Báo cáo khoa học: Acceleration of disulfide-coupled protein folding using glutathione derivatives potx

Acceleration of disulfide-coupled protein folding using
glutathione derivatives
Masaki Okumura
1,2
, Masatoshi Saiki
2,3
, Hiroshi Yamaguchi
1
and Yuji Hidaka
2
1 School of Science and Technology, Kwansei Gakuin University, Hyogo, Japan
2 Graduate School of Science and Engineering, Kinki University, Osaka, Japan
3 Department of Applied Chemistry, Faculty of Engineering, Tokyo University of Science, Yamaguchi, Japan
Introduction
The formation of the correct disulfide bonds and the
conversion of a protein into its native conformation
are the result of reversible thiol(SH) ⁄ disulfide(SS)
exchange reactions that occur during protein folding
and are thermodynamically and kinetically related to
the redox potential in the biological environment. Glu-
tathione (c-Glu-Cys-Gly), one of the most abundant
thiol compounds found in cells, plays a major role in
the formation of disulfide bonds in proteins in the
endoplasmic reticulum [1]. Oxidized glutathione
(GSSG) functions as an oxidant in the formation of
disulfide bonds in proteins and reduced glutathione
(GSH) functions as a reducing agent that cleaves mis-
bridged disulfide bonds in proteins, resulting in the
formation of the thermodynamically stable conforma-
tion of proteins in vivo [2]. Because of this, glutathione
Keywords

arginine; disulfide; folding; glutathione;
uroguanylin
Correspondence
Y. Hidaka, Graduate School of Science and
Engineering, Kinki University, 3-4-1
Kowakae, Higashi-Osaka, Osaka 577-8502,
Japan
Fax: +81 6 6723 2721
Tel: +81 6 6721 2332
E-mail:
(Received 20 October 2010, revised 18
January 2011, accepted 28 January 2011)
doi:10.1111/j.1742-4658.2011.08039.x
Protein folding occurs simultaneously with disulfide bond formation. In
general, the in vitro folding proteins containing disulfide bond(s) is carried
out in the presence of redox reagents, such as glutathione, to permit native
disulfide pairing to occur. It is well known that the formation of a disulfide
bond and the correct tertiary structure of a target protein are strongly
affected by the redox reagent used. However, little is known concerning the
role of each amino acid residue of the redox reagent, such as glutathione.
Therefore, we prepared glutathione derivatives – glutamyl-cysteinyl-argi-
nine (ECR) and arginyl-cysteinyl-glycine (RCG) – and examined their abil-
ity to facilitate protein folding using lysozyme and prouroguanylin as
model proteins. When the reduced and oxidized forms of RCG were used,
folding recovery was greater than that for a typical glutathione redox sys-
tem. This was particularly true when high protein concentrations were
employed, whereas folding recovery using ECR was similar to that of the
glutathione redox system. Kinetic analyses of the oxidative folding of prou-
roguanylin revealed that the folding velocity (K
RCG

= 3.69 · 10
)3
s
)1
)
using reduced RCG ⁄ oxidized RCG was approximately threefold higher
than that using reduced glutathione ⁄ oxidized glutathione. In addition, fold-
ing experiments using only the oxidized form of RCG or glutathione indi-
cated that prouroguanylin was converted to the native conformation more
efficiently in the case of RCG, compared with glutathione. The findings
indicate that a positively charged redox molecule is preferred to accelerate
disulfide-exchange reactions and that the RCG system is effective in medi-
ating the formation of native disulfide bonds in proteins.
Abbreviation
Arg-C, arginylendopeptidase C; ECR, glutamyl-cysteinyl-arginine; ECR
ox
, oxidized ECR; ECR
red
, reduced ECR; GSH, reduced glutathione;
GSSG, oxidized glutathione; RCG, arginyl-cysteinyl-glycine; RCG
ox
, oxidized RCG; RCG
red
, reduced RCG.
FEBS Journal 278 (2011) 1137–1144 ª 2011 The Authors Journal compilation ª 2011 FEBS 1137
is widely employed in studies of folding reactions of
disulfide-containing proteins in vitro [3,4].
Generally, folding reactions of proteins that do not
involve disulfide bond formation occur within a few
minutes. However, disulfide-containing proteins require

several hours to fold correctly, because disulfide-
exchange reactions are usually the rate-determining
step. During folding reactions, several types of mis-
bridged intermediates are often observed, and proteins
with the native conformation accumulate in a time-
dependent manner, as a result of disulfide-exchange
reactions [5]. Proteins in which cysteine residues are
involved in folding are usually folded into the native
conformation via the formation of complex disulfide
intermediates in the presence of redox reagents. Cor-
rect folding is attributed to the presence of disulfide
intermediates, which are readily converted into the
native conformation. Therefore, to achieve the native
tertiary structure of target proteins, regulation of the
formation of the proper folding intermediates is a criti-
cal issue.
A variety of redox reagents (GSH ⁄ GSSG, reduced
dithiothreitol ⁄ oxidized dithiothreitol and cysteine ⁄
cystine) can be used in protein-folding experiments
[6,7]. Recently, Beld et al. [8] reported that selenogluta-
thione, in which the cysteine residue of glutathione is
replaced with a selenocysteine, accelerated the folding
of disulfide-containing proteins by regulating the disul-
fide coupling reaction. Thus, glutathione, or derivatives
thereof, have the potential to serve as disulfide-
exchange reagents in protein-folding studies. However,
the role of each of the amino acid residues of glutathi-
one, except for the cysteine residue, in the disulfide-
coupled folding reactions is not known in detail. In
this study, to investigate the role of each amino acid

residue of glutathione, a series of glutathione analogs
– arginyl-cysteinyl-glycine (RCG) and glutamyl-cyste-
inyl-arginine (ECR) – were prepared, and their ability
to serve as a redox reagent in disulfide-exchange reac-
tions was examined. Arginine has recently been
employed in studies of protein folding [9]. It is thought
that arginine prevents the formation of nonspecific
aggregates during the folding reaction. Folding experi-
ments using lysozyme in the presence of arginine indi-
cated that arginine promoted the formation of its
native conformation, compared with other amino acids
[10], and that arginine effectively suppressed the aggre-
gation of denatured lysozyme [11], resulting in an
increase in folding yield [12]. Arginine has also been
reported to stabilize the exposed hydrophobic area of
single-chain Fv fragments during the folding reaction
[13], and the addition of both GSSG and arginine
resulted in an increase in folding recovery [14]. To
examine this aspect further, an arginine residue was
introduced into a glutathione molecule to increase the
solubility of folding intermediates in which cross-disul-
fide bond(s) are formed between the reagents and pro-
teins. In this study, except for the cysteine residue,
each amino acid residue of glutathione was systemati-
cally replaced with an arginine residue.
RCG and ECR were chemically synthesized and
their participation in disulfide-coupled folding reac-
tions of lysozyme and prouroguanylin, as model pro-
teins containing disulfide bonds, were examined.
Substituting a glutamic acid residue for an arginine

residue in glutathione dramatically affected the ability
of glutathione to function as a redox reagent for pro-
tein folding, resulting in an improved folding efficiency
of the correct tertiary structures of those proteins.
Here, we present results which demonstrate that a
positive charge on the redox molecule is preferred for
accelerating disulfide-exchange reactions in protein
folding.
Results and Discussion
Glutathione is a major reagent for regulating disulfide
bond formation of naked proteins in the endoplasmic
reticulum. It is also widely used to assist protein fold-
ing in vitro [15,16]. GSH cleaves unfavorable disulfide
bonds (solvent-exposed disulfide bonds) and GSSG
promotes the formation of disulfide bonds in proteins,
resulting in the formation of the native conforma-
tion of a protein. Glutathione (c-GSH) contains a
c-branched peptide bond between the glutamic acid
and the cysteine residues. We first investigated whether
the branched structure is required for disulfide-
exchange reactions. For this purpose, a-glutamyl-cyste-
inyl-glycine (a-GSH) was chemically synthesized and
the folding of lysozyme was examined in the presence
of a-GSH or c-GSH to estimate their reactivity. The
folding recovery for lysozyme using a-GSH was similar
to that of glutathione (Fig. S1). Therefore, RCG and
ECR were designed as a-linked peptides and then syn-
thesized.
The folding of lysozyme has been studied extensively
and it is known that the folding recovery of lyso-

zyme decreases at high protein concentrations (> 0.1
mgÆmL
)1
) [17]. To estimate the ability of RCG [both
reduced (RCG
red
) and oxidized (RCG
ox
) forms] and
ECR [both reduced (ECR
red
) and oxidized (ECR
ox
)
forms] to permit the disulfide-coupled folding of
proteins, lysozyme was folded at high concentrations
(0.1–1.6 mgÆmL
)1
). The molar ratios of the reduced
and oxidized form of the reagents were adjusted to
2 : 1 in these experiments, because the molar ratio of
Acceleration of disulfide-coupled protein folding M. Okumura et al.
1138 FEBS Journal 278 (2011) 1137–1144 ª 2011 The Authors Journal compilation ª 2011 FEBS
GSH to GSSG in the endoplasmic reticulum ranges
from 1 : 1 to 3 : 1 [2]. The fully reduced and the native
form of lysozyme were eluted at positions correspond-
ing to R and N, respectively, in Fig. 1. The HPLC
peak for lysozyme folded in the correct conformation
was assigned based on a co-elution experiment using
native lysozyme [17,18]. The folding yields of lysozyme

(cross-hatched bar in Fig. 2) in the presence of ECR
red
and ECR
ox
were similar to that for GSH ⁄ GSSG (open
bar in Fig. 2) at the protein concentrations used in this
study, indicating that the replacement of the glycine
residue with an arginine residue had no significant
effect on the reactivity of the redox reagent. On the
other hand, the folding yields using RCG
red
and
RCG
ox
(black bar in Fig. 2) were increased to 130%,
131%, 171% and 185%, compared with those using
GSH ⁄ GSSG (open bar in Fig. 2) at concentrations of
0.2, 0.4, 0.8 and 1.6 mgÆmL
)1
, respectively. The folding
yield for the correct conformation of lysozyme was
dramatically increased when RCG
red
⁄ RCG
ox
was used
at higher protein concentrations. The low folding
recovery of lysozyme at high protein concentrations is
thought to be caused by the formation of nonspecific
aggregates. It therefore appears that the arginine resi-

due introduced into the glutathione molecule prevented
nonspecific aggregation of the folding intermediates at
higher protein concentrations. However, the mecha-
nism of the RCG-mediated folding remains to be
examined.
In order to further examine the effect of RCG
red
⁄
RCG
ox
in protein folding, recombinant human prou-
roguanylin was also examined as a model protein.
Prouroguanylin contains six cysteine residues that par-
ticipate in three intramolecular disulfide bonds in the
native state (Cys41–Cys45, Cys74–Cys82 and Cys77–
Cys85) and formation of the correct disulfide pairs is
important for the complete folding recovery of the
protein [19]. The folding reactions of prouroguanylin
were examined using RCG
red
⁄ RCG
ox
or GSH ⁄ GSSG,
and the folding yields were estimated by HPLC analy-
ses. The folding recovery of prouroguanylin using the
typical GSH system was decreased at higher protein
concentrations, as reported previously [19]. The RCG
redox system resulted in improved folding yields at
higher protein concentrations, compared with the GSH
system, as shown in Fig. 3. When the RCG

red
⁄ RCG
ox
redox system was used, the folding yields of the native
conformation of prouroguanylin were increased to
114% and 143% at concentrations of 0.2 and
0.4 mgÆmL
)1
, respectively, compared with that of
GSH ⁄ GSSG. These results indicate that the
RCG
red
⁄ RCG
ox
system is an excellent redox system
Retention time (min)
01020 30
N
R
Relative absorbance at 220 nm
Fig. 1. HPLC profile of the folding of lysozyme. N and R represent
the native and reduced forms, respectively. The proteins were
eluted using a 20–60% linear gradient of CH
3
CN in 0.05% trifluoro-
acetic acid, at a rate of 1% ⁄ min and a flow rate of 1.0 mLÆmin
)1
,
and monitored at 220 nm.
Recovery of

native conformation (%)
Protein concentration (mg·mL
–1
)
100
80
60
40
20
0
0.1
0.2 0.4 0.8
1.6
Fig. 2. Folding yields of lysozyme in the presence of GSH ⁄ GSSG
(open bars), reduced (ECR
red
) and oxidized (ECR
ox
) forms of ECR
(cross-hatched bars), and reduced (RCG
red
) and oxidized (RCG
ox
)
forms of RCG (shaded bars).
0.1 0.2 0.4
Recovery of
native conformation (%)
100
80

60
40
20
0
Protein concentration (m
g
·mL
–1
)
Fig. 3. Folding yields of prouroguaylin in the presence of
GSH ⁄ GSSG (open bars) and the RCG
red
⁄ RCG
ox
(shaded bars).
M. Okumura et al. Acceleration of disulfide-coupled protein folding
FEBS Journal 278 (2011) 1137–1144 ª 2011 The Authors Journal compilation ª 2011 FEBS 1139
for the folding of prouroguanylin, as well as for the
folding of lysozyme. Therefore, it can be concluded
that RCG
red
⁄ RCG
ox
serves as an effective redox sys-
tem for disulfide-coupled protein folding. Arginine is
thought to increase the folding recovery of proteins by
suppressing the formation of aggregates. This effect
was observed when Arg was used at a concentration of
several hundred millimoles. However, our experiments
were carried out at a concentration of only a few milli-

molar concentrations of Arg. Therefore, it is likely that
the efficient folding recovery using RCG is not the
result of direct effects, such as stabilization of the
hydrophobic surface of the denatured proteins.
To address the question of how the RCG redox sys-
tem increases the folding recovery of proteins and the
nature of the mechanism responsible for this, it is nec-
essary to characterize the folding intermediates. We
hypothesized that RCG plays a role in the solubility of
the denatured protein when the RCG is linked to fold-
ing intermediates via disulfide bonds, and that the Arg
residue in the cross-disulfide-linked RCG is important
for the accumulation of proper folding intermediates
in the native conformation. Generally, it is difficult to
analyze folding intermediates of disulfide-containing
proteins at each step of the conformational transition
[20]. However, the folding intermediates of prourogu-
anylin can be separated by HPLC. We therefore
employed prouroguanylin as a model protein to ana-
lyze the folding mechanism using RCG
red
⁄ RCG
ox
. The
distribution of folding intermediates of prouroguanylin
produced using RCG
red
⁄ RCG
ox
was compared with

that of GSH ⁄ GSSG at several time-points of the fold-
ing reactions, and the folding intermediates were sepa-
rated by HPLC and analyzed using MALDI-
TOF ⁄ MS.
The folding products 0SH, 2SH, R and N, shown in
Fig. 4, represent proteins with three disulfide bonds,
including mis-bridged disulfide bonds, two disulfide
bonds, no disulfide bonds and native disulfide bonds,
respectively, as confirmed by MALDI-TOF ⁄ MS analy-
ses. The folding reactions and the correct disulfide for-
mation of prouroguanylin were complete within 24 h
for both GSH ⁄ GSSG and RCG
red
⁄ RCG
ox
. The
reduced form (R in Fig. 4A,B) of prouroguanylin was
present for a longer time in the reaction mixture that
contained GSH ⁄ GSSG compared with the reaction
mixture that contained RCG
red
⁄ RCG
ox
. The formation
and the reduction of mixed-disulfide bonds between
GSH and proteins are reversible reactions. The forma-
tion of the reduced form of proteins is predominant at
the early stage of the folding reaction because an
excess of reductant was present in the reaction mix-
ture. The formation of the correct disulfide bond(s)

then becomes predominant at the later stage in folding
because the native disulfide bond is shielded by virtue
of the fact that it is located in the interior of a protein
molecule. In our experiment, the reduced form of
prouroguanylin disappeared within 5 min (Fig. 4B) in
the reaction mixture using RCG
red
⁄ RCG
ox
but detect-
able amounts were still present in the reaction mixture
using GSH ⁄ GSSG (Fig. 4A) at that time-point. There-
fore, these results indicate that the formation of intra-
molecular disulfide bonds is rapid in the presence of
RCG
red
⁄ RCG
ox
and that the half life of folding inter-
mediates with a low solubility becomes small, resulting
in an improved folding recovery.
In order to better understand the folding mechanism
in the presence of RCG
red
⁄ RCG
ox
, the folding inter-
mediates were further analyzed in detail. The disulfide
pairing of prouroguanylin was determined using a pre-
viously reported method [21]. The intermediates 0SH

and 2SH were observed at an early stage of the folding
reaction (1 and 5 min in Fig. 4A and 1 min in
Fig. 4B). To determine the positions of the disulfide
bonds of the folding intermediates, the proteins were
separated by HPLC and digested with the endoprotein-
ase arginylendopeptidase C (Arg-C) [21]. The peptide
fragment (Thr-Ile-Ala-uroguanylin) produced, includ-
ing the mature region, was compared to chemically
synthesized peptides with three different disulfide
1 min
5 min
2 h
24 h
Refolding time
Retention time
Relative absorbance at 220 nm
Relative absorbance at 220 nm
R
R
R
N
2SH
2SH
0SH
2SH
0SH
20 min
AB
1 min
5 min

2 h
24 h
Retention time
R
N
2SH
0SH
20 min
N
N
N
Refolding time
Fig. 4. HPLC profiles of the reaction mix-
tures of prouroguanylin in the presence of
2m
M GSH ⁄ 1mM GSSG (A) and 2 mM
RCG
red
⁄ 1mM RCG
ox
(B). N, 2SH, 0SH and
R, represent the positions of prouroguanylin
with native disulfide pairing, with two thiols
and two disulfide bonds, with three disulfide
bonds and no thiol groups, and the fully
reduced form of prouroguanylin, respec-
tively.
Acceleration of disulfide-coupled protein folding M. Okumura et al.
1140 FEBS Journal 278 (2011) 1137–1144 ª 2011 The Authors Journal compilation ª 2011 FEBS
pairings using HPLC (data not shown), as described

previously [21]. The results indicated that 0SH at 1
and 5 min in Fig. 4A and 0SH at 1 min in Fig. 4B cor-
responded to prouroguanylin in which the disulfide
bonds in the mature region were mis-bridged. The for-
mation of native disulfide bonds of prouroguanylin
was observed at 5 min in the presence of RCG
red
⁄
RCG
ox
(Fig. 4B), but  20 min was required to
accomplish this in the presence of GSH ⁄ GSSG
(Fig. 4A). The mis-bridged disulfide species, 0SH and
2SH, were observed at an early stage in the folding
reaction for both systems. These results suggest that
the formation of the native disulfide bonds in prou-
roguanylin is achieved via intermediates with mis-bri-
ged disulfide bonds and that RCG
red
⁄ RCG
ox
accelerates the formation of the correct disulfide pair-
ings via disulfide-exchange reactions.
Generally, the disulfide-exchange reaction is the
rate-determining step in the disulfide-coupled folding
of proteins and is catalyzed by a free thiol group. The
kinetically trapped intermediates were observed at the
early stage of the folding reaction because the local
stability of each moiety of a protein molecule was pre-
dominantly affected at the early stage in the folding

reaction in vitro. The mis-bridged disulfide species then
disappeared, followed by the formation of native disul-
fide bonds. Therefore, the velocity of the disulfide-
exchange reaction is important in achieving correct
folding and folding recovery. The folded prouroguany-
lin with the correct disulfide bonds was observed
within 5 min in the folding reaction carried out in the
presence of RCG
red
⁄ RCG
ox
. This result indicates that
RCG
red
⁄ RCG
ox
accelerates the exchange reaction of
disulfide bonds and effectively permits prouroguanylin
to be converted into the native conformation.
To further estimate the ability of the RCG redox sys-
tem in the disulfide-coupled folding of proteins, the
folding reaction was performed under anaerobic condi-
tions using only RCG
ox
or GSSG. Under anaerobic
conditions, RCG
ox
first oxidatively reacts with the thiol
groups of proteins to form cross-disulfide bonds
between RCG and proteins. The intramolecular disul-

fide-exchange reaction mainly occurs under these condi-
tions, because RCG
red
, as an initiator of the disulfide-
exchange reaction, is absent, although trace amounts of
RCG
red
exist as a side reaction product of the forma-
tion of the cross-disulfide bond. Therefore, the intra-
molecular disulfide-exchange reaction in the disulfide-
coupled folding of proteins can be estimated by using
only its oxidative form. The oxidative folding of prou-
roguanylin was carried out in the presence of 1 mm
GSSG or 1 mm RCG
ox
at room temperature for 48 h,
and the folding products were analyzed by HPLC
(Fig. 5A,B). To estimate the ratio of disulfide isomers
of prouroguanylin, the mixtures of folding products
(peaks a and b in Fig. 5A and B, respectively) were
purified by HPLC and digested with the endoproteinase
Arg-C to release the mature region (Thr-Ile-Ala-urogu-
anylin), as described above. The digests were separated
by HPLC (Fig. 5C,D) and analyzed by MALDI-
TOF ⁄ MS. The mis-bridged disulfide peptide (I; Cys74–
Cys85 and Cys77–Cys82) and the native disulfide pep-
tide (N; Cys74–Cys82 and Cys77–Cys85) were observed
in both cases, as shown in Fig. 5C,D. The ratios of I
and N were  1 : 1 and 1 : 2 in the presence of GSSG
and RCG

ox
, respectively. This indicates that RCG
ox
promotes the formation of the native conformation of
prouroguanylin faster than that of GSSG and that this
occurs via intramolecular disulfide-exchange reactions.
Therefore, we conclude that the native conformation of
proteins can be achieved effectively using RCG
red
⁄ RC-
G
ox
, which accelerates intramolecular disulfide-
exchange reactions.
To determine the kinetic parameters for the folding
reaction using the RCG
red
⁄ RCG
ox
system, CD mea-
surements were carried out at 222 nm. The CD analysis
Relative absorbance at 220 nm
Retention time (min)
N
N
I
I
a
b
0 10 20 30 0 10 20 30 0 10 20 30 0 10 20 30

ABCD
Fig. 5. HPLC profiles of the folding of prou-
roguanylin in the presence of GSSG (A) and
RCG
ox
(B) and Arg-C digests of peaks a (C)
and b (D) in (A) and (B), respectively. N and
I represent Thr-Ile-Ala-uroguanylin with
native disulfide bonds and with mis-bridged
disulfide bonds, respectively.
M. Okumura et al. Acceleration of disulfide-coupled protein folding
FEBS Journal 278 (2011) 1137–1144 ª 2011 The Authors Journal compilation ª 2011 FEBS 1141
of the prouroguanylin structure revealed that prou-
roguanylin contains an a-helical structure (data not
shown) and that the helical structure was formed within
10 min when using RCG
red
⁄ RCG
ox
and within 30 min
when using GSH ⁄ GSSG (Fig. 6). Importantly, these
timescales corresponded to the formation of the native
disulfide bonds (Fig. 4A,B), indicating that the forma-
tion of the native disulfide bonds are closely related to
the formation of an a-helical structure. Therefore, the
folding kinetics can be estimated, based on the forma-
tion of the a-helical structure. The molar ellipticity at
222 nm, obtained for each time-point, was used to cal-
culate the rate constants, as previously described [22].
A curve-fitting method was applied to estimate the

kinetic parameters, and the results indicated that the
reaction followed single exponential kinetics. Therefore,
the rate constants for the folding of prouroguanylin
were calculated to be K
gsh
= 1.20 · 10
)3
s
)1
and
K
rcg
= 3.69 · 10
)3
s
)1
(calculation using Kaleida
Graph version 3.5) (i.e. the reaction rate in the case of
RCG
red
⁄ RCG
ox
was approximately three times faster
than that of GSH ⁄ GSSG).
Arginine has been employed in folding reactions of
proteins to prevent the formation of nonspecific aggre-
gates [9–12] and to improve folding recovery. On the
other hand, arginine has also been reported to stabilize
the exposed hydrophobic area of single-chain Fv frag-
ments [13] and to increase the folding recovery of pro-

teins [14]. The findings reported herein indicate that a
novel compound, RCG, has the ability to accelerate the
formation of the native disulfide bonds, thus resulting
in an improved folding recovery. However, the role of
the arginine residue of RCG in folding reactions
appears to be different from that in the examples
above. Arginine can suppress protein aggregation at
concentrations that exceed  100 mm in such cases [12]
but 1–2 mm RCG was used in our experiments. The
electrostatic influence of local cysteine environments on
disulfide-exchange kinetics was studied using several
types of protein fragment [23]. The cysteine residue,
when adjacent to a positively charged neighbor, reacted
more rapidly, indicating that the adjacent positive
charge accelerates disulfide formation and the exchange
reaction [23]. The mixed-disulfide intermediates of
RCG and prouroguanylin were produced at the first
step in the folding of prouroguanylin, and the mixed-
disulfide bond was then attacked by a free thiol anion
(disulfide-exchange reaction). RCG, ECR and GSH
possess net charges of +1, 0 and )1, respectively, at
neutral pH without a thiol group of the cysteine resi-
due. Local interactions between the positive (RCG)
and the negative (thiol anion) charges may be favorable
for the formation of disulfide bonds. Therefore, the
arginine residue in RCG could accelerate the disulfide-
exchange reaction by its positive charge and effectively
mediate the construction of the native tertiary structure
of proteins. However, the pI values of proteins do not
appear to be important for the reaction. Two types of

proteins were used in this study, based on electrostatic
effects. The pI values for prouroguanylin and lysozyme
are 5.5 and 11.0, respectively, indicating that prourogu-
anylin and lysozyme exist in solution as a negatively
charged protein and a positively charged protein,
respectively. Therefore, the RCG system accelerates
disulfide-exchange reactions, regardless of the net
charge of the protein being studied.
In addition, the ECR results were similar to that of
GSH. ECR (net charge zero) possesses a negative
charge at the side chain of the glutamic acid residue,
and the arginine residue is positively charged. There-
fore, it is possible that the negative charge of the glu-
tamic acid residue offsets the acceleration effect of the
positive charge of the arginine residue on disulfide-cou-
pled folding.
Arginine has been extensively studied as a protein-
folding reagent because of its ability to suppress aggre-
gation during the folding of proteins [10,12,24,25]. It is
generally thought that arginine provides stability for
the hydrophobic surfaces of folding intermediates. The
findings herein indicate an alternative role of the argi-
nine residue. The arginine residue in the glutathione
analog efficiently improved folding recovery and accel-
erated the disulfide-coupled folding of proteins. There
are only a few examples of reagents for the disulfide-
coupled folding of proteins. The disulfide-exchange
reagent, RCG, and related analogs show promise for
serving as a powerful tool for studies, not only of pro-
tein folding but also for the effective recovery of the

correct tertiary structure of target recombinant pro-
teins from the denatured state, such as inclusion bodies
in Escherichia coli cells.
Refolding time (min)
0
–5
–10
–15
[θ] x 10
–3
deg·cm
–2
·dmol
–1
a
b
0 306090
Fig. 6. Molar ellipticity at 222 nm during the folding of prourogu-
anylin. CD measurements were carried out in the presence of (a)
GSH ⁄ GSSG or RCG
red
⁄ RCG
ox
(b) at room temperature.
Acceleration of disulfide-coupled protein folding M. Okumura et al.
1142 FEBS Journal 278 (2011) 1137–1144 ª 2011 The Authors Journal compilation ª 2011 FEBS
In conclusion, replacement of the glutamic acid resi-
due with an arginine residue in the glutathione mole-
cule substantially improved its ability as a redox
reagent for protein folding. The positive charge in close

proximity to the cysteine residue of the redox molecule
effectively accelerates disulfide-exchange reactions.
Materials and methods
Materials
Glutathione, lysozyme and endoproteinase Arg-C were pur-
chased from the Peptide Institute, Inc. (Osaka, Japan), Sei-
kagaku Corporation (Tokyo, Japan) and Takara Bio Inc.
(Kyoto, Japan), respectively. RCG and ECR were synthe-
sized by Greiner Bio-One Co., Ltd. (Tokyo, Japan). All
other chemicals and solvents used were of reagent grade.
RP-HPLC
The HPLC apparatus comprised an ELITE system (Hitachi
High-Technologies Corporation, Tokyo, Japan) equipped
with an L-2400 detector and a D-2500 chromato-integrator.
Proteins were separated by RP-HPLC using a Cosmosil
5C
18
-AR-II column (8 · 250 mm; Nacalai Tesque, Inc.,
Kyoto, Japan) or a Develosil UG-5 column (4.6 · 150 mm;
Nomura Chemical Co., Ltd., Aichi, Japan). Folding yields
were estimated by the HPLC peak area at 220 nm.
MALDI-TOF
⁄
MS
The molecular mass values of proteins were determined
using a DALTONICS ultraflex spectrometer (Bruker
Japan Co., Ltd) in the positive ion mode. Mass spectro-
metric analyses of proteins and peptides were carried out
in the linear or reflector modes using 3,5-dimethoxy-4-hy-
droxycinnamic acid (Tokyo Chemical Industry Co., Ltd.,

Tokyo, Japan), and a-cyano-4-hydroxycinnamic acid
(Sigma-Aldrich Co., Tokyo, Japan) as matrices, respec-
tively. In a typical run, the lyophilized sample
( 0.1 nmol) was dissolved in 0.05% trifluoroacetic acid
aq ⁄ 50% CH
3
CN (1 lL), mixed with 1 lL of a matrix
solution (10 mgÆmL
)1
) and air-dried on the sample plate
for MALDI-TOF ⁄ MS.
CD measurements
Denatured protein (2 nmol) was dissolved in 10 mm
Tris ⁄ HCl (300 lL, pH 8.0) containing GSH ⁄ GSSG (final
concentrations: 2 mm ⁄ 1mm) or RCG
red
⁄ RCG
ox
(final con-
centrations: 2 mm ⁄ 1mm). CD measurements were carried
out in a Model J720WI CD spectrometer (JASCO Corpo-
ration, Tokyo, Japan) at room temperature using a cuvette
with 1 mm path-length.
Folding reactions of lysozyme and
prouroguanylin
The reduction of lysozyme was carried out using a previ-
ously described method, with minor modifications [11].
Briefly, lysozyme (5 mg) was dissolved in 0.1 m Tris ⁄ HCl
(pH 8.3, 1 mL) containing 20 mm dithiothreitol and 8 m
urea, and the solution was allowed to stand for 3 h at

40 °C. The reaction mixture was dialyzed against 10 mm
HCl, lyophilized and stored at )20 °C until used.
Folding reactions were carried out at several protein con-
centrations (0.1–1.6 mgÆmL
)1
of lysozyme; 0.1–0.4 mgÆmL
)1
of prouroguanylin). The denatured ⁄ reduced proteins were
dissolved in 0.1 m Tris ⁄ HCl (pH 8.0) and allowed to
undergo folding in the presence of 2 mm reductant (GSH,
RCG
red
or ECR
red
) and 1 mm oxidant (GSSG, RCG
ox
or
ECR
ox
) at room temperature for 48 h, as described previ-
ously [21]. All solutions used in the refolding experiments
were flushed with N
2
gas, and the reactions were carried
out in a sealed vial under an atmosphere of N
2
.
The kinetic experiments were performed in the same buf-
fer as described above. The reaction mixture (100-lL aliqu-
ots) was removed at several time-points, quenched with an

equivalent volume of 1 m HCl [26] and separated by
RP-HPLC. The HPLC fractions were analyzed by
MALDI-TOF ⁄ MS after lyophilization.
Digestion with endoproteinase Arg-C
The HPLC-purified proteins were dissolved in 0.1 m
Tris ⁄ HCl buffer (pH 8.0) and digested with arginylendo-
peptidase C at 37 °C for 16 h, as previously described
[19,21]. The digests were loaded onto the HPLC column
(Cosmosil 5C
18
-AR-II, 4.6 · 150 mm; Nacalai tesque, Inc.,
Kyoto, Japan) and the peptides were eluted using a linear
gradient of CH
3
CN in 0.05% trifluoroacetic acid from 15%
to 40% at a rate of 0.5% min
)1
and at a flow rate of
1.0 mLÆmin
)1
. The eluant was monitored at 220 nm.
Acknowledgements
We thank Mr Hironori Konishi (Kinki University)
and Dr Keiichi Hosokawa (Tsukuba Molecular Biol-
ogy Laboratory) for the considerable assistance and
discussion of this work. This study was supported by a
research grant (20510207) from Grant-in-Aid for Sci-
entific Research (C) and by a grant from the Naito
Foundation.
References

1 Gilbert HF (1990) Molecular and cellular aspects of
thiol-disulfide exchange. Adv Enzymol Relat Areas Mol
Biol 63, 69–172.
M. Okumura et al. Acceleration of disulfide-coupled protein folding
FEBS Journal 278 (2011) 1137–1144 ª 2011 The Authors Journal compilation ª 2011 FEBS 1143
2 Hwang C, Sinskey AJ & Lodish HF (1992) Oxidized
redox state of glutathione in the endoplasmic reticulum.
Science 257, 1496–1502.
3 Konishi Y, Ooi T & Scheraga HA (1982) Regeneration
of ribonuclease A from the reduced protein. Rate-limit-
ing steps. Biochemistry 21, 4734–4740.
4 Lyles MM & Gilbert HF (1991) Catalysis of the oxida-
tive folding of ribonuclease A by protein disulfide isom-
erase: dependence of the rate on the composition of the
redox buffer. Biochemistry 30, 613–619.
5 Arolas JL, Aviles FX, Chang JY & Ventura S (2006)
Folding of small disulfide-rich proteins: clarifying the
puzzle. Trends Biochem Sci 31 , 292–301.
6 Wedemeyer WJ, Welker E, Narayan M & Scheraga HA
(2000) Disulfide bonds and protein folding. Biochemis-
try 39, 4207–4216.
7 Chatrenet B & Chang JY (1993) The disulfide folding
pathway of hirudin elucidated by stop ⁄ go folding exper-
iments. J Biol Chem 268, 20988–20996.
8 Beld J, Woycechowsky KJ & Hilvert D (2007) Selenog-
lutathione: efficient protein folding by a diselenide. Bio-
chemistry 46, 5382–5390.
9 Arakawa T, Ejima D, Tsumoto K, Obeyama N,
Tanaka Y, Kita Y & Timasheff SN (2007) Suppression
of protein interactions by arginine: a proposed mecha-

nism of the arginine effects. Biophys Chem 127, 1–8.
10 Shiraki K, Kudou M, Fuziwara S, Imanaka T & Takagi
M (2002) Biophysical effect of amino acids on the pre-
vention of protein aggregation. J Biochem 132, 591–595.
11 Reddy KRC, Lilie H, Rudolph R & Lange C (2005)
L-Arginine increases the solubility of unfolded species
of hen egg white lysozyme. Protein Sci 14, 929–935.
12 Arakawa T & Tsumoto K (2003) The effects of arginine
on refolding of aggregated proteins: not facilitate
refolding, but suppress aggregation. Biochem Biophys
Res Commun 304, 148–152.
13 Umetsu M, Tsumoto K, Hara M, Ashish K, Goto S,
Adschiri T & Kumagai I (2003) How additives influence
the refolding of immunoglobulin-folded proteins in a
stepwise dialysis system. Spectroscopic evidence for
highly efficient refolding of a single-chain Fv fragment.
J Biol Chem 278, 8979–8987.
14 Tsumoto K, Shinoki K, Kondo H, Uchikawa M, Juji T
& Kumagai I (1998) Highly efficient recovery of func-
tional single-chain Fv fragments from inclusion bodies
overexpressed in Escherichia coli by controlled introduc-
tion of oxidizing reagents-application to a human sin-
gle-chain Fv fragment. J Immunol Methods 219,
119–129.
15 Saxena VP & Wetlaufer DB (1970) Formation of three-
dimensional structure in proteins. l. Rapid nonenzymic
reactivation of reduced lysozyme. Biochemistry 9,
5015–5023.
16 Creighton TE (1979) Intermediates in the refolding of
reduced ribonuclease A. J Mol Biol 129, 411–431.

17 Raman B, Ramakrishna T & Rao CM (1996) Refolding
of denatured and denatured ⁄ reduced lysozyme at high
concentrations. J Biol Chem 271, 17067–17072.
18 Van den Berg B, Chung EW, Robinson CV, Mateo PL
& Dobson CM (1999) The oxidative refolding of hen
lysozyme and its catalysis by protein disulfide isomer-
ase.
EMBO J 18, 4794–4803.
19 Hidaka Y, Shimono C, Ohno M, Okumura N, Knut A,
Wolf-Georg F & Shimonoshi Y (2000) Dual function of
the propeptide of prouroguanylin in the folding of the
mature peptide: disulfide-coupled folding and dimeriza-
tion. J Biol Chem 275, 25155–25162.
20 Creighton TE (1997) Protein folding coupled to
disulphide bond formation. Biol Chem 378, 731–744.
21 Hidaka Y, Ohno M, Hemmasi B, Hill O, Forssmann
WG & Shimonishi Y (1998) In vitro disulfide-coupled
folding of guanylyl cyclase-activating peptide and its
precursor proteins. Biochemistry 37, 8498–8507.
22 Ybe JA & Kahn PC (1994) Slow-folding kinetics of
ribonuclease-A by volume change and circular dichro-
ism: evidence for two independent reactions. Protein Sci
3, 638–649.
23 Snyder GH, Cennerazzo MJ, Karalis AJ & Field D
(1981) Electrostatic influence of local cysteine environ-
ments on disulfide exchange kinetics. Biochemistry 20,
6509–6519.
24 Bajorunaite E, Sereikaite J & Bumelis VA (2007) l-argi-
nine suppresses aggregation of recombinant hormones
in refolding process from E. coli inclusion bodies.

Protein J 26, 547–555.
25 Buchner J & Rudolph R (1991) Renaturation, purifica-
tion and characterization of recombinant Fab-fragments
produced in Escherichia coli. Biotechnology, 9, 157–162.
26 Weissman JS & Kim PS (1991) Reexamination of the
folding of BPTI: predominance of native intermediates.
Science 253, 1386–1393.
Supporting information
The following supplementary material is available:
Fig. S1. Folding recoveries of lysozyme using
c-GSH ⁄ GSSG (open bars) and a-GSH ⁄ GSSG (shaded
bars).
This supplementary material can be found in the
online version of this article.
Please note: As a service to our authors and readers,
this journal provides supporting information supplied
by the authors. Such materials are peer-reviewed and
may be re-organized for online delivery, but are not
copy-edited or typeset. Technical support issues arising
from supporting information (other than missing files)
should be addressed to the authors.
Acceleration of disulfide-coupled protein folding M. Okumura et al.
1144 FEBS Journal 278 (2011) 1137–1144 ª 2011 The Authors Journal compilation ª 2011 FEBS