed the side
chain protecting groups, the N-terminal Boc group, and re-
leased the crude product from the resin. After HPLC purifi-
cation, the desired glycopeptide 18 was obtained in 67%
yield relative to the final resin loading. Glycopeptide 18 and
glycopeptide thioester 19 were submitted to a native chemi-
cal ligation reaction using a ligation buffer consisting of a
4:1 mixture of NMP and 1m HEPES/6m guanidine hydro-
chloride pH 8.5, for 24 h. Purification by semi-preparative
HPLC afforded the 40-mer ligation product in 76% yield.
Subsequent treatment with a solution of 5% hydrazine con-
taining 60 mm dithiothreitol for 1 h removed the acetate
groups from the two glycans to afford the desired 40 amino
acid N-terminal fragment of EPO (20) in 95% yield
(Scheme 6).
Entry Glycosylated amino
acid
C-terminal amino acid
AA1
Isolated yield
[%]
1
2
Ser
Ser
Met
Phe
67
73
egy was amenable to the synthesis of glycopeptide thioest-
ers, our next goal was to apply this method to the synthesis
of a biologically relevant target. Construction of homogene-
ous glycoproteins is an important yet challenging endeavor,
and to date the total synthesis of a target containing a
native protein sequence has yet to be achieved. We chose to
synthesize a fragment of the therapeutic glycoprotein eryth-
ropoietin (EPO). EPO, a 166-residue glycoprotein hormone
possessing four glycosylation sites (three N-linked and one
O-linked) is produced in kidney cells and is responsible for
the regulation of red blood cell production.[54] Clinically, it is
obtained as a mixture of glycoforms by over-expression in
mammalian cells, and used for the treatment of anemia asso-
ciated with a number of illnesses.[7] Current efforts are fo-
cused on the total chemical synthesis of single glycoforms of
this medicinally significant glycoprotein for detailed biologi-
cal studies.[55–57] We embarked on the synthesis of a 40-mer
N-terminal fragment of EPO, containing two of the four gly-
cosylation sites (both N-linked glycans). This would serve as
a building block for the total synthesis of this target and as
an elegant demonstration of the side chain anchoring strat-
egy for the synthesis of large peptide and glycopeptide thio-
esters. The synthesis of the desired glycopeptide thioester
fragment (EPO 1–28) was performed on bromo-(4-methoxy-
phenyl)methyl resin. Pre-loading of the resin with Fmoc-
Ser-OAllyl (9) and peptide elongation was achieved as de-
scribed previously. Fmoc-Asn(Ac3AcNH-b-Glc)-OH was
coupled as the third amino acid, however, only one equiva-
lent relative to the resin was used to minimize waste of this
precious building block. Although the reaction time was ex-
tended to six hours, this coupling proceeded in only 60%
yield. To mitigate further loss of loading, all subsequent
amino acids were doubly coupled. After coupling Boc-Ala-
OH as the final amino acid residue, removal of the C-termi-
nal allyl group gave access to the free carboxylic acid which
was elongated in the N- to C-terminal direction by incorpo-
rating glycine thioester 5 under the HATU-mediated cou-
pling conditions previously described. Treatment of the resin
with TFA/thioanisole/triisopropylsilane/water (85:5:5:5 by
volume) removed the side chain protecting groups, the N-
terminal Boc group and cleaved the crude glycopeptide
thioester 19 from the resin. After purification by HPLC, 19
was successfully isolated. The 12-mer glycopeptide 18 was
Conclusion
In summary, an efficient strategy for the synthesis of peptide
and glycopeptide thioesters has been described. The method
relies on immobilizing the side chains of a number of amino
acids to suitable resins. In total six proteinogenic amino
acids were successfully anchored: aspartic acid and glutamic
acid to Wang or Rink amide resin (use of Rink amide leads
to asparagine and glutamine upon cleavage) and serine,
threonine, tyrosine and cysteine to bromo-(4-methoxyphe-
nyl)methyl resin. Pre-loading of allyl protected amino acids
and elongation using standard Fmoc-SPPS gave solid-sup-
ported peptides in high yields. C-terminal deallylation liber-
ated the C-terminal carboxylic acid, to which thiols could be
coupled to produce peptide thioesters containing C-terminal
asparagine, glutamine, serine, threonine, tyrosine and cys-
teine in high yields after resin cleavage. In addition to the
amino acids studied here, lysine,[49] arginine, histidine and
tryptophan could, in theory, also be immobilized to solid
supports. These strategies are currently being pursued by
our laboratory and will allow for the anchoring of all protei-
nogenic amino acids containing reactive side chains. In addi-
tion to direct coupling of thiols, and in order to broaden the
scope of the method, amino acid thioesters were also cou-
pled to the C-terminus to afford, after deprotection and
cleavage from the resin, a variety of peptide thioesters in an
efficient manner. Glycopeptide thioesters were also con-
structed using this strategy. Four amino acid thioesters (Gly,
Ala, Phe, Met) were used in these C-terminal extension
studies, and it is envisaged that the remaining four unan-
chorable amino acids (Ile, Leu, Pro and Val) could be syn-
thesized as their thioesters and incorporated in a similar
Chem. Eur. J. 2008, 14, 3620 – 3629
ꢀ 2008 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
3625

C.-H. Wong et al.
Scheme 6. Synthesis of the N-terminal fragment of the glycoprotein hormone erythropoietin (Fmoc=9-fluorenylmethoxycarbonyl; NMP=1-methyl-2-
pyrrolidinone; buffer=1m HEPES, 6m Gn·HCl pH 8.5; [M+H]+calcd =4881.41; matrix=sinapinic acid).
Experimental Section
fashion. Combined, direct coupling of a thiol or amino acid
thioester to the C-terminus of side chain anchored peptides
represents a general strategy for the synthesis of peptide
General: 1H NMR and 13C NMR were recorded on a Bruker DRX-600
spectrometer equipped with a CryoProbe operating at 600 MHz and
and glycopeptide thioesters. As such, the method should
allow access to a large majority of potential targets amena-
ble to subsequent ligation reactions for the construction of
proteins and glycoproteins. The method was successfully ap-
plied to the synthesis of a glycopeptide thioester fragment
of the therapeutic glycoprotein EPO, to exemplify its utility
in the construction of complex targets. Subsequent ligation
to a glycopeptide by native chemical ligation gave a 40
amino acid homogeneous N-terminal fragment of EPO bear-
ing two of the four glycosylation sites. Current research in
our laboratory is focused on using the side chain anchoring
strategy in combination with NCL and SAL to produce
other peptide and glycopeptide fragments of EPO, with a
view to the total synthesis of a homogeneous sample of this
therapeutic glycoprotein.
150 MHz respectively. Coupling constants (J) are reported in Hertz (Hz),
and chemical shifts are reported in parts per million (ppm) relative to tet-
ramethylsilane (TMS, 0.00 ppm). MALDI-TOF mass spectra were mea-
sured on a Voyager-DE Pro biospectrometry workstation by PerSeptive
Biosystems. Analytical HPLC was run on a Hitachi (D-7000 HPLC
system) instrument using an analytical column (Waters “XBridge BEH
130 C18”, 150”4.6 mm, 5 mm particle size or Grace Vydac “Protein &
Peptide C18”, 150”4.6 mm, 10 mm particle size, flow rate 1.5 mLminꢁ1
,
508C). Semi preparative HPLC was run on a Hitachi (D-7000 HPLC
system) instrument using a semi preparative column (Grace Vydac “Pro-
tein & Peptide C18”, 250”10 mm, 10–15 mm particle size, flow rate
4 mLminꢁ1). Preparative HPLC was run on a Hitachi (D-7000 HPLC
system) instrument using a preparative Column (Grace Vydac “Protein
&
Peptide C18”, 250”22 mm, 10–15 mm particle size, flow rate
8 mL/min). Detection of the signal was achieved with either photodiode
array or UV detector at a wavelength of l=280 nm (detection of Tyr).
For the purification and analytical traces of EPO-fragment glycopeptides,
3626
www.chemeurj.org
ꢀ 2008 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2008, 14, 3620 – 3629

Peptide and Glycopeptide Thioesters
FULL PAPER
detection was performed at l=230 nm. Additional analytical and experi-
mental details are available in the Supporting Information.
Capping: Acetic anhydride/pyridine (1:9 by volume) was added to the
resin. After 5 min the resin was washed with DMF (5”), CH2Cl2 (5”)
and DMF (5”).
Materials: Water was taken from a Milli-Q ultra pure water purification
system (Millipore corp.). DMF was purchased as peptide synthesis grade
from Alfa Aesar. Commercial reagents were purchased from Sigma-Al-
drich or Acros Organics and were used without further purification. An-
hydrous-grade solvents were purchased from Sigma-Aldrich and were
used directly. Resins, protected amino acids and PyBOP were purchased
from Novabiochem. Deuterated solvents were purchased from Cam-
bridge Isotope Laboratories Inc.
Final N-terminal deprotection: When the peptide was fully assembled,
the N-terminal Fmoc group was removed by treatment with 10% piperi-
dine/DMF (2”5 min) and the loading was determined by measuring the
absorbance of the resulting fulvene–piperidine adduct at l=302 nm.
After acetylating the N-terminus with acetic anhydride/pyridine (1:9 by
volume) for 15 min, the resin was washed with DMF (10”) and CH2Cl2
(10”) and then dried in vacuo.
General procedure for the direct thioesterification on solid support:
25 mmol of resin was swollen in dry CH2Cl2 (5 mL) for 30 min, followed
Side-chain anchoring onto Rink amide resin: Rink amide resin (100–200
mesh; 1% DVB) (290 mg, loading=0.69 mmolgꢁ1, 200 mmol) was initial-
ly washed with DMF (5”5 mL), CH2Cl2 (5”5 mL) and DMF (5”5 mL),
treated with DMF/piperidine (9:1 by volume) (2”5 min) and washed
with DMF (5”5 mL), CH2Cl2 (5”5 mL) and DMF (5”5 mL). Fmoc-
Glu-OAllyl or Fmoc-Asp-OAllyl (800 mmol) in dry DMF (4 mL) contain-
ing PyBOP (416 mg, 800 mmol) and NMM (176 mL, 1600 mmol) was pre-
activated for 4 min and then added to the resin. After two hours of shak-
ing, the resin was washed with DMF (5”5 mL), CH2Cl2 (5”5 mL) and
DMF (5”5 mL), treated with Ac2O/pyridine (1:9 by volume) for 10 min
and then washed with DMF (5”5 mL), CH2Cl2 (5”5 mL) and DMF (5”
5 mL). Treatment of the resin with 10% piperidine/DMF (2”5 min) and
measurement of the resulting fulvene–piperidine adduct at l=302 nm
showed that the yield of the side chain anchoring was quantitative.
by the addition of a solution of [Pd(PPh3)4](25 mg, 22 mmol) and Ph3SiH
A
(123 mL, 1 mmol) in dry CH2Cl2 (2 mL). The resin was shaken for 1 h and
the procedure was repeated. Afterwards, the resin was washed with
CH2Cl2 (10”5 mL), DMF (5”5 mL) and CH2Cl2 (5”5 mL). A solution
of ethyl 3-mercaptopropionate (77 mL, 600 mmol), anhydrous HOBt
(101 mg, 750 mmol), DIEA (161 mL, 938 mmol) and DIC (116 mL,
750 mmol) in CH2Cl2/DMF (1.5 mL, 4:1 by volume) was added and the
resin was shaken for 1 h. This thioesterification step was repeated before
washing the resin with CH2Cl2 (5”5 mL), DMF (5”5 mL) and CH2Cl2
(10”5 mL).
Cleavage:
A
mixture of TFA/thioanisole/triisopropylsilane/water
(85:5:5:5 by volume) was added. After 2 h, the resin was washed with
TFA (4”4 mL).
Side-chain anchoring onto Wang resin: Wang resin LL (100–200 mesh)
(454 mg, loading=0.44 mmolgꢁ1, 200 mmol) was initially washed with
DMF (5”5 mL), CH2Cl2 (5”5 mL) and DMF (5”5 mL), and then al-
lowed to swell in DMF (5 mL) for 30 min. Fmoc-Glu-OAllyl or Fmoc-
Asp-OAllyl (2 mmol, 10 equiv) were dissolved in dry CH2Cl2 (15 mL).
DIC (155 mL, 1 mmol) in dry CH2Cl2 (5 mL) was added to the above
amino acid solution at 08C and the reaction stirred for 20 min, before the
solvent was removed in vacuo. The residue was redissolved in DMF
Work-up: The combined solutions were concentrated in vacuo. The resi-
due was dissolved in water containing 30% MeCN + 0.1% TFA, puri-
fied by preparative HPLC and analyzed by MALDI-TOF/MS (matrix: a-
cyano-4-hydroxycinnamic acid).
General procedure for the C-terminal introduction of amino acid thioest-
ers: 25 mmol of resin were swollen in dry CH2Cl2 (5 mL) for 30 min, fol-
lowed by the addition of a solution of [Pd(PPh3)4](25 mg, 22 mmol) and
A
Ph3SiH (123 mL, 1 mmol) in dry CH2Cl2 (2 mL). The resin was shaken for
1 h and the procedure was repeated. The resin was subsequently washed
with CH2Cl2 (10”5 mL), DMF (5”5 mL) and CH2Cl2 (5”5 mL). A solu-
tion of amino acid thioesters 5–8 (250 mmol) and DIEA (86 mL,
500 mmol) in dry CH2Cl2 (1 mL) was added to the resin. HATU (95 mg,
250 mmol) was then added in solid form and the resin shaken for 1 h. The
coupling procedure was repeated and the resin washed with CH2Cl2 (5”
5 mL), DMF (5”5 mL) and CH2Cl2 (10”5 mL).
(5 mL) before adding to the resin.
A solution of DMAP (2.5 mg,
20 mmol) in DMF (1 mL) was added and the resin shaken for 1 h before
washing with DMF (5”5 mL), CH2Cl2 (5”5 mL) and DMF (5”5 mL).
Capping with Ac2O/pyridine (1:9 by volume) for 10 min was followed by
resin washing with DMF (5”5 mL), CH2Cl2 (5”5 mL) and DMF (5”
5 mL). Treatment of the resin with 10% piperidine/DMF (2”5 min) and
measurement of the resulting fulvene–piperidine adduct at l=302 nm
showed that the yield of the side chain anchoring was quantitative.
Cleavage:
A
mixture of TFA/thioanisole/triisopropylsilane/water
Side-chain anchoring onto bromo-(4-methoxyphenyl)methylpolystyrene:
Bromo-(4-methoxyphenyl)methylpolystyrene resin (174 mg, loading=
2.3 mmolgꢁ1, 400 mmol) was initially washed with DMF (5”5 mL) and
CH2Cl2 (10”5 mL) and the resin swelled with CH2Cl2 (5 mL) for 1 h with
the exclusion of light. A solution of the C-terminal allyl-protected Fmoc-
amino acids 7–10 (1200 mmol), DIEA (397 mL, 2400 mmol) in CH2Cl2
(4 mL) was added to the resin, which was shaken for 12–48 h at RT with
the exclusion of light. The resin was washed with CH2Cl2 (5”5 mL) and
DMF (10”5 mL). Treatment of the resin with 10% piperidine/DMF (2”
5 min) and measurement of the resulting fulvene–piperidine adduct at
l=302 nm was used to determine the loading. NB: The commercially
available resin had a very high loading, therefore pre-loadings of the
resins were conducted for times that allowed for the resin to be loaded
less than 1.4 mmolgꢁ1 to prevent resin crowding and product aggregation.
Final loadings: Fmoc-Ser-OAllyl 7 (18 h): loading=1.28 mmolgꢁ1; Fmoc-
Thr-OAllyl 8 (48 h): loading=0.79 mmolgꢁ1; Fmoc-Tyr-OAllyl 9 (46 h):
(85:5:5:5 by volume) was added. After 2 h, the resin was washed with
TFA (4”4 mL).
Work-up: The combined solutions were concentrated in vacuo. The resi-
due was dissolved in water containing 30% MeCN + 0.1% TFA, puri-
fied by preparative HPLC and analyzed by MALDI-TOF/MS (matrix: a-
cyano-4-hydroxycinnamic acid).
General procedure for the synthesis of amino acid thioesters: Boc-pro-
tected amino acids (6.00 mmol) were dissolved in anhydrous THF
(20 mL) and cooled to 08C. Ethyl 3-mercaptopropionate (0.77 mL,
6.00 mmol) was added dropwise, followed by HOBt (1.10 g, 7.20 mmol).
The solution was stirred at 08C for 15 min before the dropwise addition
of DIC (1.11 mL, 7.20 mmol). The reaction was stirred at RT for 15 h
before the urea by-product was removed by filtering through a plug of
Celite. The solvent was removed in vacuo and the resulting residue re-
dissolved in ethyl acetate (30 mL). The solution was washed with a 4%
NaHCO3 solution (3”30 mL), 1m HCl (3”30 mL), brine (2”30 mL),
dried (Na2SO4) and the solvent removed in vacuo. Purification was ach-
ieved by column chromatography (hexane/ethyl acetate 3:1) to give the
desired Boc-protected amino acid thioesters. The resulting thioester was
dissolved in ethyl acetate (15 mL) and the solution cooled to 08C before
saturating with HCl(g). The reaction was stirred at RT for 2 h, before the
solvent was removed in vacuo. The resulting residue was re-dissolved in
ethyl acetate and the solvent removed in vacuo (3”20 mL) and gave the
desired amino acid thioesters as white solids.
loading=1.02 mmolgꢁ1
0.90 mmolgꢁ1
;
Fmoc-Cys-OAllyl
10
(32 h):
loading=
.
General procedures for SPPS of peptides and glycopeptides following
the Fmoc strategy (Iterative peptide assembly)
Deprotection: The resin was treated with 10% piperidine/DMF (2”
5 min) and subsequently washed (5”DMF, 5” CH2Cl2, 5”DMF).
Amino acid coupling:
A preactivated solution of 4 equiv protected
amino acid (final concentration 0.1m in DMF) using 4 equiv PyBOP and
8 equiv NMM was added to the resin. After 30 min, the resin was washed
with DMF (5”), CH2Cl2 (5”) and DMF (5”).
Ethyl 3-(2-aminoacetylthio)propanoate hydrochloride (5): Yield=1.09 g,
79% over two steps; 1H NMR (500 MHz, CDCl3): d
=
1.33 (t, J=
Chem. Eur. J. 2008, 14, 3620 – 3629
ꢀ 2008 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
3627

C.-H. Wong et al.
7.0 Hz, 3H, CH3), 2.77 (t, J=7.0 Hz, 2H, CH2), 3.35 (t, J=7.0 Hz, 2H,
CH2), 4.18 (s, 2H, CH2), 4.22 ppm (q, J=7.0 Hz, 2H, CH2); 13C NMR
(125 MHz, CDCl3): d = 14.5, 25.1, 31.5, 48.3, 62.0, 173.0, 194.2 ppm;
HRMS (ESI-TOF): m/z: calcd for C7H14NO3S: 192.0685; found: 192.0685
[M+H]+.
7.2 Hz, 2H, CH2), 7.39 (t, J=7.2 Hz, 2H, CH2), 7.55 (t, J=6.0 Hz, 2H,
CH2), 7.75 ppm (d, J=7.8 Hz, 2H, CH2); 13C NMR (150 MHz, CDCl3): d
= 37.4, 47.1, 55.0, 66.2, 67.0, 115.5, 119.2, 119.9, 125.0, 127.0, 127.2, 127.7,
130.5, 131.3, 141.3, 143.7, 155.1, 155.7, 171.4 ppm; HRMS (ESI-TOF):
m/z: calcd for C27H25NO5: 444.1805; found: 444.1800 [M+H]+.
(S)-Ethyl 3-(2-aminopropanoylthio)propanoate hydrochloride (6):
Yield=1.12 g, 77% over two steps; 1H NMR (600 MHz, CDCl3): d =
1.24 (t, J=6.6 Hz, 3H, CH3), 1.56 (d, J=7.2 Hz, 3H, CH3), 2.68 (t, J=
6.6 Hz, 2H, CH2), 3.25 (td, J=2.4, 6.6 Hz, 2H, CH2), 4.14 (q, J=7.2, 2H,
CH2), 4.28 ppm (q, J=7.2 Hz, 1H, CH); 13C NMR (150 MHz, CDCl3): d
= 14.5, 17.6, 25.2, 34.8, 56.4, 62.0, 172.9, 198.2 ppm; HRMS (ESI-TOF):
m/z: calcd for C8H16NO3S: 206.0845; found: 206.0845 [M+H]+.
(R)-Allyl 2-(((9H-fluoren-9-yl)methoxy)carbonylamino)-3-mercaptopro-
panoate (12): Fmoc-Cys(Trt)-OH (6.9 g, 11.7 mmol) was dissolved in an-
A
hydrous DMF (75 mL) and cooled to 08C. N,N-diisopropylethylamine
(4.08 mL, 23.4 mmol) and allyl bromide (2.02 mL, 23.4 mmol) were
added dropwise and the reaction stirred at RT for 16 h, before diluting
with ethyl acetate (200 mL). The reaction was washed with water (4”
200 mL), brine (200 mL), dried (Na2SO4) and the solvent removed in
vacuo. The resulting residue was dissolved in TFA/CH2Cl2 (50 mL, 9:1 by
volume). Triisopropylsilane (10 mL) was added dropwise and the reaction
stirred at RT for 1 h. The solvent was removed in vacuo and the product
purified by column chromatography (hexane/ethyl acetate 3:1) to afford
Fmoc-Cys-OAllyl as a white solid. Yield=4.1 g, 91% over two steps; RF
(hexane/ethyl acetate 3:1)=0.30; 1H NMR (600 MHz, CDCl3): d = 3.00
(m, 2H, CH2), 4.24 (t, J=6.8 Hz, 1H, CH), 4.43 (m, 2H, CH2), 4.69 (m,
3H, CH2 + CH), 5.29 (d, J=10.2 Hz, 1H, CH), 5.35 (d, J=17.4 Hz, 1H,
CH), 5.72 (d, J=7.2 Hz, 1H, NH), 5.92 (m, 1H, CH), 7.32 (t, J=7.2 Hz,
2H, 2”CH), 7.40 (t, J=7.2 Hz, 2H, 2”CH), 7.61 (d, J=6.6 Hz, 2H, 2”
CH), 7.77 ppm (d, J=7.8 Hz, 2H, 2”CH); 13C NMR (150 MHz, CDCl3):
d = 27.1, 47.1, 55.2, 66.4, 67.0, 119.4, 120.0, 125.0, 127.0, 127.7, 131.2,
141.3, 143.6, 155.6, 169.6 ppm; HRMS (ESI-TOF): m/z: calcd for
C21H22NO4S: 384.1264; found: 384.1268 [M+H]+.
(S)-Ethyl 3-(2-amino-4-(methylthio)butanoylthio)propanoate hydrochlo-
1
ride (7): Yield=1.25 g, 69% over two steps; H NMR (600 MHz, CDCl3):
d = 1.25 (t, J=7.2 Hz, 3H, CH3), 2.12 (s, 3H, CH3), 2.17 (m, CHH), 2.25
(m, CHH), 2.65 (t, J=7.2 Hz, 2H, CH2), 2.70 (td, J=1.2, 6.6 Hz, 2H,
CH2), 3.22–3.32 (m, 2H, CH2), 4.14 (q, J=7.2, 2H, CH2), 4.38 ppm (m,
1H, CH); 13C NMR (150 MHz, CDCl3): d = 14.6, 15.0, 25.5, 29.8, 32.0,
34.7, 59.3, 62.0, 172.9, 197.3 ppm; HRMS (ESI-TOF): m/z: calcd for
C10H21NO3S2: 266.0885; found: 266.0953 [M+H]+.
(S)-Ethyl 3-(2-amino-3-phenylpropanoylthio)propanoate hydrochloride
(8): Yield=1.30 g, 68% over two steps; 1H NMR (600 MHz, CDCl3): d
= 1.24 (t, J=7.2 Hz, 3H, CH3), 2.62 (m, 2H, CH2), 3.16 (m, 2H, CH2),
3.23 (m, 2H, CH2), 4.13 (q, J=7.2 Hz, 2H, CH2), 4.47 (t, J=7.2 Hz, 1H,
CH), 7.27–7.37 ppm (m, 5H); 13C NMR (150 MHz, CDCl3): d = 14.6,
23.4, 34.7, 38.7, 61.3, 62.0, 129.1, 130.2, 130.7, 134.9, 172.9, 197.2 ppm;
HRMS (ESI-TOF): m/z: calcd for C14H20NO3S: 282.1158; found:
282.1158 [M+H]+.
General procedure for the synthesis of Fmoc-protected amino acid allyl
esters: Fmoc-protected amino acids (11.7 mmol) were dissolved in anhy-
drous DMF (75 mL) and cooled to 08C. N,N-diisopropylethylamine
(4.08 mL, 23.4 mmol) and allyl bromide (2.02 mL, 23.4 mmol) were
added dropwise and the reaction stirred at RT for 16 h, before diluting
with ethyl acetate (200 mL). The reaction was washed with water (4”
200 mL), brine (200 mL), dried (Na2SO4) and the solvent removed in
vacuo. Purification was achieved by column chromatography (hexane/
ethyl acetate 3:1) to afford the desired amino acid allyl esters as white
solids.
Acknowledgements
This work was supported by the NIH and the Skaggs Institute for Chemi-
cal Biology. S.F. is grateful for a postdoctoral fellowship provided by the
Deutsche Akademische Austauschdienst (DAAD). R.J.P. is grateful for
funding provided by the Lindemann Trust Fellowship.
(S)-Allyl 2-(((9H-fluoren-9-yl)methoxy)carbonylamino)-3-hydroxypropa-
noate (9): Yield=4.20 g, 98%; RF (hexane/ethyl acetate 3:1)=0.06;
1H NMR (600 MHz, CDCl3): d = 3.91 (d, J=10.2 Hz, 1H, CH), 4.00 (d,
J=10.2 Hz, 1H, CH), 4.21 (t, J=7.2 Hz, 1H, CH), 4.29 (m, 3H), 4.68 (d,
J=4.8 Hz, 2H, CH2), 5.24 (d, J=10.2 Hz, 1H, CH), 5.33 (d, J=17.4 Hz,
1H, CH), 5.84 (d, J=7.2 Hz, 1H, CH), 5.90 (m, 1H, CH), 7.30 (t, J=
7.2 Hz, 2H, 2”CH), 7.39 (t, J=7.2 Hz, 2H, 2”CH), 7.60 (t, J=6.0 Hz,
2H, 2”CH), 7.76 ppm (d, J=7.2 Hz, 2H, 2”CH); 13C NMR (150 MHz,
CDCl3): d = 47.1, 56.1, 63.2, 66.3, 67.2, 118.9, 119.9, 125.0, 127.0, 127.7,
131.3, 141.3, 143.6, 156.2, 170.2 ppm; HRMS (ESI-TOF): m/z: calcd for
C21H21NO5: 368.1492; found: 368.1506 [M+H]+.
[1]R. Apweiler, H. Hermjakob, N. Sharon, Biochim. Biophys. Acta
Gen. Subj. 1999, 1473, 4–8.
[2]R. G. Spiro, Glycobiology 2002, 12, 43R–56R.
[3]R. A. Dwek, Chem. Rev. 1996, 96, 683–720.
[4]A. Varki, Glycobiology 1993, 3, 97–130.
[5]D. H. Dube, C. R. Bertozzi, Nat. Rev. Drug Discovery 2005, 4, 477–
488.
[6]G. Walsh, R. Jefferis, Nat. Biotechnol. 2006, 24, 1241–1252.
[7]T. B. Drueke, F. Locatelli, N. Clyne, K. Eckardt, I. C. Macdougall,
D. Tsakiris, H. Burger, A. Scherhag, New Engl. J. Med. 2006, 355,
2071–2084.
[8]B. G. Davis, Chem. Rev. 2002, 102, 579–601.
[9]M. R. Pratt, C. R. Bertozzi, Chem. Soc. Rev. 2005, 34, 58–68.
[10]L. Liu, C. S. Bennett, C. H. Wong, Chem. Commun. 2006, 21–33.
[11]A. Brik, S. Ficht, C. H. Wong, Curr. Opin. Chem. Biol. 2006, 10,
638–644.
[12]T. Wieland, E. Bokelmann, L. Bauer, H. U. Lang, H. Lau, Justus
Liebigs Ann. Chem. 1953, 583, 129–149.
[13]P. E. Dawson, T. W. Muir, I. Clarklewis, S. B. H. Kent, Science 1994,
266, 776–779.
A
butanoate (10): Yield=4.00 g, 90%; RF (hexane/ethyl acetate 3:1)=0.12;
1H NMR (600 MHz, CDCl3): d = 1.26 (d, J=6.0 Hz, 3H, CH3), 4.24 (t,
J=7.2 Hz, 1H, CH), 4.39 (m, 4H, 2”CH2), 4.68 (d, J=5.4 Hz, 2H, CH2),
5.25 (d, J=10.8 Hz, 1H, CH), 5.34 (d, J=17.4 Hz, 1H, OH), 5.65 (d, J=
8.4 Hz, 1H, CH), 5.91 (m, 1H, CH), 7.30 (t, J=7.2 Hz, 2H, 2”CH), 7.39
(t, J=7.2 Hz, 2H, 2”CH), 7.60 (t, J=6.0 Hz, 2H, 2”CH), 7.76 ppm (d,
J=7.2 Hz, 2H, 2”CH); 13C NMR (150 MHz, CDCl3): d = 19.9, 47.1,
59.1, 66.2, 67.2, 68.0, 119.0, 120.0, 125.1, 127.1, 127.7, 131.4, 141.3, 143.7,
156.7, 170.8 ppm; HRMS (ESI-TOF): m/z: calcd for C22H23NO5:
382.1649; found: 382.1651 [M+H]+.
[14]P. E. Dawson, S. B. H. Kent, Annu. Rev. Biochem. 2000, 69, 923–
960.
(S)-Allyl
2-(((9H-fluoren-9-yl)methoxy)carbonylamino)-3-(4-hydroxy-
[15]A. Brik, Y. Y. Yang, S. Ficht, C. H. Wong, J. Am. Chem. Soc. 2006,
128, 5626–5627.
[16]A. Brik, S. Ficht, Y. Y. Yang, C. S. Bennett, C. H. Wong, J. Am.
Chem. Soc. 2006, 128, 15026–15033.
[17]Y. Y. Yang, S. Ficht, A. Brik, C. H. Wong, J. Am. Chem. Soc. 2007,
129, 7690–7701.
[18]S. Ficht, R. J. Payne, A. Brik, C. H. Wong, Angew. Chem. 2007, 119,
6079–6083; Angew. Chem. Int. Ed. 2007, 46, 5975–5979.
phenyl) propanoate (11): Yield=4.37 g, 84%; RF (hexane/ethyl acetate
3:1)=0.15; 1H NMR (600 MHz, CDCl3): d = 3.01 (dd, J=6.0, 14.4 Hz,
1H, CH), 3.07 (dd, J=6.0, 14.4 Hz, 1H, CH), 4.20 (t, J=6.6 Hz, 1H,
CH), 4.35 (dd, J=7.2, 10.8 Hz, 1H, CH), 4.42 (dd, J=7.2, 10.8 Hz, 1H,
CH), 4.62 (d, J=6.0 Hz, 2H, CH2), 4.65 (m, 1H, CH), 5.25 (d, J=
10.8 Hz, 1H, CH), 5.34 (d, J=9.0 Hz, 1H, OH), 5.91 (m, 1H, CH), 6.71
(d, J=8.4 Hz, 2H, CH2), 6.94 (d, J=8.4 Hz, 2H, CH2), 7.30 (td, J=1.2,
3628
www.chemeurj.org
ꢀ 2008 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2008, 14, 3620 – 3629

Peptide and Glycopeptide Thioesters
FULL PAPER
[19]R. J. Payne, S. Ficht, S. Tang, A. Brik, Y. Y. Yang, D. A. Case, C. H.
Wong, J. Am. Chem. Soc. 2007, 129, 13527–13536.
[20]E. Saxon, J. I. Armstrong, C. R. Bertozzi, Org. Lett. 2000, 2, 2141–
2143.
[21]J. M. Stewart, Methods Enzymol. 1997, 289, 29–44.
[22]L. E. Canne, S. M. Walker, S. B. H. Kent, Tetrahedron Lett. 1995, 36,
1217–1220.
[23]J. Kihlberg, M. Elofsson, L. A. Salvador, Methods Enzymol. 1997,
289, 221–245.
[24]D. A. Wellings, E. Atherton, Methods Enzymol. 1997, 289, 44–67.
[25]S. Biancalana, D. Hudson, M. F. Songster, S. A. Thompson, Lett.
Pept. Sci. 2001, 7(5), 291–297.
[40]N. Ollivier, J. B. Behr, O. El-Mahdi, A. Blanpain, O. Melnyk, Org.
Lett. 2005, 7, 2647–2650.
[41]Y. Shin, K. A. Winans, B. J. Backes, S. B. H. Kent, J. A. Ellman,
C. R. Bertozzi, J. Am. Chem. Soc. 1999, 121, 11684–11689.
[42]S. Mezzato, M. Schaffrath, C. Unverzagt, Angew. Chem. 2005, 117,
1677–1681; Angew. Chem. Int. Ed. 2005, 44, 1650–1654.
[43]F. Mende, O. Seitz, Angew. Chem. 2007, 119, 4661–4665; Angew.
Chem. Int. Ed. 2007, 46, 4577–4580.
[44]J. Alsina, T. S. Yokum, F. Albericio, G. Barany, J. Org. Chem. 1999,
64, 8761–8769.
[45]J. Brask, F. Albericio, K. J. Jensen, Org. Lett. 2003, 5, 2951–2953.
[46]A. Ishii, H. Hojo, Y. Nakahara, Y. Ito, Y. Nakaharai, Biosci. Biotech-
nol. Biochem. 2002, 66, 225–232.
[26]R. von Eggelkraut-Gottanka, A. Klose, A. G. Beck-Sickinger, M.
Beyermann, Tetrahedron Lett. 2003, 44, 3551–3554.
[27]S. Futaki, K. Tatsuto, Y. Shiraishi, Y. Sugiura, Biopolymers 2004, 76,
98–109.
[47]J. Tulla-Puche, G. Barany, J. Org. Chem. 2004, 69, 4101–4107.
[48]L. Z. Yan, P. Edwards, D. Flora, J. P. Mayer, Tetrahedron Lett. 2004,
45, 923–925.
[28]J. D. Warren, J. S. Miller, S. J. Keding, S. J. Danishefsky, J. Am.
Chem. Soc. 2004, 126, 6576–6578.
[29]A. B. Clippingdale, C. J. Barrow, J. D. Wade, J. Pept. Sci. 2000, 6,
225–234.
[30]X. Z. Bu, G. Y. Xie, C. W. Law, Z. H. Guo, Tetrahedron Lett. 2002,
43, 2419–2422.
[31]X. Q. Li, T. Kawakami, S. Aimoto, Tetrahedron Lett. 1998, 39, 8669–
8672.
[49]P. Wang, L. P. Miranda, Int. J. Pept. Res. Therap. 2005, 11, 117–123.
[50]Aspartic and glutamic acid thioesters can also be obtained, however,
they are not synthetically useful due to the rapid formation of cyclic
anhydrides under ligation conditions: M. Villain, H. Gaertner, P.
Botti, Eur. J. Org. Chem. 2003, 3267–3272. Lysine thioesters have
been reported to form cyclic lactams: M. C. De Koning, D. V. Fili-
ppov, G. A. van der Marel, J. H. van Boom, M. Overhand, Eur. J.
Org. Chem. 2004, 850–857.
[32]M. Lumbierres, J. M. Palomo, G. Kragol, S. Roehrs, O. Muller, H.
Waldmann, Chem. Eur. J. 2005, 11, 7405–7415.
[33]D. Swinnen, D. Hilvert, Org. Lett. 2000, 2, 2439–2442.
[34]A. Sewing, D. Hilvert, Angew. Chem. 2001, 113, 3503–3505; Angew.
Chem. Int. Ed. 2001, 40, 3395–3396.
[51]F. Albericio, L. A. Carpino, Methods Enzymol. 1997, 289, 104–126.
[52]M. Villain, H. Gaertner, P. Botti, Eur. J. Org. Chem. 2003, 3267–
3272.
[53]Novabiochem Catalog 2006/2007.
[54]W. Jelkmann, Physiol. Rev. 1992, 72, 449–489.
[35]T. J. Hogenauer, Q. Wang, A. K. Sanki, A. J. Gammon, C. H. L.
Chu, C. M. Kaneshiro, Y. Kajihara, K. Michael, Org. Biomol. Chem.
2007, 5, 759–762.
[55]B. Wu, J. H. Chen, J. D. Warren, G. Chen, Z. H. Hua, S. J. Danishef-
sky, Angew. Chem. 2006, 118, 4222–4231; Angew. Chem. Int. Ed.
2006, 45, 4116–4125.
[36]J. A. Camarero, B. J. Hackel, J. J. de Yoreo, A. R. Mitchell, J. Org.
Chem. 2004, 69, 4145–4151.
[37]G. W. Kenner, J. R. McDermot Jr. , R. C. Sheppard, J. Chem. Soc. D
1971, 636–637.
[56]B. Wu, Z. P. Tan, G. Chen, J. H. Chen, Z. H. Hua, Q. Wan, K. Ran-
ganathan, S. J. Danishefsky, Tetrahedron Lett. 2006, 47, 8009–8011.
[57]J. H. Chen, G. Chen, B. Wu, Q. Wan, Z. P. Tan, Z. H. Hua, S. J. Dan-
ishefsky, Tetrahedron Lett. 2006, 47, 8013–8016.
[38]R. Ingenito, E. Bianchi, D. Fattori, A. Pessi, J. Am. Chem. Soc.
1999, 121, 11369–11374.
Received: December 15, 2007
[39]R. Quaderer, D. Hilvert, Org. Lett. 2001, 3, 3181–3184.
Published online: February 15, 2008
Chem. Eur. J. 2008, 14, 3620 – 3629
ꢀ 2008 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
3629