Investigation of peptide thioester formation via N→Se acyl transfer

Article abstract of DOI:10.1002/psc.2469

Native chemical ligation is widely used for the convergent synthesis of proteins. The peptide thioesters required for this process can be challenging to produce, particularly when using Fmoc-based solid-phase peptide synthesis. We have previously reported a route to peptide thioesters, following Fmoc solid-phase peptide synthesis, via an N→S acyl shift that is initiated by the presence of a C-terminal cysteine residue, under mildly acidic conditions. Under typical reaction conditions, we occasionally observed significant thioester hydrolysis as a consequence of long reaction times (~48h) and sought to accelerate the reaction. Here, we present a faster route to peptide thioesters, by replacing the C-terminal cysteine residue with selenocysteine and initiating thioester formation via an N→Se acyl shift. This modification allows thioester formation to take place at lower temperatures and on shorter time scales. We also demonstrate how application of this strategy also accelerates peptide cyclization, when a linear precursor is furnished with an N-terminal cysteine and C-terminal selenocysteine.

Full text of DOI:10.1002/psc.2469

Research Article
Received: 30 July 2012
Revised: 9 October 2012
Accepted: 20 October 2012
Published online in Wiley Online Library
(wileyonlinelibrary.com) DOI 10.1002/psc.2469
Investigation of peptide thioester formation
via N!Se acyl transfer
Anna L. Adams and Derek Macmillan*
Native chemical ligation is widely used for the convergent synthesis of proteins. The peptide thioesters required for this
process can be challenging to produce, particularly when using Fmoc-based solid-phase peptide synthesis. We have
previously reported a route to peptide thioesters, following Fmoc solid-phase peptide synthesis, via an N!S acyl shift that
is initiated by the presence of a C-terminal cysteine residue, under mildly acidic conditions. Under typical reaction conditions,
we occasionally observed significant thioester hydrolysis as a consequence of long reaction times (~48 h) and sought to
accelerate the reaction. Here, we present a faster route to peptide thioesters, by replacing the C-terminal cysteine residue
with selenocysteine and initiating thioester formation via an N!Se acyl shift. This modification allows thioester formation
to take place at lower temperatures and on shorter time scales. We also demonstrate how application of this strategy also
accelerates peptide cyclization, when a linear precursor is furnished with an N-terminal cysteine and C-terminal selenocysteine.
Copyright © 2013 European Peptide Society and John Wiley & Sons, Ltd.
Keywords: native chemical ligation; selenocysteine; acyl transfer; thioester
The use of selenocysteine (Sec) as a means to improve known
reactions involving cysteine is widespread. For example, its use in
Introduction
The synthesis and semi-synthesis of proteins often rely on techni-
ques by which peptides and proteins can be ligated under mild
conditions. The advent of native chemical ligation (NCL) saw the
formation of a native peptide bond, in aqueous solution and at near
neutral pH, between two components adorned with an N-terminal
cysteine and a C-terminal thioester, respectively [1,2]. The genera-
tion of the peptide or protein C-terminal thioester is frequently
the most significant obstacle to this ligation strategy. A thioester
is generally unstable to Fmoc-based chemistry, which is a popular
method of peptide synthesis due to the common perception that
Fmoc-based chemistry is more compatible with chemically fragile
side-chain appendages. Consequently, strategies for the construc-
tion of peptide thioesters, particularly those employing Fmoc-
based chemistry, have been sought and developed [3–7].
Native peptides and proteins harboring a C-terminal cysteine
residue can spontaneously fragment to a C-terminal thioester and
cysteine in a reverse NCL-type process upon exposure to sodium
2-mercaptoethanesulfonic acid (MESNa) at mildly acidic pH [8–11].
Maximum conversion to thioester is generally observed by heating
the precursor peptide to between 50 ꢀC and 60 ꢀC for 24–48 h. In
many cases, these conditions are tolerated, although thioester
hydrolysis and peptide hydrolysis at aspartate residues (at lower
pH) have been observed as significant side reactions. Additionally,
while the reaction can differentiate between Xaa–Cys junctions,
exhibiting some preference for forming His, Cys, and Gly thioesters,
it is unlikely that we could distinguish efficiently between two such
motifs within a single peptide sequence.
NCL was reported by two groups in 2001 [12,13], and substitution
of the N-terminal cysteine for selenocysteine gave a more rapid
ligation to a peptide fragment comprising a C-terminal thioester.
This approach has since been extended to selenocysteine-
mediated cyclization of peptides via NCL [14] as well as the
semi-synthesis of proteins using expressed protein ligation
[15,16]. The selenocysteine incorporated into the peptide or pro-
tein at the ligation site can be reduced, eliminated, or alkylated at
the selenol to yield alanine, dehydroalanine, or an unnatural
amino acid, respectively [14,17,18].
Selenocysteine can readily be incorporated into synthetic
peptides during Fmoc solid-phase peptide synthesis (SPPS) as
the Se-p-methoxybenzyl (PMB) protected derivative, with PMB
deprotection and resin cleavage occurring simultaneously [19].
Compared with cysteine, selenocysteine has a similar electroneg-
ativity, but a selenolate anion (RSeꢁ) is a better nucleophile than
a thiolate (RSꢁ) at lower pH, a property partly attributable to the lower
pKa of the selenol [20] (pKA Sec–SeH = 5.24, Cys–SH = ~8 [21]).
Consequently, we hypothesized that, by introducing selenocysteine
at the C-terminus of a peptide, we could generate thioesters
more rapidly [22], proceeding via an initial N ! Se acyl shift
(Scheme 1). The product thioester would then be used in NCL
as normal. Although selenocysteine is known to be prone to dis-
elenide and selenosulfide formation, as well as deselenation
[14,18], it is hoped that the lower pH (pH < 6) might minimize
these undesirable reactions.
Although we have found this reaction operationally simple and
effective in many instances, we have been keen to accelerate
thioester formation to minimize hydrolysis, decrease heating and
heating time, and additionally to improve the selectivity of the
reaction, reducing the need for protecting groups on cysteines at
sites other than the target Xaa–Cys junction. Consequently, we
investigated selenocysteine as an acyl transfer facilitator.
*
Correspondence to: Derek Macmillan, Department of Chemistry, University
College London, 20 Gordon Street, London, WC1H 0AJ, UK. E-mail: d.macmillan
@ucl.ac.uk
Department of Chemistry, University College London, 20 Gordon Street,
London, WC1H 0AJ, UK
J. Pept. Sci. 2013
Copyright © 2013 European Peptide Society and John Wiley & Sons, Ltd.

ADAMS AND MACMILLAN
HSe
O
H3N
O
CONH2
O
H+
MESNa
SO3H
S
N
H
CONH2
1
Se
2
Scheme 1. Proposed reverse NCL employing C-terminal selenocysteine peptide 1 as the nucleophile and leaving group, resulting in thioester 2.
In this study, we prepared model thioester precursors, includ-
ing 13C-labeled Sec-containing peptides and examined thioester
formation under a range of conditions. The advantages and
disadvantages of using Sec in place of Cys became apparent in
acyl transfer processes, and an optimized procedure was
employed to effect the cyclization of an antimicrobial peptide
derived from a b-defensin.
p-Methoxybenzyl selenocysteine
Sodium borohydride (0.45 g, 11.84mmol) was added portion-wise
to an ice-cooled solution of selenocystine (0.50 g, 0.50 mmol) in
0.5-M sodium hydroxide (1.3ml). The mixture was kept in an
ice bath and stirred until the yellow color had disappeared. To
the mixture was added 2-M sodium hydroxide (3.9ml), followed
by dropwise addition of p-methoxybenzyl chloride (1.22 g,
7.77 mmol). The reaction mixture was stirred vigorously at 4 ꢀC for
4 h. The mixture was acidified with concentrated hydrochloric acid
to form a precipitate of the crude product that was collected by
filtration and washed thoroughly with diethyl ether and water to
afford a white solid (0.52 g, 1.8mmol, 60%), which was used
without further purification. TLC: Rf 0.04 (chloroform : methanol:
acetone, 8.5 : 1 : 0.5).
Materials and Methods
Materials
All reagents and solvents (excluding HPLC solvents) were obtained
from Sigma (Sigma-Aldrich Ltd. Dorset, UK) as standard laboratory
grade and used as supplied unless otherwise stated. All standard
resins and Fmoc amino acids for peptide synthesis were purchased
from Novabiochem (Merck KGaA, Darmstadt, Germany).
Fmoc-p-methoxybenzyl selenocysteine
Fmoc succinimide (0.42 g, 1.23 mmol) was dissolved in acetonitrile
(2.22 ml) and added to an ice-cooled suspension of p-methoxybenzyl
selenocysteine (0.37 g, 1.28 mmol) in a mixture of distilled water
(2.22 ml) and triethylamine (178 ml, 1.28 mmol). Further, triethyla-
mine was added (178ml 1.28 mmol) to maintain a pH above pH 8,
and the reaction was stirred at room temperature for 1 h. The
mixture was then acidified with 1-M hydrochloric acid and the crude
product extracted using ethyl acetate. The combined organic layers
were washed with 1-M hydrochloric acid followed by saturated
sodium chloride solution. The organic phase was then dried
over magnesium sulfate and filtered under gravity, and the solvent
removed under vacuum. The resulting oil was washed in n-hexane
(5ml), which was decanted, and the remaining solvent removed by
vacuum to afford the product (0.47 g, 0.91 mmol, 71%) as a white
foam. TLC: Rf 0.70 (chloroform : methanol: acetone, 8.5: 1 : 0.5).
NMR 1H: dH (500 MHz, CDCl3) 7.79–6.21 (12H, m, ArH), 4.58 (1H, m,
CHa), 3.79–3.69 (5H, Fmoc CH, Fmoc CH2, Se–CH2–Ar), 3.65 (3H, s,
OCH3), 2.93 (1H, m, CH2), 2.82 (1H, m, CH2). m/z (ESI-TOF-MS)
calculated [M+ Na]+ 534.0770, found 534.0796.
Instruments
1H NMR spectra were recorded at 500 MHz or 400 MHz and 13C NMR
spectra at 125MHz or 100MHz on Bruker (Bruker, Coventry, UK)
500 MHz and 400 MHz instruments, respectively. Chemical shifts (d)
were reported in ppm and coupling constants (J) in Hz, signals were
sharp unless stated as broad (br), s, singlet, d, doublet, t, triplet, q,
quartet, and m: multiplet. Mass spectra were obtained on Waters
(Waters MS Technology Centre, Manchester, UK) uPLC/SQD-LC series
electrospray mass spectrometer. LC-MS was performed using a gra-
dient of 1–95% acetonitrile containing 0.1% formic acid over 10 min
(flow rate of 0.6 ml/min). Semi-preparative HPLC was performed us-
ing a Phenomenex (Phenomenex, Cheshire, UK) LUNA C18 column
and a gradient of 5–60% acetonitrile containing 0.1% TFA over
45 min (flow rate of 4.0 ml/min).
Synthesis of Fmoc-Sec(PMB)-OH [19]
p-Methoxybenzyl chloride
p-Methoxybenzyl alcohol (3.90 g, 28.5 mmol) was dissolved in dry
chloroform (20 ml) and cooled to 5 ꢀC. Thionyl chloride (3.1 ml,
42.8 mmol) was added with care to keep the temperature below
10 ꢀC. The reaction mixture was allowed to warm to room
temperature and stirred for 2 h. The reaction was then basified
to pH 8.0 by slow and careful addition of saturated sodium
bicarbonate solution. The product was extracted using chloroform
(3ꢂ 30ml), and the combined organic layers were washed with
distilled water (3ꢂ 20 ml) then dried over magnesium sulfate. The
solution was filtered under gravity and the solvent removed under
vacuum to afford the crude product (4.03 g, 25.7mmol, 90%) as an
oil that was used without further purification. TLC: Rf 0.73 (30%
ethyl acetate in n-hexane). m/z (ESI-TOF-MS) calculated [MH]+
Fmoc-glycine (13C-1)
As reported in Kang et al. 2009 [23]. Glycine (13C-1) (0.25 g,
3.25 mmol) was dissolved in distilled water (3.13 ml), and triethy-
lamine (453 ml, 3.25 mmol) was added while the mixture was
continuously stirred. 9-fluorenylmethyl succinimidyl carbonate
(1.06 g, 3.155 mmol) was then dissolved with heating in acetoni-
trile (3.13 ml) and added in one portion to the amino acid
solution. Further, triethylamine (453 ml, 3.25 mmol) was added
to maintain a pH between pH 8.5 and pH 9.0. The reaction
mixture was stirred at room temperature for 0.5 h and then
concentrated in vacuo. The concentrate was poured into 1.5-M
hydrochloric acid (14 ml). The solution was then filtered under
gravity and the product extracted with DCM (3 ꢂ 6.25 ml). The
combined organic extracts were washed with 1.5-M hydrochloric
acid (6.25 ml), distilled water (6.25 ml), and a solution of saturated
aqueous sodium chloride (6.25 ml). The organic phase was dried
1
156.03364, found [MH]+ 156.03348. H NMR: dH (500 MHz, CDCl3)
7.33 (2H, d, J = 8.6, ArH), 6.89 (2H, d, J = 8.6, ArH), 4.58 (2H, s, CH2),
3.82 (3H, s, O–CH3). 13C NMR: dC (125 MHz, CDCl3) 130.18 (CH),
114.23 (CH), 55.40 (O–CH3), 46.42 (CH2).
wileyonlinelibrary.com/journal/jpepsci
Copyright © 2013 European Peptide Society and John Wiley & Sons, Ltd.
J. Pept. Sci. 2013

N!Se ACYL TRANSFER
over anhydrous magnesium sulfate and the solvent removed
in vacuo. The addition of petroleum ether (2.5 ml) afforded a
white precipitate that was filtered and dried under vacuo to a
white solid (0.81 g, 2.7 mmol, 83%). 1H NMR: (DMSO, 400 MHz)
dH/ppm: 7.90 (2H, d, J = 8.0, FmocAr–H), 7.73 (2H, d, J = 8.0, Fmoc
Ar–H), 7.62 (1H, t, J = 6.0, NH), 7.42 (2H, t, J = 8.0, Fmoc Ar–H), 7.34
(2H, t, J = 8.0, Fmoc Ar–H), 4.30 (2H, d, J = 4.0 Fmoc CH2–O), 4.23
(1H, m, Fmoc CH–CH2), 3.67 (2H, t, J = 6.0, Gly–aCH2); 13C NMR:
(DMSO, 100 MHz) dC/ppm: 172.6 (CO2H). m/z (ESI-TOF-MS)
calculated [M + Na]+ 321.0933, found 321.0928.
Table 1. Target peptide sequencesa and quantities of DTNP used
during cleavage
Sequence
DTNP (eq)
H-MEELYKSHU- Rink Amide resin
1.3
1.3
1.3
10
1b
H-MEELYKSG(13C-1)U-Rink Amide resin
H-MEELYKSG(13C-1)U- NovaSyn TGA resin
H-DTHFPICIFCCGU- Rink Amide resin
H-DTHFPIC(Acm)IFC(Acm)C(Acm)GU- Rink Amide resin
H-CRKFFARIRGGRGU- Rink Amide resin
5
aUnless otherwise stated, the resin-bound peptides carry the standard
side-chain protecting groups for Fmoc-SPPS.
bIn this case, thioanisole was omitted from the cleavage cocktail.
General Synthesis Procedures for Sec-containing Peptides
(Manual Procedures)
Fmoc-Sec(PMB)-Rink amide MBHA resin
Rink amide MBHA resin, loading = 0.56 mmol/g, (89 mg, 0.05 mmol)
was treated with 20% piperidine (v/v) in DMF (2 ml, 15 min). The
solution was then filtered and the resin exhaustively washed with
DMF and DCM. The resin was then coupled to Fmoc-Sec(PMB)-
OH (127.6mg, 0.25 mmol) with 0.45-M HBTU/HOBt (0.55 ml) and
DIPEA (75 ml) in dry DMF (0.55 ml), on a horizontal shaker
(300rpm) at room temperature for 4 h. The resin was then filtered
off and washed exhaustively with DMF and DCM. Following Fmoc
analysis to determine coupling efficiency, the washed resin
was then transferred to a reaction vessel for automated SPPS for
elongation to the desired peptide sequence.
2,20-dithiobis(5-nitropyridine) (DTNP) (see Table 1 for quantities
of DTNP added) for 5 h at room temperature with gentle stirring.
After exposure to the cleavage cocktail for 5 h, the resin was
filtered off and the filtrate added to cold diethyl ether (10ꢂ volume)
to induce precipitation of the peptide. The precipitate was
collected by centrifugation (1500 rpm, 4 ꢀC, 15 min), the super-
natant removed, and the precipitated peptide resuspended in
diethyl ether and centrifuged once more. The white precipitate
was dissolved in water and then purified by semi-preparative
reverse phase HPLC (gradient: 5–60% acetonitrile/45 min).
Fractions containing the product were identified by LC-MS and
lyophilized to afford pure peptides as white fluffy solids (Table 2).
Fmoc-Sec(PMB)-NovaSyn TGA resin
NovaSyn TGA resin, loading = 0.22 mmol/g, (128 mg, 0.05 mmol)
was allowed to swell in DCM (15 min). The resin was then filtered
off and then double coupled to Fmoc-Sec(PMB)-OH as earlier.
General procedure for thioester formation
Sec(5-Npys)-terminated peptide 4 (1 mg) was dissolved in 0.1-M
sodium phosphate buffer; pH 5.8 (0.9 ml), containing MESNa
(100 mg) and tris-carboxyethylphosphine (TCEP) (5 mg) to a total
volume of approximately 1 ml, and shaken on a Eppendorf
thermomixer (Eppendorf, Stevenage, UK) at temperatures ranging
from 40–60 ꢀC. 20-ml aliquots of reaction mixture were removed
for analysis by analytical HPLC over a period of 48 h. After 48 h, a
3-ml aliquot of each reaction mixture was subjected to LC-MS
analysis and up to three species derived from 4 were generally
observed with masses: 1161.4 Da [MH]+ corresponding to the
thioester 7, 1325.4 Da [MH]+ corresponding to the seleno-sulfide
adduct 6, and 1037.4Da [MH]+ derived from hydrolysis of 7.
Fmoc-G(13C-1)U(PMB)-Rink amide MBHA resin and Fmoc-G(13C-1)U
(PMB)-NovaSyn TGA resin
The washed and dried Fmoc-Sec(PMB)-Rink amide MBHA resin or
Fmoc-Sec(PMB)-NovaSyn TGA resin was treated with 20% piperidine
(v/v) in DMF (15 min), the solution filtered off, and the resin then
washed exhaustively with DMF and DCM. The resin was then cou-
pled to Fmoc-Gly(13C-1)-OH (75 mg, 0.25 mmol) with 0.45-M HBTU/
HOBt (0.55 ml) and DIPEA (75 ml) in dry DMF (0.55 ml), on a horizontal
shaker at room temperature for 4 h. The resin was then filtered off
and washed exhaustively with DMF and DCM. Following ninhydrin
test analysis, the washed resin was then transferred to a 433A reac-
tion for elongation to the desired peptide sequence.
General procedure for thioester formation in 13C-labeled precursors
8,10, 11, and 12
Reactions monitored by 13C NMR were prepared in a similar fashion
to that described earlier for model peptide 4. Briefly, 13C-labeled
peptides were dissolved to final concentrations of 1 mg/ml in 0.1-M
sodium phosphate buffer; pH 5.8 (prepared in D2O) containing
MESNa (10% w/v) and TCEP (0.5% w/v). 0.6-ml aliquots were
dispensed into six separate 1.5-ml Eppendorf tubes, and the reac-
tions were shaken in an Eppendorf thermomixer at 60 ꢀC for 48 h.
The contents of each Eppendorf tube, once transferred to an
NMR tube, were used to obtain an independent estimation of reac-
tion progress at 0, 1, 3, 6, 24, and 48 h by 13C NMR spectroscopy.
Automated Peptide Synthesis
Automated solid-phase synthesis was carried out on 0.05-mmol
scale, on an Applied Biosystems ABI 433A peptide synthesizer
(Applied Biosystems – Life Technologies Ltd, Paisley, UK) accord-
ing to the manufacturers’ instructions. The Fast-mocTM protocol
for SPPS was employed, using 10 equivalents of each standard
Fmoc-protected amino acid in the sequence, and each coupling
reaction was carried out using HBTU/HOBt in the presence of
DIPEA (see Table 1 for assembled sequences).
Post-Synthesis Procedures
General procedure for thioester formation in hepcidin-derived
peptides 13 and 19
Following chain assembly, the resin was then removed from the
reaction vessel, washed exhaustively with DMF and DCM, and
dried. The dry resin was treated with a cleavage cocktail (4.0 ml)
comprised of TFA (97.5% v/v) and thioanisole (2.5% v/v) with
0.1-M sodium phosphate buffer; pH 5.8 in guanidinium hydrochlo-
ride was prepared by a 10-fold dilution of 1-M sodium phosphate
buffer; pH 5.8 into 6-M guanidinium hydrochloride. Peptide species
J. Pept. Sci. 2013
Copyright © 2013 European Peptide Society and John Wiley & Sons, Ltd.
wileyonlinelibrary.com/journal/jpepsci

ADAMS AND MACMILLAN
Table 2. HPLC retention times (tR), mass spectrometry characterization, and isolated yields for Sec-containing thioester precursors
Peptide
tR/min
Calculated mass/Da (no. of 5-Npys)
Observed mass/Da (ESI-MS) [MH]+
Isolated yield (%)a
4
24.8
26.0
25.9
41.0,
42.4,
48.7
33.4
27.3
1339.4 (1)
1260.2 (1)
1261.2 (1)
2121.2 (4)
1340.8
1261.2
31
5
8
10
13A
13B
13C
19
20
1262.6
7
1810.6 (2 ꢂ 5-Npys)
1810.4 (2 ꢂ 5-Npys)
2120.8
1
2
3
1872.0 (1)
1981.2 (2)
3434.2 (Se–Se dimer)
1981.1
17
30
aIsolated yields are based on initial resin loading.
SePMB
isolated by semi-preparative HPLC and assigned, by LC-MS, as Sec
(5-Npys) or Cys(5-Npys) capped derivatives of 13 or 19 were then
dissolved to a final concentration of 1 mg/ml in the prepared
phosphate buffer solution containing MESNa (10% w/v) and TCEP
(0.5% w/v). The reaction was then shaken on an Eppendorf thermo-
mixer at 60 ꢀC for 24 h. 3-ml aliquots of the reaction mixture were
subjected to LC-MS analysis, after 6 and 24h.
O
N
CONH
3
H
i)
NO2
S
N
Se
Peptide cyclization via N!Se acyl shift
O
Peptide 20 was dissolved in 0.1-M sodium phosphate buffer (pH
5.8), with MESNa (10% w/v) and sodium ascorbate (5% w/v).
The final peptide concentration was 1 mg/ml. The reaction
mixture was degassed by freezing the sample in liquid nitrogen
and placing the frozen reaction mixture on a freeze drier for
10 min. TCEP-HCl (0.5% w/v) was then added to the thawed
sample under a blanket of nitrogen, and the reaction was shaken
on an Eppendorf thermomixer at 60 ꢀC and purified after 24 h by
semi-preparative reverse-phase HPLC. Fractions containing the cyclic
peptide were lyophilized to yield 21 as a white fluffy solid (4.3 mg,
61%), tR 24.9 min, calculated mass: 1505.8 Da, observed mass
(ESI-MS) [MH2]2+ + 753.5 Da, which deconvolutes to 1505.1 Da.
N
H
CONH2
4
ii)
iii)
iv)
SeH
Se
S
SO3H
O
O
O
SO3H
S
N
CONH2
5
N
H
CONH2
6
H
7
Scheme 2. Synthesis of a peptide thioester 7 from Sec-terminated 4.
Reagents and conditions: (i) 95% v/v TFA, 1.3 equivalent 2,20-dithiobis
(5-nitropyridine). (ii) 50-mM DTT, Na phosphate buffer; pH 7.5. (iii) 10%
w/v MESNa, 0.1-M Na phosphate buffer pH 5.8, 60 ꢀC. (iv) 10% w/v MESNa,
0.1-M Na phosphate buffer pH 5.8, 0.5% w/v TCEP-HCl, 60 ꢀC, 6 h.
Results and Discussion
For model experiments, selenocysteine was introduced to
synthetic peptides using the Fmoc-Sec(PMB)-OH building block.
Although not especially ideal, and highlighting the dearth of avail-
able orthogonal protecting groups for Sec, we envisaged that the
transformation of Sec(PMB) to the 5-nitropyridyl-2-selenosulfide
[Sec(5-Npys)] derivative 4 would enable our proposed study
(Scheme 2). Therefore, model peptide 4 (sequence: H-MEELYK-
SHU(5-Npys)-NH2), derived from the C-terminal residues of green
fluorescent protein, was obtained in 31% yield following cleavage
from Rink amide resin in the presence of DTNP [24]. We first
attempted the conversion of the 5-Npys-protected peptide to the
free selenol 5. However, upon exposure to 50-mM dithiothreitol
(DTT) at pH 7.5, 5 could only be observed in small quantities
and the reaction mixture was comprised mainly of the diselenide
after 4 h. Longer exposure to DTT resulted in decomposition to
several species after 16 h, suggesting that it may be more beneficial
to use the selenosulfide 4 directly as the thioester precursor.
Unfortunately, exposure of 4 to typical reaction conditions
(10% w/v MESNa, 0.1-M Na phosphate buffer; pH 5.8, 60 ꢀC) did
not result in thioester formation, but we instead observed quantita-
tive conversion to selenosulfide 6.
Upon introduction of TCEP-HCl to the reaction mixture, the
desired product 7 could, in fact, be observed and gratifyingly
the reaction appeared, by HPLC analysis, to be almost complete
within 6 h at 60 ꢀC. The corresponding cysteinyl peptide under-
went less than 20% conversion to 7 in the same time.
Furthermore, good conversion from 4 to 7 was observed at
lower temperatures (40–50 ꢀC) and was extremely encouraging
(Figure 1). At 40 ꢀC, thioester formation, while slow, proceeded
smoothly resulting in 73% conversion to 7 within 24 h. The reac-
tion was, however, incomplete, and a small amount (6%) of 4,
and intermediates derived from 4, remained. After 6 h, less than
11% of 7 had been hydrolyzed, and by 24 h, the amount of
hydrolyzed thioester had not increased significantly. At 50 ꢀC,
after 6 h, the rate of thioester formation appeared similar to that
at 40 ꢀC, reaching 75% conversion to 7 in 24 h; however, at this
temperature, all of 4 was consumed. 7 was more prone to hydro-
lysis at this higher temperature, and at 24 h, 14% of the peptide
had been hydrolyzed. At both temperatures, the formation of 7
was fastest within the first 6 h. In contrast, the hydrolysis of 7
occurred at a much slower rate throughout the reactions. At
60 ꢀC, maximum conversion to 7 was reached in 6 h (77%). A
wileyonlinelibrary.com/journal/jpepsci
Copyright © 2013 European Peptide Society and John Wiley & Sons, Ltd.
J. Pept. Sci. 2013

N!Se ACYL TRANSFER
(A)
The significant decrease in reaction time by using selenocysteine
rather than cysteine was verified by 13C NMR spectroscopy employ-
ing a 13C-1-labeled glycine analogue of 4; 8 (sequence: H-MEE-
LYKSG(13C-1)U-NH2, Figure 2). Conversion to the corresponding
MESNa thioester 9 (sequence: H-MEELYKSG(13C-1)-SCH2CH2SO3H)
could be easily confirmed through appearance of the characteristic
thioester signal at ~200 ppm (Figure 2(A)). Notably there was no
obvious trace of an Se-peptide intermediate, but a small signal
at 190 ppm, observed in some spectra, may be indicative of
its presence.
100
80
60
40
20
4
7
Hydrolysis
13C NMR analysis of the reaction was generally in agreement
with the HPLC data, in that the conversion of Sec(5-Npys) precur-
sor to thioester was nearly complete within 6 h and had reached
completion within 24 h. When 13C-labeling data for 8 and the
corresponding Cys-terminated peptide 12 (H-MEELYKSGC-NH2)
were compared, the rate enhancement attributable to the
presence of the Sec residue was clear (Figure 2(B)). Interestingly,
and as seen previously [9], thioester formation was initially faster with
carboxyl-terminated peptide 10 (sequence: H-MEELYKSGU-OH)
than with 8 affording 60% conversion to thioester after only
3 h. However, this reaction showed little further conversion to
thioester after this time, and the 13C NMR spectrum remained
largely unchanged after 24 h. In this case, LC-MS indicated that
a significant side reaction had occurred, resulting in a product
that lacked selenium. The molecular mass of this product
coincided exactly with that of the Ala analogue (sequence:
H-MEELYKSGA-OH) indicating the increased sensitivity of Sec
to deselenation, because no analogous side reaction is
observed in the corresponding Cys-terminated peptides under
identical reaction conditions. Although the simplified isotopic
distribution of the molecular ion in the mass spectrum was
sufficient to indicate loss of selenium, further analysis of the
1H NMR spectrum of this purified side product supported
alanine formation by the presence of a characteristic doublet
for the side-chain methyl group at 1.40 ppm. We repeated
the reaction of 10 in the presence of 5% w/v sodium ascor-
bate [25] and did not observe any deselenation. However,
once formed, 9 appeared much more susceptible to hydroly-
sis in the presence of sodium ascorbate. Here, the elimination
of one problematic side reaction only exacerbated another,
and so, the presence of sodium ascorbate was considered
poorly compatible with this process.
Interestingly, thioester hydrolysis appeared generally more
prevalent during HPLC analysis of reaction mixtures contain-
ing the His-Sec-terminated model peptide when compared
with 13C NMR analyses of Gly–Sec-terminated peptide where
it is a minor component, if observed at all. This indicated
that thioester hydrolysis was either more prevalent in His
rather than Gly thioesters or that significant hydrolysis oc-
curred during HPLC sampling of reaction mixtures. Thioester
observed at t = 0 h in HPLC monitored experiments was also
not evident in 13C NMR spectra of Gly–Sec-terminated
peptides. Thioester formation across the 13C-labeled Gly–
Sec-terminated peptide was then also subjected to HPLC
analysis and was in full agreement with the 13C NMR data,
supporting the increased reactivity of the His–Sec motif over
the Gly–Sec motif.
0
0
6
24
24
24
Time / hours
40°C
(B)
100
80
4
7
60
Hydrolysis
40
20
0
0
6
Time / hours
50°C
(C)
100
80
4
7
60
Hydrolysis
40
20
0
0
6
Time / hours
60°C
Figure 1. HPLC analysis of the formation of thioester 7 from Sec-
terminated 4. Reagents: 10% w/v MESNa, 0.1-M Na phosphate buffer
pH 5.8, 0.5% w/v TCEP-HCl, at (A) 40, (B) 50, and (C) 60ꢀC. Upon reduction
with TCEP, at t= 0h, 4 is transformed to a mixture comprising selenol 5, sele-
nosulfide 6, and diselenide, which are consumed as the reaction progresses.
The % 4 given above corresponds to the summed total integration for these
species, where present.
small quantity of 4 remained but was fully consumed after 24 h.
However, at this temperature, 7 was being hydrolyzed more
rapidly such that the total amount of 7 was reduced after 24 h,
and 27% of the hydrolysis product had accumulated. Although
hydrolysis was less prevalent over extended reaction times at lower
temperatures, the thioester was formed fastest at 60ꢀC. The
amount of hydrolysis seen at this temperature after 6 h was also
comparable with that seen after 24h at the lower temperatures.
Having confirmed that thioester formation across these Xaa–Sec
motifs can proceed to near completion in approximately 6 h, we
were keen to explore whether the reaction had the potential to
occur selectively at a single site, even in the presence of additional
unprotected Xaa–Cys motifs.
J. Pept. Sci. 2013
Copyright © 2013 European Peptide Society and John Wiley & Sons, Ltd.
wileyonlinelibrary.com/journal/jpepsci

ADAMS AND MACMILLAN
8
(A)
TCEP
0 h
1 h
6 h
9
24 h
(B)
100.0
80.0
60.0
40.0
20.0
0.0
precursor
10, H-MEELYKSGU-OH
8, H-MEELYKSGU-NH2
11, H-MEELYKSGC-OH
12, H-MEELYKSGC-NH2
0
10
20
30
40
Time /h
Figure 2. (A) Thioester formation from precursor peptide 8 monitored by 13C NMR after 0, 1, 6, and 24 h. Reagents and conditions: 1 mg/ml peptide, 10% w/v
MESNa, 0.1-M Na phosphate buffer pH 5.8 (in D2O), 0.5% w/v TCEP. (B) Comparison of thioester formation in Sec-terminated and Cys-terminated peptides.
A peptide based on a fragment of human hepcidin comprising
residues 1–12 and furnished with a C-terminal Sec(PMB) carboxa-
mide, 13, was synthesized (Figure 3). Thioesterification of the
C-terminal Sec carboxamide was examined because the synthesis
was more straightforward, and the conversion to thioester had
been observed to take place more rapidly when compared with
the C-terminal cysteinyl carboxylate. Additionally, thioester
formation, when conducted on peptide 8, had also been free from
reduction of Sec to Ala. This fragment of hepcidin contained three
unprotected cysteine residues in addition to the terminal seleno-
cysteine. Furthermore, it possessed a Cys–Cys motif, known to be
favorable to form thioesters in the presence of MESNa at pH 5.8.
By subjecting this peptide to thioester formation, it is hoped to
establish whether the faster thioester formation at Gly–Sec would
lead to selectivity being observed for this site, over the internal
Xaa–Cys sites.
Upon cleavage of this peptide from the resin, initially employ-
ing stoichiometric quantities of DTNP and then with an excess of
DTNP, a mixture of peptides protected to varying degrees with
the 5-Npys group were obtained. The occurrence of this hetero-
geneous mixture of 5-Npys-protected starting materials lowered
the isolated yield of 13 significantly. In cases where incomplete
5-Npys modification was observed, we were unable to ascertain
whether the thiols on the cysteines or the selenol group had
been protected, but we suspected that the 5-Npys group had
facilitated the formation of an intramolecular selenosulfide bond
between a cysteine residue and selenocysteine [24]. We next
subjected this mixture of peptides to the reaction conditions,
anticipating that the presence of TCEP-HCl would eradicate the
variation in the starting material(s) by reducing any disulfide
and selenosulfide bonds, as well as cleaving the 5-Npys groups.
The reaction was carried out in guanidine hydrochloride to aid
the solubility of 13, and, prior to the addition of TCEP-HCl under
nitrogen, the reaction mixture was degassed with the hope of
lowering the risk of deselenation.
At t = 0 h, and upon addition of all reagents, LC-MS indicated that
the starting peptide was indeed mostly reduced peptide, 14. After
6 h, some of the desired thioester (15) had already formed via
reaction at the Gly–Sec site on the C-terminus; however, a consider-
able quantity of unreacted 14 remained, which was in stark
contrast to the model studies. After 24 h, the reaction contained a
mixture of thioesters formed at the C-terminal Gly–Sec site, as well
as the internal Cys–Cys (16) site and remarkably at the internal
Phe–Cys site (17). A small amount of unreacted starting peptide
also remained. Unfortunately, also observed to a significant
extent was deselenated peptide, 18, where the C-terminal
Sec had been converted to Ala. Notably, after 24 h, the observed
mass (1503.9 Da, MH+) for the reduced starting material 14 was
low (calculated m/z = 1505.5Da, MH+) and the sluggish progress
of the desired reaction was possibly attributable to the formation
of a persistent intramolecular selenosulfide bond, which would also
compromise the selectivity as the reaction progressed.
From this investigation, we concluded that hepcidin-derived 13
was too ambitious for an initial substrate on which to demonstrate
the benefits of the C-terminal Sec-mediated thioester formation or
observe selectivity for thioester formation at Sec over Cys, due to its
cysteine-rich nature. Whereas thioester formation across the Ile–Cys
junction was not observed, as reported previously, a species
corresponding to the Phe-thioester 17 was particularly well repre-
sented in the combined total ion chromatogram (TIC). This raises
the possibility that direct thiolysis of this Phe–Cys motif or, intrigu-
ingly, its production form thioester-terminated 16 is also favorable.
wileyonlinelibrary.com/journal/jpepsci
Copyright © 2013 European Peptide Society and John Wiley & Sons, Ltd.
J. Pept. Sci. 2013

N!Se ACYL TRANSFER
(A)
(B)
14
15
15
(C)
17
18
14(ox)
16
Figure 3. (A) Attempted synthesis of peptide thioester 15 from Sec-terminated 14. Reagents and conditions: 10% w/v MESNa, 0.1-M Na phosphate
buffer pH 5.8, 0.5% w/v TCEP-HCl, 60 ꢀC, 24 h. (B) MS analysis showing [MH]+ of components present in the reaction after 6 h. (C) MS analysis showing
[MH]+ of components present in the reaction mixture after 24 h.
O2N
H2N
HN
NH
HN
NH2
NH
NH2
O
N
S
S
O
O
O
O
O
O
H
N
H
N
H
H
H
H
H
H
N
N
N
N
N
N
H
N
H
N
N
H
N
H
N
N
H
NH2
N
H
H
O
O
O
O
O
O
O
Se
S
HN
NH
H2N
NH
HN
NH2
NO2
20
NH2
HN
NH
O
H
N
O
NH
N
HN
H
O
NH
O
O
N
H
H2N
O
HN
HN
HN
H
NH2
N
O
HS
NH
HN
O
NH
O
O
NH
O
H
H
N
N
H2N
HN
NH
O
NH
21
O
NH2
Scheme 3. Conversion of linear thioester precursor 20 to cyclic 21. Reagents and conditions: 20, 10% w/v MESNa, 0.1-M Na phosphate buffer pH 5.8, 5%
w/v Na ascorbate, 0.5% w/v TCEP.
J. Pept. Sci. 2013
Copyright © 2013 European Peptide Society and John Wiley & Sons, Ltd.
wileyonlinelibrary.com/journal/jpepsci

ADAMS AND MACMILLAN
In an attempt to make this hepcidin fragment more compatible
with our chemistry, we decided to next protect the Cys residues
with Acm protecting groups. However, the synthesis of the Acm-
protected thioester precursor 19 was not as straightforward as it
was hoped because of the labile nature of the Acm protecting
groups in the presence of DTNP. To preserve the Acm groups, we
used 1 equivalent of DTNP and omitted thioanisole from the
cleavage cocktail [24]. The resulting peptide was isolated as a
dimer, presumably joined by an intermolecular diselenide bond,
and therefore did not possess any 5-Npys protecting groups. When
subjected to thioester formation, we did not see reduction of this
bond, which appeared inert to 0.5% w/v (17.5mM) TCEP. After
6 h, the peptide had undergone quantitative deselenation to afford
the alanine terminated product.
In light of this, we wanted to apply our reaction to a substrate that
was not quite as cysteine-rich as the fragment of hepcidin. We have
previously shown a fragment of Defb14 (sequence: H-CRKFFAR-
IRGGRGC-OH), the mouse orthologue of the human b-definsin-3
(hBD-3), to undergo cyclization by virtue of a C-terminal and an
N-terminal cysteine and subjected to thioester formation at pH
5.8. Upon conversion to the C-terminal thioester, the peptide
spontaneously underwent intramolecular NCL [26].
We isolated the desired thioester precursor containing seleno-
cysteine at the C-terminus (20, sequence: H-C(5-NPys)RKFFAR-
IRGGRGU(5-Npys)-NH2) in 30% yield and subjected it to our usual
reaction conditions (10% w/v MESNa, 0.1-M Na phosphate buffer;
pH 5.8, 0.5% w/v TCEP, 60 ꢀC). After 6 h, a small amount of
thioester and cyclic peptide could be observed by LC-MS;
however, the reaction appeared to be mainly comprised of 20
and deselenated 20 (sequence: H-CRKFFARIRGGRGA-NH2). After
24 h, the reaction contained mostly deselenated material.
To eliminate deselenation, likely initiated by the presence of air
and TCEP, we reintroduced the antioxidant sodium ascorbate to
the reaction [25], and in this case, the major product after 24 h
was the cyclic peptide 21 (Scheme 3), which we were able to isolate
in 61% yield. In contrast to what we observed in the case of peptide
10, hydrolysis of the presumed intermediate thioester (which was
enhanced in the presence of sodium ascorbate) was not observed.
This was likely to be due to the transient nature of the thioester,
which was not observed by LC-MS during the reaction.
A)
H-CRKFFARIRGGRGU-NH2
B)
H-CRKFFARIRGGRGC-NH2
325.9
419.05
100
407.2
100
t = 0 h
100
t = 0 h
100
7
2.67
2.63
2
558.56
542.7
4
837.49
813.5
4
0
0
0
0.00
0
0.00
3.00
4.00
5.00
1.00
2.00
1.00
2.00
3.00
4.00
5.00
t = 6 h
100
t = 6 h
100
2.63
2.95
2.69
2.97
0
0
0.00
0.00
1.00
3.00
4.00
100
5.00
502.63
2.00
1.00
2.00
3.00
4.00
377.27
5.00
t = 24 h
100
100
t = 24 h
100
3.06
2.91
2.61
377.20
502.70
753.54
2.76
753.82
0
0
0
0.00
0
0.00
3.00
4.00
5.00
1.00
3.00
4.00
5.00
1.00
2.00
2.00
Figure 4. Comparison of Sec-terminated (panel a) and Cys-terminated (panel b) peptides as they undergo cyclization under identical reaction conditions.
The figure shows TIC traces from LC-MS analysis of each reaction mixture at 0, 6, and 24 h. The mass spectra (inset) are obtained by combining the data across
the full-width at half maximum peak height of the most abundant peak in the TIC. (A) H-CRKFFARIRGGRGU-NH2 and (B) H-CRKFFARIRGGRGC-NH2.
wileyonlinelibrary.com/journal/jpepsci
Copyright © 2013 European Peptide Society and John Wiley & Sons, Ltd.
J. Pept. Sci. 2013

N!Se ACYL TRANSFER
3 Shin Y, Winans KA, Backes BJ, Kent SBH, Ellman JA, Bertozzi CR. Fmoc-
based synthesis of peptide-(alpha)thioesters: application to the total
chemical synthesis of a glycoprotein by native chemical ligation. J.
Am. Chem. Soc. 1999; 121(50): 11684–11689.
4 Nakamura K, Sumida M, Kawakami T, Vorherr T, Aimoto S. Generation
of an S-peptide via an N–S acyl shift reaction in a TFA solution. Bull.
Chem. Soc. Jpn. 2006; 79(11): 1773–1780.
5 Hojo H, Onuma Y, Akimoto Y, Nakahara Y. N-alkyl cysteine-assisted
thioesterification of peptides. Tetrahedron Lett. 2007; 48(1): 25–28.
6 Blanco-Canosa JB, Dawson PE. An efficient Fmoc-SPPS approach
for the generation of thioester peptide precursors for use in
native chemical ligation. Angew. Chem. Int. Ed. 2008; 47(36):
6851–6855.
7 Mende F, Seitz O. 9-Fluorenylmethoxycarbonyl-based solid-phase
synthesis of peptide alpha-thioesters. Angew. Chem. Int. Ed. 2011; 50
(6): 1232–1240.
This isolated yield of this cyclization (61%) certainly
appeared superior to the corresponding reaction of the Cys-
carboxylate-terminated peptide we had reported previously [26].
Furthermore, the reaction was essentially complete in only half of
the time, 24 h rather than 48h. For a more accurate comparison
of reaction progress, cyclization of 20 was compared with that of
the C-terminal Cys-carboxamide (sequence: H-CRKFFARIRGGRGC-
NH2,). Although the cyclization of the Sec-terminated precursor
was not complete within 6 h, it had clearly progressed further than
that of the cysteinyl peptide (Figure 4). Furthermore, the cyclic
product derived from the Cys-carboxamide was only isolated in a
yield of 20% yield after 24 h.
8 Kang J, Macmillan D. Peptide and protein thioester synthesis via N!S
acyl transfer. Org. Biomol. Chem. 2010; 8(4): 1993–2002.
9 Richardson JP, Chan CH, Blanc J, Saadi M, Macmillan D. Exploring
neoglycoprotein assembly through native chemical ligation using
neoglycopeptide thioesters prepared via N!S acyl transfer. Org.
Biomol. Chem. 2010; 8(6): 1351–1360.
10 Kang J, Richardson JP, Macmillan D. 3-Mercaptopropionic acid-medi-
ated synthesis of peptide and protein thioesters. Chem. Commun.
2009; (9): 407–409.
11 Masania J, Li J, Smerdon SJ, Macmillan D. Access to phosphoproteins
and glycoproteins through semi-synthesis, Native chemical ligation
and N!S acyl transfer. Org. Biomol. Chem. 2010; 8(22): 5113–5119.
12 Hondal RJ, Nilsson BL, Raines RT. Selenocysteine in native chemical
ligation and expressed protein ligation. J. Am. Chem. Soc. 2001; 123
(21): 5140–5141.
13 Gieselman MD, Xie LL, van der Donk WA. Synthesis of a selenocysteine-
containing peptide by native chemical ligation. Org. Lett. 2001; 3(9):
1331–1334.
14 Quaderer R, Hilvert D. Selenocysteine-mediated backbone cycliza-
tion of unprotected peptides followed by alkylation, oxidative
elimination or reduction of the selenol. Chem. Commun. 2002;
(22): 2620–2621.
15 Hondal RJ, Raines RT. Semisynthesis of proteins containing selenocysteine.
Methods Enzymol. 2002; 347: 70–83.
16 McGrath, NA; Raines, RT, Chemoselectivity in chemical biology: acyl
transfer reactions with sulfur and selenium. Acc. Chem. Res. 2011; 44
(9): 752–761.
17 Dawson PE. Native chemical ligation combined with desulfurization
and deselenization: a general strategy for chemical protein synthesis.
Isr. J. Chem. 2011; 51(8–9): 862–867.
Conclusions
We have demonstrated that incorporation of a C-terminal seleno-
cysteine residue into synthetic peptides facilitates more rapid thioe-
ster formation when compared with the corresponding cysteinyl
peptide, presumably through an initial N!Se acyl shift. To our
knowledge, this is the first time that this process has been studied
or reported. Thioester formation in a model peptide was observed
to proceed to approximately 80% conversion within 6 h at 60 ꢀC,
as judged by HPLC, LC-MS, and 13C NMR spectroscopy.
In model studies, Sec was introduced to the reaction as the Sec
(5-Npys) derivative that prevented the formation of diselenide
dimers during resin cleavage yet could be reduced to the selenol
upon addition of TCEP. In peptides containing multiple cysteine
residues, the use of DTNP gave rise to a heterogeneous mixture
of partially and fully 5-Npys-protected peptides, which compro-
mised yields, and most likely also contained an intramolecular
selenosulfide bond. Reduction of these species did allow thioester
formation to proceed but at a much slower rate and with poor
selectivity over internal Xaa–Cys motifs. It is possible that the use
of lower temperatures (40ꢀC) and strictly anoxic conditions may
yet improve selectivity in Sec-mediated thioester formation.
The addition of sodium ascorbate to prevent deselenation
during the reaction appeared to increase hydrolysis of the product
thioester. However, when the thioester is immediately consumed in
ligation, there appeared little opportunity for hydrolysis to occur.
The advantageous antioxidant properties of ascorbate could then
be combined with accelerated thioester formation in the cycliza-
tion of peptide 20 through in situ thioester formation and NCL.
The use of selenocysteine clearly presents a challenge to thioester
formation in the presence of additional protected or unprotected
cysteine residues within the peptide. However, thioester formation
through an N!Se acyl shift appears fast and efficient provided
the tendency for Sec to interact intramolecularly or intermolecularly
with additional Sec or Cys residues, or to undergo deselenation,
can be controlled. If these well-documented obstacles can be
ultimately overcome, then incorporation of C-terminal Sec poten-
tially constitutes an extremely efficient Fmoc-based approach to
peptide thioesters for use in NCL.
18 Metanis N, Keinan E, Dawson PE. Traceless ligation of cysteine
peptides using selective deselenization. Angew. Chem. Int. Ed. 2010;
49(39): 7049–7053.
19 Koide T, Itoh H, Otaka A, Yasui H, Kuroda M, Esaki N, Soda K, Fujii N.
Synthetic study on selenocystine-containing peptides. Chem. Pharm.
Bull. 1993; 41(3), 502–506.
20 Pearson RG, Sobel H, Songstad J. Nucleophilic reactivity constants
toward methyl iodide and trans-Pt(py)2Cl2. J. Am. Chem. Soc. 1968;
90(2): 319–326.
21 Huber RE, Criddle RS. Comparison of chemical properties of seleno-
cysteine and selenocystine with their sulfur analogs. Arch. Biochem.
Biophys. 1967; 122(1): 164–173.
22 Muttenthaler M, Alewood PF. Selenopeptide chemistry. J. Pept. Sci.
2008; 14(12): 1223–1239.
23 Kang J, Reynolds NL, Tyrrell C, Dorin JR, Macmillan D. Peptide thioester
synthesis through N!S acyl-transfer: application to the synthesis of a
beta-defensin. Org. Biomol. Chem. 2009; 7(23): 4918–4923.
24 Harris KM, Flemer S, Hondal RJ. Studies on deprotection of cysteine
and selenocysteine side-chain protecting groups. J. Pept. Sci. 2007;
13(2): 81–93.
25 Rohde H, Schmalisch J, Harpaz Z, Diezmann F, Seitz O. Ascorbate as an
alternative to thiol additives in native chemical ligation. Chembiochem
2011; 12(9): 1396–1400.
26 Macmillan D, De Cecco M, Reynolds NL, Santos LFA, Barran PE, Dorin
JR. Synthesis of cyclic peptides through an intramolecular amide
bond rearrangement. Chembiochem 2011; 12(14): 2133–2136.
References
1 Dawson PE, Muir TW, Clarklewis I, Kent SBH. Synthesis of proteins by
native chemical ligation. Science 1994; 266(5186): 776–779.
2 Kent SBH. Total chemical synthesis of proteins. Chem. Soc. Rev. 2009;
38(2): 338–351.
J. Pept. Sci. 2013
Copyright © 2013 European Peptide Society and John Wiley & Sons, Ltd.
wileyonlinelibrary.com/journal/jpepsci