Use of Enzymes for the Separation of Protein from Rice Flour

Abstract

Cereal Chem. 74(4):437–441 When rice flour was treated with heat stable α-amylases, the effec-tiveness of protein separation increased with increased temperature. Depending on the enzyme, treatment at 90°C for 45 min resulted in pro-tein contents of 47–65% for the insoluble fraction. Prior gelatinization enhanced the effectiveness of the enzyme reaction but was undesirable because the increased viscosity and gelation could cause difficulties in the processing operation. Follow-up treatment with other carbohydrate-hydrolyzing enzymes, such as glucoamylase, cellulase, and hemicellulase further increased the protein content up to 76% for the insoluble fraction. The subunit structure of the isolated proteins, based on electrophoretic analysis, remained practically unchanged after the treatment. The limited solubility and emulsion activity of rice protein were also unchanged.

Full text

Vol. 74, No. 4, 1997 437
MISCELLANEOUS
Use of Enzymes for the Separation of Protein from Rice Flour
FREDERICK F. SHIH1,2 and KIM DAIGLE1
ABSTRACT Cereal Chem. 74(4):437–441
When rice flour was treated with heat stable α-amylases, the effec-
tiveness of protein separation increased with increased temperature.
Depending on the enzyme, treatment at 90°C for 45 min resulted in pro-
tein contents of 47–65% for the insoluble fraction. Prior gelatinization
enhanced the effectiveness of the enzyme reaction but was undesirable
because the increased viscosity and gelation could cause difficulties in
the processing operation. Follow-up treatment with other carbohydrate-
hydrolyzing enzymes, such as glucoamylase, cellulase, and hemicellulase
further increased the protein content up to 76% for the insoluble fraction.
The subunit structure of the isolated proteins, based on electrophoretic
analysis, remained practically unchanged after the treatment. The limited
solubility and emulsion activity of rice protein were also unchanged.
Rice is the number one food crop in the world. It is also nutri-
tious and hypoallergenic, which makes rice products desirable
food ingredients. Recently, methods have been developed to sepa-
rate, modify, and utilize rice and its components for value-added
products. Morita and Kiriyama (1993) partially solubilized rice
flour with α-amylase to prepare a protein-rich product with >90%
protein. Using rice flour obtained at 70–35% milling, the treat-
ment was conducted at 97°C for 2 hr, and the product was obtained
by filtering off the solubles with boiling water. Similar enzymatic
methods have been used to produce protein-enriched rice flours
(Hansen et al 1981, Chang et al 1986, Griffin et al 1989). Protein
content of the products varied depending on the rice flour and the
processing conditions.
Of the rice components, rice protein is valuable because it is
hypoallergenic and rich in the essential amino acid lysine. The
lysine content of 3–4% (≈50% higher than that of wheat) is among
the highest in cereal proteins. However, the protein fraction,
which is 6–9% of milled rice, compared to starch at 80–85%, is
generally a by-product during the starch conversion. Only limited
information is available regarding its separation and characteriza-
tion. There is a need to study and improve the quality and quantity
of rice protein as a by-product in the processing of starch.
The present study focused on the separation of rice protein
during treatment of rice flour with carbohydrate-hydrolyzing enzymes.
The products were characterized by chemical, physical, and func-
tional properties.
MATERIALS AND METHODS
Materials
Long-grain rice flour RL100 (composition shown in Table I)
was donated by Rivland Partnership (Houston, TX). Particle-size
distributions of the flour are: 95–100% (going through 50 mesh)
55–75% (100 mesh), and 45–60% (140 mesh). Rice Protein Con-
centrate K, a reference rice protein sample (≈50% protein and
39% carbohydrate), was obtained from California Natural Prod-
ucts (Lathrop, CA). The carbohydrate-hydrolyzing enzymes Taka-
Therm L340, Diazyme L200, Cellulase TR, Hemicellulase Con-
centrate, and Cellulase AC were obtained from Solvay Enzymes
(Elkhart, IN), and Termamyl 120L was obtained from Novo
Nordisk (Danbury, CT). The proteolytic enzymes Papain 300 and
Protease 2A were obtained from International Enzymes (Troy,
VA), Pronase E was obtained from Sigma Chemical (St. Louis,
MO), and APL-440 was obtained from Solvay Enzymes (Elkhart,
IN). All other chemicals used were reagent-grade.
α-Amylase Treatment
Rice flour samples (40 g) were slurried with ≈150 mL of
deionized water, adjusted to pH 6.5, mixed with calcium chloride
(0.111 g), and diluted to the final processing weight of 200 g with
deionized water. The mixture was mechanically stirred in a flask
placed on a hot plate with automatic temperature control. Immedi-
ately before processing, 0.1% E-S (enzyme-substrate) Termamyl
120L or 0.15% E-S Taka-Therm L340 was added. The mixture
was heated quickly in ≈2 min to the desired temperature. In one
experiment using the Taka-Therm L340, the flour slurry was
gelatinized at 90°C for 30 min before the addition of the enzyme.
For all experiments, after incubation with the enzyme for the
desired duration, the mixture was adjusted to pH 4 and boiled for
5 min to inactivate the α-amylase. The mixture was then ready for
product processing and analysis or for additional enzyme treat-
ment.
For product processing and analysis, the mixture adjusted to pH 4
was diluted with water to 600 mL and then centrifuged at 9,000 × g
for 15 min at 4°C to separate the residue and the supernatant. The
residue, after being washed three times with 600 mL of water at
pH 4 and then dispersed in water again, was recovered by lyophil-
ization as the protein-rich fraction.
Additional Enzyme Treatment
After enzyme deactivation at the end of the α-amylase treat-
ment, the mixture was adjusted to pH 4 and 55°C. Selected carbo-
hydrate-hydrolyzing enzymes were then added to the mixture at
E-S ratios of: 0.18% for Diazyme L200, 0.5% for Cellulase TR,
0.2% for Cellulase AC, and 0.5% for Hemicellulase. After incu-
bation for 2 hr, the reaction products were processed and analyzed
as described earlier.
1Southern Regional Research Center, PO Box 19687, New Orleans, LA 70179.
2Corresponding author. E-mail: fshih@nola.srrc.usda.gov
Publication no. C-1997-0616-02R.
This article is in the public domain and not copyrightable. It may be freely re-
printed with customary crediting of the source. American Association of Cereal
Chemists, Inc., 1997.
TABLE I
Composition of Rice Flour
Composition g/100 g of flour
Protein (N × 5.95)a8.1 ± 0.8
Carbohydratea80.9 ± 1.9
Moisturea10.1 ± 1.8
Crude Fiberb0.6
Fatb0.4
Ashb0.6
aLaboratory-analyzed data with means of triplicates ± standard deviation.
bData from the manufacturer.

438 CEREAL CHEMISTRY
Proteolysis
Proteolytic enzymes (0.05%, E-S), APL-440, Papain 300, Pro-
nase E, or Protease 2A were added to a dispersion of protein-rich
fraction (4 g) in deionized water (80 mL). The mixture was incu-
bated for 4 hr at 50°C and neutral pH, before the enzyme was
inactivated by boiling for 5 min. After cooling to room tempera-
ture and adjusting to pH 4, the supernatant and residue were sepa-
rated by centrifugation (9,000 × g), washed, and recovered by
lyophilization.
Protein Analysis
Nitrogen content of the sample was determined using a nitrogen
analyzer (FP-428, LECO Corp., St. Joseph, MI). Protein content
in rice products was calculated as N × 5.95. Solubility test for the
protein-rich fraction was conducted using a 0.1% dispersion of the
sample adjusted to pH 2–9. After stirring for 30 min, the super-
natant was analyzed for dissolved protein (Protein Microassay
Method 3, Bio-Rad Labs, Hercules, CA). Emulsification activity
index (EAI), expressed as interfacial area/unit weight protein
(m2/g), was assessed by the turbidimetric method of Pearce and
Kinsella (1978). Subunits of the protein were analyzed by SDS-
PAGE according to Laemmli (1970). The Coomassie blue R-250
stained gels (4–20% gradient) were analyzed by densitometry and
image analysis (GDS2000 Gel Documentation System, UVP, San
Gabriel, CA).
Carbohydrate Analysis
Carbohydrate content was measured using a glucose standard
according to Dubois et al (1956). When referring to starch, the
value was multiplied by 0.9 to convert to a starch base (McCready
et al 1950). Liquefied starch and carbohydrate were calculated as
the ratio of carbohydrate found in the liquefied fraction to the total
carbohydrate of the flour. Reducing sugar was determined using a
glucose standard (AOAC 1990). Dextrose equivalent (DE) of a
converted starch was calculated as a percentage of dextrose to the
total dry substance.
Statistical Analysis
The samples were analyzed in triplicate. Data were assessed by
the one way analysis of variance (ANOVA) and Duncan’s multi-
ple range test (P < 0.05) (software ver. 6.10, SAS Institute Inc.,
Cary NC).
RESULTS AND DISCUSSION
Effect of Temperature
The α-amylases used are heat-stable enzymes with optimum
reaction temperatures up to 95°C. The effect of temperature using
Termamyl 120L on the separation of the rice protein in terms of
product composition is shown in Table II. After treatment with the
enzyme for 1 hr at 50°C, the protein content of the protein-rich
fraction was 16.2%, increased from the 9.0% (dwb) of the intact
flour. Treatment at temperatures ≥70°C were more effective. The
effectiveness of processing at 90°C was particularly pronounced,
with protein content of the protein-rich fraction up to 65%.
The effectiveness of protein separation by the α-amylase treat-
ment at high temperatures, and the lack of it at low temperatures
are believed to be related to starch gelatinization (Brooks and
Griffin 1987). Gelatinized starch is more accessible to the enzyme.
As rice starch gelatinizes at 70–75°C, a greater concentration of
gelatinized starch would be available in or above this temperature
range, resulting in more effective separation of the rice flour.
Effect of Prior Gelatinization
The Taka-Therm L340 treatment was conducted with and with-
out prior gelatinization of the rice flour. The data are shown in
Table III (Treatments Bc and B, respectively). With prior gelatini-
zation, the protein-rich fraction had a slightly higher protein con-
tent. However, the treatment without prior gelatinization was advan-
tageous because it was a one-step operation that saved time and
simplified the process by avoiding the formation of a hard-to-stir
gelling mass from the gelatinization. Therefore, unless otherwise
indicated, all enzymatic treatment of the rice flour was conducted
without prior gelatinization.
TABLE II
Effect of Temperature and Reaction Time on Product Composition for
the Treatment of Rice Flour with Termamyl 120L
Temperature (°C) Time (min) Proteina (%) Carbohydratea (%)
50 15 11.7h 58.0i
30 12.2h 54.2k
60 16.2g 53.4k
70 15 18.3f 55.9j
30 24.4e 53.5k
90 15 45.8d 39.8l
30 64.0c 24.1m
45 65.5b 19.4n
aValues are dwb. Means of three determinations. Values followed by the
same letter are not significantly different (P < 0.05).
TABLE III
Composition of Protein-Rich Fraction in the Treatment of Rice Flour
with Different Starch-Hydrolyzing Enzymes
TreatmentaProteinb (%) Carbohydrateb (%)
(A) 65.5g 19.4q
(B) 46.9k 38.6l
(B)c51.1j 35.1m
(B)c + Dz 56.0i 32.6n
(A) + TR 71.8f 9.5r
(A) + AC 75.6e 8.5r
(A) + HC 71.6f 8.8r
(A) + (AC & HC) 76.4e 9.2r
(B) + TR 62.2h 21.8p
(B) + AC 60.3h 29.4o
(B) + HC 55.2i 29.2o
aEnzyme treatments (A) and (B) with α-amylase Termamyl 120L and Taka-
Therm L340, respectively, were conducted at 90°C for 45 min. In
multienzyme treatments, the α-amylase process was followed by treatment
with the additional enzyme DZ (Diazyme L200), TR (Cellulase TR), AC
(Cellulase AC), and HC (Hemicellulose Concentrate) at 55°C for 2 hr.
bValues are dwb. Means of three determinations.Values followed by the
same letter are not significantly different (P < 0.05).
cWith prior gelatinization of the rice flour.
TABLE IV
Effect of Reaction Conditions on the Degradation and Liquefaction of
the Insoluble Fraction
TreatmentaTemp.
(°C) Time
(min) Dextrose
Equivalent Carbohydrate
Liquefiedb (%)
(A) 50 15 4.7o 23.4m
(A) 50 30 7.8n 26.6l
(A) 50 60 11.2l 43.9k
(A) 70 15 9.6m 57.4j
(A) 70 30 12.5k 73.2g
(A) 90 15 9.7m 60.2i
(A) 90 30 13.8j 78.0f
(A) 90 45 23.5h 77.7f
(A) + TR 90 45 35.5f 79.7e
(B) 90 45 20.3i 65.4h
(B)c90 45 30.1g 72.7g
(B)c + DZ 90 45 60.1e 77.9f
aSingle enzyme treatments (A) and (B) with α-amylase Termamyl 120L and
Taka-Therm L340, respectively, were conducted under the temperature and
time conditions indicated. In multienzyme treatments, the α-amylase
process was followed by treatment with the enzyme TR (Cellulase TR) or
DZ (Diazyme L200) at 55°C for 2 hr.
bValues are dwb. Means of three determinations.Values followed by the
same letter are not significantly different (P < 0.05).
cWith prior gelatinization of the rice flour.

Vol. 74, No. 4, 1997 439
Multienzyme Treatment
The combination of α-amylase and glucoamylase is particularly
effective in the hydrolysis of starch (Chen and Chang 1984, Brook
and Griffin 1987). As expected, treatment with the glucoamylase
Diazyme L200 after the α-amylosis with Taka-Term L340
(Treatment Bc + Dz, Table III) further enhanced the protein sepa-
ration. However, the increases were relatively small. Even with a
rice flour pretreated for gelatinization, the enzyme treatment pro-
duced a relatively low protein content at 56% for the protein-rich
fraction.
The combined treatments with α-amylase and glucoamylase
may be effective in the saccharification of starch and, therefore,
useful for syrup processing, but the process is not particularly
effective in the separation of protein from other components such
as the insoluble fiber. To enhance the separation, therefore, enzymes
such as cellulase and hemicellulase were added after the α-amylosis.
The results are also shown in Table III.
With the additional enzyme treatment, the protein-rich fraction
had an increase in protein and a decrease in carbohydrate. Cellulase
seemed to be more effective than hemicellulase. Of the cellulases
used, Cellulase AC was more effective than Cellulase TR. For
protein concentration in the protein-rich fraction, the best results
were achieved at a protein content of 76% with ≈9% carbohydrate
using either Cellulase AC or a mixture of Cellulase AC and
Hemicellulase Concentrate after the Termamyl 120L treatment.
Apparently, substantial amounts of fibrous material remained
attached to the protein in the insoluble fraction. The treatment
with additional enzymes only partially hydrolyzed and removed
the cellulose and hemicellulose components from the insoluble
fraction.
Food-grade cellulase and hemicellulase systems often contain
small amounts of proteolytic enzymes that could cause partial
solubilization of the rice protein. However, as the protein content
in the liquefied fraction remained low (<1.8%), the proteolytic
activity appeared to be minimal.
Protein-Rich Fraction
Samples of the protein-rich fraction were analyzed for subunit
composition. The electrophoretic profiles for selected protein-
enriched samples, including one from an industrial source, are
shown in Fig. 1. Subunits of the protein and the profile intensities
were practically identical for all laboratory-prepared samples,
indicating that the protein remained mostly intact after the treat-
ments. Less intense were the high molecular weight subunits for
the industrial sample, indicating protein degradation during
industrial processing.
Rice proteins are known to have limited solubility in water.
Treatment with heat-stable enzymes at temperature up to 90°C
could cause denaturation and cross-linking and further decrease
the solubility of the protein. As a result, proteins thus isolated
remained mostly insoluble. Fig. 2 shows the pH-solubility profiles
for selected samples of the protein-rich fraction, including one
from an industrial source as reference. Typically, the rice proteins
were more soluble in alkaline or acidic conditions with <3 µg/mL
Fig. 1. Electrophoretic profiles for selected rice protein samples from
treatments as shown in Table III. Column assignments are: A = Bc + DZ;
B = B; C = B + TR; D = A; E = A + AC; F = A + HC & AC; G = rice
protein from industrial source; H = molecular weight markers a–f
(97,400; 66,200; 45,000; 31,000; 21,500; and 14,400, respectively).
Fig. 2. Solubility of rice proteins as a function of pH. Samples of protein-
rich fractions are chosen from treatments as shown in Table III. Protein
from Treatment A ( ); protein from Treatment A + HC ( ); protein from
A + AC ( ); protein from Treatment A + AC & HC (l); and protein
reference from industrial source (∆).
Fig. 3. Emulsifying activity index (EAI) of rice proteins as a function of
pH. Samples of protein-rich fractions are chosen from treatments as
shown in Table III. Protein from Treatment A ( ); protein from
Treatment A + HC ( ); protein from A + AC ( ); protein from
Treatment A + AC & HC (l); and protein reference from industrial
source (
∆).

440 CEREAL CHEMISTRY
solubility at the isoelectric point (pH 4.6). The profiles for EAI, as
a function of pH, are shown in Fig. 3. Again, they follow the same
pattern for all samples. As the ability to emulsify oil for a protein
is normally closely related to solubility, the values of EAI for the
samples increased with increased pH and, at a given pH, they
varied only slightly among samples.
Effect of Proteolysis
The effect of proteolysis on the structure and solubility of the
rice protein was also investigated. The protein-rich fraction from
the Termamyl 120L treatment was analyzed by electrophoresis
before and after additional treatments with various food-grade
proteases. The scanning profiles of the subunits are shown in Fig. 4.
The modification affected mostly subunits with molecular weights
ranging from 20,000 to 45,000. With the treatment of APL-440,
for instance, the intensity of bands within this range of molecular
weights decreased, as compared with the control, whereas bands
with molecular weights <15,000 increased. Depending on the
protease, the changes ranged from small for Papain 300 to
relatively more extensive for Pronase E. Accordingly, solubility of
the protein increased with increased proteolysis from <1% for the
untreated control to >15% for the Pronase E treatment (Fig. 5).
Liquefied Carbohydrate Fraction
The soluble fraction of the rice flour after treatment with carbo-
hydrate-hydrolyzing enzymes contained mostly maltodextrins.
The extent of carbohydrate degradation, indicated by the reducing
value or DE, increased with increased hydrolysis. Table IV shows
the effect of reaction conditions on starch liquefaction and degra-
dation. At 50°C, liquefaction using the enzyme Termamyl 120L
was ineffective. Less than 44% of the starch and carbohydrate was
liquefied after 1 hr. Temperatures >70°C were required to achieve
extensive (>50%) starch and carbohydrate liquefaction. Depend-
ing on the enzyme, temperature, and time, carbohydrate degrada-
tion was relatively extensive with DE values up to 60 and was
relatively small with DE values down to 5. Normally, in the sepa-
ration process, extensive starch and carbohydrate hydrolysis
enhances protein isolation. However, there is a limit of ≈76%
protein enrichment during starch and carbohydrate liquefaction.
Notably, in the process involving glucoamylase, in spite of its
markedly high level of starch degradation a DE value of 60, the
protein fraction contained only 56% protein, with the balance con-
sisting of mostly insoluble fibers.
For uses other than syrup manufacture, excessively hydrolyzed
starch with DE values >15 is undesirable sometimes because of
reduced food-use functionality such as the film-forming proper-
ties. As can be seen in Table IV, effective liquefaction with low
degradation was achieved under various conditions. For instance,
the treatment with Termamyl 120L at 90°C for 30 min produced
78% liquefied carbohydrate with a DE value of 14. Thus, in addi-
tion to the highly protein-enriched insoluble product discussed
earlier, the process also generated a soluble product containing
maltodextrins with relatively low DE values. A longer reaction
time (45 min) for the same treatment resulted in a much higher
DE value of 24 for the liquefied starch, but the same protein con-
tent at 65% for the residue.
CONCLUSIONS
Protein-enriched rice product with 65% protein can be produced
by reacting regular rice flour with the α-amylase Termamyl 120L
at 90°C for 30 min. Physicochemical and functional properties of
the isolated protein remain practically unchanged. The treatment
also liquefies ≈78% of the rice carbohydrates with limited degra-
dation as indicated by the DE value of 14. Additional treatments
with carbohydrate-hydrolyzing enzymes can raise the protein
content of the protein-rich fraction to 70% and higher, but also
increase degradation of the liquefied starch to DE values >15. The
treatments are effective in hydrolyzing the starch component but
less effective in hydrolyzing the cellulose and hemicellulose. Con-
sequently, the insoluble fraction contains substantial amounts of
insoluble fibers in addition to the protein.
Fig. 4. Scanning electrophoretic profiles for rice protein after proteolysis
with no enzyme (A), APL-440 (B), Papain 300 (C), Pronase E (D), and
Protease 2A (E). The substrate was the protein-rich fraction from the
treatment of rice flour with Termamyl 120L (Treatment A, Table III).
Molecular weight markers a–f are: 97,400; 66,200; 45,000; 31,000;
21,500; and 14,400, respectively.
Fig. 5. Solubility of rice protein after proteolysis with no enzyme (A),
APL-440 (B), Papain 300 (C), Pronase E (D), and Protease 2A (E). The
substrate was the protein-rich fraction from the treatment of rice flour
with Termamyl 120L (Treatment A, Table III).

Vol. 74, No. 4, 1997 441
LITERATURE CITED
AOAC. 1990. Official Methods of Analysis, of the Association of Offi-
cial Analytical Chemists, Vol. II, 15th ed. Method 935.62: Glucose in
Sugars and Syrups, Chemical Methods. The Association: Arlington, VA.
Brooks, J. R., and Griffin, V. K. 1987. Liquefaction of rice starch from
milled rice flour using heat-stable α-amylase. J. Food Sci. 52:712-714.
Chen, W. P., and Chang, Y. C. 1984. Production of high-fructose rice
syrup and high-protein rice flour from broken rice. J. Sci. Food Agric.
35:1128-1135.
Dubois, M., Gilles, K. A., Hamilton, J. K., Rebers, P. A., and Smith, F.
1956. Calorimetric Method for Determination of Sugar and Related
Substance. Anal. Chem. 28:350-356.
Hansen, L. P., Hosek, R., Callan, M., and Jones, F. T. 1981. The devel-
opment of high-protein rice flour for early childhood feeding. Food
Technol. 35:38-42.
Laemmli, U. K. 1970. Cleavage of structural proteins during the assem-
bly of the head of bacteriophage T4. Nature 227:680-85.
McCready, R. M., Guggolz, J., Silviera, V., and Owens, H. S. 1950.
Determination of starch and amylose in vegetables. Anal. Chem.
22:1156-1158.
Morita, T., and Kiriyama, S. 1993. Mass production method for rice pro-
tein isolate and nutritional evaluation. J. Food Sci. 58:1393-1396.
Pearce, K. N., and Kinsella, J. E. 1978. Emulsifying properties of pro-
teins: Evaluation of a turbidimetric technique. J. Agric. Food Chem.
26:716-723.
[Received September 16, 1996. Accepted March 14, 1997.]