Antioxidant activities of major components of γ-oryzanol from rice bran using a linoleic acid model

Article abstract of DOI:10.1007/s11746-001-0320-1

The change in hydroperoxides of linoleic acid incubated with constant micro air flow at 37°C was used to evaluate the antioxidant activities of three major components of γ-oryzanol from rice bran (cycloartenyl ferulate, 24-methylene cycloartanyl ferulate, and campesteryl ferulate) compared with α-tocopherol and ferulic acid. The four hydroperoxide isomers of linoleic acid, 9-hydroperoxy-10-trans, 12-cis-octadecadienoic acid [9HPODE(t,c)], 9-hydroperoxy-10-trans,12-trans-octadecadienoic acid, 13-hydroperoxy-9-cis, 11-trans-octadecadienoic acid [13HPODE(c,t)], and 13-hydroperoxy-9-trans, 11-trans-octadecadienoic acid, were measured using normal-phase high-performance liquid chromatography with an ultraviolet detector. The three components of γ-oryzanol evidenced significant antioxidant activity when they were mixed with linoleic acid in a molar ratio of 1:100 and 1:250 but not in a molar ratio of 1:500 (P < 0.05). α-Tocopherol and ferulic acid also demonstrated significant antioxidant activity at all three molar ratios (P < 0.05). The highest molar ratio (1:100) of α-tocopherol, however, caused greater levels of 9HPODE(t,c) and 13HPODE(c,t) than the other two less concentrated treatments.

Full text of DOI:10.1007/s11746-001-0320-1

Antioxidant Activities of Major Components of γ-Oryzanol
from Rice Bran Using a Linoleic Acid Model
Zhimin Xu and J. Samuel Godber*
Department of Food Science, Louisiana Agricultural Experimental Station,
Louisiana State University Agricultural Center, Baton Rouge, Louisiana 70803
ABSTRACT: The change in hydroperoxides of linoleic acid in-
cubated with constant micro air flow at 37°C was used to eval-
uate the antioxidant activities of three major components of γ-
oryzanol from rice bran (cycloartenyl ferulate, 24-methylene
cycloartanyl ferulate, and campesteryl ferulate) compared with
α-tocopherol and ferulic acid. The four hydroperoxide isomers
of linoleic acid, 9-hydroperoxy-10-trans,12-cis-octadeca-
dienoic acid [9HPODE(t,c)], 9-hydroperoxy-10-trans,12-trans-
octadecadienoic acid, 13-hydroperoxy-9-cis,11-trans-octadeca-
dienoic acid [13HPODE(c,t)], and 13-hydroperoxy-9-trans,11-
trans-octadecadienoic acid, were measured using normal-phase
high-performance liquid chromatography with an ultraviolet de-
tector. The three components of γ-oryzanol evidenced signifi-
cant antioxidant activity when they were mixed with linoleic
acid in a molar ratio of 1:100 and 1:250 but not in a molar ratio
of 1:500 (P < 0.05). α-Tocopherol and ferulic acid also demon-
strated significant antioxidant activity at all three molar
ratios (P < 0.05). The highest molar ratio (1:100) of α-tocoph-
erol, however, caused greater levels of 9HPODE(t,c) and
13HPODE(c,t) than the other two less concentrated treatments.
Paper no. J9721 in JAOCS 78, 645–649 (June 2001).
KEY WORDS: Antioxidant, hydroperoxide, linoleic acid,
γ-oryzanol.
γ-Oryzanol in rice bran has been reported to lower serum cho-
lesterol (1–3). Ten components of γ-oryzanol identified in rice
bran consist of ferulic acid and triterpene derived compounds,
which are combined by an ester bond (4). Cycloartenyl feru-
late, 24-methylenecycloartanyl ferulate, and campesteryl fer-
SCHEME 1
ulate (Scheme 1) are the three major components and account
for ca. 80% of γ-oryzanol in rice bran. The nutritional func-
tion of γ-oryzanol components may be related to their antiox-
idant property because of the ferulic acid structure. Ferulic
acid is a phenolic acid antioxidant (5–8), and Ohta et al. (9)
identified ferulic acid sugar esters as active components in
corn bran hemicellulose fragments. However, the antioxidant
capacities of γ-oryzanol components are not known.
Accelerated stability methods are used extensively in eval-
uating the antioxidant activity of a compound. Usually, higher
temperatures and oxygen concentrations are used to acceler-
ate the oxidation in a lipid system. Frankel (10) reported the
limitations of these tests and indicated that the results were
unreliable when the test was done at some extreme condi-
tions. Also, these severe models may not be appropriate to
predict antioxidant activities of nutritional components in the
human body. Under more gentle conditions, however, disad-
vantages include lower sensitivity in monitoring oxidative
products and more time-consuming procedures. In this study,
a mildly accelerated oxidation model was developed in which
linoleic acid was incubated with micro air flow at 37°C, with
continuous hydroperoxides analysis by a sensitive high-per-
formance liquid chromatography (HPLC) method. The an-
tioxidant activity of the three major components of γ-oryzanol
was investigated and compared with that of α-tocopherol and
ferulic acid under identical experimental conditions.
EXPERIMENTAL PROCEDURES
Chemical and materials. All solvents were HPLC grade.
Hexane was obtained from Curtin Matheson Scientific Inc.
*To whom correspondence should be addressed. E-mail: fsgodb@lsu.edu
Copyright © 2001 by AOCS Press
645
JAOCS, Vol. 78, no. 6 (2001)

646
Z. XU AND J.S. GODBER
(Houston, TX). Isopropanol was from Mallinckrodt Baker of hydroperoxides over time was used to evaluate antioxidant
Inc. (Paris, KY), and ethyl ether and acetic acid were from activity for each tested compound.
Fisher Scientific Inc. (Fair Lawn, NJ). α-Tocopherol, ferulic
Statistical analysis. Each treatment and control were eval-
acid, and linoleic acid were purchased from Sigma Chemical uated three times in the linoleic acid model. The production
Co. (St. Louis, MO). Hydroperoxides and hydroxides of of each hydroperoxide of linoleic acid was analyzed using the
linoleic acid were from ICN Pharmaceuticals, Inc. (Costa General Linear Model procedure of the Statistical Analysis
Mesa, CA) and were stored at −80°C before use. Rice bran System (SAS Institute, Cary, NC). The rates of total produc-
was supplied from Riviana Foods, Inc. (Abbeville, LA).
tion of hydroperoxides were analyzed using the regression
Accelerated oxidation of linoleic acid. The three high-pu- model of the Statistical Analysis System. Significant differ-
rity components of γ-oryzanol, cycloartenyl ferulate, 24-meth- ence between means was considered at P < 0.05.
ylene cycloartanyl ferulate, and campesteryl ferulate, were pu-
rified from rice bran using a preparative scale reversed-phase
HPLC procedure (4). α-Tocopherol, ferulic acid, and the three
RESULTS AND DISCUSSION
high-purity γ-oryzanol components were prepared at concen- Hydroperoxides of linoleic acid. Four hydroperoxides
trations of 0.08, 0.16, and 0.40 mM in hexane and isopropanol (HPODE) and hydroxides (HODE) of linoleic acid were sep-
(9:1, vol/vol) as treatments. The control was hexane and iso- arated using normal-phase HPLC (Fig. 1). Linoleic acid is
propanol (9:1, vol/vol) without the addition of any compo- susceptible to oxidation because its structure contains 1,4-
nents. Linoleic acid solution (40 mM) was prepared by dis- pentadiene that is highly vulnerable to free radical attack (11).
solving 1.1202 g of linoleic acid in 100 mL of hexane. The The hydrogen atom abstraction from the 1,4-pentadiene of
linoleic acid solution (5 mL) and 5 mL of each treatment or linoleic acid leads to a pentadienyl radical. The pentadienyl
control solution were mixed in a 25-mL test tube. The molar radical generates 9-hydroperoxy-10-trans,12-cis-octadeca-
ratios of treatment to linoleic acid were 1:100 (0.40 mM treat- dienoic acid [9HPODE(t,c)] and 13-hydroperoxy-9-cis,11-
ment solution), 1:250 (0.16 mM treatment solution), and 1:500 trans-octadecadienoic acid [13HPODE(c,t)] after addition of
(0.08 mM treatment solution). An aliquot of 500 µL of the so- oxygen. Also, the pentadienyl radical produces a new trans
lution was transferred to an HPLC vial to determine initial pentadienyl radical during β-fragmentation. The trans penta-
concentrations of hydroperoxides. Then, the test tube was in- dienyl radical produces 9-hydroperoxy-10-trans,12-trans-oc-
cubated in a 37°C water bath. Tubing connected to com- tadecadienoic acid [9HPODE(t,t)] and 13-hydroperoxy-9-
pressed air (BOC Gases, Port Allen, LA) was passed though trans,11-trans-octadecadienoic acid [13HPODE(t,t)]. Gener-
the tube cap and into the test tube with the tip of the tubing ally, the trans structure has lower energy than the cis structure
touching the bottom of the test tube. The flow rate of air was and dominates the hydroperoxides mixture (11).
controlled at 200 µL/min. Air flow in the reaction solution pro-
Adequate resolution of each hydroperoxide was obtained
duced a tiny air bubble about every 10 s. Time of sampling after HPLC conditions were optimized. Since the structure
was scheduled at 40, 80, 120, 160, and 200 min from the be- and molecular weight of linoleic acid hydroperoxides and hy-
ginning of incubation. The volume of reaction solution re-
mained constant between each interval of sampling. In the
sampling step, 500 µL was taken from the reaction solution
and added to an HPLC vial after vortexing for 30 s.
Analysis of hydroperoxides of linoleic acid. The hydroper-
oxides of linoleic acid were determined using a normal-phase
HPLC method. The HPLC system consisted of a Waters (Mil-
ford, MA) 510 pump, a 680 automated gradient controller, a
715 ultra WISP sample processor, a Hewlett-Packard (San Fer-
nando, CA) diode-array detector, and a Baseline 810 chroma-
tography workstation (Waters, Milford, MA). A Zorbax SIL
(DuPont Co., Wilmington, DE) column was used with mobile
phase that consisted of hexane/ethyl ether/isopropanol/acetic
acid (100:15:0.1:0.1, by vol), and the flow rate (min, mL/min)
was 0–18, 1.8; 17–18, 1.8–2.0; 19–39, 2.0; 39–40, 2.0–1.8.
Total run time was 40 min. Absorbance at 234 nm was moni-
FIG. 1. Chromatogram of hydroperoxides and hydroxides of linoleic
acid at 200 min of oxidation time. 1: 13-hydroperoxy-9-cis-11-trans-
tored with the detector. The concentration change of each hy-
droperoxide of linoleic acid at every sampling time was ob-
tained by deducting the initial concentration from measured
concentration. The change of concentration indicated the per-
oxide production of linoleic acid during oxidation.
octadecadienoic acid [13HPODE(c,t)]; 2: 13-hydroperoxy-9-trans,11-
trans-octadecadienoic acid [13HPODE(t,t)]; 3: 9-hydroperoxy-10-trans-
12-cis-octadecadienoic acid [9HPODE(t,c)]; 4: 9-hydroperoxy-10-
trans,12-trans-octadecadienoic acid [9HPODE(t,t)]; 5: 13-hydroxy-9-
cis,11-trans-octadecadienoic acid [13HODE(c,t)]; 6: 13-hydroxy-9-
trans,11-trans-octadecadienoic acid [13HODE(t,t)]; 7: 9-hydroxy-10-
trans,12-cis-octadecadienoic acid [9HODE(t,c)]; 8: 9-hydroxy-10-
trans,12-trans octadecadienoic acid [9HODE(t,t)].
Total production of hydroperoxides was obtained by sum-
ming all hydroperoxides. The rate (slope) of total production
JAOCS, Vol. 78, no. 6 (2001)

ANTIOXIDANT ACTIVITIES OF γ-ORYZANOL COMPONENTS
647
droxides are similar, it is difficult to separate them individu- in a biological system. A similar analytical method and model
ally within a short retention time. The 40-min run time with were performed by Makinen et al. (12) after methyl linoleate
gradient flow rate of the mobile phase was needed to ob- was oxidized at 40°C for 24 and 48 h with or without antioxi-
tain quantitative resolution. In addition, the short wave- dants. The advantage of methyl linoleate model is that the hy-
length (234 nm) used in the ultraviolet (UV) detector was droperoxides in methyl linoleate model are not as readily de-
very close to the UV cutoff of ethyl ether, isopropanol, and composed as in linoleic acid model in longer oxidation time
acetic acid, thus the baselines of chromatograms were slightly (13). As hydroperoxides with linoleic acid are not highly sta-
noisy. The order of elution of the hydroperoxides and hydrox- ble, the linoleic acid model may only be suitable to indicate
ides of linoleic acid was 13HPODE(c,t), 13HPODE(t,t), oxidation status in the early stage of oxidation.
9HPODE(t,c), 9HPODE(t,t), 13HODE(c,t), 13HODE(t,t),
Antioxidant activity of α-tocopherol and ferulic acid at dif-
9HODE(t,c), and 9HODE(t,t). The chromatogram of hy- ferent ratios to linoleic acid. The rates of hydroperoxide pro-
droperoxides of linoleic acid is similar to that of hydroperox- duction in linoleic acid with α-tocopherol (α-tocopherol/
ides of methyl linoleate (12). Hydroxides of linoleic acid were linoleic acid molar ratio = 1:100, 1:250, and 1:500) are shown
not as abundant as that of hydroperoxides and did not change in Figures 2A, 2B, and 2C, respectively. Antioxidant activity
significantly during oxidation; thus hydroxides were not sen- of α-tocopherol was indicated by significantly lower rates of
sitive indicators of the degree of lipid oxidation in this hydroperoxide production than control. An interesting phe-
linoleic acid model system within 200 min of oxidation. Hy- nomenon was noted in that there was a higher rate of forma-
droperoxides of linoleic acid were used to evaluate oxidation tion of hydroperoxides containing cis,trans or trans,cis struc-
status of linoleic acid in this system. Makinen et al. (12) also ture at the 1:100 ratio of α-tocopherol than at 1:250 or 1:500
reported that the concentration of hydroperoxides is greater (Fig. 3), even though hydroperoxides containing trans,trans
than that of hydroxides in the methyl linoleate model.
were extensively inhibited by α-tocopherol. This agrees with
Linoleic acid model. In Table 1, the productions of hy- Makinen et al. (12), who state that α-tocopherol increases the
droperoxides of the control after 120 and 200 min of oxida- proportion of the cis,trans hydroperoxides while the forma-
tion are depicted. The four hydroperoxides increased signifi- tion of trans,trans hydroperoxides is inhibited. Jung and Min
cantly with increasing oxidation time. The amounts of (14) reported that tocopherols had significant prooxidant ef-
13HPODE(t,t) and 9HPODE(t,t) were significantly greater fects at higher concentration. In this study, high concentra-
than that of 13HPODE(c,t) and 9HPODE(t,c) during oxida- tions of α-tocopherol inhibited the production of trans,trans
tion with air. This greater production of trans,trans hydroper- hydroperoxides but appeared to elicit prooxidant activity by
oxide isomers supported the idea that the hydroperoxides con- accelerating free radical reactions that produced cis,trans and
taining a trans,trans structure have lower energy and are more trans,cis hydroperoxides. The total production of hydroper-
thermodynamically favorable for formation during free radi- oxides was higher in 1:100 ratio of α-tocopherol than either
cal attack than those having trans,cis or cis,trans (11). Dis- 1:250 or 1:500 ratio. Marinova and Yanishlieva (6) also found
criminate changes in concentration for each hydroperoxide in that the antioxidant activity of α-tocopherol is altered at dif-
the linoleic acid model as measured by HPLC demonstrated ferent temperatures. α-Tocopherol is more effective at ex-
that this system could be used in quantitative determination tending the induction period and reducing the rate of autoxi-
of the status of linoleic acid oxidation. The mild conditions dation as the temperature is increased (6).
of oxidation in this model, i.e., 37°C and micro air flow, over-
Ferulic acid exhibited antioxidant activity at all three
come the limitations of accelerated stability models that use molar ratios (ferulic acid/linoleic acid = 1:100, 1:250, and
more severe or complex oxidative conditions as discussed by 1:500) (Figs. 2A, 2B, 2C). The production of hydroperoxides
Frankel (10). The antioxidant activity of a component deter- with ferulic acid at a ratio of 1:100 was significantly lower
mined in this model may more closely reflect its real situation than that with the ratios of 1:250 or 1:500. Table 1 shows the
TABLE 1
Production of Hydroperoxides of Control, α-Tocopherol, Ferulic Acid, and the Three Major γ-Oryzanol Components Treatments
After 120 and 200 min of Oxidation (treatment component/linoleic acid molar ratio = 1:250)a
Products
13HPODE(c,t)
120
13HPODE(t,t)
120
9HPODE(t,c)
120 200
71.8 8.6 134.1 6.7
9HPODE(t,t)
120
185.5 31.5 370.8 14.8
Time (min)
200
200
200
Control
225.7 36.1
78.7 7.9
48.9 3.9
316.2 56.9
119.7 9.6
70.5 7.8
517.9 77.7
41.0 3.3
950.4 76.0
54.2 3.8
α-Tocopherol
Ferulic acid
COMP 1
COMP 2
COMP 3
50.7 6.1
24.7 1.5
78.7 6.3
61.0 7.3
26.8 2.1
60.8 4.9
34.6 2.8
100.2 15.0
151.0 22.6
346.8 28.8
292.3 31.4
261.9 28.3
244.8 39.2
473.9 53.4
417.6 31.4
362.9 41.6
182.1 25.6
169.2 26.2
163.4 29.7
252.0 44.1
292.4 37.7
250.4 42.8
86.0 8.8 126.4 15.2 106.0 14.8 157.4 26.8
75.8 9.4 112.8 10.7
89.6 9.2 113.4 13.6
84.5 13.6 136.0 19.1
73.8 14.6 106.7 19.9
aCOMP 1, cycloartenyl ferulate; COMP 2, 24-methylene cycloartanyl ferulate; COMP 3, campesteryl ferulate; 13HPODE(c,t), 13-hydroperoxy-9-cis-11-
trans-octadecadienoic acid; 13HPODE(t,t), 13-hydroperoxy-9-trans-11-trans-octadecadienoic acid; 9HPODE(t,c), 9-hydroperoxy-10-trans-12-cis-octadeca-
dienoic acid; 9HPODE(t,t), 9-hydroperoxy-10-trans-12-trans-octadecadienoic acid.
JAOCS, Vol. 78, no. 6 (2001)

648
Z. XU AND J.S. GODBER
FIG. 3. Production of hydroperoxides of linoleic acid with different
molar ratios of α-tocopherol to linoleic acid after 200 min oxidation.
Significant differences (P < 0.05) are expressed by different letters on
the bars in a cluster. For abbreviations see Figure 1.
major oryzanol components treatments after 120 and 200 min
of oxidation is listed in Table 1. The concentrations of
trans,trans hydroperoxides were much higher than that of
trans,cis and cis,trans hydroperoxides. The changes in con-
centration of individual hydroperoxides in linoleic acid mod-
els with each of the three components were similar to those
with ferulic acid. This suggests that the mechanism of antioxi-
dation of components of γ-oryzanol is the same as that of fer-
ulic acid. The antioxidant activity of these ferulic components
arises from the phenolic hydroxyl group in the ferulate portion
of their overall structure. In this study, free ferulic acid had
greater activity than did the three ferulic acid esters at the three
different ratios, especially at a lower molar ratio (1:500). The
triterpene portion of γ-oryzanol may affect its antioxidant ac-
tivity by lowering the mobility in the system due to its rela-
FIG. 2. Rates of production of total hydroperoxides of linoleic acid with
different ratios of component to linoleic acid and control. Significant
differences in rate (P < 0.05) are expressed by different letters in each
ratio. (A) 1:100 in mole; (B) 1:250 in mole; (C) 1:500 in mole. ꢀ Con-
trol; ꢁ cycloartenyl ferulate; ꢂ 24-methylene cycloartanyl ferulate; ꢃ
campesteryl ferulate; ꢂ α-tocopherol; ꢃ ferulic acid.
production of each hydroperoxide of linoleic acid in ferulic tively larger molecular structure than free ferulic acid.
acid treatment after 120 and 200 min of oxidation. Unlike α- Although the antioxidation activities of γ-oryzanol compo-
tocopherol, in which cis,trans hydroperoxide increased to a nents were lower than that of α-tocopherol in protecting
greater extent than trans,trans isomers, all of the hydroperox- against linoleic acid oxidation, the quantity of these compo-
ides with ferulic acid increased over time. The concentration nents is 10 times greater than α-tocopherol in rice bran (15).
of trans,trans hydroperoxides was higher than that of Because the structure of γ-oryzanol components is similar to
cis,trans or trans,cis hydroperoxides. It has been reported that that of cholesterol, an important component in cell mem-
the antioxidant activity of ferulic acid depends on the hydrox- branes, they may have greater availability and accessibility in
ylation (–HO) in the phenolic ring (5,8). Marinova and Yan- cell membranes than α-tocopherol in reducing oxidation stress
ishlieva (6) investigated the antioxidant activity of ferulic and maintaining functionality of cells. We are currently evalu-
acid using purified lard triacylglycerol at 25, 50, 75, and ating this possibility in model systems. It is apparent from this
100°C and found that antioxidant activity of ferulic acid re- study that the antioxidant function of γ-oryzanol components
mained constant with increasing temperature.
supports the potential nutritional value of rice bran.
Antioxidant activities of the three major components of γ-
oryzanol at different ratios to linoleic acid. Three major com- ACKNOWLEDGMENT
ponents of γ-oryzanol, cycloartenyl ferulate, 24-methylene cy-
cloartanyl ferulate, and campesteryl ferulate, significantly re-
duced the rate of hydroperoxide production at the ratios of
1:100 and 1:250 (Figs. 2A,2B) but not at 1:500 (Figs. 2C). An-
tioxidant activities of these components decreased signifi-
cantly with decreasing concentration in the linoleic acid
model. The production of each hydroperoxide in the three
Approved for publication by the Director of the Louisiana Agricul-
tural Experiment Station as manuscript number 99-21-0562.
REFERENCES
1. Gerhardt, A.L., and N.B. Gallo, Full-fat Rice Bran and Oat Bran
Similarly Reduce Hypercholesterolemia in Humans, J. Nutr.
128:865–869 (1998).
JAOCS, Vol. 78, no. 6 (2001)

ANTIOXIDANT ACTIVITIES OF γ-ORYZANOL COMPONENTS
649
2. Seetharamaiah, G.S., and N. Chandrasekhara, Studies on 10. Frankel, E.N., Recent Advances in Lipid Oxidation, J. Sci. Food
Hypocholesterolemic Activity of Rice Bran Oil, Atherosclerosis
78:219–235 (1989).
3. Sugano, M., and E. Tsuji, Rice Bran Oil and Cholesterol Metab-
olism, J. Nutr. 127:521S–524S (1997).
Agric. 54:495–500 (1991).
11. Porter, N.A., S.E. Caldwell, and K.A. Mills, Mechanisms of
Free Radical Oxidation of Unsaturated Lipids, Lipids 30:
277–290 (1995).
4. Xu, Z., and J.S. Godber, Purification and Identification of Com- 12. Makinen, M., A. Kamal-Eldin, A.-M. Lampi, and A. Hopia, Ef-
ponents of γ-Oryzanol in Rice Bran Oil, J. Agric. Food Chem.
fects of α- and γ-Tocopherols on Formation of Hydroperoxides
and Two Decomposition Products from Methyl Linoleate, J.
Am. Oil Chem. Soc. 77:801–806 (2000).
47:2724–2728 (1999).
5. Cuvelier, M.E., H. Richard, and C. Berset, Comparison of the
Antioxidative Activity of Some Acid-Phenols: Structure–Activ- 13. Hopia, A., S.W. Huang, and E.N. Frankel, Effect of α-Tocoph-
ity Relationship, Biosci. Biotech. Biochem. 56:324–330 (1992).
6. Marinova, E.M., and N.V. Yanishlieva, Effect of Temperature
erol and Trolox on the Decomposition of Methyl Linoleate Hy-
droperoxides, Lipids 31:357–365 (1996).
on the Antioxidative Action of Inhibitors in Lipid Autoxidation, 14. Jung, M.Y., and D.B. Min, Effects of α-, γ-, and β-Tocopherols
J. Sci. Food Agric. 60:313–318 (1992).
7. Marinova, E.M., and N.V. Yanishlieva, Effect of Lipid Unsatu-
on Oxidative Stability of Soybean Oil, J. Food Sci. 55:
1464–1465 (1990).
ration on the Antioxidative Activity of Some Phenolic Acids, J. 15. Shin, T., J.S. Godber, D.E. Martin, and J.H. Wells, Hydrolytic
Am. Oil Chem. Soc. 71:427–434 (1994).
8. Pratt, D.E., Water-Soluble Antioxidant Activity in Soybeans, J.
Food Sci. 37:322–323 (1972).
Stability and Changes in E Vitamers and Oryzanol of Extruded
Rice Bran During Storage, Ibid. 62:704–708 (1997).
9. Ohta, T., S. Yamasaki, Y. Egashira, and H. Sanada, Antioxida-
tive Activity of Corn Bran Hemicellulose Fragments, J. Agric.
Food Chem. 42:653–656 (1994).
[Received August 2, 2000; accepted February 22, 2001]
JAOCS, Vol. 78, no. 6 (2001)