Performances of CN-columns for the analysis of γ-oryzanol and its p-coumarate and caffeate derivatives by normal phase HPLC and a validated method of quantitation

Article abstract of DOI:10.1016/j.foodchem.2012.12.008

γ-Oryzanol is an important phytochemical used in pharmaceutical, alimentary and cosmetic preparations. The present article, for the first time, discloses the performances of NP-HPLC in separating γ-oryzanol components and develops a validated method for i

Full text of DOI:10.1016/j.foodchem.2012.12.008

Food Chemistry 138 (2013) 2079–2088
Contents lists available at SciVerse ScienceDirect
Food Chemistry
journal homepage: www.elsevier.com/locate/foodchem
Analytical Methods
Performances of CN-columns for the analysis of c-oryzanol and its p-coumarate and
caffeate derivatives by normal phase HPLC and a validated method of quantitation
⇑
Michele D’Ambrosio
Laboratory of Bio-Organic Chemistry, Department of Physics, University of Trento, via Sommarive 14, 38123 Trento, Italy
a r t i c l e i n f o
a b s t r a c t
Article history:
Received 29 March 2011
Received in revised form 28 June 2012
Accepted 7 December 2012
Available online 20 December 2012
c
-Oryzanol is an important phytochemical used in pharmaceutical, alimentary and cosmetic prepara-
tions. The present article, for the first time, discloses the performances of NP-HPLC in separating
-oryzanol components and develops a validated method for its routine quantification. The analysis is
c
performed on a cyanopropyl bonded column using the hexane/MTBE gradient elution and UV detection
at 325 nm. The method allows: the separation of steryl ferulate, p-coumarate and caffeate esters, the sep-
aration of cis- from trans-ferulate isomers, the splitting of steroid moieties into saturated and unsaturated
at the side chain. The optimised method provides excellent linear response (R2 = 0.99997), high precision
(RSD < 1.0%) and satisfactory accuracy (R⁄ = 70–86%). In conclusion, the established method presents the
details of the procedure and the experimental conditions in order to achieve the required precision and
instrumental accuracy. The method is fast and sensitive and it could be a suitable tool for quality assur-
ance and determination of origin.
Presented in part by M.D. at the 16th
International Symposium on Separation
Science, Rome, Italy, 6–10 September 2010.
Keywords:
c
-Oryzanol
Phytosterols
Ferulate esters
p-Coumarate esters
Caffeate esters
Rice bran
Ó 2012 Elsevier Ltd. All rights reserved.
Corn
Wheat
Rye
CN column
1. Introduction
c
-oryzanol appears to lower the plasmatic level of low-density
cholesterol. Anti-oxidant and free-radical scavenging properties
are attributed to the ferulate moiety of -oryzanol molecules. So
RBO and -oryzanol are used as additives in the alimentary indus-
try. -Oryzanol gives RBO a very good shelf-life and thermal stabil-
Phytosterols occur in plants either as free or conjugated forms,
the latter comprising steryl esters of fatty or phenolic acids and
steryl glycosides (Moreau, Whitaker, & Hicks, 2002). A complex
mixture of esters, between hydroxycinnamic acids (CAD) (Fig. 1,
R) and sterols based on the cholestane (Fig. 1, C) or the parent cycl-
c
c
c
ity even at frying temperature so that it acts as a natural
preservative of refrigerated cooked food during storage time. RBO
oartane (Fig. 1, A) skeletons, is collectively termed as ‘‘
c
-oryzanol’’.
and
creams.
Large number of commercial products which contain c-oryza-
nol as an ingredient, require a fast and reliable procedure of detec-
tion and quantification. This is a hard task because of the
complexity and variability of its composition. There is also the
necessity to perform analyses on different types of samples such
as cereal bran, crude or refined oils, by-products from cereal pro-
cessing, foods, etc.
c-oryzanol act as suitable natural UV filter in sun screen
The main source of -oryzanol is the bran of some grains, in
c
particular brown and red rice (Lerma-García, Herrero-Martinez,
Simó-Alfonso, Mendonça, & Ramis-Ramos, 2009). Brans from cere-
als, such as rye, wheat, triticale, corn, Job’s tears and oats, contain
less
Rice bran oil (RBO) accounts for 15–20 wt.% of rice bran and dis-
solves -oryzanol (0.9–2.9% oil) and vitamin E (0.10–0.14% oil)
(Lerma-García et al., 2009).
Interest in RBO and -oryzanol (Lerma-García et al., 2009) has
c-oryzanol than rice bran (Hakala et al., 2002; Seitz, 1989).
c
c
Several analytical methods have been proposed for the quanti-
been growing because of their technological importance and
potential health benefits. RBO is used as functional food, in fact
fication of
in samples of bran or oil (Lerma-García et al., 2009). Since the spe-
cific extinction coefficient is a known constant (Hakala et al.,
2002), the simplest way to calculate the total content of -oryzanol
in a rice bran oil sample is to measure its UV absorption. This
c-oryzanol and for the identification of its constituents
e
c
⇑
Tel.: +39 0461 281509; fax: +39 0461 281696.
E-mail address: michele.dambrosio@unitn.it
0308-8146/$ - see front matter Ó 2012 Elsevier Ltd. All rights reserved.
http://dx.doi.org/10.1016/j.foodchem.2012.12.008

2080
M. D’Ambrosio / Food Chemistry 138 (2013) 2079–2088
Fig. 1. Molecular structures of main components of
c
-oryzanol: alcohols, possessing a cholestane (C) or a cycloartane (A) skeleton, are esterified by a hydroxycinnamic acid
(R).
method doesn’t provide any information about the structure of
individual components and much care must be taken to achieve
reliable measurements. In fact, other UV absorbing compounds in
the oil matrix may interfere with the ferulate chromophore.
Another possible method (Miller & Engel, 2006) used an on-line
coupled liquid chromatography–gas chromatography (LC–GC)
technique: a not very widespread equipment. The total amount
et al., 2005) columns as a stationary phase (Table 1, B), and several
mixing of solvents in the mobile phase (Table 1, C). In this tech-
nique, generally, UV detection at 315–325 nm was applied. The
identification of ferulate esters (Table 1, E) was best obtained by
tandem liquid chromatography–mass spectrometry (LC–MS) tech-
niques (Table 1, entries 5, 6 and 8). Past reports describe the collec-
tion of each peak followed by the chemical derivatisation of those
compounds and their analysis by proton NMR or GC–MS (Table 1,
entries 2 and 4).
of c-oryzanol was measured by LC-UV detection and the relative
proportion of individual components by GC-FID detection. The
analysis takes long time and the resolution seems to be poor (Table
1, entry 1).
The chief objective was to separate
c-oryzanol components
where up to ten peaks could be well resolved in one single run (Ta-
ble 1, G). Many compounds have been identified (Akihisa et al.,
2000; Xu & Godber, 1999; Yu et al., 2007) and the most common
structures are listed (Table 2, L). By comparison of RP-HPLC chro-
matograms (Hakala et al., 2002; Norton, 1995; Seitz, 1989), the
Most efforts have been devoted to reversed-phase (RP) chroma-
tography (Table 1, entries 2–8) after multi-step procedures of
chemical (Hakala et al., 2002; Seitz, 1989; Stöggl, Huck, Wongyai,
Scherz, & Bonn, 2005) and chromatographic (Norton, 1995; Xu &
Godber, 1999) pre-treatment of the sample (Table 1, D). The anal-
yses by high pressure liquid chromatography (HPLC) were done on
different experimental conditions, using C8 (Yu, Nehus, Badger, &
Fang, 2007), C18 (Lerma-García et al., 2009) and C30 (Stöggl
composition of c-oryzanol appears to vary in different types of cer-
eal source: (i) cycloartenyl ferulates (Fig. 1, A2 and A3) are pre-
dominant in rice but almost absent in other grains, (ii)
campestanyl and sitostanyl p-coumarates (Fig. 1, C2 and C5) and
Table 1
Comparison of analytical methods of
c-oryzanol by their chromatographic performances.
A
B
C
D
E
F
G
H
I
J
K
1
2
3
4
5
6
7
8
Miller and Engel (2006)
Seitz (1989)
Norton (1995)
Xu and Godber (1999)
Hakala et al. (2002)
Stöggl et al. (2005)
Chen and Bergman (2005)
Yu et al. (2007)
Diack and Saska (1994)
Heinmann et al. (2008)
Huang and Ng (2011)
D’Ambrosio
GC
H2, 340 °C
M/W 97:3; I
chrom
chem1
chrom
chrom
chem1
chem2
None
chem3
chem4
None
GC–MS
>33
>30
ꢀ30
>40
ꢀ20
>50
>28
>70
>30
30
5
3
7
10
4
9
5
8
2
1
N
N
N
N
N
N
N
N
N
N
N
Y
3
–
20
–
22
–
2
2
–
32
2
ꢀY
–
N
–
Y
–
N
N
–
N
Y
C18, 3
C18 , 5
C18, ?
C18, 5
C30 , 3
C18, 4
C8, 5
SiO2, 4
SiO2, 5
CN, 5
1H NMR GC–MS
Reference standards
GC–MS of TMS ethers
LC–APCI-MS
LC–APCI-MS
Reference standards
LC–ESI-MS
ꢀ2
ꢀ2
1
1
1
1
2
1
1
A/B/Aa/W82:3:2:13; I
M/A/D/Aa 50:44:3:3; I
M/W/Aa 97:2:1; I
M/mtbe 8:2; I
A/M/P/W 50:45:5:5; G
A/W/Fa 20:80:0.1; G
i-octane/Ae 97.5:2.5; I
H/Ae/Aa 98:1:1; I
H/P/Ae/Aa 97:1:1:1; I
H/mtbe/Fa 97:2:1; G
9
None
None
10
11
12
2
1
>4
None
None
None
25
32
1
>2
CN, 5
1H NMR LC–MS
7
Y
A, reference; B, stationary phase, lm; C, mobile phase (A acetonitrile, Aa acetic acid, Ae ethyl acetate, B butanol, D dichloromethane, Fa formic acid, H hexane, M methanol, P
2-propanol, W water), isocratic or gradient elution; D, sample pretreatment (chem1: KOH(aq)/hexane partition, injection of components dissolved into the basic layer, chem2:
dispersion oil/(A/M/P 50:45:5) 4:96 v/v, chem3: extraction MeOH/H2O then partition CH2Cl2/H2O, chem4: conc. KOH(aq) saponification of triglycerides, injection of com-
ponents dissolved into the hexane layer, chrom: preparative column over silica gel); E, technique of peak identification; F, total elution time; G, # of peaks; H, # of CAD; I,
detection of cis-isomers; J, # of real samples quantitated; K, validation tests.

M. D’Ambrosio / Food Chemistry 138 (2013) 2079–2088
2081
Table 2
Previous and present MS data of natural and synthetic components of
c-oryzanol.
L
Ma
Nb
Ob
Pb
573
587
601⁄
615
Qb
Rb
1
2
3
4
5
6
7
8
9
24-CH2-cholesteryl
C1
573
543
557
571
585
545
559
573
547
561
559
Stigmasteryl or stigmast-7-enyl
Cycloartenyl or cycloeucalenyl
24-CH2-cycloartanyl
Campesteryl or campest-7-enyl
Sitosteryl or sitost-7-enyl
Cycloartanyl
C8 or C9
A2 or A4
A3
C3 or C4
C6 or C7
A1
484⁄
498
601
615
575
587
601
561
575
512
472⁄
486⁄
575
{589}⁄
603
{577}
591
{589}
{603}
{577}
591
Campestanyl
Stigmastanyl (sitostanyl)
C2
C5
474
488
563
577
L, steryl moieties and codes of the structures shown in Fig. 1; M, (Xu & Godber, 1999); N, (Yu et al., 2007), ferulates; O, standard ferulates at tR = 11.5 and 19.0 min; P, standard
ferulates at tR = 12.0 and 20.0 min; Q, synthetic p-coumarates; R, synthetic caffeates.
⁄Means compounds with identical molecular weight eluting as separated peaks.
{} Means compounds with different molecular weight coeluting in one peak.
a
m/z of [M]+ of steryl TMS ethers.
b
m/z of [MꢂH]- of steryl CAD esters.
sitostanyl ferulate (Fig. 1, C5) are abundant in corn, (iii) campesta-
nyl and/or sitosteryl ferulates (Fig. 1, C2 and C6) are the major ste-
ryl moieties in grains other than rice and corn.
Several studies were oriented towards quantisation (Table 1, J)
but the validation tests were not always carried out (Table 1, K).
Moreover, p-coumarate esters (Table 1, H) and minor cis isomers
(Table 1, I) co-elute with abundant ferulates and trans isomers in
RP-HPLC (Norton, 1995; Seitz, 1989; Yu et al., 2007) so that they
may affect the actual value of each integrated peak. Often the
peaks were not resolved to the baseline of the chromatogram
and reliable sample preparation and the possibility of automated
and replicated HPLC injections. In NP-HPLC, the huge amount of
apolar waxes, TAG and FFA are eluted before the analytes of inter-
est thus eliminating the need of long column rinsing and tedious
preliminary cleaning-up of samples. Therefore, our aims are: (1)
to explore the potential application of NP-HPLC in separating
c-
oryzanol components, (2) to develop and validate a simple and
informative NP-HPLC method for the routine quantification of
oryzanol.
c-
(Hakala et al., 2002); in such a case, total c-oryzanol was quanti-
tated by pooling all the peak area. Thus the separation into compo-
2. Experimental
nents turned out to be vain (Chen & Bergman, 2005). Our early
elutions for the quantitative determination of c-oryzanol revealed
2.1. Chemicals
also some experimental drawbacks in the RP-HPLC methods. In
fact, apolar stationary phases are highly retentive of triacylglyce-
rides (TAG), free fatty acids (FFA) and waxes. A proper column rins-
ing, before re-equilibration, demands to change from water
miscible to water immiscible eluents. Therefore it cannot be easily
carried out and is solvent and time consuming. An efficient clean-
up procedure in the preparation of the sample is thus mandatory
(Table 1, D) where it is again a solvent and time consuming step.
In addition, loss of analyte may occur when using partitioning by
separatory funnel (Akihisa et al., 2000; Hakala et al., 2002; Seitz,
1989; Stöggl et al., 2005; Yu et al., 2007) or preliminary normal-
phase (NP) chromatography (Norton, 1995; Xu & Godber, 1999).
Taking into account the vast number of commercial products con-
Analytical grade solvents were used for extraction procedures
and reactions (VWR International, Fontenay-sous-Bois, France).
HPLC-grade hexane, heptane (J. Matthey, Karlsruhe, Germany), 2-
propanol (J.T. Baker, Mallinckrodt, The Netherlands), decane, MTBE
and formic acid (Sigma Aldrich, Steinheim, Germany) were used for
analyses. Standards of ethyl ferulate and umbelliferone were ob-
tained from Fluka (Steinheim, Germany) and c-oryzanol was ob-
tained from Farmalabor (Canosa (BA), Italy). Reagent grade
coumaric acid, caffeic acid, sodium hydroxide, acetic anhydride,
oxalyl chloride and sodium borohydride were purchased from Sig-
ma Aldrich (Steinheim, Germany). Silica gel 60 (0.063–0.020 mm)
was employed for flash chromatography and silica gel plates 60
F254 were used for TLC and PLC (Merck, Darmstadt, Germany).
taining
c-oryzanol, the more selective extraction with alcoholic
solvents may not be always feasible (Yu et al., 2007).
Column chromatography on silica gel has been used mostly for
2.2. Instrumentations and chromatographic conditions
a semi-preparative collection of c-oryzanol devoid of TAG and FFA
(Norton, 1995; Xu & Godber, 1999). Normal-phase TLC allows the
separation of p-coumarates from ferulates (Norton, 1995; Seitz,
1989), cis from trans isomers (Akihisa et al., 2000) and might differ-
entiate between 4-dimethyl, 4-monomethyl and 4-desmethyl ste-
ryl esters (Akihisa et al., 2000). Silica columns for HPLC (Table 1,
The preparative HPLC system consisted of a Merck Hitachi mod-
el L-7100 pump, L-7400 UV detector, D-7500 integrator and Rheo-
dyne manual injector equipped with a 200
lL loop. The column
was a Luna-CN (250 ꢁ 10 mm, 5
l
m particle size, 100 Å pore size;
Phenomenex, Torrance, CA, USA); the initial mobile phase was hex-
ane/MTBE/formic acid 97:2:1 which changed linearly to hexane/
MTBE/formic acid 90:9:1 within 23 min at the flow rate of 4 mL/
min and detection at k = 325 nm.
entries 9–12) allow the splitting of
c-oryzanol components into
two closely eluting peaks but their composition wasn’t proven
(Diack & Saska, 1994). While this manuscript was being revised,
an improved method, using a NP-HPLC and a cyanopropyl column,
Analytical separations were carried out using an Agilent 1100
series LC system consisting of a binary pump, a vacuum degasser,
was published but c-oryzanol was eluted as one peak (Huang & Ng,
2011). NP- and RP-HPLC methods have been reported (Heinmann,
an autosampler with standard analytical head (100 lL), a column
Xu, Godber, & Lanfer-Marquez, 2008) to determine identical values
thermostat and a 1200 series diode array detector (DAD). The col-
of the total amount of
approach.
c
-oryzanol but the authors preferred the NP
umn oven temperature was fixed at 25 °C; the flow rate was
1.0 mL/min and the injection volume was 10
at 290, 325 and 340 nm. The HPLC instrument was equipped with
a Synergi Hydro column (150 ꢁ 4.6 mm, 4 m particle size, 80 Å
pore size; Phenomenex, Torrance, CA, USA) for the analyses of com-
lL; the DAD was set
In our opinion, RP-HPLC methods are more suitable for qualita-
tive studies of -oryzanol components and for a small number of
samples. The screening of a large number of samples requires a fast
c
l

2082
M. D’Ambrosio / Food Chemistry 138 (2013) 2079–2088
ponents in the RP mode; the eluents were 95% methanol–5% water
(A) and 100% methanol (B). The initial gradient condition was
maintained at 100% A, changing linearly to 100% B in 30 min and
at 30–32 min and 13 min as equilibration time. A representative
chromatogram is shown in Fig. 2a.
NMR spectra were acquired using a Bruker Avance 400 spec-
trometer (1H at 400 MHz). Chemical shifts were reported in ppm
on the d scale with the residual solvent signal as internal reference
(CHCl3: dH 7.26); coupling constants (J) are given in Hz.
LC–MS analyses were performed using an Agilent 1100 series LC
system interfaced to a Bruker model Esquire_LC multiple ion trap
mass spectrometer equipped with an atmospheric pressure inter-
face electrospray (API-ES) chamber. Conditions for ESI-MS analysis
then held at this eluent
B for 10 min. A Luna-CN column
(250 ꢁ 4.6 mm, 5 m particle size, 100 Å pore size; Phenomenex,
l
Torrance, CA, USA) was selected for quantitative measurements
in the NP mode; the mobile phase consisted of hexane (A) and hex-
ane/MTBE 1:1 (B), both containing 1% (v/v) formic acid, applied in
the following stepped linear gradient: 2–22% B at 0–20 min, 22–
100% B at 20–25 min; then 5 min washing at 100% B, 100–2% B
Fig. 2. NP-HPLC runs of: (a) Carnaroli rice bran sample containing a high percentage of cis-ferulates, ‘‘sat.’’ means steroid moieties possessing a saturated side chain, ‘‘unsat.’’
means steroid moieties possessing an unsaturated side chain; (b) standard -oryzanol; (c) synthetic steryl p-coumarates; (d) synthetic steryl caffeates.
c

M. D’Ambrosio / Food Chemistry 138 (2013) 2079–2088
2083
Table 3
Analytical data for the quantification of c-oryzanol.
Sample
Carnaroli rice
bran
Vialone rice
bran
wheat
bran
Storo corn
flour
Walet rye
flour
Supplement
tablet
Sunscreen
oil
1
2
3
4
5
6
Mass of sample (mg)
402.0
323.1
66.6
16.6
97.0
399.5
328.1
59.9
15.0
97.1
4.99
799.4
746.2
27.6
3.5
96.8
4.93
4000.0
3775.0
118.8
3.0
97.3
19.80
3999.6
3855.6
43.4
1.1
97.5
423.8
364.2
53.0
12.5
98.4
0.57
46.0
Mass of defatted sample residue (mg)
Mass of oil (mg)a
46.0
Yield of oil (%)
Recovery of oil + defatted sample (%)
Concentration of oil for HPLC analysis (mg/
mL)
5.01
9.98
50.00
0.00
7
8
9
cis-Ferulates (nmol/mL)
RSD%
2.08
0.25
66.73
0.47
121.46
0.85
0.41
13.32
24.24
3.78
22.78
35.5
3.30
0.35
43.58
0.39
85.55
0.35
0.66
8.73
17.14
2.39
15.92
33.7
2.5
2.55
0.17
67.01
0.38
2.24
6.37
0.52
13.60
0.45
0.29
8.51
96.8
3.5
9.74
0.12
84.44
0.11
4.54
0.12
0.49
4.26
0.23
0.09
2.91
94.9
9.9
9.73
0.21
94.09
1.06
5.77
0.21
0.98
9.43
0.58
0.07
6.41
94.2
8.9
15.31
0.33
86.83
0.06
85.47
0.08
26.70
151.45
149.07
24.38
194.99
50.4
Cholestane trans-ferulate (nmol/mL)b
10.13
0.49
77.04
0.14
0.000
0.20
1.54
10 RSD%
11 Cycloartane trans-ferulate (nmol/mL)c
12 RSD%
13 cis-Ferulates (nmol/mg oil)
14 Cholestane trans-ferulate (nmol/mg oil)
15 Cycloartane trans-ferulate (nmol/mg oil)
16
17
c
c
-Oryzanol (ꢀmg/g sample)
-Oryzanol (ꢀmg/g oil)
1.06
11.6
0.0
18 Cholestane/c-oryzanol (%, mol/mol)
19 cis-Ferulates/
c
-oryzanol (%, mol/mol)
1.1
8.2
a
b
c
‘‘Oil’’ stands for extract evaporated to dryness or sunscreen oil.
‘‘Cholestane’’ exactly means: sterols possessing a saturated side chain.
‘‘Cycloartane’’ exactly means: sterols possessing an unsaturated side chain.
of HPLC peaks in negative ion mode included a capillary voltage of
4000 V, a nebulising pressure of 30.0 psi, a drying gas flow of 7 mL/
min and a temperature of 300 °C.
funnel, diluted with water (200 mL) and shaken with hexane
(3 ꢁ 30 mL). The combined organic phase was washed with water
to the neutral pH and dried over anhydrous Na2SO4. The solvent
was finally evaporated to give the sterol mixture (420 mg,
ꢀ99 mmol, 99% yield).
2.3. Samples
A solution of caffeic acid (180 mg, 1.0 mmol) in dry pyridine
(0.3 mL) was added of acetic anhydride (0.47 mL, 5 mmol) and
then stirred at r.t. over night. The reaction mixture was gently
heated under vacuum to evaporate excess acetic anhydride and
pyridine thus obtaining quantitatively (2E)-3-[3,4-bis(acetyl-
oxy)phenyl]-2-propenoic acid (264 mg, 1.0 mmol).
Full-fat rice bran (Carnaroli and Vialone Nano varieties) were
kindly provided by ‘‘FERRON Gabriele e Maurizio snc’’ (Isola della
Scala (VR), Italy). Rye kernels (Walet variety) were provided by
‘‘Centro di Sperimentazione Agraria e Forestale’’ (Laimburg (BZ),
Italy). The other samples were purchased from a retail store. Cereal
samples were stored at ꢂ18 °C before analysis and no stabilisation
process was applied. The extracts were stored at ꢂ18 °C under
nitrogen in white bottles and vials. Samples with high levels of
cis-isomers were obtained by exposure of samples to sunlight or
by UV irradiation.
This protected caffeic acid was suspended in dry CH2Cl2 (4 mL),
stirred under N2 at r.t. and added of oxalyl chloride (2 mL, 2 mmol,
2 M in CH2Cl2). After 30 min all the solid dissolved so the solution
was evaporated to dryness in vacuo. The resulting solid was dis-
solved in dry CH2Cl2 (4 mL). Then a solution of sterols from c-oryz-
anol (660 mg, ꢀ1.56 mmol) and pyridine (0.097 mL, 1.2 mmol) in
1 mL of dry CH2Cl2 was added dropwise. The mixture was stirred
for 24 h at r.t. then filtered. The precipitate was washed with CH2-
Cl2 and discarded. Silica gel (1 spoon) was added to the clear solu-
tion and the solvent was evaporated. The desiccated slurry was
applied to a silica column and chromatographed with a gradient
of diethyl ether in hexane to isolate (2E)-3-[3,4-bis(acetyl-
oxy)phenyl]-2-propenoate of the sterols (100 mg, ꢀ0.149 mmol,
15% yield).
2.4. Sample preparation
The accurately weighed powder (Table 3, entry 1) was extracted
twice with hexane (8 mL/400 mg) at room temperature under
magnetic stirring over 3 d. The combined extracts (Table 3, entry
3) were evaporated and the residue was dissolved in hexane at a
concentration of 50.0 mg/mL. In order to obtain similar DAD re-
sponses, well centred in the calibration curve, different aliquots
NaBH4 (57 mg, 1.5 mmol) was added to a solution of those pro-
tected esters in dry THF (5 mL) and the mixture stirred for 24 h.
Then acetone (2 mL) was poured into the flask and stirred for
1 h. The solution was concentrated and the residue was applied
to a preparative plate of silica gel and eluted with hexane/ethyl
acetate 1:1. The band at Rf = 0.65 furnished pure caffeate esters
of the sterols (43 mg, ꢀ0.073 mmol, 49% yield).
of each sample (100–400
lL) were evaporated to dryness, dis-
solved with 500 L of stock solution IS (umbelliferone) in dec-
l
ane/2-propanol 96:4 and then diluted to 1.0 mL with decane
(Table 3, entry 6). Sunscreen oil was simply diluted to 50.0 mg/mL.
2.5. Syntheses of coumarate and caffeate derivatives
The above procedure was also applied to coumaric acid
(164 mg, 1.0 mmol) to prepare (2E)-3-[4-(acetyloxy)phenyl]-2-
propenoic acid (184 mg, 0.89 mmol, 89% yield). The subsequent
treatment with oxalyl chloride and one-pot esterification of the
sterols (604 mg, ꢀ1.42 mmol) gave pure (2E)-3-[4-(acetyl-
oxy)phenyl]-2-propenoates (84 mg, ꢀ0.14 mmol, 15% yield) after
separation by silica column. The deprotection step was similarly
carried out in order to obtain pure coumarate esters of the sterols
Coumarate and caffeate esters of
c-oryzanol steroid moieties
have been reported in the literature (Norton, 1995; Seitz, 1989;
Yu et al., 2007). Since the samples analysed herein were short of
such esters we prepared them by following synthesis.
Sodium hydroxide (4 g, 0.1 mol) was dissolved in water (6 mL)
and diluted with ethyl alcohol (24 mL). Standard
(600 mg, ꢀ1 mmol) was added to this solution in a 100 mL round
bottomed flask and was stirred at 80 °C for 2.5 h. After cooling at
room temperature (r.t.), the solution was poured into a separatory
c-oryzanol
from
c-oryzanol (34 mg, ꢀ0.06 mmol, 43% yield).

2084
M. D’Ambrosio / Food Chemistry 138 (2013) 2079–2088
2.6. Method validation
was performed on three different days, each time with a newly
prepared mobile phase and sample.
2.6.1. Standard solutions and calibration curves
The internal standard (IS) umbelliferone (50.7 mg) was dis-
solved in 2-propanol in 50 mL volumetric flask. Then 2.7 mL of
solution were poured in 50 mL volumetric flask, evaporated to dry-
ness and dissolved in decane/2-propanol 96:4 to prepare a stock
solution IS which was used for sample preparation. A second solu-
tion of IS stock at halved concentration and decane/2-propanol
98:2 was obtained by diluting 5.0 mL of IS stock solution with pure
decane in 10 mL volumetric flask. This IS stock was used for serial
dilutions of the calibration curve. The external standard (ES) ethyl
ferulate (52.5 mg) was dissolved in 2-propanol in 50 mL volumet-
3. Results and discussion
3.1. Isolation and structural characterisation of
components
c-oryzanol
3.1.1. Preparative NP-HPLC
A sample of standard c-oryzanol (9 mg) was dissolved in decane
and irradiated at UV light (k = 254 nm) for one hour then subjected
to NP-HPLC at semi-preparative scale; the profile of the chromato-
gram was similar to that shown in Fig. 2a and sample components
were separated into four peaks. Owing to their low intensity, the
peaks at tR = 11.5 and 12.0 min were initially collected together
and those three fractions used to acquire NMR spectra. Each of
the four peaks was also collected and then investigated by RP-
LC–ESI-MS technique.
ric flask then 100.0
evaporated to dryness and dissolved with the IS stock to prepare
the highest of seven different concentration levels. Standard
lL were poured into a 4 mL volumetric flask,
c
-
oryzanol was used for recovery studies. All stock solutions were
stored in brown bottles and vials at 4 °C.
2.6.2. Linearity
3.1.2. NMR spectroscopy
Linearity was established by injection of the seven standard
concentrations with four replicates. The calibration graph was
plotted using linear regression of the mean Signal Ratio (area ES/
The elucidation of 1H NMR spectra of those three fractions (see
Suppl. Mat., Fig. 1) proved that our chromatographic method
(Fig. 2a) allows the separation of cis- from trans-ferulates and each
group of esters is split into cholestane and cycloartane skeletons
(Fig. 1). The peak collected at tR = 20.0 min revealed diagnostic res-
onances which account for: one inductively deshielded proton
(d = 4.71, dd, J = 11.0, 4.5) at C3 and the methylene protons
(d = 0.36, d, J = 4.1; d = 0.60, d, J = 4.1) at C19 of the cyclopropane
ring (structures A), one olefinic proton (d = 5.10, tm, J = 7.0, 2.5)
at C24 and the allylic methyls C26 (d = 1.68, br.s) and C27
(d = 1.61, br.s) (structure A2), the exomethylene protons (d = 4.67,
m, J = 1.2; d = 4.72, br.s) at the C24 = CH2 and the methyls C26
and C27 (d = 1.02 and 1.03, d, J = 6.7) (structure A3) (Chen et al.,
2008). The peak at tR = 19.0 min showed: deshielded multiplets
assignable to the proton (d = 4.7–4.8, m) at C3 and many shielded
methyls at high field (d = 0.65–0.95) (structures C), the olefinic pro-
ton (d = 5.40, br.d, J = 5.0) at C6 (structures C3, C6 and C8) (Chen
et al., 2008). Both 1H NMR spectra proved that the peaks at
tR = 19.0 and 20.0 min consisted of trans-ferulate esters on the ba-
sis of characteristic chemical shifts and coupling constants: d 7.60
(d, J = 15.9, H-30), 7.08 (dd, J = 8.2, 1.8, H-90), 7.04 (d, J = 1.8, H-50),
6.91 (d, J = 8.2, H-80), 6.30 (d, J = 15.9, H-20), 3.94 (s, MeO–) (Chen
et al., 2008). The peaks at tR = 11.5 and 12.0 min were also exam-
ined by 1H NMR spectroscopy and revealed the same diagnostic
resonances attributed to the cholestane and cycloartane skeletons.
On the contrary, the proton pattern observed at low field fitted a
cis-ferulate moiety: d 7.74 (d, J = 1.8, H-50), 7.13 (dd, J = 8.2, 1.8,
H-90), 6.88 (d, J = 8.2, H-80), 6.78 (d, J = 13.0, H-30), 5.84 (d,
J = 13.0, H-20), 3.93 (s, MeO–).
area IS) vs concentration ES (nmol/lL).
2.6.3. Limits of detection and quantification
The sample solution Vialone Nano was further diluted in dec-
ane/2-propanol 98:2 in order to determine the limit of detection
(LOD) and quantification (LOQ) values which were established at
a signal-to-noise ratio S/N = 3 and S/N = 10 respectively.
2.6.4. Accuracy
Recovery tests were studied at four concentration levels (ꢀ25%,
60%, 100% and 130% of the expected oryzanol content), each sam-
ple was analysed in four replicates and the amount of c-oryzanol
was calculated from the calibration curve. Two sets of experiments
were performed: (1) by spiking oil aliquots after the extraction
procedure and (2) raw material aliquots before the extraction pro-
cedure. The oil extracted from 2010.0 mg of Carnaroli rice bran was
divided into five identical aliquots, four of which were spiked with
one each of the above defined concentration levels of c-oryzanol.
After analysis, the accuracy was assessed using the equation:
Calibration recoveryRC ð%Þ ¼ ðamount fortifiedPOST
ꢂ amount originalÞ
ꢁ 100=amount of spike
An accurate amount (one for each level) of commercial c-oryz-
anol in hexane was added to 400 mg of Carnaroli rice bran and
evaporated, the raw material was left for one hour then extracted
and analysed as described above. The accuracy was assessed using
the equation:
3.1.3. LC–MS spectroscopy
The four single peaks isolated by preparative NP-HPLC, were
investigated by RP-LC–ESI-MS technique observing negative ions.
As expected, the reversed phase chromatography (Fig. 3) splits
up each fraction obtained from normal phase into several peaks.
The chromatograms of peaks at tR = 19.0 and 20.0 min aren’t
shown because they appeared superimposable (identical pattern
and slight shifts in the retention times) to those at tR = 11.5 min
(Fig. 3a) and 12.0 min (Fig. 3b) respectively. The RP-HPLC chro-
Apparent recoveryRꢃ ð%Þ ¼ ðamount fortifiedPRE
ꢂ amount originalÞ
ꢁ 100=amount of spike
2.6.5. Precision
The repeatability of the chromatographic method was tested
using measurements of the intra- and inter-day variability. The
precision was examined using standard solutions at the concentra-
tion of 1.18 ꢁ 10ꢂ4 M which were injected six times. Coefficients of
variations were calculated for both retention times (tR) and inte-
gration area. The intra-day variability was determined by analysing
the solution within one single day; the inter-day reproducibility
matogram of standard c-oryzanol is shown in Fig. 3c.
The results of MS data are summarised in Table 2 which lists the
structures according to the elution order with increasing retention
times. Columns M (Xu & Godber, 1999) and N (Yu et al., 2007) list
the steryl moieties identified by previous authors. We used a meth-
od which parts c-oryzanol into six major peaks; ten molecular ions

M. D’Ambrosio / Food Chemistry 138 (2013) 2079–2088
2085
Fig. 3. RP-HPLC runs of: (a) peak collected at tR = 11.5 min, steryl cis-ferulates possessing a saturated side chain, (b) peak collected at tR = 12.0 min, steryl cis-ferulates
possessing an unsaturated side chain, (c) standard -oryzanol, (d) synthetic steryl p-coumarates, (e) synthetic steryl caffeates.
c

2086
M. D’Ambrosio / Food Chemistry 138 (2013) 2079–2088
were detected and assigned to the most likely structures by com-
parison with previous literature data. These compounds belonged
to two groups corresponding to each NP-HPLC fraction: peaks at
tR = 11.5 and 19.0 min (col. O) and peaks at tR = 12.0 and 20.0 min
(col. P). The relative proportion of coeluting components in braces
at entries 6, 7 and 8 (col. O) was established by integration of MS
extracted ion chromatograms thus corresponding to 49%, 19%
and 32% respectively.
NMR spectroscopy characterised those NP-HPLC fractions in
terms of major steroid skeletons and also distinguished between
cis and trans isomers. The sensitive MS technique allowed to iden-
tify minor or even trace components which constitute each chro-
matographic peak. In fact, the presence of some less abundant
compounds (Table 2, entries 1,2 and 7) demonstrates that the frac-
tions being eluted later (tR = 12.0 and 20.0 min) (Table 2, P) contain
steroid moieties which possess a double bond at the side chain
whereas the fractions being eluted sooner (tR = 11.5 and
19.0 min) (Table 2, O) contain compounds which are saturated at
the side chain and, possibly, also at the polycyclic ring system. Tak-
ing into account that compounds at entries 1, 2 and 7 (Table 2) are
between those stationary phases is endcapping of residual free
silanols.
3.3.2. Selection of detection wavelength
The detection wavelength was set at 325 nm, close to the kmax
of the ferulate chromophore. The p-coumarate moiety has strong
absorption at 290 nm and marginal absorption at 340 nm: compar-
ison of these chromatograms points to the elution of those very
minor esters (Norton, 1995; Seitz, 1989).
3.3.3. Selection of standard compounds
Ethyl ferulate was chosen as an external standard because it
possesses the identical chromophore of
and has a precise molecular weight (Yu et al., 2007). On the con-
trary, -oryzanol is a mixture of cholestane and cycloartane alkyl
c-oryzanol components
c
ferulates whose actual composition depends on its vegetable origin
and hence has no well defined molecular weight. Umbelliferone (7-
hydroxycoumarin) was chosen as an internal standard because it
shows an UV spectrum similar to the ferulate moiety, elutes far
from the peaks of interest and cannot suffer cis–trans isomerism.
Simple coumarin elutes close to ethyl ferulate and might superim-
pose to the signal of investigated compounds. However, umbellif-
erone showed low solubility in pure decane, so it was necessary
to add 2% (v/v) of 2-propanol and to restrain its concentration.
very minor components of c-oryzanol, the NMR attribution of cho-
lestane and cycloartane skeletons to the NP-HPLC fractions is still
acceptable.
3.2. Chromatography and spectroscopy of synthetic coumarate and
caffeate derivatives
3.4. Method validation
3.4.1. Linearity, LOD, LOQ and peak resolution
The regression equation of ethyl ferulate was y = 8.7136x in
which y is the signal ratio (peak area/IS area) at 325 nm and x is
Both 1H NMR spectra of coumarate and caffeate esters of the
sterols from c-oryzanol revealed the diagnostic resonances attrib-
utable to cholestane and cycloartane skeletons as described at the
Section 3.1.2. However, the proton pattern observed at low field
was in agreement with either a coumarate {d 7.61 (d, J = 15.9, H-
30), 7.44 (d, J = 8.4, H-50 and H-90), 6.84 (d, J = 8.4, H-60 and H-80),
6.31 (d, J = 15.9, H-20)} or a caffeate {d 7.50 (d, J = 15.9, H-30), 7.03
(d, J = 1.9, H-50), 6.91 (dd, J = 7.9, 1.9, H-90), 6.80 (d, J = 7.9, H-80),
6.20 (d, J = 15.9, H-20)} ester.
Synthetic coumarates and caffeates were subjected to analytical
chromatography in both NP- and RP-HPLC mode. In the normal
phase conditions, trans-ferulates eluted at tR = 17.7 and 18.5 min
(Fig. 2b), trans-p-coumarates separated into two peaks at
tR = 21.2 and 21.8 (Fig. 2c); trans-caffeates eluted as one peak at
tR = 26.8 min (Fig. 2d). In the reversed phase conditions the profile
of coumarate components (Fig. 3d) was superimposable (identical
pattern and slight shifts in the retention times) to those of ferulates
(Fig. 3c) whilst caffeate components (Fig. 3e) eluted as broad, tail-
ing peaks at tR shorter than ferulates. The molecular ions [M–H]ꢂ
listed in Table 2 (col. Q and R) were detected by LC–ESI-MS tech-
nique thus confirming the presence of p-coumarate and caffeate
esters of the native steroid moieties.
the concentration of analyte in nmol/
l
L (ꢄ mol/mL ꢄ mmol/L).
l
The determination coefficient was R2 = 0.99997 and the calibration
graph was linear in the range 8.86 ꢁ 10ꢂ7–1.18 ꢁ 10ꢂ4 M. The
developed method shows excellent linear response over a large
range of concentrations. Good sensitivity was demonstrated with
LOD and LOQ at 1.5 ꢁ 10ꢂ7 and 6.0 ꢁ 10ꢂ7 M respectively for both
cholestane and cycloartane alkyl ferulates. Under the chromato-
graphic conditions employed in the current study, the resolution
was >1.5 for the cis-ferulate pair of peaks and in the range 1.15–
1.20 for the trans-ferulate pair.
3.4.2. Accuracy
Apparent recoveries R⁄ (%) were acceptable in all tested concen-
trations: 89.9 (25%), 81.3 (60%), 71.8 (100%) and 69.3 (130%). We
expected these results because the apparent recovery is related
to the overall systematic errors of the whole analytical process
and the extraction step is not optimised. According to the sugges-
tion by Spanish analysts (Garrido-Frenich et al., 2006), we also
evaluated the term named ‘‘calibration recovery’’ which is related
to the systematic measurement errors during the quantification
process. The calibration recovery RC (%) were: 100.1 (25%), 100.7
(60%), 98.9 (100%) and 100.8 (130%). These values mean that the
chromatographic system shows a high accuracy during the quanti-
fication steps.
3.3. Quantitation of c-oryzanol components optimisation of
chromatographic conditions
3.3.1. Column choice
Silica gel was our first choice as stationary phase column, it pro-
vided a good separation of oil constituents using an isocratic elu-
tion with hexane/2-propanol 99:1. However, the column
performances soon revealed themselves to be unreliable probably
owing to deactivation of the silica surface from adsorption of water
(Diack & Saska, 1994; Hewavitharana, 2003). Subsequently, we
tried two cyanopropyl bonded stationary phases. Both gave chro-
matograms of comparable resolution and were used to elute the
sequence of injections for the calibration curve. One of them
showed a shorter dynamic range and failure of linearity at low con-
centration levels. To the best of our knowledge, the only difference
3.4.3. Precision
Repeatability was a hard task to achieve and several instrumen-
tal options were tried. To our surprise, the problem was seen to
originate from the solvent which the sample was dissolved in espe-
cially in the case of incompletely filled vials. In fact, hydrocarbons
like n-hexane, n-heptane and isooctane, possess a vapour pressure
high enough to slowly concentrate the sample during the time of
chromatographic elution thus giving an increasing area of inte-
grated peaks. The evaporation can be cut if vials pending in the
autosampler are completely filled. In order to prepare working
solutions at a stable concentration, we preferred to solve the prob-

M. D’Ambrosio / Food Chemistry 138 (2013) 2079–2088
2087
lem by dissolving samples in decane/2-propanol 98:2, the small
percentage of alcohol being necessarily added to help the dissolu-
tion of umbelliferone. Eventually, the precision at the concentra-
tion of 1.18 ꢁ 10ꢂ4 M was satisfactory because the intra- and
inter-day variability of the signal ratios were 1.03 0.004 and
1.03 0.007 with relative standard deviation (RSD) of 0.4 and
0.7% respectively; the intra- and inter-day variability of the tR were
24.7 0.04 and 24.7 0.09 with RSD of 0.2% and 0.4% respectively.
1. Total elution time is comparable or even shorter and a baseline
separation is easily obtained.
2. The quantity of each steryl moiety is partially lost but the
knowledge of the relative proportion of cholestane/cycloartane
alkyl ferulates (Table 3, entry 18) is sufficient and valuable
information. In fact, c-oryzanol contains cycloartane alkyl feru-
lates in remarkable amount when extracted from rice and neg-
ligible from other cereals.
3. Unique clear-cut separation of all the CAD esters is possible. p-
Coumarates are abundant in corn bran so both points 2 and 3
3.5. Application to cereals and commercial samples
indicate the extraction source of c-oryzanol.
4. Exclusive information with regards to cis/trans isomers is
achieved. cis-Ferulates were absent in freshly extracted oils so
they are artefacts of manipulation and storage. Their presence
We examined the bran from two varieties of rice, one variety of
wheat, the flour of corn and rye, a supplement tablet and a sun-
screen oil. The percentage of rice bran oil was similar to previous
reported values (Lerma-García et al., 2009) (Table 3, entry 4). The
moisture content can be roughly deduced (Table 3, entry 5). Each
oil was then diluted to a concentration that gives a signal ratio fall-
ing into the linear range of the calibration curve (Table 3, entry 6).
The concentration of ferulate esters in our samples was estimated
by the regression equation and is precisely expressed in nmol/mL
(Table 3, entries 7–12). The data are the mean of four replicates
and show a high precision of the HPLC system with RSD < 1 in most
cases. Taking into account the oil concentration (Table 3, entry 6),
the content of ferulate esters could be related to the mass of oil
(Table 3, entry 13–15). Assuming that the molecular weights of
cholestane and cycloartane alkyl ferulates average 583 and
609 g/mol respectively, our data were approximately changed to
provides clues about the preservation of products and of
oryzanol as their ingredient (Table 3, entry 19).
c-
5. It is evident that only NP-HPLC is promising method to be
applied for the routine quantification of a large number of sam-
ples without any preliminary step of purification.
6. In the present method, crude oils or extracts can be injected; no
long column rinsing is required so that multiple, automated
chromatographic elutions can be carried out.
In conclusion, we established a fast and sensitive method for
the quantification of
c-oryzanol and it could be a suitable tool
for quality assurance and determination of origin.
Acknowledgment
mg/g of c-oryzanol into the mass of sample or oil (Table 3, entries
16 and 17). The relative proportion of cholestane and cycloartane
components points out the vegetable origin (Table 3, entry 18).
The percentage of cis-ferulate isomers is also shown (Table 3, entry
19).
Authors thank the chef Gabriele Ferron (FERRON Gabriele e
Maurizio s.n.c.) and Dr. G. Peratoner (Centro di Sperimentazione
Agraria e Forestale) for supplying rice and rye samples. The excel-
lent technical assistance of Mr. Sandro Gadotti is greatly appreci-
ated. The author is much indebted to Dr. Petru Harghel for his
help in the syntheses. We also thank Dr. A. Luppi and one of the
referees for his suggestions on the evaporation of incompletely
filled vials. This work has been supported by Provincia Autonoma
di Trento (project CRS-2007).
3.6. Comparison of various determination methods
The performances of several chromatographic methods are out-
lined in Table 1. It is interesting to note that gradient elutions in
RP-HPLC (col. C, entries 7 and 8) have been applied to samples with
no preliminary treatment of purification (col. D) in order to wash
the column out of apolar compounds by 100% organic solvents.
The identification of peaks separated by NP chromatography has
been performed by the present author for the first time (col. E).
In general, the shorter the elution time (col. F) the lesser the num-
ber of peaks (col. G) and the worst the baseline separation. Without
using IS, the total elution time of the method proposed herein,
would have been shorter. Cereal phytosterols can be esterified by
different CAD derivatives (col. H). RP-HPLC provides the separation
of caffeate (entry 7) from ferulate which coelute with coumarate
esters. Coumarates elute as shoulders of large ferulate peaks (col.
H, entries 2 and 3) so it is difficult to achieve well resolved peaks
and their quantification is not reliable. On contrary, NP-HPLC
shows a very good resolution (Fig. 2). The author for the first time
identified the peaks of cis-ferulate isomers and reported their
retention times in the HPLC chromatogram by an NP approach
Appendix A. Supplementary data
Supplementary data associated with this article can be found, in
the online version, at http://dx.doi.org/10.1016/j.foodchem.2012.
12.008.
References
Akihisa, T., Yasukawa, K., Yamaura, M., Ukiya, M., Kimura, Y., Shimizu, N., et al.
(2000). Triterpene alcohol and sterol ferulates from rice bran and their anti-
inflammatory effects. Journal of Agricultural and Food Chemistry, 48, 2313–2319.
Chen, M. H., & Bergman, C. J. (2005). A rapid procedure for analysing rice bran
tocopherol, tocotrienol and c-oryzanol contents. Journal of Food Composition and
Analysis, 18, 139–151.
Chen, C. R., Wang, C. H., Wang, L. Y., Hong, Z. H., Chen, S. H., Ho, W. J., et al. (2008).
Supercritical carbon dioxide extraction and deacidification of rice bran oil.
Journal of Supercritical Fluids, 45, 322–331.
Diack, M., & Saska, M. (1994). Separation of vitamin E and c-oryzanol from rice bran
(col. I). In case of quantitative measurements of c-oryzanol, it is re-
ported the number of cereal samples which were analysed (col. J)
and if the validation tests were carried out (col. K).
oil by normal-phase chromatography. Journal of the American Oil Chemists’
Society, 71, 1211–1217.
Garrido-Frenich, A., Romero-González, R., Martínez-Vidal, J. L., Plaza-Bolaños, P.,
Cuadros-Rodríguez, L.,
recovery profiles using gas chromatography
&
Herrera-Abdo, M. A. (2006). Characterisation of
Triple quadrupole mass
–
spectrometry for the determination of pesticide residues in meat samples.
Journal of Chromatography A, 1133, 315–321.
4. Conclusion
Heinmann, R. J. B., Xu, Z., Godber, J. S., & Lanfer-Marquez, U. M. (2008). Tocopherols,
tocotrienols and c-oryzanol contents in japonica and indica subspecies of rice
It is rare to find in the literature examples of quantitative anal-
yses by NP chromatography, perhaps because the RP approach is
(Oryza sativa L.) cultivated in Brazil. Cereal Chemistry, 85, 243–247.
Hakala, P., Lampi, A. M., Ollilainen, V., Werner, U., Murkovic, M., Wähälä, K., et al.
(2002). Steryl phenolic acid esters in cereals and their milling fractions. Journal
of Agricultural and Food Chemistry, 50, 5300–5307.
the preferred one. As to concern
c-oryzanol, the inspection of Table
3 suggests the following advantages about the NP over RP-HPLC:

2088
M. D’Ambrosio / Food Chemistry 138 (2013) 2079–2088
Hewavitharana, A. K. (2003). Simple solutions to problems encountered in
quantitative analysis of tocopherols and tocotrienols using silica columns.
Analytical Biochemistry, 313, 342–344.
Norton, R. A. (1995). Quantitation of steryl ferulate and p-coumarate esters from
corn and rice. Lipids, 30, 269–274.
Seitz, L. M. (1989). Stanol and sterol esters of ferulic and p-coumaric acids in wheat,
corn, rye and triticale. Journal of Agricultural and Food Chemistry, 37, 662–667.
Stöggl, W., Huck, C., Wongyai, S., Scherz, H., & Bonn, G. (2005). Simultaneous
Huang, S. H.,
& Ng, L. T. (2011). An improved high-performance liquid
chromatographic method for simultaneous determination of tocopherols,
tocotrienols and
4709–4713.
c
-oryzanol in rice. Journal of Chromatography A, 1218,
determination of carotenoids, tocopherols and c-oryzanol in crude rice bran oil
by liquid chromatography coupled to diode array and mass spectrometric
detection employing silica C30 stationary phases. Journal of Separation Science,
28, 1712–1718.
Lerma-García, M. J., Herrero-Martinez, J. M., Simó-Alfonso, E. F., Mendonça, C. R. B.,
& Ramis-Ramos, G. (2009). Composition, industrial processing and applications
of rice bran
Miller, A., & Engel, K. H. (2006). Content of
c
-oryzanol. Food Chemistry, 115, 389–404.
-oryzanol and composition of steryl
Xu, Z., & Godber, J. S. (1999). Purification and identification of components of c-
c
oryzanol in rice bran oil. Journal of Agricultural and Food Chemistry, 47,
2724–2728.
Yu, S., Nehus, Z. T., Badger, T. M., & Fang, N. (2007). Quantification of Vitamin E and
ferulates in brown rice (Oryza Sativa L.) of European origin. Journal of
Agricultural and Food Chemistry, 54, 8127–8133.
Moreau, R. A., Whitaker, B. D., & Hicks, K. B. (2002). Phytosterols, phytostanols and
their conjugates in foods: Structural diversity, quantitative analysis and health-
promoting uses. Progress in Lipid Research, 41, 457–500.
c
-oryzanol components in rice germ and bran. Journal of Agricultural and Food
Chemistry, 55, 7308–7313.