Journal of Bioscience and Bioengineering

VOL. 118 No. 4, 441e447, 2014

www.elsevier.com/locate/jbiosc

Stable accumulation of seed storage proteins containing vaccine peptides in

transgenic soybean seeds
Nobuyuki Maruyama,1, * Keigo Fujiwara,1 Kazunori Yokoyama,1 Cerrone Cabanos,1 Hisakazu Hasegawa,2
Kyoko Takagi,3, 4, x Keito Nishizawa,4, xx Yuriko Uki,1 Takeshi Kawarabayashi,5 Mikio Shouji,5
Masao Ishimoto,3, 4 and Teruhiko Terakawa2
Graduate School of Agriculture, Kyoto University, Gokasho, Uji, Kyoto 611-0011, Japan,1 Hokko Chemical Industry Co., Ltd., Atsugi, Kanagawa 243-0023, Japan,2 National Institute of
Agrobiological Science, Tsukuba, Ibaraki 305-8602, Japan,3 National Agricultural Research Center for Hokkaido Region, Sapporo, Hokkaido 062-8555, Japan,4 and Graduate School of
Medicine, Hirosaki University, Hirosaki, Aomori 036-8562, Japan5
Received 29 December 2013; accepted 6 April 2014
Available online 1 May 2014

There has been a signicant increase in the use of transgenic plants for the large-scale production of pharmaceuticals
and industrial proteins. Here, we report the stable accumulation of seed storage proteins containing disease vaccine
peptides in transgenic soybean seeds. To synthesize vaccine peptides in soybean seeds, we used seed storage proteins as a
carrier and a soybean breeding line lacking major seed storage proteins as a host. Vaccine peptides were inserted into the
exible disordered regions in the A1aB1b subunit three-dimensional structure. The A1aB1b subunit containing vaccine
peptides in the disordered regions were sorted to the protein storage vacuoles where vaccine peptides are partially
cleaved by proteases. In contrast, the endoplasmic reticulum (ER)-retention type of the A1aB1b subunit containing
vaccine peptides accumulated in compartments that originated from the ER as an intact pro-form. These results indicate
that the ER may be an organelle suitable for the stable accumulation of bioactive peptides using seed storage proteins as
carriers.
2014, The Society for Biotechnology, Japan. All rights reserved.
[Key words: Seed storage protein; Three dimensional structure; Endoplasmic reticulum; Vacuole; Vaccine]

In recent years, there has been a signicant increase in the use of

transgenic plants for the large-scale production of valuable proteins, including antibodies, vaccines, other pharmaceuticals, and
industrial proteins (1). The use of plants for this purpose offers
many advantages because of their low-price and safety compared
to mammalian culture cells. Previous studies have demonstrated
the ability of plants to produce complex proteins such as secretory
antibodies, which are composed of four different polypeptide
chains covalently linked by disulde bonds (2). Furthermore,
engineered proteins containing bioactive compounds may also be
produced in crops (3,4). Another advantage of using plant seeds is
the possibility of storing pharmaceutical proteins or peptides
without requiring a temperature-regulated storage room (5). Legumes, particularly soybeans, represent promising host systems for
such molecular farming because they produce more proteins
compared to other plants.
Soybeans seeds accumulate around 40% protein by dry weight,
which is mostly composed of 7S (b-conglycinin) and 11S globulins
(glycinin). Glycinin is composed of 5 subunits. These subunits have

* Corresponding author. Tel.: 81 774 38 3763; fax: 81 884 38 3761.

E-mail address: marunobu@kais.kyoto-u.ac.jp (N. Maruyama).
x
Present address: National Agriculture and Food Research Organization Agricultural Research Center, Tsukuba, Ibaraki 305-8666, Japan.
xx
Present address: NARO Institute of Crop Science, 2-1-18 Kannondai, Tsukuba,
Ibaraki 305-8518, Japan.

been classied into 2 groups according to their amino acid sequences (group I: A1aB1b, A2B1a and A1bB2; group II: A3B4 and
A5A4B3). The sequence identity is approximately 80% and 45%
within and between groups, respectively. The three-dimensional
structures of 11S globulins have been solved (6,7). The structure of
the monomer is characterized by a core domain consisting of 2
jelly-roll beta-barrels, and 2 extended helix domains. Highly exible regions (named disordered regions I through V) are interspersed along the monomer.
In seed cells, constituent subunits are synthesized as a single
polypeptide precursor, preproglycinin. The signal polypeptide is
removed co-translationally in the endoplasmic reticulum (ER) and
the resultant proglycinin assembles into trimers. As proglycinin is
sorted into protein storage vacuoles, a specic post-translational
cleavage occurs between asparagine and glycine residues, resulting
in a mature subunit consisting of acidic (approximately 30e35 kDa)
and basic (approximately 20 kDa) chains. These chains are linked by
a disulde bond (8). The cleavage then triggers the hexamer formation of the subunits (9).
The prevalence of Alzheimers disease, which is the most common form of dementia or loss of cognitive ability, has been steadily
increasing, and worsens as it progresses and eventually leads to
death. One of the leading theories suggests that the deposition of
amyloid-b and subsequent formation of amyloid plaques in
neuronal cells trigger this neurodegenerative disease. An epitope
vaccine, PheeArgeHiseAspeSereGlyeTyr (FRHDSGY), has been

1389-1723/$ e see front matter 2014, The Society for Biotechnology, Japan. All rights reserved.
http://dx.doi.org/10.1016/j.jbiosc.2014.04.004

442

MARUYAMA ET AL.

J. BIOSCI. BIOENG.,

(kDa)

(kDa)

116.0

116.0
66.2

Pro

66.2

Pro

45.0

45.0
Acidic

35.0

Acidic

35.0
25.0

25.0
Basic
18.4

18.4

14.4

14.4

Basic

FIG. 1. SDS-PAGE and western blotting of total proteins in the control and transgenic soybean seeds accumulating A1aB1b containing vaccine peptides. (A) SDS-PAGE analysis of total
protein from transgenic soybean seeds accumulating the wild type or the mutants. (B) Soybean seed extracts were analyzed by western blotting against the anti-A1aB1b antibody.
Lane 1, Jack; 2, JQ; 3, A1aB1bWT; 4, A1aB1bM2; 5, A1aB1bM3; 6, A1aB1bM4. Pro, Acidic, and Basic indicate the pro-forms, acidic chain and basic chain of A1aB1b, respectively.

developed by McLaurin and others (10) against the amyloid protein

and has been shown to have therapeutic effects on transgenic mice
with Alzheimers disease. In this study, we aimed to develop a
transgenic soybean that accumulates vaccine peptides using Alzheimers disease vaccine peptides as a model peptide and A1aB1b,
the glycinin subunit, as its carrier. Vaccine peptide sequences were
inserted into the disordered regions of A1aB1b and their accumulation in transgenic seeds was examined. We also investigated the
accumulation of an ER-retention type of A1aB1b containing vaccine
peptides.
MATERIALS AND METHODS
Construction and introduction of transgenes into soybean
The cDNA for
A1aB1b containing vaccine sequences was prepared following the procedure
described by Nishizawa et al. (11) and Prak and Utsumi (12). The seed specic
glycinin promoter was used for all constructs. Three mutants were structured;
specically, A1aB1bM2, A1aB1bM3, and A1aB1bM4, which contained an FRHDSGY
insertion in disordered regions II, III, and IV, respectively. The vaccine peptide was
not introduced into region I because of the need to retain the correct cleavage of
the A1aB1b signal peptide. Moreover, the wild-type A1aB1b cassette was
constructed as the control (A1aB1bWT). The ER retention types of A1aB1bWT and
A1aB1bM3 were constructed by adding the ER-retention sequence
(HiseAspeGlueLeu) at the C-terminus (A1aB1bWT-HDEL and A1aB1bM3-HDEL).
The resulting plasmids were then transformed into soybean immature embryos
using the whisker-supersonic method (13) or the bombardment method (14).
Antibodies
Antibodies against A1aB1b, vaccine peptides (antibodies against
FRHDSGY peptide), and BiP derived from rabbit were used. Anti-rabbit IgG conjugated with alkaline phosphatase was used as the secondary antibody.
Extraction of total protein
Portions of mature dry seeds were scraped with
a scalpel and crushed using a Multi-beads Shocker (MB501S, Yasui Kikai, Osaka,
Japan). The soybean powder was then defatted in hexane for 10 min and centrifuged
at 3000 g for 5 min. Defatting was performed 4 times and was followed by airdrying. Twenty microliter of 1 SDS buffer [62.5 mM TriseHCl, pH 6.8, 10% (v/v)
glycerol, 5% (v/v) 2-mercaptoethanol, 2.5% (w/v) SDS] was then added to 1 mg seed
powder and vortexed for at least 1 h at room temperature. The mixture was
centrifuged at a minimum speed of 12,000 g for 15 min at 4 C, and the supernatant
was transferred into a new tube.
Extraction of soluble and insoluble fractions
Defatted seed powder
(100 mg) was added to buffer A [35 mM sodium phosphate, pH 7.4, 0.4 M NaCl, 1 mM
EDTA, 0.1 mM (p-amidinophenyl)-methanesulfonyl uoride, 1.2 mM leupeptin,
0.2 mM pepstatin A, 0.02% (w/v) NaN3]. The mixture was vortexed and mixed with
constant stirring at room temperature for 1 h and then centrifuged at a minimum
speed of 12,000 g for 15 min at 4 C. Then, the supernatant was transferred into a
new tube as the soluble fraction. The insoluble fraction was obtained by adding 1
SDS buffer to the remaining pellet, followed by vortexing, mixing and centrifugation.

SDS-PAGE and western blotting

SDS-PAGE was performed using 11%
polyacrylamide gel. Proteins were stained with Coomassie Brilliant Blue R-250.
Western blotting was performed after SDS-PAGE using 11% polyacrylamide gel according to the procedure described by Laemmli (15). The separated proteins on gels
were electrophoretically transferred to a nitrocellulose membrane (0.45 mm;
Schleicher and Schuell Inc., Dassel, Germany), and recombinant proteins were
detected with appropriate rabbit-derived anti-sera followed by goat anti-rabbit
IgGealkaline phosphatase conjugate (Promega, Madison, WI, USA).
Gel-ltration chromatography
The soluble protein fraction was applied on
to a Hi-Prep 16/60 Sephacryl S-300 HR column (GE Healthcare Life Sciences). The
ow rate was set to 0.5 ml/min using buffer A. To check for purity, the fractions were
analyzed by SDS-PAGE using 11% polyacrylamide gels. Protein concentration was
determined using a Protein Assay Rapid Kit (Wako, Osaka, Japan) with bovine serum
albumin as the standard.
Protein analysis
Protein fractions containing A1aB1bWT or A1aB1b-vaccine
peptides were initially concentrated using a trichloroacetic acid precipitation and
quantied using the 2-D Quant kit (GE Healthcare). Proteins were then blotted on a
PVDF membrane and subjected to N-terminal amino acid sequencing using a Procise
492 Protein Sequencer (Applied Biosystems).
A1aB1bM3 fragments were analyzed by mass spectrometry. The bands corresponding to A1aB1bM3 acidic chains on SDS-PAGE were digested by trypsin or V8
protease. MALDI TOF MS data was collected by using the AXIMA Performance
(Shimadzu Corporation, Kyoto, Japan).
Transmission electron microscopy
Dry soybean seeds were cut into 1.0 mm
sections and xed for 2 h in 4% (v/v) formaldehyde, and 0.05% (v/v) glutaraldehyde
solution at 4 C. Tissue sections were washed 4 times with buffer (100 mM sodium
phosphate, pH 7.2), dehydrated in a graded ethanol series and embedded in LR
White resin (London Resin, Basingstoke, UK). Ultrathin sections (70 nm) were cut
with a glass knife and placed on formvar/carbon-coated grids. The sections were
blocked with 1% (w/v) BSAePBS and then incubated for 1 h at room temperature
with anti-A1aB1b, anti-vaccine peptide, or anti-BiP antibody in 1% (w/v) BSAePBS.
The sections were washed with 1% (w/v) BSAePBS and then incubated with goat
anti-rabbit IgG conjugated to 15 nm gold particles (H L, Auro Probe EM, Amersham) in 1% (w/v) BSAePBS at room temperature. After washing with PBS, the
sections were washed twice with distilled water, stained for 5 min with 4% (w/v)
uranyl acetate, and incubated with 80 mM lead nitrate for 5 min. The grids were
examined and photographed using a transmission electron microscope (model H7100, Hitachi, Tokyo).

RESULTS
Carrier protein accumulation in the soybean line lacking
major seed storage proteins In this study, we used a soybean
breeding line, JQ, which lacked both glycinin and b-conglycinin, and
had high transformation efciency (16). The JQ line is obtained by
crossing the soybean breeding line, QF2, which lacks glycinin and
b-conglycinin, with Jack, which is a highly competent variety for

VOL. 118, 2014

A
B

STABLE ACCUMULATION OF MODIFIED SEED STORAGE PROTEINS

seeds was approximately 50 mg per 1 g seed (w11.3% for total

seed proteins) among several independent transgenic lines. The
expression of A1aB1b did not cause unfavorable effects for
soybean traits, such as seed appearance or young plant growth.

1
4
10
42
DAEFRHDSGYEVGVVIA
I

II

III

IV

A1aB1b
Repeat
number of
peptides

A1aB1bWT

A1aB1bM2

A1aB1bM3

A1aB1bM4

A1aB1bWT-HDEL

0
3

A1aB1bM3-HDEL

FRHDSGYFRHDSGYFRHDSGY

His-Asp-Glu-Leu (ER retention sequence)

A1aB1b
WT or Mutants

Gy1 Promoter

Gy1 Terminator

FIG. 2. Schematic illustration of the constructs for A1aB1b containing vaccine peptides.
(A) A partial sequence of amyloid b protein. Underlined sequences indicate the vaccine
peptides. (B) Schematic structure of the wild type and mutants of A1aB1b. Numbers in
the upper and lower parts of A1aB1b indicate the disordered regions and residue
numbers, respectively. (C) Schematic structure of the construct used in the study.

transformation, embryogenesis, and regeneration (Fig. 1, lanes 1

and 2). We conrmed that a protein extract from the seeds of JQ
did not cross-react with an antibody against A1aB1b (Fig. 1, lane
2). First, as a control line, we prepared homozygous lines of
transgenic soybean accumulating wild-type A1aB1b in JQ (Fig. 1,
lane 3; A1aB1bWT). The post-translationally processed mature
form of A1aB1b, which was composed of acidic and basic chains,
was predominantly detected in the transgenic soybean seeds that
accumulated A1aB1bWT. A small amount of the unprocessed proform of A1aB1b was also observed in the A1aB1bWT line. The
maximum accumulation level of A1aB1b in transgenic soybean

1 2 3 4

443

Total protein proles of A1aB1b containing vaccine peptides

in transgenic soybean seeds We used A1aB1b as a carrier to
accumulate the vaccine peptides in soybean seeds. A1aB1b has 5
disordered regions in its three-dimensional structure (6). We
selected the disordered regions of A1aB1b as the site for
introducing of the vaccine peptides, because these disordered
regions in the three-dimensional structure are almost identical to
the sequence-variable regions among the 11S globulin family (7).
We designated A1aB1b, which contains the vaccine peptides in
disordered regions II, III, and IV, as A1aB1bM2, A1aB1bM3 and
A1aB1bM4, respectively (Fig. 2).
The maximum accumulation levels of A1aB1b containing the
vaccine peptides (A1ab1bM2, A1ab1bM3 and A1aB1bM4) in the
transgenic seeds were comparable to or higher than that of
A1aB1bWT. As described in the above section, A1aB1bWT mainly
accumulated as a mature form. A1ab1bM2 and A1ab1bM3 also
accumulated as mature forms of A1aB1b containing the vaccine
peptides (Fig. 1, lanes 4 and 5). The mobility of acidic chains on SDSPAGE was slightly slower compared to that of A1aB1bWT because
of the insertion of the FRHDSGY sequence. In addition to these
bands, A1aB1bM2 and A1ab1bM3 had some partially cleaved
products, possibly of the acidic chain, which were not found in
A1aB1bWT. A1aB1bM4 also exhibited the mature and pro-forms of
the protein. A larger amount of pro-form was present in A1aB1bM4
compared to A1aB1bM2 and A1aB1bM3 (Fig. 1, lane 6).
Solubility and self-assembly of A1aB1b containing vaccine
peptides in transgenic soybean seeds
To evaluate the solubility of A1aB1b containing vaccine peptides, we separated the total
proteins of seeds into soluble and insoluble fractions and subjected

Pro
Acidic

A1aB1bM2

Basic

Pro

A1aB1bM3

Abs 280

Acidic
Basic

Pro

A1aB1bM4

Acidic

Basic

40

80 120 160
Elution time (min)

200

240
Elution time (min)

FIG. 3. Self-assembly of A1aB1b containing vaccine peptides. (A) Gel-ltration elution prole of the soluble fractions using the Hi-Prep 16/60 Sephacryl S-300 HR column. Numbers
indicates the positions of the standards: 1, Blue dextran (2000 kDa); 2, thyroglobulin (669 kDa); 3, ferritin (440 kDa); 4, aldorase (158 kDa); 5, conalbumin (75 kDa). Arrows indicate
the peaks corresponding to A1aB1b containing vaccine peptides. (B) Western blotting against the anti-A1aB1b antibody of fractions by gel-ltration chromatography. Pro, Acidic and
Basic indicate the pro-forms, acidic chain and basic chain of A1aB1b, respectively.

444

MARUYAMA ET AL.

J. BIOSCI. BIOENG.,

them to western blot analysis using anti-A1aB1b antibody (Fig. S1).

Most of the mature forms of A1aB1bWT were soluble. The mature
forms of the A1aB1bM2, A1aB1bM3 and A1aB1bM4 lines were
detected in the soluble fractions, even though their pro-forms
were detected in the insoluble fractions. A signicant amount of
pro-form was mainly found in the insoluble fraction, especially, in
A1aB1bM4.
Furthermore, we examined the self-assembly of A1aB1b containing the vaccine peptides in the soluble fractions using gelltration chromatography and compared it with that of A1aB1bWT
(Fig. 3). Peaks corresponding to A1aB1bWT were eluted at
approximately 110 min. The fractions for these peaks corresponded
to the size of the hexamer form of A1aB1b (approximate molecular
weight of 320 kDa) when compared with the protein molecular
weight standards (data not shown). In contrast, the mature forms of
A1aB1bM2, A1ab1bM3 and A1aB1bM4 in the soluble fractions also
had elution times of approximately 110 min, indicating that these
mature proteins formed the hexameric structure. These results
indicate that the addition of the vaccine peptides into the disordered regions of A1aB1b did not inuence the self-assembly of the
mature forms. A part of the pro-form of mutants was detected in
the fractions near void volumes, indicating that the unprocessed
form of A1aB1b containing the vaccine peptides formed a soluble
aggregate.
Analysis of the post-translational modication of A1aB1b
containing vaccine peptides Although the mature form of
A1aB1bWT was mainly represented by acidic and basic chains,
some extra bands were observed in all mutants (Fig. 1). By Nterminal protein sequencing of the fractions containing
A1aB1bWT on gel-ltration chromatography, we conrmed that 2
major bands of A1aB1bWT (Fig. S2, band nos. 1 and 2) matched
with those of the acidic and basic chains. In contrast, band no. 3
of A1aB1bM2, the size of which was close to that of the acidic
chain of A1aB1bWT, corresponded to the N-terminal sequence of
the acidic chain. Band no. 4 of A1aB1bM2 was positioned
between the acidic and basic chains, and was characterized by
the presence of heterozygous N-terminal sequences, indicating
that it may not solely consist of specic cleaved products of the
acidic chain. Band no. 5 of A1aB1bM2 corresponded to the Nterminal sequence of the acidic chain. These results indicate that
the cleavage of the acidic chain of A1aB1bM2 into 2 bands might
be caused by the introduction of the vaccine peptides into this
disordered region. The band no. 8 of A1aB1bM3 also indicates the
heterozygous sequences similar to the band no. 4 of A1aB1bM2.
Although the molecular weight of an acidic chain of A1aB1bM4
predicted from the primary structure is larger than that of
A1aB1bWT, the mobility of the band corresponding to the acidic
chain of A1aB1bM4 was almost similar to that of A1aB1bWT on
SDS-PAGE (Fig. S2). Further, we conrmed the N-terminal
sequence of the band no. 9 was identical to that of A1aB1bM4,
meaning the acidic chain of A1aB1bM4 was cleaved near its Cterminus. Therefore, we predicted that the cleavage of A1aB1bM4
occurred within the bioactive peptide sequence. These results
indicate that the vaccine peptide sequences appeared to favor an
additional cleavage in transgenic soybean seeds. This
phenomenon was further corroborated by the western blot
analysis of the same samples using an anti-vaccine peptide (data
not shown), wherein the band pattern was similar to the western
blot analysis using anti-A1aB1b antibody. This result indicated
that the vaccine peptides remained attached to the acidic chain of
A1aB1b and were not completely degraded. Furthermore, we
prepared the A1aB1bM3 fragments from the seed extracts of
transgenic soybean seeds by afnity purication and examined
the fragments of the A1aB1bM3 acidic chains (band nos. 7e1 and
7e2 in Fig. S2). In the mass spectrum of the A1aB1bM3 acidic

FIG. 4. Transmission microscopy of transgenic soybean seeds of A1aB1b containing the

vaccine peptide. Sections prepared from mature seeds of A1aB1bWT (A), A1aB1bM2
(B) and A1aB1bM3 (C) were treated with the anti-A1aB1b antibody. PSV, protein
storage vacuole; OB, oil body. Scale bar: 1.0 mm.

chains treated with trypsin or V8 protease, fragments containing

the vaccine peptides (fragments 1 and 2 in Fig. S3) were
observed. Therefore, we concluded that the vaccine peptides
remained in the cleaved fragments of the A1aB1bM3 acidic
chains, even though A1aB1bM3 was cleaved at the position
where the vaccine peptides were introduced in transgenic
soybean seeds.
Localization of A1aB1b containing vaccine peptides in
transgenic soybean seeds
Endogenous 11S globulins are
mainly sorted into the protein storage vacuoles of soybean seeds
(17). We assessed the effect of introducing the vaccine peptide on
the localization of A1aB1b by immuno-electron microscopy. As
expected, A1aB1bWT in transgenic soybean seeds was sorted into
protein storage vacuoles, which was indicated by the deposition
of gold particles conjugated with the anti-A1aB1b antibody

VOL. 118, 2014

STABLE ACCUMULATION OF MODIFIED SEED STORAGE PROTEINS

445

B
(kDa)

(kDa)
94.6

94.6

71.3

71.3

45.1

45.1

32.2

32.2

25.8

25.8

17.2

17.2

FIG. 5. SDS-PAGE and western blotting of total proteins of transgenic soybean seeds
accumulating the modied A1aB1b with ER-retention sequence. (A) SDS-PAGE analysis
of total protein from transgenic soybean seeds. (B) Soybean seed extracts were
analyzed by western blotting against the anti-A1aB1b antibody. Lane 1, JQ; 2,
A1aB1bWT-HDEL; 3, A1aB1bM3-HDEL.

(Fig. 4). Similar to A1aB1bWT, A1aB1bM2 and A1aB1bM3 were also

mainly deposited into the protein storage vacuoles (Fig. 4).
However, characteristic compartments were observed in
transgenic soybean seeds accumulating A1aB1bM4. The antiA1aB1b antibody recognized the characteristic compartments in
addition to the protein storage vacuoles (Fig. S4), resulting in
A1aB1bM4 accumulating in both the protein storage vacuole and
the characteristic structures. Immuno-electron microscopy with
an antibody against vaccine peptides produced similar results
(Fig. S4). We examined the characteristic structures in transgenic
soybean seeds of A1aB1bM4 by using an antibody against BiP,
which is an ER chaperone protein. The characteristic
compartments were clearly detected by anti-BiP antibody
(Fig. S4). Moreover, we observed a ribosome-like structure at the
periphery of the characteristic structures. These results indicate
that the characteristic structures accumulating A1aB1bM4
originated from the ER.
Characteristic compartments formation by ER-retention
type of A1aB1b and prevention of the vaccine peptide
degradation
We examined the effect of adding the known ERretention sequence HiseAspeGlueLeu to the C-terminus of A1aB1b
(A1aB1bWT-HDEL in Fig. 2) using a soybean breeding line, JQ, on
the accumulation behavior of A1aB1b containing the vaccine
peptides in transgenic soybean seeds. A1aB1bWT-HDEL mainly
accumulated as a pro-form (Fig. 5). We examined its solubility
and self-assembly state. In a sequential extraction experiment, we
found that A1aB1bWT-HDEL was present as a soluble form
(Fig. S5). Using the same gel-ltration chromatography procedure
as
for
A1aB1bM2eM4,
A1aB1bWT-HDEL
was
eluted
approximately 10 min later (at about 120 min) compared to the
wild-type (at about 110 min), indicating that it forms as a trimer
(of only approximately 160 kDa). Immuno-electron microscopy
showed that A1aB1bWT-HDEL was deposited into the
compartments that originated from the ER (Fig. 6A), which are
morphologically similar to those observed in transgenic soybean
seeds of A1aB1bM4. Furthermore, we prepared a transgenic

FIG. 6. Transmission microscopy of transgenic soybean seeds of the modied A1aB1b

with ER-retention sequence. Sections were prepared from mature seeds of A1aB1bWTHDEL (A) and A1aB1bM3-HDEL (B). The sections were treated with anti-BiP (A) or antiA1aB1b antibody (B). PSV, protein storage vacuole; EB, ER body (the compartments
originated from the ER); OB, oil body. Scale bars: 1.0 mm (A and B).

soybean expressing A1aB1bM3 with an ER-retention signal

sequence (A1aB1bM3-HDEL in Fig. 2B). A1aB1bM3-HDEL also
mainly accumulated as a soluble pro-form, similar to A1aB1bWTHDEL. The compartments originating from the ER were also
observed by electron microscopy (Fig. 6). These results indicate
that the cleavage of the A1aB1bM3-HDEL vaccine peptides could
be prevented by retaining them in the ER, and that the retention
of seed storage proteins in the ER induces the formation of the
characteristic structures.
DISCUSSION
Soybean seed storage protein decient lines are suitable
hosts for pharmaceutical protein production
Plant seeds are
a promising source for the production of pharmaceutical proteins,
with the advantages of low-cost, ease of shipping and safety. Soybean seeds have high protein content, representing approximately
40% dry weight. Therefore, soybean is an attractive source for
producing pharmaceutical proteins in plant seeds. In previous
studies, expression levels of human coagulation factor IX and human growth hormone in transgenic soybean seeds of a normal
variety represented up to 0.23% and 2.9% of total soluble seed
protein content, respectively (18,19). To develop an oral vaccine, the
heat labile toxin of enterotoxigenic Escherichia coli, which
accumulates up to 2.4% of the total soluble seed protein content
in transgenic soybean seeds, has been used (20). b-Conglycinin
contains bioactive peptides, and is composed of up to 0.2% of the
protein extract of transgenic soybean seeds, when using a
standard variety (Jack) (11). In a study using a model reporter

446

MARUYAMA ET AL.

J. BIOSCI. BIOENG.,

protein, a green uorescent protein fused with the ER-retention

sequence accumulated up to >7% in transgenic soybean seeds
lacking b-conglycinin a and a0 subunits (21).
Recently, it has been reported that rice lines lacking seed storage
protein accumulate higher amounts of foreign proteins (22,23). In
this study, we used a line lacking all subunits of glycinin and bconglycinin, but which also had high transformation efciency (JQ).
A1aB1bWT accumulated a maximum of about 11% of total soluble
proteins in the transgenic soybean seeds of JQ. This level was
remarkably higher compared to those obtained in previous studies.
Furthermore, we conrmed that the A1aB1b containing the vaccine
peptides accumulated similar or higher levels of total soluble proteins compared to that of A1aB1bWT. Therefore, we anticipate that
the soybean line lacking major seed storage proteins could be
widely used for pharmaceutical protein production.

that the expression in A1aB1bM4 induces the formation of the

compartments originating from the ER, and that the insoluble
aggregates trigger its formation. Soluble A1aB1bWT-HDEL and
A1aB1bM3-HDEL also induced the formation of the
compartments that originated from the ER, in a manner similar to
A1aB1bM4. The A1aB1b containing the vaccine peptides in
A1aB1bM3-HDEL was not cleaved in transgenic soybean seeds at
the sequences of the vaccine peptides. Proteases cleaving the
vaccine peptides are assumed to be located and exert their
activity in the protein storage vacuoles. Therefore, the ER may
represent a promising organelle for the stable accumulation of
vaccine peptides using seed storage protein as a carrier.
Supplementary data to this article can be found online at http://
dx.doi.org/10.1016/j.jbiosc.2014.04.004.

Cleavage of the vaccine peptides in transgenic soybean

seeds
A1aB1b containing the vaccine peptides (A1aB1bM2-M4)
was cleaved in transgenic soybean seeds. Cleavage occurred at the
sites of the insertion of the peptides in addition to a site in
A1aB1bWT. Wild-type glycinin is processed at the sequences between AsneGly of the disordered regions IV by the vacuolar processing enzyme (VPE), a cysteine protease. VPE puried from
soybean seeds exhibits specicity for Asn at the N-terminal side of
the cleavage sequences but little specicity on the type of amino
acid residues at the C-terminal side. Recently, it has been reported
that VPE cleaves at the C-terminal of Asp and Asn (24,25). In
Arabidopsis, 3 VPEs (aVPE, bVPE, and gVPE) play a role in the
processing of storage proteins (12S globulin and 2S albumin),
with bVPE being the most essential (26). Experiments with
Arabidopsis mutant lines of the VPEs indicate that aspartic
proteinases (APs) are also involved in the processing of seed
storage proteins. Two genes for soybean APs are known (27,28).
AP-1 is highly expressed in soybean seed tissues. It has been
reported that APs in Arabidopsis seeds cleave at the sequences of
PheeGlu and PheeAsp and that Arabidopsis AP A1 cleaves with
the oxidized insulin B-chain, which is a model substrate, at the
sequences of LeueTyr, PheePhe, and PheeTyr (29). Based on
previous reports (29,30), we infer that plant APs typically cleave
sequences between 2 residues with hydrophobic chains (Leu, Ile,
Val, and Phe) or sequences adjacent to a hydrophobic residue.
The vaccine peptide sequence is FRHDSGY and a tandem triple
repeat of this sequence was used in this study. VPE probably does
not efciently cleave the vaccine peptides, because of the lack of an
Asn residue. However, there is a possibility that VPE cleaves at the
sites between Asp and Ser, given that the VPE of some plant species
cleave at the C-terminal of Asp. In contrast, the sequence of TyrePhe in the vaccine peptides may be the site of cleavage by APs,
because APs cleave sequences between 2 residues with hydrophobic chains. In the mass spectrometry experiments, we identied
fragments indicating several C-terminal sequences. These results
indicate that APs are involved in the cleavages of A1aB1b containing vaccine peptides and that proteases other than APs and VPEs
also contribute to these cleavages.

ACKNOWLEDGMENTS

ER may be a promising organelle for the stable

accumulation of vaccine peptides in seed cells
The mature
and pro-forms of A1aB1b containing vaccine peptides were
detected in transgenic soybean seeds (Fig. 1). The pro-forms of
A1aB1bM2, A1aB1bM3 and A1aB1bM4 were detected in insoluble
fractions, with more insoluble pro-form occurring in A1aB1bM4
compared to A1aB1bM2 and A1aB1bM3 (Fig. S1). Using
transmission electron microscopy, it was observed that
A1aB1bM4 accumulated in both the compartments that
originated from the ER and in the protein storage vacuoles,
whereas A1aB1bM2 and A1aB1bM3 mainly accumulated in the
protein storage vacuoles (Fig. 4 and Fig. S4). These results indicate

This study was supported in part by grants from the Development of Fundamental Technologies for Production of High-Value
Materials Using Transgenic Plants project from the Ministry of
Economy, Trade, and Industry (to N.M., M.I., and T.T.) and from
JSPS KAKENHI Grant number 24658286 (to N.M.). We thank Prof.
Shigeru Utsumi (Kyoto University) for contributing to the design of
this study and Prof. Reiko Urade (Kyoto University) for the positive
encouragement. We also thank Ms. M. Sawada (Kyoto University),
Ms. E. Okuda (Kyoto University), Ms. M. Tokuda (Hokko Chemical
Industry) and Ms. M. Yamashita (Hokko Chemical Industry) for
technical assistance. We thank the Development and Assessment of
Sustainable Humanosphere (DASH) system and Prof. Kazufumi
Yazaki of Kyoto University for supporting the development of the
transgenic crops.

References
1. Daniell, H., Singh, N. D., Mason, H., and Streateld, S. J.: Plant-made vaccine
antigens and biopharmaceuticals, Trends Plant Sci., 14, 669e679 (2009).
2. Orzez, D., Granell, A., and Blzquez, M. A.: Manufacturing antibodies in the
plant cell, Biotechnol. J., 4, 1712e1724 (2009).
3. Maruyama, N., Mikami, B., and Utsumi, S.: The development of transgenic
crops to improve human health by advanced utilization of seed storage proteins, Biosci. Biotechnol. Biochem., 75, 823e828 (2011).
4. Wakasa, Y., Takagi, H., Hirose, S., Yang, L., Saeki, M., Nishimura, T.,
Kaminuma, O., Hiroi, T., and Takaiwa, F.: Oral immunotherapy with transgenic rice seed containing destructed Japanese cedar pollen allergens, Cry j 1
and Cry j 2, against Japanese cedarpollinosis, Plant Biotechnol. J., 11, 66e76
(2013).
5. Khan, I., Twyman, R. M., Arcalis, E., and Stoger, E.: Using storage organelles
for the accumulation and encapsulation of recombinant proteins, Biotechnol. J.,
7, 1099e1108 (2012).
6. Adachi, M., Takenaka, Y., Gidamis, A. B., Mikami, B., and Utsumi, S.: Crystal
structure of soybean proglycinin A1aB1b homotrimer, J. Mol. Biol., 305,
291e305 (2001).
7. Tandang-Silvas, M. R., Fukuda, T., Fukuda, C., Prak, K., Cabanos, C.,
Kimura, A., Itoh, T., Mikami, B., Utsumi, S., and Maruyama, N.: Conservation
and divergence on plant seed 11S globulins based on crystal structures, Biochim. Biophys. Acta, 1804, 1432e1442 (2010).
8. Staswick, P. E., Hermodson, M. A., and Nielsen, N. C.: Identication of the
cystines which link the acidic and basic components of the glycinin subunits,
J. Biol. Chem., 259, 13431e13435 (1984).
9. Jung, R., Scott, M. P., Nam, Y. W., Beaman, T. W., Bassuner, R., Saalbach, I.,
Muntz, K., and Nielsen, N. C.: The role of proteolysis in the processing and
assembly of 11S seed globulins, Plant Cell, 10, 343e357 (1998).
10. McLaurin, J., Cecal, R., Kierstead, M. E., Tian, X., Phinney, A. L., Manea, M.,
French, J. E., Lambermon, M. H., Darabie, A. A., Brown, M. E., and other 8
authors: Therapeutically effective antibodies against amyloid-beta peptide
target amyloid-beta residues 4-10 and inhibit cytotoxicity and brillogenesis,
Nat. Med., 8, 1263e1269 (2002).
11. Nishizawa, K., Kita, A., Doi, C., Yamada, Y., Ohinata, K., Yoshikawa, M., and
Ishimoto, M.: Accumulation of the bioactive peptides, novokinin, LPYPR and
rubiscolin, in seeds of genetically modied soybean, Biosci. Biotechnol. Biochem., 72, 3301e3305 (2008).

VOL. 118, 2014

STABLE ACCUMULATION OF MODIFIED SEED STORAGE PROTEINS

12. Prak, K. and Utsumi, S.: Production of a bioactive peptide (IIAEK) in Escherichia
coli using soybean proglycinin A1ab1b as a carrier, J. Agric. Food Chem., 57,
3792e3799 (2009).
13. Terakawa, T., Hasegawa, H., and Yamaguchi, M.: Efcient whisker-mediated
gene transformation in a combination with supersonic treatment, Breed. Sci.,
55, 465e468 (2005).
14. Khalafalla, M. M., Mizanur, R. S., Nakamoto, Y., Wakasa, K., and
Ishimoto, M.: Optimization of particle bombardment condition by monitoring
of transient sGFP(S65T) expression in transformed soybean, Breed. Sci., 5,
257e263 (2005).
15. Laemmli, U. K.: Cleavage of structural proteins during the assembly of the
head of bacteriophage T4, Nature, 227, 680e685 (1970).
16. Kita, Y., Nishizawa, K., Takahashi, M., Kitayama, M., and Ishimoto, M.: Genetic improvement of the somatic embryogenesis and regeneration in soybean
and transformation of the improved breeding lines, Plant Cell Rep., 26,
439e447 (2007).
17. Maruyama, N., Mun, L. C., Tatsuhara, M., Sawada, M., Ishimoto, M., and
Utsumi, S.: Multiple vacuolar sorting determinants exist in soybean 11S
globulin, Plant Cell, 18, 1253e1273 (2006).
18. Cunha, N. B., Murad, A. M., Cipriano, T. M., Arajo, A. C., Arago, A. J.,
Leite, A., Vianna, G. R., McPhee, T. R., Souza, G. H., Waters, M. J., and
Rech, E. L.: Expression of functional recombinant human growth hormone in
transgenic soybean seeds, Transgenic Res., 20, 811e826 (2011).
19. Cunha, N. B., Murad, A. M., Ramos, G. L., Maranho, A. Q., Brgido, M. M.,
Arajo, A. C., Lacorte, C., Arago, F. J., Covas, D. T., Fontes, A. M., and other 3
authors: Accumulation of functional recombinant human coagulation factor IX
in transgenic soybean seeds, Transgenic Res., 20, 841e855 (2011).
20. Moravec, T., Schmidt, M. A., Herman, E. M., and Woodford-Thomas, T.: Production of Escherichia coli heat labile toxin (LT) B subunit in soybean seed and
analysis of its immunogenicity as an oral vaccine, Vaccine, 25, 1647e1657 (2007).
21. Schmidt, M. A. and Herman, E. M.: Proteome rebalancing in soybean seeds can
be exploited to enhance foreign protein accumulation, Plant Biotechnol. J., 6,
832e842 (2008).

447

22. Tada, Y., Utsumi, S., and Takaiwa, F.: Foreign gene products can be enhanced
by introduction into low storage protein mutants, Plant Biotechnol. J., 1,
411e422 (2003).
23. Cabanos, C., Ekyo, A., Amari, Y., Kato, N., Kuroda, M., Nagaoka, S.,
Takaiwa, F., Utsumi, S., and Maruyama, N.: High-level production of lactostatin, a hypocholesterolemic peptide, in transgenic rice using soybean A1aB1b
as carrier, Transgenic Res., 22, 621e629 (2013).
24. Hiraiwa, N., Nishimura, M., and Hara-Nishimura, I.: Vacuolar processing
enzyme is self-catalytically activated by sequential removal of the C-terminal
and N-terminal propeptides, FEBS Lett., 447, 213e216 (1999).
25. Bosch, M., Poulter, N. S., Perry, R.,M., Wilkins, K., and Franklin-Tong, W. E.:
Characterization of a legumain/vacuolar processing enzyme and YVADase activity in Papaver pollen, Plant Mol. Biol., 74, 381e393 (2010).
26. Shimada, T., Yamada, K., Kataoka, M., Nakaune, S., Koumoto, Y.,
Kuroyanagi, M., Tabata, S., Kato, T., Shinozaki, K., Seki, M., and other 4
authors: Vacuolar processing enzymes are essential for proper processing of
seed storage proteins in Arabidopsis thaliana, J. Biol. Chem., 278, 32292e32299
(2003).
27. Terauchi, K., Asakura, T., Nishizawa, N. K., Matsumoto, I., and Abe, K.:
Characterization of the genes for two soybean aspartic proteinases and analysis of their different tissue-dependent expression, Planta, 218, 947e957
(2004).
28. Terauchi, K., Asakura, T., Ueda, H., Tamura, T., Tamura, K., Matsumoto, I.,
Misaka, T., Hara-Nishimura, I., and Abe, K.: Plant-specic insertions in the
soybean aspartic proteinases, soyAP1 and soyAP2, perform different functions
of vacuolar targeting, J. Plant Physiol., 163, 856e862 (2006).
29. Mazorra-Manzano, M. A., Tanaka, T., Dee, D. R., and Yada, R. Y.: Structurefunction characterization of the recombinant aspartic proteinase A1 from
Arabidopsis thaliana, Phytochemistry, 71, 515e523 (2010).
30. Kervinen, J. and Wlodawer, A.: Phytepsin, pp. 118e124, in: Barrett, A. J.,
Rawlings, N. D., and Salvesen, G. (Eds.), Handbook of proteolytic enzymes, vol. 1,
3rd ed. Elsevier Academic Press, London (2013).