Enzymatic propyl gallate synthesis in solvent-free system: Optimization by response surface methodology

Article abstract of DOI:10.1016/j.molcatb.2010.08.013

The ability of a non-commercial immobilized Staphylococcus xylosus lipase to catalyze the esterification of propanol with gallic acid was investigated and the antioxidant as well as the antimicrobial activities of the ester formed were evaluated. The response surface methodology, based on a three variables Box-Behnken design (reaction temperature, enzyme amount and 1-propanol/gallic acid molar ratio), was used to optimize the experimental conditions of propylgallate synthesis. The maximum conversion yield (90% ±3.5) was obtained by using 400 IU of immobilized lipase and a propanol/gallic acid at a molar ratio of 160 at 52 °C. The obtained ester was characterized by spectroscopic methods, NMR and FTIR. The antioxidant activity of propyl gallate was evaluated and compared to the synthetic classical antioxidants, BHA and ascorbic acid, taken as references. In addition, the antimicrobial activity of the propyl gallate was tested against S. xylosus, Escherchia coli and Staphylococcus aureus using disc diffusion and macrodilution methods. Our results show that the synthesized propyl gallate ester presents a higher antioxidant and antimicrobial power than the parent gallic acid as well as the synthetic classical antioxidants.

Full text of DOI:10.1016/j.molcatb.2010.08.013

Journal of Molecular Catalysis B: Enzymatic 67 (2010) 242–250
Contents lists available at ScienceDirect
Journal of Molecular Catalysis B: Enzymatic
journal homepage: www.elsevier.com/locate/molcatb
Enzymatic propyl gallate synthesis in solvent-free system:
Optimization by response surface methodology
Ahlem Bouaziza,1, Habib Horchani , Nadia Ben Salem , Ali Chaari ,
a,1
a
a
b
a
a,∗
Moncef Chaabouni , Youssef Gargouri , Adel Sayari
Laboratoire de Biochimie et de Génie Enzymatique des Lipases, ENIS, BP1173, Sfax 3038, Tunisia
Laboratoire de Chimie Industrielle, ENIS, BP1173, Sfax 3038, Tunisia
a
b
a r t i c l e i n f o
a b s t r a c t
Article history:
Received 22 June 2010
Received in revised form 21 August 2010
Accepted 27 August 2010
Available online 8 October 2010
The ability of a non-commercial immobilized Staphylococcus xylosus lipase to catalyze the esterification
of propanol with gallic acid was investigated and the antioxidant as well as the antimicrobial activi-
ties of the ester formed were evaluated. The response surface methodology, based on a three variables
Box–Behnken design (reaction temperature, enzyme amount and 1-propanol/gallic acid molar ratio),
was used to optimize the experimental conditions of propylgallate synthesis. The maximum conversion
yield (90% ± 3.5) was obtained by using 400 IU of immobilized lipase and a propanol/gallic acid at a molar
Keywords:
◦
ratio of 160 at 52 C. The obtained ester was characterized by spectroscopic methods, NMR and FTIR. The
Immobilized Staphylococcus xylosus lipase
Enzymatic synthesis
Response surface methodology
Propyl gallate
Antioxidant
Antibacterial
antioxidant activity of propyl gallate was evaluated and compared to the synthetic classical antioxidants,
BHA and ascorbic acid, taken as references. In addition, the antimicrobial activity of the propyl gallate was
tested against S. xylosus, Escherchia coli and Staphylococcus aureus using disc diffusion and macrodilution
methods. Our results show that the synthesized propyl gallate ester presents a higher antioxidant and
antimicrobial power than the parent gallic acid as well as the synthetic classical antioxidants.
©
2010 Elsevier B.V. All rights reserved.
1
. Introduction
The common synthetic phenolic compounds used as food additives
in many countries include butylated hydroxylanisole (BHA), buty-
lated hydroxyltoluene (BHT), propyl gallate (PG), octyl gallate and
dodecyl gallate. However, these antioxidants have been proved to
exert different degrees of adverse and toxic effects upon consump-
tion [2,3]. Thus to protect people, the quantity of these compounds
present in food preparations is strictly controlled in many coun-
tries worldwide. The regulatory levels are generally based on the
estimation of an acceptable daily intake for a healthy individual.
Propyl gallate (PG, propyl 3,4,5-trihydroxybenzoate) is used as
antioxidant in processed food, cosmetics and food packing mate-
rials in order to prevent rancidity and spoilage. According to the
US Food and Drug Administration list, PG is also used to preserve
and stabilize medicinal preparations [4]. It is an allowed additive
in the European Union and in many others countries. Because of its
prevalent usage, the potential toxicity of PG has been investigated
in vivo [5,6] and in vitro to assess various toxicological properties,
i.e. mutagenicity [7] and cytogenetic effects [8]. Although PG has a
low toxicity, it has various effects on tissue and cell functions.
Several studies have demonstrated the benefits of PG as an
antioxidant [6,9,10], as a chemopreventive agent [11,12] and as
an anti-inflammatory agent [13]. For instance, PG is an efficient
protector of liver cells from lipid peroxidation by oxygen radi-
cals [6]. Furthermore, it is almost as effective as the vitamin E
analogue, Trolox and it is even more effective than several water
Antioxidants belong to a specific class of chemical compounds
that have the ability to prevent or to reduce the rate of oxidative
reactions of substrates in food, pharmaceuticals, nutraceuticals and
other consumer goods. These compounds, such as ascorbic acid, are
either naturally present or deliberately added. The addition of artifi-
cial antioxidants to maintain food quality and nutritional values has
been routinely carried out for several decades in the food industry.
This food treatment technology is particularly important in con-
trolling the oxidation of lipids, which may result in the changes
of the odour of edible oil and fat-containing food products and
may turn them rancid. The mechanism of action of antioxidants
is to interfere with the formation of free radicals that propa-
gate oxidation [1]. Phenolic compounds that have unique chemical
characteristics which form low-energy radicals through stable res-
onance hybrids are considered to be good antioxidants for food.
Abbreviations: SXL2, Staphylococcus xylosus lipase 2; SXL2i, immobilized
Staphylococcus xylosus lipase 2; IU, Enzymatic International Unit; CaCO3, cal-
cium carbonate; HPLC, High Performance Liquid Chromatography; BHA, butylated
hydroxyanisole; PG, propyl gallate.
∗ Corresponding author. Tel.: +216 74 675 055; fax: +216 74 675 055.
E-mail address: adelsayari@yahoo.fr (A. Sayari).
They have participated equally to this work.
1
1
381-1177/$ – see front matter © 2010 Elsevier B.V. All rights reserved.
doi:10.1016/j.molcatb.2010.08.013

A. Bouaziz et al. / Journal of Molecular Catalysis B: Enzymatic 67 (2010) 242–250
243
Fig. 1. HPLC chromatogram of the polyphenols extracts of seeds grapes: (1) shibinic acid; (2) gallic acid; (3) protocatechic acid; (4) hydroxybenzoic acid; (5) coumaric acid;
6) cinnamic acid.
(
soluble antioxidants such as ascorbate [6]. Traditionally, this kind
of compounds has been either isolated from natural sources or pro-
duced by chemical synthesis. However, the use of chemical process
is being progressively restricted for many applications. Such later
methods rely on the use of sophisticated experimental techniques
and organic solvents which are prohibited for industrial applica-
tions, especially in food industries. The use of extracellular lipases
as catalysts for esters production has a great potential. In fact, using
a biocatalyst excludes the disadvantages of the chemical process
by producing compounds which are highly pure with less or no
downstream operations. Moreover, the use of a non-commercial
lipase in an immobilized form would overcome the technological
difficulties of production. Thus, it might be interesting to test the
performance of new lipases in the synthesis of propyl gallate by
direct esterification in solvent-free system. It is the purpose of this
research to develop a non-commercial lipase-mediated process for
the conversion of gallic acid and 1-propanol into propyl gallate.
The effects of various reaction parameters (the lipase amount, the
olive oil emulsion as a substrate [15]. The activity was expressed as
units per mL of enzymatic solution. One-unit (IU) of lipase activity
corresponds to one ␮mol of fatty acid released per min at pH 8.5
◦
and at 55 C.
2.3. Polyphenols extraction
Polyphenols were extracted, isolated and purified according to
Nawaz et al. with a slight modification [16]. Ethanol, mixed with
water, was selected as the best organic solvent used to improve
the solubility of the bioactive components [17]. The maximum
extraction yield was obtained when using a mixture of 50% of
ethanol and 50% of water (v/v). The following protocol was used
to the sample preparation. About 40 g of milled grape seeds were
mixed with 200 mL of 50% ethanol solution and let for 1 h in the
dark. The top phase was filtered through Whatman filter paper
and the residue was resuspended in 150 mL of 95% of ethanol solu-
tion. The latter extraction step was repeated three times in order
to extract the total polyphenols which are present in the sam-
ple. It’s Noteworthy that during the first extraction, the majority
of the polyphenols were removed. The identification of polyphe-
nols and the purification of gallic acid were carried out by HPLC
(Fig. 1). The concentrations were determined by calculating the
HPLC peak areas which are proportional to the amount of analytes
in a peak and presented as the mean of two determinations which
were highly repeatable. Six compounds were identified (cinnamic,
coumaric, protocatechic, shibinic, hydroxybenzoic and gallic acid)
on the basis of their retention time compared to those considered
as standards (Sigma Aldrich, 99% puriss). One can note from Fig. 1
that gallic acid is the major compound detected. It represents more
than 90% of the total polyphenols extracted from grape seeds. About
2 ± 0.2 mg of purified gallic acid were obtained from 1 g of grape
seeds. The purified gallic acid was used to synthesize propylgallate.
1
-propanol/gallic acid molar ratio, the reaction temperature) were
optimized using response surface methodology. The antioxidant
and the antibacterial activities of the propyl gallate were evaluated.
2
. Materials and methods
2.1. Reagents
Whole grape seeds were collected in Tunisia and milled to a
fine powder. Gallic acid and propyl gallate (98% of purety) were
purchased from Fluka (Buchs, Switzerland). The Hexane, chloro-
form, methanol were purchased from Prolabo (Paris, France). The
calcium carbonate (CaCO ) was obtained from Pharmacia (Uppsala,
3
Sweden). Molecular sieve dehydrate with indicator for drying sol-
vents 3 A˚ was purchased from Fluka (Buchs, Switzerland). All other
used reagents are from analytical grade.
2
.4. Enzymatic esterification
2
2
.2. Lipases
The propyl gallate was obtained by the esterification of gallic
.2.1. Production and immobilization of lipase
acid with 1-propanol using immobilized S. xylosus lipase.
Staphylococcus xylosus lipase was produced and immobilized
The esterification reactions were carried out in screw-capped
flasks at various 1-propanol/gallic acid molar ratios (100–160) and
with variable amounts of immobilized lipase (100–160 IU). The
reaction mixture was incubated in an orbital shaker (Certomat
onto CaCO , as previously described [14].
3
2
.2.2. Lipase hydrolytic activity
The lipase activity was titrimetrically measured with a pH-Stat
◦
H/HK, Germany, Melsungen) at various temperatures (45–55 C)
(
Metrohm, Herisau, Switzerland) under optimal conditions using
at 200 rpm.

2
44
A. Bouaziz et al. / Journal of Molecular Catalysis B: Enzymatic 67 (2010) 242–250
Table 1
with y = yˆ + e and where yˆ is the estimated response function; y, the
measured response; e, the error, b , b , b and b , the estimated
Range of variables for the experimental design.
0
j
jk
jj
Variables
Levels
model coefficients. Xj are the coded variables which take the levels
1 and +1 when natural variables take their low and high levels,
respectively. In this study, the generation and the data treatment
of the Box–Behnken design are performed using the experimental
design software NemrodW [23].
−
−
1
0
+1
55
500
160
◦
X1: temperature ( C)
X2: enzyme amount (UI)
X3: 1-propanol/gallic acid molar ratio
45
100
100
50
300
130
2
2
.6. Reaction products analysis
A control without enzyme was run in parallel under the same
conditions. Aliquots of the reaction mixture were periodically with-
drawn. The immobilized enzyme was removed by centrifugation
at 8000 rpm for 5 min. The residual acid content was determined
by titrating with 0.07 N sodium hydroxide using phenolphthalein
solution as an indicator and 3 mL of ethanol. The conversion yield
of propyl gallate ester was calculated on the basis of the amount of
the consumed acid [18].
.6.1. HPLC analysis
In order to purify propyl gallate and to analyse the extracted
polyphenols, we have used an analytical Shimadzu C-18 column
composed of an LC-10ATvp pump and an SPD-10Avp detector. The
used mobile phase was 0.1% formic acid in water (A) versus 80%
acetonitrile in water (B) for a total running time of 15 min. The
absorbance was measured at 275 nm. The measurement of the area
of the new peak appeared after esterification allows us to confirm
the conversion yield obtained. About 100 ␮g of purified propyl gal-
late was obtained after HPLC analysis described above (data not
shown).
2.5. Experimental design
The optimization of the experimental conditions of a propyl
gallate synthesis was achieved by using the Response Surface
Methodology (RSM) [19]. In this work, a quadratic polynomial
model was set up in order to study the empirical relationship
between conversion yield and three controlled factors namely:
2
.6.2. FTIR and NMR analysis
In order to check the esterification process, we used FTIR NEXUS
spectrophotometer (Nicoleit, Madison, WIS, USA) showing the shift
of the carbonyl of carboxylic acid group to the carbonyl of ester
group. The sample was mixed with KBr. FTIR spectra were acquired
after 32 scans between 4000 and 400 cm with spectral resolution
of 4 cm
X , temperature; X , amount of immobilized lipase and X , 1-
1
2
3
propanol/gallic acid molar ratio. In this part of the paper, we
briefly discuss the principles governing the construction and analy-
sis of the experimental design. To estimate the model coefficients,
a three-variable Box–Behnken design, which requires 16 experi-
ments (Table 1), is carried out in the cubic experimental domain.
The experimental points are located in the middle of a cube ridges
−1
−1
.
The structure of the propyl gallate ester was also confirmed by
C NMR (Madison, USA). Then, samples were dissolved in CDCl3
13
containing trace amounts of tetramethylsilane which was used as
an internal chemical shift reference to indicate, in parts per million
(
[
12 experiments) and at the center of the cube (4 experiments)
20,21]. The model parameters are estimated by a least square fit-
(ppm), the difference of the resonance frequency.
ting of the model to experimental results obtained in the design
points. The adequacy of the model is tested using four check points
2.7. Antioxidant activity in refined soy oil
[
21,22].
Table 1 shows the selected variable levels. Table 2 shows the
Propyl gallate, gallic acid, ascorbic acid and butylated hydrox-
yanisole (BHA) were dissolved (1 mg/mL) in ethanol and added to
refined soy oil at 200 ␮g/g as previously described [24]. The pheno-
lic antioxidants were mixed with oil by stirring for 5 min. Samples
three-variable replicate Box–Behnken design used for fitting the
following second-order mathematical model.
◦
were stored in the dark at 70 C. The stability of oil was evaluated
yˆ = b + b X + b X + b X + b11X12 + b22X22 + b33X32
0
1
1
2
2
3 3
by the measurement of the conjugated dienes CD (K232), triene
formation CT (K270) and the peroxide value (PV).
+
b12X X + b X X + b X X
1
2
13
1
3
23 2 3
Table 2
Three-variable Box–Behnken experimental design.
Run
Variable values
Esterification yield
X1
X2
X3
Experimental
Predicted
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
−1(45)
1(55)
−1(45)
1(55)
−1(45)
1(55)
−1(45)
1(55)
0(50)
0(50)
0(50)
0(50)
0(50)
0(50)
0(50)
0(50)
−0.5774
0.5774
0
−1(100)
−1(100)
1(500)
1(500)
0(300)
0(300)
0(300)
0(300)
−1(100)
1(500)
−1(100)
1(500)
0(300)
0(300)
0(300)
0(300)
−0.3333
0.3333
0.6667
0
0(130)
0(130)
0(130)
35.40
49.30
77.90
82.00
69.10
74.30
78.70
83.90
42.60
73.80
60.00
89.40
82.00
85.30
82.70
84.70
64.91
72.76
82.64
86.42
38.17
50.17
77.02
79.22
66.42
73.52
79.47
86.57
42.5
77.35
56.45
89.50
83.67
83.67
83.67
83.67
69.98
75.02
86.23
87.82
0(130)
−1(100)
−1(100)
1(160)
1(160)
−1(100)
−1(100)
1(160)
1(160)
0(130)
0(130)
0(130)
0(130)
−0.2357
−0.2357
−0.2357
−0.7071
1
1
1
1
1
1
1
1
1
1
2
0

A. Bouaziz et al. / Journal of Molecular Catalysis B: Enzymatic 67 (2010) 242–250
245
Table 3
Analysis of variance.
Source of variation
Sum of squares
Degrees of freedom
Mean square
Ratio
Significance
0.0106***
Regression
Residuals
Total
3991.72
65.11
4056.84
0.984
9
6
15
443.525
10.853
40.867
2
R
*
**
Significant at the level 99.9%.
2
.7.1. Measurement of conjugated diene (CD) and triene (CT)
Conjugated diene (CD) and conjugated triene (CT) are suited for
enumerating the number of viable cells by plating serial dilutions
in the nutrient agar medium.
measuring the oxidative state of oil [25]. Refined oil (0.1 g) stored
◦
at 70 C was dissolved in 25 mL of hexane and the absorbance was
3. Results and discussion
measured at 232 and 270 nm, using hexane as a blank. All measure-
ments were performed in duplicate.
3.1. Optimization of conversion yield
2.7.2. Measurement of peroxide value (PV)
3.1.1. Preliminary study
Peroxide value was determined according to the method of
Low and Ng [26]. The oil sample (1 g) was dissolved in 25 mL of
a chloroform and acetic acid mixture (2/3, v/v). To this mixture,
We studied the ability of immobilized S. xylosus lipase, produced
in our laboratory, to synthesize the propyl gallate. A preliminary
study carried out with seven variables using the One Variable At
a Time method (OVAT) [20] showed that three variables had sig-
nificant effects on the response. These variables were the reaction
temperature (X1), the amount of immobilized lipase (X2) and the
1-propanol: gallic acid molar ratio (X3). The levels of the three other
variables having only a small effect on the esterification yield were
fixed as follows: reaction time: 6 h, molecular sieve: 0%, stirring
speed: 200 rpm. Hexane (2 mL) was chosen as a suitable solvent for
both gallic acid and propanol solubilization.
1
mL of saturated potassium iodide solution was added followed
by the addition of 75 mL of distilled water. The peroxide value was
determined by titrating the iodine liberated from potassium iodide
with standardised 0.01 N sodium thiosulfate solution using 0.5 mL
of starch solution (1%, w/v) as indicator. The peroxide value was
expressed as milli equivalents of free iodine per kg of lipid. All
measurements were performed in duplicate.
2
.8. Antimicrobial activity
3
3
.1.2. Experimental design results
.1.2.1. Experimental conditions. In order to define the explored
Antimicrobial activity of propyl gallate and gallic was tested
experimental domain, the level values of variables were chosen in
such a way that their limits were as wide as possible while all those
experiments were done (Table 1). The results are indicated in the
matrix design (Table 2).
against Escherichia coli and S. xylosus using disc diffusion and broth
macrodilution methods. Ampicillin was used as a standard refer-
ence. Propyl gallate and gallic acid were dissolved in ethanol at a
final concentration of 21 and 17 mg/mL, respectively. The solutions
were sterilized by the microfilter sterile. One-hundred microlitres
8
3.1.2.2. Model equation. The coefficients of the postulated model
were calculated on the basis of the experimental responses (with-
out including the checked points). The fitted model that expressed
in coded variables was represented by the following equation.
(
2 × 10 CFU/mL) of suspension were spread on the solid media and
grown for 24 h then spread over the solid surface of nutrient agar
medium. Filter paper discs (6 mm of diameter), loaded with propyl
gallate, gallic acid and ampicillin were placed on the surface of
◦
nutrient agar and incubated at 37 C for 24 h. Then, the diameters
2
1
2
2
yˆ = 83.6 + 3.5 X + 16.9 X + 6.5 X − 6.2 X − 16.2 X2 − 0.9 X3
1
2
3
of inhibition zones were measured. All experiments were done in
triplicates.
− 2.4 X1X2 + X1X3 − 0.4 X2X3
2
.8.1. Determination of minimum inhibitory concentration (MIC)
3
.1.2.3. Analysis of variance and validation of the model. The good
and minimum bactericidal concentration (MBC)
quality of the fitted model was attested with the analysis of the
variance (ANOVA) as demonstrated in Table 3. Indeed, this table
shows that the sum of squares related to the regression was sta-
tistically significant when using the F-test at a 99.9% probability
level, which suggests that the variation accounted for by the model
was significantly greater than the residual variation. Likewise, the
coefficient of multiple determination of the polynomial model
Minimum inhibitory concentration (MIC) and minimum bacte-
ricidal concentration (MBC) were determined in LB broth using a
macrobroth dilution method as described by the Clinical and Lab-
oratory Standards Institute (CLSI) (2007). The final concentration
of bacteria in each macrobroth dilution tube was approximately
6
6
.8 × 10 CFU/mL of LB.
The MIC was defined as the lowest concentration of the phe-
2
termed R , indicates that more than 98% of the variability in the
nolic compound that resulted in no visible growth after 24 h of
incubation at 37 C.
The MBC was defined as the lowest concentration of the propyl
gallate or the gallic acid at which inoculated microorganism was
completely killed. Ampicillin was used as a positive control. All
experiments were done in triplicates.
response could be explained by the second-order polynomial pre-
dicted equation given above. In this respect, the validity of the
model has been established by comparing the obtained results at
the four check points (experiments 17–20) with the predicted val-
ues (Table 2). These results seem to confirm the validity of the
model.
◦
2.8.2. Influence of propyl gallate in bacteria viability
3.1.3. Interpretation of the response surface model
Samples were withdrawn at selected time points and serial dilu-
The coordinates of the stationary point of the isoresponse curves
corresponding to a maximum of conversion yield are: temperature
of the reaction, 52 C; immobilized lipase, 400 IU; 1-propanol/gallic
tions were performed in sterile LB liquid before these samples were
spread onto LB agar medium in Petri dishes. Bacterial growth was
◦
◦
followed by incubation for 24 h at 37 C and was determined by
acid molar ratio, 160.

2
46
A. Bouaziz et al. / Journal of Molecular Catalysis B: Enzymatic 67 (2010) 242–250
Fig. 2. Contour plots and response surface plot showing: (A) The effects of temperature (X1) and the 1-propanol/gallic acid molar ratio (X3) and their mutual interaction on
the esterification yield at a constant enzyme amount (300 IU). (B) The effects of 1-propanol/gallic acid molar ratio (X3) and enzyme amount (X2) and their mutual interaction
◦
on the esterification yield. The temperature was fixed at 50 C. (C) The effects of the amount of immobilized lipase (X2) and temperature (X1) and their mutual interaction on
the esterification yield at a constant 1-propanol/gallic acid molar ratio of 160.

A. Bouaziz et al. / Journal of Molecular Catalysis B: Enzymatic 67 (2010) 242–250
247
1
00
(
A)
8
6
4
0
0
0
9
0
0
0
0
0
0
0
Immobilized lipase
Free lipase
8
7
6
5
4
3
2
1
Carbonyl group
20
0
1
750
0
0
0
1
500
0
500
1000
1500
2000
2500
3000
3500
4000
4500
wavennumbers (cm-1)
0
2
4
6
8
10
110
Time (h)
(B)
100
Fig. 3. The time course of propyl gallate synthesis using free and immobilized lipase.
The experimental conditions were: 400 IU of enzyme, 1-propanol/gallic acid molar
90
ratio of 160 dissolved in 2 mL hexane. The reaction mixture was stirred at 200 rpm at
80
◦
5
2
C. A control without enzyme was carried out under the same condition. Experi-
1750
ments were carried out in triplicates.
70
1
500
60
The response surfaces and the isoresponse curves are plotted
in function of two factors while the third was maintained con-
stant at its main level (Fig. 2A–C). These figures show that the
conversion yield was highly enhanced (90%) by increasing the 1-
propanol/gallic acid molar ratio (over 160). These results are in
agreement with those obtained by Xiao and Yong who stipulate that
the stability of lipases is strongly affected when gallic acid is used
as substrate [27]. Also, the contour plots in Fig. 2B and C emphasize
the important role of temperature and quantity of enzymes. Indeed,
the conversion yield decreased dramatically at a low temperature.
Also, high enzyme levels (above 400 IU) decreased the conversion
yield dramatically. This result can be explained by the fact that at
high amount of lipase, the molecules of the enzyme can aggregate
together and the active site cannot be exposed to the substrates.
In order to confirm these results, three independent, com-
plementary experiments were carried out under the optimal
conditions (coordinates of the stationary point given above).
Indeed, the time course of the propyl gallate synthesis by free
and immobilized S. xylosus lipase was made under optimal con-
ditions (Fig. 3). From the presented data, one can note that the
conversion yield of the propyl gallate synthesis increases rapidly to
reach its maximum (90 ± 3.5%) after 4 h of incubation time. Thus,
50
4
3
2
0
0
0
0
500
1000
1500
2000
2500
3000
3500
4000
4500
wavenumbers (cm-1)
Fig. 5. FTIR spectra of the propl gallate adsorbed to CaCO3 before (A) and after (B)
the wash step by the ethanol. The carbonyl group of the ester was marked by the
arrow.
the obtained experimental conversion yield was very close to the
predicted value estimated (96.86 ± 3.70%). Noteworthy, the con-
version yield of the obtained propyl gallate synthesis when using
immobilized SXL, is higher than that obtained by free lipase. This
result is also in agreement with that obtained by Ghamgui et al.
[28] and Horchani et al. [29] who state that the immobilization
of staphylococcal lipases into CaCO3 increases both their stabil-
ity at high temperature and their ability to tolerate the presence
of organic solvents. In addition, one can relate the low conver-
sion yield obtained with free enzyme and its inactivation at the
Fig. 4. HPLC analysis of the reaction mixture obtained after the esterification with free lipase (A). Propyl gallate profil found in the washing flow of the CaCO3 support (B).

2
48
A. Bouaziz et al. / Journal of Molecular Catalysis B: Enzymatic 67 (2010) 242–250
interface of the biphasic media. In order to confirm these findings,
a reaction mixture was carried out using free lipase at different
concentrations of molecular sieve (data not shown). An increase
in the conversion yield was observed reaching 70% after 4 h of
incubation.
reaction mixture before and after esterification (data not shown),
showing that the intensity of the band appeared at 168.6 ppm,
which corresponds the carbonyl group of the synthesized ester,
decreased significantly after washing CaCO with ethanol (data not
3
shown).
3
3
.2. Analysis of the reactions products
3
.3. Antioxidant activity
.2.1. HPLC analysis
The HPLC analysis of the reaction mixture obtained after esterfi-
In order to determine its antioxidant power, the propyl gal-
late was added to the refined soy oil at 200 ␮g/g and the results
were compared to BHA with ascorbic acid and gallic acid used as
standard antioxidants. The oxidation of refined oil, stored at 70 C,
fication (Fig. 4) shows paradoxically the presence of propyl gallate
only when free lipase is used (Fig. 4A). This apparent contradiction
could be related to the adsorption of the ester to CaCO . To confirm
◦
3
was monitored by measuring conjugated diene (CD) and triene
formation (CT) (Fig. 6A and B). The conjugated diene, measured
at 232 nm, determines the primary product of oil oxidation. The
conjugated triene, measured at 270 nm, enquires the secondary
oxidation products such as aldehydes and ketones, and hence may
be an indication of the formation of the final products of oil per-
oxidation. From the data presented in Fig. 6A and B, a significant
increase was observed in CD and CT for controlled refined soy
oil. The addition of antioxidants exhibited prominent antioxidant
activity. Results indicated clearly that the propyl gallate shows a
high antioxidant activity compared to that of BHA (Fig. 6A and B).
So, the diminution of the absorbance at 232 nm and 272 nm due to
the formation of conjugated diene (CD) and triene, after the addi-
tion of propyl gallate, confirms the high protective effect of propyl
gallate against oxidation of refined soy oil.
this hypothesis, the reaction mixture was removed by centrifuga-
tion and the CaCO3 was washed twice with 1 mL of ethanol. The
HPLC analysis of the washing flow showed the presence of a single
peak corresponding to propyl gallate (Fig. 4B) emerging at the same
time as the pure commercial propyl gallate used as reference.
3
.2.2. FTIR measurement and 13C NMR
The chemical structure of propyl gallate was confirmed by
FTIR by analyzing the CaCO3 before (Fig. 5A) and after (Fig. 5B)
−
1
being washed by ethanol. Typical stretches at 1750 cm
1
and at
−1
500 cm corresponding to the carbonyl ester and the aromatic
groups, respectively, were observed before CaCO3 washing with
ethanol. The intensity of these bands decreased significantly after
the wash step with ethanol. This result confirms the fact that the
propyl gallate was synthesized successfully and was adsorbed to
The antioxidant power of propyl gallate was also tested by mea-
suring the peroxide value after adding various concentrations of
13
CaCO . The same result was observed after C NMR analysis of the
3
100
12
(
A)
(B)
90
80
70
60
50
40
30
20
10
0
1
0
8
6
4
2
0
0
10
20
30
40
0
10
20
30
40
Time (days)
Time (days)
800
700
600
500
400
300
200
100
0
(
C)
0
5
10
15
20
25
30
35
Time (days)
Fig. 6. Antioxidant activities of propyl gallate (o) and gallic acid (ꢀ) tested by the measurement of conjugated dienes (A), conjugated trienes (B) and peroxide values (C) of
◦
refined soy oil during storage at 70 C. For each test, negative (ꢀ) and positive controls using BHA (ꢁ), ascorbic acid (ꢂ) were run in parallel.

A. Bouaziz et al. / Journal of Molecular Catalysis B: Enzymatic 67 (2010) 242–250
249
Table 4
Antimicrobial activity of the propyl gallate, gallic acid and ampicillin.
Microorganisms
Propylgallate
MBCb
Gallic acid
DDa
Ampicillin
DDc
DDa
MICb
MICb
MBCb
MICb
MBCb
Staphylococcus xylosus
Staphylococcus aureus
Escherichia coli
31
28
36
1
2
0.5
1
3
0.5
24
22
25
1
2
1
2
4
2
35
30
20
0.02
0.2
0.02
0.05
0.2
0.04
DD diameter of zone of inhibition (mm) including disc diameter of 6 mm.
a
Tested at concentration of 1.5 mg/disc.
Values given as mg/ml.
Tested at concentration of 10 ␮g/disc.
b
c
propyl gallate (Fig. 6C). In fact, the measurement of the primary oil
oxidation is widely used and it indicates the amount of peroxides
formed in fats and oil during oxidation [30]. Changes in peroxide
values are presented in Fig. 6C. Peroxide value of the controlled
refined soy oil increased from 6 to 670 meq/kg after 32 days of
microorganisms by blocking respiration and nucleic acid synthesis
[32]. This antibacterial activity of the hydrophobic propyl gallate
molecule is probably due to its increased interaction with the cyto-
plasmic membrane, which results in the inhibition of the oxygen
consumption and the disruption of the membrane located in the
respiratory chain [33].
◦
incubation at 70 C. After adding 300 ppm of propyl gallate, the
peroxide values decreased significantly to reach 115 meq/kg. The
analytical investigations we have explained so far reveal that the
synthesized propyl gallate is an excellent antioxidant for refined
soy oil. Its activity is comparable to that of BHA and much higher
than that of ascorbic acid, both of which are used as standard
antioxidants.
Acknowledgments
This work is a part of a doctoral thesis by Ahlem BOUAZIZ
whose research was supported financially by “Ministère de
l’Enseignement Supérieur et de la Recherche scientifique-Tunisia”
through a grant to “Laboratoire de Biochimie et de Génie Enzyma-
tique des Lipases-ENIS”.
3.4. Antimicrobial activity
The antimicrobial activity of the propyl gallate and gallic acid
References
was evaluated against Gram-positive (S. xylosus and S. aureus)
and Gram-negative (E. coli) bacteria using disc diffusion and broth
macrodilution methods. The disc diameters of the inhibition zone
[
[
1] J. Karovicova, P. Simko, J. Chromatogr. A 882 (2000) 271–281.
2] D.B. Clayson, F. Iversion, E. Nera, E. Lok, C. Rogers, C. Rodrigues, Food Chem.
Toxicol. 24 (1986) 1171–1182.
(
DD), the minimum inhibitory concentrations (MIC) and the min-
[
[
3] N. Ito, M. Hirose, S. Fukushima, H. Tsuda, T. Shirai, H. Tatematsu, Food Chem.
Toxicol. 24 (1986) 1071–1082.
4] J.W. Daniel, Xenobiotica 16 (1986) 1073–1078.
imum bactericidal concentration (MBC) of both the propyl gallate
and gallic acid against the susceptible pathogenic organisms are
shown in Table 4. The antimicrobial activity was compared to
ampicillin, a currently used antibiotic of broader spectrum (Gram-
positive and Gram-negative bacteria). According to the results
presented in Table 4, one can note that propyl gallate as well as
gallic acid showed a good inhibitory effects against Gram-positive
and Gram-negative bacteria species being tested. The diameters of
the inhibition zone of the newly synthesized propyl gallate (1.5 mg)
were: 36 mm for E. coli, 31 mm for S. xylosus and 28 mm for S. aureus
The MIC values of propyl gallate were on the range of 1 mg/mL
on S. xylosus and 0.5 mg/mL on E. coli. We found also that gallic
acid and propyl gallate showed the same MIC values against Gram-
positive bacteria. Yet, the propyl gallate showed better inhibitory
effects on Gram-negative bacteria (MIC = 500 ␮g/mL) than the gallic
acid (MIC = 1000 ␮g/mL)
[5] J.C. Dacre, Food Cosmet. Toxicol. 12 (1974) 125–129.
[
6] T.W. Wu, K.P. Funy, L.H. Zeny, J. Wu, H. Nakamura, Biochem. Pharmacol. 48
1994) 419–422.
7] P.M. Rosin, F. Stich, Cancer Lett. 9 (1980) 299–305.
(
[
[8] K.M. Abdo, J.E. Huff, J.K. Haseman, C.J. Alden, Food Chem. Toxicol. 24 (1986)
1091–1097.
[
9] J.R. Reddan, F.J. Giblin, M. Sevilla, V. Padgaonkar, C.D. Dziedzic, R.V. Leverenz,
C.I. Misra, S.J. Chang, T.J. Pena, Exp. Eye Res. 76 (2003) 49–59.
[10] S. Raghavan, H.J. Hultin, Food Biochem. 33 (2005) 163–175.
[11] M. Hirose, S. Takahashi, K. Ogawa, M. Futakuchi, T. Shirai, Food Chem. Toxicol.
37 (1999) 985–992.
[
12] K. Karthikeyan, B.R. Sarala Bai, K. Gauthaman, S. Niranjali Devaraj, J. Pharm.
Pharmacol. 57 (2005) 67–73.
[13] W.K. Jeon, B.C. Kim, Biochem. Biophys. Res. Commun. 361 (2007) 645–650.
[
14] H. Horchani, M. Chaâbouni, Y. Gargouri, A. Sayari, Carbohydr. Polym. 79 (2010)
66–474.
4
[
15] Y. Gargouri, R. Julien, A. Sugihara, L. Sarda, R. Verger, Biochim. Biophys. Acta
795 (1984) 326–331.
[
[
[
16] H. Nawaz, J. Shi, G. Mittal, Y. Kukudo, Sep. Purif. Technol. 48 (2006) 176–181.
17] V. Nwuha, J. Food Eng. 44 (2000) 233–238.
18] X. Wu, J. Li, X. Li, C.L. Hsieh, P.M. Burgers, M.R. Lieber, Nucleic Acids Res. 24
(1996) 2036–2043.
The highest effect of propyl gallate on Gram-negative more than
on Gram-positive bacteria is probably due to the more complex cell
wall structure of the Gram-positive bacteria which has additional
lipopolysaccharides on the outer surface, which generally reduces
the ability of most antibiotics and extracts to penetrate the bacterial
cells [31].
[
[
19] L.A. Sarabia, M.C. Ortiz, Chem. Biochem. Data Anal. 1 (2009) 345–390.
20] R. Carlson, Design and Optimization in Organic Synthesis, New York, 1992, pp.
26–27.
[21] J. Goupy, Plan d’expériences pour surface de réponses, DUNOD, Paris, 1999, pp.
253–293.
The bacterial effect of propyl gallate against S. xylosus was
confirmed by the time kill curve experiment (data not shown).
Cultures of a Gram-positive (S. xylosus) and a Gram-negative (E.
[
22] G.A. Lewis, G.A. Mathieu, T.R. Phan, Pharmaceutical Experimental Design, Mar-
cel Dekker Inc., New York, 1999.
[23] D. Matheiu, J. Nony, T.R. Phan, NEMROD-W Software, LPRAI, Marseille, 2000.
[24] W. Della, Y. Wong, C. Mak, C. Sze, W. Yao, J. Food Compos. Anal. 19 (2006)
6
coli), with a cell density of 6.8 × 10 CFU/mL, were exposed to dif-
784–791.
ferent concentrations of propyl gallate. The number of viable cells
was determined after different periods of incubation. It was found
that propyl gallate reduced the number of viable Gram-positive
or Gram-negative cells rapidly and within the 1 h. No viable cells
were detected after being exposed to 1 mg/mL of propyl gallate. In
fact it has been reported that propyl gallate inhibits the growth of
[
25] L. McGinely, in: J.B. Rossel, J.L.R. Printed (Eds.), Analysis of Oil Seeds, Fat and
Fatty Food, Elsevier, Applied Science, New York, 1991, pp. 440–470.
[26] L.K. Low, C.S. Ng, in: H. Hasegawa (Ed.), Laboratory Manual on Analytical Meth-
ods and Procedures for Fish and Fish Products, Marine Fisheries Research
Department, Southeast Asian Fisheries Development Center, Singapore, 1978,
pp. C7.1–C7.3.
[27] W.Y. Xiao, Q.L. Yong, J. Mol. Catal. B: Enzyme 40 (2006) 44–50.

2
50
A. Bouaziz et al. / Journal of Molecular Catalysis B: Enzymatic 67 (2010) 242–250
[
[
28] H. Ghamgui, M. Karra-Châabouni, Y. Gargouri, Enzyme Microb. Technol. 35
2004) 355–363.
29] H. Horchani, S. Ouertani, Y. Gargouri, A. Sayari, J. Mol. Catal. B: Enzyme 61 (2009)
94–201.
[30] O. Gulcan, H. Bedia, J. Food Eng. 79 (2007) 1391–1396.
(
[31] H.P. Rang, M. Dale, Pharmacology, Churchill, Edinburgh, 1987.
[32] I. Boyd, E.G. Beveridge, Microscopy 24 (1979) 173–184.
[33] I. Kubo, P. Xiao, K. Fujita, Bioorg. Med. Chem. Lett. 12 (2002) 113–116.
1