Solid-phase synthesis of peptide thioacids through hydrothiolysis of resin-bound peptide thioesters

Article abstract of DOI:10.1016/j.tetlet.2008.08.018

We report that solid-phase hydrothiolysis is an efficient method to convert resin-bound peptide thioesters to thioacids in aqueous buffer by using a total PEG-based resin. Also demonstrated is the use of the so-prepared peptide thioacids in chemoselective amide bond formation reactions.

Full text of DOI:10.1016/j.tetlet.2008.08.018

Tetrahedron Letters 49 (2008) 6122–6125
Contents lists available at ScienceDirect
Tetrahedron Letters
journal homepage: www.elsevier.com/locate/tetlet
Solid-phase synthesis of peptide thioacids through hydrothiolysis
of resin-bound peptide thioesters
*
Xiaohong Zhang, Xiao-Wei Lu, Chuan-Fa Liu
Division of Chemical Biology and Biotechnology, School of Biological Sciences, Nanyang Technological University, 60 Nanyang Drive, Singapore 637551, Singapore
a r t i c l e i n f o
a b s t r a c t
Article history:
We report that solid-phase hydrothiolysis is an efficient method to convert resin-bound peptide thio-
esters to thioacids in aqueous buffer by using a total PEG-based resin. Also demonstrated is the use of
the so-prepared peptide thioacids in chemoselective amide bond formation reactions.
Ó 2008 Elsevier Ltd. All rights reserved.
Received 26 June 2008
Revised 29 July 2008
Accepted 5 August 2008
Available online 8 August 2008
The advent of solid-phase synthesis marked a paradigm shift in
the practice of synthetic chemistry.1 Apt for the preparation of
oligomeric compounds such as peptides and oligonucleotides,2,3
the solid-phase synthesis methodology is also widely used for
the synthesis of small molecules,4 and has been the main platform
technology for combinatorial chemistry.5–7 The versatility of this
methodology makes almost any chemical reaction adaptable to
solid-phase settings, and the critical task often lies with whether
or not one can find a suitable solid support and/or a linker to meet
a particular synthetic need. Previously, we reported a simple
hydrothiolysis reaction for the synthesis of peptide thioacids8—an
important class of compounds with increasingly recognized syn-
thetic value.9–11 When conducted in aqueous buffer at alkaline
pH, hydrothiolysis of a peptide thioester by hydrosulfide ions gives
rise to the corresponding peptide thioacid in near quantitative
yield.8 We reasoned that, provided with a solid support that has
good swelling properties in aqueous media, it should also be pos-
sible to conduct solid-phase hydrothiolysis of resin-bound peptide
thioesters.
CM
HS(CH2)2CO-NH-CH2
SPPS
O
CM
Fully protected peptide
C
S(CH2)2CO-NH-CH2
1) Side-chain
deprotection
O
CM
peptide
peptide
C
S(CH2)2CO-NH-CH2
2) hydrothiolytic
HS, H2O
cleavage
O
CM
C SH + HS(CH2)2CO-NH-CH2
Scheme 1. Synthesis of peptide thioacids by solid-phase hydrothiolysis.
CM = ChemMatrix resin.
The traditional approach to preparing peptide thioacids is based
on the benzhydryl linker chemistry originally developed by Blake
et al.12–14 This approach requires the use of Boc SPPS, and the yield
is usually low, because the supernucleophilicity and low pKa value
of the thioacid functionality make it highly susceptible to nucleo-
philic reactions with cationic electrophiles released from the vari-
ous side-chain protecting groups during the final HF deprotection
and cleavage stage. Such a problem would not exist with a solid-
phase hydrothiolysis method, because, through the use of a non
acid-labile linker, it is possible to conduct the final deprotection-
cleavage in two steps: (1) the acid-mediated deprotection to
remove all the side-chain protecting groups and, (2) hydrothiolytic
cleavage of the resin-bound peptide thioester (Scheme 1) to give
the desired peptide thioacid.
Since the hydrothiolysis reaction needs to be conducted in
aqueous media, the solid support must have good swelling ability
in water. Our first attempt to use a poly(ethylene glycol)-grafted
polystyrene (PEG-PS) resin, the TentaGel resin, gave very poor
results, with the thiolytic cleavage yield being about 30% at best.8
This unsatisfactory result is apparently due to insufficient water-
swelling of the resin in which the polystyrene core matrix is highly
hydrophobic. We therefore turned our attention to a recently
developed, totally PEG-based ChemMatrix resin (CM resin).15 CM
resin has a high degree of cross-linking, and exhibits good loading
capacity and mechanical stability.16,17 Most importantly, it is well-
solvated in a broad range of nonpolar and polar solvents, including
water.16,17 CM resin has a swelling volume in water of ꢀ11 mL/g.17
This makes it an excellent candidate resin for solid phase-
supported reactions in aqueous media. Herein, we report on the
use of CM resin for the synthesis of peptide thioacids through
* Corresponding author. Tel.: +65 6316 2867; fax: +65 6791 3856.
E-mail address: cfliu@ntu.edu.sg (C.-F. Liu).
0040-4039/$ - see front matter Ó 2008 Elsevier Ltd. All rights reserved.
doi:10.1016/j.tetlet.2008.08.018

X. Zhang et al. / Tetrahedron Letters 49 (2008) 6122–6125
6123
hydrothiolysis of resin-bound peptide thioesters. We also demon-
strate the synthetic utility of the prepared peptide thioacids in spe-
cific amidation with azide compounds and in thioacid capture
ligation.
First, thiol-functionalized CM resin was prepared by coupling
Trt-SCH2CH2COOH onto commercial aminomethyl CM resin fol-
lowed by detritylation with TFA. Peptides were then assembled
on the derived resin using typical SPPS protocols. Although Boc
chemistry is the preferred method for synthesizing peptide thioes-
ters, Fmoc chemistry can also be used if one employs Aimoto’s
Fmoc deprotection method to minimize aminolytic cleavage of
the thioester linkage.18 Thus, peptide 1 (Table 1) was synthesized
by Fmoc chemistry using the Fmoc-deprotection mixture of
1-methylpyrrolidine (25%), hexamethyleneimine (2%), HOBt (2%)
in NMP/DMSO (1:1). Peptides 2–6 were synthesized by using Boc
chemistry. For peptides 4–6, the side-chain protecting group for
Arg was Mts. Final deprotection was carried out using standard
TFA or TFMSA/TFA protocols for Fmoc- or Boc-chemistry synthe-
sized peptides, respectively, to remove all the side-chain protecting
groups without cleaving the peptide from the resin. The side-chain
deprotected peptide-resin was washed with TFA and acetonitrile,
and used immediately for the cleavage with hydrosulfide ions.
Reaction conditions for the solid-phase hydrothiolysis reaction
on the resin-bound peptide thioesters were similar to those used
in solution reactions reported earlier.8 Either ammonium sulfide
or sodium sulfide was used as the source of hydrosulfide ions.
Good hydrothiolytic cleavage yields were achieved for all the CM
resin-bound peptide thioesters (Table 1). As seen from Table 1,
these peptides contain a diverse set of C-terminal residues, from
the smallest Gly to the hindered Ile. For instance, thioacid peptide
2, which has an Ile residue at the C-terminus, was obtained in
excellent yield after a prolonged cleavage reaction time. For most
other peptides, the reaction reached its maximum yield after about
3 h, as seen in a simple kinetic study on peptide 4 (Fig. 1).
The cleaved crude products were, in general, of good purity as
shown by analytical HPLC analysis. Figure 2 shows representative
HPLC profiles of two crude thioacid peptides (Fig. 2A for peptide
4 and Fig. 2B for peptide 3). It should be noted that peptide 3
was synthesized on recycled mercaptopropionylaminomethyl CM
resin as this thiol-derived CM resin is regenerated after thiolytic
cleavage and therefore can be reused. Before reuse, the recycled
resin was treated with 5% hydrazine in DMF (2.5% mercapto-
ethanol) for 30 min to ensure complete removal of any uncleaved
peptide. The purity of the crude peptide depends solely on the effi-
Figure 1. Cleavage yield of thioacid peptide 4 as a function of time. See Table 1
footnote for cleavage conditions.
Figure 2. HPLC profiles of crude thioacid peptides 4 and 3 cleaved from CM resin.
(A) Thioacid peptide 4 after 5 h cleavage. (B) Thioacid peptide 3 after 3 h cleavage.
Peptide 3 was synthesized on recycled CM resin. HPLC linear gradient: 0–60% of
buffer B (90% CH3CN in H2O containing 0.05% TFA) in buffer A (H2O containing 0.05%
TFA) in 30 min on an analytical C18 column (Jupiter, 250 Â 4.6 mm).
ciency of SPPS. No S-alkylation side products were observed from
HPLC and MS analysis, presumably because all the reactive species
that would be derived from the side-chain protecting groups had
already been removed and washed away in the previous step. It
is worth noting that the solid-phase hydrothiolysis reaction can
be conducted under denatured conditions, which are highly desir-
able for hydrophobic peptides that are prone to aggregation. In this
study, thioacid peptides 2, 3, and 4 were prepared in the presence
of 6 M guanidine–HCl.
Table 1
Synthesis of peptide thioacids through hydrothiolysis of CM resin-bound peptide
thioesters19
It is recommended that, once the peptide resins are fully depro-
tected, they should be cleaved with the thiol reagent as soon as
possible, as we have noticed a decrease in the cleavage yield if
the peptide resin is stored over time. This is likely due to interchain
interactions and aggregation of the resin-bound peptide. In such
cases, the deprotected peptide resin must be soaked in TFA for
15–20 min to re-swell the resin thoroughly and break up any inter-
winding peptide chains and then washed with acetonitrile and
DMSO before thiolytic cleavage.
Peptide
Sequence
Cleavage yielda (%)
MWb calcd
m/zb found
1
2
3
4
5
6
ARTKQTARKSTGc
GIGDPVTCLKSGAId
VGLFEDTNLd,f
RLLLPGELAd
68 (2 h)
85 (8 h)
70 (3 h)
67 (5 h)
80 (3 h)
80 (3 h)
1319.7
1345.7
1022.5
996.6
1125.6
678.3
1320.8
1347.4
1023.4
997.8
1127
APKRYKANYe
DSARAGSc
679.6
All resin-bound peptide thioesters are of the general structure: peptide-CO-
SCH2CH2CO-NHCH2-CM resin.
a
Cleavage yield was calculated using the quantitative ninhydrin test (absorbance
The thioacid functionality is a very soft and powerful nucleo-
phile, which makes it useful in organic synthesis. For instance,
the reaction between a thioacid and an azide has received renewed
attention recently, because it can be used for facile preparation of
complex amides. A thorough mechanistic study has suggested that
the reaction proceeds through a prior capture step leading to the
formation of a thiatriazoline intermediate, which decomposes to
give the amide product.20 Notably, the reaction was reported to
at 570 nm) on peptide-resin before and after cleavage; values in brackets are the
cleavage time.
b
Calculated isotopic molecular weight and found ([M+H]+) m/z value of the
peptide thioacid products.
c
Cleavage conditions: 0.2 M (NH4)2S, 0.3 M HEPES buffer, pH 8.6, rt.
Cleavage conditions: 0.2 M (NH4)2S, 0.3 M HEPES buffer, 6 M guanidine–HCl, pH
d
8.6, rt.
e
Cleavage conditions: 0.1 M Na2S, 0.3 M HEPES buffer, pH 7.6.
Synthesized on recycled HSCH2CH2CO-NHCH2-CM resin.
f

6124
X. Zhang et al. / Tetrahedron Letters 49 (2008) 6122–6125
be especially efficient with electron-deficient azides such as sulfo-
nazides.10 Although so far thioacid/azide coupling has only been
demonstrated with small thioacids such as thioacetic acid and N-
protected amino thioacid compounds,10,11 its unique reaction fea-
tures imply that it should have the required chemical orthogonal-
ity to work in a chemoselective manner with large and complex
thioacid compounds containing other functional groups, such as
the peptide thioacids listed in Table 1. Therefore, to demonstrate
the synthetic utility of the prepared peptide thioacids, peptide 4,
H-ArgLeuLeuLeuProGlyGluLeuAla-COSH, was subjected to amida-
tion with tosyl azide in wet methanol in the presence of 2,6-
lutidine.21 HPLC analysis showed a clean and near quantitative
conversion (>95%) of the thioacid peptide to the desired N-peptidyl
sulfonamide product after 2 h reaction at 23 °C (Fig. 3). Similarly,
reaction of peptide 5, H-AlaProLysArgTyrLysAlaAsnTyr-COSH, with
tosyl azide in wet methanol was complete in 2 h to give the
expected N-acyl-sulfonamide (m/z [M+H]+ found: 1263.5, MW
calcd: 1262.6) in excellent yield (data not shown). It should be
noted that the reaction was conducted in a very dilute solution
(ca. 0.2 mM) of the thioacid peptide. Clearly, the high reaction
efficiency in such a dilute solution can only be explained by a
mechanism of the prior capture type proposed earlier by Williams
and co-workers20 The reaction appeared highly chemoselective
despite the presence of a number of side-chain functional groups
and a free N-terminal amine in the two peptides. These results
further validate the utility of the thioacid/azide amidation reaction
as a new method to introduce a C-terminal peptide modification
with, for example, a biophysical probe.
Peptide thioacids are also the key building blocks for ‘mini’ thiol
capture ligation (or thioacid capture ligation).9 The key element of
this method consists of specific capture of a C-ter thioacid of the
first peptide by an activated disulfide from an Npys-modified
N-ter Cys side chain of the second peptide to form an acyl disulfide
intermediate, which undergoes rapid intramolecular acylation to
generate an amide bond.9 The final product with a native Cys
residue at the ligation site is obtained after a simple thiolytic
reduction. Figure 4 shows the thioacid capture ligation reaction
between thioacid peptide 4 and an Npys-modified cysteinyl pep-
tide (in ꢀ2-fold molar excess), which gave a very good yield of
the 35-residue ligation product after 10 min of reaction.22 Of all
the chemoselective ligation methods developed thus far, the
thioacid-capture ligation technique is probably the most efficient
in terms of reaction rate. The main drawback of this method is
Figure 4. HPLC monitoring of ligation between H-RLLLPGELA-COSH and Npys-CA-
26. Trace 1, peak a is Npys-CA-26 or H-C(Npys)AIHAK(ac)RVTIMPKDIQLARRIR-
GERA-COOH (m/z [M+3H]3+ found: 1067.8, MW calcd: 3197.7). Trace 2, peak b is H-
RLLLPGELA-COSH. Trace 3, TCEP-reduced reaction mixture after 20 min ligation.
Peak c is reduced CA-26 (m/z [M+2H]2+ found: 1523.6, MW calcd: 3043.7). Peak d is
the hydrolysis product of the peptide thioacid (m/z [M+H]+ found: 981.8, MW calcd:
980.6). Peak e is the ligation product, H-RLLLPGELACAIHAK(ac)RVTIMPK-DIQL-
ARRIRGERA-COOH (m/z [M+3H]3+ found: 1337.2, MW calcd: 4006.3). HPLC linear
gradient: 0–40% buffer B (90% CH3CN in H2O containing 0.05% TFA) in buffer A (H2O
containing 0.05% TFA) for 40 min.
associated with the difficulty in obtaining the thioacid building
blocks. The development of the solid-phase hydrothiolysis tech-
nique for peptide thioacid synthesis will therefore make the thio-
acid capture ligation a more useful method for protein synthesis.
We have demonstrated herein that solid phase-supported
hydrothiolysis is an efficient method to convert a CM resin-bound
peptide thioester to its corresponding thioacid in aqueous media.
The simplicity of this new method should make peptide thioacids
easily available for use in organic synthesis by using, for example,
thioacid/azide coupling and thioacid capture ligation. It should also
further stimulate the development of new synthetic methods
based on the use of peptide thioacids.
Acknowledgments
The authors thank the Ministry of Education (MoE) of Singapore
and Singapore Heart Foundation for financial support as well as
Nanyang Technological University.
References and notes
1. Merrifield, R. B. J. Am. Chem. Soc. 1963, 85, 2149–2154.
2. Marshall, G. R. J. Pept. Sci. 2003, 9, 534–544.
3. Caruthers, M. H. Science 1985, 230, 281–285.
4. Früchtel, J. S.; Jung, G. Angew. Chem., Int. Ed. 1996, 35, 17–42.
5. Lam, K. S.; Salmon, S. E.; Hersh, E. M.; Hruby, V. J.; Kazmierski, W. M.; Knapp, R.
J. Nature 1991, 353, 82–84.
6. Houghten, R. A. Proc. Natl. Acad. Sci. U.S.A. 1985, 82, 5131–5135.
7. Thompson, L. A.; Ellman, J. A. Chem. Rev. 1996, 96, 555–600.
8. Tan, X.-H.; Zhang, X.; Yang, R.; Liu, C.-F. ChemBioChem 2008, 9, 1052–1056.
9. Liu, C. F.; Rao, C.; Tam, J. P. Tetrahedron Lett. 1996, 37, 933–936.
10. Shangguan, N.; Katukojvala, S.; Greenberg, R.; Williams, L. J. J. Am. Chem. Soc.
2003, 125, 7754–7755.
11. Merkx, R.; Brouwer, A. J.; Rijkers, D. T. S.; Liskamp, R. M. J. Org. Lett. 2005, 7,
1125–1128.
12. Blake, J. Int. J. Pept. Protein Res. 1981, 17, 273–274.
13. Gaertner, H.; Villain, M.; Botti, P.; Canne, L. Tetrahedron Lett. 2004, 45, 2239–
2241.
14. Goldstein, A. S.; Gelb, M. H. Tetrahedron Lett. 2000, 41, 2797–2800.
15. Cote, S. PCT Int. Appl. 2005, WO 2005012277 AI 20050210 CAN 142: 198979
AN 2005: 120909.
16. Camperi, S. A.; Marani, M. M.; Iannucci, N. B.; Cote, S.; Albericio, F.; Cascone, O.
Tetrahedron Lett. 2005, 46, 1561–1564.
17. Garcia-Martin, F.; Quintanar-Audelo, M.; Garcia-Ramos, Y.; Cruz, L. J.; Gravel,
C.; Furic, R.; Cote, S.; Tulla-Puche, J.; Albericio, F. J. Comb. Chem. 2006, 8, 213–
220.
Figure 3. HPLC monitoring of the reaction of peptide thioacid 4 (H-RLLLPGELA-
COSH) with tosyl azide. Trace 1—the thioacid peptide (peak a; m/z [M+H]+ found:
997.7; MW calcd: 996.6). Trace 2—reaction mixture at 1 h. Peak b: reaction product
H-RLLLPGELA-CONH-SO2-Ph-CH3 (m/z [M+H]+ found: 1135.8, MW calcd:1133.6).
Peak c: p-MePh-SO2-N3 in excess. Trace 3—reaction mixture at 2 h. HPLC gradient:
0–80% buffer
B (90% CH3CN in H2O containing 0.05% TFA) in buffer A (H2O
containing 0.05% TFA) for 40 min.

X. Zhang et al. / Tetrahedron Letters 49 (2008) 6122–6125
6125
18. Li, X.; Kawakami, T.; Aimoto, S. Tetrahedron Lett. 1998, 39, 8669–8672.
19. For a typical reaction, 25–50 mg of peptide-resin was suspended in 1 mL of
hydrothiolysis buffer in an Eppendorf vial with gentle shaking. The reaction
was stopped by acidification with 20% TFA/H2O with ice cooling. The mixture
was subjected to HPLC purification to give the purified thioacid peptide. All
operations should be performed in a well-ventilated fume hood.
20. Kolakowski, R. V.; Shangguan, N.; Sauers, R. R.; Williams, L. J. J. Am. Chem. Soc.
2006, 128, 5695–5702.
21. HPLC purified peptide thioacid (ca. 0.25 mM) was mixed with tosyl azide (ca.
2 mM) and 2,6-lutidine (ca. 2 mM) in wet methanol, and the reaction mixture
was monitored by analytical HPLC.
22. Thioacid peptide 4 (ca. 0.5 mM) was mixed with Npys-modified CA-26 (ca.
1 mM) in CH3CN/H2O (0.05% TFA, pH 2). A yellow color developed immediately.
The pH of the reaction solution was then adjusted to about 6 with 0.2 M
phosphate buffer. TCEP was added after 10 min for thiolytic reduction, and the
reaction mixture was subjected to HPLC purification.