Kinetic studies of cholesterol oxidation as inhibited by stearylamine during heating

Article abstract of DOI:10.1021/jf048951+

The formation of cholesterol oxidation products (COPs) during heating in the presence of stearylamine at 140°C was analyzed by high-performance liquid chromatography (HPLC) and kinetically studied by use of nonlinear regression models. Results indicated that the COPs concentration increased with increasing heating time, and stearylamine was shown to reduce both oxidation and degradation rates of cholesterol. Without stearylamine, the highest rate constant (per hour) was observed for epoxidation (545.4), followed by free radical chain reaction (251.0), reduction (147.3), dehydration (95.8), triol dehydrogenation (4.7), degradation (0.34), triol formation (0.31), and dehydrogenation (0.13). With stearylamine, the epoxidation and free radical chain reaction rates could be reduced by about 800- and 3.4-fold, respectively, and triol formation during oxidation could be completely inhibited. In addition, the reactions for reduction, dehydration, degradation, and dehydrogenation could proceed slower in the presence of stearylamine. The kinetic model developed in this study can be used to predict the inhibition of COPs formation by stearylamine during heating of cholesterol.

Full text of DOI:10.1021/jf048951+

7132 J. Agric. Food Chem. 2004, 52, 7132−7138
Kinetic Studies of Cholesterol Oxidation as Inhibited by
Stearylamine during Heating
JOHN-TUNG CHIEN, DONG-YONG HUANG, AND BING-HUEI CHEN*
Department of Nutrition and Food Science, Fu Jen University, Taipei, Taiwan 242, Republic of China
The formation of cholesterol oxidation products (COPs) during heating in the presence of stearylamine
at 140 °C was analyzed by high-performance liquid chromatography (HPLC) and kinetically studied
by use of nonlinear regression models. Results indicated that the COPs concentration increased
with increasing heating time, and stearylamine was shown to reduce both oxidation and degradation
rates of cholesterol. Without stearylamine, the highest rate constant (per hour) was observed for
epoxidation (545.4), followed by free radical chain reaction (251.0), reduction (147.3), dehydration
(95.8), triol dehydrogenation (4.7), degradation (0.34), triol formation (0.31), and dehydrogenation
(0.13). With stearylamine, the epoxidation and free radical chain reaction rates could be reduced by
about 800- and 3.4-fold, respectively, and triol formation during oxidation could be completely inhibited.
In addition, the reactions for reduction, dehydration, degradation, and dehydrogenation could proceed
slower in the presence of stearylamine. The kinetic model developed in this study can be used to
predict the inhibition of COPs formation by stearylamine during heating of cholesterol.
KEYWORDS: Cholesterol oxidation; stearylamine; heating; kinetics; HPLC
INTRODUCTION
this study was undertaken to determine the kinetics of cholesterol
oxidation with or without stearylamine during heating.
Food products such as meats and eggs are rich in cholesterol,
which can be susceptible to formation of cholesterol oxidation
products (COPs) during processing (1). Many reports have
shown that most COPs are mutagenic and carcinogenic (2, 3).
Thus, the inhibition of COPs formed during heating of
cholesterol is important. To date, the oxidation pathways of
cholesterol have been well documented (4), and the mechanism
of cholesterol oxidation is reported to be similar to that of lipid
oxidation; that is, it can undergo a series of free radical chain
reactions to form COPs and other degradation products (5). The
authors also pointed out that the cholesterol oxidation can be
initiated by the second-order reaction, and the first-order
reactions follow afterward (5).
The addition of lysine to soy proteins (6) or cysteine or
alanine to fish oils (7, 8) has been shown to prevent lipid
oxidation during thermal processing. Likewise, stearylamine has
long been incorporated into cationic liposomes, which can be
used to encapsulate various materials as possible delivery
vehicles for drugs, enzymes, antibodies and gene (9-11). The
mechanism of primary and secondary amines for exhibiting
antioxidative activity in soybean oil has been attributed to the
protective effect of oxidized lipid/amine reaction (12). Thus,
understanding the reactions that lead to inhibition of COPs
formation will provide a basis for developing food products with
better quality. As there is no information available about the
inhibition of COPs as affected by amino-containing compounds,
MATERIALS AND METHODS
Materials. Stearylamine, cholesterol and COPs standards, 7R- and
7â-hydroxycholesterol, 7-ketocholesterol, 5,6R- and 5,6â-epoxycho-
lesterol, and 5R-cholestane-3â,5,6â-triol were purchased from Sigma
Chemical Co. (St. Louis, MO) and Steraloids Inc. (Wilton, NH). These
standards were used without further purification. Spraying reagent N,N-
dimethyl-p-phenylenediamine dihydrochloride was also from Sigma.
TLC adsorbent silica gel 60 GF254, reagent benzoyl peroxide, and
paraffin oil were from Merck Co. (Darmstadt, Germany). The analytical-
grade solvents lauryl alcohol, chloroform, benzene, methanol, and ethyl
acetate were obtained from Merck Co. Acetone, acetic acid, and sulfuric
acid were from Hau-Fong Co. (Taipei, Taiwan). The HPLC-grade
solvents n-hexane and 2-propanol were from Mallinckrodt Co. (Paris,
KY).
Instrumentation. The HPLC instrument consists of a Jasco PU-
980 pump (Tokyo, Japan), a Jasco 830 refractive index detector, and
a SIC Chromatocoder 12 integrator (System Instruments Co., Tokyo,
Japan). A Lichrospher 100 CN column (244 × 4.0 mm i.d.) containing
5-µm packing material was from Merck Co. (Darmstadt, Germany). A
CHEN-WIN computer software system (Shuen-Hua Co., Taipei,
Taiwan) was used to process data.
Heating of Cholesterol. A mixture containing 0.2 mL of paraffin
oil and 0.08 mL of lauryl alcohol was poured into a 100-mL round-
bottom flask of a rotary evaporator, which was preheated in a silicon-
oil bath (145-150 °C) for 5 min with nitrogen flushed into the flask
at the same time. Then 100 mg of cholesterol standard was added alone
or mixed with 70 mg of stearylamine to the flask. A thin film of the
mixture was heated for 2 min so that the internal temperature could
reach 140 °C. This temperature was selected to avoid thermal
* Corresponding author: telephone 886-2-29031111, ext 3626; fax 886-
2-29021215; e-mail nutr1007@mails.fju.edu.tw.
10.1021/jf048951+ CCC: $27.50
© 2004 American Chemical Society
Published on Web 10/16/2004

Kinetic Studies of Cholesterol Oxidation
J. Agric. Food Chem., Vol. 52, No. 23, 2004 7133
degradation of cholesterol during 2-min preheating and facilitate
formation of COPs afterward. In the alcoholic system, a significant
thermal degradation of cholesterol was found at 150 °C or above. In
addition, this temperature can be used for frying of food products (low-
temperature frying). Meanwhile, oxygen was pumped through at a rate
of 10 mL/min and the heating times, 0, 10, 20, 40, 60, and 120 min,
started to count. The dissolved oxygen content was not calculated but
was assumed to be constant during heating. This is important to make
the kinetic study of cholesterol oxidation much easier as the number
of variables can be reduced. After heating, the flask was cooled in dry
ice to terminate the reaction.
Purification of Cholesterol and COPs. Twenty milliliters of
n-hexane-2-propanol (3/2 v/v) was added to the flask and the mixture
was vigorously shaken for 3 min to dissolve cholesterol and COPs.
The solution was subjected to centrifugation at 26000g for 5 min, after
which the upper phase was collected and poured into a centrifuged
tube. Twelve milliliters of water was added and the mixture was
centrifuged again for 5 min. The upper layer was collected and
evaporated to dryness for purification. The residue was dissolved in 1
mL of hexane/1,2-dichloroethane (1:1 v/v) and poured into a NH2
cartridge. Initially 5 mL of hexane was added to remove impurities
such as hydrocarbons, cholesterol esters and triglycerides. COPs were
next eluted with hexane/1,2-dichloroethane/2-propanol (50/30/15
v/v/v). The eluate was evaporated to dryness at 35 °C, and the residue
was dissolved in 1 mL of hexane/2-propanol (95/5 v/v) and filtered
through a 0.2-µm membrane filter for HPLC analysis. Duplicate
treatments and triplicate analyses were performed, and the data were
subjected to nonlinear regression analysis by use of SAS (13). A high
recovery of 97-100% was obtained when both cholesterol and COPs
standards were subjected to the same extraction and purification method.
TLC Analysis of COPs: (a) Preparation of Wurster Dye. The
Wurster dye was prepared by a method described by Smith and Hill
(14). Briefly, 1 g of N,N-dimethyl-p-phenylenediamine dihydrochloride
was dissolved in 100 mL of 50% methanol solution (in water), and the
mixture was shaken thoroughly. Glacial acetic acid (1 mL) was added,
and the solution was poured into a glass vial and stored at -20 °C
until use.
Figure 1. Pathways of cholesterol degradation and oxidation. A
)
cholesterol; A′ ) 7-OOH (7-hydroperoxycholesterol); B
hydroxycholesterol); C 7-Keto (7-ketocholesterol); D
products; E 5,6-EP (5,6-epoxycholesterol); O
one; S 5,6-epoxycholesterol stearylamine reaction products; T
cholestan-3 ,5 ,6 -triol; k1 k8, k3 , k6
rate constants. Thermal degradation, A
B or C; epoxidation: other oxidized products.
)
)
7-OH (7-
degraded
)
)
)
cholestan-3
â
,5R
,6-
)
)
−
â
R
â
∼
′
′
, and k7′ ) the corresponding
D; C-7 oxidation, A
f
f A′ f
A
f E f
iterative methods until the convergence of the best-fitted parameters
were met. The rate constant (per hour) and correlation coefficient of
each kinetic equation of cholesterol oxidation and degradation were
determined. The precision of the parameters of the kinetic equations
was also assessed.
(b) Separation of COPs by TLC. Development of the TLC plate
was carried out in a glass tank lined with a filter paper and equilibrated
for 30 min with 200 mL of benzene/ethyl acetate (60/40 v/v) (15) prior
to development. A 10-µL volume of extract was spotted on the glass
plate with a micropipet. The chromatogram was developed for a distance
of 16 cm at ambient temperature, after which the plate was dried in an
oven at 110 °C for 10 min and sprayed with 50% H2SO4, and the color
development of COPs under UV radiation at 254 nm was observed.
Both 7R- and 7â-hydroperoxycholesterol bands were identified and
quantified by the Wurster dye method as described in a previous study
(16).
HPLC Analyses of COPs. An isocratic mobile phase of n-hexane/
2-propanol (95/5 v/v) with flow rate 1.0 mL/min and refractive index
detection (sensitivity 16 × 10-5 RIU) was used (5). The injection
volume was 100 µL, and two LiChrospher 100 CN columns controlled
at 30 °C were used for separation of COPs. The various COPs were
identified by comparison of retention times of unknown peaks with
reference standards and addition of standards to sample for cochro-
matography, as well as collection of eluates for TLC analysis.
Because of absence of a suitable internal standard, each COP was
quantified by an external calibration method. Eight concentrations of
each COP ranging from 104 to 107 ppm were injected onto an HPLC
column, and the calibration curve for each COP standard was obtained
by plotting concentration against area. The regression equations and
correlation coefficients (r2) were calculated on a CHEN-WIN computer
software system. Each COP was quantified by a method as described
by Chien et al. (5). The eluates of 7R- and 7â-hydroperoxycholesterol
were collected individually and then quantified by the Wurster dye
method (16).
RESULTS AND DISCUSSION
Percentage Changes of Cholesterol during Heating. Due
to the complexity of food itself, it is difficult to assess the
formation mechanism of COPs in real food systems. The
formation of COPs during heating of cholesterol at 150 °C has
been kinetically studied by nonlinear regression models (5).
However, the effect of stearylamine on formation or inhibition
of COPs remains uncertain. Theoretically, primary and second-
ary amines are able to react with oxidized lipids to form oxidized
lipid/amino acid reaction products, which provide protection
against lipid oxidation (12, 17).
Owing to the poor solubility of the mixture of stearylamine,
cholesterol, and COPs in water, a solvent system of lauryl
alcohol/paraffin oil (2/5 v/v) was used instead. Our preliminary
results showed that cholesterol oxidation in this solvent system
was too fast to be monitored when temperature reached 160 °C
and above. After several studies, the most appropriate temper-
ature for kinetic study of cholesterol was found to be 140 °C.
Furthermore, with nitrogen flushing, no significant oxidation
of cholesterol in this solvent system was observed during 2-min
preheating at 140 °C.
Figure 1 shows the major reaction pathways and rate
constants (per hour) of cholesterol oxidation and degradation
based on a study by Chien et al. (5). Similar to that in an aqueous
environment, 5,6R- or 5,6â-epoxycholesterol may undergo
alcoholysis to form 5R-cholestane-3â,5,6â-triol (18), which can
be further dehydrogenated to cholestan-3â,5R,6-one. In the
presence of stearylamine, â-hydroxylamine, one of the reaction
products between 5,6R- or 5,6â-epoxycholesterol and stearyl-
Kinetic Analyses of COPs. The various concentration changes of
cholesterol and COPs during heating of cholesterol were subjected to
statistical analysis by a nonlinear regression procedure (13). All the
rate constants of a nonlinear model were estimated by the least-squares
method with a NLIN (nonlinear regression) procedure-Marquardt

7134 J. Agric. Food Chem., Vol. 52, No. 23, 2004
Chien et al.
Figure 2. Percentage changes of cholesterol during heating at 140 °C.
Mean of experimental data for (
line ( ) for 0 40 min of heating and (
stearylamine and the best-fitting line (
standard errors are within 2.7% of the means (n
b
) cholesterol only and the best-fitting
) cholesterol in the presence of
) for 0 60 min of heating. All the
3).
s
−
O
s
−
±
)
amine, is probably formed through amination of epoxides (19,
20). The possible reaction for amination of epoxides is shown
below:
Figure 3. HPLC chromatograms of COPs formation during heating of
cholesterol in the absence of stearylamine at 140 °C.
Figure 2 shows the percentage changes of cholesterol during
heating at 140 °C for up to 2 h. A loss of cholesterol by 16.2%
was found after 40-min heating, and a further decrease of 86.3%
occurred after prolonged heating for 2 h. In contrast, in the
presence of stearylamine, degradation of cholesterol proceeded
more slowly over the 1-h heating period. However, the residual
percentage of cholesterol drastically dropped to 14.5% after 2-h
heating. This outcome may be attributed to the antioxidative
activity of the reaction products of stearylamine with 5,6R- or
5,6â-epoxycholesterol. Several authors also demonstrated that
some amines may react with alkenals or epoxyalkenals to
produce antioxidant compounds (12, 17). In a previous study,
Chien et al. (5) reported that approximately 33.3% cholesterol
remained after dry heating of cholesterol at 150 °C for 30 min.
This result showed a higher loss of cholesterol, which may be
explained by the heating conditions such as temperature or
solvent system.
Figures 3 and 4 show the HPLC chromatograms of COPs
formed during heating at 140 °C in the absence and presence
of stearylamine, respectively. A total of 7 COPs, including 7R-
and 7â-hydroperoxycholesterol, 5,6R- and 5,6â-epoxycholes-
terol, 7-ketocholesterol, 7R- and 7â-hydroxycholesterol, and 5R-
cholestane-3â,5,6â-triol were adequately resolved within 40 min.
Two columns were used and the separation efficiency could be
substantially improved. Nevertheless, a prolonged retention time
could occur for cholesterol and various COPs. Without stear-
ylamine, the COPs were formed gradually during the initial
heating period and sharply increased after 40-min heating, and
a plateau was achieved in 2 h. Conversely, with stearylamine,
the total percentages of COPs showed a minor increase over a
2-h heating period (Figure 5). This phenomenon further
demonstrated that stearylamine exerts a strong antioxidative
effect on cholesterol oxidation during high-temperature heating.
Alaiz et al. (17) also reported that n-octylamine possesses
antioxidative activity when added to soybean oil during heating.
Kinetic Study of Cholesterol Oxidation during Heating.
Due to the complexity of COPs formation during heating, only
Figure 4. HPLC chromatograms of COPs formation during heating of
cholesterol in the presence of stearylamine at 140 C.
°
five major COPs, 7-OOH (7R- and 7â-hydroperoxycholesterol),
7-ketocholesterol, 7-OH (7R- and 7â-hydroxycholesterol), 5,6-
EP (5,6R- and 5,6â-epoxycholesterol), and triol (5R-cholestane-
3â,5,6â-triol), as well as cholesterol and its degraded products
were selected for kinetic study. According to several previous
studies (5, 21-23), cholesterol heated in the presence of oxygen
proceeds in two major pathwayssdegradation and oxidation.
For the oxidative pathway, reactions can be divided into two
major routes of C-7 oxidation and epoxidation (Figure 1). The
oxidation reaction involved the attack of free radicals on
cholesterol to result in formation of 7-OOH, which belongs to
the second-order pathway, and the rate equation for 7-OOH
formation was shown as in a previous study (5). According to
the C-7 oxidation route in the reaction pathways of cholesterol
during heating (Figure 1), 7-OOH can be further degraded to
7-OH and 7-keto through the first-order reaction (5, 24). The

Kinetic Studies of Cholesterol Oxidation
J. Agric. Food Chem., Vol. 52, No. 23, 2004 7135
of stearylamine could be considered as a constant to [E] in the
equation as shown below:
d[S]
dt
) k7[SA]0[E] ) k7′[Ε]
(5)
where [S] is the concentrations of reaction products between
5,6-EP and stearylamine and [SA]0 is the concentration of
stearylamine at time 0, respectively. k7 and k7′ are the reaction
rate constants (per hour).
The concentration of 7-OOH formed in the initial 20-min
heating was found to be much lower than cholesterol concentra-
tion without stearylamine (1400[A′] = [A]) or with stearylamine
(88 128[A′] = [A]). The equation of thermal degradation of
cholesterol could thus be written as -d[A]/dt ) k5[A]. The
integration of this equation gave
Figure 5. Total percentage of COPs formation during heating of cholesterol
at 140
of the means (n
All the standard errors are within
°
C. (
b
) Cholesterol only. All the standard errors are within
3). ( ) Cholesterol in the presence of stearylamine.
5.9% of the means (n 3).
±4.8%
)
O
±
)
[A]t ) [A]0e-k t
(6)
5
rate equation (eq 1) for 7-OOH formation was thus modified
as follows:
Equation 1 is in the form of a Bernoulli equation (n ) 2).
By transformation of a variable [A′], the following equation
was obtained by integration:
d[A′]
dt
[A′]
) k1 1 -
[A′] - k2[A′] - k3[A′]
(
)
[A′max
]
-1
M2
M1
M2
M1
1
[A′]0
1
[A′]t )
+
-
e-M t
(7)
(
)
k1
[
]
) (k1 - k2 - k3)[A′] -
[A′]2
(1)
[A′]max
where M1 ) k1 - k2 - k3 and M2 ) k1/[A′]max. By substitution
of eq 7 into equations for d[B]/dt and d[C]/dt, the differential
equations for [B]t and [C]t could be integrated as shown in a
previous study (5). Substituting eqs 6 and 7 into eq 2, we could
obtain eq 8 by integration:
where [A′] is percentage concentration of 7-OOH and [A′]max
is the maximum attainable concentration of 7-OOH prior to
degradation; k1, k2, k3, and t are the reaction rate constants (per
hour) and time, respectively. In addition, the rate equations of
C-7 oxidation routes B and C were also shown in a previous
study (5).
In the epoxidation route, the formation of 5,6-EP from
cholesterol could be judged from the concentrations of both
cholesterol and 7-OOH (5, 21). Thus, the rate equations for both
5,6-EP formation and cholesterol depletion could be respectively
written as follows:
e(k ′-k )s
6
5
t
[E]t ) [Ε]0e-k ′t + k4[A]0e-k ′t
ds
6
6
∫
0 M2
M1
M2
1
1
+
-
e-M s
[
]
M1
[A′]0
(8)
Equation 9 and 10 were respectively obtained by integration
after substitution of eq 8 into eqs 4 and 5:
d[E]
dt
) k4[A][A′] - k6′[Ε]
(2)
(3)
[T]t )
e[k ′-k ]s
6
5
t
υ
d[A]
dt
k6′k4[A]0e-k t e(k -k ′)υ
ds dυ
8
8
6
-
) (k1 + k4)[A][A′] + k5[A]
∫
∫
0
o M2
M1
M2
1
e-M s
1
+
+
-
(
)
[
]
M1
[A′]0
where [A] and [E] are the percentage concentrations of
cholesterol and 5,6-EP, respectively. k4 and k5 are the reaction
rate constants (per hour). According to the study of autoxidation
of cholesterol in the suspended oil (4), hydrolysis of 5,6-EP
leads to formation of triol. Since there is no water present in
the solvent system, alcoholysis of 5,6-EP is most likely to occur
in our study (18). Thus, the rate equation can be written as
follows:
(9)
e(k ′-k )υ
7
5
t
r
[S] ) k ′k [A] e-k ′r
dυ dr
7
0∫ ∫
t
7
4
0 M2
M2
0
1
1
-
e-M υ
[
]
M1
M1
[A′]0
(10)
where [A]0, [A′]0, [B]0, [C]0, [E]0, and [T]0 are percentage
concentrations of cholesterol, 7-OOH, 7-OH, 7-keto, 5,6-EP,
and triol, respectively, at heating time 0 min. Likewise, [A]t,
[A′]t, [B]t, [C]t, [E]t, [S]t, and [T]t are percentage concentrations
of cholesterol, 7-OOH, 7-OH, 7-keto, 5,6-EP, 5,6-EP-stearyl-
amine reaction products, and triol at heating time t min, and r,
s, t, and υ are time.
For the above equations, only eq 6 can obtain an exact
integration. As the exact integrations for eqs 7-10 are not
possible, we can only get a better approximation to the integral
term in each equation by using a quadratic polynomial of
d[T]
dt
) k6[L]0[Ε] - k8[T] ) k6′[Ε] - k8[T]
(4)
where [T] and [L]0 are the percentage concentrations of triol
and of lauryl alcohol at time 0, respectively. k6, k6′, and k8 are
the reaction rate constants (per hour). Likewise, the formation
of reaction products between 5,6-EP and stearylamine could
be regarded as a function of concentrations of stearylamine and
5,6-EP. Since the initial concentration of stearylamine at time
0 min is 566 times higher than that of 5,6-EP, the concentration

7136 J. Agric. Food Chem., Vol. 52, No. 23, 2004
Chien et al.
Table 1. Rate Constants of Cholesterol Degradation and Oxidation with or without Stearylamine Addition
Cb
Sb
-1
-1
rate equationsa
reactions
K (h
)
R2
K (h
)
R2
k1
98
free radical chain reaction
251.0
±
±
±
±
±
±
±
±
0.2
0.96
74.3
±
±
±
±
±
±
0.0
0.77
Chol
7-OOH
k2
reduction
147.3
95.8
0.13
545.4
0.34
0.31
4.70
0.1
0.96
0.96
0.88
0.99
0.92
0.99
0.73
c
59.7
4.74
2.49
0.69
0.19
0.0
0.77
0.77
0.45
0.73
0.91
c
7-OOH
9
8
7-OH
k3
dehydration
dehydrogenation
epoxidation
degradation
alcoholysis
14.6
0.51
135.2
0.05
0.14
0.40
0.0
7-OOH
98
7-keto
k3
0.26
0.25
0.02
7-OH
Chol
Chol
9
′8 7-keto
k4
9
8
5,6-EP
k5
98
D
k6
9
5,6-EP
′8 triol
c
k8
dehydrogenation
amination
c
c
triol
98
O
k7
5,6-EP
9
′8
S
c
1.51
±
0.20
0.73
a Chol (cholesterol), 7-OOH (7
(5,6-epoxycholesterol), triol (5 -cholestan-3
k7
, and k8 are the rate constants. b C, cholesterol only. S, cholesterol in the presence of stearylamine. c Not detected.
R
- or 7
â
â
-hydroperoxycholesterol), 7-OH (7
R
- or 7
â
-hydroxycholesterol), 7-keto (7-ketocholesterol), D (degraded products), 5,6-EP
stearylamine reaction products), and k1 k5, k3 , k6′,
R
,5 ,6 -triol), O (cholestan-3 ,5 -diol-6-one), S (5,6-epoxycholesterol
R
â
â
R
−
∼
′
′
Figure 6. Comparison of predicted lines and experimental data for total
COPs and C-7 route formed during heating at 140 C. Mean of
experimental data ( ); best-fitting lines ( ); data-point-connecting line
‚‚‚). All the standard errors are within 4.8% of the means (n 3).
Figure 7. Comparison of predicted lines and experimental data for total
COPs and C-7 route formed during heating at 140 C in the presence of
stearylamine. Mean of experimental data ( ); best-fitting lines ( ); data-
point-connecting line (‚‚‚). All the standard errors are within 5.9% of
the means (n 3).
°
°
b
s
O
s
±
(
±
)
)
Taylor’s series. The experimental data were thus analyzed in
sequence and fitted best to the corresponding nonlinear regres-
sion model.
degradation and oxidation during heating at 140 °C. Without
stearylamine, the r2 values of all the reactions were higher than
0.88, with the exception of dehydrogenation of triol (r2 ) 0.73).
Likewise, with stearylamine, the r2 values of all the reactions
were higher than 0.73, with the exception of dehydrogenation
of 7-OH (r2 ) 0.45). Each COP was further plotted on the basis
of the corresponding rate equation and the rate constants listed
in Table 1, which are shown in Figures 6-8. Without stear-
ylamine, all the curves fit well with the data points (Figure 6).
However, in the presence of stearylamine, only the curve of
7-OH fit well with the data points (Figure 7). Neither curves
of 7-OOH nor 7-keto fit well for the heating times 1 and 2 h
(Figure 7). The above results suggested that eq 6 could only
successfully estimate the degradation of cholesterol during the
For computing k5 in eq 6 by use of a linear regression model,
the data with or without stearylamine were obtained by
subtracting total COPs formed (Figure 5) from total loss of
cholesterol (Figure 2) over a 60- or 40-min heating period,
respectively. k5 values were found to be 0.19 ( 0.02 with
stearylamine and 0.34 ( 0.05 without stearylamine, with
correlation coefficients (r2) 0.91 and 0.92, respectively, and the
corresponding best-fitting lines are shown in Figure 2. The rate
constants in eqs 7-10 were estimated by using the least-squares
method with a nonlinear Marquardt iterative method until the
convergences of best-fitted parameters were met. Table 1 shows
the rate constants of the reaction pathways of cholesterol

Kinetic Studies of Cholesterol Oxidation
J. Agric. Food Chem., Vol. 52, No. 23, 2004 7137
cholesterol is heated in an alcoholic solvent and a lower
temperature is employed.
Table 1 also shows the rate constants of cholesterol degrada-
tion and oxidation in the presence of stearylamine during
heating. The rate constants (per hour) are in the following
order: k1 > k2 > k3 > k3′ > k7′ > k4 > k5. Compared to the
control treatment, all the rate constants decreased substantially,
especially for the epoxidation reaction (k4), which decreased
by 800-fold. The incorporation of stearylamine not only reduces
the COPs formation but also completely inhibits triol production
during heating. However, we have to point out here that we do
not have a good explanation for the data points of 7-OOH and
7-keto in Figure 7 for heating times 1 and 2 h. It may be
postulated that, with stearylamine, epoxidation proceeds faster
than C-7 oxidation as shown in Figure 8. Also, â-hydroxylamine
may be formed from the amination reaction between epoxide
and stearylamine, which may expedite the consumption of 5,6-
EP. The k2 value was found to be about 12.6 times higher than
the k3 value, implying that dehydration is less susceptible to
proceed than reduction. Similar results were reported by Nielsen
et al. (25), who studied the cholesterol oxidation in a hetero-
geneous system initiated by water-soluble radicals. Despite the
small k6′ value, the level of triol sharply rose after 40-min
heating (Figure 8). However, the formation of triol in the
presence of stearylamine remained extremely slow even up to
2-h heating (Figure 8). This is probably because 5,6-epoxides
reacted with stearylamine rapidly to form â-hydroxylamine, as
reported by Ingold (19). In Figures 6-8, the curves fit well
with the experimental data based on the oxidation of cholesterol
during heating. These results suggested that the kinetic models
developed in this study can be used to predict the concentration
changes of COPs during heating of cholesterol with or without
addition of stearylamine.
Figure 8. Comparison of predicted lines and experimental data for 5,6-
epoxycholesterol and triol formed during heating at 140
experimental data for cholesterol only ( ) and for cholesterol in the
presence of stearylamine ( ); best-fitting lines ( ). All the standard errors
are within 5.4% of the means (n 3).
°C. Mean of
b
O
s
±
)
initial 40-min heating period and eqs 7-10 could be used to
predict the concentration changes of COPs during the initiation
and propagation periods of oxidation.
Of the various reactions, the r2 values of the free radical chain
reaction and epoxidation were 0.96 and 0.99, respectively, in
the absence of stearylamine. It can be inferred that both reactions
for formation of 7-OOH and 5,6-epoxide from cholesterol are
second-order. For the reduction and dehydration reactions, the
r2 values of both were 0.96. This result also demonstrated that,
in the absence of stearylamine, the reactions for formation of
7-OH and 7-keto were first-order. The rate constant for
dehydrogenation of 7-OH to 7-keto is too small to be signifi-
cantly analyzed by using a nonlinear simulation. The formation
of triol by alcoholysis can be further inferred from eq 9 with
high correlation coefficient (0.99), revealing that this reaction
is pseudo-first-order.
LITERATURE CITED
(1) Paniangvait, P.; King, A. J.; Jones, A. D.; German, B. G.
Cholesterol oxides in foods of animal origin. J. Food Sci. 1995,
60, 1159-1174.
(2) Ansari, G. A. S.; Walker, R. D.; Smart, V. B.; Smith, L. L.
Further investigations of mutagenic cholesterol preparation. Food
Chem. Toxicol. 1982, 20, 35-41.
(3) Watanabe, K.; Nakamura; Hosono, A. Mutagenic activity of heat-
induced chilesterol degredation products. J. Food Sci. 1988,
53(6), 1913-1916.
(4) Maerker, G. Cholesterol autoxidation-current status. J. Am. Oil
Chem. Soc. 1987, 64(3), 388-392.
(5) Chien, J. T.; Wang, H. C.; Chen, B. H. Kinetic model of the
cholesterol oxidation during heating. J. Agric. Food Chem. 1998,
46(7), 2572-2577.
(6) Kritchevsky, D. Dietary protein, lipid metabolism and athero-
sclerosis. J. Am. Oil Chem. Soc. 1978, 64, 1167-1171.
(7) Li, Y. J. Processing induced changes in fish lipid with emphasis
on cholesterol. Dissertation Abstr. Int. B: Sci. Eng. 1992, 52(9),
4536-4537.
(8) Nawar, W. W.; Kim, S. K.; Vajdi, M. Measurement of oxidative
interactions of cholesterol. J. Am. Oil Chem. Soc. 1991, 68(7),
496-498.
(9) Yen, G. C.; Kao, H. H. Antioxidative effect of biogenic amine
on the peroxidation of linoleic acid. Biosci., Biotechnol., Bio-
chem. 1993, 57(1), 115-116.
It was also observed that the k4 value is very large without
stearylamine addition, implying that the 5,6-EP concentration
increased sharply as soon as 7-OOH was formed during the
initial period of oxidation (Table 1). It can be postulated that
most 7-OOH was converted to 5,6-EP during heating, and thus
7-OOH concentrations were maintained at a low level (0.2%)
over a 40-min heating period. This is an initial maximum
attainable concentration for 7-OOH ([A′]max in eq 7). However,
in practice it is difficult for 7-OOH to maintain at this level,
because it could be further degraded to form 7-OH or 7-keto in
the C-7 route and to triol in the epoxidation route. Therefore,
the 7-OOH level could be increased to about 0.8% in 2-h heating
(Figure 6). By comparing the rate constants (per hour) of the
major reaction pathways of cholesterol oxidation, it can be found
that k4 > k1> k2 > k3 > k8 > k5 > k6′ > k3′. This order seemed
to be different from that by Chien et al. (5), who reported that
the rate constants of COPs formed during dry heating of
cholesterol at 150 °C were k1 ≈ k4 ≈ k3 ≈ k2 ≈ k3′. This
difference can be attributed to the fact that, in our study,
(10) Aoki, H.; Tottroi, T.; Sakurai, F.; Fuji, K.; Miyajima, K. Effects
of positive charge density on the liposomal surface on disposition
kinetics of liposomes in rats. Int. J. Pharm. 1997, 156, 163-
174.
(11) Kokkona, M.; Kallinteri, P.; Fatouros, D.; Antimisiaris, S. G.
Stability of SUV liposomes in the presence of cholate salts and
pancreatic lipases: effect of lipid composition. Eur. J. Pharm.
Sci. 2000, 9, 245-252.

7138 J. Agric. Food Chem., Vol. 52, No. 23, 2004
Chien et al.
(12) Alaiz, M.; Barraga´n, S. Reaction of a lysyl residue analogue
with (E)-2-octenal. Chem. Phys. Lipids 1995, 75(1), 43-49.
(13) SAS. SAS Procedures and SAS/Graph User’s Guide, version 6;
Software Release 8.2 (TS2M0) of the SAS System for Microsoft
Windows; SAS Institute, Inc.: Cary, NC, 2003.
(20) March, J. Chapter 10, Aliphatic nucleophilic substitution: Reac-
tion 0-51, Amination of epoxides. AdVanced organic chemistry,
2nd ed.; McGraw-Hill: New York, 1977; p 381.
(21) Smith, L. L. Cholesterol autoxidation 1981-1986. Chem. Phys.
Lipids 1987, 44, 87-125.
(14) Smith, L. L.; Hill, F. L. Detection of sterol hydroperoxides on
thin layer chromatoplates by means of the wurster dyes. J.
Chromatogr. 1972, 66, 101-109.
(15) Maerker, G.; Bunick, F. J. Cholesterol oxides I: isolation and
determination of some cholesterol oxidation products. J. Am. Oil
Chem. Soc. 1986, 63(6), 767-771.
(22) Park, S. W.; Addis, P. B. Identification and quantitation
estimation of oxidized cholesterol derivatives in heated tallow.
J. Agric. Food Chem. 1986, 34(4), 653-659.
(23) Park, S. W.; Addis, P. B. Further investigation of oxidized
cholesterol derivatives in heated fats. J. Food Sci. 1986, 51(5),
1380-1381.
(16) Chien, J. T.; Lu, Y. F.; Hu, P. C.; Chen, B. H. Cholesterol
photooxidation as affected by combination of riboflavin and fatty
acids methyl esters. Food Chem. 2003, 81, 421-431.
(17) Alaiz, M.; Zamora, R.; Hidalgo, F. J. Contribution of the
formation of oxidized lipid/amino acid reaction products to the
protective role of amino acid in oils and fats. J. Agric. Food
Chem. 1996, 44(7), 1890-1895.
(24) Yan, P. S.; White, P. J. Linalyl acetate and methyl silicone effects
on cholesterol and triglyceride oxidation in heated lard. J. Am.
Oil Chem. Soc. 1990, 68(10), 763-768.
(25) Nielsen, J. H.; Olsen, C. E.; Skibsted, L. H. Cholesterol oxidation
in a heterogeneous system initiated by water-soluble radicals.
Food Chem. 1996, 56(1), 33-37.
(18) March, J. Chapter 10, Aliphatic nucleophilic substitution: Reac-
tion 0-20, Alcoholysis of epoxides. AdVanced organic chemistry,
2nd ed.; McGraw-Hill: New York, 1977; pp 360-361.
(19) Ingold, C. K. Structure and mechanism in organic chemistry,
2nd ed.; Cornell University Press: Ithaca, NY, 1969; pp 457-
463.
Received for review June 28, 2004. Revised manuscript received August
31, 2004. Accepted September 5, 2004. This study was supported by a
grant (NSC-90-2313-B-030-008) from the National Science Council,
Taiwan, R.O.C.
JF048951+