Analyses of methionine sulfoxide reductase activities towards free and peptidyl methionine sulfoxides

Article abstract of DOI:10.1016/j.abb.2012.07.009

There have been insufficient kinetic data that enable a direct comparison between free and peptide methionine sulfoxide reductase activities of either MsrB or MsrA. In this study, we determined the kinetic parameters of mammalian and yeast MsrBs and MsrAs for the reduction of both free methionine sulfoxide (Met-O) and peptidyl Met-O under the same assay conditions. Catalytic efficiency of mammalian and yeast MsrBs towards free Met-O was >2000-fold lower than that of yeast fRMsr, which is specific for free Met-R-O. The ratio of free to peptide Msr activity in MsrBs was 1:20-40. In contrast, mammalian and yeast MsrAs reduced free Met-O much more efficiently than MsrBs. Their k_cat values were 40-500-fold greater than those of the corresponding MsrBs. The ratio of free to peptide Msr activity was 1:0.8 in yeast MsrA, indicating that this enzyme can reduce free Met-O as efficiently as peptidyl Met-O. In addition, we analyzed the in vivo free Msr activities of MsrBs and MsrAs in yeast cells using a growth complementation assay. Mammalian and yeast MsrBs, as well as the corresponding MsrAs, had apparent in vivo free Msr activities. The in vivo free Msr activities of MsrBs and MsrAs agreed with their in vitro activities.

Full text of DOI:10.1016/j.abb.2012.07.009

Archives of Biochemistry and Biophysics 527 (2012) 1–5
Contents lists available at SciVerse ScienceDirect
Archives of Biochemistry and Biophysics
journal homepage: www.elsevier.com/locate/yabbi
Analyses of methionine sulfoxide reductase activities towards free
and peptidyl methionine sulfoxides
a
b
a,
⇑
Geun-Hee Kwak , Kwang Yeon Hwang , Hwa-Young Kim
a
Department of Biochemistry and Molecular Biology, Yeungnam University College of Medicine, Daegu 705-717, Republic of Korea
Division of Biotechnology, College of Life Sciences & Biotechnology, Korea University, Seoul 136-701, Republic of Korea
b
a r t i c l e i n f o
a b s t r a c t
Article history:
There have been insufficient kinetic data that enable a direct comparison between free and peptide
methionine sulfoxide reductase activities of either MsrB or MsrA. In this study, we determined the kinetic
parameters of mammalian and yeast MsrBs and MsrAs for the reduction of both free methionine sulfox-
ide (Met-O) and peptidyl Met-O under the same assay conditions. Catalytic efficiency of mammalian and
yeast MsrBs towards free Met-O was >2000-fold lower than that of yeast fRMsr, which is specific for free
Met-R-O. The ratio of free to peptide Msr activity in MsrBs was 1:20–40. In contrast, mammalian and
yeast MsrAs reduced free Met-O much more efficiently than MsrBs. Their kcat values were 40–500-fold
greater than those of the corresponding MsrBs. The ratio of free to peptide Msr activity was 1:0.8 in yeast
MsrA, indicating that this enzyme can reduce free Met-O as efficiently as peptidyl Met-O. In addition, we
analyzed the in vivo free Msr activities of MsrBs and MsrAs in yeast cells using a growth complementation
assay. Mammalian and yeast MsrBs, as well as the corresponding MsrAs, had apparent in vivo free Msr
activities. The in vivo free Msr activities of MsrBs and MsrAs agreed with their in vitro activities.
Ó 2012 Elsevier Inc. All rights reserved.
Received 29 April 2012
and in revised form 19 July 2012
Available online 31 July 2012
Keywords:
Methionine sulfoxide reductase
MsrA
MsrB
Free Msr activity
Kinetic analysis
Growth complementation
Introduction
and subsequently the Msr enzymes are regenerated for the next
cycle of reaction. In the case of Msrs lacking resolving Cys, the sul-
Methionine sulfoxide reductases (Msrs) constitute an important
defense system to protect cells against oxidative stress [1,2]. They
catalyze the reduction of free and protein-based methionine sulf-
oxide (Met-O) to methionine. Two distinct Msr families have
evolved for the reduction of oxidized methionine with a substrate
stereospecificity [3,4]. MsrA is specific for the reduction of the S-
form of Met-O (Met-S-O), whereas MsrB only reduces the R-epimer
of Met-O (Met-R-O). In addition, a new type of Msr, designated
fRMsr, was recently discovered [5]. This enzyme catalyzes the
reduction of only free form of Met-R-O. It cannot reduce protein-
based Met-R-O residues. MsrA and MsrB are present in all three
kingdoms of life from bacteria to humans. However, the occurrence
of fRMsr is limited to unicellular organisms [6]. Multicellular
organisms including mammals do not contain this enzyme.
Catalytic mechanisms of Msr families are similar, and employ a
sulfenic acid chemistry [7]. A catalytic Cys residue attacks the sulf-
oxide moiety of Met-O, and then is oxidized to the Cys sulfenic acid
with concomitant release of the Met product. Another Cys (resolv-
ing Cys) interacts with the sulfenic acid of catalytic Cys to form a
disulfide bond. The disulfide bond is finally reduced by a reductant
fenic acid intermediate can be directly reduced by the reductants
[8]. Thioredoxin (Trx) is generally thought to be an in vivo reduc-
tant, while dithiothreitol (DTT) can be used for an in vitro
reductant.
Although MsrA is active on peptidyl Met-S-O, it also has a sig-
nificant reduction activity towards free Met-S-O [9]. In addition
to free Met-S-O, MsrA can reduce other methyl sulfoxide com-
pounds including dimethyl sulfoxide and sulindac [9–11]. In con-
trast to MsrA, MsrB appears to act mainly on peptidyl Met-R-O
with very weak activity towards free Met-R-O. MsrB from Esche-
richia coli is much less efficient (1000-fold lower) at reducing free
Met-O than E. coli MsrA [4]. In addition, MsrB has been reported to
be unable to reduce methyl sulfoxide compounds that are reduc-
ible by MsrA [11,12]. However, the kinetic data able to directly
compare free Msr activity to peptide Msr activity of either MsrA
or MsrB remain largely unknown.
One MsrA and three MsrBs (referred to MsrB1, MsrB2, and
MsrB3) exist in mammalian cells [13]. Among the three MsrBs,
MsrB1 is a selenoenzyme, in which selenocysteine replaces Cys
in the active site, while MsrB2 and MsrB3 are Cys-containing
homologs. In humans, there are two MsrB3 forms, MsrB3A and
MsrB3B, generated by alternative splicing of the first exon [13].
We previously verified that all the three mammalian MsrBs can
reduce free Met-R-O when assayed with a qualitative method
⇑
Corresponding author. Address: Department of Biochemistry and Molecular
Biology, Yeungnam University College of Medicine, 170 Hyeonchungno, Namgu,
Daegu 705-717, Republic of Korea. Fax: +82 53 654 6651.
E-mail address: hykim@ynu.ac.kr (H.-Y. Kim).
0
003-9861/$ - see front matter Ó 2012 Elsevier Inc. All rights reserved.
http://dx.doi.org/10.1016/j.abb.2012.07.009

2
G.-H. Kwak et al. / Archives of Biochemistry and Biophysics 527 (2012) 1–5
[
13]. However, a study that employed growth complementation
Measurement of free Msr activity of yeast MsrB with various Trxs
experiments showed that human SK-Hep1 cells are unable to re-
duce free Met-R-O, whereas they are capable of reducing free
Met-S-O [14]. Thus, it was of interest to quantitatively determine
the activities of MsrB and MsrA towards free Met-O and to com-
pare those activities each other.
The reaction mixture (200
phate (pH 7.5), 50 mM NaCl, 2 mM free Met-(R,S)-O, 30
0.2 mM NADPH, 0.1 mM EDTA, 14 g human Trx reductase 1, and
10 g of various Trxs. The activity was measured following the
l
l) contained 50 mM sodium phos-
lg MsrB,
l
l
In this work, we determined the kinetic parameters of mamma-
lian MsrA and MsrBs for the reduction of both free and peptidyl
Met-O. In parallel, we also analyzed the kinetic properties of yeast
MsrA and MsrB. We used yeast fRMsr to compare the free Msr
activities of MsrAs and MsrBs. In addition, we analyzed and com-
pared the in vivo free Msr activities of MsrAs and MsrBs using a
growth complementation assay in yeast cells.
above assay protocol.
Yeast constructs and transformation
Mouse MsrB2(24–175) lacking N-terminal signal peptide and
human MsrB3B were PCR-amplified using p423-based constructs
[
13] and cloned into SpeI/HindIII sites of p425 GPD yeast vector.
The N-terminal signal peptide of MsrB2 was removed to make the
protein be expressed in the cytosol of yeast cells. The human
MsrB3B is also expressed in the cytosol of yeast cells as previously
described [18]. Mouse MsrA and S. cerevisiae MsrB in p423-based
constructs [18] were cut with BamHI/XhoI and cloned into the same
sites of p425 GPD vector. Constructs of p425-based S. cerevisiae
MsrA and fRMsr have been described elsewhere [16,19]. The
p425-based constructs were transformed into S. cerevisiae cells defi-
cient in triple msrA/msrB/fRmsr genes (MATa his3 leu2 met15 ura3
Materials and methods
Preparation of MsrA, MsrB and fRMsr enzymes
Mouse MsrA, mouse MsrB2, and human MsrB3A(32–191) lack-
ing N-terminal signal peptide were prepared as described previ-
ously [13,15]. Saccharomyces cerevisiae fRMsr and MsrA were also
prepared as described previously [12,16]. S. cerevisiae MsrB was
PCR-cloned into NdeI/XhoI sites of pET28a vector. The resulting
construct was named pET28-yMsrB and coded for the full-length
of yeast MsrB with an N-terminal His-tag derived from the vector.
The yeast MsrB was overexpressed in E. coli BL21(DE3) and purified
using a Talon-metal affinity resin (Clontech). All purified Msr pro-
teins were analyzed by sodium dodecyl sulfate–polyacrylamide gel
electrophoresis (SDS–PAGE) to check the purity.
D
msrA::URA3 DmsrB::KAN DfRmsr::HIS3) [6] using the lithium ace-
tate method. Transformants were selected for leucine prototrophy.
Growth complementation assay
The triple msrA/msrB/fRmsr-deleted S. cerevisiae cells containing
p425 vector only or p425-based Msr constructs were grown aero-
bically at 30 °C in yeast nitrogen base minimal medium supple-
mented with 2% glucose (YNBD). The overnight cultures were
each adjusted to an optical density of 2.5, 0.25, 0.025, and 0.0025
Preparation of Trxs
at 600 nm via serial dilution. Each diluted sample (5 ll) was spot-
S. cerevisiae cytosolic Trx1 (yTrx1), human cytosolic Trx1
ted onto YNBD agar medium in the presence of 0.14 mM Met or
0.28 mM Met-(R,S)-O. The spotted plates were incubated at 30 °C
and the cell growth was monitored. At least two biological repli-
cates were done.
(
hTrx1), and rat mitochondrial Trx2 (rTrx2) were prepared as
described previously [8,16]. S. cerevisiae mitochondrial Trx3
yTrx3; YCR083W) was PCR-cloned into NdeI/XhoI sites of pET21b
(
and the resulting construct was named pET21-yTrx3. The yTrx3
was overexpressed in E. coli BL21(DE3) and purified using the
Talon-metal affinity resin. The purity of all purified Trxs was
analyzed by SDS–PAGE.
Western blot analysis
The triple msrA/msrB/fRmsr-deleted S. cerevisiae cells containing
p425 vector only or p425-based Msr constructs were grown aero-
bically at 30 °C in YNBD medium containing Met. Whole cell ex-
tracts were prepared by alkali treatment and boiling as described
elsewhere [18]. Polyclonal antibodies against mouse MsrB2, hu-
man MsrB3, mouse MsrA, or yeast fRMsr were used for Western
blot analysis.
Assays for free and peptide Msr activities and kinetic analysis
Free Msr activity was assayed in the Trx-dependent reaction.
The reaction mixture (200
(
l
l) contained 50 mM sodium phosphate
pH 7.5), 50 mM NaCl, 0.2 mM NADPH, 0.1 mM EDTA, 10 g yTrx1,
g human Trx reductase 1, 0.1–4 mM free Met-(R,S)-O (Sig-
g MsrBs or 10 g MsrAs or 1
l
and 14
l
Measurement of peptide Msr activity from yeast cell extracts
ma–Aldrich), and either 30–50
l
l
lg
fRMsr. The mixture of (R,S)-Met-O has also been used for the ki-
netic studies of MsrA, MsrB, and fRMsr enzymes as described else-
where [4,17]. The reactions were carried out at 25 °C for 5 min, and
a decrease in absorbance of NADPH at 340 nm was monitored
using a spectrophotometer UV-160A (Shimadzu). A control for nor-
malization purposes was the reaction mixture without Msr en-
zyme. Peptide Msr activity was measured by the above assay
The triple deletion mutant S. cerevisiae cells containing p425
vector only or p425-based Msr constructs were grown aerobically
for 2 days at 30 °C in YNBD medium containing Met. Yeast cell ex-
tracts were prepared as previously described using glass beads
[
18]. The reaction mixture (100
phate (pH 7.5), 50 mM NaCl, 20 mM DTT, 200
for MsrA or dabsyl-Met-R-O for MsrB, and 300
l
l) contained 50 mM sodium phos-
M dabsyl-Met-S-O
g crude protein.
l
l
procedure with the reaction mixture (200
ll) containing 50 mM
The reaction was carried out at 37 °C for 30 min. The product
Met was analyzed by high-performance liquid chromatography
sodium phosphate (pH 7.5), 50 mM NaCl, 0.2 mM NADPH,
0
0
.1 mM EDTA, 10
.025–0.4 mM either dabsyl-Met-S-O for MsrA or dabsyl-Met-R-O
g MsrA or 10 g MsrB. Enzyme activity
was calculated using a molar extinction coefficient of NADPH
lg yTrx1, and 14 lg human Trx reductase 1,
(
HPLC) as described elsewhere [20].
for MsrB, and either 3
l
l
Results and discussion
À1
À1
(
6220 M cm ) and expressed as nmole of oxidized NADPH per
min. K and kcat values were determined by non-linear regression
using GraphPad Prism 5 software.
m
We first determined the kinetic parameters of the mammalian
MsrBs, MsrB2 and MsrB3, using a mixture of free Met-(R,S)-O for

G.-H. Kwak et al. / Archives of Biochemistry and Biophysics 527 (2012) 1–5
3
the reduction of free Met-O and compared them with those of
yeast fRMsr (Fig. 1; Table 1). The mixture of (R,S)-Met-O, although
not the perfect choice, has been used in similar studies [4,17]. It is
well known that MsrB only acts on the R-form. It cannot reduce the
S-form at all [4,21]. The kcat values of MsrB2 and MsrB3 were 5400-
and 4500-fold lower, respectively, than that of fRMsr. Catalytic effi-
analysis (Fig. 2A). Also, the functionality of MsrB2 and MsrB3
was proven by an enzyme assay for peptide Msr activity
(Fig. 2B). As expected, the yeast cells containing p425 vector only
did not grow in the Met-(R,S)-O medium, while they could grow
in the Met medium (Fig. 3A). Cells expressing MsrB2 did not grow
in the Met-(R,S)-O medium during a 2-day-incubation, while cells
expressing MsrB3 (or fRMsr) showed apparent growth at this cul-
ture time. However, the growth of cells expressing MsrB2 seemed
to begin in the Met-(R,S)-O medium at a 3-day-incubation and was
apparently supported at a 6-day-incubation. The data indicate that
both MsrB2 and MsrB3 have in vivo reduction ability towards free
Met-O.
In the growth complementation assay, cells expressing MsrB3
grew much slower than cells expressing fRMsr in the Met-(R,S)-O
medium. The growth rate of MsrB3-expressing cells appeared to
be at least two orders of magnitude less than that of fRMsr-
expressing cells, as judged from the spotting assay. This difference
in growth between MsrB3- and fRMsr-expressing cells agreed well
with the kinetic data, which showed a 2800-fold lower catalytic
efficiency of MsrB3 compared to fRMsr. Also, in the Met-(R,S)-O
medium there was a big difference in growth between MsrB2-
and MsrB3-expressing yeast cells, although the catalytic efficien-
cies of both enzymes were similar (Table 1). The growth rate of
MsrB2-expressing cells was approximately three orders of magni-
tude less than that of MsrB3-expressing cells (Fig. 3A). However,
the peptide Msr activity from the cell extract of MsrB2-expressing
cells was found to be even higher (5-fold) than that from the cell
extract of MsrB3-expressing cells (Fig. 2B). It should be noted that
the both transgenic MsrB2 and MsrB3 are localized in the cytosol of
yeast cells [18] and cytosolic yeast Trx1 was used in the in vitro
free Msr activity assay condition. Therefore, it is unlikely that this
big difference in growth is due to different access to the reductant
m
ciency (kcat/K ) of the two mammalian MsrB enzymes was >2400-
fold lower than that of fRMsr, indicating much lower ability to re-
duce free Met-O. To compare the free Msr activities of MsrB2 and
MsrB3 with their peptide Msr activities, we next analyzed the ki-
netic parameters of these enzymes for the reduction of peptidyl
Met-R-O (Table 1). For a comparison, the peptide Msr activity
was measured under the same assay conditions for the measure-
ment of free Msr activity. As anticipated, yeast fRMsr had no activ-
ity towards peptidyl Met-R-O. K
m
value of MsrB2 towards peptidyl
Met-O was 2-fold lower than that towards free Met-O, while K
m
value of MsrB3 towards peptidyl Met-O was 4-fold greater than
that towards free Met-O. Catalytic efficiencies of MsrB2 and MsrB3
to reduce peptidyl Met-O were 35- and 19-fold higher, respec-
tively, than those for the reduction of free Met-O. Taken together,
the data clearly indicate that mammalian MsrB2 and MsrB2 have
very weak activity towards free Met-O, relative to the activity to-
wards peptidyl Met-O.
We next performed in vivo analysis of free Msr activities of
MsrB2 and MsrB3 via a growth complementation assay. We gener-
ated p425-based MsrB2- and MsrB3-expressing constructs and
separately transformed the constructs into a triple msrA/msrB/
fRmsr-deleted S. cerevisiae strain, which was unable to grow in a
Met-O medium but could grow in a Met medium [6]. We also
transformed p425 empty vector and fRMsr-expressing construct
into the triple deletion mutant cells. The overexpression of MsrB2
and MsrB3 as well as fRMsr was confirmed by Western blot
Fig. 1. Kinetic analysis of free Msr activities of mammalian MsrBs and MsrA. A mixture of Met-(R,S)-O was used for measuring the free Msr activities of the purified enzymes.
cat and K values were calculated from non-linear regression. The K values should be divided by 2 because of use of the mixture of (R,S)-Met-O. The kcat and K values are
summarized in Table 1.
k
m
m
m

4
G.-H. Kwak et al. / Archives of Biochemistry and Biophysics 527 (2012) 1–5
Table 1
Kinetic parameters of free and peptide Msr activities.
Protein Free Msr activity
Peptide Msr activity
Ratio of free to peptide Msr activitya
À1
À1
kcat (min
)
K
m[free Met-O]
k
cat/K
m
k
cat (min
)
K
m[dabsyl-Met-O]
kcat/K
m
(mM min )
À1
À1
À1
À1
(
l
M)
(mM min
)
(l
M)
MsrB2
MsrB3
yMsrB
fRMsr
MsrA
0.05 ± 0.01
0.06 ± 0.002 50 ± 5
0.06 ± 0.002 51 ± 4
270 ± 20
2.6 ± 0.2
30 ± 4
36 ± 17
1.4
1.2
1.2
3380
7.4
0.73 ± 0.06
4.4 ± 0.5
1.5 ± 0.1
0
5.3 ± 1.3
3.3 ± 0.2
15 ± 6
190 ± 20
30 ± 9
–
51 ± 38
24 ± 7
49
23
50
–
104
138
1:35
1:19
1:42
–
1:14
1:0.8
80 ± 27
350 ± 60
180 ± 80
yMsrA
170
A mixture of Met-(R,S)-O was used for measuring the free Msr activities of MsrB, fRMsr, and MsrA. Dabsyl-Met-R-O and -Met-S-O were used for measuring for the peptide Msr
activities of MsrB and MsrA, respectively. All assays were performed with purified enzymes. kcat and K values were calculated from non-linear regression. The Km[free Met-O]
m
values were obtained by dividing by 2 because of use of the mixture of (R,S)-Met-O.
a
Catalytic efficiency towards free Met-O to that towards peptidyl Met-O. MsrB2, mouse MsrB2; MsrB3, human MsrB3; yMsrB, S. cerevisiae MsrB; fRMsr, S. cerevisiae fRMsr;
MsrA, mouse MsrA; yMsrA, S. cerevisiae MsrA.
Msr activity of yeast MsrB via the growth complementation assay.
The expression of yeast MsrB in the triple deletion mutant cells
was confirmed by peptide Msr assay using cell extracts (Fig. 2B).
As shown in Fig. 3B, cells expressing yeast MsrB grew much slower
in the Met-(R,S)-O medium, compared to cells expressing fRMsr.
The growth rate of yeast MsrB-expressing cells was approximately
two orders of magnitude less than that of fRMsr-expressing cells.
This in vivo analysis of free Msr activity agreed well with the ki-
netic data. The growth of yeast MsrB-expressing cells was similar
to that of MsrB3-expressing cells (Fig. 3).
Yeast MsrB is targeted to mitochondria [22]. Thus, we specu-
lated that a mitochondrial Trx form may be a better reductant
for the yeast MsrB than a cytosolic form. We prepared cytosolic
and mitochondrial Trxs from both yeast and mammalian cells
and used them for measuring free Msr activity of yeast MsrB. The
specific free Msr activity was found to be higher (2–4-fold) with
the mitochondrial forms of Trx than with the cytosolic forms (Ta-
ble 2), indicating the preference of mitochondrial Trx to cytosolic
Trx in the yeast MsrB’s free Met-O reduction reaction.
We then determined the kinetic parameters of mammalian and
Fig. 2. Western blot analysis (A) and measurement of peptide Msr activity (B) of
recombinant MsrAs and MsrBs in yeast cells. Cell extracts were prepared from S.
cerevisiae cells deficient in msrA/msrB/fRmsr genes and containing p425 vector only
or indicated Msr constructs cultured for 2 days in YNBD medium containing Met.
Peptide Msr activity from cell extracts was measured in DTT-dependent reaction
using HPLC analysis. EV, p425 empty vector; MsrB2, mouse MsrB2; MsrB3, human
MsrB3; fRMsr, S. cerevisiae fRMsr; yMsrB, S. cerevisiae MsrB; MsrA, mouse MsrA;
yMsrA, S. cerevisiae MsrA; ND, not detectable.
yeast MsrAs for the reduction of both free and peptidyl Met-O
(Fig. 1; Table 1). The kcat value of mouse MsrA towards free Met-
O was 100-fold lower than that of fRMsr, but this value was much
greater (>40-fold) than those of mammalian and yeast MsrBs. In
the case of yeast MsrA, its kcat value towards free Met-O was
>500-fold greater than those of MsrBs. Catalytic efficiencies of
mouse and yeast MsrAs were >5- and >120-fold higher, respec-
tively, than those of MsrBs. The data indicate that mammalian
and yeast MsrAs are much more efficient at reducing free Met-O
than the corresponding MsrBs, as previously characterized in
E. coli MsrA and MsrB forms [4]. Catalytic efficiency of mouse MsrA
towards free Met-O was 14-fold lower than that towards peptidyl
Met-O. However, interestingly, catalytic efficiency of yeast MsrA
was found to be even greater than that towards peptidyl Met-O.
Together, the kinetic data suggest that MsrAs have an efficient free
Msr activity. The efficient reduction ability of MsrAs towards free
Met-O was observed profoundly in the growth complementation
experiments. Growth of both mouse and yeast MsrA-expressing
cells in the Met-(R,S)-O medium was highly comparable to that
of fRMsr-expressing cells (Fig. 3B). Mouse MsrA-expressing cells
grew just slightly slower, compared to fRMsr-expressing cells;
the growth of yeast MsrA-expressing cells was similar to that of
fRMsr-expressing cells.
Trx1. However, it is possible that there are other reductants that
preferably reduce MsrB3 over MsrB2 in yeast cells. Taken together,
our data suggest that MsrB2 is much less efficient at reducing free
Met-O in vivo than MsrB3 for unknown reasons.
An earlier study has shown that human SK-Hep1 cells are un-
able to grow in the free Met-R-O medium [14]. However, the pres-
ent study clearly demonstrates the in vivo reduction abilities of
MsrB2 and MsrB3 towards free Met-R-O in yeast cells. This discrep-
ancy is likely due to the much lower expression levels of both
endogenous MsrB2 and MsrB3 enzymes in SK-Hep1 cells compared
to those in yeast cells overexpressing MsrB2 or MsrB3. In addition,
the much lower free Msr activities of MsrB2 and MsrB3, when
compared to MsrA as described below, will contribute to the
growth inability of mammalian cells on the free Met-R-O medium.
We also separately measured kinetic parameters of yeast MsrB
for determining its free and peptide Msr activities (Table 1). Cata-
lytic efficiency of the yeast MsrB towards free Met-O was similar to
those of MsrB2 and MsrB2, displaying a 2800-fold lower value than
yeast fRMsr. As expected, catalytic efficiency towards peptidyl
Met-O was much higher (42-fold) than that towards free Met-O.
Thus, these data support the evidence that MsrB enzyme is poorly
active on free Met-O. We also conducted the in vivo analysis of free
The foregoing results lead us to directly compare free Msr activ-
ity with peptide Msr activity in MsrBs and MsrAs. The kinetic data
revealed that mammalian and yeast MsrBs were much less effi-
cient (20–40-fold) at reducing free Met-O than peptidyl Met-O.
In contrast, the corresponding MsrAs reduced free Met-O much
more efficiently than MsrBs. The turnover number of MsrAs was

G.-H. Kwak et al. / Archives of Biochemistry and Biophysics 527 (2012) 1–5
5
Fig. 3. Growth complementation assays. Cell growth of triple msrA/msrB/fRmsr-deleted S. cerevisiae cells containing p425 vector only or indicated Msr constructs was
analyzed in the Met (left) or Met-(R,S)-O (right) medium via serial dilution. The spotted cells were incubated at 30 °C. (A) Growth complementation of mammalian MsrBs. The
pictures were taken after 2, 3, and 6 days of incubation. (B) Growth complementation assay of mammalian and yeast MsrAs and yeast MsrB. The pictures were taken after
2
days of incubation. EV, p425 empty vector; MsrB2, mouse MsrB2; MsrB3, human MsrB3; fRMsr, S. cerevisiae fRMsr; yMsrA, S. cerevisiae MsrA; MsrA, mouse MsrA; yMsrB, S.
cerevisiae MsrB.
Table 2
References
Specific activity of yeast MsrB towards free Met-O with
various Trxs.
[
[
[
1] J. Moskovitz, Biochim. Biophys. Acta 1703 (2005) 213–219.
2] H.Y. Kim, V.N. Gladyshev, Biochem. J. 407 (2007) 321–329.
3] N. Brot, L. Weissbach, J. Werth, H. Weissbach, Proc. Natl. Acad. Sci. USA 78
Trx
Activity (nmol/min/mg)
(
1981) 2155–2158.
hTrx1
yTrx1
rTrx2
yTrx3
0.5 ± 0.3
0.4 ± 0.2
0.8 ± 0.4
1.6 ± 0.2
[
[
[
[
4] R. Grimaud, B. Ezraty, J.K. Mitchell, D. Lafitte, C. Briand, P.J. Derrick, F. Barras, J.
Biol. Chem. 276 (2001) 48915–48920.
5] Z. Lin, L.C. Johnson, H. Weissbach, N. Brot, M.O. Lively, W.T. Lowther, Proc. Natl.
Acad. Sci. USA 104 (2007) 9597–9602.
6] D.T. Le, B.C. Lee, S.M. Marino, Y. Zhang, D.E. Fomenko, A. Kaya, E. Hacioglu, G.H.
Kwak, A. Koc, H.Y. Kim, V.N. Gladyshev, J. Biol. Chem. 284 (2009) 4354–4364.
7] B.C. Lee, A. Dikiy, H.Y. Kim, V.N. Gladyshev, Biochim. Biophys. Acta 1790 (2009)
Free Msr activity was measured using various Trxs and free
Met-(R,S)-O as described in Materials and methods section.
All assays were performed with purified enzymes. hTrx1,
human cytosolic Trx1; yTrx1, S. cerevisiae cytosolic Trx1;
rTrx2, rat mitochondrial Trx2; yTrx3, S. cerevisiae mito-
chondrial Trx3.
1
471–1477.
[
[
8] H.Y. Kim, J.R. Kim, Biochem. Biophys. Res. Commun. 371 (2008) 490–494.
9] J. Moskovitz, H. Weissbach, N. Brot, Proc. Natl. Acad. Sci. USA 93 (1996) 2095–
2
099.
[
[
10] F. Etienne, L. Resnick, D. Sagher, N. Brot, H. Weissbach, Biochem. Biophys. Res.
Commun. 312 (2003) 1005–1010.
11] B.C. Lee, D.E. Fomenko, V.N. Gladyshev, ACS Chem. Biol. 6 (2011) 1029–1035.
4
0–500-fold greater than that of MsrBs. Yeast MsrA reduced free
Met-O as efficiently as it reduced peptidyl Met-O. In addition,
use of a growth complementation assay demonstrated in vivo free
Msr activity in the mammalian and yeast MsrBs, as well as the cor-
responding MsrAs. The in vivo free Met-O reduction activities of
MsrBs and MsrAs agreed with, but not wholly supported their
in vitro activities. There may be a different type of in vivo reducing
system between MsrA and MsrB, attributed by their different sub-
cellular localization.
[12] G.H. Kwak, S.H. Choi, J.R. Kim, H.Y. Kim, BMB Rep. 42 (2009) 580–585.
[
[
[
13] H.Y. Kim, V.N. Gladyshev, Mol. Biol. Cell 15 (2004) 1055–1064.
14] B.C. Lee, D.T. Le, V.N. Gladyshev, J. Biol. Chem. 283 (2008) 28361–28369.
15] H.Y. Kim, V.N. Gladyshev, Biochemistry 44 (2005) 8059–8067.
[16] G.H. Kwak, M.J. Kim, H.Y. Kim, Biochem. Biophys. Res. Commun. 395 (2010)
12–415.
17] A. Gruez, M. Libiad, S. Boschi-Muller, G. Branlant, J. Biol. Chem. 285 (2010)
5033–25043.
[18] G.H. Kwak, J.R. Kim, H.Y. Kim, BMB Rep. 42 (2009) 113–118.
4
[
2
[
[
[
[
19] A. Koc, A.P. Gasch, J.C. Rutherford, H.Y. Kim, V.N. Gladyshev, Proc. Natl. Acad.
Sci. USA 101 (2004) 7999–8004.
20] R.A. Kumar, A. Koc, R.L. Cerny, V.N. Gladyshev, J. Biol. Chem. 277 (2002)
37527–37535.
21] G.V. Kryukov, R.A. Kumar, A. Koc, Z. Sun, V.N. Gladyshev, Proc. Natl. Acad. Sci.
USA 99 (2002) 4245–4250.
22] A. Kaya, A. Koc, B.C. Lee, D.E. Fomenko, M. Rederstorff, A. Krol, A. Lescure, V.N.
Gladyshev, Biochemistry 49 (2010) 8618–8625.
Acknowledgment
This work was supported by a Grant from the National Research
Foundation of Korea (2011-0028166).