MINISTRY OF EDUCATION AND

MINISTRY OF AGRICULTURE

TRAINING

AND RURAL DEVELOPMENT

VIETNAM ACADEMY OF AGRICULTURAL SCIENCES
----------------------------------------

CHU DUC HA

EXPRESSION PROFILES OF THE GENES ENCODING ENZYME
METHIONINE SULFOXIDE REDUCTASE IN ARABIDOPSIS AND
SOYBEAN UNDER HIGH SALT AND DROUGHT STRESSES

Specialization: Biotechnology
Code:

9420201

SUMMARY OF THE PH.D. THESIS

HA NOI – 2018


The research work was conducted at:
VIETNAM ACADEMY OF AGRICULTURAL SCIENCES

Supervisors: 1. Dr. Le Tien Dung

2. Dr. Pham Thi Ly Thu

Critic 1:

Critic 2:

Critic 3:

The thesis will be presented in the PhD dissertation committees of
Vietnam Academy of Agricultural Sciences at:
Vietnam Academy of Agricultural Sciences
At ................. in ................ 2018

This thesis can be referred at:
1. Vietnam National Library
2. The Library of Vietnam Academy of Agricultural Sciences


1
INTRODUCTION
1. The rationale of the thesis
Adverse environmental conditions, including biotic and abiotic stresses,
can have negative effects on the growth and development of the plants.
Literally, the major mechanism of the attack(s) of stress(es) could be explained
by the high accumulation of the reactive oxygen species (ROS) in the plant
cells. To respond to these stresses, the number of regulatory and functional
proteins were determined to play important roles in the regulation of the
growth and development of the plant. Among them, enzyme methionine
sulfoxide reductase (MSR), including methionine-S-sulfoxide Reductase
(MSRA) and methionine-R-sulfoxide reductase (MSRB), is well-known to

function in the reduction of the MetO into Met. Previously, MSRs were
determined to involve in the regulation of various biological processes, including
the stress response, in the plant. Therefore, the study on the MSR could get
insight into the mechanism of the adaptation of plant to oxidative stress(es) and
provide the candidate MSR-coding genes for the improvement of the stress
tolerance in the plant by genetic engineering.
To address these questions, the model plant Arabidopsis thaliana and
soybean (Glycine max) were used to carry out the experiments in the Ph.D. thesis,
namely "Expression analysis of the genes encoding methionine sulfoxide
reductase in the salinity and drought stresses in Arabidopsis and soybean".
2. The purpose of the thesis
The aims of this Ph.D. thesis are to analyze the roles of enzyme MSRs in
the plant response to abiotic stress(es) and to figure out the potential proteins
that need to repaired by MSR in plant cells. To answer this question, these
contents were described as below:
 Identification and structural characterization of the Methionine-rich
proteins in the model plant Arabidopsis thaliana.
 Identification and structural characterization of the Methionine-rich
proteins in soybean.
 Identification and characterization of the genes encoding methionine
sulfoxide reductase in soybean.
 Expression analysis of the genes encoding methionine sulfoxide reductase


2
under the drought and salinity conditions in soybean plants.
3. Scientific and realistic meaning of the thesis
Results obtained from the Ph.D. thesis were an intensive understanding of
the role(s) of the methionine-rich proteins in A. thaliana, which get insight into
the stress response in the plant via the methionine oxidation pathway.

Additionally, our identification and functional characterization of the genes
encoding methionine sulfoxide reductase(s) in soybean could provide a basic
understanding of the role(s) of these enzymes in the stress response, and thus,
suggested the candidate genes for further genetic engineering for the
improvement of the stress tolerance of chickpea plants.
4. The contributions of the thesis
This Ph.D. thesis was systematically analyzed the methionine-rich proteins
in the model plant A. thaliana and soybean, thus, here is the first study in
Vietnam. Furthermore, the stress-responsive methionine-rich protein encoded
genes found in this thesis could provide reliable evidence of the protein
susceptible to the methionine oxidation in the plant.
More significantly, the identification, annotation and characterization of the
MSR gene family in soybean are the first reports in Vietnam. Together with the
previous study of GmMSRB, this thesis has successfully revealed the function
of genes encoding MSR in soybean. Interestingly, this thesis also provided
several stress-responsive MSR gene for genetic engineering for the
improvement of the stress tolerance in plants.
5. The structure of the thesis
The Ph.D. thesis was designed in 106 pages (excluding the Reference and
Appendix) as following this order: Introduction (3 pages), Section I:
Literature Review (34 pages), Section 2: Materials and Methods (13 pages),
Section 3: Results and Discussion (54 pages), Conclusions and
Recommendation (1 page). There were 186 references, including 8
Vietnamese publications and 178 international publications were cited in the
Ph.D. thesis. Additionally, there were 15 tables, 29 figures, 7 appendix were
presented in the Ph.D. thesis. More importantly, 4 research articles were
published based on the results of the Ph.D. thesis.


3

SECTION 1
LITERATURE REVIEW
The Ph.D. thesis were cited various references and summarized them in 4
major contents as following: The impacts of the adverse environmental
conditions on the plants; Function(s) of the enzyme methionine sulfoxide
reductase and methionine oxidation in plants; Methionine and the methioninerich proteins in plants; Potential application of the enzyme methionine
sulfoxide reductase in the improvement of the stress tolerance in plants.
Briefly, the accumulation of the ROS in the organelles is well-established
to cause the serious damages to the macromolecules in the plant cells [127]. To
adopt on the oxidative stress(es), many mechanisms were developed with the
involvements of various distinct protein groups, such as enzymic and
nonenzymic antioxidants [77]. Among them, the repair of the protein oxidation
was the important process that was attracted from the research community [47,
80, 83]. For instance, approximately 68% macromolecules which easily oxidized
by the ROS in the plant cell are reported as proteins [138]. A hypothesis was
thought that the oxidation of the methionine residues in the polypeptide
sequences was a defense mechanism of the plant cells. To address this question,
three questions were raised that, how many proteins are susceptible with the
methionine oxidation in the plant cells; what their general characteristics are; and
how these proteins-encoded genes were responded under stress conditions.
On the other hand, the methionine oxidation could be repaired by the
involvement of enzyme methionine sulfoxide reductases (MSRs), including two
enzyme families,
[145]. Until now, the genes encoding MSR
were studied in various plant species, including A. thaliana [145] and several
important crops, such as rice [65], tomato [41, 41], maize [186]. Recently, an effort
has been made to identify the MSR gene family in soja [159]. Previously, Le et al
also reported 5 genes encoding GmMSRB in soybean [99]. Unfortunately, no
information of the GmMSRA gene family in soybean was reported. Therefore, it is
interesting to raise three questions that how many genes are encoding MSRA in

soybean; what are their typical characteristics; how this gene family was expanded
during the evolution and how these genes respond to stress conditions.
To sum up, all major contents above were presented in Section 1. Based on
the understanding provided in the literature review, it is strongly believed that
the study on MSR and MRP in soybean are very important.


4
SECTION 2
MATERIALS AND METHODS
2.1. Materials
2.1.1. Data for the computational approach
The genome and proteome of A. thaliana, ecotype Col-0 [91].
The genome and proteome of soybean 'Williams 82' [149].
2.1.2. Materials for the experimental approach
A. thaliana ecotype Col-0 seeds were provided from RIKEN CSRS - Japan.
Arabidopsis seeds overexpressing At3G55240 were provided from RIKEN
CSRS - Japan as previously reported [75].
Soybean cultivar 'Williams 82' seeds were stored in Agricultural Genetics
Institute - Vietnam.
2.2. Duration and place of the study
2.2.1. Duration
The Ph.D. thesis was carried out from 2014 to 2017.
2.2.2. Place
International Laboratory for Cassava Molecular Breeding, National Key
Laboratory for Plant Biotechnology, Agricultural Genetics Institute, Pham Van
Dong Road, North Tu Liem District, Ha Noi, Viet Nam.
Department of Molecular Biology, Agricultural Genetics Institute, Pham
Van Dong Road, North Tu Liem District, Ha Noi, Viet Nam.
Stress Adaptation Research Unit, RIKEN CSRS, Japan.

2.3. Methods
2.3.1. Computational approach
a. Identification of the genes encoding the methionine-rich protein in
Arabidopsis
The current proteome database of A. thaliana [91] was obtained to screen
all proteins by a javascript with the criteria as follows: protein length > 95
amino acid residues and Met > 6 %.
b. Identification of the genes encoding the methionine-rich protein in
soybean
The current proteome database of soybean [145] was used to screen all
proteins by a javascript with the criteria as follows: protein length > 95 amino
acid residues and Met > 6 %.


5
c. Functional classification of the genes encoding MRP in Arabidopsis and
soybean
A list of identifiers of genes encoding MRP in Arabidopsis and soybean
was obtained to analyze in the MAPMAN software [168].
d. Structural characterization of the MRP in Arabidopsis and soybean
Amino acid compositions of MRPs were calculated by BioEDIT tool [68].
The lengths of MRPs were identified in NCBI. The subcellular localization of
MRPs was predicted by ChloroP [49, 50], WoLF PSORT [73], CELLO [184]
and Blast2GO Basic [35].
e. Prediction of the cis-regulatory elements in the promoter regions of the
genes encoding MRPs in Arabidopsis
An 1000-bp-upstream region in the promoter region of genes encoding
MRPs in Arabidopsis was used to analyze by BioEDIT [68] to identify the
presence(s) of the abscisic acid responsive element (ABRE) [84, 133], MYB
recognition site (MYBR) and MYC recognition site (MYCR) [173].

f. Expression patterns of genes encoding MRP in Arabidopsis
Expression patterns of genes encoding MRP in A. thaliana were analyzed
in the normal condition based on the microarray database namely GDS416
[26] and an RNA-seq database available in TraVA web-based tool [86], and
under stress conditions based on the microarray databases [122, 123].
g. Expression patterns of genes encoding MRP in soybean
Expression patterns of genes encoding MRP in soybean were analyzed in
the normal condition based on the public information in Missouri University
[104] and a microarray under drought stress [96].
h. Identification of the genes encoding MSRA in soybean
At5G61640, AtMSRA1 in A. thaliana [145], was used as a seed sequence
to BlastP against the proteome database of soybean [149] in the Phytozome
[60]. The candidate proteins were then confirmed in the Pfam. The gene,
protein and locus identifiers, chromosomal distribution were identified in
Phytozome [60] and NCBI [149].
i. Prediction of the gene duplication events
The gene duplication events were predicted by the Plant Genome Duplication
Database website [100]. The identity of duplicated genes was calculated by the
ClustalX tool [92]. The number of nonsynonymous substitutions per
nonsynonymous site (Ka) and the number of synonymous substitutions per


6
synonymous site (Ks) were predicted by the DNAsp software [105].
j. Structural characterization of MSRA in soybean
The lengths of MSRAs were identified in NCBI [149]. The molecular weight
and theoretical isoelectric point (pI) were calculated in Expasy [58]. The
subcellular localization of MSRAs was predicted by TargetP [49-51]. The
multiple alignments of the conserved domains were analyzed by ClustalX [169].
Unrooted phylogenetic trees were constructed based on the Neighbor-Joining

method in MEGA [89].
k. Prediction of the cis-regulatory elements in the promoter regions of the
genes encoding MSR in soybean
The presence(s) of the cis-regulatory elements were identified in the 1000bp-upstream promoter regions of the genes encoding MSR in soybean by using
PlantCARE [101].
l. Expression patterns of the genes encoding MSRs in soybean
Expression patterns of the MSR genes were analyzed in the normal condition
and stress conditions based on the public transcriptome database [24, 96].
m. Design of primers for the qRT-PCR
Primers for the quantitative realtime - PCR (qRT-PCR) were designed by
using Primer3 [147]. The specific primers were identified based on the
multiple alignments of the genome of soybean [149]. Fbox was used as a
reference gene in this study [97].
2.3.2. Experimental approach
a. Evaluation of the morphology of the transgenic Arabidopsis line
The transgenic Arabidopsis line and the control were sowed in the petri
dishes containing ½
( urashige-Skoog) agar with hygromycine [75], in
the normal condition with the photoperiod 16 h light/8 h dark, temperature 24
± 2 oC [36].
b. Arabidopsis plant treatment and sample collection
The transgenic Arabidopsis seeds were germinated on the petri dishes
containing ½
with NaCl 175 mM. Similarly, the 12-day-old plants were
transferred to the petri dishes containing ½
with CdCl2 750 µ . The
paraquat treatment of the transgenic line was performed as previously
described [182].
c. Soybean plant treatment and sample collection
Soybean plants were treated with drought stress as following exactly



7
previous study [98]. For instance, the 12-day-old plants were kept on the bench
at the indicating time, 0 - 2 - 10 h. The controls were kept under the
hydroponic condition at 2 - 10 h.
For the salinity and ABA treatments, 12-day-old plants were treated in the
solution containing NaCl 250 mM and ABA 100 μ , respecti ely. The plants
were then kept in the normal condition at the indicating time, 0 - 2 - 10 h. The
controls were kept under the hydroponic condition at 2 - 10 h.
d. Total RNA isolation and the cDNA synthesis
Total RNA isolation was carried out based on the TRIZol-based approach
(ThermoFisher Scientific, USA). For instance, total RNA was treated with the
Ambion Turbo DNAse I kit (Ther
isher cienti ic,
a
). Treated RNA
was then used to synthesize cDNA using ReverTra Ace qPCR RT kit (Toyobo,
Japan). The procedures were followed the previous study [98].
e. The qRT-PCR and the data analysis
The qRT-PCR steps were carried out by using the Stratagen MX3000P
system (Agilent Technologies, USA) with the Thunderbird SYBR qPCR Mix
kit (Toyobo, Japan) as previously described [98].
The Δ-CT method was used to analyze the data as previously reported [98].
SECTION 3
RESULTS AND DISCUSSION
3.1. Identification and characterization of the methioinine-rich proteins in
Arabidopsis
3.1.1. Identification, annotation of the genes encoding methioinine-rich
proteins in Arabidopsis

A total of 121 MRP was identified in this study. All MRPs were noted to
have the gene identifiers in the current genome assembly, no MRP gene was
annotated in the unplaced scaffolds. Particularly, the most of AtMRP genes were
distributed on five chromosomes, while only two genes were noted to located on
the mitochrondial. To get insight into the AtMRPs, these genes were then used for
functional classification based on the TAIR10 reference.
3.1.2. Functional classification of the MRPs in Arabidopsis
Our results showed that AtMRP genes may involve in various biological
processes in Arabidopsis plants. Particularly, 20 AtMRP genes were found to
be associated with the transcription regulation. Among them, several genes
were noted to encode the transcription factors. For instance, At4G34590


8
encodes leucine zipper 11, containing 6.33% Met, which was characterized to
involve in the growth of the root system via the auxin signaling in A. thaliana
[180]. At3G23050, encodes a member in the Indole-3-acetic acid family, was
well-established to repress the expression of auxin-induced genes in plants.
Previously, At3G23050 was also known to involve in various biological
processes related to the auxin signaling pathway, such as the light
responsiveness [148], root development [137] and stress response.
Additionally, a number of MRP genes in Arabidopsis was also functionally
categorized into some important biological processes, such as protein
modification (12 AtMRPs - 11%), signaling (6 AtMRPs - 5%), metal
transportation (6 AtMRPs - 5%), RNA processing (4 AtMRPs - 4%), cell cycle
(3 AtMRPs - 3%), metal handling (2 AtMRPs - 2%), development and stress
response shared 1 AtMRP - 1%. On the other hand, 56 AtMRPs have been not
annotated in any specific functional categories. Our results provided a list of
potential proteins for further functionally characterizations.
3.1.3. Structural characterization of MRP in Arabidopsis

As the results, a total of 23, 26 and 16 ABREs, MYBRs and MYCRs has
been identified in the promoter regions of MRP genes in A. thaliana. There
were 65 CREs were found to locate on the promoter regions of 121 MRP genes
(0.54 CRE per MRP gene). More importantly, the presence(s) of ABRE in
promoter regions of MRP genes indicated that these genes might be involved
in the ABA-dependent manner, while the occurrence(s) of MYBR and MYCR
revealed that these MRP genes may respond to the abiotic stress [173]. To sum
up, our prediction strongly suggested the majority of MRP genes may involve
in the stress response via the ABA-dependent and/or -independent pathways.
It is also very important to know the subcellular localizations of MRPs as
suggesting their roles in the cells. Our results suggested that 21 MRPs might
be distributed in the chloroplast via the WolF PSORT and/or ChloroP tools,
while 9 MRPs could be found in the mitochondria through the WolF PSORT
and/or CELLO software. These predictions were also confirmed by the
Blast2GO software.
3.1.4. Expression patterns of the genes encoding MRPs in Arabidopsis
Based on the gene identifier, expression patterns of 49 genes, covering
40.5% AtMRPs, have been found in three major organs, including leaves,
shoots and flowers in Arabidopsis plants in the normal condition [26]. Among
them, six MRP genes were noted to strongly expressed in all organs. For instance,


9
three of the six genes are At3G05220 encodes protein related to the metal
transportation, At4G35070 encodes enzyme S-ribonuclease and At2G46600.
Two of the six genes, namely At1G67350 and At5G39570 were still unknown
function, while At4G32470 encodes enzyme cytochrome ubiquinol oxidase also
highly expressed in both leaves, shoots and flowers in the normal condition.
According to the TraVa database, the majority of MRP genes was found to
highly expressed in at least one major organ, especially the flowers and flower

buds in the mature stage in plants.
Expression of the approximately 50% MRP genes was found to be
significantly altered under the salinity condition [122] and/or the drought
treatment [123]. Particularly, 11 and 17 genes were up- and down-regulated by at
least two-folds under the salinity treatment, respectively. Under the drought
treatment, 23 and 16 genes were found to be induced and reduced, respectively.
Six and three genes were up- and down-regulated in both stress conditions,
respectively. Interestingly, we also found that the majority of these genes encoded
the proteins which are known to be specific in plants. Among them, At3G59900
was noted to be induced by 10.7-fold and reduced by -2.6-fold under drought and
salinity conditions, respectively (Table 3.4).
Table 3.4. Expression patterns of MRP genes under
drought and salinity conditions
#

Gene
name

Met
(%)

1
2
3
4
5
6
7
8
9

10

At1G32560
At1G33860
At3G55240
At3G59900
At3G62090
At4G12334
At4G33467
At4G34590
At5G42325
At5G67390

6,02
8,55
6,12
6,20
6,38
6,25
8,91
6,33
6,03
7,43

Fold-change
under drought
treatment
135,3
2,4
-60,3

10,7
64,6
-9,8
337,5
8,3
2,7
-4,2

Fold-change
under salinity
treatment
3,3
2,2
-26,9
-2,6
2,3
-3,0
6,2
3,3
2,4
-4,1

Gene annotation
AtLEA4-1
Unknown function
Unknown function
Unknown function
bHLH
Cytochrome P450
Unknown function

bZIP11
IIS
Unknown function

Note: The red and blue colors indicated the up- and downregulated genes, respectively.
In this study, At4G33467 was recognized to be highest induced in the
drought condition by approximately 330-fold (Table 3.4). Another


10
interesting gene was At1G32560, encoding a member in the Late
embryogenesis abundant family, which was up-regulated by approximately
135-fold under the drought condition. Previously, these proteins were well characterized to involve in the water-limited responsiveness in A. thaliana
[37, 125]. At3G62090 encoding a member in the bHLH family was also
noted to induced by 64-fold and two-fold in drought and salinity conditions.
Recently, bHLH transcription factors were found to regulate the stress
response, especially the light responsiveness in the germination stage [132].
On the other hand, At3G55240 was strongest reduced in both drought and
salinity conditions by approximately -60 and -26-fold, respectively (Table
3.4). Interestingly, this gene has been the still unknown function, even
Ichikawa et al. (2006) reported the 'pseudo-etiolation in light' phenomenon in
A. thaliana [75].
3.1.5. Evaluation the physiological responses of transgenic Arabidopsis lines
overexpressing At3G55240 to abiotic stresses
In this study, At3G55240 which was highest down-regulated in both drought
and salinity conditions was selected as a candidate gene for further functional
characterization. In the normal condition, the transgenic line (namely RBC1) had
the lighter green color in leaves, significantly different as compared with the Col-0
control [75].


Day 1

Day 2

Day 3

Day 4

Day 5

Day 6

Day 7

Day 8

Figure 3.8. Evaluation of the physiological responses of transgenic
Arabidopsis line and control plants to salinity condition


11
Next, the transgenic line was treated in the medium containing 175 mM
NaCl. As the results, both of the RBC1 and control plants were sensitive with
the high salinity condition. From the 6th to 8th day, the survival rate of the
transgenic plants was significantly lowered than the controls (Figure 3.8).

Day 1

Day 2


Day 5

Day 6

Day 3

Day 7

Day 4

Day 8

Figure 3.9. Evaluation of the physiological responses of transgenic
Arabidopsis line and control plants to CdCl2 treatment
Under the CdCl2 750 µ treatment, leaves of the transgenic plants were
observed to transfer into the yellow. After 5 days of the treatment, the RBC1
line was obviously more sensitive than the control plants (figure 3.9). Thus,
our results indicated that overexpression of At3G55240 showed the sensitivity
to CdCl2 in A. thaliana.

Figure 3.10. Evaluation of the physiological responses of transgenic
Arabidopsis line and control plants to paraquat treatment


12
Under the paraquat treatment, overexpression of At3G55240 also exhibited the
sensitivity in leaves as compared with controls (Figure 3.10). The transgenic leaves
were found to lose the chlorophyll and be whitening faster than the control plants
after 24 h of paraquat treatment. These observations were also confirmed in all
paraquat treatments (Figure 3.10).

Taken together, overexpression of At3G55240 showed the sensitivity to the
abiotic stresses, including salinity, CdCl2 and paraquat treatments. In all
experiments, the survival rates of the transgenic line were significantly lower
than the control plants.
3.2. Identification and characterization of the methioinine-rich proteins in
soybean
3.2.1. Identification, annotation of the genes encoding methioinine-rich
proteins in soybean
According to the proteome of the soybean [149], a total of 213 MRPs has
been identified. All MRPs have the annotations, no gene encoding MRP was
found in the unplaced scaffolds. We also found that these MRP genes were
located on the chromosomes with an uneven ratio, whereas no gene was found
in the cytosol.
3.2.2. Functional classification of the MRPs in soybean
As the results, GmMRP genes were predicted to involve in various
biological processes in the cells. For instance, these processes could be metal
handling, RNA processing, stress response, protein modification, transcription
regulation, signaling, cell cycle, development, metal transportation and lipid
processing. Additionally, the approximately 43% MRP genes in soybean have
been still the unknown function. Thus, it would be very interesting to raise a
question that these genes may involve in the stress response in soybean plants
or not? Furthermore, the unknown-function MRP in soybean also could be
used as the raw materials for further functional characterization.
3.2.3. Expression patterns of the genes encoding MRPs in soybean
Based on the transcriptome published by Libault et al. (2010), 49 MRP
genes were under the limitation of the detection in nine major organs in
soybean plants in the normal condition [104]. The remaining MRP genes tend
to highly expressed in at least one major organ in the plant in the normal
condition.
Among them, Glyma13g03910, encoding MRP related to the signaling

pathway in the cell, was noted to specifically express in roots, shoot apical
meristems (SAM), and green pods. Two genes, Glyma10g29710 and


13
Glyma20g37600, encoding the proteins involved in the metal handling, also
showed the high accumulation in the SAM. Recently, one neighbor gene
Atriplex canescens was well-established to involve in the metal resistance and
abiotic stress(es) response in plant [158]. Glyma18g44300, encoding lipid
transporter, was specific in the root hairs and SAM, suggesting that lipid might
be accumulated in the root hairs in the early germination stage. Another example
was Glyma15g05510, highly expressed in flowers, while its close-relationship
gene, At1G25275 was known to involve in the light responsiveness in A.
thaliana [120]. Our analysis suggested that Glyma15g05510 might involve in the
flowering time in soybean via the light responsiveness.
Next, expression patterns of MRP genes were analyzed based on the public
microarray in the drought condition [96]. Our analysis showed that 11 and 12
genes were reduced and induced in leaves V6 in drought condition,
respectively. 24 and six genes were up- and down-regulated in leaves R2 in
drought condition, respectively. Among them, a total of 13 MRP genes were
responsive in both leaves V6 and R2 under the drought condition (Table 3.7).
Table 3.7. Expression patterns of MRP genes in leaves V6 and R2 in
soybean under the drought condition
#
1
2
3
4
5
6

7
8
9
10
11
12
13
1

Gene name
Glyma01g15910
Glyma01g15930
Glyma02g10620
Glyma03g32740
Glyma04g37040
Glyma06g39910
Glyma10g30380
Glyma15g05510
Glyma16g02510
Glyma19g43580
Glyma20g00780
Glyma20g22280
Glyma20g36730

Met

L

Annotation


8,08
6,56
7,22
6,04
7,91
10,34
7,43
7,37
7,26
6,70
6,69
6,59
7,89

100
458
98
481
140
117
149
96
125
210
285
426
153

Unknown function
UNE10

Unknown function
PIF1
Calmodulin 38
Calmodulin
Calmodulin 5
Unknown function
Calcium binding protein
GIF, GIF1, AN3
Homeodomain
PIF3
Calmodulin 5

Drought condition
LeavesV6
Leaves R2
3,63
4,96
-20,34
-3,87
-44,63
-4,04
-2,19
-2,02
15,03
40,08
3,12
4,14
7,50
5,27
2,93

2,41
2,05
4,63
-2,01
2,42
-3,03
-2,36
2,25
2,99
3,06
2,29

Met content (%),2Protein size (aa),3Public microarray database according to
[96]. The red and blue colors indicated the up- and down-regulated genes,
respectively.
Among 13 genes described in Table 3.7, Glyma04g37040 encoding
calmodulin-binding protein was highest expressed in leaves R2 and V6 by
approximately 40-fold and 15-fold. Glyma02g10620 encoding a 98-amino-cid-


14
residue protein was reduced in leaves V6 and R2 by approximately 44- and
four-fold (Table 3.7). Additionally, Glyma19g43580 was reduced in leaves V6
but induced in leaves R2 under the drought condition. At5G28640 in
Arabidopsis, a neighbor gene with Glyma19g43580, was recently
characterized to associate with the sugar homeostasis, and thus, involved in the
cell differentiation in the leaves [114, 115]. We hypothesized that the high
accumulation of Glyma19g43580 in leaves under the normal condition might
boost the sugar level in these tissues as a mechanism of the drought escape of
the soybean plants. It also very interestingly to raise a question of the functions

and roles of the MRP genes in the stress response in soybean plants.
3.3. In silico analysis of the genes encoding MSR in soybean
3.3.1. Genome-wide identification of the genes encoding MSR in soybean
As the results, a total of 7 genes encoding MSRA has been found in the
soybean genome. As compared with MSRAs in plant species, 5 genes were
found in A. thaliana [145], while 4 MSRA genes were identified in rice [65].
Our analysis indicated that MSRA genes are a multiple gene family in plant
species.
3.3.2. Gene duplication in the MSRA gene family in soybean
Based on the multiple alignments, three duplication events have been found
in the MSRA gene family in soybean. The identity of the duplicated genes
varies from 73.76% (GmMSRA4/A7) to 96.00% (GmMSRA2/A5), while the
pair of GmMSRA1/A6 shared the homology level of more than 90%.
Additionally, three duplicated pairs were distributed on the different
chromosomes in the soybean genome. For instance, GmMSRA1/A6 were
located on the Chr 2 and 16, GmMSRA2/A5 were mapped on the Chr2 and 14,
and GmMSRA4/A7 were located on Chr 8 and 18. Our results indicated that
these genes were occurred from the segmental duplication events in the
different chromosomes. Previously, Le et al. (2013) also reported two
segmental duplication events in the MSRB gene family in soybean. Particularly,
GmMSRB2/B5 and GmMSRB3/B4 were located on the Chr 13/15, respectively
[99]. Taken together, our results revealed that the expansion of MSR gene
family in soybean (including MSRA and MSRB) might be mostly explained by
the segmental duplication events.
Furthermore, the Ka/Ks ratios of three duplicated pairs were less than 1,
suggesting that the natural selection prevented the point mutations occurred in
the MSRA gene family in soybean as previously described [103].


15

3.3.3. Structural characterization of the MSRA family in soybean
The protein sizes of MSRAs were ranged from 194 (MSRA3) to 266 amino
acid residues (MSRA2 and MSRA5). The molecular weights of MSRAs vary
from 21.63 (MSRA3) to 29.87 kDa (MSRA2), while the pI was from 5.01 to
8.78, with the average pI was approximately 6.77. It should be remembered
that the pI of a protein could suggest its subcellular localization [14]. The pI
values of
1, - , - , -A7 were acidic, suggesting that these
proteins might be located on the cytosol. Interestingly, the pI values of
GmMSRA2 and GmMSRA5 were base, predicting that two proteins might be
distributed on the membranes of the organelles (Table 3.10).
Table 3.10. General characteristics of the MSRAs in soybean
Gene name
GmMSRA1
GmMSRA2
GmMSRA3
GmMSRA4
GmMSRA5
GmMSRA6
GmMSRA7

Gene identifier
Glyma02g05550
Glyma02g46020
Glyma04g36480
Glyma08g42790
Glyma14g02705
Glyma16g24130
Glyma18g11110


Size
250
266
194
203
266
250
203

mW
26.17
29.87
21.63
22.78
29.69
28.11
22.76

pI
6.14
8.78
5.01
6.52
8.74
6.39
5.84

SL
S*
C*

−
−
C*
S*
M

Note: S: Secretory pathway; C: Chloroplast; M: Mitochondrion. -: Unknown
place; *: Reliability.
Prediction by TargetP revealed that GmMSRA1 and GmMSRA6 might be
located on the secretory pathway, while GmMSRA2 and GmMSRA5 might be
distributed on the chloroplast. It should be noted that chloroplast is the
organelle that ROS was highly accumulated under the abiotic stress(es).
Therefore, GmMSRA2 and GmMSRA5 might involve in the reduction of the
MetO, and GmMSRA1 and GmMSRA6 might be transported into many
organelles to function on the repair of MetO in the cells.
Next, the phylogenetic tree was constructed based on the full-length protein
sequences of MSRAs in soybean and A. thaliana [145]. As the results, the
MSRA family in soybean could be classified into 2 groups (Figure 3.13).
Particularly, there are 5 members in group 1, including MSRA2, -A3, -A4, -A5
and -A7, while group 2 contains MSRA1 and MSRA6. Most of the members
in the group 1 should share similar structural characteristics. The sizes of
GmMSRA2 and GmMSRA5 were 266 amino acid residues with the mWs vary
from 29.69 ~ 29.87 kDa, and the pI values are also similar (7.84 ~7.87).


16

Figure 3.13. A phylogenetic tree of the MSRA in soybean and A. thaliana
The conserved domains of MSRAs in soybean could be classified into 2
groups as in the phylogenetic tree (Figure 3.13). All members in group 1

shared the Cys residue with the catalytic function and two Cys residues with
the resolving function. The conserved domain was F[G/A]AGCW[G/S][V/A]E
as previously confirmed in AtMSRA1, -A2, -A3 and -A4 in A. thaliana [145].
On the other hand, the conserved domain of group 2, including GmMSRA1, A6 and AtMSRA5 did not have the catalytic and resolving Cys residues.
3.4. Structural characterization of the MSRA gene family in soybean
3.4.1. Prediction of the cis-regulatory elements in the promoter regions of
MSRA gene family in soybean
Previously, five genes encoding MSRB were reported in the soybean genome
[99]. Thus, MSR genes, including 7 and 5 genes encoding MSRA and MSRB have
been genome-widely identified in soybean. To get insight into the gene function
and the regulation, the presence(s) of the CREs in the promoter regions of MSR
genes were analyzed by using the PlantCARE web-based tool [101].
Several typical core elements, such as TATA-box, CAAT-box were found
in the promoter sequences of all MSR genes. Generally, the CREs found in this
study could be classified into three functional categories, including the CREs
related to the light responsiveness, the CREs specific in the tissues and the
CREs related to the stress and/or hormone responsiveness.
3.4.2. Expression patterns of the MSR genes in soybean
In the normal condition [104], expression profiles of the most of MSR
genes were found, excluding GmMSRA5 had no information. The majority of
MSR genes were strongly and/or specifically expressed in at least one organ in


17
soybean plants. Among them, MSRB1 and MSRB3 were noted to highly
expressed in leaves, while MSRA3 were accumulated in the nodules. The high
accumulation of MSR in all major organs in soybean plants indicated that this
enzyme family may function on the repair of the protein oxidation.
Furthermore, MSRA3, -B1 and -B3 were highly expressed in leaves and
nodules which were predicted to accumulate the ROS during the adverse

environmental conditions (Figure 3.15).

Figure 3.15. Expression patterns of MSR genes during the development
84RH, 120RH: Root hairs havested after 84 and 120 hours of the
germination.; SAM: Shoot apical meristerm; F: Flower; GP: Green pods; N:
Nodules; L: Leaves; R: Roots; RT: Root tips; PKRM values were presented by
the heatmap
Under the drought condition [96], expression levels of all MSR genes were
significantly changed. For instance, five and one genes were induced and
reduced in leaves V6 and/or R2 under the drought treatment, respectively. In
the leaves V6, MSRB3 and MSRA3 were up- and down-regulated under the
drought condition, respectively. On the other hand, GmMSRA4, -A7, -B2 and B5, and GmMSRA3 were induced and reduced in the leaves R2 under the
drought condition, respectively (Figure 3.16A).
Under the salinity treatment [24], the majority of MSR genes, excluding
MSRA5, has the expression levels in the public RNA-seq data (Figure 3.16B).
Several MSR genes were significantly altered in roots under the salinity


18
condition. As compared with the controls, MSRA2 and MSRB1 were downregulated, whereas MSRA4, -B2 -B5 were noted to up-regulated in roots.

Figure 3.16. Expression patterns of the MSR genes under the (A) drought
and (B) salinity conditions in soybean
V6 and R2: Leave samples at V6 and R2 stages were treated in the drought
condition; R-Na1hr, R-Na6hr and R-Na10hr: Root samples were havested at 1,
6 and 10 hours after salinity treatment.
3.4.3. Experimental validation of the expression levels of MSRB in soybean
In the normal condition, MSRB1 was validated as the highest expressed
gene by the qPCR. For instance, MSRB1 was strongly expressed in leaves in
all stages in the normal condition (Figure 3.17). However, MSRB1 was not

clearly expressed in the remaining organs. Previously, MSRB1 was predicted
to distribute on the chloroplast [99]. Take together, our results revealed that
MSRB1 was specific in leaves and may function on the oxidative stress
response in the chroloplasts in leave tissues.

Figure 3.17. Validation of the expression patterns of MSRB genes in the
normal condition
Nine tissue samples, including leaves and roots of the seedlings, leaves V6 and
R2, seeds R5 and R7, roots and root tips R2 were havested to examine the


19
expression profiles of MSR genes by using qRT-PCR.
Additionally, MSRB5 was lowest expressed in all tissue samples. Our
previous analysis also revealed that not many stress-responsive CREs were
found in the promoter region of MSRB5 gene. Expression patterns of MSRB5
were also not significantly altered in nine major organs in the normal condition
[104] (Figure 3.15). Taken together, our data indicated that MSRB5 might play
the less important role as compared with other member of MSRB gene family
in soybean in the normal condition.

Figure 3.18. Validation of the expression patterns of MSRB genes in the
drought condition
(A): Validation of the expression profiles of MSRB genes in shoot samples under
the drought treatment after 0 - 2 - 10 hours. (B): Validation of the expression
profiles of the MSRB genes in root samples under the drought treatment after 0 - 2
- 10 hours. (C): Validation of the expression profiles of the MSRB genes in leaves
V6 under the drought and well-watered (control) conditions. (D): Validation of
the expression profiles of the MSRB genes in leaves R2 under the drought and
well-watered (control) conditions. (E): Validation of the expression profiles of the



20
MSRB genes in root tips under the drought treatment.
Under the dehydration treatment, MSRB2 and MSRB5 were validated to
up-regulate in leaves, whereas the remaining three MSR genes were not
significantly altered (Figure 3.18A). For instance, no MSR genes were found
to transcriptionally changed in roots under the dehydration treatment (Figure
3.18B). Interestingly, MSRB3 and two genes, MSRB2 and GmMSRB5 were
up-regulated in leaves V6 and R2 under the dehydration treatment,
respectively (Figure 3.18D), whereas no MSRB genes have been found to be
responsive in roots (Figure 3.18E). As our previous discussion, expression
profiles of the MSRB genes did not significantly change in roots in the
normal condition (Figure 3.15). Taken together, our data indicated that
MSRBs did not highly accumulate in root in the normal and dehydration
conditions.

Figure 3.19. Validation of the expression patterns of MSRB genes in the
salinity condition
Expression profiles of MSRB genes in shoots (A) and roots (B) under the salinity
condition after 0 - 2 - 10 hours. Shoots have been havested at 0 (S0s), 2 (S2s) and
10 hourse (S10s) after the treatment. Roots have been obtained at 0 (R0s), 2 (R2s)
and 10 hours (R10s) after the treatment
Our qRT-PCR confirmation showed that MSRB genes were up-regulated
after 2 h of the salinity treatment but seemed to reduced after 10 h of the
treatment. Among them, MSRB1 was noted to reduce in shoots after 10 h of
the treatment (Figure 3.19A). MSRB5 was validated to up-regulate after 2 h
of the treatment but down-regulate after 10 h of the treatment. Our data
indicated that the majority of the MSRB genes, excluding MSRB2 seemed to
up-regulated according to the duration of the treatment. MSRB5 was

significantly up-regulated in roots after 10 h of the salinity treatment (Figure


21
3.19B).

Figure 3.20. Validation of the expression patterns of MSRB genes in the
ABA condition
(A): Expression profiles of MSRB genes in shoots under the ABA treatment.
(B): Expression profiles of MSRB genes in rootss under the ABA treatment.
To address the question of how MSRB genes respond to the adverse
environmental conditions via the ABA-dependent and/or -independent
pathways, the samples were havested to treat with ABA and used for the qPCR.
As provided in Figure 3.20A, all MSRB genes were tend to up-regulate in
shoots, even no genes were responsive to the ABA. This phenomenon was also
recorded in the roots treated with ABA. Interestingly, MSRB2 was found to
highest induce in roots under the ABA treatment ( ình .20 ).
Table 3.11. Summary of the expression levels of MSRB genes in soybean in
the normal condition


22
Red color indicated the high-regulated genes.
Among them, MSRB1 was recently reported to locate on the chloroplast
[99]. Microarray analysis also revealed that MSRB1 was highly expressed in
leaves and tend to induced in green pods [104]. Our validation also confirmed
the high expression levels of MSRB1 in leaves in the normal condition.
Additionally, MSRB3, predicted to locate on the chloroplast [99], was strongly
expressed in leaves, roots and green pods in the normal condition.
Experimental validation confirmed that MSRB3 was up-regulated in leaves as

our prediction. This phenomenon was also recorded in MSRB4 as this gene
was highly expressed in green pods and leaves in the normal condition (Table
3.11). Two remaining MSRB, including MSRB2 and MSRB5 were not
specific in any major organs in soybean plants in normal condition. Taken
together, MSRB1, -B3 and -B4 were strongly expressed in leaves, suggesting
these members might function on the repair of MetO in leave tissues. Two
genes, MSRB2 and MSRB5 might play less important role as compared with
other members of MSRB genes.
Table 3.12. Summary of the expression profiles of MSRB genes in soybean
under the drought condition

Red and blue colors indicated the up- and down-regulated gene. ABRE:
Abscisic acid responsive element. [TC] n: TC-rich repeats.


23
Of our interest, our analyses also provided some reliable indication based
on the in silico and experimental analyses. The summary of the expression
profiles of MSRB genes in soybean under the drought condition was
represented in Table 3.12. For instance, MSRB3 was predicted and confirmed
to be responsive in leaves V6 under the drought condition (Figure 3.18C) but
not altered by ABA (Figure 3.20).
Table 3.13. Summary of the expression profiles of MSRB genes in soybean
under the salinity condition

Red and blue colors indicated the up- and down-regulated gene. ABRE:
Abscisic acid responsive element. [TC] n: TC-rich repeats.
Our results indicated that expression level of MSRB4 did not significantly
change in any conditions, even this gene seemed to highly express in green
pods and leaves in the normal condition. Thus, MSRB4 was not responsive in

the drought and salinity treatments. Our microarray analyses and qRT-PCR
validations strongly confirmed that MSRB genes did not significantly change
in roots under the adverse envinromental conditions. MSRB5 gene was not
noticeably expressed in the normal condition but was considered as the most
responsive gene in the stress condition(s). MSRB1 was the highest expressed
gene in the normal condition, but was reduced in the salinity treatment. In this
study, we provided a list of candidate MSR genes for further studies for the
improvement of the stress tolerance in soybean plants.