Application of 2,2′-dipyridyl disulfide-mediated thiazolidine ring-opening reaction to glycoprotein synthesis: Total chemical synthesis of evasin-3

Article abstract of DOI:10.1002/psc.3290

Thiazolidine ring-opening reaction is one of the key steps in protein chemical synthesis via sequential native chemical ligation strategy. We recently developed a novel thiazolidine ring-opening reaction with 2,2′-dipyridyl disulfide (DPDS). In order to i

Full text of DOI:10.1002/psc.3290

Received: 29 July 2020
DOI: 10.1002/psc.3290
Revised: 10 September 2020
Accepted: 22 September 2020
R E S E A R C H A R T I C L E
0
Application of 2,2 -dipyridyl disulfide-mediated thiazolidine
ring-opening reaction to glycoprotein synthesis: Total chemical
synthesis of evasin-3
1
2
Hidekazu Katayama
| Koji Nagata
1
Department of Applied Biochemistry, School
of Engineering, Tokai University, Hiratsuka,
Japan
Thiazolidine ring-opening reaction is one of the key steps in protein chemical synthe-
sis via sequential native chemical ligation strategy. We recently developed a novel
2
Department of Applied Biological Chemistry,
0
thiazolidine ring-opening reaction with 2,2 -dipyridyl disulfide (DPDS). In order to
Graduate School of Agricultural and Life
Sciences, The University of Tokyo, Bunkyo,
Japan
investigate the applicability of this reaction to glycoprotein synthesis, we synthesized
evasin-3, a cysteine-rich glycoprotein with chemokine-binding ability originally found
in tick saliva. The sequence of evasin-3 was divided into three segments, and these
segments were separately synthesized with the ordinary solid-phase peptide synthe-
sis method. After the first ligation of middle and C-terminal segments, thiazolidine
used as a protecting group of Cys residue at the N-terminus of the middle segment
was converted to Cys with DPDS. In this thiazolidine ring-opening reaction, DPDS
treatment did not affect the N-linked glycan moiety. After the second ligation with
the N-terminal segment and the refolding reaction, evasin-3 could be obtained in
good yield. The synthetic evasin-3 showed the binding ability specifically to CXCL
chemokines. These results clearly indicate that this DPDS method is useful for glyco-
protein synthesis.
Correspondence
Hidekazu Katayama, Department of Applied
Biochemistry, School of Engineering, Tokai
University, 4-1-1 Kitakaname, Hiratsuka,
Kanagawa 259-1292, Japan.
Email: katay@tokai-u.jp
Funding information
KAKENHI, Grant/Award Number: 18 K05113
K E Y W O R D S
0
2
,2 -dipyridyl disulfide, evasin-3, native chemical ligation, thiazolidine
1
|
INTRODUCTION
middle segment, the N-terminal Cys residue or C-terminal thioester
functionality should be protected. Among the Cys protecting groups,
thiazolidine-4-carboxilic acid (Thz) has been widely used as a Cys sur-
Protein chemical synthesis is an indispensable tool to promote biologi-
cal and biochemical researches. Because the ordinary solid-phase pep-
tide synthesis (SPPS) method developed by Merrifield is usually
4
rogate. It is well known that the N-terminal Thz can be easily
converted to Cys with the treatment of methoxyamine under weakly
acidic conditions, and the newly generated N-terminal Cys can be
afforded to the NCL with the N-terminal segment having α-thioester
1
limited to the length of 40–50 amino acid residues, ligation methods
of peptides that are prepared with SPPS are widely used for protein
chemical synthesis. In the native chemical ligation (NCL), one of the
peptide ligation methods, a peptide α-thioester and an N-terminally
cysteinyl peptide can be site-specifically condensed without any
4
functionality. However, the complete conversion of Thz to Cys with
methoxyamine generally requires a reaction time over 8 h. To shorten
the reaction time, water-soluble palladium(II) complex-mediated ring-
2
protecting group. The NCL method provides an efficient means to
2 2
opening method was developed. In this method, [Pd (allyl)Cl] or PdCl
generate proteins, and many proteins have been synthesized by this
was used in the presence of 4-mercaptophenylacetic acid (MPAA) and
tris(2-carboxyethyl)phosphine (TCEP), and the Thz could be
3
method.
5
When three or more peptide segments are condensed through
NCL strategy, an N-terminally cysteinyl peptide α-thioester should be
used as a middle segment. To prevent the self-cyclization of the
completely converted to Cys within 15 min. On the other hand, it is
problematic that palladium(II) complexes also removed other Cys
protecting groups such as acetamidomethyl (Acm) and
J Pep Sci. 2020;e3290.
wileyonlinelibrary.com/journal/psc
© 2020 European Peptide Society and John Wiley & Sons, Ltd.
1 of 10
https://doi.org/10.1002/psc.3290

2
of 10
KATAYAMA AND NAGATA
6,7
allyloxycarbonylaminomethyl (Allocam) groups.
More recently,
8
copper-mediated Thz deprotection method has been reported,
whereas the compatibility of this method with the other Cys
protecting groups still remains unclear.
0
In our previous study, we reported that 2,2 -dipyridyl disulfide
(
(
DPDS) was a good reagent for the Thz ring-opening reaction
9
Figure 1). This reaction proceeds under weakly acidic conditions,
and the reaction mechanism is presumed as follows: the sulfur atom
of Thz is protonated, and the ring is opened; the generated sulfanyl
group is immediately pyridylsulfenylated with DPDS, and then the
FIGURE 1 Ring-opening reaction of thiazolidine (Thz) derivative
with 2,2 -dipyridyl disulfide (DPDS)
0
FIGURE 2 Primary structure of evasin-3. Solid lines and arrowheads indicate presumed disulfide bonds and the ligation site, respectively.
Numbers represent the positions of Cys residues
FIGURE 3 Synthetic procedure of
linear evasin-3 (6). Reaction conditions: A,
NCL buffer, room temperature, 2 h; B, (1)
0
ꢀ
2
,2 -dipyridyl disulfide (DPDS), 50 C, 2 h;
2) TCEP; C, NCL buffer, room
temperature for 7 h
(

KATAYAMA AND NAGATA
3 of 10
9
imine structure is hydrolyzed. It has been shown that this
deprotection method does not affect the Cys protecting groups such
as p-methoxybenzyl group and disulfide bond and is complete within
2
h. Thus, this method might be an alternative deprotection method
for Thz for protein chemical synthesis. On the other hand, it still
remains unclear whether this method can be applicable to the chemi-
cal synthesis of modified proteins such as glycoproteins.
Evasin-3 is a small glycoprotein found in tick saliva and consists
of 66 amino acid residues and two N-linked glycan moieties
10
(
Figure 2). Evasin-3 has six Cys residues at the positions of 22, 26,
3
2
3, 37, 39, and 50, forming three disulfide bonds connected between
11,12
2 and 37, 26 and 39, and 33 and 50.
Evasin-3 could bind
11
specifically to vertebrate CXCL proteins, suppressing inflammation.
Chemical syntheses of a truncated evasin-3 (17–56) and of a
12–14
fluorophore-labeled full-length evasin-3 have been reported,
although the evasin-3 having N-linked glycans has not yet been
synthesized. In this paper, we show that the DPDS-mediated Thz
deprotection method is applicable to glycoprotein synthesis through
the total chemical synthesis of glycosylated evasin-3.
2
2
|
RESULTS AND DISCUSSION
.1
|
Chemical synthesis of evasin-3 with the
refolding method
At first, to examine the applicability of the DPDS method for the
chemical synthesis of Cys-rich glycoprotein, we tried to synthesize
evasin-3 without Cys protecting group. The synthetic procedure is
shown in Figure 3. Evasin-3 sequence was divided into three
segments (1–3) corresponding to 37–66, 22–36, and 1–21 of
evasin-3, and these segments were prepared by the
FIGURE 4 Reversed-phase HPLC chromatograms in the synthesis
of linear evasin-3 (6). A, After DPDS treatment at 50 C for 2 h; B,
ꢀ
initial reaction mixture of second NCL; C, the reaction mixture after
second NCL (7 h). Elution conditions: column, Inertsil ODS-3
9
-fluorenylmethoxycarbonyl (Fmoc)-based SPPS method. To prepare
(
4.6 mmϕ × 150 mm) at a flow rate of 1 ml/min. *, DPDS; **, MPAA
the C-terminal thioester functionality, we used the N-alkyl-cysteine
15
22
(
NAC)-assisted method. At the Cys position, Thz was used as the
Cys surrogate. Recent progress of chemical glycoprotein synthesis
enables the chemoenzymatic transglycosylation reaction using
N-acetylglucosaminylated asparagine [Asn(GlcNAc)] residue and
sialylglycopeptide (SGP) as glycosyl acceptor and donor, respec-
N-terminal Thz was completely converted to Cys, intrachain disulfide
bond(s) were randomly connected in the evasin-3 (22–66) segment,
and the remaining Cys residues were pyridylsulfenylated. This mixture
was confirmed to be converted to the linear peptide 5 with TCEP
treatment, and no significant side product was observed on the
RP-HPLC chromatogram (Figure S2). Therefore, the mixture was
directly used to the next NCL reaction without purification.
16
tively. Therefore, GlcNAc-attached evasin-3 was synthesized as a
prototype of glycoprotein, and the Asn(GlcNAc) derivative with acid-
labile protecting groups developed in our previous study was used at
17
the putative N-glycosylated Asn positions.
Peptides 1 and 2 were condensed with the NCL method, giving
evasin-3 (22–66) fragment (4) in 83% yield (Figure S1). To open the
N-terminal Thz ring of peptide 4, DPDS was treated in a 50% acetoni-
After lyophilization, an equimolar amount of peptide 3 was added
to the mixture and dissolved in the NCL buffer. As expected,
intrachain disulfide bonds were immediately reduced, and a single
peak corresponding to the (22–66) segment (5) was observed on the
RP-HPLC chromatogram. The peptide condensation reaction with
NCL was almost complete within 7 h. During the reaction, no signifi-
cant side reaction was observed, and the desired full-length linear
evasin-3 (6) was obtained in 53% yield. Finally, the disulfide bonds
were formed in a redox buffer containing reduced and oxidized forms
trile aqueous solution containing 0.1% trifluoroacetic acid (TFA) at
ꢀ
5
0 C. After 2 h, the reaction mixture was analyzed with reversed-
phase (RP)-high-performance liquid chromatography (HPLC) and mass
spectrum measurement. On the HPLC chromatogram, many peaks
were observed (Figure 4). In the mass spectral analysis, all these peak
materials gave molecular ion peaks corresponding to Thz ring-opened
peptide having two or four pyridylsulfenyl groups, indicating that the
ꢀ
of glutathione. After the reaction at 4 C for 24 h, the folded evasin-3

4
of 10
KATAYAMA AND NAGATA
FIGURE 5 Synthetic procedure of evasin-3 (15) with regioselective disulfide formation reactions. Reaction conditions: A, NCL buffer, room
0
ꢀ
temperature, 2 h; B, (1) 2,2 -dipyridyl disulfide (DPDS), 50 C, 2 h; (2) TCEP; C, NCL buffer, room temperature for 20 h; D, 10% DMSO/50 mM
phosphate buffer (pH 7.0), room temperature, 24 h; E, I /HCl/CH OH/H O, room temperature, 1 h; F, 5% DMSO/TFA, room temperature, 30 min
2
3
2

KATAYAMA AND NAGATA
5 of 10
26
39
33
50
(
7) was obtained in 81% yield (Figure S3). Thus, the Thz ring-opening
at Cys and Cys , and Cys and Cys were protected with the
t
reaction with DPDS was shown to be applicable to the chemical syn-
thesis of Cys-rich glycoprotein through three-segments sequential
NCL method.
tert-butyl (Bu ) group.
Peptides 8 and 9 were condensed with the NCL method, giving
evasin-3 (22–66) fragment (10) in 64% yield (Figure S4). To open the
N-terminal Thz ring of peptide 10, DPDS was treated in 50% acetoni-
ꢀ
trile aqueous solution containing 0.1% TFA at 50 C. Thz was almost
2
.2
|
Chemical synthesis of evasin-3 with
completely converted to S-pyridylsulfenylated Cys residue within 2 h.
37,
regioselective disulfide formation reactions
During this reaction, the side chain of Cys which had no protecting
group, was also S-pyridylsulfenylated, giving peptide 11. Unfortu-
nately, peptide 11 appeared as a shoulder of the peak corresponding
to DPDS on the RP-HPLC chromatogram (Figure 5), which was con-
firmed by mass spectral analysis. At the same time, a small amount of
In order to investigate whether our DPDS method could be applicable
to the glycoprotein synthesis with the regioselective disulfide forma-
tion reactions, we tried to synthesize evasin-3 by the alternative syn-
thetic procedure using Cys protecting groups. The synthetic
procedure is shown in Figure 5. The C-terminal and middle segments
(
8 and 9) were synthesized by the Fmoc-SPPS. To achieve the
regioselective formation of disulfide bonds, the Acm group was used
FIGURE 7 Comparison of two synthetic evasin-3 (7 and 15). A,
Circular dichroism spectra of synthetic evasin-3 proteins. Dashed and
solid lines indicate the spectra of refolded evasin-3 7 and
regioselectively disulfide-formed evasin-3 15, respectively. B, C,
Reversed-phase HPLC elution profiles of the synthetic evasin-3.
Elution conditions: column, Inertsil ODS-3 (4.6 mmϕ × 150 mm) at a
flow rate of 1 ml/min
FIGURE 6 Reversed-phase HPLC chromatograms in the synthesis
of linear evasin-3 (12). A, After DPDS treatment at 50 C for 2 h; B,
ꢀ
initial reaction mixture of second NCL; C, the reaction mixture after
second NCL (20 h). Elution conditions: column, Inertsil ODS-3
(
4.6 mmϕ × 150 mm) at a flow rate of 1 ml/min. *, DPDS; **, MPAA;
**, evasin-3 (22–66) fragment without S-pyridylsulfenyl group
*

6
of 10
KATAYAMA AND NAGATA
0
22
37
peptide 11 having an intrachain disulfide bond between Cys -Cys
were found. Only the major peak with shorter retention time was col-
lected, and the evasin-3 having three disulfide bonds (15) was
obtained in 32% yield.
was observed on the RP-HPLC chromatogram. Different from the
Pd-mediated deprotection method, which was known to remove Acm
6
groups, no deprotection of the other Cys protecting groups were
In contrast to the refolded evasin-3 (7), which gave a single peak
on RP-HPLC, the purified peptide 15 gave two peaks with different
retention times from that of 7 (Figure 7), demonstrating that purified
peptide 15 consisted of two tautomeric conformers. The truncated
evasin-3 composed of 17–56 region has been synthesized with the
observed during the Thz ring-opening reaction with DPDS, indicating
t
that the DPDS method was compatible with Acm and Bu groups.
0
The mixture of peptides 11 and 11 was directly used to the sec-
ond NCL reaction. An equimolar amount of peptide 3 was added to
the mixture and dissolved in the NCL buffer. The S-pyridylsulfenyl
groups were immediately removed by TCEP, and a single peak
corresponding to the (22–66) segment was observed on the RP-HPLC
chromatogram (Figure 6). The second NCL reaction was almost com-
plete within 20 h, and the desired linear evasin-3 (12) was obtained in
12
oxidative folding method, and the disulfide bond arrangement in this
peptide was determined to be the same as that in the full-length
14
evasin-3. These findings supported that the refolded evasin-3 pro-
tein 7 obtained in this study had disulfide bonds at the desired posi-
tions and may have different conformation from that of peptide 15.
7
7% yield.
The three disulfide bonds were regioselectively formed. The first
22
37
disulfide bond between Cys and Cys was formed by treating
2.3
|
Conformational and functional analyses of
dimethyl sulfoxide (DMSO) in an aqueous buffer, giving peptide 13 in
evasin-3
26
39
7
5% yield (Figure S5). Next, Acm groups at Cys and Cys were oxi-
datively removed by iodine oxidation under acidic conditions, forming
To check the conformation of peptides 7 and 15, the circular dichro-
ism (CD) spectra of these peptides were measured. As shown in
Figure 7, these peptides showed the spectral patterns completely dif-
ferent from each other. Using the CD data, the secondary structure
the second disulfide bond, and the desired peptide 14 was obtained in
8
5
0% yield. The final disulfide bond was formed with the treatment of
18
% DMSO/TFA. Unexpectedly, on the RP-HPLC chromatogram of
molecular ion peak
corresponding to the desired product in the mass spectral analysis
19,20
crude material, two peaks showing
a
contents were estimated with BeStSel.
The refolded evasin-3 (7)
was estimated to be rich in β-structure (38.7%) but to have no α-helix.
FIGURE 8 Surface plasmon resonance biosensor analysis of evasin-3 proteins (7 and 15) upon binding to immobilized human CXCL
chemokines. A, 7 and CXCL-1; B, 7 and CXCL-8; C, 15 and CXCL-1; D, 15 and CXCL-8. Concentration series with twofold dilution, raging 10 to
0
.625 μM, were examined

KATAYAMA AND NAGATA
7 of 10
On the other hand, the estimated secondary structure contents of
synthetic evasin-3 with regioselective disulfide bond formation reac-
tions (15) were 8.4% α-helix and 26.0% β-structure. These results indi-
cated that the two synthetic evasin-3 (7 and 15) have tertiary
structures different from each other. The X-ray crystal structure of
evasin-3 revealed that the secondary structure composition of
an Autoflex spectrometer (Brucker, Germany) or a JMS-S3000 spec-
trometer (JEOL, Tokyo, Japan). The synthetic peptides were quanti-
fied with amino acid analysis, which was performed using a LaChrom
amino acid analyzer (Hitachi, Tokyo, Japan) after hydrolysis with a
ꢀ
6 M HCl solution at 150 C for 2 h in a vacuum-sealed tube. CD spec-
tra were measured with a Jasco J-820 spectropolarimeter (JASCO,
Tokyo, Japan) at room temperature with a 2-mm path length cell using
a phosphate buffer (50 mM, pH 7.0) as a solvent.
14
evasin-3 is 6.1% α-helix and 36.4% β-structure. These values do not
coincide well with the estimated ones of peptides 7 and 15, although
the β-structure content peptide 7 (38.7%) was close to that of the
native one.
Evasin-3 is known to have a binding ability specifically to CXCL
4.2
|
Evasin-3 (37–66) 1
10
chemokine. In order to evaluate the function of two synthetic
evasin-3 (7 and 15), the binding abilities to CXCL-1 and CXCL-8 were
examined with the surface plasmon resonance (SPR) biosensor analy-
Fmoc-Arg(Pbf)-Wang resin (0.28 mmol/g, 0.179 g, 0.05 mmol) was
swelled in 1-methyl-2-pyrrolidinone (NMP) for 30 min and was
treated with 20% piperidine/NMP for 5 and 15 min. After washing
with NMP, Fmoc-Arg(Pbf)-OBt, which was prepared by mixing
sis. As a result, the refolded peptide 7 could bind to CXCL-1 and
7
CXCL-8 with K
D
values of 7.76 × 10− M and 8.76 × 10−9 M, respec-
0
tively (Figure 8). These values were larger than those shown in the
Fmoc-Arg (Pbf)-OH (0.20 mmol), 1 M N,N -dicyclohexylcarbodiimide
1
0
previous report. The difference in these values probably resulted
from the difference in the method of SPR experiments. In contrast,
peptide 15 did not bind to CXCL chemokines. These results indicated
that the synthetic evasin-3 with refolding reaction (7) had the native
conformation and that the synthetic evasin-3 with regioselective
disulfide bond formation reactions (15) did not fold properly. Evasin-3
might not be able to be synthesized by the regioselective disulfide
bond formations, at least in this formation order, but could be pre-
pared with the refolding method in good yield.
(DCC)/NMP (0.3 ml), and 1 M 1-hydroxybenzotriazole (HOBt)/NMP
(0.3 ml) at room temperature for 30 min, was added and the reaction
ꢀ
mixture was mixed with a vortex at 50 C for 1 h. The resin was
washed with NMP and 50% dichloromethane/methanol, treated with
2
10% acetic anhydride (Ac O)/5% N,N-diisopropylethylamine
(DIEA)/NMP for 5 min, and washed with NMP. The peptide chain was
elongated in essentially the same manner as described above, and H-
Cys(Trt)-Phe-Cys(Trt)-Gly-Leu-Leu-Gly-Gln(Trt)-Asn(Trt)-Lys(Boc)-Lys
t
(Boc)-Gly-His(Trt)-Cys(Trt)-Tyr(Bu )-Lys(Boc)-Ile-Ile-Gly-Asn[GlcNAc(b
t
t
enzylidene, Boc)]-Leu-Ser(Bu )-Gly-Glu(OBu )-Pro-Pro-Val-Val-Arg(P
bf)-Arg(Pbf)-resin (0.449 g) was obtained. A part of the resin (21 mg)
3
|
CONCLUSION
2
was treated with a TFA cocktail (TFA/H O/phenol/thioanisole/
triisopropylsilane, 82.5/5/5/5/2.5, 0.3 ml) at room temperature for
2 h, and then the peptide was precipitated with diethyl ether. After
washing twice with ether, the precipitate was dried in vacuo. The
crude peptide was purified by RP-HPLC on an Inertsil ODS-3 column
(GL Science, Tokyo, Japan) with a linear gradient of acetonitrile con-
taining 0.1% TFA to give peptide 1 (804 nmol, 34% yield). MALDI-
To investigate the applicability of DPDS-mediated thiazolidine ring-
opening reaction to glycoprotein synthesis, we synthesized evasin-3.
DPDS treatment did not affect the N-linked glycan moiety, and
evasin-3 could be obtained by the sequential NCL reaction followed
by a refolding reaction in a redox buffer. The synthetic evasin-3
possessed a binding ability to CXCL chemokines. On the other hand,
we also tried to synthesize evasin-3 with regioselective disulfide bond
formation reactions. Cys protecting groups such as Acm and tert-butyl
groups were retained in thiazolidine ring-opening reaction with DPDS.
Unfortunately, evasin-3 might be unsuitable for the synthetic model
on the sequential disulfide bond formation reactions and the synthetic
evasin-3 did not have the proper conformation, but the applicability
of the DPDS method to sequential disulfide bond formation could
be shown.
+
TOF mass, found: m/z 3491.1, calcd: 3492.1 for (M + H) . Amino acid
analysis:
5
Asp2.00Ser0.81Glu2.04Pro1.88Gly Val1.80Ile1.58Leu2.99Tyr1.01
Phe1.04Lys3.00His1.00Arg2.04
.
2
2
4.3
|
Thz -Evasin-3 (22–36)-thioester 2
Starting from Fmoc-Rink Amide MBHA resin (0.39 mmol/g, 0.128 g,
0.05 mmol), Fmoc-Arg(Pbf)-OH was introduced to the resin by DCC-
HOBt method as described above. Another Fmoc-Arg(Pbf)-OH was
introduced to the resin, and Fmoc-Arg(Pbf)-Arg(Pbf)-NH-resin was
obtained. This rein was treated with 20% piperidine/NMP for 5 and
4
4
|
EXPERIMENTAL PROCEDURE
General
15 min. After washing with dichloromethane, Fmoc-Gly-(Et)Cys(Trt)-
.1
|
OBt, which was prepared by mixing Fmoc-Gly-(Et)Cys(Trt)-OH
0
(
0.10 mmol), N,N -diisopropylcarbodiimide (DIC, 23 μl, 0.15 mmol),
17
t
Fmoc-Asn[GlcNAc(benzylidene, Boc)]-OH,
Fmoc-Ser(Bu )-(Et)Cys
HOBt (20 mg, 0.15 mmol), and dichloromethane (0.4 ml) at room tem-
perature for 30 min, was added, and the reaction mixture was mixed
with a vortex at room temperature for 30 min. Another DIC (23 μl,
21
21
(
Trt)-OH,
and Fmoc-Gly-(Et)Cys(Trt)-OH
were prepared as
described previously. MALDI-TOF mass spectra were recorded using

8
of 10
KATAYAMA AND NAGATA
0
.15 mmol) was added, and the reaction mixture was mixed with a
Asp4.65Thr0.82Ser1.59Glu4.69Pro2.70Gly
7
Val1.88Ile1.47Leu3.00Tyr1.09Phe1.
vortex at room temperature overnight, giving Fmoc-Gly-(Et)Cys(Trt)-
Arg(Pbf)-Arg(Pbf)-NH-resin. The peptide chain was further elongated
with the DCC-HOBt method described above, and Boc-Thz-Asn(Trt)-
13Lys3.94His1.15Arg2.08
.
t
t
Lys(Boc)-Asn[GlcNAc(benzylidene, Boc)]-Cys(Trt)-Thr(Bu )-Ser(Bu )-
4.6
|
Linear evasin-3 6
t
t
Gly-Gln(Trt)-Asn(Trt)-Glu(OBu )-Cys(Trt)-Pro-Glu(OBu )-Gly-(Et)Cys
Trt)-Arg(Pbf)-Arg(Pbf)-NH-resin (0.319 g) was obtained. A part of the
(
Peptide 4 (210 nmol) was dissolved in 20 mM DPDS/0.1% TFA/50%
acetonitrile aq. (210 μl), and the solution was mixed with a vortex at
resin (20 mg) was treated with a TFA cocktail (0.3 ml) at room temper-
ature for 2 h, and then the peptide was precipitated with diethyl
ether. After washing twice with ether, the precipitate was dried in
vacuo. The crude peptide was dissolved in 50% acetonitrile/5% acetic
acid aqueous solution (950 μl), and 3-mercaptopropionic acid (MPA,
ꢀ
50 C for 2 h. Peptide 3 (210 nmol) was added to the solution, and the
mixture was lyophilized. The residue was dissolved in 6 M guanidine-
HCl/40 mM TCEP/60 mM MPAA/100 mM phosphate buffer (pH 7.4,
250 μl), and the solution was gently mixed at room temperature for
7 h. The crude peptide was separated by RP-HPLC on an Inertsil
ODS-3 column with a linear gradient of acetonitrile containing 0.1%
TFA to give peptide 6 (111 nmol, 53% yield). MALDI-TOF mass,
5
0 μl) was added. The mixture was gently mixed at room temperature
for 3 days and purified by RP-HPLC on an Inertsil ODS-3 column with
a linear gradient of acetonitrile containing 0.1% TFA to give peptide
+
2
(309 nmol, 9.7% yield). MALDI-TOF mass, found: m/z 1886.8, calcd:
found: m/z 7411.1, calcd: 7412.2 for (M + H) . Amino acid analysis:
+
1
886.7 for (M + H) . Amino acid analysis: Asp2.53Thr0.81Ser0.83Glu2.68
Asp7.82Thr2.48Ser4.30Glu5.43Pro3.73Gly9.74Ala0.40Val4.35Ile2.55Leu
1.11Phe2.18Lys4.30His1.11Arg3.16
4
Tyr
Pro0.93Gly
2
Lys0.91
.
.
4
.4
|
Evasin-3 (1–21)-thioester 3
4.7
|
Refolded evasin-3 7
Starting from Fmoc-Rink Amide MBHA resin (0.39 mmol/g, 0.128 g,
Peptide 6 (111 nmol) was dissolved in 8 M urea aq. (1.5 ml) and
t
0
.05 mmol), Fmoc-Ser(Bu )-(Et)Cys(Trt)-Arg(Pbf)-Arg(Pbf)-NH-resin
diluted with an ice-cooled buffer containing 0.6 M Tris/2.4 mM glu-
t
ꢀ
was obtained using Fmoc-Ser(Bu )-(Et)Cys(Trt)-OH by essentially the
same manner as described above. The peptide chain was further
elongated with the DCC-HOBt method described above, and H-Leu-
tathione (reduced form) (7.5 ml). After stirring the solution at 4 C
for 15 min, glutathione (oxidized form, 2.8 mg, the final concentra-
ꢀ
tion at 1 mM) was added. The solution was further stirred at 4 C
t
t
t
t
t
t
Val-Ser(Bu )-Thr(Bu )-Ile-Glu(OBu )-Ser(Bu )-Arg(Pbf)-Thr(Bu )-Ser(Bu )
overnight. The solution was then acidified by adding acetic acid
(270 μl) and separated by RP-HPLC on an Inertsil ODS-3 column
with a linear gradient of acetonitrile containing 0.1% TFA to give
t
t
t
-
Gly-Asp(OBu )-(Dmb)Gly-Ala-Asp(OBu )-Asn(Trt)-Phe-Asp(OBu )-Val-
t
Val-Ser(Bu )-(Et)Cys(Trt)-Arg(Pbf)-Arg(Pbf)-NH-resin (0.351 g) was
obtained. A part of the resin (104 mg) was treated with a TFA cocktail
peptide 7 (90.2 nmol, 81% yield). MALDI-TOF mass, found: m/z
+
(
1.5 ml) at room temperature for 2 h, and then the peptide was precip-
7406.9, calcd: 7406.2 for (M
+
H) . Amino acid analysis:
itated with diethyl ether. After washing twice with ether, the precipi-
tate was dried in vacuo. The crude peptide was dissolved in 6 M urea
Asp8.86Thr2.60Ser4.65Glu5.91Pro3.17Gly10.30Ala0.44Val4.56Ile2.61Leu
Tyr0.92Phe2.06Lys4.02His1.03Arg3.08
4
.
2
dissolved in 50% acetonitrile/5% acetic acid/H O (4.75 ml), and MPA
(
250 μl) was added. The mixture was gently mixed at room tempera-
3
9
50
t
ture for 7 days and purified by RP-HPLC on an Inertsil ODS-3 column
with a linear gradient of acetonitrile containing 0.1% TFA to give pep-
4.8
|
Cys (Acm), Cys (Bu )-Evasin-3 (37–66) 8
tide 3 (1.86 μmol, 13% yield). MALDI-TOF mass, found: m/z 2257.0,
Peptide 8 was prepared essentially according to the method for
peptide 1 and obtained in 23% yield. MALDI-TOF mass, found: m/z
+
calcd: 2257.0 for (M
+
H) . Amino acid analysis: Asp4.01
+
2
Thr1.76Ser3.26Glu1.04Gly Ala1.01Val2.48Ile1.06Leu0.94Phe1.05Arg1.08
.
3619.1, calcd: 3619.3 for (M + H) . Amino acid analysis: Asp1.99
Ser0.80Glu1.96Pro1.95Gly
His1.01Arg2.04
5
Val1.78Ile1.49Leu3.05Tyr1.04Phe1.02Lys2.97
.
22
4
.5
|
Thz -Evasin-3 (22–66) 4
2
2
26
33
t
Peptides 1 and 2 (308 nmol each) were dissolved in 6 M guanidine-
4.9
|
Thz , Cys (Acm), Cys (Bu )-Evasin-3
HCl/10 mM TCEP/60 mM MPAA/100 mM phosphate buffer (pH 7.4,
(22–36)-thioester 9
3
1
10 μl), and the solution was gently mixed at room temperature for
h. The crude peptide was separated by RP-HPLC on an Inertsil
Peptide 9 was prepared essentially according to the method for
ODS-3 column with a linear gradient of acetonitrile containing 0.1%
peptide 2, and obtained in 16% yield. MALDI-TOF mass, found: m/z
+
TFA to give peptide 4 (256 nmol, 83% yield). MALDI-TOF mass,
2013.8, calcd: 2013.8 for (M
+
H) . Amino acid analysis:
+
found: m/z 5272.9, calcd: 5272.9 for (M + H) . Amino acid analysis:
2
Asp2.63Thr0.82Ser0.81Glu2.56Pro1.08Gly Lys0.94.

KATAYAMA AND NAGATA
9 of 10
2
2
26,39
33,50
t
4
.10
|
Thz , Cys
(Acm), Cys
(Bu )-Evasin-3
4.14
|
Evasin-3 with three disulfide bonds 15
(
22–66) 10
Peptide 14 (52.7 nmol) was dissolved in 5% DMSO/TFA (500 μl), and
the solution was mixed with a vortex at room temperature for 30 min.
The crude peptide was precipitated with diethyl ether, washed twice
with ether, and dried under vacuum. The residue was separated by
RP-HPLC on an Inertsil ODS-3 column with a linear gradient of aceto-
nitrile containing 0.1% TFA to give peptide 15 (17.0 nmol, 32% yield).
Peptides 8 and 9 (388 nmol each) were ligated in the same manner as
for peptide 4, and peptide 10 was obtained in 64% yield (250 nmol).
+
MALDI-TOF mass, found: m/z 5527.1, calcd: 5527.3 for (M + H) .
Amino
acid
analysis:
Asp4.86Thr0.89Ser1.69Glu4.83Pro3.00Gly
7
Val1.87Ile1.46Leu2.99Tyr1.02Phe1.04Lys3.88His0.98Arg2.11
.
+
MALDI-TOF mass, found: m/z 7406.1, calcd: 7406.1 for (M + H) .
Amino acid analysis: Asp8.79Thr2.56Ser4.52Glu5.91Pro2.99Gly9.74
26,39
33,50
t
4
.11
|
Cys
(Acm), Cys
(Bu )-Evasin-3 12
4
Ala0.47Val4.41Ile2.87Leu Tyr0.99Phe2.20Lys4.01His0.99Arg3.01.
Peptide 10 (198 nmol) was treated with 20 mM DPDS/0.1%
ꢀ
TFA/50% acetonitrile aq. (200 μl) at 50 C for 2 h. Peptide 3 (198 nmol)
4.15
|
Surface plasmon resonance
was added to the solution, and the mixture was lyophilized. The resi-
due was dissolved in 6 M guanidine-HCl/40 mM TCEP/60 mM
MPAA/100 mM phosphate buffer (pH 7.4, 200 μl), and the solution
was gently mixed at room temperature for 20 h. The crude peptide
was separated by RP-HPLC on an Inertsil ODS-3 column with a linear
gradient of acetonitrile containing 0.1% TFA to give peptide 12
Affinity assays of the synthetic evasin-3 were performed with BIAcore
T200 SPR system (GE Healthcare). All assays were carried out in PBS
(pH 7.4) as a running buffer. Human CXCL-1 and CXCL-8 purchased
from PEPROTECH were immobilized on Sensor Chip CM5
(GE Healthcare) according to the manufacturer's protocol. Concentra-
tion series of evasin-3 with twofold dilution, raging 10 to 0.625 μM,
were examined. Association and dissociation kinetics were monitored
at a flow rate of 30 μl/min for 125 s each. The data were analyzed
using BIAcore T200 software.
(
153 nmol, 77% yield). MALDI-TOF mass, found: m/z 7666.2, calcd:
+
7
666.5 for (M + H) . Amino acid analysis: Asp8.35Thr2.50Ser4.19
Glu5.54Pro3.86Gly9.81Ala0.48Val4.23Ile2.29Leu
His1.09Arg3.18
4
Tyr1.02Phe2.07Lys4.30
.
ACKNOWLEDGEMENTS
2
6,39
33,50
t
4
.12
|
Cys
(Acm), Cys
(Bu )-Evasin-3 with
This work was supported in part by KAKENHI (grant number
18 K05113).
one disulfide bond 13
Peptide 12 (153 nmol) was dissolved in 50 mM phosphate buffer
900 μl, pH 7.0), and DMSO (100 μl) was added. The mixture was
CONFLICT OF INTEREST
(
The authors declare no conflicts of interest with the contents of this
article.
gently mixed at room temperature for 24 h and separated by
RP-HPLC on an Inertsil ODS-3 column with a linear gradient of
acetonitrile containing 0.1% TFA to give peptide 13 (116 nmol,
AUTHOR CONTRIBUTIONS
7
5% yield). MALDI-TOF mass, found: m/z 7665.0, calcd: 7664.5
H. K.: Conceptualization, investigation, writing-original draft, editing,
funding acquisition; K. N.: investigation, editing.
+
for (M + H) . Amino acid analysis: Asp8.97Thr2.66Ser4.80Glu5.97Pro
3 4
.05Gly9.80Ala0.44Val4.36Ile2.16Leu Tyr1.00Phe2.11Lys4.23His1.11Arg3.11
.
ORCID
Hidekazu Katayama
https://orcid.org/0000-0003-3367-9147
33,50
t
4
.13
|
Cys
(Bu )-Evasin-3 14
REFERENCES
1
2
3
.
.
.
Merrifield RB. Solid phase peptide synthesis. I. The synthesis of a
Peptide 13 (115 nmol) was dissolved in H
2
O (1.2 ml), and added
tetrapeptide. J Am Chem Soc. 1963;85:2149-2154.
dropwisely to methanol (4.8 ml) containing 20 mM iodine/methanol
72 μl) and conc. HCl aq. (12 μl). After mixing with a vortex at room
Dawson PE, Muir TW, Clark-Lewis I, Kent SB. Synthesis of proteins
by native chemical ligation. Science. 1994;266(5186):776-779.
Conibear AC, Watson EE, Payne RJ, Becker CFW. Native chemical
ligation in protein synthesis and semi-synthesis. Chem Soc Rev. 2018;
(
temperature for 1 h, the reaction was quenched by adding ascorbic
acid aqueous solution. Peptide 14 was obtained by RP-HPLC purifica-
tion using an Inertsil ODS-3 column with a linear gradient of acetoni-
trile containing 0.1% TFA in 80% yield (91.5 nmol). MALDI-TOF mass,
4
7(24):9046-9068.
4
.
Bang D, Kent SBH. A one-pot total synthesis of crambin. Angew Chem
Int Ed Engl. 2004;43(19):2534-2538.
+
5. Jbara M, Maity SK, Seenaiah M, Brik A. Palladium mediated rapid
deprotection of N-terminal cysteine under native chemical ligation
conditions for the efficient preparation of synthetically challenging
proteins. J Am Chem Soc. 2016;138(15):5069-5075.
found: m/z 7520.3, calcd: 7520.4 for (M + H) . Amino acid analysis:
Asp8.70Thr2.62Ser4.78Glu5.78Pro3.54Gly9.65Ala0.47Val4.23Ile2.55Leu
Tyr1.10Phe2.18Lys4.07His1.01Arg3.01
4
.

1
0 of 10
KATAYAMA AND NAGATA
6
.
Maity SK, Jbara M, Laps S, Brik A. Efficient palladium-assisted one-
pot deprotection of (acetamidomethyl)cysteine following native
chemical ligation and/or desulfurization to expedite chemical protein
synthesis. Angew Chem Int Ed. 2016;55(28):8108-8112.
16. Fairbanks AJ. The ENGases: versatile biocatalysts for the production
of homogeneous N-linked glycopeptides and glycoproteins. Chem Soc
Rev. 2017;46(16):5128-5146.
17. Katayama H. Facile preparation of the N-acetylglucosaminylated
asparagine derivative with TFA-sensitive protecting groups useful for
solid-phase glycopeptide synthesis. J Pept Sci. 2015;21(9):696-699.
18. Otaka A, Koide T, Shide A, Fujii N. Application of dimethylsulphoxide
(DMSO) /trifluoroacetic acid (TFA) oxidation to the synthesis of
cystine-containing peptide. Tetrahedron Lett. 1991;32:1223-1226.
19. Micsonai A, Wien F, Kernya L, et al. Accurate secondary structure
prediction and fold recognition for circular dichroism spectroscopy.
Proc Natl Acad Sci U S A. 2015;112(24):E3095-E3103.
7
.
.
Kondasinghe TD, Saraha HY, Odeeshoa SB, Stockdill JL. Direct
palladium-mediated on-resin disulfide formation from Allocam protec-
ted peptides. Org Biomol Chem. 2017;14:2914-2918.
8
Naruse N, Kobayashi D, Ohkawachi K, Shigenaga A, Otaka A.
Copper-mediated deprotection of thiazolidine and selenazolidine
derivatives applied to native chemical ligation. J Org Chem. 2020;85
(
3):1425-1433.
9
.
Katayama H, Morisue S. A novel ring opening reaction of peptide
0
N-terminal thiazolidine with 2,2 -dipyridyl disulfide (DPDS) efficient
20. Micsonai A, Wien F, Bulyáki É, et al. BeStSel: a web server for
accurate protein secondary structure prediction and fold recognition
from the circular dichroism spectra. Nucleic Acids Res. 2018;46(W1):
W315-W322.
for protein chemical synthesis. Tetrahedron. 2017;73:3541-3547.
1
1
0. Déruaz M, Frauenschuh A, Alessandri AL, et al. Ticks produce highly
selective chemokine binding proteins with antiinflammatory activity.
J Exp Med. 2008;205(9):2019-2031.
21. Hojo H, Kobayashi H, Ubagai R, et al. Efficient preparation of
Fmoc-aminoacyl-N-ethylcysteine unit, a key device for the synthesis
of peptide thioesters. Org Biomol Chem. 2011;9(19):6807-6813.
1. Galan K, Dallegri F, Carbone F, Proudfoot AE, Vuilleumier N,
Frossard JL. Treatment with evasin-3 abrogates neutrophil-mediated
inflammation in mouse acute pancreatitis. Eur J Clin Invest. 2014;44:
940-950.
1
2. Denisov SS, Ippel JH, Mans BJ, Dijkgraaf I, Hackeng TM. SecScan: a
general approach for mapping disulfide bonds in synthetic and
recombinant peptides and proteins. Chem Commun. 2019;55(10):
SUPPORTING INFORMATION
Additional supporting information may be found online in the
Supporting Information section at the end of this article.
1374-1377.
1
1
3. Lee AW, Deruaz M, Lynch C, et al. A knottin scaffold directs the
CXC-chemokine–binding specificity of tick evasins. J Biol Chem.
How to cite this article: Katayama H, Nagata K. Application of
2019;294(29):11199-11212.
0
4. Denisov SS, Heinzmann ACA, Vajen T, et al. Tick saliva protein
evasin-3 allows for visualization of inflammation in arteries through
interactions with CXC-type chemokines deposited on activated
endothelium. Bioconjug Chem. 2020;31(3):948-955.
2,2 -dipyridyl disulfide-mediated thiazolidine ring-opening
reaction to glycoprotein synthesis: Total chemical synthesis of
evasin-3. J Pep Sci. 2020;e3290. https://doi.org/10.1002/
psc.3290
1
5. Hojo H, Onuma Y, Akimoto Y, Nakahara Y, Nakahara Y. N-Alkyl
cysteine-assisted thioesterification of peptides. Tetrahedron Lett.
2007;48:25-28.