Accelerated Protein Synthesis via One-Pot Ligation-Deselenization Chemistry

Article abstract of DOI:10.1016/j.chempr.2017.04.003

Peptide ligation chemistry has revolutionized protein science by facilitating access to synthetic proteins. Here, we describe the development of additive-free ligation-deselenization chemistry at β-selenoaspartate and γ-selenoglutamate that enables the generation of native polypeptide products on unprecedented timescales. The deselenization step is chemoselective in the presence of unprotected selenocysteine, which is highlighted in the synthesis of selenoprotein K. The power of the methodology is also showcased through the synthesis of three tick-derived thrombin-inhibiting proteins, each of which were assembled, purified, and isolated for biological assays within a few hours. The methodology described here should serve as a powerful means of accessing synthetic proteins, including therapeutic leads, in the future.

Full text of DOI:10.1016/j.chempr.2017.04.003

Article
Accelerated Protein Synthesis via One-Pot
Ligation-Deselenization Chemistry
Nicholas J. Mitchell, Jessica
Sayers, Sameer S. Kulkarni, ...,
Bun Chan, Leo Radom, Richard
J. Payne
richard.payne@sydney.edu.au
HIGHLIGHTS
Rapid one-pot ligation-
deselenization at
b-selenoaspartate and
g-selenoglutamate
Methodology enables chemical
protein synthesis on
unprecedented timescales
Synthesis of Selenoprotein K
through chemoselective
deselenization of
b-selenoaspartate
Synthesis, purification, and
quantification of thrombin
inhibitory proteins in 3 hr
Here, we describe the development of a highly efficient one-pot ligation-
deselenization technology at aspartate and glutamate that enables the synthesis
of polypeptides and proteins on unprecedented timescales. The power of the
methodology is showcased through the rapid assembly of three thrombin-
inhibiting tick-derived proteins as well as the synthesis of the 21 kDa homodimeric
selenoprotein K. This work lays the foundation for the facile synthesis of a range of
bioactive polypeptides and proteins in the future.
Mitchell et al., Chem 2, 703–715
May 11, 2017 ª 2017 Elsevier Inc.
http://dx.doi.org/10.1016/j.chempr.2017.04.003

Article
Accelerated Protein Synthesis via
One-Pot Ligation-Deselenization Chemistry
Nicholas J. Mitchell,1,5 Jessica Sayers,1,5 Sameer S. Kulkarni,1 Daniel Clayton,1 Anna M. Goldys,1
Jorge Ripoll-Rozada,2,3 Pedro Jose´ Barbosa Pereira,2,3 Bun Chan,1,4 Leo Radom,1
and Richard J. Payne1,6,
*
SUMMARY
The Bigger Picture
A renaissance in the use of
Peptide ligation chemistry has revolutionized protein science by facilitating ac-
cess to synthetic proteins. Here, we describe the development of additive-free
ligation-deselenization chemistry at b-selenoaspartate and g-selenoglutamate
that enables the generation of native polypeptide products on unprecedented
timescales. The deselenization step is chemoselective in the presence of unpro-
tected selenocysteine, which is highlighted in the synthesis of selenoprotein K.
The power of the methodology is also showcased through the synthesis of three
tick-derived thrombin-inhibiting proteins, each of which were assembled, puri-
fied, and isolated for biological assays within a few hours. The methodology
described here should serve as a powerful means of accessing synthetic pro-
teins, including therapeutic leads, in the future.
proteins as therapeutic agents has
prompted the need for novel
technologies to rapidly generate
these biomolecules. The advent of
peptide ligation technologies has
revolutionized protein science by
facilitating access to proteins with
tailor-made modifications to
maximize specificity and activity.
However, one of the key problems
remaining in the field is the
unacceptably long time required
for the assembly of protein
INTRODUCTION
Native chemical ligation has revolutionized the field of protein science by facilitating
access to native, modified, and designer biomolecules for interrogative studies on
structure and/or function.1,2 The need for a cysteine residue (the least abundant pro-
teinogenic amino acid) on the N terminus of one of the reacting peptide fragments
has motivated the development of b-, g-, and d-thiolated variants of other amino
acids,3-14 as well as thiol-containing auxiliaries that can be used as Cys surrogates
in ligation chemistry.15-20 After ligation reactions at these residues, the thiol auxiliary
is desulfurized (usually by means of radical-based protocols21) to yield native poly-
peptide products (Scheme 1A). However, this transformation is not chemoselective
in the presence of other unprotected cysteine residues that might be found else-
where in the sequence. This limitation of desulfurization chemistry has led to expan-
sion of the native chemical ligation transformation to peptides bearing the 21st
amino acid selenocysteine (Sec),22-24 as well as selenoamino acids (specifically
Pro25 and Phe26 to date). The key advantage of carrying out ligation chemistry at se-
lenoamino acids rather than thioamino acids is that chemoselective deselenization
can be performed under mild conditions (typically with a phosphine reductant and
a hydrogen-atom source) that do not affect unprotected Cys residues.22 We27 and
Dery et al.28 have also independently demonstrated that ligation products can be
subjected to oxidative deselenization to afford Ser in place of Sec at the ligation
junction.
targets. Here, we describe the
development of a ligation-
deselenization technology at
aspartate and glutamate that
enables the synthesis of native
polypeptides and proteins on
unprecedented timescales.
Importantly, both the ligation and
the deselenization reactions are
chemoselective, proceed rapidly
and cleanly, and are compatible
for use in a one-pot regime,
features that are showcased
through the assembly of four small
protein targets. The methodology
should find use as a powerful
means of rapidly accessing
therapeutic proteins in the future.
Although there are clear benefits to using Sec or other selenoamino acids in native
chemical ligation, the rates of these reactions in the presence of an aryl thiol as a
reductant are generally slower than would be anticipated given the significantly
enhanced nucleophilicity of selenolates (in comparison with the corresponding
Chem 2, 703–715, May 11, 2017 ª 2017 Elsevier Inc. 703

Thiol additive
pH 6.0-8.0
1-48 hr
R3
O
A
X
HS R3
R1
O
H
N
SR2
peptide 1
peptide 1
peptide 2
peptide 1
+
peptide 2
N
H
N
H
H2N
O
R1
Desulfurization
ca. 6 hr
O
O
X
Se
2
O
O
Additive-free
pH 2.5-7.5
1-55 min
Deselenization
<1 min
peptide 2
X
B
H2N
O
R1
H
O
O
n
OR
N
peptide 2
peptide 1
+
SePh
N
N
H
O
O
R1
H
O
O
X
n = 1 or 2
Se
2
X = SH or Se
2
peptide 2
H2N
O
Scheme 1. Ligation-Desulfurization and Ligation-Deselenization Methods
(A) Native chemical ligation-desulfurization.
(B) The one-pot additive-free diselenide-selenoester ligation-deselenization reaction at (b-Se)-Asp
and (g-Se)-Glu reported here. When X = Se, Sec often exists as an intramolecular diselenide with
(b-Se)-Asp and (g-Se)-Glu before deselenization.
thiolates used in native chemical ligation). These slow reaction rates can be rational-
ized by the low redox potential of Sec (ꢀ381 mV),29 coupled with the weak reductive
power of aryl thiols, which provides a low steady-state concentration of selenolate
during the ligation reaction.23 One solution is to use stronger reducing agents,
e.g., a phosphine. However, this promotes homolysis of the weak C–Se bond of
Sec, a transformation that has been exploited for the chemoselective deselenization
of Sec to Ala in the presence of free Cys (see above).22,30 Reddy et al.31 have recently
reported a workable solution to this deleterious side reaction at Sec via the use of
tris(2-carboxyethyl)phosphine (TCEP) in the presence of ascorbate as a radical
trap. However, this approach prevents in situ deselenization chemistry without prior
removal of ascorbate from the reaction mixture.
We have recently reported that peptides possessing an N-terminal selenocystine
moiety (the oxidized form of Sec) can be ligated to peptides bearing a C-terminal
aryl selenoester in aqueous buffer without the use of any additives in the reaction.32
These ligation reactions were demonstrated to have unparalleled reaction rates
(1–10 min) in comparison with the corresponding thiol-mediated ligation reactions
(1–48 hr) and could be coupled with in situ radical deselenization (through the addi-
tion of TCEP and DTT) to afford native peptide products in excellent yields. Although
this new additive-free methodology provides a significant advance in ligation tech-
nology, a synthetic bottleneck still exists in the deselenization step, which typically
requires 6–16 hr to reach completion. Clearly, enhancing the rate of the deseleniza-
tion step would raise the possibility of generating native polypeptides and proteins
on unprecedented timescales. Here, we demonstrate that peptides possessing
N-terminal b-selenoaspartate ([b-Se]-Asp) or g-selenoglutamate ([g-Se]-Glu) resi-
dues can facilitate rapid and efficient additive-free ligation reactions with peptide
selenoesters (Scheme 1B). Crucially, these ligations can be coupled with one-pot
deselenization reactions that proceed cleanly in under a minute. The rapid nature
of both the additive-free ligation and deselenization reactions at Asp and Glu en-
ables the preparation of proteins in minutes, a feature that we highlight through
the synthesis of three tick-derived protein thrombin inhibitors.
1School of Chemistry, The University of Sydney,
Sydney, NSW 2006, Australia
2IBMC - Instituto de Biologia Molecular e Celular,
Universidade do Porto, 4200-135 Porto, Portugal
3Instituto de Investigac¸ao e Inovac¸ao em Saude,
˜
˜
´
Universidade do Porto, 4200-135 Porto, Portugal
4Graduate School of Engineering, Nagasaki
University, Bunkyo 1-14, Nagasaki 852-8521,
Japan
RESULTS
5These authors contributed equally
6Lead Contact
Synthesis of b-Selenoaspartate and g-Selenoglutamate Building Blocks and
Incorporation into Peptides
*Correspondence:
richard.payne@sydney.edu.au
The synthesis of a suitably protected (b-Se)-Asp building block began with electro-
philic selenylation chemistry, analogous to the sulfenylation transformation we
http://dx.doi.org/10.1016/j.chempr.2017.04.003
704 Chem 2, 703–715, May 11, 2017

A
AgNO3
PhSO2Na
acetone/H2O
O
S
O
Se
Se
2
60%
1
OMe
OMe
2
O
O
O
O
Pd(PPh3)4
PhSiH3
CH2Cl2, RT
(i) LHMDS,
THF, -78°C
(ii) 1
B
PMBSe
BocHN
PMBSe
OtBu
OtBu
OtBu
O
OH
O
BocHN
BocHN
86%
81%
O
O
3
5 (85:15 syn/anti)
4 (85:15 syn/anti)
O
OtBu
O
OtBu
O
OtBu
C
Pd(PPh3)4
PhSiH3
CH2Cl2, RT
(i) LHMDS,
THF, -78°C
(ii) 1
PMBSe
PMBSe
O
OH
O
BocHN
BocHN
BocHN
89%
89%
O
7
O
O
6
8
Scheme 2. Synthesis of b-Selenoaspartate and g-Selenoglutamate Building Blocks
Synthesis of (A) selenylating reagent 1, (B) PMB-protected b-selenyl aspartate 5, and (C) PMB-
protected g-selenyl glutamate 6.
recently reported in the synthesis of thioamino acids.4,13,14 Initially, our intention was
to incorporate a 2,4,6-trimethoxybenzyl (Tmb)-protected selenol unit into the target
amino acid. However, the instability of the Tmb-protected selenosulfonate pre-
cluded isolation. We next investigated the preparation of a less electron-rich seleny-
lating reagent, namely p-methoxybenzyl (PMB) selenosulfonate 1. This reagent was
prepared by treatment of diselenide 2 with AgNO3 in the presence of sodium ben-
zenesulfinate (Scheme 2A and Figures S103–S106). Selenylation was next effected
through the addition of selenosulfonate 1 to the dianion of orthogonally protected
Asp 3 at low temperature and provided b-Se amino acid 4 in 81% yield as an 85:15
(syn/anti) mixture of diastereomers (inseparable by flash column chromatography
but separable by C18 reverse-phase high-performance liquid chromatography
[HPLC]; see below and Scheme 2B). Finally, allyl ester deprotection was facilitated
by treatment with (tetrakis)triphenylphosphine palladium(0) and phenylsilane to
afford the desired building block 5 (85:15 syn/anti) in excellent yield (Figures
S107–S110). Similarly, suitably protected (g-Se)-Glu building block 6 could be
prepared by selenylation of Boc-Glu(OtBu)-OAll 7 and subsequent allyl ester de-
protection in excellent yield (Scheme 2C). On this occasion, 6 was prepared as a sin-
gle diastereoisomer (2S,4S as determined by NMR spectroscopy) as a result of the
exquisite stereoselectivity of the selenylation reaction on the Glu substrate (Figures
S111–S114).
The protected selenylated aspartate building block 5 was next incorporated into the
N terminus of a model peptide by Fmoc-strategy solid-phase peptide synthesis
(SPPS) to afford resin-bound 9 (see Scheme 3A and Supplemental Information).
After acidolytic side-chain deprotection and cleavage from the resin, oxidative de-
protection of the PMB selenoether-protecting group (20% DMSO in trifluoroacetic
acid [TFA]), and purification by reverse-phase HPLC, the peptide diselenide dimer
10 was isolated in 44% yield according to the original resin loading. The acidity of
the b-proton of 5 led to epimerization during coupling to afford a 1:1 (syn/anti)
mixture of diastereomers of the final peptide product 10 (Figures S1 and S2). This
was inconsequential to the purity of the final product because the stereocenter is
destined for removal through deselenization after the ligation event (Figure S3).
Chem 2, 703–715, May 11, 2017 705

A
B
Scheme 3. Synthesis of Model Peptides Incorporating N-Terminal Selenoamino Acids
Synthesis of model peptides (A) 10 bearing an N-terminal (b-Se)-Asp and (B) 12 bearing an
N-terminal (g-Se)-Glu.
We also demonstrated that the nature of the stereochemistry at the b-center does
not lead to significant differences in ligation rate. Specifically, when HPLC-separated
syn and anti diastereomers of 4 (Scheme 2B) were deprotected and subjected to an
additive-free diselenide-selenoester ligation reaction, both reached completion at
similar time points (see Supplemental Information for details). Selenylated gluta-
mate 6 could also be coupled to the N terminus of a resin-bound peptide to afford
11, which was subsequently subjected to acidolytic deprotection and cleavage from
the resin (Scheme 3B). On this occasion, PMB deprotection was best effected by
2,20-dithiobis(5-nitropyridine) (DTNP; 5.2 equiv) in TFA to afford the corresponding
selenyl sulfide, which could be removed by treatment with a solution of 40 mM ascor-
bate in phosphate buffer to provide the desired peptide target in 36% yield accord-
ing to the original resin loading (Figure S4). In contrast to selenopeptide 10, 12 was
produced as a single diastereomer (Figure S5).
One-Pot Additive-Free Diselenide-Selenoester Ligation-Deselenization
Reactions at (b-Se)-Asp and (g-Se)-Glu
With model diselenide dimer peptides 10 and 12 in hand, we next explored additive-
free ligation reactions with a range of model selenoesters (13–19). Specifically, 10 or
12 and a given peptide selenoester (see Supplemental Experimental Procedures for
selenoester synthesis and Figures S6–S12 for characterization data) were simply
dissolved in 6 M Gdn,HCl and 0.1 M phosphate buffer at a final pH of 6.2–6.5
(with no additives or pH adjustment necessary). Gratifyingly, all reactions proceeded
cleanly to afford the desired ligation products in excellent yield, as judged by ultra-
performance liquid chromatography-mass spectrometry (UPLC-MS) analysis. As
reported in our previous investigations on selenocystine-selenoester ligation,32 mul-
tiple products are observed from the ligation reaction (denoted as I–III for ligation at
(b-Se)-Asp in Table 1). Whereas symmetric diselenide I is usually observed as the
major product, together with a small amount (<10%) of asymmetric diselenide prod-
uct II, product selenoester III is observed when an excess of the selenoester frag-
ment is used in the reaction and/or when selenoesters bearing sterically hindered
C-terminal residues, e.g., Val and Leu, are used. However, the ratio of these prod-
ucts is inconsequential for the overall efficiency of the reaction because these
converge into the single native polypeptide product after deselenization. The rates
of the additive-free diselenide-selenoester ligation reactions at both (b-Se)-Asp and
(g-Se)-Glu were rapid, such that all reactions reached completion between 5 and
55 min (as judged by UPLC-MS analysis and by precipitation of diphenyldisele-
nide [DPDS], a visual prompt for completion of the reactions). Notably, ligations at
706 Chem 2, 703–715, May 11, 2017

Table 1. Scope of the One-Pot Additive-Free Diselenide-Selenoester Ligation Deselenization at
(b-Se)-Asp and (g-Se)-Glu
2
O
Se
O
1. Additive-free diselenide-
selenoester ligation
SPGYS-NH2
O
H2N
O
Ac-LYRANX-SePh (13-19)
O
10
OR
O
5-55 min
n
SPGYS-NH2
Ac
-LYRANX
O
N
2. In situ deselenization
H
O
<1 min
Se
20-26: n = 1
27-33: n = 2
2
SPGYS-NH2
12
H2N
O
2
O
PhSe
Se
O
O
O
Se
Ac-LYRANX
Se
O
O
O
O
SPGYS-NH2
SPGYS-NH2
Ac-LYRANX
Ac-LYRANX
SPGYS-NH2
III
Ac
-LYRANX
N
N
N
H
H
II
H
I
O
O
NOT ISOLATED
Entry
Peptide
Diselenide
Selenoester (X)
Ligation Time (min)
Yield for One-Pot Ligation
Deselenization
1
10
10
10
10
10
10
10
12
12
12
12
12
12
12
13: X = A
14: X = S
15: X = F
16: X = Y
17: X = M
18: X = L
19: X = V
13: X = A
14: X = S
15: X = F
16: X = Y
17: X = M
18: X = L
19: X = V
5a
20: 64%
21: 57%
22: 53%
23: 64%
24: 65%
25: 72%
26: 58%
27: 77%
28: 76%
29: 91%
30: 60%
31: 67%
32: 67%
33: 52%
2
10a
10a
5a
3
4
5
5a
6
15a
55b
10a
10a
10a
15a
15a
10a
45b
7
8
9
10
11
12
13
14
Conditions of additive-free ligation: 2.5 mM final concentration of diselenide dimer in 6 M Gdn,HCl, and
0.1 M Na2HPO4 (pH 7.2; reduced to 6.2–6.5 upon addition to peptide fragments). Conditions of one-pot
deselenization: hexane extraction (35) followed by the addition of 0.25 M TCEP, 25 mM DTT in 6 M
Gdn,HCl, and 0.1 M Na2HPO4 (pH 5–6). Products from the additive-free ligation at (b-Se)-Asp I–III (not
isolated) are shown in the box.
a0.5 equiv of H-(b-Se)DSPGYS-NH2 dimer or H-(g-Se)ESPGYS-NH2 dimer to 1.3 equiv of selenoester.
b0.5 equiv of H-(b-Se)DSPGYS-NH2 dimer or H-(g-Se)ESPGYS-NH2 dimer to 2.0 equiv of selenoester.
selenoesters bearing C-terminal Ala (13), Ser (14), Phe (15), Tyr (16), Met (17), and
Leu (18) were complete in 5–15 min, significantly faster than native chemical ligation
at Cys and the homologous b-mercapto-Asp4 and g-mercapto-Glu.14 Furthermore,
ligation with selenoester 19 with a sterically hindered C-terminal Val residue was
complete in 55 min at Asp and 45 min at Glu, again significantly faster than the thi-
olate equivalents (b-mercapto-Asp4 and g-mercapto-Glu14; 16 hr).
Upon completion of the additive-free ligations, crude reaction mixtures were sub-
jected to in situ deselenization. This involved extraction of the precipitated DPDS
with hexane and subsequent treatment with TCEP (50 equiv) and DTT (5 equiv).
Remarkably, the deselenization of (b-Se)-Asp and (g-Se)-Glu proceeded cleanly
Chem 2, 703–715, May 11, 2017 707

and to completion (in all cases) within 60 s, whereas deselenization at Sec
requires 4–6 hr (see Figures S13–S40 for data). Interestingly, the deselenization
was also complete within a minute at a wide pH range (1.8–7.0; see Figures S54–
S58). The resulting products from the one-pot additive-free diselenide-selenoester
ligation-deselenization reactions were subsequently purified by reverse-phase
HPLC to afford the native peptide products 20–26 for ligations at (b-Se)-Asp
and 27–33 for ligations at (g-Se)-Glu in excellent yield (52%–91%) over the two
steps (Table 1 and Figures S13–S40). Importantly, the exceptional rates of the addi-
tive-free ligation and deselenization reactions at (b-Se)-Asp and (g-Se)-Glu enable
access to native peptides in minutes (including at sterically hindered junctions), a
timescale that, to our knowledge, cannot be achieved with currently available
techniques.
Ligation-Deselenization Using Additives
Very recently, it has been reported that native chemical ligation reactions at Sec
(using peptide thioesters as the acyl donor) can be performed in the presence of
the reductant TCEP (which normally facilitates deselenization) through the addition
of ascorbic acid.31 We were therefore interested in assessing whether (b-Se)-Asp
could be used in additive ligations through a native chemical ligation pathway. To
this end, we reacted 10 with peptide selenoester 19 in the presence of 50 mM
TCEP and 100 mM ascorbic acid in ligation buffer.31 Unfortunately, these conditions
resulted in complete deselenization of starting diselenide dimer peptide 10 with no
detectable ligation product (Figures S41 and S42). Given that these conditions pre-
vent deselenization of Sec-containing peptides (see Reddy et al.31 and Figures S43
and S44), this observation reflects the increased lability of the C–Se bond in the
(b-Se)-Asp moiety (see above). We therefore sought to optimize additive conditions
by using an alternative radical trap, namely DPDS, in place of ascorbic acid. Opti-
mized conditions involved ligation between 10 and 19 at a saturating concentration
of DPDS (250 mM) and 25 mM TCEP (Figures S45–S47). Although a significant
amount of the desired product was formed (in deselenized form), deselenization
of starting peptide 10 could not be prevented entirely, thus lowering the overall
yield of the reaction.
Mechanistic Insight into the Rapid Deselenization Reaction
Having established that ligation reactions at (b-Se)-Asp and (g-Se)-Glu must be
performed strictly under an additive-free regime, we moved to further explore the
exceptional rate enhancement observed for deselenization at these amino acids
over Sec (see Supplemental Information). The current mechanistic model for desele-
nization using a phosphine (e.g., TCEP) and a hydrogen-atom source (e.g., DTT) in-
vokes an initial reduction of the diselenide (or selenyl sulfide for native chemical
ligation) to afford a selenol (see A in Scheme 4), which could serve as a precursor
to a small amount of Se-centered radical B.22,28 It is also feasible that B could
be generated from the starting diselenide.33 Regardless of the pathway to the sele-
nium-centered radical, B could react rapidly with the phosphine to generate the
phosphorus-centered radical species C. C–Se bond homolysis of C would generate
a b-carbon-centered radical D and phosphine selenide E (proposed to be a key
driving force for the reaction). Hydrogen-atom abstraction by the b-carbon-centered
radical D could then produce the native amino acid F. Interestingly, we have demon-
strated that deselenization reactions proceed to completion (and at similar rates) at
(b-Se)-Asp even in the absence of an H-atom source such as DTT (Figure S59). This
suggests that H-atom abstraction could be possible from selenol A (produced by
TCEP reduction of the starting diselenide), which would regenerate B and propagate
a radical chain as depicted in Scheme 4.
708 Chem 2, 703–715, May 11, 2017

Scheme 4. Putative Pathways for the Deselenization of Selenoamino Acids
ꢀ
For TCEP, R = CH2CH2CO2
.
To help understand the significant rate enhancement observed for deselenization,
initially at (b-Se)-Asp, we probed the energies associated with the formation of
the proposed intermediates with computational quantum chemistry calculations.
These were performed with the Gaussian 09 program34 with the species depicted
in Scheme 4 as models (Table S1). Gas-phase energies were obtained at the
DSD-PBEP86/aug0-cc-pVTZ level,35 and the effect of solvation was incorporated
through the SMD continuum model at the M05-2X/6-31G(d) level. Our calculated
energies corresponding to the individual steps for the pathways shown in Scheme 4,
together with a schematic energy profile, are included in Tables S1 and S2 and Fig-
ure S115. Thus, the barrier calculated for C / D for the unsubstituted system
ꢀ
(X = H) is 23.4 kJ molꢀ1, whereas that for the aspartate derivative (X = CO2
)
is ꢀ5.3 kJ molꢀ1, the negative value of which indicates that this modification essen-
tially removes the barrier (Table S1 and Figure S115). This can be attributed to
stabilization of the electron-deficient radical center by the anionic b-carboxylate
moiety. Corresponding calculated spin densities for species involved in the
conversion of C to D are consistent with the energy data (Figure S116). In addition,
abstraction of the H atom from DTT by carbon-centered radical D is predicted to
have a lower barrier of 5.3 kJ molꢀ1 for (b-Se)-Asp (X = CO2ꢀ) than Sec (X = H,
10.9 kJ molꢀ1). We performed the analogous computational experiments for the
proposed pathway in Scheme 4 with (g-Se)-Glu (Tables S1 and S2 and Figure S115).
Perhaps unsurprisingly, the key barrier for cleavage of the C–Se bond in TCEP
adduct C to afford carbon-centered radical D was very similar to that calculated
for the Asp homolog (Table S1 and Figure S115), which agrees with our experimental
observation, i.e., that deselenization at both (b-Se)-Asp and (g-Se)-Glu is rapid.
Selectivity of Deselenization
Given the expedient nature of the deselenization step, we were interested in
probing whether selective deselenization of (b-Se)-Asp and (g-Se)-Glu would be
possible without concomitant conversion of Sec to Ala. Toward this end, we first
demonstrated that (b-Se)-Asp and (g-Se)-Glu could be selectively deselenized in a
model peptide that also possessed an unprotected Sec (Figures S48–S53). Having
demonstrated the chemoselectivity on a model system, we next embarked on the
assembly of a protein target that would benefit from the chemoselective deseleniza-
tion transformation. For this purpose, we chose the 93 amino acid protein seleno-
protein K (SelK) 34 (residues 2–94), possessing a Sec residue at position 92 that
cannot be used for assembling the protein via ligation because of its proximity to
Chem 2, 703–715, May 11, 2017 709

A
Se
O
Se
OOC
H2N
O
H
60
N
+
SelK (2-59)
SePh
GR-OH
H
SelK (62-91)
N
92
61
H
O
36
35
1. Additive-free
ligation:
HO
1. Additive-free ligation: 6 M Gdn•HCl
0.1 M Na2HPO4, pH 6.0, 15 min
6 M Gdn•HCl,
62% over 2 steps
0.1 M Na2HPO4,
pH 6.0, 15 min
2. Hydrazinolysis:
2 vol.% hydrazine
3. Chemoselective
deselenization:
aq. TCEP (50 equiv),
pH 5.1, 2 min
2. Hydrazinolysis: 2 vol.% hydrazine
Se
O
Se
OOC
O
H
N
60
GR-OH
SelK (62-91)
N
92
SelK (2-59)
H
N
61
H
H
O
37
HO
Chemoselective deselenization:
40%
over 3 steps
TCEP (50 equiv), 6 M Gdn•HCl
84%
0.1 M Na2HPO4, pH 5.0, 2 min
2
COO
Se
O
H
N
60
Mtb CM (42-83)
SelK (62-91)
GR-OH
SelK (2-59)
N
H
N
92
61
H
H
O
O
Selenoprotein K (SelK) 34
HO
B
C
Scheme 5. Synthesis of Selenoprotein K
(A) Synthesis of homodimeric SelK via additive-free diselenide-selenoester ligation at (b-Se)-Asp
and subsequent chemoselective deselenization in the presence of native and unprotected Sec92.
(B) HPLC trace of purified synthetic SelK after the one-pot ligation-deselenization protocol
(26.6 min at room temperature, l = 220 nm).
(C) Electrospray ionization mass spectrum of synthetic SelK homodimer 34 (20,954 Da).
the C terminus. Unlike most selenoproteins that possess selenosulfide linkages, SelK
exists as a homodimer linked via an intermolecular diselenide at Sec92.36 Although
the exact biological role of SelK has not been established, Sec92 and the high redox
potential of the intermolecular diselenide bond are thought to be important for pro-
tein function.37 Assembly of peptide selenoester 35 and peptide 36 bearing an
N-terminal (b-Se)-Asp and an internal Sec residue (linked as an intramolecular dise-
lenide) was first performed by Fmoc-SPPS methods (Scheme 5, Supplemental Infor-
mation, and Figures S60–S63). Peptide 36 was prepared with two norleucines
substituted for methionine (Met) residues owing to significant but incomplete oxida-
tion of the thioether side chains of Met during the acidic deprotection and cleavage
conditions of the peptide fragments that complicated analysis. We performed liga-
tion under the additive-free conditions by simply dissolving the two fragments in
aqueous ligation buffer (adjusted to pH 6.0) by using a 2-fold excess of selenoester
35. Pleasingly, the reaction proceeded to completion in 15 min (as judged by UPLC-
MS analysis) to afford a mixture of intramolecular diselenide 37 and the selenoester-
linked ligation product. After hydrazinolysis of the unproductive selenoesters, 37
was afforded as the exclusive ligation product (Figure S64). The intramolecular dis-
elenide ligation product 37 was subsequently isolated in 62% yield after reverse-
phase HPLC (Figures S65 and S66). Gratifyingly, treatment of 37 with TCEP, in the
absence of DTT, for 2 min led to chemoselective deselenization of the (b-Se)-Asp
without any observed Sec deselenization. Purification subsequently afforded SelK
34 in 84% yield (Figure S67). Mass spectrometric analysis confirmed that 34 was
710 Chem 2, 703–715, May 11, 2017

isolated as the homodimer with a molecular weight of 21 kDa (Figure S68).38 Having
successfully showcased the chemoselective deselenization in the synthesis of SelK
with intermediary purification, we next attempted to rapidly access the selenopro-
tein by using a one-pot protocol (Scheme 5A). Toward this end, rapid additive-
free ligation between 35 and 36, followed by in situ hydrazine treatment and chemo-
selective deselenization via treatment with TCEP, provided SelK together with the
acyl hydrazide of selenoester fragment 35 (Figure S69). After reverse-phase HPLC
purification, homodimeric SelK 34 was isolated in 40% yield and in excellent purity
over the three synthetic steps (Schemes 5B and 5C and Figures S70 and S71).
One-Pot Synthesis of the Hyalomins via Ligation Deselenization
In our previously described selenocystine-selenoester ligation,32 the deselenization
of Sec was the synthetic bottleneck, requiring 4–6 hr to reach completion. This step
prevented full exploitation of the exceptional rate of the additive-free ligation reac-
tion. We envisaged that the increase in deselenization rate at (b-Se)-Asp, coupled
with the fast additive-free ligation rates at this residue, would provide a unique
means to access target proteins by chemical synthesis within hours rather than
days, a feature that was showcased in the synthesis of SelK. To investigate this
possibility further, we sought to prepare a selection of small thrombin-inhibiting
hyalomin proteins (hyalomin-2, hyalomin-3, and hyalomin-4) by using our methodol-
ogy. The hyalomins are a family of four cysteine-free proteins produced within the
salivary glands of the tick Hyalomma marginatum rufipes that support the blood-
feeding behavior of the organism.39 The absence of Ala residues at a site that would
permit assembly through other ligation approaches, together with the wealth of
acidic residues within the sequences (pI < 4), made these amenable to assembly
with our ligation technology. Disconnection of hyalomins 2–4 (38–40) was made at
Asp residues close to the middle of the sequences. The C-terminal diselenide dimer
peptide fragments 41–43 were synthesized on 2-chlorotrityl chloride resin via Fmoc-
SPPS with the incorporation of (b-Se)-Asp building block 5 at the N terminus (Figures
S72–S77). N-terminal peptides were also prepared on 2-chlorotrityl chloride resin
via Fmoc-SPPS and converted to C-terminal phenylselenoesters 44–46 (Figures
S78–S83).
With the requisite fragments in hand, we next set out to prepare each of the hyalo-
min targets with an emphasis on synthesis, purification, and isolation within a short
time frame. Toward this end, diselenide dimer fragments 41–43 (1 equiv on the basis
of the monomeric peptide) were reacted with peptide selenoesters 44–46 (1.5 equiv
for 44, 1.6 equiv for 46, and 2 equiv for 45) in ligation buffer (pH adjusted to 6.2) un-
der additive-free conditions (Scheme 6). Gratifyingly, the reactions proceeded in just
2 min to afford the desired ligation products as a mixture of the symmetrical disele-
nide dimer and product bearing a selenoester linkage (as judged by UPLC-MS anal-
ysis). Subsequent deselenization was effected via the addition of TCEP (50 equiv)
and DTT (5 equiv) and, like the model systems, proceeded smoothly and to comple-
tion within 1 min (Figures S84–S89 and S94). The DTT was added to the reaction to
thiolyze the product selenoester to the corresponding selenol, which could be sub-
sequently deselenized with TCEP. The crude hyalomins were purified by reverse-
phase HPLC over a 30 min gradient, and fractions containing the desired protein
were analyzed with a UPLC-MS system with a gradient of 3 min, thus allowing
numerous samples to be run over a short period. The relevant fractions were pooled
and the solvent removed on a Genevac solvent evaporation system over 1 hr at 50ꢁC
and 3 mbar. The pure proteins were re-dissolved in distilled water and the concen-
tration confirmed by NanoDrop UV-visible spectrophotometry to determine con-
centrations of the protein solutions (for direct use in thrombin inhibitory assays).
Chem 2, 703–715, May 11, 2017 711

1. Additive-free ligation: 6 M Gdn•HCl,
0.1 M Na2HPO4, pH 6.2, 2 min
2. DPDS removal: hexane extraction
3. Deselenization: TCEP (50 equiv),
DTT (5 equiv), pH 5.0, 1 min
A
O
H
23
N
Hyal2 (1-22)
SePh
H
COO
Hyal2 (25-58)
O
44
H
23
Mtb CM (42-83)
OH
N
4. (i) HPLC, (ii) concentration,
+
Hyal2 (1-22)
H
N
24
(iii) quantification
H
Se
COO
O
2
65% isolated yield
3 hr total for all steps
Hyal2 (25-58)
OH
Hyalomin 2 (38)*
H2N
24
O
41
1. Additive-free ligation: 6 M Gdn•HCl,
0.1 M Na2HPO4, pH 6.2, 2 min
2. DPDS removal: hexane extraction
3. Deselenization: TCEP (50 equiv),
DTT (5 equiv), pH 5.0, 1 min
B
O
H
25
N
Hyal3 (1-24)
SePh
H
COO
O
OH
H
45
4. (i) HPLC, (ii) concentration,
25
Hyal3 (27-59)
Mtb CM (42-83)
OH
N
Hyal3 (1-24)
H
(iii) quantification
N
26
+
H
O
Se
COO
OH
67% isolated yield
2
3 hr total for all steps
Hyalomin 3 (39)
Hyal3 (27-59)
OH
H2N
26
O
42
1. Additive-free ligation: 6 M Gdn•HCl,
0.1 M Na2HPO4, pH 6.2, 2 min
2. DPDS removal: hexane extraction
3. Deselenization: TCEP (50 equiv),
DTT (5 equiv), pH 5.0, 1 min
C
O
O
H
25
N
Hyal4 (1-24)
SePh
NH2
H
COO
O
H
46
4. (i) HPLC, (ii) concentration,
25
Hyal4 (27-51)
Mtb CM (42-83)
OH
N
Hyal4 (1-24)
H
N
(iii) quantification
26
+
H
O
Se
COO
66% isolated yield
NH2
2
3 hr total for all steps
Hyal4 (27-51)
OH
O
H2N
26
Hyalomin 4 (40)
O
43
D
E
[M+5H]5+
[M+6H]6+
[M+7H]7+
[M+4H]4+
Time (min)
Time (min)
Scheme 6. Rapid Assembly of Hyalomin-2, Hyalomin-3, and Hyalomin-4 via One-Pot Ligation-Deselenization Chemistry
(A–C) Synthesis of thrombin-inhibiting proteins (A) hyalomin 2, (B) hyalomin 3, and (C) hyalomin 4 via one-pot additive-free diselenide-selenoester
ligation at (b-Se)-Asp. All products were synthesized, purified, characterized, and quantified within a 3 hr time period. *Hyalomin 2 (38) was produced
with Met18 oxidized. The sulfoxide could be reduced in a subsequent step to afford the native protein in 94% yield (see Supplemental Information for
details).
(D) Crude HPLC of one-pot ligation deselenization to afford hyalomin 2 (38) (23.5 min at room temperature, l = 280 nm).
(E) HPLC of purified hyalomin 2 (38) (23.5 min at room temperature, l = 280 nm). Inset: the electrospray ionization mass spectrum of hyalomin 2 (38).
After the one-pot diselenide-selenoester ligation deselenization and purification,
the hyalomins were isolated in excellent yields (65%–67%). After ligation of the
peptide fragments, purification, characterization, solvent removal, and quantifica-
tion of the pure proteins, each of the hyalomins were generated within an impres-
sively brief 3 hr period in >98% purity (Figures S84–S98). Finally, having accessed
712 Chem 2, 703–715, May 11, 2017

hyalomins 2–4 (38–40), we next assessed the activity of the synthetic proteins as
inhibitors of human thrombin (see Supplemental Information for details). Both hya-
lomin 2 (38) and hyalomin 3 (39) proved to be extremely potent thrombin inhibitors
with inhibition constants (Ki) of 1.24 G 0.05 nM and 14.73 G 0.64 nM, respectively
(Figures S99–S102). Surprisingly, hyalomin 4 (40) exhibited weaker inhibitory activity
(half maximal inhibitory concentration = 20 mM), possibly because the C-terminal tail
of this protein is significantly shorter than those of 38 and 39. Future work in our lab-
oratories will focus on determining the three-dimensional structure of thrombin-
hyalomin complexes to elucidate their inhibitory binding mode.
DISCUSSION
In summary, we have developed a short and efficient synthesis of suitably protected
(b-Se)-Asp and (g-Se)-Glu building blocks. We have demonstrated that these seleny-
lated amino acids can be incorporated into resin-bound peptides and facilitate
rapid, additive-free ligation reactions with peptide selenoesters. After the ligation
event, and without purification, deselenization of the b- and g-seleno auxiliaries
was smoothly effected within 1 min in all cases. The impressive rates of deseleniza-
tion at (b-Se)-Asp were highlighted in the synthesis of SelK, whereby the b-seleno
auxiliary on Asp could be chemoselectively deselenized in the presence of Sec to
afford the native selenoprotein. Furthermore, the rapid rates of both the ligation
and deselenization steps were showcased in the one-pot assembly of three
thrombin-inhibiting proteins from the hyalomin family, which could all be assem-
bled, purified, and isolated within 3 hr for immediate assessment in thrombin inhibi-
tion assays. The simplicity, efficiency, and speed of the ligation-deselenization
chemistry described here should see the technology applied to the synthesis of
numerous protein targets and protein libraries.
EXPERIMENTAL PROCEDURES
General Procedure for One-Pot Additive-Free Diselenide-Selenoester
Ligation Deselenization
The diselenide dimer peptide 10 [H-(b-Se)DSPGYS-NH2]2 (2.0 mg, 1.4 mmol, 2.8 mmol
of H-(b-Se)DSPGYS-NH2 monomer) or diselenide dimer 12 [H-(g-Se)ESPGYS-NH2]2
(1.0 mg, 0.7 mmol, 1.4 mmol of H-(g-Se)ESPGYS-NH2 monomer) and a selenoester
Ac-LYRANX-SePh (13–19) (3.6–5.6 mmol for reaction with 10 or 1.8–2.8 mmol for reac-
tion with 12) were separately dissolved in ligation buffer (6 M Gdn,HCl and 0.1 M
Na2HPO4 [pH 7.1]) to a concentration of 10 mM (with respect to the selenopeptide
fragment) and 13 mM (with respect to the selenoester fragment or 20 mM in the ex-
amples with Ac-LYRANV-SePh). The selenoester solution was added in one portion to
the solution of diselenide in an Eppendorf tube, and the reaction mixture was left at
room temperature with intermittent agitation. Analytical HPLC-mass spectrometry
(HPLC-MS) analysis indicated consumption of the diselenide and formation of liga-
tion products at the times indicated in Table 1. The ligation reaction mixture was
washed with an equal volume of hexane (35) for the removal of DPDS and sparged
with nitrogen for 5–10 min. Separately, a solution of TCEP (0.25 M) and DTT
(25 mM) was prepared in ligation buffer, and the pH was adjusted to 5–6. An equal
volume of the TCEP-DTT solution was added in one portion to the ligation reaction
mixture, and the reaction was left for 5 min (after 1 min of HPLC-MS analysis had indi-
cated that the ligation products had already been consumed and the deselenized
peptide had been formed). The reaction mixture was diluted with water containing
0.1% TFA, purified by preparative HPLC, and lyophilized to give the native peptide
product as a white solid (see Supplemental Information for UPLC data of crude reac-
tions and characterization data of purified peptide products).
Chem 2, 703–715, May 11, 2017 713

SUPPLEMENTAL INFORMATION
Supplemental Information includes Supplemental Experimental Procedures, 116
figures, and 2 tables and can be found with this article online at http://dx.doi.org/
10.1016/j.chempr.2017.04.003.
AUTHOR CONTRIBUTIONS
Conceptualization, P.J.B.P., L.R., and R.J.P.; Methodology, N.J.M., J.S., S.S.K., D.C.,
A.M.G., J.R.-R., B.C., and R.J.P.; Investigation, N.J.M., J.S., S.S.K., D.C., A.M.G.,
J.R.-R., and B.C.; Writing – Original Draft, N.J.M., L.R., and R.J.P. Writing – Review &
Editing, N.J.M., J.S., S.S.K., D.C., A.M.G., J.R.-R., B.C., P.J.B.P., L.R., and R.J.P.;
Funding Acquisition, P.J.B.P., L.R., and R.J.P.; Supervision, P.J.B.P., L.R., and R.J.P.
ACKNOWLEDGMENTS
We acknowledge funding from Australian Research Council Discovery Projects
(DP160101324 and DP150101425), a Northcote Scholarship for PhD funding (J.S.),
Fundac¸a˜o para a Cieˆ ncia e a Tecnologia (Portugal) through a postdoctoral fellowship
(SFRH/BPD/108004/2015 to J.R.-R.), and generous grants for supercomputer time
from the National Computational Infrastructure National Facility, Intersect Australia
Ltd., and the RIKEN Advanced Center for Computing and Communication and Cen-
ter for Integrative Medical Sciences in Japan (Q16266). We would also like to thank
Professor Chris Easton (Australian National University) for helpful discussions.
Received: October 3, 2016
Revised: March 11, 2017
Accepted: April 11, 2017
Published: May 11, 2017
REFERENCES AND NOTES
1. Dawson, P.E., Muir, T.W., Clark-Lewis, I., and
Kent, S.B.H. (1994). Synthesis of proteins by
native chemical ligation. Science 266, 776–779.
9. Yang, R., Pasunooti, K.K., Li, F., Liu, X.W., and
Liu, C.F. (2009). Dual native chemical ligation at
lysine. J. Am. Chem. Soc. 131, 13592–13593.
ligation–desulfurization chemistry. Aust. J.
Chem. 68, 521–537.
17. Malins, L.R., and Payne, R.J. (2014). Recent
extensions to native chemical ligation for the
chemical synthesis of peptides and proteins.
Curr. Opin. Chem. Biol. 22, 70–78.
2. Kent, S.B.H. (2009). Total chemical synthesis of
proteins. Chem. Soc. Rev. 38, 338–351.
10. Harpaz, Z., Siman, P., Kumar, K.S., and Brik, A.
(2010). Protein synthesis assisted by native
chemical ligation at leucine. Chembiochem 11,
1232–1235.
3. Malins, L.R., Cergol, K.M., and Payne, R.J.
(2013). Peptide ligation-desulfurization
chemistry at arginine. Chembiochem 14,
559–563.
18. Hackenberger, C.P.R., and Schwarzer, D.
(2008). Chemoselective ligation and
modification strategies for peptides and
proteins. Angew. Chem. Int. Ed. 47, 10030–
10074.
11. Shang, S., Tan, Z., Dong, S., and Danishefsky,
S.J. (2011). An advance in proline ligation.
J. Am. Chem. Soc. 133, 10784–10786.
4. Thompson, R.E., Chan, B., Radom, L., Jolliffe,
K.A., and Payne, R.J. (2013). Chemoselective
peptide ligation-desulfurization at aspartate.
Angew. Chem. Int. Ed. 52, 9723–9727.
12. Malins, L.R., Cergol, K.M., and Payne, R.J.
(2014). Chemoselective sulfenylation and
peptide ligation at tryptophan. Chem. Sci. 5,
260–266.
19. Payne, R.J., and Wong, C.H. (2010). Advances
in chemical ligation strategies for the synthesis
of glycopeptides and glycoproteins. Chem.
Commun. 46, 21–43.
5. Dawson, P.E. (2011). Native chemical ligation
combined with desulfurization and
deselenization: a general strategy for chemical
protein synthesis. Isr. J. Chem. 51, 862–867.
13. Cergol, K.M., Thompson, R.E., Malins, L.R.,
Turner, P., and Payne, R.J. (2014). One-pot
peptide ligation-desulfurization at glutamate.
Org. Lett. 16, 290–293.
20. Loibl, S.F., Harpaz, Z., and Seitz, O. (2015). A
type of auxiliary for native chemical peptide
ligation beyond cysteine and glycine junctions.
Angew. Chem. Int. Ed. 54, 15055–15059.
6. Crich, D., and Banerjee, A. (2007). Native
chemical ligation at phenylalanine. J. Am.
Chem. Soc. 129, 10064–10065.
14. Sayers, J., Thompson, R.E., Perry, K.J., Malins,
L.R., and Payne, R.J. (2015). Thiazolidine-
protected b-thiol asparagine: applications in
one-pot ligation–desulfurization chemistry.
Org. Lett. 17, 4902–4905.
21. Wan, Q., and Danishefsky, S.J. (2007). Free-
radical-based, specific desulfurization of
cysteine: a powerful advance in the synthesis of
polypeptides and glycopolypeptides. Angew.
Chem. Int. Ed. 46, 9248–9252.
7. Chen, J., Wang, P., Zhu, J., Wan, Q., and
Danishefsky, S.J. (2010). A program for ligation
at threonine sites: application to the controlled
total synthesis of glycopeptides. Tetrahedron
66, 2277–2283.
15. Bondalapati, S., Jbara, M., and Brik, A. (2016).
Expanding the chemical toolbox for the
synthesis of large and uniquely modified
proteins. Nat. Chem. 8, 407–418.
8. Chen, J., Wan, Q., Yuan, Y., Zhu, J., and
Danishefsky, S.J. (2008). Native chemical
ligation at valine. Angew. Chem. Int. Ed. 47,
8521–8524.
22. Metanis, N., Keinan, E., and Dawson, P.E.
(2010). Traceless ligation of cysteine peptides
using selective deselenization. Angew. Chem.
Int. Ed. 49, 7049–7053.
16. Malins, L.R., and Payne, R.J. (2015). Synthetic
amino acids for applications in peptide
714 Chem 2, 703–715, May 11, 2017

23. Gieselman, M.D., Xie, L., and van Der Donk,
W.A. (2001). Synthesis of a selenocysteine-
containing peptide by native chemical ligation.
Org. Lett. 3, 1331–1334.
29. Besse, D., Siedler, F., Diercks, T., Kessler,
H., and Moroder, L. (1997). The redox
potential of selenocystine in unconstrained
cyclic peptides. Angew. Chem. Int. Ed. 36,
883–885.
Petersson, G.A., et al. (2009). Gaussian 09,
Revision C.03 (Gaussian, Inc.).
35. Kozuch, S., and Martin, J.M.L. (2011). DSD-
PBEP86: in search of the best double-hybrid
DFT with spin-component scaled MP2 and
dispersion corrections. Phys. Chem. Chem.
Phys. 13, 20104–20107.
24. Quaderer, R., Sewing, A., and Hilvert, D. (2001).
Selenocysteine-mediated native chemical
ligation. Helv. Chim. Acta 84, 1197–1206.
30. Malins, L.R., Mitchell, N.J., and Payne, R.J.
(2014). Peptide ligation chemistry at selenol
amino acids. J. Pept. Sci. 20, 64–77.
36. Liu, J., Zhang, Z., and Rozovsky, S. (2014).
Selenoprotein K form an intermolecular
diselenide bond with unusually high
redox potential. FEBS Lett. 588, 3311–
3321.
25. Townsend, S.D., Tan, Z., Dong, S., Shang, S.,
Brailsford, J.A., and Danishefsky, S.J. (2012).
Advances in proline ligation. J. Am. Chem. Soc.
134, 3912–3916.
31. Reddy, P.S., Dery, S., and Metanis, N. (2016).
Chemical synthesis of proteins with non-
strategically placed cysteines using
selenazolidine and selective deselenization.
Angew. Chem. Int. Ed. 55, 992–995.
26. Malins, L.R., and Payne, R.J. (2012). Synthesis
and utility of b-selenol-phenylalanine for native
chemical ligation–deselenization chemistry.
Org. Lett. 14, 3142–3145.
37. Liu, J., and Rozovsky, S. (2015). Membrane-
bound selenoproteins. Antioxid. Redox Signal.
23, 795–813.
32. Mitchell, N.J., Malins, L.R., Liu, X., Thompson,
R.E., Chan, B., Radom, L., and Payne, R.J.
(2015). Rapid additive-free selenocystine-
selenoester peptide ligation. J. Am. Chem.
Soc. 137, 14011–14014.
27. Malins, L.R., Mitchell, N.J., McGowan, S., and
Payne, R.J. (2015). Oxidative deselenization of
selenocysteine: applications for programmed
ligation at serine. Angew. Chem. Int. Ed. 54,
12716–12721.
38. Liu, J., Srinivasan, P., Pham, D.N., and
Rozovsky, S. (2012). Expression and purification
of the membrane enzyme selenoprotein K.
Protein Expr. Purif 86, 27–34.
33. Ji, S., Cao, W., Yu, Y., and Xu, H. (2014).
Dynamic diselenide bonds: exchange
39. Jablonka, W., Kotsyfakis, M., Mizurini, D.M.,
Monteiro, R.Q., Lukszo, J., Drake, S.K.,
Ribeiro, J.M.C., and Andersen, J.F. (2015).
Identification and mechanistic analysis of a
novel tick-derived inhibitor of thrombin. PLoS
One 10, e0133991.
reaction induced by visible light without
catalysis. Angew. Chem. Int. Ed. 53, 6781–6785.
28. Dery, S., Reddy, P.S., Dery, L., Mousa, R.,
Dardashti, R.N., and Metanis, N. (2015).
Insights into the deselenization of
selenocysteine into alanine and serine. Chem.
Sci. 6, 6207–6212.
34. Frisch, M.J., Trucks, G.W., Schlegel, H.B.,
Scuseria, G.E., Robb, M.A., Cheeseman, J.R.,
Scalmani, G., Barone, V., Mennucci, B.,
Chem 2, 703–715, May 11, 2017 715