Bis(2-sulfanylethyl)amido peptides enable native chemical ligation at proline and minimize deletion side-product formation

Article abstract of DOI:10.1021/ol402678a

Native chemical ligation of C-terminal peptidyl prolyl alkylthioesters with N-terminal cysteinyl peptides usually exhibits poor kinetic rates compared to other C-terminal amino acid residues. It is shown here that the reaction is accompanied by the formation of a deletion side product which is minimized by using a bis(2-sulfanylethyl)amido (SEA) thioester surrogate at a mildly acidic pH.

Full text of DOI:10.1021/ol402678a

ORGANIC
LETTERS
2013
Vol. 15, No. 21
5516–5519
Bis(2-sulfanylethyl)amido Peptides
Enable Native Chemical Ligation at Proline
and Minimize Deletion Side-Product
Formation
Laurent Raibaut, Phillip Seeberger, and Oleg Melnyk*
UMR CNRS 8161, Pasteur Institute of Lille, 59021 Lille, France
oleg.melnyk@ibl.fr
Received September 16, 2013
ABSTRACT
Native chemical ligation of C-terminal peptidyl prolyl alkylthioesters with N-terminal cysteinyl peptides usually exhibits poor kinetic rates
compared to other C-terminal amino acid residues. It is shown here that the reaction is accompanied by the formation of a deletion side product
which is minimized by using a bis(2-sulfanylethyl)amido (SEA) thioester surrogate at a mildly acidic pH.
Modern protein chemical synthesis usually involves the
sequential ligation of unprotected peptide segments in
aqueous solution.1 Severalchemoselective reactions enable
the formation of a peptide bond between two unprotected
peptide segments.2 Among these reactions, nativechemical
ligation (NCL),2a which is based on the coupling of a
C-terminal peptide thioester with an N-terminal cysteinyl
peptide, is undoubtedly the most popular reaction for
protein total synthesis.3
limitations of this reaction.4 In particular, early studies
identified leucine, threonine, valine, isoleucine, and
proline as the slowest reacting C-terminal amino acid
residues.4a Among these, proline displayed the slowest
kinetic rates and poorest ligation yields (∼10%) after
48 h. Recent studies suggest that the poor reactivity of
peptidyl prolyl thioesters could be due to electronic effects
rather than to steric effects. Indeed, in a trans Xaa-Pro
bond, the π*-orbital of the prolyl thioester carbon can
R
The extensive use of NCL for proteintotal synthesisover
almost 20 years has shed some light on the scope and
interact with the n-orbital of the preceding N carbonyl
oxygen.5 This interaction might reduce the electrophilicity
and accessibility of prolyl thioesters and, thus, explain
the difficulty of forming a Pro-Cys bond using the NCL
reaction.
(1) Raibaut, L.; Ollivier, N.; Melnyk, O. Chem. Soc. Rev. 2012, 41,
7001–7015.
(2) (a) Dawson, P. E.; Muir, T. W.; Clark-Lewis, I.; Kent, S. B.
Science 1994, 266, 776–779. (b) Bode, J. W.; Fox, R. M.; Baucom, K. D.
Angew. Chem., Int. Ed. 2006, 45, 1248–1252. (c) Saxon, E.; Armstrong,
J. I.; Bertozzi, C. R. Org. Lett. 2000, 2, 2141–2143. (d) Nilsson, B. L.;
Kiessling, L. L.; Raines, R. T. Org. Lett. 2000, 2, 1939–1941. (e)
Pattabiraman, V. R.; Bode, J. W. Nature 2011, 480, 471–479. (6) Zhang,
Y.; Xu, C.; Lam, H. Y.; Lee, C. L.; Li, X. Proc. Natl. Acad. Sci. U.S.A.
2013, 110, 6657–6662.
Besides the work of Alewood and co-workers on
C-terminal peptidyl prolyl selenoesters,6 previous studies
in the field examined mainly the reactivity of peptidyl
prolyl thioesters derived from simple alkyl thiols such as
3-mercaptopropionic acid (MPA). We thus explored the
(3) Hackenberger, C. P.; Schwarzer, D. Angew. Chem., Int. Ed. 2008,
47, 10030–10074.
(5) Such an interaction is significant for Xaa-Pro bonds in proteins;
see: Bartlett, G. J.; Choudhary, A.; Raines, R. T.; Woolfson, D. N. Nat.
Chem. Biol. 2010, 6, 615–620.
(6) Durek, T.; Alewood, P. F. Angew. Chem., Int. Ed. 2011, 50,
12042–12045.
(4) (a) Hackeng, T. M.; Griffin, J. H.; Dawson, P. E. Proc. Natl.
Acad. Sci. U.S.A. 1999, 96, 10068–10073. (b) Pollock, S. B.; Kent,
S. B. H. Chem. Commun. 2011, 47, 2342–2344. (c) Dang, B.; Kubota,
T.; Mandal, K.; Bezanilla, F.; Kent, S. B. J. Am. Chem. Soc. 2013, 135,
11911–11919.
r
10.1021/ol402678a
Published on Web 10/11/2013
2013 American Chemical Society

reactivity of other types prolyl thioesters in NCL reactions
with the aim to solve this difficult problem.
Cys peptide CILKEPVHGV-NH2 4b, showing little influ-
ence of the peptide length or composition on the occurrence
of this side reaction (Table 1, compare entries 1 and 3; see
also Figure 1).
In this study we examined the reactivity of bis(2-
sulfanylethyl)amido (SEA)7 peptides 1, thiazolidine thioe-
ster (TT) peptides8 2, and MPA-derived peptide thioesters
3 (Scheme 1). Peptides 1À3 were produced starting from
SEA polystyrene resin using Fmoc-SPPS (Scheme 1).
Previous work showed that SEA peptides 1 featuring a
C-terminal residue other than proline could be converted
efficiently into thiazolidine thioester peptides8 or MPA
thioesters9 at 37 °C using an excess of MPA or glyoxylic
acid respectively at acidic pH. However, these conditions
were unsuccessful with proline as the C-terminal residue.
We reasoned that this lack of reactivity might be due to
a slow N,S-acyl shift of the prolyl SEA group at 37 °C
compared to other amino acid residues. Interestingly,
heating the mixture at 65 °C to accelerate N,S-acyl shift
solved the problem and cleanly yielded thiazolidine thioe-
ster peptide 2b or MPA thioesters 3aÀd in good yield.
Scheme 2. NCL with Peptide Prolyl Thioesters 1À3
Scheme 1. Synthesis of Peptide Prolyl Thioesters 1À3
The formation of deletion side-product 6 might proceed
through the transient peptidyl diketopiperazine intermedi-
ate 7 (Scheme 2). Displacement of the diketopiperazine
moiety by 4-mercaptophenylacetic acid (MPAA) used
in excess during the NCL reaction might furnish the
arylthioester 9, which is expected to react with Cys peptide
4 to yield the deletion side-product 6. This mechanism is
supported by (i) the detection of the diketopiperazine 8 by
LC-MS in the ligation mixtures (see Figures S5, S6 in the
Supporting Information), (ii) the ability of imides of type 7
to react with thiols,10 and (iii) the propensity of Xaa-Pro
dipeptidyl esters to form diketopiperazines.11
The other experiments described in Table 1 show that
with MPA thioesters 3 the formation of the deletion side-
product 6 (i) decreases with the bulkiness of the Xaa
residue (Table 1, entry 3 for Ala, entry 9 for Gly, entry 11
for Ile), Gly being particularly prone to deletion side-
product 6 formation, and (ii) increases with the pH of the
reaction since the amount of side-product 6b formed at pH
7.8 is twice those observed at pH 6.8 (Table 1, entries 2À4).
We next examined the reactivity of SEA peptides 1bÀd
(Xaa = Ala, Gly, Ile). The study of the effect of the pH on
the rate of ligation of prolyl SEA peptide 1b with Cys
peptide 4b reveals that SEA ligation is accelerated signifi-
cantly by decreasing the pH from 7.1 to 5.5 (Figure 1d).12
We thus reasoned that decreasing the pH of the ligation
with prolyl SEA peptides would increase the rate of the
We first examined the ligation of MPA thioesters 3 with
N-terminal cysteinyl peptides 4 since no detailed kinetic
data are available in the literature for the NCL reaction
with peptidyl prolyl alkylthioesters (Scheme 2). The use of
standard conditions for the NCL reaction yielded the target
peptide 5 but also unexpectedly two amino acid deletion
side-product 6 sometimes in significant amounts (Scheme 2).
In particular, formation of peptide LYRCFRANK 6a
(∼8.8% by HPLC; see Figure S8 in the Supporting
Information) was observed with model thioester LYRAP-
MPA 3a in reaction with CFRANK 4a (Table 1, entry 1).
Similar kinetic rates and the amount of deletion side-product
ILKEPVHCILKEPVHGV-NH2 6b were observed during
the ligation of thioester peptide ILKEPVHAP-MPA 3b with
(7) (a) 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. (b) Ollivier, N.; Dheur, J.;
Mhidia, R.; Blanpain, A.; Melnyk, O. Org. Lett. 2010, 12, 5238–5241. (c)
Hou, W.; Zhang, X.; Li, F.; Liu, C. F. Org. Lett. 2011, 13, 386–389.
(8) (a) Dheur, J.; Ollivier, N.; Melnyk, O. Org. Lett. 2011, 13, 1560–
1563. (b) El-Mahdi, O.; Melnyk, O. Bioconjugate Chem. 2013, 24, 735–
765.
(10) Blanco-Canosa, J. B.; Dawson, P. E. Angew. Chem., Int. Ed.
2008, 47, 6851–5.
(11) (a) Goolcharran, C.; Borchardt, R. T. J. Pharm. Sci. 1998, 87,
283–8. (b) Fischer, P. M. J. Pept. Sci. 2003, 9, 9–35.
(12) Decreasing the pH from 7.1 to 5.5 accelerated also the ligation of
peptide ILKEPVHGA-SEA 1e with Cys peptide 4b, but reduced the rate
of NCL reaction between peptide ILKEPVHGA-MPA 3e and 4b (see
Figure S9 in the Supporting Information).
(9) Dheur, J.; Ollivier, N.; Vallin, A.; Melnyk, O. J. Org. Chem. 2011,
76, 3194–3202.
Org. Lett., Vol. 15, No. 21, 2013
5517

Table 1. Kinetic Data for the Ligation of Peptidyl Prolyl Thioesters 1À3a
peptide prolyl
temp
HPLC yield HPLC yield ligation product deletion product
entry
thioester
Cys peptide2
4a: CFRANK
(°C) pH
for 5 (%)
for 6 (%)
5: t1/2 (h)
6: t1/2 (h)
1
3a LYRAP-MPA
37
37
37
37
65
65
37
65
37
65
37
65
65
7.1 67.6
6.8 65
8.8
4.2
6.8
8.6
16.9
48
5a: 20.2
5b:17.1
5b: 16.9
5b: 18.3
5b: ndc
5b: nd
6a: 25.7
6b: 14
2
3b ILKEPVHAP-MPA 4b: CILKEPVHGV-NH2
3b ILKEPVHAP-MPA 4b
3b ILKEPVHAP-MPA 4b
3b ILKEPVHAP-MPA 4b
3b ILKEPVHAP-MPA 4b
2b ILKEPVHAP- TTX 4b
1b ILKEPVHAP- SEA 4b
3c ILKEPVHGP-MPA 4b
1c ILKEPVHGP- SEA 4b
3d ILKEPVHIP-MPAd 4b
1d ILKEPVHIP- SEAd 4b
1d ILKEPVHIP- SEAd 4a
3
7.1 60
6b: 16.1
6b: 12.8
6b: nd
4
7.8 66
5
5.5 62.6
7.1 28
6
6b: nd
7
7.1 68.4
5.5 76.5 (36b)
7.1 38
3.25
3.3
22.4
15
5b: 2.84
5b: 12.4
5c: nd
6b: 2.63
6b: 19.8
6b: nd
8
9
10
11
12
13
5.5 63.7
7.2 75.2
5.5 78.3
5.5 85.8 (42b,d
5c: nd
6b: nd
<1
5d: 15.2
5d: 12.7
5e: nd
6b: nd
<1
6b: nd
6c: nde
)
<1
a All the ligation reactions showed typical first-order kinetic profiles (see Figure 1). b HPLC purified. c nd: not determined. d D-Pro content: 0.19% for
1d, 0.58% for 3d, and 1.1% for peptide ILKEPVHIPCFRANK 5e (see Supporting Information). e 6c corresponds to peptide ILKEPVHCFRANK.
Figure 1. Reaction of SEA peptide 1b, TT peptide 2b, or MPA peptide 3b with Cys peptide 4b. (a, b, d) Kinetic data (HPLC, UV
detection at 215 nm). The data were modeled using the first-order rate law (continuous curves); see also Figure S7 in the Supporting
Information. (c) HPLC profiles (C18 column, UV detection at 215 nm).
5518
Org. Lett., Vol. 15, No. 21, 2013

ligation, while minimizing the base-catalyzed process lead-
ing to deletion side-product 6 formation. Since ligation of
prolyl SEA peptides 1 with Cys peptide 4 showed poor
kinetic rates at 37 °C, the ligations were performed at 65 °C
to accelerate the N,S-acyl shift process as in the case for the
synthesis of TT and MPA thioesters 2 and 3. Satisfactorily,
the reaction of prolyl SEA peptides 1b,c (Xaa = Ala, Gly)
with Cys peptide 4b resulted in higher ligation yields
and lower deletion side-product levels in comparison
with MPA analogs 3b,c (Table 1, compare entries 3,9
and 8,10). The data in Table 1 show that, as in the case
for MPA thioesters 3, the amount of deletion side-product
6 with prolyl SEA peptides 1 increased in the order
Ile < Ala < Gly.
Scheme 3. SEA Kinetically Controled Ligation
In a control experiment, we reacted MPA thioester
3b with Cys peptide 4b at pH 5.5 or 7.1 at 65 °C, the
temperature used for prolyl SEA peptides. At pH 7.1, the
deletion side-product 6b (48%) dominated over the liga-
tion product 5b (28%, Table 1, entry 6). At pH 5.5, it still
represented a 17% yield (Table 1, entry 5) compared to
only a 3.3% yield with the SEA peptide analog 1b (Table 1,
entry 8). Overall, these data show that SEA peptides 1 are
significantly more resistant to deletion side-product for-
mation than MPA thioesters 3 and therefore constitute a
potential alternative to the formation of Pro-Cys peptide
bonds.
Finally, the NCL reaction of thiazolidine thioester pep-
tide 2b with Cys peptide 4b proceeded 6 times faster (t1/2
2.8 h) than for MPA thioester 3b (t1/2 16.9 h). The fact that
the yield of the deletion side-product was only 3.3%
suggests that thioesters of type 2 are also potentially useful
derivatives for the formation of Pro-Cys peptide bonds.
In the last part of this work, we sought to exploit the fact
that prolyl SEA peptides are slow reacting systems at 37 °C
in comparison with otherC-terminalresidues for designing
a kinetically controlled one-pot three-segment assembly
process (Scheme 3).13 The first ligation step between alanyl
SEA peptide 1e and Cys peptide 1f was performed at 37 °C
and cleanly yielded prolyl SEA peptide 10 (see Figure S11
in the Supporting Information). In particular, cyclization
of peptide segment 1f was not observed. Then, the third
Cys peptide segment 4b was added to the mixture and
the temperature was raised to 65 °C to trigger the second
ligation step and the formation of target peptide 11 (29%
overall yield).
Inconclusion, wereport here thatadeletionside product
is formed during the NCL reaction of peptidyl prolyl
thioesters with cysteinyl peptides. This side reaction is
significant for peptide thioesters derived from MPA which
are often used in the field of protein total synthesis. It can
be minimized with MPA thioesters by lowering the pH
down to 6.8. The level of side-product formation and the
rate of ligation vary also significantly with the peptide
thioester structure. Thiazolidine thioester peptides are
significantly more reactive then MPA thioesters at 37 °C,
but significantly less susceptible to deletion side-product
formation. Prolyl SEA peptides are slow reacting systems
at 37 °C but ligate efficiently at pH 5.5 and 65 °C. In this
case, the level of deletion side product is reduced compared
to MPA-derived thioesters. The dependence of prolyl SEA
peptide reactivity on the temperature opens novel synthetic
opportunities as illustrated by the design of a kinetically
controlled one-pot three-segment assembly process.
Supporting Information Available. Experimental pro-
cedures and characterization data for all compounds.
This material is available free of charge via the Internet
at http://pubs.acs.org.
(13) (a) Bang, D.; Pentelute, B. L.; Kent, S. B. Angew. Chem., Int. Ed.
2006, 45, 3985–8. (b) Zheng, J. S.; Cui, H. K.; Fang, G. M.; Xi, W. X.;
Liu, L. ChemBioChem 2010, 11, 511–5. (c) Sato, K.; Shigenaga, A.;
Tsuji, K.; Tsuda, S.; Sumikawa, Y.; Sakamoto, K.; Otaka, A. Chem-
BioChem 2011, 12, 1840–4.
The authors declare no competing financial interest.
Org. Lett., Vol. 15, No. 21, 2013
5519