Solid-phase Staudinger ligation from a novel core-shell-type resin: A tool for facile condensation of small peptide fragments

Article abstract of DOI:10.1021/ol0530629

Solid-phase Staudinger ligation of small peptides was performed on a novel core-shell-type resin. Solid-phase Staudinger ligation was mediated by synthetic solid-supported phosphinothiol, which was readily prepared by a straightforward synthetic route. This protocol afforded final peptide products in excellent yields and purities and thus could provide the opportunity to facilitate a simple manipulation for condensation of peptide fragments. In particular, the resulting resin could be recycled in a successful manner.

Full text of DOI:10.1021/ol0530629

ORGANIC
LETTERS
2006
Vol. 8, No. 6
1149-1151
Solid-Phase Staudinger Ligation from a
Novel Core-Shell-Type Resin: A Tool for
Facile Condensation of Small Peptide
Fragments
Hanyoung Kim,† Jin Ku Cho,† Saburo Aimoto,‡ and Yoon-Sik Lee*,†
School of Chemical and Biological Engineering, Seoul National UniVersity,
Kwanak-Gu, Shilim-Dong, San 56-1, Seoul 151-742, Korea, and
Institute for Protein Research, Osaka UniVersity, 3-2 Yamadaoka,
Suita, Osaka 565-0871, Japan
yslee@snu.ac.kr
Received December 18, 2005
ABSTRACT
Solid-phase Staudinger ligation of small peptides was performed on a novel core-shell-type resin. Solid-phase Staudinger ligation was mediated
by synthetic solid-supported phosphinothiol, which was readily prepared by a straightforward synthetic route. This protocol afforded final
peptide products in excellent yields and purities and thus could provide the opportunity to facilitate a simple manipulation for condensation
of peptide fragments. In particular, the resulting resin could be recycled in a successful manner.
The chemical synthesis approach to protein engineering has
attracted a great deal of attention because of its generality
and flexibility in the incorporation of noncoded moieties.1
For the total chemical synthesis of proteins, the ligation of
synthetic peptides is an essential step, and the most common
route of accomplishing this is native chemical ligation, which
is widely utilized for the preparation of peptides containing
over 50 residues.2 In a typical native chemical ligation, the
thiolate of an N-terminal cysteine residue of one peptide
attacks the C-terminal thioester of the other peptide, ulti-
mately leading to the synthesis of the final peptide through
S f N acyl migration (Figure 1a). However, a fatal drawback
of this method is the requirement that there be a cysteine
residue at the ligation junction. To overcome this limitation,
more general ligation strategies have been proposed,3 and
in this respect, the protocol based on the Staudinger reaction
(2) (a) Dawson, P. E.; Muir, T. W.; Clark-Lewis, I.; Kent, S. B. H.
Science 1994, 266, 776. (b) Tam, J. P.; Yu, Q. T.; Miao, Z. W. Biopolymers
(Pept. Sci.) 1999, 51, 311. (c) Dawson, P. E.; Kent, S. B. H. Annu. ReV.
Biochem. 2000, 69, 923. (d) Borgia, J. A.; Fields, G. B. Trends Biotechnol.
2000, 18, 243.
(3) (a) Canne, L. E.; Bark, St. J.; Kent, S. B. H. J. Am. Chem. Soc. 1996,
118, 5891. (b) Offer, J.; Dawson, P. E. Org. Lett. 2000, 2, 23. (c) Marinzia,
C.; Barkb, S. J.; Offer, J.; Dawson, P. E. Bioorg. Med. Chem. 2001, 9,
2323. (d) P. Botti, M. R. Carrasco, S. B. H. Kent, Tetrahedron Lett. 2001,
42, 1831. (e) Kawakami, T.; Akaji, K.; Aimoto, S. Org. Lett. 2001, 3, 1403.
(f) Offer, J.; Boddy, C. N. C.; Dawson, P. E. J. Am. Chem. Soc. 2002, 124,
4642. (g) Kawakami, T.; Aimoto, S. Tetrahedron Lett. 2003, 44, 6059.
† Seoul National University.
‡ Osaka University.
(1) (a) Schno¨lzer, M.; Kent, S. B. H. Science 1992, 256, 221. (b) Baca,
M.; Kent, S. B. H. Tetrahedron 2000, 56, 9503. (c) Pal, G.; Santamaria,
F.; Kossiakoff, A. A.; Lu, W. Protein Expres. Purif. 2003, 29, 185. (d)
Kent, S. B. H. Curr. Opin. Biotech. 2004, 15, 607.
10.1021/ol0530629 CCC: $33.50
© 2006 American Chemical Society
Published on Web 02/25/2006

Recently, we developed a highly cross-linked core-shell
type (HiCore) resin from aminomethyl polystyrene (AM PS)
resin.6 Its highly cross-linked rigid core and poly(ethylene
glycol) (PEG) shell provide it with chemical and physical
stability and facilitate the easy access of large sized molecules
to the functional groups of the resin, which are located on
the surface of the resin beads (Figure 2b).
Figure 1. Schematic rationale of peptide fragment condensation
via (a) native chemical ligation and (b) Staudinger ligation.
Figure 2. Cross-sectional images of the FITC-coupled resins
viewed through a confocal microscope: (a) TentaGel resin (0.30
mmol NH2/g); (b) HiCore resin (0.32 mmol NH2/g).
has particularly interesting features for the formation of
peptide bonds.4 The Staudinger ligation does not rely on the
presence of a cysteine or other specific residue at the
N-terminus of the peptide fragment. In addition, this method
is traceless, in the sense that no residual groups from the
phosphinothiol involved as a mediator remain in the peptide
product. This reaction probably proceeds via the intramo-
lecular rearrangement of an iminophosphorane intermediate
to give an amidophosphonium salt, which is in turn hydro-
lyzed to yield an amide and o-(diphenylphosphinyl)ben-
zenethiol (Figure 1b). There are two types of phosphinothiol
which can be used to mediate Staudinger ligation (1 and 2
in Figure 1). Since the isolated yields of the Staudinger
ligations using 1 are too low in some cases, 2 was developed
more recently.4c,5 Although 2 has shown an appropriate level
of performance in various ligation reactions, all of the
synthetic steps for 2 are complicated and the overall yields
are unsatisfactory. Furthermore, additional purification steps
are required in order to isolate both the thioester precursor
prior to ligation and the final peptide obtained after ligation.
These factors prompted us to exploit a readily removable
and reusable solid-supported phosphinothiol for Staudinger
ligation. Also, given that, in this case, Staudinger ligation
would be performed in the aqueous phase, water-compatible
TentaGel resin and CLEAR resin were chosen as candidate
supports. Following the coupling of 4-bromophenylacetic
acid onto the amino-functionalized resin, an excess of
magnesium was added to the reaction mixture to form resin-
bound phenylmagnesium bromide, which could act as a
Grignard reagent. However, the resins failed to endure the
harsh reaction conditions. In particular, during the insertion
of magnesium on CLEAR resin, unknown products were
released, which were probably the decomposition products
of the resin backbone.
As anticipated, when the HiCore resin coupled with
4-bromophenylacetic acid was treated with magnesium, no
severe damage was observed on the resin beads. Subse-
quently, the reaction between 3 and chloromethylphosphonic
dichloride followed by another solid-phase Grignard reaction
with phenylmagnesium bromide were carried out to afford
resin-bound phosphinoxide 4.7 After displacement with
thioacetic acid in the presence of TEA, the resulting
phosphinoxide was reduced by using trichlorosilane, and,
finally the remaining acetyl group was removed by hydrazi-
nolysis to furnish the target solid-supported phosphinothiol
5 (Scheme 1).
Scheme 1. Preparation of Solid-Supported Phosphinothiol 5a
a Key: (a) 4-bromophenylacetic acid, BOP, DIPEA, NMP; (b)
Mg, THF, reflux; (c) chloromethylphosphonic dichloride, THF,
reflux; (d) PhMgBr, THF, reflux; (e) thioacetic acid, TEA, reflux;
(f) SiHCl3, THF, reflux; (g) NH2NH2‚H2O, DMF, 60 °C.
However, the final loading level (0.063 mmol/g) of the
resin-bound phosphinothiol was substantially decreased, as
compared to the initial loading level (0.31 mmol/g) of the
amino group. This led us to search for a method of
minimizing the matrix effect of the polymer backbone in
(4) (a) Saxon, E.; Armstrong, J. I.; Bertozzi, C. R. Org. Lett. 2000, 2,
2141. (b) Nilsson, B. L.; Kiessling, L. L.; Raines, R. T. Org. Lett. 2000, 2,
1939. (c) Nilsson, B. L.; Kiessling, L. L.; Raines, R. T. Org. Lett. 2001, 3,
9.
(6) Kim, H.; Cho, J. K.; Chung, W. J.; Lee, Y. S. Org. Lett. 2004, 6,
3273.
(5) Janssen, M. J. In The Chemistry of Carboxylic Acids and Esters;
Patai, S., Eds.; Interscience Publishers: New York, 1969; p 730.
(7) To prevent undesired cross-linking, excess of chloromethylphosphonic
dichloride (5 equiv) was used.
1150
Org. Lett., Vol. 8, No. 6, 2006

the magnesium insertion step (Scheme 1b)), which occurs
between the two solid phases, i.e., magnesium and the resin.
Therefore, spacers with four different lengths were introduced
onto the HiCore resin by using ꢀ-aminocaproic acid (E) and/
or â-alanine (B), and the supported phosphinothiols were
prepared on each spacer-attached resin. We found that the
longest BEBE spacer exhibited the best result (the stepwise
coupling yield for BEBE spacer was 92%, overall yield 65%;
see Table 1).8
in quantitative yields. Regardless of the properties of the
azido acids, the purities of the ligation products were
excellent, thus dispensing for the need for any chromatog-
raphy step (Figure 3a). Another noticeable advantage of the
Table 1. Spacer Effect on Final Loading Level of Thiol
Groups
spacer
no spacer
0.063
E
BE
EBE
0.14
BEBE
0.17
mmol of SH/ga
0.091
0.12
a The loading level of the thiol groups on resin was determined by trityl
titration and confirmed by the Fmoc photometric test after the coupling of
Fmoc amino acid.
Figure 3. (a) HPLC traces and MALDI-TOF MS of representative
crude products obtained by solid-phase Staudinger ligation: Fmoc-
Phe-Phe-Ala-OH (red, m/z 628.48 [M + Na]+, 644.43 [M + K]+,
668.39 [M - H + Na + K]+); Fmoc-Phe-Phe-Met-OH (blue, m/z
688.33 [M + Na]+, 704.24 [M + K]+); (b) HPLC profiles of the
crude product (Fmoc-Phe-Phe-Ile-OH) obtained using regenerated
resin-bound phosphinothiol: first reuse (red), second reuse (green),
fifth reuse (blue).
As the other component for the Staudinger ligation, azido
acids were synthesized using the diazo transfer reaction with
triflyl azide.9 Triflyl azide was prepared by the biphasic
reaction of sodium azide and triflyl anhydride and then
combined with 15 different amino acids, viz. Gly, Ala, Leu,
Ile, Val, Phe, Asn, Gln, Asp(tBu), Ser(tBu), Thr(tBu), Met,
Tyr(tBu), Lys(tBoc), and His, respectively.10 Solid-phase
Staudinger ligation was performed with 5. Initially, the
dipeptide (Fmoc-Phe-Phe-OH) was assembled by solid-phase
peptide synthesis on 5 (Fmoc-strategy). Then, solid-phase
Staudinger ligations were performed between 5 and the
synthetic azido acids (Scheme 2). To minimize the contami-
solid-phase Staudinger ligation is the reuse of the resin-bound
phosphinothiol. The resulting resin-bound phosphine oxide
7 can be readily recovered from the ligation product by sim-
ple filtration and reduced with an excess of trichlorosilane.
The loading levels of the thiol group of the regenerated resin
5 were equal to the original loading level. Resin 5 was re-
cycled five times and showed consistent performance in terms
of the ligation yield and the purity of the product (Figure 3b).
In conclusion, we successfully performed the solid-phase
Staudinger ligation on HiCore resin, which was recycled five
times without any change in performance. Due to its physical
and chemical stability, the backbone structure of the HiCore
resin remained unchanged under harsh reaction conditions.
Moreover, the surface localized functionalities allowed for
the Staudinger ligation to be easily accomplished in the solid
phase.
Scheme 2. Strategy for Solid-Phase Staudinger Ligation Using
Solid-Supported Phosphinothiol and Recycling of the Resin
Acknowledgment. This work was supported by a grant
of the Korea Health 21 R&D Project, Ministry of Health &
Welfare, Republic of Korea (A050432).
Supporting Information Available: Experimental pro-
cedures for the preparation of triflyl azide, azido acids, solid
supported phophinothiol, solid-phase Staudinger ligation of
small peptides and their analytical data. This material is
available free of charge via the Internet at http://pubs.acs.org.
OL0530629
(8) The rather low overall yield (65%) might be due to the proton
exchange between the resin-bound Grignard reagent and the relatively acidic
protons in the amide bonds of the spacer, which caused the phenylmag-
nesium bromide to decompose to phenyl group. However, this side reaction
did not influence the subsequent reaction steps.
(9) For detailed experimental conditions for preparation of 15 azido acids
and their characterization data, see the Supporting Information.
(10) Fmoc-Phe-Phe-OH was selected considering the sensitive detection
on TLC and HPLC.
nation of the released peptides, <1 equiv of azido acid was
added to 6, and the reaction mixture was stirred until azido
acid was no longer observed on TLC.
All of the reactions were completed within 24-40 h in
THF/H2O (3:1) at rt, and the final tripeptides were obtained
Org. Lett., Vol. 8, No. 6, 2006
1151