Peptide ligation via side-chain auxiliary

Article abstract of DOI:10.1039/b718945a

A new peptide ligation strategy based on a side-chain auxiliary was developed; the auxiliary is fairly simple and can be removed, without product isolation, under basic conditions. The Royal Society of Chemistry.

Full text of DOI:10.1039/b718945a

View Article Online / Journal Homepage / Table of Contents for this issue
COMMUNICATION
www.rsc.org/chemcomm | ChemComm
Peptide ligation via side-chain auxiliaryw z
Marina-Yamit Lutsky, Natalia Nepomniaschiy and Ashraf Brik*
Received (in Cambridge, UK) 10th December 2007, Accepted 14th January 2008
First published as an Advance Article on the web 31st January 2008
DOI: 10.1039/b718945a
A new peptide ligation strategy based on a side-chain auxiliary
was developed; the auxiliary is fairly simple and can be removed,
without product isolation, under basic conditions.
removal of the sugar was not efficient, providing the desired
peptide in a very low yield (10–15%). Moreover, using SAL in
which the sugar moiety serves as a removable auxiliary for the
synthesis of peptides rather than glycopeptides would be an
expensive and tedious method.
Intramolecular acyl transfer to form an amide bond is the
basis of several peptide ligation strategies. This step is mark-
edly accelerated by a high local concentration of the reactive
groups i.e. a proximity effect. Wieland and Brenner1 were the
first to show the importance of an entropic activation/proxi-
mity effect to peptide bond formation by intramolecular acyl
transfer. The pioneering work of the Kemp group on the
intramolecular O - N acyl transfer in peptide segment
coupling using the dibenzofuran auxiliary brought these con-
cepts into practice for the synthesis of large peptides.2 Un-
arguably, in large polypeptide synthesis, native chemical
ligation (NCL) and expressed protein ligation (EPL), employ-
ing S - N acyl transfer, have been the most effective ligation
methods.3
In principle, a removable auxiliary with similar character-
istics to the sugar moiety can be attached to one of the amino
acid side-chains in order to assist peptide ligation. Here we
report that the removable thiol auxiliary based on cyclohexane
and cyclopentane molecules can indeed mimic the sugar role in
SAL and facilitate S - N acyl transfer for peptide ligation.
Our design principle of utilizing a side-chain auxiliary to
assist peptide ligation is described in Scheme 1.
A
thiol-modified cyclohexane is attached through an ester bond
with the carboxylate side-chain of an aspartic acid (Asp) or
glutamic acid (Glu) residue.
The modified cyclohexane is meant to play two crucial roles.
First, to allow capture of the thioester peptide through the
transthioesterification step. Second, it will serve to position the
N-terminal amine and the acyl group in proximity so rapid
and chemoselective S - N acyl transfer can occur, hence
acting as a rigid template for facilitating acyl transfer. As in
SAL, the peptide is extended with an extra amino acid, next to
the residue bearing the auxiliary, to allow efficient ligation.
Finally, the auxiliary can be removed through a saponification
step to generate the unmodified peptide.
Mimicking the cysteine function in NCL/EPL with a remo-
vable auxiliary is a promising approach in assisting S - N
acyl transfer at various amino acid junctions. In this regard,
peptides with N-linked auxiliaries such as 1-phenylethanethiol
and 2-mercaptobenzyl have been investigated and successfully
applied to the synthesis of large peptides that correspond to
folded proteins.4 In these elegant examples the attachment of
the auxiliary to the N-terminal peptide generates a secondary
amine, which is engaged in the S - N acyl transfer. On the
other hand, this leads to an increased steric hindrance at the
transition state, rendering efficient ligation mainly at the
Gly–Gly junction.5
To test our strategy, we first synthesized building block 1,
starting from the racemic mixture of trans-amino cyclohexa-
nol, in order to include it close to the N-terminal peptide
(ESIw). Peptide synthesis was carried out on Fmoc-Rink amide
resin employing routine HBTU–DIEA coupling conditions
(Scheme 2A). Upon completion of coupling building block
1, the resin was split into four portions in which different
amino acids (Gly, Ala, His, Asp) were coupled to the free
N-terminal peptide to generate peptides 3a–3d as diastereo-
meric mixtures in 55–65% yield (ESIw).
Recently, we introduced a new glycopeptide ligation ap-
proach named sugar-assisted ligation (SAL).6 In SAL, instead
of anchoring an auxiliary to the N-terminal peptide, we take
advantage of the existing sugar by slightly modifying its
acetamido group on the C2 position to facilitate the ligation.6
This in turn allows the primary amine to be engaged in the acyl
transfer in contrast to previous methods where a secondary
amine is employed. Following SAL, the sugar moiety can be
cleaved enzymatically using PNGase A, affording the peptide
structure.6b However, in our previous studies, the enzymatic
To test the ligation reaction, peptides were dissolved at a
final concentration of 5 mM in 6 M guanidineꢀHCl, pH 7.5–8
(1 : 1.2 molar ratio of peptide to thioester). The ligation
reactions were performed at 37 1C and the progress of each
reaction was monitored using analytical HPLC and MALDI-
TOF. Upon completion of the ligation reaction, the pH was
increased to 10 by addition of 1 M NaOH. The reaction was
left at this pH for 5 min, at room temperature, during which a
full hydrolysis of the auxiliary was achieved (Fig. 1B). Nota-
bly, the auxiliary removal is done in situ in contrast to other
studies, including SAL, where the ligation product is isolated
first and then subjected to auxiliary removal.4
Department of Chemistry, Ben Gurion University, Beer Sheva, Israel
84105. E-mail: abrik@bgu.ac.il; Fax: (+972) 08-647-2944;
Tel: (+972) 08-6461195
w Electronic supplementary information (ESI) available: Synthesis of
auxiliaries, peptide synthesis and ligation, HPLC and mass spectro-
metric analysis of precursors and products. See DOI: 10.1039/
b718945a
z Dedicated to Professor Ehud Keinan on the occasion of his 60th
birthday.
ꢁc
This journal is The Royal Society of Chemistry 2008
Chem. Commun., 2008, 1229–1231 | 1229

View Article Online
Scheme 2 Peptides bearing side-chain auxiliaries. (A) Aspartic acid
and glutamic acid bearing the trans-cyclohexane-based auxiliary and
its incorporation into model peptides. (B) Peptides containing
cis-cyclohexane- and cyclopentane-based auxiliaries. (C) Synthesis of
peptide with the auxiliary attached to a serine side-chain.
Since the connectivity between the auxiliary and side-chain
of aspartic acid is an ester, aspartamide formation is a
competing reaction.8 Careful analysis of the ligation reaction
(Entries 2 and 9, Table 1) revealed that after 24 h reaction
time, B10% of the ligation product undergoes aspartamide
formation (peak c, Fig. 1A). However, this side product, which
generates isoaspartyl residue, increased during ester hydrolysis
to 20% of the final ligation product, as we found after careful
Scheme 1 Proposed mechanism for ligation via side-chain auxiliary.
R1, R3 are variations of amino side-chains, R2 = –CH2CH2CONH2.
The rates reported in Table 1 provided the following in-
sights when compared to SAL. (1) Similar to SAL and in
contrast to NCL,7 the ligation reaction proceeds through S -
N acyl transfer as the rate determining step (ESIw). (2) The
ligation reaction tolerates different amino acids at the N-
terminal of peptide-(auxiliary) with preference for amino acids
that bear side-chains serving as a general base in the ligation
pathway (Entries 2 and 3, Table 1) and those that are sterically
less hindered (Entry 1, Table 1). (3) The observed ligation rate
is affected by the C-terminal amino acid of the peptide
thioester (Entries 8, 9 and 10, Table 1). (4) The ligation is
chemoselective and compatible with the presence of amine and
hydoxyl groups. In this regard, similar ligation results were
obtained when the thioester peptide included Lys residues
(ESIw). (5) Despite the changes in the ligation junctions, the
desired product was observed as the major peak by HPLC
analysis. Preparative HPLC of the ligation reactions gave the
desired peptides, after auxiliary removal, in 60–70% overall
yield (Table 1).
Fig. 1 Representative analytical HPLC traces/(MALDI-TOF/MS)
of ligation reactions (Entry 2, Table 1): peak a, thioester hydrolysis
with the expected mass of 579.27 ꢂ 0.2 (1.2 equiv. of peptide thioester
was used in the ligation reaction); peak b, unreacting thioester with the
expected mass of 666.32 ꢂ 0.2; peak c, byproduct generated from
aspartamide formation with the expected mass of 1268.6 ꢂ 0.2; peak d,
ligation product with the expected mass of 1457.6 ꢂ 0.2 Da. (B)
Ligation reaction after auxiliary removal (pH 10, 5 min): peak a,
hydrolyzed thioester; peak b, ligation product with the expected mass
of 1286.4 ꢂ 0.2 Da.
ꢁc
This journal is The Royal Society of Chemistry 2008
1230 | Chem. Commun., 2008, 1229–1231

View Article Online
Table 1 Scope of peptide ligation via side-chain auxiliary
desired peptide-(auxiliary) was fully accomplished on the solid
support (Scheme 2C). Peptide 7, bearing the Gly extension,
was ligated under similar conditions to a C-terminal glycine
thioester. In this case, the ligation went to completion at a
slightly faster rate (6 h) compared to ligation with peptide 3a
(8 h) (ESIw). Auxiliary removal was accomplished under
similar conditions to give the product in 70% yield. A similar
trend was observed when the N-terminal glycine in peptide 7
was substituted with alanine and compared with peptide 3b.
Here, the reaction was completed in 16 h compared to 24 h
with peptide 3b. Although in the serine-(auxiliary) and aspar-
tic acid-(auxiliary) the reactions use a similar ring size for the
acyl transfer, the ester bond in peptide 7 is reversed compared
to the one in peptide 3, which seems to favorably affect the S
- N acyl transfer reaction. Interestingly, when mercaptoace-
tic acid was attached directly to the serine side-chain, without
the cyclohexane moiety, only 15–20% of the ligation product
was observed, despite extending the reaction time to 24 h. In
this example, the reaction proceeds through an 11-membered
ring transition state, which is similar to the ring size of the
transition state in the dibenzofuran auxiliary.2 While the ring
size is an important factor, the rigidity of the template/
auxiliary to position the nucleophilic amine to the acyl group
is crucial.9
Obsd
Entry -AA2- -AA1- -AA2-AA1- t/ha mass
Isolated yield
(%)b
1
2
3
4
5
6
7
8
9
10
a
Gly Gly
Gly Asp
Gly His
Gly Ala
Gly-Gly
Gly-Asp
Gly-His
Gly-Ala
His-Gly
His-His
His-Ala
Ala-Gly
Ala-His
Ala-Asp
B8 1228.7 ꢂ 0.2 68
B6 1286.4 ꢂ 0.2 70
B6 1308.3 ꢂ 0.2 68
B24 1242.6 ꢂ 0.2 60
B12 1308.6 ꢂ 0.2 60
B8 1388.4 ꢂ 0.2 62
B24 1322.5 ꢂ 0.2 59
B36 1242.5 ꢂ 0.2 62
B30 1322.5 ꢂ 0.2 62
B24 1300.5 ꢂ 0.2 63
His
His
His
Gly
His
Ala
Ala Gly
Ala His
Ala Asp
t represents the time at which over 80% of the peptide-(auxiliary)
was consumed with the major peak being the desired product.
b
Reported yield is for ligation and auxiliary removal steps. R =
–CH2CH2CONH2.
separation of the ligation product (ESIw). In a peptide sequence
that contains -Asp-AA-, where AA is a sterically hindered
amino acid, aspartamide formation should decrease substan-
tially and should not be observed if AA is proline. In principle,
the use of backbone protection or cleavage of the ligation
product with HF would eliminate this side reaction, regardless
of the nature of the aspartyl a-carboxy amide bond.8
In summary, we have shown that attaching a removable
auxiliary to the side-chain of an amino acid to allow for the
primary amine to be involved in the ligation reaction results in
higher flexibility at the ligation junction. Moreover, three
amino acid side-chains successfully served to anchor the
auxiliary, thereby expanding the scope of peptide ligation
beyond the cysteine residue.
In SAL, we found that regardless of the glycopeptide type
(a-O-linked, b-O-linked and N-linked glycopeptide),6 the liga-
tion rates were similar. To evaluate the effect of the substi-
tuents’ configuration on our new auxiliary, we compared the
rates of ligation for peptides bearing the trans- and cis-
cyclohexane-based auxiliaries. We also prepared a similar
peptide sequence bearing the cyclopentane-based auxiliary 6,
to examine the effect of the ring type on the ligation rate
(Scheme 2B). The three peptides with the Gly extension, 3a, 5,
6, were reacted with the C-terminal glycine thioester under
similar ligation conditions. Notably, our results showed no
differences in the ligation rates between the three types of
auxiliaries. This could be explained by the large ring that the
reaction uses to undergo S - N acyl transfer, in which the
changes in the configuration and ring type of the different
auxiliaries can be tolerated.
We are grateful to the Marc Rich Foundation and the Israel
Science Foundation for financial support. This work was also
supported by the Marie Curie International Reintegration
Grant (MIRG-CT-2007-046374).
Notes and references
1 (a) T. Wieland, E. Bokelmann, L. Bauer, H. U. Lang and H. Lau,
Justus Liebigs Ann. Chem., 1953, 583, 129–149; (b) M. Brenner, J. P.
Zimmermann, J. Wehrmuller, P. Quitt, A. Hardtmann, W. Schneider
¨
and U. Beglinger, Helv. Chim. Acta, 1957, 40, 1497–1517.
2 D. S. Kemp and R. I. Carey, J. Org. Chem., 1993, 58, 2216–2222.
3 (a) P. E. Dawson, T. W. Muir, I. Clark-Lewis and S. B. H. Kent,
Science, 1994, 266, 776–779; (b) T. W. Muir, D. Sondhi and P. A.
Cole, Proc. Natl. Acad. Sci. U. S. A., 1998, 95, 6705–6710; (c) Y.-A.
Lu and J. P. Tam, Org. Lett., 2005, 7, 5003–5006.
4 (a) D. W. Low, M. G. Hill, M. R. Carrasco, S. B. H. Kent and P.
Botti, Proc. Natl. Acad. Sci. U. S. A., 2001, 98, 6554–6559; (b) J.
Offer, C. Boddy and P. E. Dawson, J. Am. Chem. Soc., 2002, 124,
4642–4646.
5 (a) D. Macmillan and D. W. Anderson, Org. Lett., 2004, 6,
4659–4662; (b) B. Wu, J. Chen, J. D. Warren, G. Chen, Z. Hua
and S. J. Danishefsky, Angew. Chem., Int. Ed., 2006, 45,
4116–4125.
6 (a) A. Brik, Y.-Y. Yang, S. Ficht and C.-H. Wong, J. Am. Chem.
Soc., 2006, 128, 5626–5627; (b) A. Brik, S. Ficht, Y.-Y. Yang and
C.-H. Wong, J. Am. Chem. Soc., 2006, 128, 15026–15033; (c) Y.-Y.
Yang, S. Ficht, A. Brik and C.-H. Wong, J. Am. Chem. Soc., 2007,
129, 7690–7701.
The successes with the auxiliary anchored to the side-chain
of aspartic acid prompted us to explore the attachment of our
auxiliary to the glutamic acid side-chain. Similar synthetic
schemes were adopted to prepare the necessary building block
and peptide-(auxiliary). Peptide 4a bearing the glycine exten-
sion was ligated to a C-terminal glycine thioester. Interest-
ingly, the time of the reaction increased to 12 h compared to
8 h with peptide 3a. The slower rate could be explained by the
increase in the flexibility of the glutamic acid side-chain due to
the extra methylene group, which leads to a less favorable
S - N acyl transfer.
7 E. C. B. Johnson and S. B. H. Kent, J. Am. Chem. Soc., 2006, 128,
6640–6646.
8 M. Bodanszky and J. Z. Kwei, Int. J. Pept. Protein Res., 1978, 12,
69–74.
9 R. J. Payne, S. Ficht, S. Tang, A. Brik, Y.-Y. Yang, D. A. Case
and C.-H. Wong, J. Am. Chem. Soc., 2007, 129, 13527–13536.
Ligation assisted by an auxiliary which is attached to the
serine side-chain was also examined. The synthesis of the
ꢁc
This journal is The Royal Society of Chemistry 2008
Chem. Commun., 2008, 1229–1231 | 1231