Enhancement in the synthesis of novel feruloyl lipids (feruloyl butyryl glycerides) by enzymatic biotransformation using response surface methodology

Article abstract of DOI:10.1021/jf802478r

Response surface methodology was successfully employed to optimize lipase-catalyzed synthesis of feruloyl butyryl glycerides (FBGs). The effects of the reaction parameters, including the reaction time, reaction temperature, enzyme concentration, substrate

Full text of DOI:10.1021/jf802478r

J. Agric. Food Chem. 2008, 56, 11493–11498 11493
Enhancement in the Synthesis of Novel Feruloyl
Lipids (Feruloyl Butyryl Glycerides) by Enzymatic
Biotransformation Using Response Surface
Methodology
†
†
‡
,†
YAN ZHENG, JING QUAN, XIN NING, AND LI-MIN ZHU*
College of Chemistry, Chemical Engineering and Biotechnology, Donghua University, Shanghai
01620, and Kimberly-Clark (China) Company, Ltd., Shanghai 200001, People’s Republic of China
2
Response surface methodology was successfully employed to optimize lipase-catalyzed synthesis
of feruloyl butyryl glycerides (FBGs). The effects of the reaction parameters, including the reaction
time, reaction temperature, enzyme concentration, substrate molar ratio, and water activity, and the
interaction parameters were examined. The analysis suggested that the conversion of the FBGs
was significantly (p < 0.05) affected by independent factors of reaction time, reaction temperature,
substrate molar ratio, and water activity as well as interactive terms of reaction temperature/reaction
time, reaction temperature/enzyme concentration, substrate molar ratio/reaction temperature, water
activity/reaction temperature, reaction time/enzyme concentration, and enzyme concentration/water
activity. The highest conversion yield of FBGs was 81.2% at the following optimized reaction conditions:
reaction temperature of 53.6 °C, reaction time of 5.5 days, enzyme concentration of 50.8 mg/mL,
water activity of 0.14, and substrate molar ratio of 2.9. The conversion is higher as compared to that
at the conditions before optimization.
KEYWORDS: Feruloyl butyryl glycerides; lipase; response surface methodology; transesterification;
optimization
INTRODUCTION
while the butyric acid moiety is a traditional antitumor
agent (2, 12) (Figures 1 and 2).
Ferulic acid (FA) is a monophenolic phenylpropanoid present
widely in the plant kingdom. Recently, it is of growing particular
interest because of its potential biological properties, such as
antioxidant, anti-inflammatory, antiviral, and UV filter pro-
perties (1-3). However, due to its hydrophilic nature, applica-
tions for this natural compound in oil-based food processing,
pharmaceuticals, and cosmetics are limited (4, 5). To overcome
this limit, modification of FA via its esterification with aliphatic
alcohol has been widely investigated (6-9). Nevertheless, to
our knowledge, there are few reports on the modification of
FA with triglyceride (5, 10).
Chemical synthesis of this kind of FA esters is usually
performed with basic or acidic catalysts under reflux. However,
with the steadily growing “natural” demand, the synthesis of
such esters by lipase-catalyzed reactions under mild conditions
has received much attention. Particularly, the development of
an optimum enzymatic synthesis procedure to improve the yield
of conversion would be more attractive for the manufacturers
(
13).
The earlier work in our laboratory showed that the results
were of general interest for the lipase-catalyzed transesterifi-
Ferulyl butyryl glycerides (FBGs), obtained by lipase-
catalyzed transesterification of ethyl ferulate (EF) with tributyrin
(
TB), is a novel feruloylated lipid. It is composed of 1(3)-
feruloyl monobutyryl glyceride (FMG) and 1(3)-feruloyl dibu-
tyryl glyceride (FDG) and can exhibit higher solubility and
nutritional quality than free ferulic acid in a hydrophobic
medium (11). Moreover, it is a potential bifunctional compound,
because the ferulic acid moiety works as a natural antioxidant,
*
To whom correspondence should be addressed. Phone/fax: +86
2
1 67792655. E-mail: lzhu@dhu.edu.cn.
†
Donghua University.
Kimberly-Clark (China) Co., Ltd.
Figure 1. Reaction route of lipase-catalyzed transesterification of EF and
TB.
‡
1
0.1021/jf802478r CCC: $40.75  2008 American Chemical Society
Published on Web 11/11/2008

11494 J. Agric. Food Chem., Vol. 56, No. 23, 2008
Zheng et al.
Table 1. Coded Levels for Independent Factors Used in the Experimental
Design
factor
symbol level -2 level -1 level 0 level 1 level 2
reaction temp (°C)
reaction time (day)
enzyme concn (mg/mL)
TB:EF molar ratio
aw
x1
x2
x3
x4
x5
40
3
20
1
45
4
30
2
50
5
40
3
55
6
50
4
60
7
60
5
0.05
0.10
0.15
0.20
0.25
According to previous reports (4, 5), the total conversion of FBGs
was determined most accurately by measuring the residual EF and FA
peak areas by HPLC. The sum of all ferulate species peak areas
remained constant over the time course of the reaction, allowing
accurate FDG and FMG yields to be calculated as the percentage ratio
of the FDG and FMG peak areas to the residual EF and FA peak
areas.
w
Water Activity Pre-Equilibration. Water activity (a ) is an
important consideration for biocatalysis in a nonaqueous medium. It is
a better parameter than the water content to determine the amount of
water associated with the enzyme and thereby to truly correlate with
the enzyme activity (17). Among the available methods for controlling
a , the simple and convenient way is pre-equilibration of the reaction
w
components in the presence of saturated salt solutions, and the method
is particularly suitable for the reaction (18).
Figure 2. Time course of the lipase-catalyzed synthesis of FBGs.
Conditions: temperature, 50 °C; water activity, 0.15; molar ratio of TB to
EF, 3; enzyme loading, 40 mg/mL. Vertical bars represent the standard
deviation for each data point. Each sample is significantly different (n )
3, p < 0.05).
cation of EF with TB for the preparation of FBGs (11). The
present investigation was aimed to enhance our knowledge about
the reaction parameters affecting lipase-catalyzed synthesis of
FBGs and to optimize the process for enzymatic synthesis.
Response surface methodology (RSM) is an efficient statisti-
cal tool for optimization of multiple variables and to predict
the best performance conditions for the target value (14, 15). It
is also a statistical technique useful for designing experiments,
building models, and analyzing the effects of independent
variables (14, 16). Thus, RSM was employed to optimize the
synthesis of FBGs via lipase-catalyzed transesterification in this
research.
The water activity values of toluene, enzyme, and substrate were
adjusted before the reaction was started by the following method:
toluene, EF, TB, and Novozyme 435 were separately incubated in a
chamber containing a desirable saturated salt solution (18), and the
system was allowed to reach equilibrium for at least 7 days at 25 °C
for the desired a
w
. Molecular sieves were used to generate the nearly
) 0.05). The following salts were used in
) 0.10), KAc, MgCl (a ) 0.15), CuCl (a
) 0.25). The water activity of the pre-equilibrated
anhydrous condition (a
this work: LiCl (a
.20), and NaCl (a
w
w
2
w
2
w
)
0
w
system was measured by a Karl Fischer titrator (Mettler-Toledo,
DL31).
Experimental Design. RSM was employed to analyze the operating
conditions of lipase transesterification between ethyl ferulate and
tributyrin to obtain a high conversion of feruloyl butyryl glycerides.
The software Design-Experiment 6.0 (Stat-Ease) was used to design
and regress the experimental data. The experimental design was carried
out by five chosen independent variables (factors) with five levels
MATERIALS AND METHODS
Materials. Novozyme 435 (Candida antartica lipase immobilized
on polyacrylic resin, with an activity of 10 000 propyl laurate units
(
Table 1). The investigated factors were the following: reaction
temperature (°C), reaction time (day), enzyme concentration (mg/mL),
substrate molar ratio (TB:EF), and a . For each factor, the experimental
range and central point were based on the results of preliminary trials
11).
Statistical Analysis. The effect of the reaction time was performed
(
(
PLU)/g of solid enzyme) was purchased from Novozymes A/S
Bagsvaerd, Denmark). FA (purity >99%) and EF (purity >99%) were
w
purchased from Suzhou Chang Tong Chemical Co., Ltd. (Suzhou,
China). TB (purity >99%) was purchased from Tao He Chemical Co.,
Ltd. (Shanghai, China).
All solvents were of analytical grade and were dried by activated 4
Å molecular sieves before use. All other reagents used were of high
purity and were commercially available unless otherwise noted.
Enzymatic Reactions. All enzymatic reactions were carried out in
a temperature-controlled incubator shaker at 210 rpm. The procedure
for the enzymatic transesterification of FBGs in toluene was described
as an example (1). The reactions were performed by adding Novozyme
(
in triplicate, and the level of significance was set at p < 0.05. Analysis
of variance (ANOVA) and significant differences among the means
were tested by one-way ANOVA, using SPSS (version 13.0 for
Windows, SPSS Inc., Chicago, IL).
The experimental FBG data (Table 1) were best-fitted to the second-
order polynomial model (p < 0.05) by multiple regression after manual
evaluation (p for the first-order model and third-order model was 0.93
and 0.24, respectively, which indicated that both these models were
insignificant). The model proposed for the response of Y fitted a second-
order polynomial equation as follows:
4
35 (30 mg/mL), EF (0.5 mmol), and TB (1 mmol) in toluene (3 mL)
and shaking the reaction mixtures at 45 °C. Samples of the biotrans-
formation were withdrawn at different times and analyzed by high-
performance liquid chromatography (HPLC).
HPLC Analysis. Analytical HPLC was performed using a Waters
5
5
4
5
5
10 with an Inertsil Ph-3 column (4.6 mm i.d. × 250 mm, 5 µm, GL
2
Y ) ꢀ +
ꢀ x +
ꢀ x +
ꢀ x x
i j
(1)
Sciences, Japan), with a dual-absorbance detector (Waters 2487) at 325
nm. The mobile phase was solution A (water containing 0.1% acetic
acid) and solution B (100% methanol), in all cases at 1 mL/min flow
of A/B (30:70, v/v). Analysis was carried out at room temperature.
The solvents were filtered using Whatman 0.45 µm nylon membrane
filters (Sigma-Aldrich) and degassed using a Thermo Separation
Products SCM 1000 membrane degasser.
0
∑
i
i
∑
ii
i
∑ ∑ ii
i)1
i)1
i)1 j)i+1
where ꢀ0,
term, ꢀi
ꢀ
i
, ꢀii, and ꢀij are regression coefficients (ꢀ
is a linear effect term, ꢀii is a squared effect term, and ꢀij and
is an interaction effect term) and Y is the predicted response value.
0
is a constant
RESULTS AND DISCUSSION
A 10 µL sample was removed from the reaction mixture at set time
intervals during the reaction and further diluted 100-fold with methanol.
The sample injection volume was 10 µL.
Effect of the Reaction Time. The selection of reaction time
range needs to be extremely precise in fractional factorial design;

Enhancement in the Synthesis of FBGs by RSM
J. Agric. Food Chem., Vol. 56, No. 23, 2008 11495
Table 2. Experimental Design and Results of the CCD Design
distribution (%)
trial
T (°C)
t (days)
enzyme concn (mg/mL)
TB:EF
aw
response Y (%)
FDG
FMG
1
2
3
4
5
6
7
8
9
45 (-1)
55 (1)
45 (-1)
55 (1)
45 (-1)
55 (1)
45 (-1)
55 (1)
45 (-1)
55 (1)
45 (-1)
55 (1)
45 (-1)
55 (1)
45 (-1)
55 (1)
40 (-2)
60 (2)
50 (0)
50 (0)
50 (0)
50 (0)
50 (0)
50 (0)
50 (0)
50 (0)
50 (0)
50 (0)
50 (0)
50 (0)
50 (0)
50 (0)
4 (-1)
4 (-1)
6 (1)
30 (-1)
30 (-1)
30 (-1)
30 (-1)
50 (1)
50 (1)
50 (1)
50 (1)
30 (-1)
30 (-1)
30 (-1)
30 (-1)
50 (1)
50 (1)
50 (1)
50 (1)
40 (0)
40 (0)
40 (0)
60 (-2)
60 (2)
40 (0)
2.0 (-1)
2.0 (-1)
2.0 (-1)
2.0 (-1)
2.0 (-1)
2.0 (-1)
2.0 (-1)
2.0 (-1)
4.0 (1)
4.0 (1)
4.0 (1)
4.0 (1)
4.0 (1)
4.0 (1)
4.0 (1)
4.0 (1)
3.0 (0)
3.0 (0)
3.0 (0)
3.0 (0)
3.0 (0)
3.0 (0)
1.0 (-2)
1.0 (2)
3.0 (0)
3.0 (0)
3.0 (0)
3.0 (0)
3.0 (0)
3.0 (0)
3.0 (0)
3.0 (0)
0.2 (1)
64.51
60.43
62.58
60.92
70.32
56.02
70.13
55.32
30.12
69.26
64.12
71.59
38.11
67.13
50.33
72.64
52.19
71.98
62.56
73.34
60.57
63.22
60.93
50.09
60.24
63.54
67.06
64.89
67.12
68.98
69.29
70.12
30.43
32.97
36.54
30.54
34.53
28.77
35.65
27.65
12.67
33.90
21.14
32.45
18.32
33.45
20.65
36.98
32.12
34.65
21.89
34.98
32.89
29.05
27.15
19.65
28.98
30.91
16.94
12.54
16.21
18.43
17.11
18.54
33.15
26.59
24.76
28.68
34.91
26.11
32.97
25.14
11.98
33.95
40.23
37.86
17.43
31.56
27.62
34.31
18.65
35.91
38.76
36.32
25.87
31.97
32.09
29.91
30.34
30.54
49.54
51.11
49.87
48.75
50.92
50.21
0.1 (-1)
0.1 (-1)
0.2 (1)
0.1 (-1)
0.2 (1)
0.2 (1)
0.1 (-1)
0.1 (-1)
0.2 (1)
0.2 (1)
0.1 (-1)
0.2 (1)
6 (1)
4 (-1)
4 (-1)
6 (1)
6 (1)
4 (-1)
4 (-1)
6 (1)
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
6 (1)
4 (-1)
4 (-1)
6 (1)
6 (1)
5 (0)
0.1 (-1)
0.1 (-1)
0.2 (1)
0.15 (0)
0.15 (0)
0.15 (0)
0.15 (0)
0.15 (0)
0.15 (0)
0.15 (0)
0.15 (0)
0.05 (-2)
0.25 (2)
0.15 (0)
0.15 (0)
0.15 (0)
0.15 (0)
0.15 (0)
0.15 (0)
5 (0)
3 (-2)
7 (2)
5 (0)
5 (0)
5 (0)
5 (0)
5 (0)
5 (0)
5 (0)
5 (0)
5 (0)
5 (0)
5 (0)
40 (0)
40 (0)
40 (0)
40 (0)
40 (0)
40 (0)
40 (0)
40 (0)
40 (0)
40 (0)
5 (0)
Table 3. Regression Coefficients and Significance (p < 0.05) after
Backward Elimination for Conversion of FBGs
otherwise, the optimal condition of synthesis may not be located
inside the experimental region through the analyses of statistics
and contour plots. Figure 1 shows time course for the lipase-
catalyzed synthesis of FBGs at 50 °C. The percent molar
conversion increased to 70.6% (p < 0.05) at 5 days, and there
was no significant increase after 5 days. Therefore, the range
of reaction time was chosen in the range of 4-6 days in the
experiment.
a
variable
coefficient
p value
intercept
67.93
4.28
3.05
0.074
-2.44
1.44
-1.48
-0.014
-1.53
-3.12
-1.53
-2.28
-1.17
8.30
<0.0001
<0.0001
<0.0001
0.8430
<0.0001
0.0003
0.0009
0.9658
0.0007
<0.0001
0.0007
0.0003
0.0040
<0.0001
0.0211
0.0278
<0.0001
0.0023
0.7790
0.0001
0.0103
x1
x2
x3
x4
x5
2
Model Fitting. The combination of RSM and five-
factor-five-level central composite experimental design
x1
2
x2
2
x3
(
CCD) was employed to investigate the relationship between
2
x4
the variables and the conversion of FBGs. CCD, which is a
2
x5
x x
k
2
factorial design with star points and center points, was
1
2
x1x3
x1x4
x1x5
x2x3
x x
used to fit a full second-order polynomial model. According
to statistics theory (19), a CCD design of five factors
consisted of 32 experiments, including 15 factorial points
-1.20
-1.13
3.52
(
cubic point) and 11 axial points (star point) as well as 6
2
4
replicates at the center point. The pure error was evaluated
by six replications (trials 27-32) at the center of the design.
The results at each point based on the experimental design
are shown in Table 2. In addition, the distribution of FDG
and FMG under the experimental condition is also given.
The coded values of each factor in parentheses correspond
to the real values of the factor levels. For each factor, a
conventional level was set to zero as a coded level.
x2x5
x3x4
x3x5
x4x5
1.76
-0.64
-2.52
1.38
a
See Table 1.
indicated in Table 3, the independent variables (x1, x2, x4, x5),
interactions (x1x4, x1x2, x1x4, x1x3, x1x5, x4x5, x2x3, x2x4, x2x5, x3x5),
2
2
2
2
Among the various treatments, the greatest conversion
and quadratic terms (x1 , x3 , x4 , x5 ) were significant (p <
2
(
3
78.3%) was from trial 20 and the smallest conversion (only
0.1%) was from trial 9. Coefficients of a full model were
0.0001). The independent variable x3, the quadratic term x2 ,
and the interaction term x3x4 did not produce a significant effect
on the conversion within the designed intervals (p > 0.5) and
were removed from the original model. Thus, the final second-
order polynomial equation is
evaluation by regression analysis and tested for their signifi-
cance. The insignificant coefficients were eliminated stepwise
on the basis of the p value after the coefficients were tested. As

11496 J. Agric. Food Chem., Vol. 56, No. 23, 2008
Zheng et al.
conversion (%) ) 67.93 + 4.28x + 3.05x - 2.44x +
for regression of the model were significant (p < 0.0001),
meaning that the models were statistically good, and the models
had no lack of fit of the 95% level of significance. As a result,
this well-fitting model for conversion of FBGs was successfully
established.
1
2
4
2
2
2
2
1
.44x - 1.48x - 1.53x - 3.12x - 1.53x -
5 1 3 4 5
2
.28x x - 1.17x x + 8.30x x - 1.20x x - 1.13x x +
1 2 1 3 1 4 1 5 2 3
3
.52x x + 1.76x x - 2.52x x + 1.38x x (2)
2 4 2 5 3 5 4 5
Mutual Effect of the Parameters. A clear interpretation of
the effects of variable interaction on the conversion of FBGs
can be seen in the response contour plots (Figure 3).
Figure 3a depicts the effects of water activity and substrate
molar ratio and their mutual interaction on the conversion of
reaction at 5 days, 50 °C, and 40 mg/mL enzyme concentration.
Many research groups have reported that water activity presented
in the reaction system was one of the most significant factors
where xi is the coded value of each factor.
ANOVA and Adequacy Test of the Model. According to
the ANOVA, the coefficient of determination (R ) of the model
was 0.9881, which indicates that the model is suitable to
represent the real relationships among the selected reaction
parameters (Table 4). Table 4 also shows that the probabilities
2
Table 4. ANOVA for the Quadratic Model for the Conversion of FBGs
(
20). On one hand, a certain amount of water is essential to
source
model
sum of squares
DF
mean square
F value
p > F
keep the lipase activity, and almost all kinds of lipase have their
own optimum aw (21); on the other hand, an excess of water in
the reaction media will inhibit the synthetic reaction while
promoting the hydrolysis of the acylated products (22) and the
acyl donors (23). As presented in Figure 3a, both the higher
water activity (aw > 0.18) and lower water activity (aw < 0.13)
had negative effects on the conversion of reaction. Also, the
2906.71
34.99
16.52
18.40
2941.36
0.9881
20
11
6
5
31
145.34
3.17
2.75
45.78
<0.0001
residual
lack of fit
pure error
total
0.75
0.6372
3.68
2
R
a
a
2
R ) 0.9881.
Figure 3. Response surface plots: (a) effect of the water activity and substrate molar ratio and their mutual interaction on production of FBGs, with other
parameters (water activity, substrate molar ratio, and enzyme concentration) being set constant at the 0 level; (b) effect of the reaction time and reaction
temperature and their mutual interaction on prodution of FBGs, with other parameters (water activity, reaction time, and enzyme concentration) being set
constant at the 0 level; (c) effect of the substrate molar ratio and reaction temperature and their mutual interaction on production of FBGs, with other
parameters (water activity, reaction time, and enzyme concentration) being set constant at the 0 level; (d) effect of the enzyme concentration and
reaction temperature and their mutual interaction on production of FBGs, with other parameters (water activity, reaction time, and substrate molar ratio)
being set constant at the 0 level.

Enhancement in the Synthesis of FBGs by RSM
J. Agric. Food Chem., Vol. 56, No. 23, 2008 11497
effect of the substrate molar ratio on the conversion was
significant (p < 0.05). The conversion increased as the substrate
molar ratio was increased from 1 to 3; however, when the
substrate molar ratio is larger than 3, it could lead to a decrease
in conversion. These results indicate that the excess of TB would
lead to an increased limitation of the substrate-enzyme reaction
sites, which leads to a decrease in conversion. It was concluded
that a high conversion could be obtained by combining aw )
Table 5. Optimum Conditions Founded by the Model for the Conversion of
FBGs
run 1
53.62
5.5
50.76
2.91
0.14
run 2
53.1
reaction temp (°C)
reaction time (day)
enzyme concn (mg/mL)
TB:EF molar ratio
aw
6
36.25
3.12
0.20
predicted value (%)
experimental value (%) ( SD
79.92
81.21 ( 0.7
64.69
62.04 ( 0.9
0
.15 and a substrate molar ratio of TB to EF of 3.
a
Figure 3b shows the effect of the reaction time and reaction
temperature on the conversion of the reaction with a 40 mg/
mL enzyme concentration, a substrate molar ratio of TB to EF
of 3, and aw ) 0.15. As presented in Figure 3b, an increasing
reaction temperature could raise the conversion of the reaction
strongly. It was indicated that a high temperature can activate
the substrate molecules, reduce the viscosity of the reaction,
and lead to a higher reaction rate. Also, within the given range
a
Standard deviation of triplicate determinations from different experiments.
2
7-30). Thus, it could be concluded that the optimization of
lipase-catalyzed synthesis of FBGs by Novozyme 435 was
successfully developed by RSM. The experimental conditions
allowed a quantitative and maximum conversion of FBGs.
LITERATURE CITED
(4-6 days) of reaction time, the conversion yield of the reaction
increased almost linearly with an increase in the reaction time.
This phenomenon has also been reported by Dossat et al. (24).
Not surprisingly, the conversion of the reaction peaked with a
maximum conversion of 69.4% at 55 °C and 6 days. From the
analysis of the response surface plots, both the reaction
temperature and reaction time exhibited a positive influence on
the response surface.
(
1) Rice Evans, C. A.; Miller, N. J.; Paganga, G. Antioxidant activity
applying an improved ABTS radical cation decolorization assay.
Free Radical Biol. Med. 1996, 26, 1231–1237.
(2) Murakami, A.; Nakamura, Y.; Koshimizu, K.; Takahashi, D.;
Matsumoto, K.; Hagihara, K.; Taniguchi, H.; Nomura, E.; Hosoda,
A.; Tsuno, T.; Maruta, Y.; Kim, H. W.; Kawabata, K.; Ohigashi,
H. FA15, a hydrophobic derivative of ferulic acid, suppresses
inflammatory responses and skin tumor promotion: Comparison
with ferulic acid. Cancer Lett. 2002, 180, 121–129.
Figure 3c denotes the effect of the substrate molar ratio and
reaction temperature on the conversion of the reaction at 5 days,
a 40 mg/mL enzyme concentration, and aw ) 0.15. The
interaction term of the substrate molar ratio and reaction
temperature played an important role in the process of the
reaction as was evident from its first-order effect (p < 0.05).
When the reaction temperature and substrate molar ratio
increased to 55 °C and 4, respectively, the conversion of the
reaction achieved was 72.7%. It must be pointed out that, at
low temperature (<50 °C), conversion declined with increasing
substrate molar ratio, which could be attributed to the fact that
there was great resistance of the viscosity of the reaction caused
by a high amount of TB when the temperature was low.
Figure 3d shows the enzyme concentration and temperature
effect on the conversion of the reaction at 5 days, a substrate
molar ratio of TB to EF of 3, and aw ) 0.15. It was observed
that the interaction between the reaction temperature and enzyme
concentration obviously affects the conversion, and this was
also confirmed by the p value (p ) 0.004). At any given enzyme
concentration in the investigated range (30-50 mg/mL), an
increase in the temperature from 45 to 55 °C could lead to an
increase in conversion. The reaction with a high reaction
temperature (55 °C) and enzyme amount (50 mg/mL) favored
maximal conversion, which indicated the same report from Novo
Nordisk Bioindustrials that the enzyme is stable in an organic
solvent at higher temperature (40-60 °C).
Model Verification. Within the experimental range studied,
optimum conditions for synthesis of FBGs were predicted using
the optimization function of Design Expert. Two sets of
predicted reaction conditions are presented in Table 5 along
with their predicted and actual values. To validate the predicted
results, experiments using the improved formula were per-
formed, and the observed values are shown in Table 5. On the
basis of the solution given by the design, two runs of
experiments were established at the fixed conditions. The
experimental values were found to be in reasonable agreement
with the predicted ones, which confirmed the validity and
adequacy of the predicted models. Moreover, the conversion is
higher as compared to the conditions before optimization (trials
(
3) Lin, F. H.; Lin, J. Y.; Gupta, R. D.; Tournas, J. A.; Burch, J. A.;
Selim, M. A.; Monteiro-Riviere, N. A.; Grichnik, J. M.; Zielinski,
J.; Pinnell, S. R. Ferulic acid stabilizes a solution of vitamins C
and E and doubles its photoprotection of skin. J. InVest. Dermatol.
2
005, 125, 826–832.
(
(
4) Compton, D. L.; Laszlo, J. A.; Berhow, M. A. Lipase-catalyzed
synthesis of ferulate esters. J. Am. Oil Chem. Soc. 2000, 77, 513–
5
19.
5) Laszlo, J. A.; Compton, D. L.; Eller, F. J.; Taylor, S. L.; Isbell,
T. A. Packed-bed bioreactor synthesis of feruloylated monoacyl-
and diacylglycerols: Clean production of a “green” sunscreen.
Green Chem. 2003, 5, 382–386.
(6) Sabally, K.; Karboune, S.; StLouis, R.; Kermasha, S. Lipase-
catalyzed transesterification of trilinolein or trilinolenin with
selected phenolic acids. J. Am. Oil Chem. Soc. 2006, 83, 101–
1
07.
(7) Figueroa-Espinoza, M. C.; Villeneuve, P. Phenolic acids enzymatic
lipophilization. J. Agric. Food Chem. 2005, 53, 2779–2787.
(
8) Vosmann, L.; Weitkamp, P.; Weber, N. Solvent-free lipase-
catalyzed preparation of long-chain alkyl phenylpropanoates and
phenylpropyl alkanoates. J. Agric. Food Chem. 2006, 54, 2969–
2976.
(9) Laszlo, J. A.; Compton, D. L. Enzymatic glycerolysis and
transesterification of vegetable oil for enhanced production of
feruloylated glycerols. J. Am. Oil Chem. Soc. 2006, 83, 765–770.
10) Sun, S. D.; Shan, L.; Jin, Q. Z.; Liu, Y. F.; Wang, X. G. Solvent-
free synthesis of glyceryl ferulate using a commercial microbial
lipase. Biotechnol. Lett. 2007, 29, 945–949.
11) Zheng, Y.; Quan, J.; Zhu, L. M.; Jiang, B.; Nie, H. L. Optimization
of selective lipase-catalyzed feruloylated monoacyglycerols by
response surface methodology. J. Am. Oil Chem. Soc. 2008, 85,
635–639.
(12) Heerdt, B. G.; Houston, M. A.; Anthony, G. M.; Augenlicht, L. H.
Initiation of growth arrest and apoptosis of MCF-7 mammary
carcinoma cells by tributyrin, a triglyceride analogue of the short-
chain fatty acid butyrate, is associated with mitochondrial activity.
Cancer Res. 1997, 59, 1854–1591.
13) Chang, S. W.; Shaw, J. F.; Shieh, C. H.; Shieh, C. J. Optimal
formation of hexyl laurate by lipozyme IM-77 in solvent-free
system. J. Agric. Food. Chem. 2006, 54, 7125–7129.
(
(
(
(14) Garrido-Vidal, D.; Pizarro, C.; Gonzalez-Saiz, J. M. Study of
process variables in industrial acetic fermentation by a continuous

11498 J. Agric. Food Chem., Vol. 56, No. 23, 2008
Zheng et al.
pilot fermentor and response surfaces. Biotechnol. Prog. 2003,
(22) Caro, Y.; Pina, M.; Turon, F.; Guilbert, S.; Mougeot, E.; Fetsch,
D. V. Plant lipases: Biocatalyst aqueous environment in relation
to optimal catalytic activity in lipase-catalyzed synthesis reactions.
Biotechnol. Bioeng. 2002, 77, 693–703.
(23) Weber, H. K.; Weber, H.; Kazlauskas, R. J. Watching lipase-
catalyzed acylations using 1H NMR: Competing hydrolysis of
vinyl acetate in dry organic solvents. Tetrahedron: Asymmetry
1
9, 1468–1479.
(
(
(
(
15) Zanto, E. J.; Al-Muhtaseb, S. A.; Ritter, J. A. Sol-gel-derived
carbon aerogels and xerogels: Design of experiments approach
to materials synthesis. Ind. Eng. Chem. Res. 2002, 41, 3151–3162.
16) Li, Q. H.; Fu, C. L. Application of response surface methodology
for extraction optimization of germinant pumpkin seeds protein.
Food Chem. 2005, 92, 701–706.
17) Wehtje, E.; Kaur, J.; Adlercreutz, P.; Chand, S.; Mattiasson, B.
Water activity control in enzymatic esterification processes.
Enzyme Microb. Technol. 1997, 21, 502–510.
18) Goderis, H. L.; Ampe, G.; Feytn, M. P.; Fouwe, B. L.; Guffens,
W. M.; VanCauwenbergh, S. M.; Tobback, P. P. Lipase-catalyzed
ester exchange reactions in organic media with controlled humid-
ity. Biotechnol. Bioeng. 1987, 30, 258–266.
1
999, 10, 2635–2638.
(
24) Dossat, V.; Combers, D.; Marty, A. Efficient lipase catalysed
production of a lubricant and surfactant formulation using a
continuous solvent-free process. J. Biotechnol. 2002, 97, 117–
1
24.
Received for review August 10, 2008. Revised manuscript received
October 22, 2008. Accepted October 22, 2008. This work was financially
supported by the National Nature Science Foundation of China (Grant
No. 50773009), the Program for Changjiang Scholars and Innovative
Research Team in University (Grant No. IRT0526), the Biomedical
Textile Materials “111 Project” from the Ministry of Education of
People’s Republic of China (Grant No. B07024), and Grant No.
(
19) Deming, S. N. Quality by designspart 5. CHEMTECH 1990, 20,
1
18–120.
(
20) Yang, K.; Wang, Y. J.; Kuo, M. I. Effects of substrate pretreatment
and water activity on lipase-catalyzed cellulose acetylation in
organic media. Biotechnol. Prog. 2004, 20, 1053–1061.
21) Zhang, D. H.; Bai, S.; Sun, Y. Lipase-catalyzed regioselective
synthesis of monoester of pyridoxine (vitamin B6) in acetonitrile.
Food Chem. 2007, 102, 1012–1019.
08JC1400600 from the Science and Technology Commission of Shang-
(
hai Municipality.
JF802478R