Trimethoxyphenylthio as a highly labile replacement for tert-butylthio cysteine protection in fmoc solid phase synthesis

Article abstract of DOI:10.1021/ol3025499

Trimethoxyphenylthio (S-Tmp) is described as a novel cysteine protecting group in Fmoc solid phase peptide synthesis replacing the difficult to remove tert-butylthio. S-Tmp and dimethoxyphenylthio (S-Dmp) were successfully used for cysteine protection in a variety of peptides. Moreover, both groups can be removed in 5 min with mild reducing agents. S-Tmp is recommended for cysteine protection, as it yields crude peptides of high purity.

Full text of DOI:10.1021/ol3025499

ORGANIC
LETTERS
2012
Vol. 14, No. 21
5468–5471
Trimethoxyphenylthio as a Highly Labile
Replacement for tert-Butylthio Cysteine
Protection in Fmoc Solid Phase Synthesis
Tobias M. Postma,†,‡ Matthieu Giraud,§ and Fernando Albericio*,†,‡, ,^
Institute for Research in Biomedicine, Barcelona, 08028, Spain, CIBER-BBN,
08028-Barcelona, Spain, Department of Organic Chemistry, University of Barcelona,
08028, Barcelona, Spain, Lonza Ltd., Visp, VS 3930, Switzerland, and School of
Chemistry, University of KwaZulu Natal, 4001-Durban, South Africa
fernando.albericio@irbbarcelona.org
Received September 16, 2012
ABSTRACT
Trimethoxyphenylthio (S-Tmp) is described as a novel cysteine protecting group in Fmoc solid phase peptide synthesis replacing the difficult to
remove tert-butylthio. S-Tmp and dimethoxyphenylthio (S-Dmp) were successfully used for cysteine protection in a variety of peptides. Moreover,
both groups can be removed in 5 min with mild reducing agents. S-Tmp is recommended for cysteine protection, as it yields crude peptides of high
purity.
Multiple disulfide containing peptides are ubiquitous in
nature and therapeutically relevant because of their selec-
tiveandpotent bioactivities(e.g., conotoxins, cyclotides).1,2
The synthesis of multiple disulfide containing peptides
requires the use of orthogonal cysteine (Cys) protection
strategies to ensure the correct disulfide connectivity.3,4
Unfortunately, there is a lack of orthogonal cysteine pro-
tecting groups that can be used in routine SPPS under mild
conditions.5
The concept of Cys protecting groups labile to mild
reducing agentsishighlypromising dueto orthogonality to
other Cys protecting groups. The commercial Cys protect-
ing group tert-butylthio (StBu) is orthogonal to all other
Cys protecting groups due to its mild deprotection
conditions.6 This protecting group can be removed with
mild reducing agents (e.g., thiols or phosphines) and it is
stable to piperidine, hence compatible with Fmoc/tBu
peptide synthesis. However, owing to exceedingly long
deprotection times (4À24 h), StBu cannot be used in routine
SPPS. As a result of the proximity of bulky protecting
groups and sensitivity to certain amino acid sequences
prone to folding, in some cases this protecting group has
proven to be very difficult or even impossible to remove.7,8
Thus, in previous studies on Linaclotide, a 14-residue
peptide containing three disulfide bonds, we observed that
StBu groups on Cys2-Cys10 were not possible to remove.7
Denis and Trifilieff used harsh conditions, heating a pepti-
dyl resin to 135 ꢀC for 24 h; however, they achieved only
† Institute for Research in Biomedicine.
‡ CIBER-BBN.
§ University of Barcelona.
Lonza Ltd.
^ University of KwaZulu Natal.
(1) Livett, B. G.; Gayler, K. R.; Khalil, Z. Curr. Med. Chem. 2004, 11,
1715–1723.
ꢀ
(2) Craik, D. J.; Cemazar, M.; Wang, C. K. L.; Daly, N. L. Peptide
ꢀ
Sci. 2006, 84, 250–266.
(3) Andreu, D.; Albericio, F.; Sole, N. A.; Munson, M. C.; Ferrer,
M.; Barany, G. Methods in Molecular Biology: Peptide Synthesis Pro-
tocols; Pennington, M. W., Dunn, B. M., Eds.; Humana Press, Inc.: Totowa,
NJ, 1994; Vol. 45, pp 91À169.
(6) Weber, U.; Hartter, P. Hoppe-Seyler’s Z. Physiol. Chem. 1970,
351, 1384–1388.
(4) Liu, H.; Boudreau, M. A.; Zheng, J.; Whittal, R. M.; Austin, P.;
Roskelley, C. D.; Roberge, M.; Andersen, R. J.; Vederas, J. C. J. Am.
Chem. Soc. 2010, 132, 1486–1487.
ꢁ
(7) Gongora-Benıtez, M.; Tulla-Puche, J.; Paradıs-Bas, M.; Werbitzky,
´ ´
ꢁ
O.; Giraud, M.; Albericio, F. Peptide Sci. 2011, 96, 69–80.
(8) Dennis, B.; Trifilieff, E. J. Pept. Sci. 2000, 6, 372–377.
(5) Isidro-Llobet, A.; Alvarez, M.; Albericio, F. Chem. Rev. 2009,
109, 2455–2504.
r
10.1021/ol3025499
Published on Web 10/17/2012
2012 American Chemical Society

partial StBu deprotection.8 Additionally, reports of desul-
furization of StBu protected Cys to dehydroalanine, by
means of prolonged exposure to reducing agents, illustrates
the limitations of this protecting group.9
Given the importance of having a Cys protecting group
removable by reducing agents to fulfill the orthogonal
scheme for Cys and the significant limitations of StBu,
we have addressed novel reduction labile Cys protecting
groups. As scaffolds for the preparation of mixed disul-
fidesweinitially studied phenyland benzyl derivatives. The
former were not stable to base while the latter were clearly
not stable to acid (data not shown). The benzyl derivatives
were discarded, and the phenyl group was modified to
contain alkoxy groups on the 2,6 positions in order to
study the balance between high lability to reducing agents
and base stability (Scheme 1).
Figure 1. Tripeptides protected with StBu, S-Dmp, and S-Tmp.
protected with S-Dmp, S-Tmp, and StBu were incorpo-
rated in the model tripeptides by SPPS, using diisopro-
pylcarbodiimide (DIC) and Oxyma Pure with 5 min of
preactivation to prevent racemization of the Cysresidue.12
Scheme 1. Synthesis of Fmoc-Cys(S-Dmp)-OH (3) and
Fmoc-Cys(S-Tmp)-OH (4)
Table 1. On-Resin Deprotection of Model Tripeptidesa
peptidyl-
resin
deprotection time
(min) with BME
deprotection time
(min) with DTT
5 (StBu)
180
5
5% after 24 h
6 (S-Dmp)
7 (S-Tmp)
8 (StBu)
5
5
5
360
5
2% after 24 h
9 (S-Dmp)
10 (S-Tmp)
5
5
5
a β-Mercaptoethanol (BME), Dithiothreitol (DTT). Deprotection
conditions: 0.1 M NMM, BME (20%), or DTT (5%) in DMF.
The synthesis of Fmoc-Cys(S-Dmp)-OH and Fmoc-
Cys(S-Tmp)-OH is shown in Scheme 1. 2,6-Dimethox-
ythiophenol was prepared following the literature, and
2,4,6-trimethoxythiophenol was synthesized with minor
modifications of the same procedure.10 The key reaction
to the mixed disulfide containing Cys was inspired by a
reaction used by Kraus and Jeon where N-chlorosuccin-
imide (NCS) reacts with thiophenol to form a highly
reactive sulfenyl chloride.11
Fmoc-Cys(S-Dmp)-OH was prepared by forming a
sulfenyl chloride from 2,6-dimethoxythiophenol. This sul-
fenyl chloride was subsequently added to a solution of
Fmoc-Cys-OH, where nucleophilic attack on the sulfenyl
chloride by Cys yields the mixed disulfide. The sulfenyl
chloride of 2,4,6-trimethoxythiophenol was unstable
at À30 ꢀC and had to be formed in the presence of Fmoc-
Cys-OH at À78 ꢀC to obtain Fmoc-Cys(S-Tmp)-OH.
In order to compare the efficiency of S-Dmp and S-Tmp
to StBu, and SPPS compatibility, we prepared several
model tripeptides (Figure 1). The model tripeptide Fmoc-
Ala-Cys-Ala-NH2 was already used in our laboratory to
assess the Cys protecting group on resin. Cys residues
The stability of the protecting groups to piperidine and
trifluoroacetic acid (TFA), conditions used in routine
SPPS, was evaluated. All protecting groups were found to
be stable to 20% piperidine/DMF for 4 h, which is suffi-
cient for routine applications. The peptides were cleaved
from the resin with 95% TFA for 1 h at rt. The protecting
groups were stable to these conditions.
Subsequently, the lability of these groups to reducing
agents was studied. We found that the most efficient
deprotection mixtures contained N-methylmorpholine
(NMM) (0.1 M) and either the malodorous β-mercapto-
ethanol (BME) (20%) or nonmalodorous dithiothreitol
(DTT) (5%) in DMF.13 Both deprotection mixtures
achieved quantitative removal of S-Dmp and S-Tmp
from Fmoc-Ala-Cys(PG)-Leu-NH2 model tripeptides
in 5 min. In contrast, 3 h were required for quantitative
StBu removal using BME and practically no deprotec-
tion occurred with deprotection mixtures containing DTT
(Table 1).
(9) Rijkers, D. T. S.; Kruijtzer, J. A. W.; Killian, J. A.; Liskamp,
R. M. J. Tetrahedron Lett. 2005, 46, 3341–3345.
ꢁ
(12) Subiros-Funosas, R.; Prohens, R.; Barbas, R.; El-Faham, A.;
Albericio, F. Chem.;Eur. J. 2009, 15, 9394–9403.
(10) Wada, M.; Natsume, S.; Suzuki, S.; Akira, U.; Nakamura, M.;
Hayase, S.; Erabi, T. J. Organomet. Chem. 1997, 548, 223–227.
(11) Kraus, G. A.; Jeon, I. Tetrahedron 2005, 61, 2111–2116.
(13) No on-resin reduction was observed with TCEP in DMF and
DMF/H2O mixtures (low DMF solubility). We chose not to work with
tributylphosphine to avoid possible desulfurization of Cys (ref 9).
Org. Lett., Vol. 14, No. 21, 2012
5469

Scheme 2. Oxytocin Synthesis from Linear Oxytocin Protected with StBu (a), S-Dmp (b), and S-Tmp (c) on a Rink Amide AM Resina
a StBu removal: 0.1 M NMM in BME/DMF (1:4). S-Dmp/S-Tmp removal: 0.1 M NMM in DTT/DMF (5:95).
The deprotection efficiency for the removal of StBu,
S-Dmp, and S-Tmp, when flanked by bulky trityl (Trt)
protecting groups, was studied. The model tripeptide
Fmoc-Asn(Trt)-Cys(PG)-Asn(Trt)-NH2 was used to de-
termine whether bulky protecting groups influence depro-
tection times. The resin containing the peptides was treated
with the BME deprotection mixture. The deprotection
time for StBu doubled, while those for S-Dmp and S-Tmp
remained unchanged. On the other hand when the DTT
mixture was used, both S-Dmp and S-Tmp were removed
very gently in 5 min, while only 5% of StBu was removed
after 24 h. For subsequent on-resin deprotection, we applied
3 Â 5 min treatments to ensure quantitative removal of
S-Dmp and S-Tmp.
These promising results led ustoexamine a longer model
peptide. For this purpose, we chose oxytocin,14 an exten-
sively studied nonapeptide containing two Cys residues
and currently used as a drug to induce labor. A pair of
Cys residues protected with StBu, S-Dmp, or S-Tmp were
incorporated into oxytocin. Treatment of the peptidyl-
resin with 20% piperidine/DMF followed by cleavage with
TFA/TIS/H2O (95:2.5:2.5) led to the unexpected mono-
deprotection of S-Dmp (13%) and S-Tmp (8%). This
process was not observed during the studies with the model
tripeptides, and the mechanism of deprotection was un-
clear. However, we suspected that the monodeprotection
was caused by a high concentration of TFA.
requires a cleavage mixture containing 95% TFA, whereas
1À5% TFA is sufficient to achieve peptide cleavage from
Sieber Amide resin. The peptidyl-resin containing oxytocin
protected with S-Dmp or S-Tmp was treated with 20%
piperidine/DMF followed by cleavage with 1% TFA. Sub-
sequent analysis showed no monodeprotection for either
protecting groups, thereby corroborating the instability to
high TFA concentrations. The partial instability of these
protecting groups to high TFA concentrations is not detri-
mental for their use in peptide synthesis because in multiple
disulfide containing peptides these groups are predominantly
removed on-resin. Due to their deprotection mechanism,
S-Dmp and S-Tmp must be removed first in orthogonal Cys
protecting group strategies; otherwise the reducing agents
required for deprotection would reduce the other disulfide
bonds present. Hence, these protecting groups are removed
on-resin prior to cleavage. Thus, in SPPS, the end of the
synthesis involves removal of the Fmoc group, removal of
the Cys protecting groups (S-Dmp/S-Tmp), and then either
oxidation followed by cleavage or cleavage with subsequent
oxidation in solution.
Linear oxytocin protected with StBu, S-Dmp, or S-Tmp
was deprotected on-resin (3 Â 5 min, 0.1 M NMM in 5%
DTT/DMF) for S-Dmp and S-Tmp, whereas 8 h (0.1 M
NMM in 20% BME/DMF) were needed for StBu depro-
tection (Scheme 2). The linear peptides were cleaved from
the resin and oxidized in 5% DMSO in H2O/CH3CN (3:1)
for 24 h. The peptides protected with StBu, S-Dmp, and
S-Tmp were obtained in 72%, 86%, and 94% purity after
lyophilization. Given that the best results were obtained
with S-Tmp, we focused on this protecting group.
To confirm this notion, oxytocin was resynthesized with
protecting group S-Dmp or S-Tmp on a Sieber Amide Resin.
Previously, a Rink Amide resin was used. This support
(14) Viero, C.; Shibuya, I.; Kitamura, N.; Verkhratsky, A.; Fujihara,
H.; Katoh, A.; Ueta, Y.; Zingg, H. H.; Chvatal, A.; Sykova, E.;
Dayanithi, G. CNS Neurosci. Ther. 2010, 16, 138–156.
To further demonstrate the applicability of S-Tmp, we
synthesizedthe18-residueT22 peptidecontaining4S-Tmp
5470
Org. Lett., Vol. 14, No. 21, 2012

Scheme 3. Synthesis of T22 Using S-Tmp Protected Cys on a ChemMatrix Resin
protected Cys residues.15 In a recent study we demon-
strated that the T22 peptide folds into its native conforma-
tion by oxidative folding.16 This peptide contains five
arginine and three lysine residues, thus rendering it highly
hydrophilic. We were unable to properly synthesize the
peptide on a polystyrene based Rink Amide resin and
required a hydrophilic polyethylene glycol based Chem-
Matrix resin for this purpose. Following peptide elonga-
tion, S-Tmp was removed on-resin (3 Â 5 min, 0.1 M
NMM, 5% DTT in DMF) and cleaved from the resin with
TFA/TIS/H2O (95:2.5:2.5) for 3 h to ensure complete de-
protection of the five Pbf protecting groups (Scheme 3).
The linear peptide was oxidized in solution using 5%
DMSO in water at pH 8.0 for 24 h at rt and subsequently
lyophilized to give crude T22 in 78% purity.
In comparison to StBu, S-Dmp and S-Tmp proved to be
superior Cys protecting groups. They are labile to mild
reducing agents and have rapid deprotection times of
5 min, which contrasts with the hours required for StBu
removal. No changes were observed in the deprotection
times when the protecting group was flanked by bulky Trt
groups. In concusion, we recommend the use of S-Tmp
over S-Dmp because it showed greater stability, gave the
purest product in the synthesis of oxytocin, and was
successfully used in T22 synthesis.
ꢁ
Acknowledgment. We are grateful to Miriam Gongora-
Benıtez from the Institute for Research in Biomedicine
´
for academic input. We thank the European community -
FP7-PEOPLE-ITN-2008 for financial support. We also
thank Dr. Thomas Bruckdorfer from IRIS Biotech GmbH
for fruitful discussions.
Both S-Dmp and S-Tmp are stable to base, SPPS
coupling conditions, and a low concentration of TFA.
Supporting Information Available. Experimental de-
tails and spectroscopic data. This material is available
free of charge via the Internet at http://pubs.acs.org.
(15) Nakashima, H.; Masuda, M.; Murakami, T.; Koyanagi, Y.;
Matsumoto, A.; Fujii, N.; Yamamoto, N. Antimicrob. Agents Che-
mother. 1992, 36, 1249–1255.
ꢁ
(16) Manuscript accepted: Gongora-Benıtez, M.; Basso, A.;
´
Bruckdorfer, T.; Royo, M.; Tulla-Puche, J.; Albericio, F. Chem.;Eur.
J. DOI: 10.1002/chem.201201370.
The authors declare no competing financial interest.
Org. Lett., Vol. 14, No. 21, 2012
5471