Biologia 66/3: 387—394, 2011

Section Cellular and Molecular Biology

DOI: 10.2478/s11756-011-0053-y
Review

In vitro synthesis of glycogen: the structure, properties, and

physiological function of enzymatically-synthesized glycogen

Hideki Kajiura1,2, Hiroki Takata2, Tsunehisa Akiyama2, Ryo Kakutani2, Takashi

Furuyashiki2, Iwao Kojima2, Toshiaki Harui1 & Takashi Kuriki2
1
Development Department, Food Material Division, Glico Foods Co., Ltd., 7–16 Kasuga-cho, Takatsuki, Osaka 569-0053,
Japan; e-mail: kajiura-hideki@gf.glico.co.jp
2
Institute of Health Sciences, Ezaki Glico Co., Ltd., 4-5-6, Utajima, Nishiyodogawa-ku, Osaka 555-8502, Japan

Abstract: This review describes a new enzymatic method for in vitro glycogen synthesis and its structure and properties. In
this method, short-chain amylose is used as the substrate for branching enzymes (BE, EC 2.4.1.18). Although a kidney bean
BE and Bacillus cereus BE could not synthesize high-molecular weight glucan, BEs from 6 other bacterial sources produced
enzymatically synthesized glycogen (ESG). The BE from Aquifex aeolicus was the most suitable for the production of
glycogen with a weight-average molecular weight (Mw ) of 3,000–30,000 k. The molecular weight of the ESG is controllable
by changing the concentration of the substrate amylose. Furthermore, the addition of amylomaltase (AM, EC 2.4.1.25)
significantly enhanced the efficiency of this process, and the yield of ESG reached approximately 65%. Typical preparations
of ESG obtained by this method were subjected to structural analyses. The average chain length, interior chain length, and
exterior chain length of the ESGs were 8.2–11.6, 2.0–3.3, and 4.2–7.6, respectively. Transmission electron microscopy and
intrinsic viscosity measurement showed that the ESG molecules formed spherical particles. Unlike starch, the ESGs were
barely degraded by pullulanase. Solutions of ESG were opalescent (milky-white and slightly bluish), and gave a reddish-
brown color on the addition of iodine. These analyses revealed that ESG shares similar molecular shapes and solution
properties with natural-source glycogen. Moreover, ESG had macrophage-stimulating activity and its activity depends on
the molecular weight of ESG. We successfully achieved large scale production of ESG. ESG could lead to new industrial
applications, such as in the food, chemical, and pharmaceutical fields.
Key words: glycogen; branching enzyme; enzymatic synthesis; α-glucan; polysaccharide; structure.
Abbreviations: AM, amylomaltase; BE, branching enzyme; DP, degree of polymerization; ESG, enzymatically synthesized
glycogen; GP, α-glucan phosphorylase; HPSEC-MALLS-RI, high-performance size-exclusion chromatography with a multi-
angle laser light scattering photometer and a differential refractive index detector; IAM, isoamylase; NSG, natural-source
glycogen; SP, sucrose phosphorylase; TEM, transmission electron microscopy.

Glycogen
Glycogen is the storage form of glucose in animals,
fungi, bacteria, yeasts, and archaea. The polysaccharide
is a highly branched (1→4)(1→6)-linked α-d-glucan
with a high molecular weight (106 –109) (Geddes 1986).
Based on the results of electron microscopy, it has been
shown that glycogen consists of spherical particles with
diameters of 20–40 nm (β-particles), and that the β-
particles often associate into much larger α-particles
(∼200 nm). The molecular weight of an individual
β-particle has been shown to be approximately 107
(Geddes 1986). Glycogen aqueous solution is opalescent
(milky-white and slightly bluish), and gives a reddish-
brown color on the addition of iodine. Amylopectin, a
major component of starch, is also a highly branched
(1→4)(1→6)-linked α-d-glucan. However, the degree of
branching, namely the number of (1→6)-α-d-glucosidic
linkages, in glycogen is about twice that in amylopectin.
A glycogen-type polymer, referred to as phytoglycogen,
has also been isolated from several higher plants. In an-
imal tissues, glycogen is synthesized from UDP-glucose
via the cooperative action of glycogenin (EC 2.4.1.186),
glycogen synthase (EC 2.4.1.11), and branching en-
zyme (BE, 1,4-α-d-glucan:1,4-α-d-glucan 6-α-d-(1,4-α-
glucano)-transferase, EC 2.4.1.18).
Glycogen isolated from animal tissues or shellfish
has been used to stimulate the exudation of phagocytic
cells into the peritoneal cavities of experimental ani-
mals in immunological studies (Thorpe & Marcus 1967;
Mantovani 1981). It has also been reported that glyco-
gen has an antitumor effect, probably through its im-
munomodulating activity, and that this activity seems
to depend on the fine structure of glycogen (Takaya et
al. 1998; Takaya 2000). Glycogen has long been consid-
ered to have health benefits as a food ingredient. As re-
vealed below, we demonstrated that enzymatically syn-
thesized glycogen (ESG) has a stimulating activity on
immunocompetent cells, such as macrophages. Glyco-
gen has also been used as a raw material in the cosmetic


c 2011 Institute of Molecular Biology, Slovak Academy of Sciences
388
industry and as a carrier to enhance the yield of DNA
during precipitation with organic solvent (Heath et al.
2001).
In vitro synthesis of glycogen
Extraction of glycogen from natural resources such as
animal tissues or shellfishes is laborious and costly pro-
cess. Therefore, in vitro synthesis may be possible an-
swer to obtain glycogen in industrial scale. Actually,
the in vitro synthesis of glycogen has been success-
fully achieved by the combined action of α-glucan phos-
phorylase (GP, EC 2.4.1.1) and BE using glucose-1-
phosphate (G-1-P) as a substrate (GP-BE method)
about 70 years ago. (Cori & Cori 1943). At that time,
enzymes were extracted from animal tissues and costly.
Additionally, GP from animal tissues is allosteric en-
zyme and need to be phosphorylated or added an acti-
vator (glucose-6-phosphate) (Fukui et al. 1982). How-
ever, microbial or plant enzymes have been developed
and used for the in vitro synthesis by many researchers
(e.g., Tolmasky & Krisman 1987; Fujii et al. 2003;
van der Vlist et al. 2008). The ESG produced by
the GP-BE method has properties similar to those of
the natural-source glycogen (NSG) (Kitahata & Okada
1988). UDP-glucose, a natural substrate for in vivo
glycogen synthesis, was also used as substrate by us-
ing glycogen synthase and BE as catalysts (Parodi et
al. 1969). Although both G-1-P and UDP-glucose are
expensive for industrial applications, the former could
be obtained from the inexpensive substrate sucrose by
using sucrose phosphorylase (SP, EC 2.4.1.7) (SP-GP-
BE method) (Ohdan et al. 2006). Furthermore, we have
demonstrated that ESG produced by the SP-GP-BE
method has immunomodulating activity (Ryoyama et
al. 2004). However, the yield of glycogen from sucrose
was less than 40%, and a more efficient method is re-
quired to meet the demand for glycogen for industrial
applications.
In the GP-BE and SP-GP-BE methods, GP elon-
gates a chain of α-(1→4)-glucan and BE introduces
branch points in the growing chains, mimicking the in
Table 1. Weight-average molecular weight and yield of α-glucans produced by the action of several BEs on two amyloses.a
Source of BE
Amylose A
Mw (k)
kidney bean N.D.b
Bacillus cereus N.D.
Escherichia coli 3600
Bacillus caldovelox 5910
Bacillus caldolyticus 36300
Bacillus stearothermophilus 2620
Aquifex aeolicus 9370
Rhodothermus obamensis 10300
a A reaction mixture (1 mL) containing 5 mg of Amylose A or Amylose AS-10, 50 mM buffer (optimum pH) and 200 U of various
BEs was incubated at optimum temperature for each enzyme for 24 h.
b Not detected by HPSEC-MALLS-RI.
H. Kajiura et al.
vivo synthesis of glycogen by glycogen synthase and BE.
Then, a question has come up. Can we synthesize glyco-
gen simply using amylose as a substrate? The published
results to date have all been negative: BE can synthe-
size branched glucans but their molecular weights are
much lower than that of glycogen. For example, Praznik
et al. (1992) reported that the action of potato BE on
amylose with a weight-average molecular weight (Mw )
of 67.6 k resulted in a branched product with a Mw of
33.5 k. Boyer et al. (1982) analyzed the action of maize
BE isozyme I by gel filtration chromatography and ob-
served that the product was eluted more slowly than the
substrate. Kitamura (1996), using light-scattering anal-
yses, also reported that potato BE brought about a de-
crease in the molecular weight of the substrate amylose.
We have reported that Bacillus cereus BE produced
branched glucan with a Mw of 50 k from amylose with
Mw of 320 k (Takata et al. 2005). The decreases in the
molecular weight of the substrate can be attributable to
a cyclization reaction by BE; it has been demonstrated
that BE catalyzes a cyclization reaction in addition to
a branching reaction in vitro. As a result of cycliza-
tion, the molecular weight of the substrate should be
decreased. By contrast, an inter-chain branching reac-
tion results in both larger and smaller molecules if two
molecules of the same size are used as a substrate. An
intra-chain branching reaction should not change the
molecular weight. However, most studies have used BE
from plant sources and relatively long-chain amyloses
were used as the substrate. To evaluate the possibility
of glycogen synthesis from amylose, we tested the ac-
tivity of eight types of BE from various sources using
relatively short-chain amyloses as the substrates. We
found that some BEs from bacterial sources can syn-
thesize glycogen (ESG). We then compared some of the
properties of the ESG with those of NSG. Furthermore,
we found that ESG has physiological function.
Actions of BEs from several sources on amyloses
To evaluate the possibility of producing glycogen from
amylose, we tested the actions of BEs from eight sources
Substrate
Amylose AS10
Yield (%) Mw (k) Yield (%)
N.D. 185 16.0
N.D. 87 25.5
3.9 3900 25.2
4.5 10200 23.7
1.2 30400 5.3
6.7 5760 32.0
10.1 20400 59.0
31.2 45300 66.7
In vitro synthesis of glycogen 389

on short-chain amyloses using high-performance size-

exclusion chromatography with a multi-angle laser light
scattering photometer and a differential refractive index
detector (HPSEC-MALLS-RI) (Kajiura et al. 2008).
The substrates used were Amylose A (Mn , number-
average molecular weight, 2.9 k; Nacalai Tesque, Inc.,
Kyoto, Japan) and Amylose AS-10 (Mw , 10 k; Nakano
Vinegar Co., Ltd., Aichi, Japan). In this analysis sys-
tem, the peak for short-chain amylose given by the RI-
detector overlapped the peak for the salts in the re-
action buffer. However, if glucan with a high molecu-
lar weight (>1,000 k) is produced, its peak could be
easily detected. Table 1 summarizes the results. In the
reaction mixtures including kidney bean BE and Bacil-
lus cereus BE, no high molecular weight glucans were
detected. However, glucans with Mw more than 1,000
k were detected in the mixtures including other BEs.
Rhodothermus obamensis BE gave glucans with an ex-
tremely high molecular weight, whereas Aquifex aeoli-
cus BE produced glucan that was similar in size to the
β-particles of NSG. A. aeolicus BE was selected for fur-
ther analyses.
The high-molecular-weight glucan produced by
this method (Fig. 1) can be considered to be ESG, be-
cause the properties and structures of the products were
similar to those of NSGs as described later.
From a practical standpoint, Amylose A is prefer-
able to Amylose AS-10, since the former can be easily
obtained by treating starch with isoamylase (IAM), al-
though the yield of ESG from Amylose A was lower
than that from Amylose AS-10.

Effect of amylomaltase

In order to solve the problem of low yield of ESG, we

added amylomaltase (AM, EC 2.4.1.25) from Thermus
aquaticus to the reaction mixture. The addition of AM
to the reaction mixture significantly improved the yield
of ESG (Kajiura et al. 2008). The yield reached 64%
under optimal conditions.
How to control the molecular weight of ESG
After debranching of partially degraded starch, mal-
todextrin (DE 8–9.5), by using IAM, we used it at
various concentrations as the substrate for BEs in the
synthesis of ESGs (Kajiura et al. 2008). At a higher
substrate concentration, the Mw values of ESGs were
lower. By changing the concentration of substrate, we
can control the Mw of ESGs.
Structures of ESG
We synthesized ESGs (A-I) under various reaction con-
ditions, and compared the properties and structures of
the ESGs with those of NSGs (Table 2) (Kajiura et al.
2008, 2010). The fine structural parameters of glyco-
gen were the average chain length, the exterior chain
length, and the interior chain length (Hata et al. 1984;
Manners 1991). The structures of NSGs depend on the
Fig. 1. Glycogen synthesis by the IAM-BE-AM method. ↓, α-1,6-
glucosidic linkage; –, α-1,4-glucosidic linkage; ◦, glucosyl residue;
◦
, glucosyl residue with reducing end; , the α-1,4 linkage cleaved
by BE; , the glucosyl residue used as an acceptor; n, number
of amylose molecules initially present in the reaction mixture; m,
number of synthetic glycogen molecules. Some glucosyl residues
connected by α-1,4 linkages are indicated with broken lines. (A)
Short-chain amyloses are firstly prepared from starch or mal-
todextrin by using IAM. (B) A singly-branched molecules are
produced by the action of BE on short-chain amyloses. (C) The
branched molecule is preferentially used as a substrate for BE
in the second reaction. (D) Branching reactions are repeated to
produce high-molecular-weight glucans (glycogens).
source and extraction method, and therefore they show
structural heterogeneity (Table 2). The average chain
lengths of ESGs (A-I) tended to be slightly shorter than
those of NSGs, whereas exterior chain lengths and in-
terior chain lengths were within the variations of the
values for NSGs (Table 2). On the other hand, the
unit-chain distributions of ESGs and NSGs as analyzed
by high-performance anion exchange chromatography
after IAM treatment were slightly different from each
other (Takata et al. 2009): NSGs have larger amounts
of long unit chains (degree of polymerization (DP): 25–
390 H. Kajiura et al.

Fig. 2. Comparison of aqueous solutions of α-glucans. (a) Waxy corn starch (0.1 g) was gelatinized in 10 mL of distilled water by
heating at 100 ◦C for 30 min. The same amounts of maltodextrin (DE 8–9.5), NSG from rabbit liver and ESG-F were each dissolved
in 10 mL of distilled water. (b) These solutions (1 mL) were added 10 µL of iodine reagent (52 mg of I2 and 520 mg of KI in 10 mL
of water).

Table 2. Structural parameters of ESGs and NSGs.

Sample name
or NSG source Mw (k) CLa ECLb ICLc

ESG A 2720 9.3 5.3 3.0

ESG B 4960 11.6 7.6 3.0
ESG C 7080 9.3 6.3 2.0
ESG D 7940 8.5 5.0 2.5
ESG E 9720 9.1 6.1 2.0
ESG F 11800 8.2 4.2 3.0
ESG G 13900 8.7 4.8 2.8
ESG H 15000 8.6 5.0 2.6
ESG I 28700 8.9 4.6 3.3

Oyster 6010 10.2 6.6 2.7

Slipper limpet 4830 9.0 5.3 2.7
Mussel 5120 10.6 6.5 3.1
Bovine liver 2030 11.9 7.0 3.9
Rabbit liver 15700 12.5 7.7 3.9
Sweet corn 19800 13.3 8.2 4.1
Fig. 3. 5% (w/v) aqueous solutions of ESGs.
a Number-average chain length, CL = [Total sugars as glucose]/
[non-reducing end residues].
b Exterior chain length, ECL = CL×β-amylolysis (%)/100 + 2.
c Interior chain length, ICL = CL – ECL – 1.
ure 2b shows the comparison of the solutions with the
addition of iodine. The solutions of amylopectin and
maltodextrin turned dark blue and dark purple colors,
35) than ESGs. In other words, ESGs have narrower respectively. By contrast, the solutions of NSG from
unit-chain distributions than NSGs. rabbit liver and ESG-F turned a reddish-brown color.
NSG and ESG showed similar color reaction with io-
Appearance of aqueous solution dine.
The larger the molecular size, the deeper were
Figure 2a shows a comparison of the solutions of amy- the opalescent colors of the aqueous solutions of ESGs
lopectin, maltodextrin, NSG from rabbit liver and ESG- (Fig. 3). The colors of the NSG solutions also depended
F. The solution of maltodextrin was clear/colorless, on their molecular sizes (data not shown). The concen-
whereas the solution of amylopectin was cloudy. By con- tration of ESG also affected the depth of color (Fig. 4).
trast, the solutions of glycogens (NSG from rabbit liver In the concentration range of 1–10% (w/v), the higher
and ESG-F) were opalescent, as described by Manners the concentration, the deeper was the color. However,
(1991). All ESGs listed in Table 2 were easily soluble in in the range of 10–20% (w/v), the color of the solution
water and the aqueous solutions were also opalescent, became lighter. Similar phenomena were also observed
resembling NSGs (data not shown). Furthermore, Fig- for the NSG solutions (data not shown).
In vitro synthesis of glycogen
Fig. 4. 1, 5, 10 and 20% (w/v) aqueous solutions of ESG-F.
Fig. 5. TEM images of ESG-C and NSG from mussel. Scale bars:
100 nm.
Image of transmission electron microscopy
(TEM)
According to TEM images (Fig. 5), both ESG and NSG
have a slightly distorted circular shape, with diameters
of 20–40 nm (Kajiura et al. 2010). The shapes and sizes
were in good agreement with those of β-particles pre-
viously reported by Hata et al. (1983). In the image of
NSG, a few large and irregular shaped particles with di-
391
ameters of around 80–100 nm were found. These large
particles may be aggregates of smaller particles, since
it is known that the β-particles sometimes associate to
form larger α-particles (60–200 nm) with a rosette-like
appearance (Kjoelberg et al. 1963; Manners 1991). In
the image of ESG, aggregated forms (α-particles) of
glycogen were not found.
Intrinsic viscosity of ESG
We determined the intrinsic viscosities of glycogens by
using a capillary viscometer of Ubbelohde type. The
intrinsic viscosities of ESGs and NSGs were very low
(around 8 mL/g), and showed no molecular weight de-
pendence (Kajiura et al. 2010). By contrast, Kitamura
et al. (1989) and Kitamura (1996) demonstrated that
the intrinsic viscosities of amylose and amylopectin in-
crease with their molecular weights. The intrinsic vis-
cosity result suggested that ESG molecules behave as
hard spheres in solution, resembling the behavior of
NSG.
Digestibility of ESG
Digestibilities of maltodextrin, NSGs and ESGs were
investigated according to an in vitro digestion method
(Okada et al. 1990; Hashimoto et al. 2006) (Tables 3, 4).
All polysaccharides were not digested by artificial gas-
tric juice. Glycogens were digested by human salivary
α-amylase, porcine pancreatic α-amylase, and rat small
intestinal mucosa enzymes, the hydrolysis rates were
46–62%, 44–55%, and 60–75%, respectively. Maltodex-
trin was found to be digested easier than glycogens,
as expected from its lower degree of branching. The di-
gestibility of ESG was seemed to be lowest among these
polysaccharides.
Susceptibility of ESG to pullulanase and α-
amylase
Pullulanase (EC 3.2.1.41) hydrolyzes α-1,6 linkages of
α-1,4/1,6 glucans such as pullulan, starch, and glyco-
gen. However, it has been shown that the enzyme
only partially hydrolyzes glycogen, trimming its out-
ermost branches (Kitahata 1995). Glycogen synthe-
sized by the GP-BE method was shown to be resis-
tant to pullulanase (Boyer & Preiss 1977; Kitahata
& Okada 1988). Therefore, we tested the pullulanase
susceptibilities of the ESG and NSG (Fig. 6) (Ka-
jiura et al. 2008, 2010). Waxy corn starch was easily
degraded into small fragments by pullulanase. How-
ever, NSG from oyster and the ESG-B were barely de-
graded. This result indicates that the ESG was simi-
lar to NSG, but was unlike amylopectin. Furthermore,
we tested the susceptibility of ESG by excess amounts
of α-amylase (Takata et al. 2009). The final products
of α-amylase hydrolysis of NSG were glucose, maltose,
maltotriose, branched oligosaccharides with DP≥4, and
highly branched macrodextrin molecules with molecu-
lar weights of up to 10 k. In the final products of α-
392 H. Kajiura et al.
Table 3. Digestive conditions.

Reaction conditions

Substrate concentration (%) U/g-substrate pH ◦C Time (min)

Human salivary α-amylase 5 160 6.0 37 30

Artificial gastric juice (0.02M HCl-KCl) 0.7 2.0 37 100
Porcine pancreatic α-amylase 0.5 400 6.6 37 360
Rat small intestinal mucosa enzymes 0.5 43 6.6 37 180

Table 4. Digestibility of glycogen.a

Reducing sugar increased (%)

Human salivary Artificial Porcine pancreatic Rat small intestinal

Sample α-amylase gastric juice α-amylase mucosa enzymes

Maltodextrin (DE 2–5) 67.2 0.0 55.5 77.7

NSG (oyster) 50.9 0.0 43.8 74.5
NSG (sweet corn) 61.8 0.0 55.3 72.2
ESG-A 46.6 0.0 43.6 60.2
ESG-C 46.0 0.0 44.2 62.9
a Reducing sugars were measured by the dinitrosalicylic acid method (Takata et al. 2009).

Fig. 7. Macrophage-stimulating activity of ESG. RAW264.7 was

cultured with ESGs or lipopolysaccharide (LPS) in the presence
of interferon γ. After cultivation for 48 h, NO concentration in
the culture supernatants was determined.

Fig. 6. Susceptibility of α-glucan to pullulanase. ESG-B (), NSG
from oyster (), and gelatinized waxy corn starch ( ) were treated
with pullulanase. Mw values of hydrolyzates were analyzed by
HPSEC-MALLS-RI.
amylase hydrolysis of ESGs, we detected much larger
macrodextrins (molecular weight >1,000 k). In contrast
to NSG, oligosaccharides with DP 4-7 could not be de-
tected in the ESG hydrolysates. These results suggest
that the α-1,6 linkages in ESG molecules are more reg-
ularly distributed than those in NSG molecules. ESG is
synthesized in much simpler circumstances than NSG,
in which no trimming reactions occur during the syn-
thetic process. This simplicity can result in a regular in-
ternal structure without long spans between α-1,6 link-
ages.
Physiological function of ESG
Glycogen is well known as the energy and carbon re-
serves in animal cells and microorganisms. In addition
to this role, some reports have suggested that glyco-
gen has an immunological activity (Thorpe & Marcus
1967; Mantovani 1981; Takaya et al. 1998). However,
strong scientific evidence has not been obtained for
the immunomodulating activity. Indeed, many scien-
tists have been annoyed by the lack of reproducibility
of their experimental results. This lack of reproducibil-
ity may be because their samples of glycogen have been
extracted from natural resources, as (1) the effect of
trace amounts of other materials cannot be ruled out,
and (2) important characteristics of each glycogen sam-
ple, such as the average molecular mass and the chain
length, are quite different depending on the source and
purification procedures. We have clarified the immuno-
logical activity of glycogen by using completely pure
ESG with very uniform characteristics (Kakutani et
al. 2007, 2008). First, we examined the immunestim-
ulating effect of ESG on NO (nitric oxide) produc-
tion using RAW264.7, a murine macrophage cell line.
The result showed glycogen enhanced NO production
of RAW264.7 (Fig. 7). The immunostimulatory activity
In vitro synthesis of glycogen
of glycogen relates closely to its molecular weight and
reaches to the highest at around 5,000 k-6,500 k. Next,
the peritoneal-exudate cells collected from C3H/HeJ
mice, Toll-like receptor 4 mutants, were also activated
by ESGs with similar profiles as RAW264.7 (Kakutani
et al. 2007). Furthermore, we demonstrated by flow cy-
tometry that biotinylated ESGs bound the macrophage
cell line (Kakutani et al. 2007, 2008). These results
strongly suggested that glycogen functions not only as
a fuel reservoir, but also as a signaling factor in vivo.
Conclusion and future prospects
We have developed a novel method (IAM-BE-AM
method) for producing glycogen from short-chain amy-
lose by using BE. This method has two advantages com-
pared with the SP-GP-BE method, in which sucrose is
used as a starting material. First, the yield of glyco-
gen is higher. The yield of glycogen by the SP-GP-BE
method was less than 40%, whereas that by the IAM-
BE-AM method reached 65% under the optimal condi-
tions. Second, the molecular weight of glycogen is con-
trollable within the range of 3,000–30,000 k by changing
the concentration of substrate.
The structures and properties of the ESGs were
quite similar to those of NSGs. Despite these simi-
larities, there were two differences (the unit-chain dis-
tribution and final products by α-amylase hydrolysis)
between the ESG and NSG. Furthermore, we have
demonstrated that ESG has immunomodulating activ-
ity. We hypothesize that the macromolecule fraction af-
ter partial hydrolysis of ESG that reaches the intesti-
nal tract stimulates immunocompetent cells resulting
in the health benefits. Investigations in vivo are now in
progress to gain a further understanding of the effect of
orally administered glycogen.
Acknowledgements
We are deeply grateful to Dr. S. Kitamura of Osaka Pre-
fecture University for his kind help in performing structural
analyses and testing the solution properties of ESG. We are
also indebted to Drs. H. Matsui and H. Ito of Hokkaido Uni-
versity for providing the plasmid used to produce the BE of
Phaseolus vulgaris L. We also thank Drs. N. Ohno and Y.
Adachi of Tokyo University of Pharmacy and Life Science
for their advice and valuable discussions during the course of
the immunological studies. This work was supported in part
by grants from the “Program to develop new technology to
promote the agriculture, forestry, fisheries and food indus-
tries through collaboration among industry, academia and
the government” and “Research and development projects
to promote new policies in agriculture, forestry and fish-
eries” of the Ministry of Agriculture, Forestry and Fisheries,
Japan.
References
Boyer C. & Preiss J. 1977. Biosynthesis of bacterial glycogen:
purification and properties of the Escherichia coli B α-1,4–
glucan: α-1,4–glucan 6–glycosyltransferase. Biochemistry 16:
3693–3699.
393
Boyer C.D., Simpson E.K. G. & Damewood P.A. 1982. The pos-
sible relationship of starch and phytoglycogen in sweet corn.
II. The role of branching enzyme I. Starch/Stärke 34: 81–85.
Cori G.T. & Cori C.F. 1943. Crystalline muscle phosphorylase.
IV. Formation of glycogen. J. Biol. Chem. 151: 57–63.
Fujii K., Takata H., Yanase M., Terada Y., Ohdan K., Takaha T.,
Okada S. & Kuriki T. 2003. Bioengineering and application of
novel glucose polymers. Biocatal. Biotransform. 21: 167–172.
Fukui T., Shimomura S. & Nakano K. 1982. Potato and rabbit
muscle phosphorylases: comparative studies on the structure,
function and regulation of regulatory and nonregulatory en-
zymes. Mol. Cell. Biochem. 42: 129–144.
Geddes R. 1986. Glycogen: a metabolic viewpoint. Biosci. Rep.
6: 415–428.
Hashimoto T., Kurose M., Oku K., Nishimoto T., Chaen H.,
Fukuda S. & Tsujisaka Y. 2006. Digestibility and suppressive
effect on rats’ body fat accumulation of cyclic tetrasaccharide.
J. Appl. Glycosci. 53: 233–239.
Hata K., Hata M., Hata M. & Matsuda K. 1983. The structures
of shellfish glycogens I. J. Jpn. Soc. Starch Sci. 30: 88–94.
Hata K., Hata M., Hata M. & Matsuda K. 1984. A proposed
model of glycogen particle. J. Jpn. Soc. Starch Sci. 31: 146–
155.
Heath E.M., Morken N.W., Campbell K.A., Tkach D., Boyd E.A.
& Strom D.A. 2001. Use of buccal cells collected in mouth-
wash as a source of DNA for clinical testing. Arch. Pathol.
Lab. Med. 125: 127–133.
Kajiura H., Kakutani R., Akiyama T., Takata H. & Kuriki T.
2008. A novel enzymatic process for glycogen production. Bio-
catal. Biotransform. 26: 133–140.
Kajiura H., Takata H., Kuriki T. & Kitamura S. 2010. Structure
and solution properties of enzymatically synthesized glyco-
gen. Carbohydr. Res. 345: 817–824.
Kakutani R., Adachi Y., Kajiura H., Takata H., Kuriki T. & Ohno
N. 2007. Relationship between structure and immunostimu-
lating activity of enzymatically synthesized glycogen. Carbo-
hydr Res. 342: 2371–2379.
Kakutani R., Adachi Y., Kajiura H., Takata H., Ohno N. &
Kuriki T. 2008. Stimulation of macrophage by enzymatically
synthesized glycogen: the relationship between structure and
biological activity. Biocatal. Biotransform. 26: 152–160.
Kitahata S. 1995. Debranching enzymes (isoamylase, pullu-
lanase), pp. 18–27. In: The Amylase Research Society of
Japan (ed.), Enzyme Chemistry and Molecular Biology of
Amylases and Related Enzymes, CRC Press, Boca Raton.
Kitahata S. & Okada S. 1988. Branching enzymes, pp. 143–154.
In: The Amylase Research Society of Japan (ed.), Handbook
of Amylase and Related Enzymes, Pergamon Press, Oxford.
Kitamura S. 1996. Starch polymers, natural and synthetic, pp.
7915–7922. In: Salamone J.C. (ed.), Polymeric Materials En-
cyclopedia, CRC Press, Boca Raton.
Kitamura S., Kobayashi K., Tanahashi H., Ozaki T. & Kuge T.
1989. On the Mark-Houwink-Sakurada equation for amylose
in aqueous solvents. (Dilute solution properties of starch re-
lated polysaccharides. Part 1) Denpun Kagaku 36: 303–309.
Kjoelberg O., Manners D.J. & Wright A. 1963. α-1,4–Glucosans.
XVII. The molecular structure of some glycogens. Comp.
Biochem. Physiol. 34: 353–365.
Manners D.J. 1991. Recent developments in our understanding
of glycogen structure. Carbohydr. Polym. 16: 37–82.
Mantovani B. 1981. Phagocytosis of immune complexes mediated
by IgM and C3 receptors by macrophages from mice treated
with glycogen. J Immunol. 126: 127–130.
Ohdan K., Fujii K., Yanase M., Takaha T. & Kuriki, T. 2006.
Enzymatic synthesis of amylose. Biocatal. Biotransform. 24:
77–81.
Okada K., Yoneyama M., Mandai T., Aga H., Sakai S. & Ichikawa
T. 1990. Digestion and fermentation of pullulan. J. Jpn. Soc.
Nutr. Food Sci. 43: 23–29. (in Japanese)
Parodi A.J., Krisman C.R., Leloir L.F. & Mordoh J. 1967. Prop-
erties of synthetic and native liver glycogen. Arch. Biochem.
Biophys. 121: 769–778.
Praznik W., Rammesmayer G. & Spies T. 1992. Characterization
of the (1→4)-α-D-glucan-branching 6-glycosyltransferase by
394
in vitro synthesis of branched starch polysaccharides. Carbo-
hydr. Res. 227: 171–182.
Ryoyama K., Kidachi Y., Yamaguchi H., Kajiura H. & Takata
H. 2004. Anti-tumor activity of an enzymatically synthesized
α-1,6 branched α-1,4-glucan, glycogen. Biosci. Biotechnol.
Biochem. 68: 2332–2340.
Takata H., Kajiura H., Furuyashiki T., Kakutani R. & Kuriki
T. 2009. Fine structural properties of natural and synthetic
glycogens. Carbohydr. Res. 344: 654–659.
Takata H., Kato T., Takagi M. & Imanaka T. 2005. Cyclization
reaction catalyzed by Bacillus cereus branching enzyme, and
the structure of cyclic glucan produced by the enzyme from
amylose. J. Appl. Glycosci. 52: 359–365.
Takaya Y. 2000. Biological activities of natural resources around
us are now in the limelight. Yakugaku Zasshi 120: 1075–1089.
Takaya Y., Uchisawa H., Ichinohe H., Sasaki J., Ishida K. & Mat-
sue H. 1998. Antitumor glycogen from scallops and the in-
terrelationship of structure and antitumor activity. J. Mar.
Biotechnol. 6: 208–213.
H. Kajiura et al.
Thorpe B. D. & Marcus S. 1967. Phagocytosis and intracellu-
lar fate of Pasteurella tularensis: in vitro effects of exudate
stimulants and streptomycin on phagocytic cells. J Reticu-
loendothel Soc. 4: 10–23.
Tolmasky D. S. & Krisman C. R. 1987. The degree of branching in
(α1,4)-(α1,6)-linked glucopolysaccharides is dependent on in-
trinsic properties of the branching enzymes. Eur. J. Biochem.
168: 393–397.
van der Vlist J., Palomo Reixach M., van der Maarel M., Di-
jkhuizen L., Schouten A.J. & Loos K. 2008. Synthesis of
branched polyglucans by the tandem action of potato phos-
phorylase and Deinococcus geothermalis glycogen branching
enzyme. Macromol. Rapid Commun. 29: 1293–1297.
Received December 29, 2011
Accepted March 11, 2011