Content uploaded by Ana M. Ibanez
Author content
All content in this area was uploaded by Ana M. Ibanez on Aug 16, 2020
Content may be subject to copyright.
In Myoung Park
Ana Maria Ibáñez
Charles F. Shoemaker
Department of Food Science
& Technology,
University of California,
Davis, CA, USA
Rice Starch Molecular Size and its Relationship with
Amylose Content
The composition and starch molecular structure of eight rice varieties were studied.
Waxy and non-waxy (long-, medium-, and short-grain) rice varieties from California
and Texas were used. The amylose contents were measured using the Concanavalin A
method and were found to be related to the type of rice: waxy <1.0%, short and me-
dium grain 8.7–15.4%, and long grain 17.1–19.9%. The weight-average molar masses
(Mw) of the starches varied from 0.52 to 1.966108g/mol. As would be expected, a
higher Mwof rice starch correlated to lower amylose content. The range of Mwof
amylopectin was 0.82 to 2.50 6108g/mol, and there was also a negative correlation of
amylopectin Mwwith amylose content. Amylose Mwranged from 2.20 to 8.316105g/
mol. After debranching the amylopectin with isoamylase, the weight-average degree
of polymerization (DPw) for the long-chain fraction correlated positively with a
higher amylose content. California and Texas varieties were significantly different in
their amylose content, starch Mw(short- and medium-grain only), and amylopectin
Mw(p,0.05).
Keyword: Starch structure; Amylose content; Weight-average molar mass; z-Average
radius of gyration; Amylopectin branch chain length distribution
1 Introduction
Starch, the major storage polysaccharide, is a polymeric
mixture of predominantly linear (a-1?4) amylose and
highly branched amylopectin (a-1?6) glucan molecules
[1]. The ratio of amylose to amylopectin and the branching
properties of the amylopectin molecules of a rice starch
can affect the physical, textural, and pasting properties
during the cooking of rice and rice starch [2–5]. Rice
starches contain varied amylose contents; waxy, 0–2%
and non-waxy, 5–33% [2], and the molar masses of amy-
lopectins are about 100 times higher than those of amy-
lose [6–9]. The rice starch granule is one of the smallest
starch granules (2–5 mm diameter). The fine size and
unique white color of rice starch has led to its application
as a cosmetic powder, laundry stiffening agent, paper
coating powder, sugar coating in confectionery applica-
tions, and alternative ingredient for fats. In addition, rice
starch has various applications in food industries as an
allergy-free ingredient [2, 6, 10].
Static multi-angle laser light scattering (MALLS) in con-
junction with high-performance size-exclusion chroma-
tography (HPSEC) can be used to determine the molar
mass and mean square radius of gyration of polymers in
solution without the need for calibration standards [11].
This system has been used to measure starch weight-
average molar mass (Mw) [12–17], amylose Mw[18, 19]
and amylopectin Mw[12, 17, 20] from different cereal
sources. Furthermore, this method has been applied to
determine the weight-average degree of polymerization
(DPw) of the side chains of amylopectin [14, 15, 21, 22].
High-performance anion-exchange chromatography
equipped with a pulsed amperometric detector (HPAEC-
PAD) has been also used for effective separation of low
DP chains, but its sensitivity falls off with chains above a
DP of 80 [23].
This study focused on enhancing knowledge of rice
starch structure by studying a set of diverse rice types
from two growing locations, California (CA) and Texas
(TX). The following properties were investigated 1) chem-
ical composition of rice flour and starch, 2) the molar
mass and size of starch, amylopectin and amylose, and 3)
the DP of amylopectin branches.
2 Materials and Methods
2.1 Materials
One waxy rice (Calmochi 101) and seven non-waxy rices
(Bengal, Cocodrie, Koshihikari, L-205, M-202, M-401, S-
102) were studied. They were cultivated in two regions of
the United States, CA and TX during the 2003 cropping
year. The CA rice was provided by the California Coop-
Correspondence: In Myoung Park, Food Science & Technology
Department, University of California, Davis, CA 95616, USA.
Phone:11-530-752-7347, Fax: 11-530-752-4759, e-mail;
ipark@ucdavis.edu.
Starch/Stärke 59 (2007) 69–77 69
©2007 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim www.starch-journal.com
Research Paper
DOI 10.1002/star.200600568
70 I. M. Park et al. Starch/Stärke 59 (2007) 69–77
erative Rice Experimental Station, Biggs, CA, and the
Texas rice was provided by the USDA-ARS Rice Quality
Laboratory, Beaumont, TX. The TX paddy rices were
shipped to the California Cooperative Rice Experimental
Station. The CA and TX paddy rice were shelled using a
McGill Sheller (Houston, TX, USA) and then immediately
milled using a Yamamoto VP31T whitener (Yamamoto
Corp., Ltd., Yamagata, Japan), registering about 30 on a
Kett whiteness meter (C-300–3, Kett Electric Lab, Tokyo,
Japan) for short-, medium- and long-grain types and 40
for the waxy type. The outcome of the final whiteness was
different according to variety because long, medium and
short grains have a naturally higher appearance of white-
ness than the waxy type grain. The milled rice was ground
to rice flour using a UDY Cyclotec Cyclone mill (Tecator,
Höganäs, Sweden) with 0.5 mm mesh screen.
2.2 Starch isolation
Starch was isolated from milled rice using pronase (alka-
line protease; 45,000 U/g, Calbiochem, San Diego, CA,
USA) from Streptomyces griceus based on the method of
Biliaderis and Juliano [8]. One hundred grams of milled
rice were soaked in three times their volume of water for
3 h. The samples were then blended for 3 min in a food
processor. This dispersion was then passed through 100,
200 and 400 mesh screens consecutively and then the
collected wet-milled flour was centrifuged and the super-
natant was discarded. Next, the flour was treated with five
times its volume of 0.2% pronase in 0.03 M phosphate
buffer, pH 7.4, at 377C with 0.02% sodium azide. After
24 h, the dispersion was centrifuged. A fresh pronase so-
lution was then added, and the suspension was incu-
bated for another 24 h and centrifuged. Next, the non-
starch lipids were removed from the recovered starch by
treating the wet isolate twice with five times its volume of
water-saturated 1-butanol (WSB: 36% water: 64% buta-
nol) for a total incubation time of 24 h. The starch granules
were recovered by centrifugation at 2,0006gfor 10 min.
After washing with water, the final sample was freeze-
dried and sieved with a 100 mesh screen. Freeze-dried
rice starches were spread out on an aluminum plate and
equilibrated to ambient relative humidity (<50% RH) for
two weeks then stored in plastic bottles at room temper-
ature.
2.3 Determination of amylose content
The amylose content of the isolated rice starches was
determined using an amylose/amylopectin assay kit
(Megazyme, Wicklow, Ireland) based on the Con A meth-
od [24].
2.4 Starch and amylopectin sample preparation
for HPSEC
Samples for the HPSEC analysis were prepared following
the method used by Yokoyama et al. [16], except for the
final sample concentration selected as 0.4% (w/w).
Starch was added to the dimethyl sulfoxide (DMSO)/
50 mM LiBr solvent and then heated at 957C for 15 min, in
a Reactitherm stirring heating module (Pierce Chemical
Co., Rockford, IL, USA). After heating, the samples were
cooled and continuously stirred overnight and then cen-
trifuged for 10 min at 10,000 rpm (13,5006g) (Eppendorf,
model 5415C, Westbury, NY, USA) and filtered through a
1.2 mm nylon syringe membrane filter (Scientific Resour-
ces Inc., North Brunswick, NJ, USA) prior to analysis.
2.5 Fractionation of amylopectin
Amylopectin from isolated starch was fractionated using
a size-exclusion chromatography column (2.5 cm
6100 cm) with Sepharose CL-2B (exclusion range Mr
16105–261071, Pharmacia, Piscataway, NJ, USA) fol-
lowing the method of Han and Hamaker [25]. Starch
samples were prepared by first dispersing in DMSO
(concurrently with heating) and then recovered by pre-
cipitation by the addition of ethanol [26]. Rice starches
(25–28 mg) samples were weighed and solubilized in 1 mL
DMSO for 15 min at 957C. After cooling the samples, 6 mL
of ethanol was added to precipitate the starch. The mix-
tures were centrifuged at 2,0006gfor 5 min and the
supernatant was discarded. The sample tubes were
turned upside down on tissue paper for draining for
10 min. The precipitates were then redissolved in 7.5 mL
deionized distilled water and boiled for 15 min with stir-
ring. After filtering the sample solution through 0.45 mm
nylon membrane filter (Millipore, Bedford, MA, USA), the
sample solution was loaded to a column at a flow rate of
0.4 mL/min in descending mode, and 3 mL fractions were
collected. Aliquots from the fractionation (0.5mL) were
diluted in distilled water to 2.5 mL, and then 0.2 mL of
0.025 M potassium iodide/iodine solution was added.
The elution profiles were analyzed for blue value with
iodine solution at 630 nm. Amylopectin fractions were
collected, freeze-dried and later used for the measure-
ment of amylopectin Mw.
2.6 Enzymatic debranching of amylopectin and
sample preparation for HPSEC
Rice starches were debranched using a modified method
based on that of Bradbury and Bello [9], and Ward et al.
[27]. The starch samples were prepared by dissolving
50 mg of starch in 5 mL of distilled water and heated at
©2007 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim www.starch-journal.com
Starch/Stärke 59 (2007) 69–77 Rice Starch Molecular Size and Amylose Content 71
957C for 15min. The tubes were then placed in a water
bath at 457C to cool down. After cooling to 457C, 5 mL of
40 mM acetate buffer (pH 3.8) and 25 mL of the iso-
amylase from Pseudomonas amyloderamosa (59,000 U/
mg, Hayashibara Biochemical Labs, Okayama, Japan)
was added to each tube for debranching. After 24 h, the
enzyme reaction was stopped by heating the tubes at
1007C for 10 min. The debranched solution was cen-
trifuged and the supernatant was filtered through a
0.45 mm nylon syringe membrane filter (Scientific
Resources Inc., North Brunswick, NJ, USA). The filtrate
was then injected into the HPSEC system. The chroma-
togram was analyzed for Mw’s of amylose and the amylo-
pectin branches.
2.7 High-performance size-exclusion
chromatography
The HPSEC system consisted of an HP 1050 pump and
auto-injector (Hewlett Packard, Valley Forge, PA, USA)
fitted with a 100 mL injection loop. The system also
employed a MALLS detector (Dawn DSP-F, Wyatt Tech.,
Santa Barbara, CA, USA) with a He-Ne laser source (l=
632.8 nm), a K-5 flow cell, and a differential refractometer
detector (RI) (model ERC-7512, ERMA Inc., Tokyo,
Japan).
2.7.1 Determination of Mwof starch and
amylopectin
For starch and amylopectin samples, a Plgel Mixed-A
(Polymer Labs., Amherst, MA, USA) column was used
with DMSO (HPLC grade, Sigma Chemical Co., St. Louis,
MO, USA) with 50 mM LiBr (Fisher Scientific, Fair Lawn,
NJ, USA) as a mobile phase with a flow rate of 0.4mL/
min. A column bank of four Styragel columns (Styragel
two-HMW 7, HMW 6E and HMW 2; Waters Co., Milford,
MA, USA) was also used for comparing the resolution with
that of the Plgel Mixed-A column. The Plgel Mixed-A and
Styragel HMW columns had high-porosity 10 mm frits and
20 mm particles which were designed to minimize poly-
mer shear effects. A refractive index value of 1.4785 and a
dn/dcvalue of 0.066 for starch in DMSO/50 mM LiBr so-
lution were used for the molecular weight calculations.
2.7.2 Determination of Mwof amylose and the
debranched amylopectin
For the determination of the Mwof amylose and the amy-
lopectin branches, four aqueous SEC columns (Ultra-
hydrogel 120, 250, 500 and 1000; Waters Co., Milford,
MA, USA) with a guard column (Ultrahydrogel, 7567.5,
Millipore Co., USA) were used. The mobile phase was
0.1 M aqueous sodium nitrate with 0.02% sodium azide,
as a preservative, with a 0.6mL/min flow rate. A refractive
index of 1.3340 set at a dn/dcvalue of 0.160 was used for
the analysis of starch in H2O/NaNO3. The columns were
maintained at 407C with a column heater for both the
organic solvent and aqueous SEC experiments.
2.8 Data treatment
The data analysis methods were those used by Yokoyama
et al. [16]. Astra software (Version 4.7.07, Wyatt Technol-
ogy, Santa Barbara, CA, USA) was used for data analysis.
The Mwand z-average radius of gyration (Rz) for starch
and AP samples was calculated using the second-order
Berry method [7]. The Mwof amylose and the amylopectin
branches were analyzed using the first-order Debye
method [28].
2.9 Statistical analysis
Data were statistically analyzed by the general linear
models procedure (GLM), the difference between the
means of the samples were analyzed by the least signifi-
cant difference (LSD) test, at a probability level of 0.05,
and Pearson correlation coefficient, with the Statistical
Analysis System software (version 8., SAS Institute Inc.,
Cary , NC, USA).
3 Results and Discussion
3.1 Amylose content of rice starch
After analyzing the amylose content with the Con A
method, long-grain rice varieties were found to contain
17.1–19.9% amylose, while short- and medium-grain
varieties contained 8.7–15.4%, and the waxy rice con-
tained an average of 1.0%. The M-401, medium-grain
variety grown in CA, showed an unexpected low value of
amylose of 8.7%.
3.2 Starch Mw
The purified starches were analyzed by HPSEC-MALLS
to determine weight-average values of their molar mas-
ses (Mw) and z-average radius of gyration (Rz). SEC col-
umns with large packaging particles (20 mm) and large
pore sized frits (10 mm) were used to minimize shear deg-
radation of the amylopectins which have high molar
masses.
©2007 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim www.starch-journal.com
72 I. M. Park et al. Starch/Stärke 59 (2007) 69–77
Initial SEC chromatographs with the four Waters Styragel
HMW columns using DMSO/LiBr did not provide resolu-
tion of amylopectin and amylose (Fig. 1). The lowest Mw’s
observed from the light scattering analysis of the latter
part of the chromatograph were considerably higher than
that expected for amylose. For the long-grain variety of
Cocodrie from CA, the lowest Mwwas 1006106g/mol in
the SEC chromatogram. This was considerably higher
than previously reported values of amyloses that have
ranged from 16105to 36106g/mol [29]. As there was no
resolution of amylopectin from amylose; this would sug-
gest that the apparent high Mwvalues for the lower Mw
species of amylose may have been for a mixture of amy-
lopectin and amylose. In order to measure the Mw’s of the
amylopectins and amyloses of the rice varieties, fraction-
ation of these two components was necessary before Mw
analysis using SEC-MALLS.
For the measurement of the Mwof the rice starches, one
Plgel Mixed-A column SEC analysis with DMSO/LiBr as
the mobile phase was used to save time and solvent. This
single column provided overall measurements of Mwof
the starches which were in agreement with the four col-
umn bank. The average starch Mw’s of the long, medium/
short, and waxy rice starches were 0.786108, 1.186108,
and 1.876108g/mol, respectively. Average Rzvalues
were 189nm, 207 nm, and 224 nm, respectively for the
same grain types (Tab. 1). The Berry second-order fitting
method was used for calculating the starch Mwbecause a
small change in the slope has a dramatic effect on the
value of the y-intercept of the large polymers [16, 17, 22].
Previous studies have reported a wide range of Mwfrom
different grain sources; waxy maize was 6.5 or 596107g/
mol depending on dispersing conditions or 206107g/
mol, Mwof potato starches ranged from 2.6 to 0.26107or
5.46107g/mol, and wheat starch showed an Mwof
7.16107g/mol with different sample preparation steps
and different solvent systems by an HPSEC system [12,
13, 30].
As was expected, there was a negative correlation be-
tween the amylose content and the Mwof the rice starch-
es. The starch Mwdecreased as the amylose content
Fig. 1. SEC-chromatograph of CA-Bengal starch with
four columns showing the lack of resolution between
amylopectin and amylose. The mass was detected with
an RI detector or the molar mass was detected with a
MALLS detector.
Tab. 1. Weight-average molar mass (Mw) and z-average radius of gyration (Rz) of starch, amylopectin and amylose.
Rice variety Location Grain
type
Starch Amylopectin Amylose
Mw1
[g/mol6108]
Rz’ [nm] Mw
[g/mol6108]
Rz[nm] Mw
[g/mol6105]
Rz[nm]
Cocodrie CA L 0.59 60.01 177 61.1 1.19 60.07 252 66.2 3.96 60.98 33.4 60.50
Cocodrie TX L 0.52 60.02 177 61.0 1.12 60.01 258 68.8 2.78 60.30 33.3 60.20
L-205 CA L 1.00 60.03 194 60.9 0.82 60.22 184 613.2 5.88 60.25 36.1 61.20
L-205 TX L 0.99 60.03 209 60.2 1.24 60.56 224 68.6 3.00 60.55 32.2 60.80
Bengal CA M 1.15 60.03 210 61.7 1.65 60.35 281 623.2 3.59 60.14 34.6 61.20
Bengal TX M 1.46 60.05 209 63.3 2.32 60.44 311 612.1 3.51 60.42 34.4 61.30
M-202 CA M 1.01 60.07 202 60.5 1.45 60.15 254 60.6 3.44 60.88 35.6 60.57
M-202 TX M 1.08 60.05 198 63.9 1.64 60.15 251 61.5 3.35 60.53 35.3 60.35
M-401 CA M 1.35 60.11 215 61.4 2.13 60.50 270 61.1 4.31 60.53 36.5 60.49
M-401 TX M 1.23 60.03 207 61.5 2.05 60.08 292 64.2 4.27 60.53 33.1 61.48
Koshihikari CA S 1.05 60.03 200 63.3 1.26 60.02 240 64.3 4.16 60.01 34.7 61.78
Koshihikari TX S 1.12 60.06 206 61.5 2.04 60.17 283 612.0 3.09 60.01 33.9 62.25
S-102 CA S 1.05 60.00 205 61.1 1.65 60.22 260 61.4 3.85 60.05 34.1 63.35
S-102 TX S 1.28 60.04 216 62.0 1.73 60.11 269 63.5 3.80 60.89 32.1 63.80
Calmochi-101 CA W 1.85 60.09 222 61.5 2.50 60.37 319 610.1 N/S N/S
Calmochi-101 TX W 1.89 60.00 225 60.0 2.44 60.45 304 64.2 N/S N/S
1Average from at least two replicates 6standard deviation of each sample.
©2007 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim www.starch-journal.com
Starch/Stärke 59 (2007) 69–77 Rice Starch Molecular Size and Amylose Content 73
increased (r=20.924). Waxy rice contained 2% or less
amylose, and had the largest Mw’s of starch. Long-grain
rice had the highest amylose content and showed smal-
lest Mw. When the Mw’s of starches of short and medium
grains were statistically analyzed, the Mw’s of the TX vari-
eties were higher than those of the corresponding CA
varieties, however, the waxy and long-grain rice varieties
showed no significant difference in starch Mwbetween
growth locations (p,0.05).
3.3 Amylopectin Mw
Failing adequate SEC resolution, fractionation of amylo-
pectin from amylose was necessary before measure-
ments of the Mwof the amylopectins with SEC-MALLS
system. Fractionation of the rice starch samples was
conducted with gel permeation chromatography (GPC)
with Sepharose CL-2B. A chromatograph of normal rice
starch (Cocodrie-CA, 18.6% amylose) showed two peaks
(Fig. 2a), amylopectin (peak 1) and amylose (peak 2). In
contrast, a chromatograph for a waxy rice starch (Cal-
mochi-CA, 1.0% amylose, Fig. 2b) showed only one
peak, which corresponded to the same elution volume as
the first peak for the normal rice starch. The first fractions,
peak 1, for all the rice varieties were collected and freeze-
dried for HPSEC-MALLS analysis of the Mwand Rz. The
recovery of the peak 2 fractions provided an insufficient
amount of sample for HPSEC analysis.
From HPSEC-MALLS analysis, the average Mwvalues of
peak 1, amylopectin, were found to be 1.10 6108g/mol,
1.81 6108g/mol and 2.47 6108g/mol for the long-, short/
medium-, and waxy grain samples, respectively (Tab. 2).
The average Rzvalues were 229nm, 271nm and 311nm,
respectively for the same samples. There was a negative
correlation between amylose content and amylopectin Mw
(r=20.819). As amylose content increased the amylo-
pectin Mwdecreased. Waxy rice varieties showed the
highest Mwand long-grain varieties showed the smallest
Mw. This relationship is similar to the correlation found be-
tween the starch Mwand amylose content. Yoo and Jane
[17] reported similar results, they found an Mwof amylo-
pectin from waxy maize (8.36108g/mol) and waxy wheat
(5.26108g/mol) which was higher than that of normal
maize (4.96108g/mol) and wheat (3.16108g/mol).
There was a correlation between the Rzand Mwof amy-
lopectin (Fig. 3; slope = 0.393, r= 0.906). The value of the
slope 0.393 suggested that the shape of the dispersed
species was between a highly compact spherical and
branched form. Wyatt [11] stated that a slope less than
0.33 indicates a compact homogeneous spherical mole-
cule; whereas a slope less than 0.50 indicates a species
of a branched molecule.
Fig. 2. Sepharose CL-2B gel permeation profiles of rice
starches for (a) normal rice starches (Cocodrie-CA) (b)
waxy rice starches (Calmochi-CA).
Peak 1; amylopectin, Peak 2; amylose fraction.
Tab. 2. Amylose content of rice starches.
Rice variety Location Grain
type
Starch
Amylose1
content [%]
Average
of % AM2
Cocodrie CA L 18.6 60.44
Cocodrie TX L 19.9 60.63
L-205 CA L 17.1 60.53
L-205 TX L 17.6 60.85 18.3
Bengal CA M 13.8 61.50
Bengal TX M 11.0 60.85
M-202 CA M 14.2 60.65
M-202 TX M 15.4 60.29
M-401 CA M 8.7 60.29
M-401 TX M 10.4 60.29 12.3
Koshihikari CA S 14.6 60.35
Koshihikari TX S 12.6 60.30
S-102 CA S 15.8 60.35
S-102 TX S 13.1 6.0.21 14.0
Calmochi-101 CA W 1.0 60.28
Calmochi-101 TX W 0.9 60.03 1.0
1Determined by Concanavalin A method (Gibson et al.
1997).
2Average for each grain type with at least three repli-
cates 6standard deviation of each sample.
©2007 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim www.starch-journal.com
74 I. M. Park et al. Starch/Stärke 59 (2007) 69–77
Fig. 3. Relationships between the weight-average molar
mass (Mw) and z-average radius of gyration (Rz) of rice
starch amylopectins.
As for the starch Mw, the amylopectin Mwof short and
medium grains were statistically higher in the TX varieties
than in the CA varieties. Comparing the Mwof starch with
amylopectin shows the Mwof amylopectin is larger than
the Mwof starch as would be expected, except for the
L-205 variety grown in CA.
3.4 Amylose Mw
The fractionation of amylose from amylopectin was per-
formed by hydrolyzing the amylopectin at its branch
points with isoamylase, and then using aqueous HPSEC
analysis. Isoamylase hydrolyzes amylopectin and amy-
lose molecules at their a-(1?6) branch points, which
produces a lower Mwdistribution of linear chains. An iso-
amylase treated sample from a non-waxy rice starch
(Cocodrie-CA, 18.6% amylose) showed three peaks
(Fig. 4b); the first (FAm) was the amylose and the second
and third (F1 and F2 is assigned for second peaks, F3 is
assigned for third peak) were the amylopectin branches;
the same treatment of a waxy rice starch (Calmochi-CA,
1.0% amylose) showed only the second and third peaks
(Fig. 4a).
The Mw’s of the first peaks range from 2.786105to
5.886105g/mol among the rice varieties (Tab. 2). Takeda
et al. [31] reported that from rice starch the amylose
weight-average DPwwas 2750–3320. Suortti et al. [29]
reported the range of amylose Mwwas 0.5 to 3.06106g/
mol from various starch sources with a post-column
iodine addition detector that selectively detected amy-
lose from laser light scattering. Radosta et al. [18] report-
ed Mwof amylose as 0.1 to 366106g/mol, using a
HPSEC-MALLS system. This would suggest that peak 1
Fig. 4. The detailed RI-Chromatograms of amylopectin
fractions using isoamylase treated rice starch from (a)
Waxy (Calmochi) (b) Long-grain type (Cocodrie) grown in
CA. FAm, amylose fraction; F1 Long branch chain of
amylopectin; F2, Short branch fraction; and F3, oligo-
saccharide fraction.
was amylose from the rice starches and the Mwof these
amyloses spanned a narrower range than previous
reported amylose Mwvalues. We can hypothesize that a
25–50% number fraction of a-(1?6) branches in the
amylose molecule of rice starch can reduce the Mwof
debranched amylose compared to native amylose, a
mixture of linear and branched amylose.
There was an apparent trend of higher amylose Mw’s of
CA rice varieties compared to the corresponding TX vari-
eties but the difference was not significant. This study
also showed that the Mwof amylose was not significantly
related with amylose content (p,0.05). The results are in
good agreement with previous studies showing that dif-
ferences in molecular structure of amylose exist in differ-
ent botanical sources whereas the structure of amylose
from same botanical origin showed no significant differ-
ence [31–34].
3.5 A comparison of the Mwfrom the direct
measurement of whole starch and
calculated value of the amylopectin and
amylose fractions
In order to compare the analysis of the Mwof the starches
and those of the fractionated components, the Mwof
amylopectins and amyloses mixtures was estimated by
©2007 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim www.starch-journal.com
Starch/Stärke 59 (2007) 69–77 Rice Starch Molecular Size and Amylose Content 75
using the fractionated Mw’s (Tab. 1) and amounts of each
fraction in the various rice varieties (Tab. 2). The Mwis
defined as:
Mw¼PwiMi
Pwi
In this case Mwcan be calculated by
Mw¼ð100 AmÞMAp
wþAmMAm
w
100
where Am is the percent amylose (Tab. 1) and MwAp and
MwAm are the molar masses of amylopectin and amylose,
respectively, of the rice varieties.
The measured Mw’s of the starches by HPSEC-MALLS
were lower than the calculated values corresponding to
the AP/AM fractions with one exception for the L-205
from CA (Tab. 3). However, the differences are within the
range of reported values for Mw’s of starches.
3.6 Isoamylase treated debranched-
amylopectin
The analysis of the DPwof the amylopectin branches was
made from the SEC-MALLS chromatographs of the de-
branched starches (Fig. 4). There was a small RI peak and
large LS peak (not shown) for amylose containing sam-
ples as the first fraction to be eluted. The branches
appeared as the second and third peak. The first of these
two was separated into two fractions for further analysis.
The largest Mwfraction (F1) was defined from the begin-
ning of the elution point 32.5mL to 35.5 mL which corre-
sponded to an inflection of the multicomponent peak. The
location of the inflection point of the shoulder was
assumed constant at 35.5 mL for all samples. The second
fraction (F2) was taken from 35.5 mL to the minimum point
of the later two components, 38.5 mL. The third fraction
(F3) was defined from this minimum point 38.5 mL to
41 mL which corresponded to the third peak (Fig. 4). From
the light scattering analysis for FAm (amylose peak), F1,
F2, and F3, the DPwlimits of each fraction were 18518–
303, 303–36, 36–7, and 7–1, respectively. The DPwvalues
of 7 and 1 were determined by extrapolating the light
scattering Mwdata to lower Mwas light scattering is not
accurate in this range.
The average DPw’s of F1 fractions were 76, 68 and 60 for
long, medium/short, and waxy, respectively (Tab. 4). The
correlation coefficient between the amylose content and
the Mwor DPwof F1 was 0.942, which is consistent with
previous studies [21, 34]. This suggests that the average
chains in this fraction of amylopectin branches from rice
varieties of high amylose contents have longer amylo-
Tab. 3. Weight-average molar masses of the rice vari-
eties from direct measurement of rice starch and
the predicted values from a mixture of amylo-
pectin and amylose.
Rice Variety Location MwStarcha
[106g/mol]
Mwcombined
AP and AMb
[106g/mol]
Cocodrie CA 59.0 96.9
TX 52.0 89.8
L-205 CA 100.0 68.1
TX 99.0 102.2
Bengal CA 115.0 142.3
TX 146.0 206.5
M-202 CA 101.0 124.5
TX 108.0 138.8
M-401 CA 135.0 194.5
TX 123.0 183.7
Koshihikari CA 105.0 107.7
TX 112.0 178.3
S-102 CA 105.0 139.0
TX 128.0 150.4
Calmochi-101 CA 185.0 250.0
TX 189.0 244.0
aThe measured value of Mwof the starch from Tab. 2
bThe calculated value of Mwfrom the Mw’s of amylo-
pectin (AP) and amylose (AM) values
from Tab. 2 using Mw¼ð100 AmÞMAp
wþAmMAm
w
100
pectin branches than the amylopectin branches of low
amylose rice varieties. From the DPwvalues, the F1
could be categorized into three groups; waxy, short and
medium, and long grain (p,0.05), which are the same
groups as those for the amylose content and the starch
Mwand AP Mw. The average DPwof F2 were within the
same range of 20–22 (Tab. 4), which did not show any
significant differences among grain types and growing
locations (p,0.05) as well as no relationship with
amylose content. The F2 chains are believed to be
confined within a single cluster. The similarity of the F2
DPwvalues among all the rice varieties suggests that
the structure of single clusters among the varieties were
similar. The mass of the F1 fractions was 16–18% and
F2 was 82–84%, and did not vary among varieties and
locations (Tab. 4). The F2 branches make up the double
helices within cluster, which are regularly packed into
crystalline structure and would relate to crystalline
thickness [3, 35].
As the amylose content increased, the F1 DPwof the
debranched amylopectin increased as well, and the Mw
amylopectin decreased.
©2007 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim www.starch-journal.com
76 I. M. Park et al. Starch/Stärke 59 (2007) 69–77
Tab. 4. Distribution of mass concentration, molar percentage and branch chain weight molar mass (Mw) and degree of
polymerization (DPw) of isoamylase treated rice starch.
Rice variety Location Grain
type
Mass concentration Molar percentage F12F23
F11F2’ F1 F2
M
wDPw4Ave5
M
wDPw4Ave5
Cocodrie CA L 0.947 4.367 17.8 82.2 1.27 60.10578 35.02 60.00 22
Cocodrie TX L 0.957 4.682 17.0 83.0 1.24 60.12 77 34.22 60.01 21
L-205 CA L 0.938 4.455 17.4 82.6 1.16 60.08 72 34.32 6001 21
L-205 TX L 0.938 4.357 17.7 82.3 1.25 60.10 77 76 34.29 60.00 22 21
Bengal CA M 0.989 4.525 17.9 82.1 1.10 60.10 68 33.82 60.01 21
Bengal TX M 0.975 4.677 17.3 82.7 1.06 60.08 65 33.78 60.01 21
M-202 CA M 0.981 4.894 16.7 83.3 1.19 60.10 74 34.75 60.01 21
M-202 TX M 1.087 4.876 18.2 81.8 1.13 60.11 70 34.35 60.02 21
M-401 CA M 0.999 4.813 17.2 82.8 1.06 60.03 65 32.32 60.05 20
M-401 TX M 0.990 4.792 17.1 82.9 1.01 60.07 62 67 33.45 60.00 21 21
Koshihikari CA S 0.976 4.835 16.8 83.2 1.14 60.11 71 35.42 60.03 22
Koshihikari TX S 0.997 4.856 17.0 83.0 1.06 60.07 65 33.31 60.02 21
S-102 CA S 0.985 4.767 17.1 82.9 1.18 60.09 73 35.02 60.00 22
S-102 TX S 0.955 4.544 17.4 82.6 1.11 60.10 69 68 33.54 60.01 21 21
Calmochi-101 CA W 1.133 5.167 18.0 82.0 0.95 60.04 59 32.82 60.01 20
Calmochi-101 TX W 1.136 5.098 18.2 81.8 0.98 60.04 60 60 33.45 60.01 21 20
1mass concentration of F1[61024g]; 2mass concentration of F2[61024g]: Each value was calculated.
2First fraction of isoamylase treated amylopectin [Mw:6104].
3Second fraction of isoamylase treated amylopectin [Mw:6102].
4Degree of polymerization.
5Average from each grain type with at least two replicates of each sample 6standard deviation of each sample.
4 Conclusions
The short- and medium-grain rices, which were grown in
CA contained a higher level of amylose than those grown
in TX, comparing the different cultivation locations (CA
and TX). Also, those samples grown in CA had a lower
starch Mwand amylopectin Mwin short- and medium-
grain varieties than those samples grown in TX (p,0.05).
The combined information above gives insight into rice
starch crystalline structure. As amylose content
increased, chain length and F1 (DPw) of debranched
amylopectin was higher, on the contrary Mwof starch and
amylopectin decreased. This study showed that samples
with higher amylose contents had larger amounts of long
chains of amylopectin compared to the low amylose or
waxy samples. Also, higher amylose content rice starch
samples contained longer branches of amylopectin.
Based on our study, the different amount and size of long-
branch chain of amylopectin had an influence on com-
pactness of starch structure. Furthermore it may also
have an effect on the Mwof the rice starch and AP. From
this study, the amylose content was closely correlated to
the DPwof long-branch chain fraction (F1) of amylopectin,
Mwof starch and AP but not Mwof amylose and short-
branch chain fraction (F2) of amylopectin.
Acknowledgements
We gratefully acknowledge the financial support from The
Rice Foundation. We also give thanks to the Dr. K.
McKenzie and Ms A. Noble, the California Rice Experi-
mental Station, and the USDA-ARS Rice Quality Labora-
tory, Beaumont, TX, for providing the rice samples and
advice.
References
[1] B. O. Juliano: Polysaccharides, proteins, and lipids of rice,
American Association of Cereal Chemists, Inc., St. Paul, MN,
1985.
[2] E. T. Champagne: Rice starch composition and character-
istics. Cereal Food World.1996,41, 833–838.
[3] J. Jane, Y. Y. Chen, L. F. Lee, A. E. McPherson, K. S. Wong, M.
Radosavljevic, T. Kasemsuwan: Effects of amylopectin
branch chain length and amylose content on the gelatiniza-
tion and pasting properties of starch. Cereal Chem. 1999,76,
629–637.
[4] B. O. Juliano: Structure, chemistry, and function of the rice
grain and its fractions. Cereal Food World 1992,37, 772.
[5] M. Ramesh, S. Z. Ali, K. R. Bhattacharya: Structure of rice
starch and its relation to cooked-rice texture. Carbohydr.
Polym. 1999,38, 337–347.
©2007 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim www.starch-journal.com
Starch/Stärke 59 (2007) 69–77 Rice Starch Molecular Size and Amylose Content 77
[6] J. N. BeMiller, R. L. Whistler: Carbohydrate, in Food Chem-
istry (Ed. O. R. Fennema) Marcel Dekker, Inc, New York,
1996.
[7] G. C. Berry: Thermodynamic and conformational properties
of polystyrene. I. Light-scattering studies on dilute solutions
of linear polystyrenes. J. Chem. Phys. 1966,44, 4550–4564.
[8] C. G. Biliaderis, B. O. Juliano, Thermal and mechanical-
properties of concentrated rice starch gels of varying com-
position. Food Chem. 1993,48, 243–250.
[9] A. G. W. Bradbury, A. B. T. Bello: Determination of molecular-
size distribution of starch and debranched starch by a single
procedure using high-performance size-exclusion chroma-
tography. Cereal Chem. 1993,70, 543–547.
[10] F. S. Frederick: Rice proteins, in Rice: Chemistry and Tech-
nology (Ed. E. T. Champagne) American Association of Cer-
eal Chemists, St. Paul, MN, 2004.
[11] P. J. Wyatt: Light-scattering and the absolute characteriza-
tion of macromolecules. Anal. Chim. Acta 1993,272, 1–40.
[12] L. A. Bello-Perez, P. Roger, B. Baud, P. Colonna: Macro-
molecular features of starches determined by aqueous high-
performance size exclusion chromatography. J. Cereal Sci.
1998,27, 267–278.
[13] M. L. Fishman, L. Rodriguez, H. K. Chau: Molar masses and
sizes of starches by high-performance size-exclusion chro-
matography with on-line multi-angle laser light scattering
detection. J. Agric. Food Chem. 1996,44, 3182–3188.
[14] J. P. Mua, D. S. Jackson, Fine structure of corn amylose and
amylopectin fractions with various molecular weights. J.
Agric. Food Chem. 1997,45, 3840–3847.
[15] M. H. Ong, K. Jumel, P. F. Tokarczuk, J. M. V. Blanshard, S. E.
Harding: Simultaneous determinations of the molecular-
weight distributions of amyloses and the fine-structures of
amylopectins of native starches. Carbohydr. Res. 1994,260,
99–117.
[16] W. Yokoyama, J. J. Renner-Nantz, C. F. Shoemaker: Starch
molecular mass and size by size-exclusion chromatography
in DMSO-LiBr coupled with multiple angle laser light scat-
tering. Cereal Chem. 1998,75, 530–535.
[17] S. H. Yoo, J. L. Jane: Molecular weights and gyration radii of
amylopectins determined by high-performance size-exclu-
sion chromatography equipped with multi-angle laser-light
scattering and refractive index detectors. Carbohydr. Polym.
2002,49, 307–314.
[18] S. Radosta, M. Haberer, W. Vorwerg: Molecular character-
istics of amylose and starch in dimethyl sulfoxide. Bioma-
cromol. 2001,2, 970–978.
[19] P. Roger, P. Colonna: Molecular weight distribution of amy-
lose fractions obtained by aqueous leaching of corn starch.
Int. J. Biol. Macromol. 1996,19, 51–61.
[20] C. M. Durrani, A. M. Donald: Shape, molecular weight dis-
tribution and viscosity of amylopectin in dilute solution.
Carbohydr. Polym. 2000,41, 207–217.
[21] N. W. H. Cheetham, L. P. Tao: The effects of amylose content
on the molecular size of amylose, and on the distribution of
amylopectin chain length in maize starches. Carbohydr.
Polym. 1997,33, 251–261.
[22] S. H. Yoo, J. L. Jane: Structural and physical characteristics
of waxy and other wheat starches. Carbohydr. Polym. 2002,
49, 297–305.
[23] I. Hanashiro, J. Abe, S. Hizukuri: A periodic distribution of
the chain length of amylopectin as revealed by high-perfor-
mance anion-exchange chromatography. Carbohydr. Res.
1996,283, 151–159.
[24] T. S. Gibson, V. A. Solah, B. V. McCleary: A procedure to
measure amylose in cereal starches and flours with con-
canavalin A. J. Cereal Sci. 1997,25, 111–119.
[25] X. Z. Han, B. R. Hamaker: Amylopectin fine structure and
rice starch paste breakdown. J. Cereal Sci. 2001,34, 279–
284.
[26] S. H. Yun, N. K. Matheson: Estimationof amylose content of
starches after precipitation of amylopectin by concanavalin-
A. Starch/Stärke 1990,42, 302–305.
[27] K. E. J. Ward, R. C. Hoseney, P. A. Seib: Retrogradation of
amylopectin from maize and wheat starches. Cereal Chem.
1994,71, 150–155.
[28] P. Debye: Molecular-weight determination by light scatter-
ing. J. Phys. Colloid Chem. 1947,51, 18–32.
[29] T. Suortti, M. V. Gorenstein, P. Roger: Determination of the
molecular mass of amylose. J. Chromatogr. A 1998,828,
515–521.
[30] M. M. Millard, F. R. Dintzis, J. L. Willett, J. A. Klavons, Light-
scattering molecular weights and intrinsic viscosities of
processed waxy maize starches in 90% dimethyl sulfoxide
and H2O. Cereal Chem. 1997,74, 687–691.
[31] Y. Takeda, S. Hizukuri, B. O. Juliano: Purification and struc-
ture of amylose from rice starch. Carbohydr. Res. 1986,148,
299–308.
[32] P. A. Seib: Wheat starch: isolation, structure and properties.
Oyo Toshitsu Kagaku 1994,41, 49–69.
[33] C. Takeda, Y. Takeda, S. Hizukuri: Structure of amylomaize
amylose. Cereal Chem. 1989,66 (1), 22–25.
[34] Y. Takeda, S. Hizukuri, C. Takeda, A. Suzuki: Structures of
branched molecules of amyloses of various origins, and
molar fractions of branched and unbranched molecules.
Carbohydr. Res. 1987,165, 139–145.
[35] A. Buleon, P. Colonna, V. Planchot, S. Ball: Starch granules:
structure and biosynthesis. Int. J. Biol. Macromol. 1998,23,
85–112.
(Received: October 12, 2006)
(Revised: December 18, 2006)
(Accepted: December 18, 2006)
©2007 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim www.starch-journal.com
