Hydrolysis of fish oil by lipases immobilized inside porous supports

Article abstract of DOI:10.1007/s11746-010-1728-1

A new assay was designed to measure the release of omega-3 acids [eicosapentaenoic acid (EPA) and docosahexaenoic acid (DHA)] from the hydrolysis of sardine oil by lipases immobilized inside porous supports. A biphasic system was used containing the fish

Full text of DOI:10.1007/s11746-010-1728-1

J Am Oil Chem Soc (2011) 88:819–826
DOI 10.1007/s11746-010-1728-1
ORIGINAL PAPER
Hydrolysis of Fish Oil by Lipases Immobilized Inside
Porous Supports
•
Gloria Fernandez-Lorente Carolina Pizarro Dolores Lopez-Vela
•
•
´
´
•
•
•
Lorena Betancor Alfonso V. Carrascosa Benevides Pessela
Jose M. Guisan
Received: 11 December 2009 / Revised: 16 July 2010 / Accepted: 17 November 2010 / Published online: 14 December 2010
Ó AOCS 2010
Abstract A new assay was designed to measure the
release of omega-3 acids [eicosapentaenoic acid (EPA) and
docosahexaenoic acid (DHA)] from the hydrolysis of sar-
dine oil by lipases immobilized inside porous supports. A
biphasic system was used containing the fish oil dissolved
in the organic phase and the immobilized lipase suspended
in the aqueous phase. The assay was optimized by using a
very active derivative of Rhizomucor miehei lipase (RML)
adsorbed onto octyl-Sepharose. Standard reaction condi-
tions were: (a) an organic phase composed by 30/70 (v:v)
of oil in cyclohexane, (b) an aqueous phase containing
50 mM methyl-cyclodextrin in 10 mM Tris buffer at pH
7.0. The whole reaction system was incubated at 25 °C.
Under these conditions, up to 2% of the oil is partitioned
into the aqueous phase and most of the 95% of released
acids were partitioned into the organic phase. The organic
phase was analyzed by RP-HPLC (UV detection at
215 nm) and even very low concentrations (e.g., 0.05 mM)
of released omega-3 fatty acid could be detected with a
precision higher than 99%. Three different lipases adsorbed
on octyl-Sepharose were compared: Candida antarctica
lipase-fraction B (CALB), Thermomyces lanuginosa lipase
(TLL) and RML. The three enzyme derivatives were very
active. However, most active and selective towards poly-
unsaturated fatty acids (PUFA) versus oleic plus palmitic
acids (a fourfold factor) was CALB. On the other hand,
the most selective derivatives towards EPA versus DHA
(a 4.5-fold factor) were TLL and RML derivatives.
Keywords Enzymatic release of omega-3 acids Á
Selective enzymatic release of EPA Á Solubilization of oil
by randomly methylated b-cyclodextrins
Introduction
The release of omega-3 fatty acids (e.g., EPA and DHA)
from fish oil may mean an interesting first step in the
preparation of highly enriched triglycerides (70–90% of
content in one or both polyunsaturated fatty acids, PUFA).
These triglycerides have been described as excellent
functional ingredients [1]. Recently, PUFA have meant a
breakthrough amongst health care professionals of the
beneficial effects of omega-3 fatty acids derived from fish
oils—mainly consisting of docosahexaenoic acid (DHA)
plus eicosapentaenoic acid (EPA) mixtures. DHA is
required in high levels in the brain and retina as a physi-
ologically essential nutrient to provide for optimal neuronal
functioning (learning ability, mental development) and
visual acuity, in the early stages of the life [2]. On the other
hand, EPA is considered to have beneficial effects in the
prevention of cardiovascular diseases in adults [3, 4]. In
general, both PUFA have been reported useful in pre-
venting a significant number of health disorders [5–10].
The enzymatic release of PUFA may have important
advantages over the chemical release: mild reaction condi-
tions, absence of undesirable byproducts, etc. Furthermore,
´
C. Pizarro Á D. Lopez-Vela Á J. M. Guisan (&)
Instituto de Catalisis, CSIC, Campus UAM-Cantoblanco,
28049 Madrid, Spain
e-mail: jmguisan@icp.csic.es
´
G. Fernandez-Lorente Á A. V. Carrascosa Á B. Pessela
Instituto de Fermentaciones Industriales, CSIC,
c/ Juan de la Cierva 3, 28006 Madrid, Spain
L. Betancor
Laboratorio de Biotecnologıa, Universidad ORT Uruguay,
´
Cuareim 1451, Montevideo, Uruguay
123

820
J Am Oil Chem Soc (2011) 88:819–826
the use of immobilized lipases may have additional tech-
nological and economical advantages for the hydrolysis.
Immobilization of enzymes inside porous supports is an
interesting challenge in enzyme biotechnology. Commer-
cially available supports with excellent mechanical prop-
erties and high enzyme loading capacity can be selected. In
this way, a very active enzyme catalyst can be used in any
type of industrial reactor (stirred tank, packed bed, etc.). In
addition, since the easy workability of porous supports,
some immobilization protocols have been used to improve
enzyme properties (e.g., stabilization by a very intense
multipoint covalent immobilization [11], hyperactivation
of lipases by immobilization and post-immobilization
techniques [12, 13], reactivation of partially inactivated
lipases [14], etc.)
(e.g., oil interfaces, solvent interfaces, small bubbles of
oxygen, etc.).
On the other hand, a rapid evaluation of the intrinsic
properties of new lipases on these oily substrates (activity,
stability, selectivity) may be directly tested using immo-
bilized derivatives even before any previous purification
process. Developing this partial but very rapid and highly
sensitive assay for such purposes may become of great
interest.
Using biphasic systems (water/immiscible solvents) may
be interesting for the above mentioned assays. Oil will be
partitioned into the organic phase and the enzymatic cat-
alyst will be in the aqueous phase only acting on isolated
oil molecules that are partitioned into the aqueous phase.
Pure oils are very viscous and difficult to work with, but
mixtures oil/organic solvents (e.g., cyclohexane) could
simplify the reaction design. In the case of fish oil, their
high content in PUFA (as acyl chains forming part of tri-
acylglycerol in the oil) could facilitate the aqueous/organic
partition of oil molecules. Additionally, released PUFA
could be easily detected by RP-HPLC with UV detector.
The use of cyclodextrins to increase the solubility of apolar
compounds in water has been widely utilized [15, 16].
Randomly methylated b-cyclodextrins (RMBCD) seem to
be much more effective than unmodified b-cyclodextrins in
order to improve the solubility of hydrophobic compounds
in the aqueous phase [17, 18]. A degree of substitution
(DS) of 1.8 has been reported as the optimal one. [19].
Hydrolysis of fish oils by lipases immobilized inside
porous supports is an outstanding example of lipase
application in biotechnology. These oils from marine origin
are the most important source in omega-3-PUFA in nature,
hence their importance in food technology to make new
food additives. In nature, soluble lipases suffer interfacial
activation by oil drops. In contrast, using immobilized
lipases on porous materials, the enzyme can only hydrolyze
the oil molecules partitioned into the aqueous phase
(Fig. 1). However, this limitation is compensated for by the
plethora of advantages that immobilization gives to
enzymes in order to be used in industrial biotransforma-
tions. From a practical point of view, immobilized lipases
can be re-used for a number of reaction cycles and lipases
(inside a porous structure) cannot be inactivated by
hydrophobic interfaces present in strongly stirred reactors
Materials and Methods
Materials
Lipases from CAL-B (Novozym 525L), Rhizomucor miehei
(Palatase) and Thermomyces lanuginosa (Lypozyme TL)
were generously donated by Novozymes (Bagsvaerd,
Denmark). Randomly methylated BCD (RMBCD, methyl-
cyclodextrin) was purchase from Cyclodextrin Resource,
CTD Inc. (High Springs, FL, USA). Triton X-100,
p-nitrophenyl butyrate (pNPB), docosahexaenoic acid
(DHA) and eicosapentaenoic acid (EPA) were obtained
from Sigma Chemical Co. (St. Louis, MO, USA). Octyl
SepharoseTM CL-4B was purchased from GE Healthcare
(Uppsala, Sweden). Sardine Oil was generously donated by
C
A
B
D
´
BTSA, Biotecnologıas Aplicadas, S.L. (Madrid, Spain).
Other reagents and solvents were of analytical or HPLC
grade.
Fig. 1 Some special features of the hydrolysis of oils are strongly
stirred biphasic systems catalyzed by lipases immobilized on porous
supports. a Immobilized lipase is only acting on fully water soluble
oil molecules. b There is a low partition between oil in the organic
phase and oil in the aqueous phase. c Gas bubbles promoted by strong
stirring are not able to penetrate the porous structure of the catalyst.
d Drops of the organic phase are not able to penetrate inside the
porous structure of the catalyst
Enzyme Activity Assay
Enzyme activity was measured by monitoring the increase
in absorbance at 348 nm produced by the release of
123

J Am Oil Chem Soc (2011) 88:819–826
821
p-nitrophenol in the hydrolysis of 0.4 mM p-nitrophenyl
butyrate (pNPB) in 25 mM sodium phosphate at pH 7 and
25 °C. To initialize the reaction, 0.1 mL of lipase solution
or suspension was added to 2.5 mL of substrate solution.
An International Unit of pNPB activity is defined as the
amount of enzyme necessary to hydrolyze 1 lmol of
pNPB/min (IU) under the above described conditions.
initiated by adding different amounts (from 0.1 up to 1 g)
of lipase derivative and the reaction suspension was stirred
at 150 rpm. A pH–stat Mettler Toledo DL50 graphic was
used to maintain the pH at a constant value during the
reactions. The concentration of free fatty acids was deter-
mined at various times by HPLC–ELSD and HPLC–UV
methods.
Immobilization of Lipases on Octyl-Sepharose
Supports
Analysis of Polyunsaturated Free Fatty
Acids (PUFA) by HPLC–UV
Ten grams of octyl-Sepharose were added to 900 mL of
5 mM sodium phosphate buffer at pH 7 containing 250 mg
of commercial extracts of lipase (containing a 35% of
RML, a 65% of TLL and a 95% of CALB) and the sus-
pension was incubated at 25 °C. A blank of each soluble
enzyme with the same concentration was incubated under
the same experiment conditions. The activity of the blanks
was fully preserved for 4 h. In contrast, the activity of the
supernatant of the immobilization suspension decreased
down to less than 5%. That is, in all cases, more than 95%
of the lipase enzyme was immobilized. After 4 h, the
immobilized preparation was filtered and washed with
distilled water. The percentage of recovered activity of
immobilized derivatives was very low because of mass
transfer limitations. Hence, the derivatives were broken (to
reduce particle size down to 1 lm) by vigorous magnetic
stirring. After this treatment the recovered activities
(towards pNPB) were 120% for CALB derivative, 1,000%
for RML derivative and 2,000% for TLL derivative. The
same hyperactivation of lipases adsorbed on octyl-
Sepharose was also observed for very low loaded deriva-
tives (e.g., 0.1 mg of lipase per gram of support) and this
hyperactivation is due to interfacial activation of lipases on
hydrophobic supports [20–22]. The activity of low loaded
derivatives in the absence of diffusional problems were
2.4 IU/wet gram of CALB, 39 IU/wet gram of TLL and
9 IU/wet gram of RML. Lipase derivatives were stored in
wet form after a final filtration with 10 mM of Tris–Cl pH
7.0 containing 0.02% of sodium azide in order to prevent
microbial contamination.
After a given time, aliquots of 0.1 mL of organic phase
were withdrawn and dissolved in 0.8 mL of acetonitrile.
The organic phase was easily separated (in less than 5 min)
from the aqueous phase when stirring of the biphasic sys-
tem was stopped. The unsaturated fatty acids produced
were analyzed by RP-HPLC [Spectra Physic SP 100 cou-
pled with an UV detector Spectra Physic SP 8450 (Spectra
Physics, Santa Clara, CA. USA)] using a Kromasil C8
(15 cm 9 0.4 cm) column. Products were eluted at a flow
rate of 1.0 mL/min using acetonitrile-10 mM Tris–OH
buffer at pH 3 (70:30, v/v) and UV detection performed at
215 nm. The retention times for the unsaturated fatty acids
were: 9.4 min for EPA and 13.5 min for DHA. These
enzymatically produced PUFA were compared to their
corresponding pure commercial standards.
Analysis of Free Fatty Acids by HPLC–ELSD
After a given time, aliquots of 0.1 mL of the organic phase
were withdrawn and dissolved in 0.8 mL of propanol. The
fatty acids produced were analyzed by RP-HPLC (Spectra
Physic SP 100 coupled to a ELSD 2000 evaporative laser
light scattering detector). N2 was used as the nebulizing gas
at a flow of 3 mL/min, and a nebulizing temperature of
30 °C. A Kromasil C18 (25 cm 9 0.4 cm) column was
used and a stepwise elution was performed using solvent A
(acetonitrile/water, 98:2, v/v), solvent B (acetonitrile/pro-
panol, 95:5, v/v) and solvent C (pure acetonitrile). The
stepwise elution was performed as follows: 0–6.5 min:
100% A; 6.5–7.5 min: 50% A and 50% B; 7.6–15 min:
100% B; 15.1–18 min: 50% B and 50% C. 18.1–22 min:
100% C. The flow rate was set to 1 mL/min. The retention
times for the 4 main fatty acids [EPA, DHA, oleic acid
(OA) and palmitic acid (PA)] were between 5 and
13.5 min. The column was thermostated at 40 °C.
All experiments describe here and in the next sections
were carried out in triplicate with an error lower than 5%.
The values used in tables and figures were the average
value of the three determinations.
Hydrolysis of Sardine Oil
Partition of Sardine Oil into the Aqueous Phase
The hydrolysis of sardine oil was performed in an organic/
aqueous two-phase system. The procedure was as follows:
5 mL of cyclohexane, 5 mL of Tris–Cl buffer (0.1 M) pH
7, and 0.5 mL of sardine oil were placed in a reactor and
pre-incubated at 25 °C for 30 min. The reaction then was
Sardine oil exhibit a high absorbance in UV detection
because of the presence of around 30% PUFA in its structure.
In this way, by using a very hydrophobic mobile phase (98%
acetonitrile/2% water) and a short and lowly hydrophobic
123

822
J Am Oil Chem Soc (2011) 88:819–826
column (Kromasil C4-15 cm) a number of peaks were
obtained from RP-HPLC with RT comprised between 15 and
25 min. The total area of these peaks was proportional to the
amount of injected oil. For example, an injected sample of
20 lL of 0.5 mM of sardine oil (at 1 ml/min) gave a total
area of triglycerides of 20,000,000. On the other hand, these
peaks completely disappeared after total saponification of
the oil.
samples. By using pure commercial EPA and DHA dif-
ferent standard solutions with different concentration of
each omega-3 fatty acids in cyclohexane were prepared.
Then, 20-lL samples of each standard solution were ana-
lyzed by RP-HPLC with UV detection at a constant 1 ml/
min flow rate. Integration areas were perfectly proportional
to PUFA concentrations (Fig. 2). The absorbance of DHA
was a 50% higher than such one from EPA. UV–HPLC
chromatograms of aliquots of the organic phase of the
reaction mixture were compared to the corresponding
calibration curves and hence the rates of hydrolysis
(EPA ? DHA) and the EPA/DHA ratios could be easily
calculated.
A biphasic system was generated by mixing 20 ml of
aqueous phase (different buffer and methyl-cyclodextrin
concentration) and 20 ml of organic phase (1:1 oil/cyclo-
hexane, 0.54 M of oil). The biphasic mixture was vigor-
ously stirred for 1 h and then stirring was stopped. Both
phases were rapidly separated (in less than 5 min). The
organic phase was diluted 1,000-fold with mobile phase
and 20 lL of the new solution were analyzed by HPLC.
The aqueous was diluted 5-fold with mobile phase and
20 lL of the new solution were analyzed by HPLC. The
ratio between the concentration of sardine oil in the
aqueous phase and the concentration of sardine oil in
the organic phase could be easily calculated.
DHA exhibited an absorbance 50% higher than EPA at
215 nm (Table 2). In contrast, a similar concentration of
oleic acid absorbed around 1% in comparison to PUFA.
Besides, the polyunsaturated structure of PUFA makes
them less apolar and hence the retention times in
RP-HPLC under isocratic conditions are relatively low
(Table 2).
Therefore, using an isocratic HPLC under the previously
mentioned conditions, a rapid, easy and highly precise
analysis of omega-3 fatty acids released by enzymatic
hydrolysis of fish oil could be designed. For instance, when
analyzing a sample of 0.1 mM DHA utilizing a 1 ml/min
flow the peak area was very high (over 4,800,000 10,000
units with a very low baseline noise) (Table 2). On the
other hand, sardine oil was not spontaneously hydrolyzed
at pH 6.0 for 24 h. Thus, EPA and DHA conversions lower
than 0.2% in 24 h (concentration of 0.05 mM of PUFA in
the organic phase) were easily detected. A rapid pre-
liminary evaluation of very lowly active immobilized
lipases was performed by this protocol. In addition to that,
the enzyme selectivity (EPA/DHA ratio) can also be
measured using this protocol because of the high resolution
to separate both EPA and DHA using HPLC.
Results and Discussion
Detection of Traces of PUFA by Using Isocratic
RP-HPLC with UV Detection
Polyunsaturated fatty acids (PUFA) exhibited a high
absorbance at low wavelengths in the UV region. EPA
exhibited a very high absorbance at 210, 215 and 220 nm
(Table 1). A wavelength of 215 was selected in order to
minimize interference by impurities present in sardine oil
Table 1 UV detection of eicosapentanoic acid (EPA) eluted from a
RP-HPLC system
k (nm)
Peak area (EPA)
Use of HPLC with a Light Scattering Detector
for the Detection of Traces of Fatty Acids
210
215
220
8,130,482 (100%)
3,224,882 (40%)
128,049 (15%)
A ‘‘light scattering’’ detector (ELSD) was used to analyze
the ratios between polyunsaturated fatty acids and the
abundant oleic and palmitic acids also present in fish oil
Mobile phase: (70/30), acetonitrile: 10 mM ammonium phosphate pH
3.0, flow rate: 1 ml/min, injected sample: 20 lL of 0.1 mM EPA
Fig. 2 UV response of EPA
and DHA at different
concentrations. The response is
expressed as absorbance units
observed for the corresponding
peaks in the RP-HPLC
70
60
50
40
30
20
10
0
70
EPA
DHA
60
50
40
30
20
10
0
chromatograms. Injected
samples: 20 lL of solutions
with different concentrations
(from 0.125 to 2 mM) of PUFA
0
0.25 0.5 0.75
1
1.25 1.5 1.75
2
0
0.25 0.5 0.75
1
1.25 1.5 1.75
2
[mM]
[mM]
123

J Am Oil Chem Soc (2011) 88:819–826
823
oil concentrations) the biphasic systems water-immiscible
cosolvents (e.g., cyclohexane) was approached. Firstly, it
was proposed 1:1 ratio aqueous/organic phase where the
aqueous phase contained the lipase immobilized inside
porous structures (e.g., agarose gels). In this case, the
enzyme inside the porous could only hydrolyze the oil
molecules present in the aqueous phase of the system.
The biphasic system and the reaction conditions were
optimized by using a very active RML derivative con-
taining 5 mg of RML adsorbed per ml of octyl-Sepharose
4BCL and showing 10-fold hyperactivation due to stabil-
ization of the active (open) form of the lipase [25].
Table 2 UV detection of different fatty acids separated in a RP-
HPLC system
Fatty acid
Peak area, units
Retention time (min)
Oleic acid
EPA
35,899
3,224,882
4,807,103
21.5
9.4
DHA
13.5
Mobil phase: (70/30), acetonitrile: 10 mM ammonium phosphate pH
3.0, flow rate: 1 ml/min, injected sample: 20 lL of 0.1 mM fatty acid,
wavelength: 215 nm
Table 3 Effect of buffer concentration on the rate of hydrolysis of
sardine oil catalyzed by immobilized RML
Aqueous phase
Activity
Selectivity
Partitioning of the Released Fatty Acids
10 mM Tris–Cl buffer
100 mM Tris–Cl buffer
8.91
3.67
4.85
3.03
The partition of a DHA standard in a biphasic system 1:1 at
different pH values was studied. Over a wide range of pH
(pH 5–8) the non ionized acid was mainly found in the
organic phase. Similar values were obtained with EPA.
Since, it has been reported for PUFA a pK around 9–10
[26], pH 7 seems to be suitable for performing the hydro-
lysis of fish oils with a number of lipases independently of
their corresponding optimal pH. In this sense, the product
to analyze (PUFA) can be directly withdrawn from the
organic phase. Another advantage of this system is the low
concentration of released acids in the aqueous phase which
would avoid product inhibitions issues and/or saponifica-
tion processes even at high hydrolytic conversions.
A water/cyclohexane biphasic reactor at pH 7.0 and 25 °C was used
Activity is expressed as micromoles of PUFA (EPA ? DHA)
released per minute and per gram of immobilized enzyme
Selectivity is expressed as the ratio between EPA and DHA
[23]. The detection of oleic and palmitic acids (at high
concentrations, e.g., 0.5–1 mM) was much more sensitive
using this detector than the UV one and hence they were
easily compared to the PUFA. Now, a much more apolar
mobile phase was used in order to decrease the retention
time of oleic and palmitic acids (RT of 9 min approxi-
mately). Anyway, representation of the absorbance versus
concentration follows a complex sigmoidal trend and hence
calibration curves have to be used to calculate the exact
concentration of each fatty acid (Fig. 3) [24].
Optimization of the Reaction Conditions
The composition of the aqueous phase was also optimized
(Tables 3, 4) observing two highlightable effects; low
buffer concentration contributed to increase the reaction
rate and added methyl-cyclodextrin to the aqueous phase
had a positive effect in the hydrolytic activity. Considering
those two beneficial effects induced by both low pH and
methyl-cyclodextrin, lipase activity increased from 3.67 U
at high buffer concentrations to 20.14 U at low ones. This
activity improvement seems to be related to a better
Design of a Biphasic Reactor Water-Cosolvent
for the Hydrolysis of Fish Oil Using Lipases
Immobilized on Porous Supports
Oils are very viscous and difficult to handle (they become
adsorbed on the walls of the containers, on moderately
hydrophobic supports or on the stirring devices). Thus, in
order to make oil manipulation feasible (especially at low
Fig. 3 ‘‘Light-scattering’’
response of EPA and DHA at
different concentrations.
Response is expressed as
absorbance units observed for
the corresponding peaks in the
RP-HPLC chromatograms.
Injected samples: 20 lL of
solutions with different
30
25
20
15
10
5
30
EPA
DHA
25
20
15
10
5
concentrations of PUFA (from
0.125 to 3 mM)
0
0
0
0.5
1
1.5
2
2.5
3
0
0.5
1
1.5
2
2.5
3
[mM]
[mM]
123

824
J Am Oil Chem Soc (2011) 88:819–826
Table 4 Effect of methyl-cyclodextrin concentration on the rate of
hydrolysis of sardine oil catalyzed by immobilized RML
Table 6 Effect of the concentration of sardine oil in the organic
phase on the hydrolysis catalyzed by RML derivatives
Methyl-cyclodextrin (mM)
Activitya
EPA/DHA ratiob
Sardine oil (M)
Activity
Selectivity
0
8.91
10.53
14.63
20.14
4.95
6.34
5.43
5.31
0.11
0.32
0.54
20.14
68.67
109.8
4.95
4.47
5.1
10
25
50
A biphasic water/cyclohexane reactor at pH 6.0 and 25 °C was used
A water/cyclohexane biphasic reactor at pH 7.0 and 25 °C was used.
The presence of different concentrations of methyl-cyclodextrin in the
aqueous phase was analyzed
Activity is expressed as micromoles of PUFA (EPA plus DHA)
released per minute and per gram of immobilized enzyme. Selectivity
is expressed as the ratio between EPA and DHA
a
Activity is expressed as micromoles of PUFA (EPA plus DHA)
released per minute and per gram of immobilized enzyme
b
Selectivity is expressed as the ratio between EPA and DHA
partition of the oil in the aqueous phase from 0.15% in
100 mM Tris–Cl up to 2% in 20 mM Tris–Cl and 50 mM
methyl-cyclodextrin.
On the other hand, the rate of hydrolysis improved when
increasing the pH and curiously it also improved the
selectivity EPA/DHA (Table 5). This result might be
related to the intrinsic properties of RML. This result is
also quite interesting for showing a possible modulation of
activity-selectivity of lipases by using different experi-
mental conditions. However, pH 7 is here proposed as the
standard pH for a hydrolytic assay useful for most of
lipases because of activity-stability binomials.
Fig. 4 Effect of the amount of immobilized lipase on hydrolytic
rates. Activity is expressed as micromoles of PUFA (EPA plus DHA)
released per minute. The amount of immobilized lipase is expressed
as the weight of catalyst added to 10 mL of the biphasic reactor (from
0.1 to 1 g of biocatalyst). RML derivatives were evaluated
The effect of oil concentration in the organic phase on
the hydrolytic activity was also tested. From this study it
can be observed that concentrations of up to 50% of fish oil
enhances the lipase activity, probably because of the higher
concentration of oil in the aqueous phase, increasing the
substrate availability to be used by the lipase in that phase
(Table 6). Optimal reaction conditions were reached at
30% of fish oil, keeping low viscosity, making oils easier to
work with at the same time the lipase showing a high
hydrolytic activity. Under these suitable reaction condi-
tions the rate of reaction was proportional to the amount of
catalyst used (Fig. 4).
Hydrolysis of Sardine Oil Using Three Different
Immobilized Lipases
Using octyl-Sepharose derivatives of three different lipases
the hydrolysis of sardine oil under standard reaction con-
ditions was studied. In addition to the HPLC–UV analysis
to follow the rate of PUFA release and the selectivity EPA/
DHA, HPLC-light scattering analysis was also carried out
to determine lipase selectivity towards polyunsaturated
fatty acids and towards oleic acid (OA) (monounsaturated
fatty acid d) and palmitic acid (PA) (a saturated fatty acid).
The comparison between PUFA and OA and PA was made
because OA and PA are very abundant in many oils. In this
way, the validity of the release of PUFA for the accurate
measurement of lipolytic activity of lipases could be
checked. RML and TLL were the most selective enzymes
towards EPA with regard to DHA (EPA/DHA ratio of 4.5)
and CALB exhibited a low EPA/DHA ratio but a high
EPA ? DHA/oleic (OA) ? palmitic acids (PA) ratio
(Table 7). Anyway, the rate of the release of PUFA and the
rates of release of oleic or palmitic acids were of the same
order of magnitude for these three lipases.
Table 5 Effect of pH on the hydrolysis of sardine oil catalyzed by
immobilized RML
pH
Activity
Selectivity
5
6
7
8
7.84
20.14
21.58
28.35
2.46
4.95
5.52
8.01
A water/cyclohexane biphasic reactor at 25 °C was used. The effect
of PH in the aqueous phase was analyzed. Acetate buffer was used at
pH 5.0 and 6.0, Tris–Cl at pH 7.0 and borate at pH 8.0. In all cases,
buffer concentration was 10 mM
In this paper the interfacial activation is directly pro-
moted on the support [21] and the possible inactivation
by hydrophobic interfaces was avoided because of the
Activity is expressed as micromoles of PUFA (EPA plus DHA)
released per minute and per gram of immobilized enzyme
Selectivity is expressed as the ratio between EPA and DHA
123

J Am Oil Chem Soc (2011) 88:819–826
825
pore size (e.g., 10–15% agarose gels) and consequently a
much higher enzyme loading capacity. In this way, lipase
loadings of 100–200 mg of pure enzyme per ml of support
would be obtained. Activities of these derivatives would be
more than tenfold higher than those reported here (e.g.,
intrinsic activities around 2,300 IU of released PUFA/ml of
immobilized RML).
Table 7 Hydrolysis of sardine oil catalyzed by three commercial
lipases immobilized by adsorption on octyl-Sepharose
Lipase
Activity
Selectivity 1
Selectivity 2
RML
TLL
69
53
4.47
4.45
1.5
1.22
0.95
4.2
CALB
105
A biphasic system 1:1 cyclohexane water was used. Organic phase:
30/70 of oil/cyclohexane. Aqueous phase: 10 mM Tris–Cl buffer,
50 mM methyl-cyclodextrin at pH 7.0. Reaction temperature: 25 °C
Conclusions
Activity is expressed as micromoles of PUFA (EPA plus DHA)
released per minute and per gram of immobilized enzyme
The standard conditions for an easy assay of sardine oil
hydrolysis by lipases immobilized on porous supports
were the following: a biphasic system at 25 °C, where the
organic phase was composed of a 30/70 mixture of oil
and cyclohexane and an aqueous phase containing 50 mM
methyl-cyclodextrin in 10 mM TRIS buffer at pH 7.0.
Under these conditions any derivative of any lipase (even
immobilized from crude extracts) can be easily assayed:
conversions as low as 0.2% of hydrolysis in 24 h can be
easily detected. In addition, the initial rate of PUFA
hydrolysis (release of DHA and EPA) and the selectivity
of the hydrolysis (EPA/DHA ratio) can be easily
calculated.
Selectivity 1 is expressed as the EPA/DHA ratio
Selectivity 2 is expressed as the EPA ? DHA/OA ? PA
protection of the immobilized enzymes inside the porous
structures [27, 28]. In addition to this hyperactivation on
hydrophobic supports, other hyperactivation protocols
during and after immobilization have been reported [29].
Catalytic Activity of High Loaded Lipase Derivatives
By using the synthetic substrate (pNPB) the intrinsic
activity of the highly loaded derivatives was very high
6,330 IU/ml of immobilized CALB, 6,500 IU/ml of
immobilized RML and 15,000 IU/ml of immobilized TLL.
Intrinsic activity was measured after breaking the deriva-
tives under a very vigorous magnetic stirring for 2 h. In this
way particle sizes were reduced down to less than 1 lm
and the broken derivatives do not exhibit diffusional lim-
itations. The amount of commercial lipase preparations was
the same for the three derivatives (25 mg of protein per ml
of support). However, the purity of the commercial lipase
samples was different (qualitatively shown by SDS-
PAGE). CALB had a purity higher than 95%, TLL had a
purity of 65% and RML had a purity of 35%. In this way,
the specific activity was even much higher for immobilized
TLL.
The use of fish oil as a substrate has a great practical
interest because of the PUFA released, being the first step
in synthesizing new key functional ingredients (triglycer-
ides highly enriched with PUFA). From a more basic point
of view, hydrolysis of fish oil (e.g., from sardines) may also
be a very good test of lipolytic activities of a number of
lipases: (a) sardine oil has a high percentage of the main
types of fatty acids (PUFA, monounsaturated and satu-
rated); (b) TLL, CALB and RML (and probably a number
of other lipases) hydrolyze all kinds of fatty acids within
the same order of magnitude and thus the release of PUFA
can provide a fair semi-quantitative measurement of the
whole lipolytic activity of different enzymes; (c) PUFA can
be easily detected by UV–HPLC (at 215 nm) with a very
high sensitivity and this fact would allow a simple test of
even poorly active derivatives (e.g., new crude extracts) at
very low conversion values (e.g., 0.2%) over a long time
(e.g., 24 h), (d) at higher conversions, and using a slightly
more complex system (light scattering-HPLC) the release
of saturated and monounsaturated fatty acids could also be
easily detected.
Highly loaded lipase derivatives also exhibited interesting
activities for the release of PUFA from sardine oil: 69
(RML), 50 (TLL) and 104 (CALB) IU (micromole of
released PUFA per minute) per ml of immobilized deriva-
tive. These good results were obtained in spite of using
impure commercial samples of lipases and commercial
hydrophobic supports with very large pore size octyl-
Sepharose 4BCL. If the purity of commercial preparations
was taken into consideration, the specific activity of the
different immobilized lipases were: 11.5 IU/mg of immo-
bilized RML, 4.2 IU/mg of immobilized TLL and 5.2 IU/mg
of immobilized CALB.
Acknowledgments This work was sponsored by the Spanish Min-
istry of Science and Innovation (project AGL-2009-07526) and the
Comunidad Autonoma de Madrid (Project S0505/PPQ/03449). We
gratefully recognize the Spanish Ministry of Science and Innovation
´
for the ‘‘Ramon y Cajal’’ contract for Dr. Fernandez-Lorente. We
thank the Spanish Ministry of Science and Innovation (MICINN)
grant Consolider INGENIO 2010 CSD2007-00063 FUN-C-FOOD
and the Comunidad de Madrid (CAM) ALIBIRD-S2009/AGR-1469
for financial support.
Interestingly, lipases are small proteins so that they
could be immobilized on supports having a much smaller
123

826
J Am Oil Chem Soc (2011) 88:819–826
16. Strickley RG (2004) Solubilizing excipients in oral and injectable
formulations. Pharm Res 21:201–230
References
17. Fava F, Bertin L, Fedi S, Zannoni D (2003) Methyl-b-cyclo-
dextrin-enhanced solubilization and aerobic biodegradation of
polychlorinated biphenyls in two aged-contaminated soils. Bio-
technol Bioeng 81(4):381–390
18. Fava F, Ciccotosto VF (2002) Effects of randomly methylated-
b-cyclodextrins (RAMEB) on the bioavailability and aerobic
biodegradation of polychlorinated biphenyls in three pristine soils
1. Fernandez L, Banuelos O, Zafra A, Ronchel C, Perez-Victoria I,
Morales JC, Velasco J, Adrio JL (2008) Alteration of substrate
specificity of Galactomyces geotrichum BT107 lipase I on eico-
sapentaenoic acid-rich triglycerides. Biocatal Biotransform
26:296–305
2. Heird WC (2001) The role of polyunsaturated fatty acids in term
and preterm infants and breastfeeding mothers. Pediatr Clin
North Am 48:173–188
3. Demaison L, Moreau D (2002) Dietary n-3 polyunsaturated fatty
acids and coronary heart disease-related mortality: a possible
mechanism of action. Cell Mol Life Sci 59:463–477
4. Saremi A, Arora R (2009) The utility of omega-3 fatty acids in
cardiovascular disease. Am J Ther 16:421–436
spiked with
58(3):393–399
a transformer oil. Appl Microb Biotechnol
19. Cui Y, Wang C, Mao J, Yu Y (2010) A facile and practical
approach to randomly methylated b-cyclodextrin. J Chem Tech-
nol Biotechnol 85:248–251
´
´
20. Palomo JM, Mun˜oz G, Fernandez-Lorente G, Mateo C, Fernan-
dez-Lafuente R, Guisan JM (2002) Interfacial adsorption of
´
5. Antypa N, Van Der Does AJW, Smelt AHM, Rogers RD (2009)
Omega-3 fatty acids (fish-oil) and depression-related cognition in
healthy volunteers. J Psychopharmacol 23:831–840
lipases on very hydrophobic support (octadecyl-Sepabeads):
Immobilization, hyperactivation and stabilization of the open
form of lipases. J Mol Catal B 19:279–286
6. Bougnoux P, Hajjaji N, Maheo K, Couet C, Chevalier S (2010)
Fatty acids and breast cancer: sensitization to treatments and
prevention of metastatic re-growth. Prog Lipid Res 49:76–86
7. Montgomery P, Richardson AJ (2008) Omega-3 fatty acids for
bipolar disorder. Cochrane Database of Systematic Reviews, 2
8. Siddiqui RA, Harvey KA, Zaloga GP (2008) Modulation of
enzymatic activities by n-3 polyunsaturated fatty acids to support
cardiovascular health. J Nutr Biochem 19:417–437
´
´
21. Fernandez-Lafuente R, Armisen P, Sabuquillo P, Fernandez-
Lorente G, Guisan JM (1998) Immobilization of lipases by
´
selective adsorption on hydrophobic supports. Chem Phys Lipids
93:185–197
22. Aucoin MG, Erhardt FA, Legge RL (2004) Hyperactivation of
Rhizomucor miehei lipase by hydrophobic xerogels. Biotechnol
Bioeng 85:647–655
´
23. Gamez-Meza N, Higuera-Ciapara I, Calderon De La Barca AM,
9. GISSI-HF Investigators (2008) Effect of n-3 polyunsaturated
fatty acids in patients with chronic heart failure (the GISSI-HF
´
´
Vazquez-Moreno L, Noriega-Rodrıguez J, Angulo-Guerrero O
(1999) Seasonal variation in the fatty acid composition and
quality of sardine oil from Sardinops sagax caeruleus of the Gulf
of California. Lipids 34:639–642
trial): a randomised, double-blind, placebo-controlled trial.
Lancet 372:1223–1230
10. Riediger ND, Othman RA, Suh M, Moghadasian MH (2009) A
systemic review of the roles of n-3 fatty acids in health and
disease. J Am Diet Assoc 109:668–679
24. Rombaut R, De Clercq N, Foubert I, Dewettinck K (2009)
Triacylglycerol analysis of fats and oils by evaporative light
scattering detection. J Am Oil Chem Soc 86(1):19–25
25. Fernandez-Lorente G, Cabrera Z, Godoy C, Fernandez-Lafuente
R, Palomo JM, Guisan JM (2008) Interfacially activated lipases
against hydrophobic supports: effect of the support nature on the
biocatalytic properties. Process Biochem 43:1061–1067
26. Kanicky JR, Shah DO (2002) Effect of degree, type, position of
unsaturation on the pKa of long-chain fatty acids. J Colloid
Interface Sci 256(1):201–207
´
11. Mateo C, Palomo JM, Fuentes M, Betancor L, Grazu V, Lopez-
Gallego F, Pessela BCC, Hidalgo A, Fernandez-Lorente G,
´
´
´
Fernandez-Lafuente R, Guisan JM (2006) Glyoxyl agarose: a
fully inert and hydrophilic support for immobilization and high
stabilization of proteins. Enzyme Microb Technol 39:274–280
´
12. Bastida A, Sabuquillo P, Armisen P, Fernandez-Lafuente R,
Huguet J, Guisan JM (1998) A single step purification, immo-
´
bilization, and hyperactivation of lipases via interfacial adsorp-
tion on strongly hydrophobic supports. Biotechnol Bioeng
58:486–493
27. Bes MT, Gomez-Moreno C, Guisan JM, Fernandez-Lafuente R
(1995) Selective oxidation: stabilisation by multipoint attachment
of ferredoxin NADP ? reductase, an interesting cofactor recy-
cling enzyme. J Mol Catal A 98:161–169
´
13. Godoy CA, Fernandez-Lorente G, Felice M, De las Rivas B,
Guisan JM, Palomo JM (2010) Dramatic increase in activity of
´
´
28. Betancor L, Lopez-Gallego F, Hidalgo A, Alonso-Morales N,
immobilized derivatives of lipase from Geobacillus thermocate-
nolatus. Synergistic effect of different additives and site-directed
chemical modification. J Mol Catal B: Enzym (in press)
14. Rodrigues RC, Bolivar JM, Volpato G, Filice M, Godoy C,
Fernandez-Lafuente R, Guisan JM (2009) Improved reactivation
of immobilized-stabilized lipase from Thermomyces lanuginosus
by its coating with highly hydrophilic polymers. J Biotechnol
144:113–119
´
´
Fuentes M, Fernandez-Lafuente R, Guisan JM (2004) Prevention
of interfacial inactivation of enzymes by coating the enzyme
surface with dextran-aldehyde. J Biotechnol 110:201–207
´
29. Godoy C, de las Rivas B, Filice M, Fernandez-Lorente G, Guisan
JM, Palomo, JM (2009) Enhanced activity of an immobilized
lipase promoted by site- directed chemical modification with
polymers. Process Biochem 45:534–541
15. Loftsson T, Brewster ME (1996) Pharmaceutical applications of
cyclodextrins. 1. Drug solubilization and stabilization. J Pharm
Sci 85:1017–1025
123