Synthesis of peptide thioacids at neutral pH using bis(2-sulfanylethyl) amido peptide precursors

Article abstract of DOI:10.1021/ol402601j

Reaction of bis(2-sulfanylethyl)amido (SEA) peptides with triisopropylsilylthiol in water at neutral pH yields peptide thiocarboxylates. An alkylthioester derived from β-alanine was used to trap the released bis(2-sulfanylethyl)amine and displace the equi

Full text of DOI:10.1021/ol402601j

ORGANIC
LETTERS
XXXX
Vol. XX, No. XX
000–000
Synthesis of Peptide Thioacids at Neutral
pH Using Bis(2-sulfanylethyl)amido
Peptide Precursors
Silvain L. Pira, Emmanuelle Boll, and Oleg Melnyk*
UMR CNRS 8161, Pasteur Institute of Lille 59021 Lille, France
oleg.melnyk@ibl.fr.
Received September 9, 2013
ABSTRACT
Reaction of bis(2-sulfanylethyl)amido (SEA) peptides with triisopropylsilylthiol in water at neutral pH yields peptide thiocarboxylates. An
alkylthioester derived from β-alanine was used to trap the released bis(2-sulfanylethyl)amine and displace the equilibrium toward the peptide
thiocarboxylate.
C-terminal peptide thioacids are used in a variety of
useful chemoselective reactions and therefore constitute
an important class of peptide derivatives.1 For example,
selective activation of C-terminal peptide thioacids by
silver ion,2a Sanger and Mukaiyama reagents,2b arylsul-
fonamides,2c bromoacetic acid,2d or oxidants2e enables the
block synthesis of peptides. The nucleophilic properties
of the thioacid group toward bromoalkyl,3a disulfide,3b
isocyanate or isothiocyanate,3c aziridine,3d isonitrile,3e or
azide functionalities4 were also used for designing various
useful native or peptidomimetic amide bond-forming
reactions.
The development of synthetic methods giving access to
C-terminal peptide thioacids is thus an important goal. It is
also challenging because of the sensitivity of thioacids to
nucleophiles, oxidants, and various electrophilic reagents.
Peptide thioacids can be produced by the Boc5a,b or Fmoc5c
solid-phase peptide synthesis (SPPS) methods. The thio-
ester linkage linking the peptide to the solid support
was cleaved with strong acids5 or by hydrothiolysis5c,d in
aqueous base to release the peptide thioacid in solution. An
alternative is to use soluble peptide thioester precursors
to generate the peptide thioacid by hydrothiolysis6 or by
base-induced elimination from S-(9-fluorenylmethyl)7 or
(1) Sasaki, K.; Aubry, S.; Crich, D. Phosphorus, Sulfur, Silicon 2011,
186, 1005–1018.
(2) (a) Blake, J.; Li, C. H. Proc. Natl. Acad. Sci. U.S.A. 1981, 78,
4055–4058. (b) Crich, D.; Sharma, I. Angew. Chem., Int. Ed. 2009, 48,
2355–2358. (c) Crich, D.; Sharma, I. Angew. Chem., Int. Ed. 2009, 48,
7591–7594. (d) Canne, L. E.; Botti, P.; Simon, R. J.; Chen, Y.; Dennis,
E. A.; Kent, S. B. H. J. Am. Chem. Soc. 1999, 121, 8720–8727. (e) Wang,
P.; Danishefsky, S. J. J. Am. Chem. Soc. 2010, 132, 17045–17051.
(3) (a) Tam, J. P.; Lu, Y. A.; Liu, C. F.; Shao, J. Proc. Natl. Acad. Sci.
U.S.A. 1995, 92, 12485–12489. (b) Liu, C.-F.; Rao, C.; Tam, J. P.
Tetrahedron Lett. 1996, 37, 933–936. (c) Crich, D.; Sasaki, K. Org. Lett.
2009, 11, 3514–3517. (d) Assem, N.; Yudin, A. K. Nat. Protoc. 2012, 7,
1327–1334. (e) Wang, T.; Danishefsky, S. J. Proc. Natl. Acad. Sci. U.S.A.
2013, 110, 11708–11713.
(4) (a) Kolakowski, R. V.; Shangguan, N.; Sauers, R. R.; Williams,
L. J. J. Am. Chem. Soc. 2006, 128, 5695–5702. (b) Fazio, F.; Wong, C.-H.
Tetrahedron Lett. 2003, 44, 9083–9085. (c) Barlett, K. N.; Kolakowski,
R. V.; Katukojvala, S.; Williams, L. J. Org. Lett. 2006, 8, 823–826. (d)
Merkx, R.; Brouwer, A. J.; Rijkers, D. T.; Liskamp, R. M. Org. Lett.
2005, 7, 1125–1128. (e) Zhang, X.; Li, F.; Lu, X. W.; Liu, C. F.
Bioconjugate Chem. 2009, 20, 197–200. (f) Mhidia, R.; Beziere, N.;
Blanpain, A.; Pommery, N.; Melnyk, O. Org. Lett. 2010, 12, 3982–
3985. (g) Mhidia, R.; Boll, E.; Fecourt, F.; Ermolenko, M.; Ollivier, N.;
Sasaki, K.; Crich, D.; Delpech, B.; Melnyk, O. Bioorg. Med. Chem. 2013,
21, 3479–3485.
(5) (a) Canne, L. E.; Walker, S. M.; Kent, S. B. H. Tetrahedron Lett.
1995, 36, 1217–1220. (b) Gaertner, H.; Villain, M.; Botti, P.; Canne, L.
Tetrahedron Lett. 2004, 45, 2239–2241. (c) Zhang, X.; Lu, X.-W.; Liu,
C.-F. Tetrahedron Lett. 2008, 49, 6122–6125. (d) Tsuji, K.; Shigenaga,
A.; Sumikawa, Y.; Tanegashima, K.; Sato, K.; Aihara, K.; Hara, T.;
Otaka, A. Bioorg. Med. Chem. 2011, 19, 4014–4020.
(6) (a) Zhang, X.; Li, F.; Lu, X. W.; Liu, C. F. Bioconjugate Chem.
2009, 20, 197–200. (b) Tan, X. H.; Zhang, X.; Yang, R.; Liu, C. F.
ChemBioChem 2008, 9, 1052–1056.
(7) Crich, D.; Sana, K.; Guo, S. Org. Lett. 2007, 9, 4423–4426. For a
recent adaptation of 9-fluorenylmethyl esters to the synthesis of peptide
ꢀ
selenocarboxylates see: Fecourt, F.; Delpech, B.; Melnyk, O.; Crich, D.
Org. Lett. 2013, 15, 3758–3761.
r
10.1021/ol402601j
XXXX American Chemical Society

S-(2-cyanoethyl)8 thiocarboxylates. The need for a milder
access to peptide thioacids led us to develop a method
allowing the synthesis of these useful peptide derivatives in
water at neutral pH.
Scheme 1. Conversion of SEA Peptides into Peptide Thioacids
by Reaction with Triisopropylsilyl Thiol
The principle of the method is described in Scheme 1 and
exploits the masked thioester properties of bis(2-
sulfanylethyl)amido (SEA) peptides 1,9 which can be
produced by Fmoc SPPS. Reduction of SEA peptide 1
with tris(2-carboxyethyl)phosphine (TCEP) at neutral pH
generates dithiol peptide 2, which equilibrates in situ with
the transient thioester peptide 3. At the beginning of this
work, we expected that hydrothiolysis of peptide 3 with
NaSH could yield directly target thiocarboxylate 6. How-
ever, reaction of model peptide 1a with an excess of NaSH
in pH 7 sodium phosphate buffer in the presence of TCEP
and 4-mercaptophenylacetic acid10 (MPAA) to catalyze
the exchange reaction failed to give any peptide thiocar-
boxylate 6a. We next examined if triisopropylsilylthiol
could serve as a hydrogen sulfide anion surrogate. Indeed,
we reasoned that the thiolÀthioester exchange reaction
between peptide 3 and triisopropylsilylthiol could yield
peptide thiocarboxylate 6 after in situ hydrolysis of
intermediate triisopropylsilyl thioester 4. Gratifyingly,
reaction of peptide 1a with triisopropylsilylthiol, TCEP,
and MPAA at pH 7 yielded thiocarboxylate 6a, albeit with
an HPLC yield of only ∼30À35% after 20 h (Figure 1a).11
Note that the conversion and rate of thiocarboxylate
formation was significantly lower in the absence of
MPAA.12
(ii) the fact that the latter does not react with peptide 2 to
give thiocarboxylate 6 as discussed before, and (iii) the
concentration of triisopropylsilylthiol is limited. Under
these conditions, the cyclic process highlighted in red in
Scheme 1 is fueled with triisopropylsilylthiol and produces
hydrogen sulfide ion, which presumably escapes from the
reaction mixture as hydrogen sulfide (pKa 7.05). It runs
until the concentration of triisopropylsilylthiol is not high
enough to sustain the production of thiocarboxylate 6.
Interestingly, feeding the reaction again with triisopropyl-
silylthiol after 49 h allowed us to maintain the proportion
of peptide thiocarboxylate 6a in the mixture, in accordance
with the proposed mechanism (Figure 1a). We noticed also
that prolonged reaction times resulted in the appearance of
an organic phase, which by sequestering triisopropylsi-
lylthiol might amplify the decrease of triisopropylsilylthiol
concentration in the aqueous phase and thus the reversal of
the reaction. The use of guanidinium hydrochloride as an
additive(3.7 M) in addition tot-BuOHsolvedthisproblem
and retarded the reversal of peptide thiocarboxylate
formation (Figure 1a).
With this knowledge, we reasoned that capturing amino
thiol 5 would favor the formation of target thiocarboxylate
6. We used for this the capacity of alkylthioesters 7 to react
with β-amino thiol 5 to give amides 8 by analogy with the
native chemical ligation (NCL) of peptide alkylthioesters
with Cys peptides.14 Interestingly, the use of 2.3 equiv of
Gly-S(CH2)2CO2H 7a15 resulted in a significant increase
in the yield of thiocarboxylate 6a (Figure 1b).16 LCÀMS
LCÀMS analysis of the reaction mixture in the presence
of MPAA showed essentially two peaks corresponding
to unreacted dithiol peptide 2a and thiocarboxylate 6a.
Unexpectedly, longer reaction times resulted in the reverse
process, that is an increase of peptide 2a proportion in the
mixture to the detriment of thiocarboxylate 6a (Figure 1a).
This phenomenon can be understood by considering (i)
the capacity of thiocarboxylate 6 to react efficiently with
amino thiol 5 to give peptide 2 plus hydrogen sulfide ion,13
(8) Raz, R.; Rademann, J. Org. Lett. 2012, 14, 5038–5041.
(9) (a) Raibaut, L.; Adihou, H.; Desmet, R.; Delmas, A. F.; Aucagne,
V.; Melnyk, O. Chem. Sci. 2013, 4, 4061–4066. (b) Raibaut, L.; Vicogne,
J.; Leclercq, B.; Drobecq, H.; Desmet, R.; Melnyk, O. Bioorg. Med.
Chem. 2013, 21, 3486–3494. (c) Ollivier, N.; Vicogne, J.; Vallin, A.;
Drobecq, H.; Desmet, R.; El-Mahdi, O.; Leclercq, B.; Goormachtigh,
G.; Fafeur, V.; Melnyk, O. Angew. Chem., Int. Ed. 2012, 51, 209–213. (d)
Boll, E.; Dheur, J.; Drobecq, H.; Melnyk, O. Org. Lett. 2012, 14, 2222–
2225. (e) Dheur, J.; Ollivier, N.; Vallin, A.; Melnyk, O. J. Org. Chem.
2011, 76, 3194–3202. (f) Dheur, J.; Ollivier, N.; Melnyk, O. Org. Lett.
2011, 13, 1560–1563. (g) Ollivier, N.; Dheur, J.; Mhidia, R.; Blanpain,
A.; Melnyk, O. Org. Lett. 2010, 12, 5238–5241.
(10) Johnson, E. C.; Kent, S. B. J. Am. Chem. Soc. 2006, 128, 6640–
6646.
(11) Triisopropylsilyl thioester 4a was not detected by LCÀMS in the
reaction mixture. The presence of thiocarboxylate 6a was demonstrated
by alkylation experiments with iodoacetamide and LCÀMS detection of
the formed peptide thioester ILKEPVHGA-SCH2CONH2.
(12) In the absence of MPAA, we also observed the formation of side
products arising from the desulfurization of dithiol peptide 2a.
(13) Reaction of purified peptide 6a (4.8 mM) with amino thiol 5
(4.8 mM) at pH 7 (30% t-BuOH) and 37 °C in the presence of TCEP
(30 mM) and MPAA (60 mM) cleanly furnished peptide 2a with a half-
life of ∼15 h (see Figure S29 in the Supporting Information). As
discussed in ref 2e, the capacity of thioacids to acylate amines might
proceed through the oxidation of the thioacid into an diacyl disulfide by
traces of molecular oxygen.
(14) Dawson, P. E.; Muir, T. W.; Clark-Lewis, I.; Kent, S. B. Science
1994, 266, 776–779.
(15) Thioesters 7aÀc were prepared according to the method of Hojo
et al. Hojo, H.; Aimoto, S. Bull. Soc. Chem. Jpn 1991, 64, 111–117.
(16) A similar kinetic profile was obtained by using 3.3 equiv of 7a.
B
Org. Lett., Vol. XX, No. XX, XXXX

kinetic profile for peptides 11 and 12 is dictated by the
3-aminopropionamide structure of Asp(SEA) or βAla-
SEA moieties rather than by their position within the
peptide structure. The important consequence of these
observations is that βAla-SEA amide 8c is probably a poor
amino thiol 5 donor compared to Gly-SEA amide 8a.
Figure 2. Time course of the ligation of peptides 9, 11, or 12 with
CILKEPVHGV-NH2 at pH 7 and 37 °C (TCEP 200 mM,
MPAA 200 mM).
We thus evaluated the relevance of βAla-S(CH2)2CO2H
7c as trapping agent. Gratifyingly, the use of 2.3 equiv of 7c
led to a net increase in the yield of thiocarboxylate 6a
compared to Gly analogue 7a. The best yield was obtained
after 48 h of reaction at 37 °C (Figure 1b).
Figure 1. Time course of the conversion of SEA peptide 1a into
thiocarboxylate 6a (detection at 215 nm).
These experimental conditions proved to be efficient for
different C-terminal amino acids such as Ala, Arg, Leu,
Ser, and Tyr as shown in Table 1. In each case, LCÀMS
analysis of the reaction mixtures showed the formation of
amide 8c in parallel to those of thiocarboxylates 6aÀe
(Figure 3). Using these conditions, thiocarboxylate 6a was
isolated in 45% yield (52% corrected, entry 1, Table 1).
Note that the purification of thiocarboxylates 6aÀe was
carried out by HPLC using triethylamineÀacetate pH 6.5
buffer, which to the contrary of 0.1% TFA water/acetonitrile
eluent preserves the peptide thiocarboxylates from partial
degradation during the lyophilization step.17 Figure 3 shows
the LCÀMS analysis of the crude thiocarboxylate 6e as a
typical example.
analysis of the mixture showed also the formation of amide
8ainaccordancewiththemechanismproposedinScheme1
(see the Supporting Information). In contrast, no signifi-
cant improvement in the yield of thiocarboxylate 6a was
observedbyusingVal-S(CH2)2CO2H 7bastrapping agent.
The fact that valine thioester 7b is a poor trapping agent
compared to glycine analogue 7a is due presumably to the
reduced accessibility of the valine carbonyl group. Given
the reversibility of amide 8 formation (Scheme 1), we
reasoned that the ideal thioester trapping agent 7 should
react efficiently with amino thiol 5 such as Gly thioester 7a,
but the formed amide 8 should be a poor amino thiol 5
donor.
Kinetic studies with C-terminal SEA peptides estab-
lished that ligation with Cys peptides proceeded faster
for Gly as C-terminal residue than for other amino acid
residues.9g Figure 2 shows the time course of the ligation of
peptide ILKEPVHGG-SEA 9 with CILKEPVHGV-NH2
10 at pH 7 in the presence of TCEP and MPAA (200 mM
each). Inother words, SEA peptides featuringa C-terminal
Gly residue are good amino thiol 5 donors. In contrast, the
reactivity of peptide Ac-Asp(SEA)-ILKEPVHGA-NH2
11 featuring a SEA group on the side chain of an aspartate
residue was significantly altered, maybe due to the lack of
an amido group in R position relative to the SEA carbonyl
group (Figure 2).9d We also noticed that the kinetic rate
for the reaction of peptide ILKEPVHG-βAla-SEA 12
with Cys peptide 10 was very similar to those obtained
for Asp(SEA) peptide 11. These results show that the
Table 1. Synthesis of Thiocarboxylates 6aÀea
entry
thiocarboxylate
X
isolated yield (%)
1
2
3
4
5
6a
6b
6c
6d
6e
Ala
Arg
Leu
Ser
Tyr
45 (52% corrected)
44
27
32
47
a Peptide 1 (4.1 mM), thioester 7c (9.4 mM), TCEP (27 mM), and
MPAA (56 mM) in 0.1 M pH 7.1 sodium phosphate buffer, 30%
tBuOH, 3.7 M Gdn.HCl, 37 °C under nitrogen atmosphere.
(17) Contrary to thiocarboxylates, thioacids hydrolyze upon storage;
see ref 3d and: Mitin, Yu, V.; Zapevalova, N. P. Int. J. Pept. Protein Res.
1990, 35, 352–356.
Org. Lett., Vol. XX, No. XX, XXXX
C

Figure 3. Synthesis of peptide ILKEPVHGY-SH 6e. LCÀMS analysis of the crude reaction mixture (after extraction of MPAA).
In conclusion, C-terminal SEA peptides can be con-
verted into the corresponding thiocarboxylates at pH 7 by
reaction with triisopropylsilylthiol in the presence of an
Scheme 2. Synthesis of Imide 14
alkylthioester derivative of β-alanine. The reaction is very
mild and racemization free and should facilitate the access
to these highly valuable peptide derivatives.
Acknowledgment. We are grateful for the financial support
from ANR (ANR- 11-BS07-022-01). We also thank Laurent
Raibaut (CNRS, France) and Elisabeth Lissy (Louisiana State
Finally, the functionality of thiocarboxylates 6 was
verified by the chemoselective imide ligation of peptide
University) for performing some kinetic experiments.
6e with azidoformiate 13 at pH 2.5 to give imide 14 (64%
Supporting Information Available. Experimental pro-
cedures and characterization data for all compounds.
isolated, Scheme 2).4f,g Moreover, chiral GCÀMS analysis
of the imide ligation product 14 after acid hydrolysis
This material is available free of charge via the Internet
at http://pubs.acs.org.
indicated a D-Tyr content of only 0.28% and thus that
racemization of C-terminal Tyr residue was negligible for
both the exchange and imide ligation process.
The authors declare no competing financial interest.
D
Org. Lett., Vol. XX, No. XX, XXXX