Investigation of Cysteine as an Activator of Side-Chain N→S Acyl Transfer and Tail-to-Side-Chain Cyclization

Article abstract of DOI:10.1055/s-0036-1590797

N→S Acyl transfer is a popular method for the postsynthesis production of peptide C _α -thioesters for use in native chemical ligation and for the synthesis of head-to-tail cyclic peptides. Meanwhile thioester formation at the side chain of aspartic or glutamic acids, leading to tail-to-side-chain-cyclized species, is less common. Herein we explore the potential for cysteine to function as a latent thioester when appended to the side chain of glutamic acid. Initial insights gained through study of C-terminal β-alanine as a model for the increased chain length were ultimately applied to peptide macrocyclization. Our results emphasize the increased barrier to acyl transfer at the glutamic acid side chain and indicate how a slow reaction, facilitated by cysteine itself, may be accelerated by fine-tuning of the stereoelectronic environment..

Full text of DOI:10.1055/s-0036-1590797

SYNLETT
0
9
3
6
-
5
2
1
4
1
4
3
7
-
2
0
9
6
© Georg Thieme Verlag Stuttgart · New York
2017, 28, A–F
cluster
A
en
D. J. Castillo-Pazos, D. Macmillan
Cluster
Synlett
Investigation of Cysteine as an Activator of Side-Chain N→S Acyl
Transfer and Tail-to-Side-Chain Cyclization
Durbis J. Castillo-Pazos
SH
NH
O
SH
H
Derek Macmillan*
0
0
0
0
0-
0
0
3
4-
4
1
2
0-
3
1
4
N
CONH2
O
R
HN
X
HN
CO2H
Department of Chemistry, University College London,
20 Gordon Street, London, WC1H 0AJ, UK
d.macmillan@ucl.ac.uk
O
O
HN
O
O
O
HN
N
S
N
X
O
NH
S
O
N
Published as part of the Cluster Recent Advances in
Protein and Peptide Synthesis
NH
O
N
N
H
?
OH
H
N
O
N
O
H
R = H/ Me
X = H/ F
Received: 03.05.2017
Accepted after revision: 22.05.2017
Published online: 14.07.2017
mainly studied the N→S acyl transfer activity of native pep-
tide sequences, where cysteine itself promotes the reaction
(Scheme 1, b).
DOI: 10.1055/s-0036-1590797; Art ID: st-2017-w0322-c
Abstract N→S Acyl transfer is a popular method for the postsynthesis
production of peptide Cα-thioesters for use in native chemical ligation
and for the synthesis of head-to-tail cyclic peptides. Meanwhile thio-
ester formation at the side chain of aspartic or glutamic acids, leading
to tail-to-side-chain-cyclized species, is less common. Herein we ex-
plore the potential for cysteine to function as a latent thioester when
appended to the side chain of glutamic acid. Initial insights gained
through study of C-terminal β-alanine as a model for the increased
chain length were ultimately applied to peptide macrocyclization. Our
results emphasize the increased barrier to acyl transfer at the glutamic
acid side chain and indicate how a slow reaction, facilitated by cysteine
itself, may be accelerated by fine-tuning of the stereoelectronic envi-
ronment.
Key words cyclic peptides, native chemical ligation, acyl transfer
Scheme 1 a) Head-to tail peptide cyclization enabled by NCL. b) A C-
terminal cysteine residue promotes N→S acyl transfer, providing tran-
sient access to the required thioester for cyclization via NCL. c) Tail-to-
side-chain cyclization using a thioester precursor installed on an amino
acid side chain.
Due to the increasing importance of cyclic peptides as
robust pharmaceutical scaffolds for otherwise metabolical-
ly fragile bioactive peptides,1 a vast number of chemical and
biological methods have been reported for their synthesis.
Despite the multitude of methods accessible to the synthet-
ic chemist,2 chemoselective ligation technologies continue
to lead the way in the quest for scalable and cost-effective
protocols.3 The use of native chemical ligation (NCL,
Scheme 1, a) to effect cyclization of a linear precursor,
adorned with a C-terminal thioester and N-terminal cyste-
ine, is arguably the most common approach since it can be
conducted in water, using unprotected peptides, and em-
ploy precursors of synthetic or biological origin. Postsyn-
thesis N→S acyl transfer is becoming a common method for
production of the required thioester component, and sever-
al devices have been developed around the β-amino thiol
scaffold of cysteine (Cys) to help facilitate it.4 We have
Having demonstrated the cysteine-promoted head-to-
tail cyclization using synthetic5 and biologically6 produced
precursors we turned our attention to more challenging to-
pologies such as the tail-to-side-chain cyclization depicted
in Scheme 1 (c). This mode of cyclization is often found in
lariat (lasso) peptides, which have not yet proved amenable
to chemical synthesis,7 and is particularly interesting be-
cause a side-chain thioester cannot be introduced by using
an intein. Furthermore, conducting thioester formation in
aqueous solution using fully unprotected, or tethered, lariat
precursors may allow the peptide to adopt a more native
conformation prior to cyclization. Melnyk and co-workers
previously showed that acyl transfer, when conducted at
© Georg Thieme Verlag Stuttgart · New York — Synlett 2017, 28, A–F

B
D. J. Castillo-Pazos, D. Macmillan
Cluster
Synlett
the side chain of aspartic acid (Asp) or glutamic acid (Glu),
was slow relative to C-terminal thioester formation, when
using a bis-2-sulfanylethylamino (bis-SEA) functionalized
peptide.8 The difference in reactivity also formed the basis
for kinetically controlled thioester formation.9 We were
keen to establish whether slow thioester formation was
also observed at the side chain of Asp/Glu when cysteine
was used to promote the initial N→S acyl shift and, if so,
what could be done to remedy it.
which served as models for the increased chain length.11
We were pleasantly surprised to find that thioester 6b
could be cleanly formed from β-alanine terminated 6a.
However, formation of 6b was significantly slower when
compared with the corresponding Gly-Cys-OH motif 5a,
taking up to 72 hours to reach >70% conversion at 60 °C.
Although not surprising, since the Melnyk group report-
ed reaction times of 144 hours for thioesterification/cy-
clization (at 37 °C) with a side-chain SEA motif,8 this raised
concerns about the use of cysteine itself as a facilitator N→S
acyl shift when applied to the amino acid side chain and re-
emphasized the important contribution of the α-amino
group of the adjacent nonscissile amide in mediating the
reactivity of native peptide sequences. Keen to reconstitute
glycine-like reactivity at β-alanine-terminated peptide we
additionally explored the use of model peptide 7a contain-
ing α-methyl cysteine because it is more susceptible to thi-
olysis than cysteine itself and, as a commercially available
amino acid, it is simple to introduce relative to common al-
ternatives.12 Furthermore, since the desired dipeptide mo-
tifs that would need to be introduced in order to effect side-
chain activation (e.g., 3) are not naturally occurring there
was no need to retain a strict cysteine scaffold. The model
reaction employing 7a was extremely encouraging showing
almost complete consumption of the starting material
within six hours.
We postulated that the reduction in reaction rate, when
using β-Ala, was likely due to the increased distance be-
tween the reaction site and the proximal electron-with-
drawing amide nitrogen atom. To explore this further we
examined whether reactivity could be restored through in-
troduction of an electronegative atom, such as fluorine,
placed α to the carbonyl group of the scissile amide bond
(Scheme 3). Consequently we prepared fluorinated pep-
tides 8a and 9a. Monofluorinated 8a was prepared as an
epimeric mixture from racemic 2-fluoro-β-alanine. The
epimers were almost completely separated by HPLC, ap-
peared configurationally stable under the reaction condi-
tions, and both behaved similarly in reactions. Analysis of
reaction mixtures after 24 hours indicated that the fluoro-
β-alanine residues installed at the C-terminus indeed accel-
SH
SH
NH2
SH
SH
CO2H
O
HN
NH
CO2H
O
CO2H
O
O
NH
NH
n
N
N
H
H
H2N
CO2H
H2N
CO2H
O
O
1
2
3a (n = 0)
3b (n = 1)
4
Figure 1 Desired peptide motifs derived from Asp (1) or Glu (2) could
be introduced using e.g. 3a,b and an expanded genetic code. Dipeptide
4 has already been genetically encoded, enabling side-chain ubiquitina-
tion using NCL.
Specifically, we set out to examine the N→S acyl transfer
behavior of branched dipeptide motifs 1 and 2 (Figure 1)
within the context of a synthetic peptide sequence. If suc-
cessful the hope would be that these should be genetically
encodeable via building blocks such as 3a,b. Interestingly
several branched dipeptides including 4,10 which enables
site-specific ubiquitination via NCL with ubiquitin thio-
esters, have already been genetically encoded in bacteria
employing orthogonal pyrrolysyl tRNA/tRNA synthetase
pairs suggesting that this route could well be plausible.
To establish how dramatically the additional methylene
groups between the side-chain carboxyl group of, for exam-
ple, glutamic acid and the peptide backbone influence the
reactivity of cysteine towards N→S acyl shift, we first com-
pared the reactivity of glycine terminated model peptide 5a
with β-alanine terminated peptides 6a and 7a (Scheme 2),
Scheme 2 Comparison of MESNa thioester formation across Gly-Cys-OH (H-MEELKYSGC-OH, 5a), βAla-Cys-OH (H-MEELYKSβAC-OH, 6a), and βAla-
αMeCys-OH (H-MEELYKSβAαMeC-OH, 7a) terminated peptides after 6 h at 60 °C.
© Georg Thieme Verlag Stuttgart · New York — Synlett 2017, 28, A–F

C
D. J. Castillo-Pazos, D. Macmillan
Cluster
Synlett
erated the rate of thioester formation relative to β-alanine
itself. However, what also became evident was that thio-
esters 8b and 9b were unstable and more prone to hydroly-
sis. After 24 hours the initially formed thioester product 8b
was almost completely hydrolyzed at 60 °C. Hydrolysis
could be reduced, but not abolished, by conducting the re-
action at 50 °C or 40 °C (Scheme 3, a) but, when using 9a, a
thioester product could not be observed at all. However, the
reactive nature of the 2-fluoro-β-Ala-Cys-OH motif
prompted us to investigate the reaction further since we
were confident that, by intercepting the intermediate with
a nucleophilic species such as hydrazine or creating a more
stable product via in situ cyclization,5d we could ultimately
create a potentially useful process.
Then, as expected, 8a and 9a formed acyl hydrazides 8c
and 9c cleanly and near quantitatively at 50 °C after 24
hours when the reaction mixture was supplemented with
5% w/v hydrazinium acetate,13 appearing superior both β-
Ala-Cys-OH and Gly-Cys-OH terminated peptides (Scheme
3, b). Curious as to whether the successful outcome indicat-
ed that cyclization may prove more productive than thio-
ester formation a model peptide corresponding to an ana-
logue of the plasmin inhibitory peptide, agardhipeptin A,14
was readily assembled. As with 8a, the racemic monofluori-
nated β-Ala residue was employed but, in contrast to 8a, the
agardhipeptin A analogue 10a did not yield an easily sepa-
rable mixture of epimers. Since this was considered to be of
little consequence for the purpose of the experiment the
progress of the reaction was monitored using the summed
peak areas of each epimer for the starting material and
product. In the cyclization reaction only the linear precur-
sor containing a monofluorinated β-Ala residue gave rise to
a cyclic peptide (Scheme 4) whereas the precursor contain-
ing the difluorinated β-Ala showed only hydrolysis under
identical reaction conditions.
Overall, the results of model experiments using β-ala-
nine and fluorinated derivatives were extremely enlighten-
ing showing, not only how an additional methylene group
placed between the scissile amide bond and the adjacent α-
amino group reduced reactivity, but additionally how reac-
tivity could be restored by introduction of a single fluorine
atom. Whilst thioesters 8b and 9b appeared unstable, we
demonstrated how they could be efficiently trapped by
added nucleophiles such as hydrazine, which may well find
new applications in the future. However, regardless of how
encouraging the cyclization of 10a appeared, we were not
keen to progress the monofluorinated β-alanine derivatives
as acyl transfer precursors. Since the fluorine atom would
be retained by the product, a scalable and enantioselective
synthesis of a corresponding monofluorinated glutamic/as-
partic acid would need to be conducted. Although enantio-
selective routes to monofluorinated glutamic acids exist
from hydroxyproline,15 pyroglutamic acid derivatives,16 and
more recently from the γ-aldehyde,17 their synthesis is dif-
ficult to scale, employing expensive and/or toxic metals and
catalysts. Furthermore, having already achieved acceptable
results with 6a and improved results with 7a we felt suffi-
ciently confident to explore the reactivity of the γ-glutamyl
cysteine containing peptides themselves. Due to the poten-
tial for aspartimide formation upon activation of aspartic
acid derivative 1 we initially focused on the synthesis and
reactivity of the γ-glutamyl cysteine motif 2.
Previous investigations employing cysteine as the sole
facilitator of acyl transfer indicated that, for optimal reac-
tivity, the branching cysteine residue should possess a free
carboxyl group.18
Scheme 3 Cysteine-promoted thioester and hydrazide formation at glycine (5a), β-alanine (6a), and fluorinated analogues (8a, 9a). a) After 24 h at 50
°C in the absence of hydrazine. b) After 24 at 50 °C in the presence of hydrazine. * Corresponds to the hydrolysis product as identified by LC–MS.
© Georg Thieme Verlag Stuttgart · New York — Synlett 2017, 28, A–F

D
D. J. Castillo-Pazos, D. Macmillan
Cluster
Synlett
Scheme 4 Cyclization of agardhipeptin A analogues in 0.7 M MESNa at 60 °C from linear precursors H-CHGWPWGX-OH, where X = Gly (10a) or 2-
fluoro β-alanine (11a)
Consequently it was important to differentiate the α-
carboxyl groups of cysteine and glutamic acid in the di-
peptide building block such that the α-carboxyl group of
glutamic acid could be unmasked for solid-phase peptide
synthesis while the α-carboxyl group of cysteine remained
protected (Scheme 5). Suitably protected dipeptide build-
ing blocks 16a,b were readily prepared from available H-
Cys-OH, or H-αMeCys-OH, and Fmoc-Glu-OAll. Briefly, the
free amino acids were converted into the STrt/Nα-Fmoc-
protected derivatives 13a,b by overnight reaction with tri-
tyl chloride in DMF followed by reaction of the crude prod-
uct with Fmoc succinimide in a 1:1 mixture of water and
dioxane in the presence of excess Na2CO3 [Fmoc-Cys(Trt)-
OH is also commercially available].12 Introduction of the
cysteinyl Cα tert-butyl ester was next introduced according
to the method reported by Schmidt,19 using tert-butyl 2,2,2-
trichloroacetimidate (TBTA) in the presence of BF3·OEt2 to
afford fully the protected amino acid. Subsequent Fmoc
deprotection with piperidine in CH2Cl2 liberated the free α-
amino group for coupling to the γ-carboxyl of commercially
available Fmoc-Glu-OAll, which was efficiently completed
using pyBOP as coupling reagent in the presence of DIPEA.20
NH
, BF3⋅OEt2
i)
STrt
STrt
SH
Cl3C
OtBu
i) Trt-Cl, DMF
ii) Fmoc-OSu
R
R
R
H3N
12a, R = H
ii)
piperidine, MeOH
FmocHN
CO2H
H2N
CO2tBu
CO2H
13a, R = H
13b, R = Me
14a, R = H
14b, R = Me
12b, R = Me
STrt
STrt
R
R
HN
CO2tBu
HN
CO2tBu
O
O
Fmoc-Glu-OAll
pyBOP,
Pd(PPh3)4 (10 mol%)
morpholine (1.5 equiv)
DIPEA
O
OH
FmocHN
FmocHN
O
O
15a, R = H (85%)
15b, R = Me (53%)
16a, R = H (88%)
16b, R = Me (61%)
Scheme 5 Synthesis of suitably protected branched dipeptides 16a and 16b
© Georg Thieme Verlag Stuttgart · New York — Synlett 2017, 28, A–F

E
D. J. Castillo-Pazos, D. Macmillan
Cluster
Synlett
Scheme 6 a) Cyclization of branched precursor peptides 17 and 18 at 50 °C in the presence of 0.7 M MESNa. b) HPLC monitoring of the cyclization of
18 to afford 19 over 24 h at 50 °C. The peak marked with an asterisk is the MESNa thioester intermediate.
The desired building blocks, Fmoc-Glu[Cys(Trt)OtBu]-OH
(16a) and Fmoc-Glu[αMeCys(Trt)OtBu]-OH (16b) were ulti-
mately prepared upon palladium(0)-catalyzed cleavage of
the allyl ester from 15a,b.21 Whether using Cys or αMeCys
identical procedures could be used throughout and, with
the exception of coupling to the glutamic acid derivative
and allyl deprotection, yields for the αMeCys derivatives
were comparable.
as our own examination of β-Ala terminated model pep-
tides this was not particularly unexpected, however, reac-
tivity could be reconstituted by employing more reactive
acyl-transfer precursors 7a and 18 containing recently de-
scribed α-methylcysteine. Interestingly, researchers have
mainly sought to elevate the reactivity of thioester precur-
sors towards N→S acyl transfer by modification of the cys-
teine-like C-terminus. In contrast here we have shown that
N→S acyl transfer can also be enhanced by electronic acti-
vation of the adjacent residue offering new opportunities,
more generally, to promote controlled fragmentation of
peptides under mild conditions. Installation of a single α-
fluorine atom conferred Gly-Cys-like properties on other-
wise sluggish β-Ala-Cys-terminated peptides, whereas in-
troduction of a geminal difluoro motif appeared too reac-
tive to support thioester formation. Nevertheless fluorinat-
ed amino acid building blocks (FlXaa) may yet serve as
useful promoters of N→S acyl transfer across FlXaa-Cys mo-
tifs in future.
Overall the selective activation of the glutamic acid side
chain using unprotected peptides in aqueous solution, and
in the context of McJ25, paves the way to a potentially via-
ble route to lariat peptides. Selective acyl-transfer reactions,
likely working in combination with a tethering strategy to
template the lariat fold,24 may ultimately bring this fascinat-
ing class of peptides under the routine command of the
synthetic chemist.
Dipeptides 16a and 16b were next incorporated in place
of Glu8 into the nine N-terminal residues of antimicrobial
lariat peptide microcin J25 to afford 17 and 18, respectively
(Scheme 6). Although this partial sequence is incapable of
forming the characteristic lariat structure of microcin J25,22
it served as a suitable model system to test the tail-to-side-
chain cyclization. In agreement with preceding results cy-
clization of 17 similarly showed that the desired peptide
macrocycle could be formed in greater that 70 % yield (as
judged by HPLC) over 72 hours at 50 °C, validating the use
of β-alanine as a model system. While encouraging, we ulti-
mately believed that the long reaction time at elevated tem-
perature would be generally detrimental to the formation
of side-chain thioesters and tail-to-side-chain cyclization
reactions in more complex samples. Encouragingly peptide
18 also behaved similarly to the preceding model experi-
ments. While the reaction to form 1923 took nearer 24
hours to reach 92% conversion at 50 °C (compared with ca.
six hours at 60 °C in β-Ala experiments) it appeared that
thioester formation had advanced significantly within six
hours, as indicated by the appearance and subsequent con-
sumption of the MESNa thioester intermediate (marked
with an asterisk in Scheme 6, b). It is likely that NCL cycliza-
tion under nonideal conditions (pH <7) is the reason for the
accumulation of the MESNa thioester.
Acknowledgment
The authors acknowledge Dr. John Offer for helpful discussions and
providing a sample of Fmoc-αMeCys(Trt)-OH.
In conclusion cysteine, when installed at the side-chain
carboxyl group of glutamic acid, was found to be a willing
facilitator of N→S acyl transfer, albeit with reduced efficien-
cy when compared with presentation at the C-terminus.
Based on previous work by Melnyk and co-workers, as well
Supporting Information
Supporting information for this article is available online at
https://doi.org/10.1055/s-0036-1590797.
S
u
p
p
ortiInfogrmoaitn
S
u
p
p
o
nrtogI
f
rmoaitn
© Georg Thieme Verlag Stuttgart · New York — Synlett 2017, 28, A–F

F
D. J. Castillo-Pazos, D. Macmillan
Cluster
Synlett
References and Notes
(15) Hudlický, M. J. Fluorine Chem. 1993, 60, 193.
(16) Konas, D. W.; Coward, J. K. J. Org. Chem. 2001, 66, 8831.
(17) Buckingham, F.; Kirjavainen, A. K.; Forsback, S.; Krzyczmonik,
A.; Keller, T.; Newington, I. M.; Glaser, M.; Luthra, S. K.; Solin, O.;
Gouverneur, V. Angew. Chem. Int. Ed. 2015, 13564.
(18) Richardson, J. P.; Chan, C.-H.; Blanc, J.; Saadi, M.; Macmillan, D.
Org. Biomol. Chem. 2010, 8, 1351.
(19) Haag, T.; Hughes, R. A.; Ritter, G.; Schmidt, R. R. Eur. J. Org. Chem.
2007, 6016.
(20) Synthesis of Dipeptides 15a,b
(1) (a) Northfield, S. E.; Wang, C. K.; Schroeder, C. I.; Durek, T.; Kan,
M.-W.; Swedberg, J. E.; Craik, D. J. Eur J. Med. Chem. 2014, 77,
248. (b) Poth, A. G.; Chan, L. Y.; Craik, D. J. Pept. Sci. 2013, 100,
480. (c) Young, T. S.; Young, D. D.; Ahmad, I.; Louis, J. M.;
Benkovic, S. J.; Schultz, P. G. Proc. Natl. Acad. Sci. U.S.A. 2011,
108, 11052.
(2) (a) White, C. J.; Yudin, A. K. Nat. Chem. 2011, 3, 509. (b) Yudin, A.
K. Chem. Sci. 2015, 6, 30.
(3) (a) Camarero, J. A.; Cotton, G. J.; Adeva, A.; Muir, T. W. J. Peptide
Res. 1998, 51, 303. (b) Tam, J. P.; Lu, Y.-A. Tetrahedron Lett. 1997,
38, 5599. (c) Tavassoli, A.; Benkovic, S. J. Nat. Protoc. 2007, 2,
1126. (d) Jagadish, K.; Gould, A.; Borra, R.; Majumder, S.;
Mushtaq, Z.; Shekhtman, A.; Camarero, J. A. Angew. Chem. Int.
Ed. 2015, 54, 8390. (e) Nguyen, G. K. T.; Qiu, Y.; Cao, Y.; Hemu,
X.; Liu, C.-F.; Tam, J. P. Nat. Protoc. 2016, 11, 1977. (f) Palei, S.;
Mootz, H. D. ChemBioChem. 2016, 17, 378. (g) Jia, X.; Kwon, S.;
Wang, C.-I. A.; Huang, Y.-H.; Chan, L. Y.; Tan, C. C.; Rosengren, K.
J.; Mulvenna, J. P.; Schroeder, C. I.; Craik, D. J. J. Biol. Chem. 2014,
289, 6627. (h) Harris, K. S.; Durek, T.; Kaas, Q.; Poth, A. G.;
Gilding, E. K.; Conlan, B. F.; Saska, I.; Daly, N. L.; van der
Weerden, N. L.; Craik, D. J.; Anderson, M. A. Nat. Commun. 2015,
6, 10199. (i) Iwane, Y.; Hitomi, A.; Murakami, H.; Katoh, T.; Goto,
Y.; Suga, H. Nat. Chem. 2016, 8, 317. (j) Passioura, T.; Suga, H.
Chem. Commun. 2017, 53, 1931.
(4) (a) Macmillan, D. Synlett 2017, 28, in press; DOI: 10.1055/s-
0036-1588789. (b) Melnyk, O.; Agouridas, V. Curr. Opin. Chem.
Biol. 2014, 22, 137. (c) Zheng, J.-S.; Tang, S.; Huang, Y.-C.; Liu, L.
Acc. Chem. Res. 2013, 46, 2475.
(5) (a) Adams, A. L.; Macmillan, D. J. Pept. Sci. 2013, 19, 65. (b) Chen,
W.; Kinsler, V. A.; Macmillan, D.; Di, W.-L. PLoS One 2016, 11,
e0166268. (c) Macmillan, D.; De Cecco, M.; Reynolds, N. L.;
Santos, L. F. A.; Barran, P. E.; Dorin, J. R. ChemBioChem. 2011, 12,
2133. (d) Cowper, B.; Shariff, L.; Chen, W.; Gibson, S. M.; Di, W.-
L.; Macmillan, D. Org. Biomol. Chem. 2015, 13, 7469.
(6) (a) Cowper, B.; Craik, D. J.; Macmillan, D. ChemBioChem. 2013,
14, 809. (b) Shariff, L.; Zhu, Y.; Cowper, B.; Di, W.-L.; Macmillan,
D. Tetrahedron 2014, 70, 7675.
(7) (a) Hegemann, J. D.; Zimmermann, M.; Xie, X.; Marahiel, M. A.
Acc. Chem. Res. 2015, 48, 1909. (b) Al Toma, R. S.; Kuthning, A.;
Exner, M. P.; Denisiuk, A.; Ziegler, J.; Budisa, N.; Süssmuth, R. D.
ChemBioChem. 2015, 16, 503.
(8) Boll, E.; Dheur, J.; Drobecq, H.; Melnyk, O. Org. Lett. 2012, 14,
2222.
(9) (a) Boll, E.; Drobecq, H.; Lissy, E.; Cantrelle, F.-X.; Melnyk, O. Org.
Lett. 2016, 18, 3842. (b) Raibaut, L.; Drobecq, H.; Melnyk, O. Org.
Lett. 2015, 17, 3636.
Cysteine derivative 14a (0.437 g, 1.04 mmol) was dissolved in
anhydrous CH2Cl2 (6.0 mL). Fmoc-Glu-OAll (0.512 g, 1.25 mmol)
and pyBOP (0.595 g, 1.5 mmol) were added, followed by DIPEA
(0.362 mL, 2.1 mmol), and the reaction mixture was stirred at
r.t. for 1 h. The reaction mixture was then diluted with EtOAc
(50 mL) and washed with sat. aq KHSO4 (1 × 15 mL) and sat. aq
NaHCO4 (2 × 15 mL). The organic phase was separated, dried
(MgSO4), and evaporated to afford the crude product as a white
foam. Purification by flash column chromatography over silica,
eluent toluene–EtOAc (5:1) afforded 15a (0.719 g; 85%) as a
white foam.
Dipeptide 15a
1H NMR (500 MHz, CDCl3): δ = 7.76 (2 H, d, J = 8.0 Hz, 2 × ArH),
7.62 (2 H, d, J = 7.4 Hz, 2 × ArH), 7.44–7.15 (ca. 19 H, m (over-
lapped by CDCl3 signal), ArH ), 6.15 (1 H, d, J = 7.6 Hz, CH allyl),
5.93–5.85 (1 H, m, CH allyl), 5.74 (1 H, d, J = 8.0 Hz, CH allyl),
5.35–5.24 (2 H, 2 × d, 2 × NH), 4.64 (2 H, s (br), CH2-allyl), 4.52–
4.50 (1 H, m, CHα), 4.43–4.38 (3 H, m, CHα, CH2-Fmoc), 4.23 (1 H,
t, J = 7.0 Hz, CH-Fmoc), 2.79–2.54 (2 H, m CH2β-cys), 2.33–1.97
(4 H, m, CH2β-Glu and CH2γ-Glu), 1.45 (9 H, s, tBu). ESI+-MS: m/z
calced for C49H50N2O7S: 810.33; found: 811.5 Da [MH]+ and
833.5 Da [MNa]+.
Dipeptide 15b
Procedure as above, 53% isolated as a white foam. 1H NMR (300
MHz, CDCl3): δ = 7.78 (2 H, d, J = 7.5 Hz, 2 × ArH), 7.61 (2 H, d, J =
7.4 Hz, 2 × ArH), 7.42–7.17 (ca. 19 H, m (overlapped by CDCl3
signal), ArH ), 6.39 (1 H, s (br), CH allyl), 5.86–5.83 (2 H, m, 2 ×
CH allyl), 5.35–5.21 (2 H, 2 × d, 2 × NH), 4.63 (2 H, d (br), CH2-
allyl) 4.42–4.37 (3 H, m, CHα-Glu, CH2-Fmoc) 4.23 (1 H, t (br),
CH-Fmoc), 3.04 (1 H, d, J = 11.3 Hz, half CH2β-αMeCys), 2.60 (1
H, d, J = 11.3 Hz, half CH2β-αMeCys), 2.36–2.05 (4 H, m, CH2β-Glu
and CH2γ-Glu), 1.45 (12 H, s (br), CH3, tBu). ESI+-MS: m/z calcd
for C50H52N2O7S: 824.35; found: 825.5 Da [MH]+ and 847.5 Da
[MNa]+.
(21) Hang, J.; Li, H.; Deng, L. Org. Lett. 2002, 4, 3321.
(22) Bayro, M. J.; Mukhopadhyay, J.; Swapna, G. V. T.; Huang, J. Y.;
Ma, L.-C.; Sineva, E.; Dawson, P. E.; Montelione, G. T.; Ebright, R.
H. J. Am. Chem. Soc. 2003, 125, 12382.
(10) Nguyen, D. P.; Elliott, T.; Holt, M.; Muir, T. W.; Chin, J. W. J. Am.
Chem. Soc. 2011, 133, 11418.
(11) See Supporting Information for full experimental details.
(12) Burlina, F.; Papageorgiou, G.; Morris, C.; White, P. D.; Offer, J.
Chem. Sci. 2014, 5, 766.
(13) (a) Adams, A. L.; Cowper, B.; Morgan, R. E.; Premdjee, B.;
Caddick, S.; Macmillan, D. Angew. Chem. Int. Ed. 2013, 52, 13062.
(b) Li, Y.-M.; Li, Y.-T.; Pan, M.; Kong, X.-Q.; Huang, Y.-C.; Hong,
Z.-Y.; Liu, L. Angew. Chem. Int. Ed. 2014, 53, 2198. (c) Zheng, J.-S.;
Tang, S.; Guo, Y.; Chang, H.-N.; Liu, L. ChemBioChem. 2012, 13,
542. (d) Fang, G.-M.; Li, Y.-M.; Shen, F.; Huang, Y.-C.; Li, J.-B.; Lin,
Y.; Cui, H.-K.; Liu, L. Angew. Chem. Int. Ed. 2011, 50, 7645.
(14) Shin, H. J.; Matsuda, H.; Murakami, M.; Yamaguchi, K. Tetrahe-
dron 1996, 52, 13129.
(23) Lyophilized peptide 18 was dissolved in ddH2O to a final con-
centration of 2 mg/mL; 0.5 mL was transferred to a sterile
Eppendorf tube and water (0.3 mL) was added followed by 1 M
sodium phosphate buffer (pH 5.8, 0.1 mL), TCEP·HCl (5 mg), and
sodium 2-mercaptoethanesulfonate (MESNa, 0.1 g). The reac-
tion mixture was shaken (700 rpm) in an Eppendorf thermo-
mixer at 50 °C for 24 h. The cyclic peptide was then purified
from the reaction mixture by preparative HPLC; tR= 26.5 min,
and lyophilized to afford the pure product as fluffy white solid.
ESI+-MS: m/z calcd: 912.4; found: 913.2 Da [MH]+.
(24) Sohma, Y.; Hua, Q.-X.; Whittaker, J.; Weiss, M. A.; Kent, S. B. H.
Angew. Chem. Int. Ed. 2010, 49, 5489.
© Georg Thieme Verlag Stuttgart · New York — Synlett 2017, 28, A–F