Native chemical ligation at valine

Article abstract of DOI:10.1002/anie.200801590

(Chemical Equation Presented) Peptide ligation at hydrophobic sites is possible with a ligation-desulfurization strategy in which penicillamine serves as a precursor of valine. The β,β-dimethylcysteine peptides reacted surprisingly fast in native chemical ligation reactions. Even the sterically crowded and unpolar Leu-Val bond can be formed in high yield.

Full text of DOI:10.1002/anie.200801590

Angewandte
Chemie
DOI: 10.1002/anie.200801590
Peptide Ligation without Cysteine
Native Chemical Ligation at Valine
Christian Haase, Heike Rohde, and Oliver Seitz*
Among the techniques employed for coupling peptide seg-
ments,[1] native chemical ligation is among the most useful.[2]
Since its discovery it has become a powerful tool for the
chemical synthesis of proteins including labeled or posttrans-
lationally modified proteins and proteins that contain non-
proteinogenic amino acids.[3] In the native chemical ligation
an unprotected C-terminal peptide thioester reacts with an
unprotected N-terminal cysteine residue. The requirement for
the rare amino acid cysteine limits the applicability in the
synthesis of naturally occurring proteins. Several approaches
have been developed to allow access to other, more common
ligation sites.[4]
In the extended native chemical ligation the cysteine
structure is mimicked by means of a removable auxiliary
group that is attached to the N terminus of the peptide
fragment.[5] Typically, electron-rich aromatic ring systems are
included in the auxiliary structure to facilitate acidolytic
removal subsequent to the ligation.[5e–k] The formation of
glycine–glycine peptide bonds usually proceeds without
problems. However, the reactivity of the secondary amine
rapidly decreases with increasing steric demand at the ligation
site. Thus, ligation at bulky amino acids such as valine or
isoleucine has not been achieved. Recently, sugar-assisted
ligation has been reported, which is of particular interest for
glycopeptide synthesis.[6]
A conceptually different approach avoids the usage of less
reactive secondary amines and draws upon the coupling of
amino acids that contain sulfanyl groups in the side chain
which are removed after ligation. For example, cysteine was
used in a conventional native chemical ligation followed by
conversion to the abundant alanine by using an excess of
metal reagents.[7] The undesired desulfurization of other
cysteine residues can be avoided by using protecting
groups.[7b] Recently, the repertoire of this ligation–desulfur-
ization approach was extended to phenylalanine.[8] The
required b-sulfanylphenylalanine was prepared in a multistep
synthesis.
Scheme 1. Native chemical ligation at valine. The b,b-dimethylcysteine
at the N terminus of the penicillyl peptide II induces the thiol exchange
with peptide I (step a) to form thioester intermediate III. Subsequent
S!N acyl transfer furnishes the peptide bond in IV (step b). Desulfur-
ization (step c) provides the valyl peptide V.
surprisingly fast despite the steric shielding of the methyl
groups in the vicinity of the sulfanyl function. Even Leu–Val
ligation sites, which appear in hydrophobic peptide segments,
are accessible. We also present an improved method for
achieving metal-free desulfurization and show applications in
the synthesis of valine-containing peptides.
Initially, we anticipated that the steric demand of the
methyl groups in penicillamine would present a challenge in
ligation-like reactions. Hence we first scrutinized the kinetics
of reactions between model peptides that have been studied
previously.[9] The penicillaminyl model peptide Pen-Arg-Ala-
Glu-Tyr-Ser-NH2 (1, Scheme 2) was prepared by Fmoc-based
In this communication we describe the use of penicill-
amine (Pen) as a precursor of valine in the ligation–
desulfurization strategy (Scheme 1). b,b-Dimethylcysteine is
commercially available with various protecting-group pat-
terns suitable for routine solid-phase synthesis of peptides.
We found that the ligation at penicillamine proceeded
Scheme 2. Peptides 1–10 used in the study of the penicillamine
ligation.
solid-phase synthesis, which included the coupling of a Boc/
Trt-protected penicillamine building block (see the Support-
ing Information). The first experiments explored ligations at
glycine as a sterically less demanding C-terminal amino acid
in the peptide thioester Leu-Tyr-Lys-Ala-Gly-SR (2). The
ligations were performed at a 5 mm concentration of the
peptides in aqueous sodium phosphate buffer which con-
tained 6m guanidinium hydrochloride (GnHCl) and 50 mm
triscarboxyethylphosphine (TCEP) as the denaturing and
reducing agent, respectively. Sodium sulfanylethanesulfonate,
[*] C. Haase, H. Rohde, Prof. Dr. O. Seitz
Humboldt-Universität zu Berlin
Institut fürChemie
Brook-Taylor-Strasse 2, 12489 Berlin (Germany)
Fax : (+49)30-2093-7266
E-mail: oliver.seitz@chemie.hu-berlin.de
Supporting information for this article is available on the WWW
underhttp://dx.doi.org/10.1002/anie.200801590.
Angew. Chem. Int. Ed. 2008, 47, 6807 –6810
ꢀ 2008 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
6807

Communications
benzylthiol, and/or thiophenol were added to induce the
furnishing 60% ligation product already after 1 h and 87%
ligation yield after 12 h. To exclude the possibility of a direct
reaction between the thioester moiety and the a-amino
function of the penicillaminyl peptide 1, the corresponding
valyl peptide 10 was synthesized and incubated with the
peptide thioester 2 under the optimized ligation conditions
(see Figure S2 in the Supporting Information). The formation
of a ligation product was not observed, highlighting the
importance of the thiol group in penicillamine.
formation of reactive thioesters in situ. The course of the
ligation reaction was monitored by HPLC-MS analysis, which
provided an accurate means to detect all the reaction products
(Figure 1). The fastest initial ligation rates were achieved
Encouraged by these results we proceeded to explore
ligations with sterically more demanding peptide thioesters.
The peptidylhistidine thioester in Leu-Tyr-Lys-Ala-His-SR
(3) underwent facile reaction (Figure 2) and provided the
ligation product 12 in 70% yield (Table 1). However, small
amounts (max. 8%) of an epimerized ligation product also
formed (see Figure S3in the Supporting Information). This
behavior of histidine thioesters, high reactivity in both native
chemical ligation and racemization reactions, was described
previously.[9] The segment coupling of the methionine thio-
ester in peptide 4 and pencillaminylpeptide 1 proceeded
smoothly, yielding 65% ligation product after 24 h. Careful
HPLC analysis revealed that racemization occurred to a small
extent (less than 4%), which could be further reduced to 2%
when the reaction was performed at pH 7.5 rather than pH 8.5
(see Table S1 in the Supporting Information). We then
investigated the sterically demanding coupling of the leucine
thioester in 5 with 1. Surprisingly, the initial ligation rate was
only six times less than the ligation rate of 1 with the glycine
thioester in 2. At extended reaction times (48 h) the ligation
product 14 formed in 70% yield. This could be further
improved to 80% yield when the leucine thioester was used in
twofold excess. Racemization was not observed.
The versatility of the penicillamine-mediated ligation was
demonstrated in the synthesis of the two longer peptide
sequences 15 and 16 from sections of the signal transduction
proteins STAT-1 and Syk kinase, respectively. The synthesis of
the 176–197 segment 15 of STAT-1 was performed as
described for the model peptides 11–14. HPLC analysis
showed that the reaction of peptide thioester 6 with penicillyl
peptide 7, added in 1.4-fold excess, resulted in near-quanti-
tative formation of the ligation product 15 (Figure S6 in the
Supporting Information). The coupling of peptide thioester 8
with penicillaminyl peptide 9 provided the 22-mer Syk kinase
Figure 1. Scope of the penicillamine-mediated ligation. A) HPLC trace
analysis of a representative ligation of 1 with 2 aftera reaction time of
2 h. (*: hydrolyzed thioester). B) Time course of product formation in
ligations of 1 with peptide thioesters 2, 3, 4, or 5. Ligation conditions:
5 mm 1, 5 mm thioester 2, 3, 4 or 5, 6m GnHCl, 100 mm NaH2PO4,
50 mm TCEP, 5% PhSH, pH 8.5, 378C. After3, 6, 9, and 12 h further
aliquots of buffersolution (8 vol% of the initial reaction volume)
containing 0.5m TCEP, 6m GnHCl, 200 mm NaH2PO4, and PhSH were
added.
when the aqueous phosphate buffer contained 5% thiophenol
as sole thiol additive. HPLC analysis indicated the formation
of the symmetric disulfide and the disappearance of thiophe-
nol during the course of the reaction. Hence, further aliquots
of TCEP and thiophenol were added after 3, 6, 9, and 12 h.
Under these conditions, the ligation was remarkably fast,
Table 1: Yields of the ligations and the desulfurization.
Penicillyl
peptide
Peptide
thioester
Ligation
Desulfurization product
Desulfurization yield [%]
Product/yield [%]/reaction time [h]
metal-based
metal-free
1
1
1
1
7
2
3
4
5
6
8
2
LYKAGPenRAEYS 11/87/12
LYKAHPenRAEYS 12/70/24
LYKAMPenRAEYS 13/65/24
LYKAGVRAEYS 17
LYKAHVRAEYS 18
LYKAMVRAEYS 19
LYKALVRAEYS 20
LKKPFNRPQGVQPKTGPFEDLK 21
TLQNREHETNGVAKSDQKQEQL 22
–
61
–
–
98
93
77
79
72
91
–
LYKALPenRAEYS 14/70[a] and 82[b]/48
TLQNREHETNGPenAKSDQKQEQL 15/78[c]/24
LKKPFNRPQGPenQPKTGPFEDLK 16/87[c]/24
LYKAGVRAEYS 17/0[e]
–
0[d]
54
–
9
10
[a] After48 h with 1 equiv 5. [b] after48 h with 2 equiv 5. [c] Penicillyl peptides were used in 1.3- to 1.4-fold excess. [d] After a reaction time of 8 h neither
starting material nor product could be isolated. [e] Direct fragment coupling was not observed. The one letter codes for the amino acids are given in
http://www.chem.qmul.ac.uk/iupac/AminoAcid/AA1n2.html
6808 www.angewandte.org
ꢀ 2008 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Angew. Chem. Int. Ed. 2008, 47, 6807 –6810

Angewandte
Chemie
peptide 16 in nearly quantitative yield, when 9 was added in
1.3fold excess.
The next goal was to convert the penicillyl residues in the
formed ligation products to valine residues. We first
attempted sulfur removal by applying known metal-based
methods of desulfurization. The penicillyl ligation products
were dissolved in 20% acetic acid and treated with a large
excess of Raney nickel. Desulfurization yields were moderate
and amounted to 61% for the model peptide 4 and 54% for
the Syk kinase peptide 16 despite prolonged reaction times.
Attempts to remove the thiol group in the STAT-1 peptide 15
failed. Neither was it possible to achieve the sulfur removal
nor could the starting material be recovered. Apparently, the
peptide material remained adsorbed onto the metal surface
despite several attempted extractions with trifluoroacetic
acid.
Very recently, Wan and Danishefsky described a method
for the metal-free desulfurization of peptides,[10] in which a
water-soluble radical starter first abstracts a hydrogen atom
from the cysteine thiol group which is then reduced with
TCEP to form an alanyl radical.[11] This alkyl radical receives
a hydrogen atom either from unreacted cysteinyl peptide or
from EtSH and tBuSH, which are added to accelerate product
formation. When we applied the published reaction condi-
tions (VA-044, TCEP, EtSH, and tBuSH) to penicillyl peptide
11 we observed that the desulfurized peptide 17 was formed;
however, it was accompanied by several by-products. We
assumed that hydride-transfer reactions would proceed less
effectively in penicillamine than in cysteine. To accelerate
thiyl radical formation, we increased the reaction temper-
ature and the amount of radical starter VA-044. Furthermore,
glutathione was used as the hydrogen source as we reasoned
that this powerful hydrogen donor[12] may react faster with the
formed valyl radical than malodorous EtSH and/or tBuSH.
The reactions were performed at peptide concentrations of
1–2 mm in an aqueous 100 mm phosphate buffer adjusted to
pH 6.5 which contained 250 mm TCEP, 200 mm VA-044,
40 mm glutathione, and 3m GnHCl. The desulfurization
reactions proceeded smoothly and went to completion
within 2.5 h (Figure 2). The penicillamine-containing peptides
11 and 16 were converted to the corresponding valine-
containing peptides 17 and 22 in 98% and 91% yield,
respectively (Table 1). Interestingly, even the STAT-1 peptide
segment 15, which resisted metal-induced thiol removal, was
successfully desulfurized (!21). The desulfurization of the
penicillamine peptides 12 and 14 was also straightforward. In
the case of the methionine-containing peptide 13 oxidation of
the thioether moiety concomitantly occurred when the
reaction was performed at 658C. The oxidation of the
methionine side chain was insignificant at a reaction temper-
ature of 378C. Under these conditions conversion into the
valine peptide 19 was complete after 6 h.
Figure 2. Metal-free desulfurization of penicillamine-containing peptide
Leu-Tyr-Lys-Ala-Gly-Pen-Arg-Ala-Glu-Tyr-Ser-NH2 (11). HPLC traces
A) before and B) after 150 min reaction time. The insets show the ESI
mass spectra, which provide ample evidence of desulfurization.
Conditions: 5 mm 11, 3m GnHCl, 100 mm NaH2PO4, 200 mm VA-044,
250 mm TCEP, 40 mm glutathione, 658C, pH 6.5.
Gly–Gly, Gly–Ala, Gly–Gln, Gly–Asp, Gly–His, Ala–Gly,
Lys–Gly, His–Gly, Phe–Gly, Pro–Gly, His–His, His–Ala, Ala–
His, and Ala–Asp peptide bonds has been demonstrated.[5]
The application of these methods requires access to non-
commercial building blocks. Frequently occurring Xaa–Ala
sites can be formed by means of the native chemical ligation/
desulfurization approach.[7] One advantage of this two-step
method is that only commonly applied amino acid derivatives
are used. The synthesis of b-sulfanylphenylalanine has
enabled ligations at Xaa–Phe sites.[8] The work described
here demonstrates the use of penicillamine (Pen) as a
precursor to valine in the ligation–desulfurization approach.
Valine is a frequently occurring amino acid (6.6% content).
The availability of suitably protected penicillamine building
blocks and in particular the feasibility to establish ligations at
hydrophobic peptide segments are considered as advanta-
geous.
The presented ligation reactions of thioesters terminated
by glycine, histidine, methionine, and leucine residues suggest
a broad applicability of the method, which may allow almost
general access to Xaa–Val peptide bonds. Future experiments
should reveal whether sterically crowded Val–Val or Ile–Val
peptide bonds can be formed. The demonstrated access to
Leu–Val ligation sites may provide interesting opportunities
in the synthesis of transmembrane proteins. However, we
wish to note that care should be taken to avoid epimerization
of reactive peptide thioesters during long reaction times. This,
however, applies to all reactions that involve peptide thio-
esters. The presented ligation–desulfurization approach also
included an optimized metal-free method for removing thiol
groups in peptides. Glutathione was used as the hydrogen
The native chemical ligation is among the few chemical
methods, if not the only one, that has enabled the synthesis of
proteins of a complexity that can usually be obtained only by
applying recombinant techniques.[13] In its original form,
native chemical ligation provides access to Xaa–Cys (Xaa =
any amino acid except proline) sites. Auxialiary-based
methods have extended the scope, and the formation of
Angew. Chem. Int. Ed. 2008, 47, 6807 –6810
ꢀ 2008 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.angewandte.org 6809

Communications
7460 – 7462; e) T. Kawakami, K. Akaji, S. Aimoto, Org. Lett.
2001, 3, 1403– 1405; f) P. Botti, M. R. Carrasco, S. B. H. Kent,
Tetrahedron Lett. 2001, 42, 1831 – 1833; g) J. Offer, C. N. C.
Boddy, P. E. Dawson, J. Am. Chem. Soc. 2002, 124, 4642 – 4646;
h) C. Marinzi, J. Offer, R. Longhi, P. E. Dawson, Bioorg. Med.
Chem. 2004, 12, 2749 – 2757; i) S. Tchertchian, O. Hartley, P.
Botti, J. Org. Chem. 2004, 69, 9208 – 9214; j) D. Macmillan, D. W.
Anderson, Org. Lett. 2004, 6, 4659 – 4662; k) B. Wu, J. Chen, J. D.
Warren, G. Chen, Z. Hua, S. J. Danishefsky, Angew. Chem. 2006,
118, 4222 – 4231; Angew. Chem. Int. Ed. 2006, 45, 4116 – 4125;
l) M. Y. Lutsky, N. Nepomniaschiy, A. Brik, Chem. Commun.
2008, 1229 – 1231.
source in the radical desulfurization rather than the less
effective and malodorous thiols EtSH and tBuSH. Consider-
ing the abundant availability of this inexpensive reagent in
bioorganic and biological chemistry laboratories, we propose
glutathione as the reagent of choice in metal-free desulfur-
ization reactions.
In conclusion, we have expanded the scope of native
chemical ligation by gaining access to hydrophobic ligation
sites. Careful optimization of both ligation and reduction
conditions allowed the use of the b-sulfanyl amino acid
penicillamine in the ligation–desulfurization approach. Nei-
ther special building blocks nor are risky reagents are
required; this may encourage the application of this new
option in peptide synthesis.
[6] a) A. Brik, Y. Y. Yang, S. Ficht, C. H. Wong, J. Am. Chem. Soc.
2006, 128, 5626 – 5627; b) S. Ficht, R. J. Payne, A. Brik, C. H.
Wong, Angew. Chem. 2007, 119, 6079 – 6083; Angew. Chem. Int.
Ed. 2007, 46, 5975 – 5979; c) 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
Received: April 4, 2008
[7] a) L. Z. Yan, P. E. Dawson, J. Am. Chem. Soc. 2001, 123, 526 –
533; b) B. L. Pentelute, S. B. H. Kent, Org. Lett. 2007, 9, 687 –
690.
[8] a) P. Botti, S. Tchertchian, WO 2006/133962, 2006; b) D. Crich,
A. Banerjee, J. Am. Chem. Soc. 2007, 129, 10064 – 10065.
[9] T. M. Hackeng, J. H. Griffin, P. E. Dawson, Proc. Natl. Acad. Sci.
USA 1999, 96, 10068 – 10073.
Published online: July 14, 2008
Keywords: native chemical ligation · penicillamine ·
.
peptide ligation · valine
[1] B. L. Nilsson, M. B. Soellner, R. T. Raines, Annu. Rev. Biophys.
Biomol. Struct. 2005, 34, 91 – 118.
[10] a) Q. Wan, S. J. Danishefsky, Angew. Chem. 2007, 119, 9408 –
9412; Angew. Chem. Int. Ed. 2007, 46, 9248 – 9252.
[2] a) T. Wieland, E. Bokelmann, L. Bauer, H. U. Lang, H. Lau,
Justus Liebigs Ann. Chem. 1953, 583, 129 – 149; b) P. E. Dawson,
T. W. Muir, I. Clark-Lewis, S. B. H. Kent, Science 1994, 266, 776 –
779.
[3] P. E. Dawson, S. B. H. Kent, Annu. Rev. Biochem. 2000, 69, 923–
960.
[11] a) C. Walling, R. Rabinowitz, J. Am. Chem. Soc. 1957, 79, 5326;
b) A. Gonzꢀlez, G. Valencia, Tetrahedron: Asymmetry 1998, 9,
2761 – 2764; c) J. Cuesta, G. Arsequell, G. Valencia, A. Gonzꢀlez,
Tetrahedron: Asymmetry 1999, 10, 2643– 2646; d) G. Arsequell,
A. Gonzꢀlez, G. Valencia, Tetrahedron Lett. 2001, 42, 2685 –
2687.
[12] a) C. Chatgilialoglu, Helv. Chim. Acta 2006, 89, 2387 – 2398;
b) C. Chatgilialoglu, C. Ferreri, R. Bazzanini, M. Guerra, S. Y.
Choi, C. J. Emanuel, J. H. Horner, M. Newcomb, J. Am. Chem.
Soc. 2000, 122, 9525 – 9533.
[4] C. Haase, O. Seitz, Angew. Chem. 2008, 120, 1575 – 1579; Angew.
Chem. Int. Ed. 2008, 47, 1553– 1556.
[5] a) L. E. Canne, S. J. Bark, S. B. H. Kent, J. Am. Chem. Soc. 1996,
118, 5891 – 5896; b) J. Offer, P. E. Dawson, Org. Lett. 2000, 2, 23 –
26; c) C. Marinzi, S. J. Bark, J. Offer, P. E. Dawson, Bioorg. Med.
Chem. 2001, 9, 2323 – 2328; d) G. Chen, J. D. Warren, J. Chen. B.
Wu, Q. Wan, S. J. Danishefsky, J. Am. Chem. Soc. 2006, 128,
[13] V. Y. Torbeev, S. B. H. Kent, Angew. Chem. 2007, 119, 1697 –
1700; Angew. Chem. Int. Ed. 2007, 46, 1667 – 1670.
6810 www.angewandte.org
ꢀ 2008 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Angew. Chem. Int. Ed. 2008, 47, 6807 –6810