Esterification of R/S-ketoprofen with 2-propanol as reactant and solvent catalyzed by Novozym 435 at selected conditions

Article abstract of DOI:10.1016/j.molcatb.2012.06.016

The enzymatic esterification of R/S-ketoprofen with 2-propanol catalyzed with the commercial biocatalyst Novozym 435 is addressed in this investigation. The low reaction rate registered in this reaction was investigated in terms of the effect of the alcohol on the physicochemical- enzymatic stability of the biocatalyst and the interaction of the substrates with the catalytic triad at a molecular level. The effect of contacting 2-propanol:H₂O mixture on Novozym 435 was investigated at 45 °C for an extended period of time (8 days). The mixture dissolves the polymethylmethacrylate (PMMA) that constitutes the support of the Candida antarctica B lipase (CALB). Additionally, the alcohol diffuses into the biocatalyst's beads remaining strongly adsorbed (the alcohol desorption is evidenced only upon heating at 187 °C) and altering the inner texture of the biocatalyst's beads. Additionally, 2-propanol modifies the secondary structure of the enzyme by decreasing the β-sheet contribution and increasing the β-turn structure. The molecular modeling of the interaction of R/S-ketoprofen and 2-propanol with the catalytic triad of the lipase provides evidences that the secondary alcohol exerts an important steric hindrance for the reaction to proceed.

Full text of DOI:10.1016/j.molcatb.2012.06.016

Journal of Molecular Catalysis B: Enzymatic 83 (2012) 108–119
Contents lists available at SciVerse ScienceDirect
Journal of Molecular Catalysis B: Enzymatic
journal homepage: www.elsevier.com/locate/molcatb
Esterification of R/S-ketoprofen with 2-propanol as reactant and solvent
catalyzed by Novozym® 435 at selected conditions
María Victoria Toledoa, Carla Joséa, Sebastián E. Collinsb, Rita D. Bonettoa, María Luján Ferreirac,
Laura E. Brianda,∗
a Centro de Investigación y Desarrollo en Ciencias Aplicadas-Dr. Jorge J. Ronco, Universidad Nacional de La Plata, CONICET, CCT La Plata, Calle 47 No 257, B1900AJK La Plata, Buenos
Aires, Argentina
b Instituto de Desarrollo Tecnológico para la Industria Química INTEC, Universidad Nacional del Litoral-CONICET, Güemes 3450, 3000 Santa Fe, Santa Fe, Argentina
c PLAPIQUI-UNS-CONICET Camino La Carrindanga, Km 7, CC 717, 8000 Bahía Blanca, Argentina
a r t i c l e i n f o
a b s t r a c t
Article history:
The enzymatic esterification of R/S-ketoprofen with 2-propanol catalyzed with the commercial biocat-
alyst Novozym® 435 is addressed in this investigation. The low reaction rate registered in this reaction
was investigated in terms of the effect of the alcohol on the physicochemical–enzymatic stability of the
biocatalyst and the interaction of the substrates with the catalytic triad at a molecular level.
The effect of contacting 2-propanol:H2O mixture on Novozym® 435 was investigated at 45 ◦C for an
extended period of time (8 days). The mixture dissolves the polymethylmethacrylate (PMMA) that con-
stitutes the support of the Candida antarctica B lipase (CALB). Additionally, the alcohol diffuses into the
biocatalyst’s beads remaining strongly adsorbed (the alcohol desorption is evidenced only upon heating
at 187 ◦C) and altering the inner texture of the biocatalyst’s beads. Additionally, 2-propanol modifies the
secondary structure of the enzyme by decreasing the ␤-sheet contribution and increasing the ␤-turn
structure. The molecular modeling of the interaction of R/S-ketoprofen and 2-propanol with the catalytic
triad of the lipase provides evidences that the secondary alcohol exerts an important steric hindrance for
the reaction to proceed.
Received 1 June 2012
Accepted 24 June 2012
Available online 21 July 2012
Keywords:
Novozym® 435
2-Propanol
R/S-ketoprofen
Enantiomer
Esterification
© 2012 Elsevier B.V. All rights reserved.
1. Introduction
while the R(−)-enantiomer can be used as a toothpaste additive
to prevent periodontal disease [4,9]. There is always a danger that
Among the racemic drugs, the 2-arylpropionic acids (the
“profen” family) are an important group of non-steroidal anti-
inflammatory pharmaceuticals (NSAIDs), widely used as racemic
mixtures to control the symptoms of arthritis and related con-
nective tissue diseases [1]. R/S-ketoprofen (2-(3-benzoylphenyl)-
propionic acid), one of the most useful NSAIDs, has received
attention since the past two decades and its anti-inflammatory
effect is approximately 160 times the anti-inflammatory potency
of aspirin on a unit weight basis [2–4]. It has analgesic, anti-
inflammatory and antipyretic properties, and is effective for
the treatment of rheumatoid arthritis, osteoarthritis and anky-
losing spondylitis [5–7]. Commercially, ketoprofen is marketed
and administrated as a racemic mixture of “R” and “S” enan-
tiomers, which are equivalent on a unit weight basis. However,
(S)-ketoprofen and (R)-ketoprofen display significantly different
pharmacologic activities and benefits [8]. S (+)-enantiomer has the
therapeutic activity of reducing inflammation and relieving pains
one enantiomer of a drug may possess the desired activity and the
other, although inactive in producing the desired activity, may pos-
sess extraneous and even harmful pharmacologic properties [7,10].
A major reason for the use of mixtures of enantiomers remains the
cost of separation of the enantiomers which exceeds the poten-
tial advantage of a possible increase in the activity. Therefore, it
has been a key issue in pharmaceutical industry to obtain the pure,
active form of ketoprofen [7].
An alternative enzymatic route in the preparation of enantiop-
ure (S)-ketoprofen using lipase as biocatalyst is gaining importance
because of their remarkable properties (e.g. stereo and substrate
specificity), milder reaction conditions reduction in the energy
requirements and the production cost [11]. Novozym® 435 pro-
duced by Novozymes Co., is the most widely tested biocatalyst in
the esterification of profens. This catalyst is composed of Candida
antarctica lipase B (CALB) physically immobilized within beads of
a macroporous resin. Among the commercially available biocata-
lysts, there is no doubt that Novozym® 435 is the most commonly
heterogeneous biocatalyst used in the enantioselective synthesis of
optically active alcohols, amines and carboxylic acids. Novozym®
435 is the commercial immobilized C. antarctica lipase B (CALB)
∗
Corresponding author. Tel.: +54 221 4 211353/210711.
E-mail address: briand@quimica.unlp.edu.ar (L.E. Briand).
1381-1177/$ – see front matter © 2012 Elsevier B.V. All rights reserved.
http://dx.doi.org/10.1016/j.molcatb.2012.06.016

M.V. Toledo et al. / Journal of Molecular Catalysis B: Enzymatic 83 (2012) 108–119
109
2.2.2. Quantification of the protein desorbed out of Novozym®
435
produced by submerged fermentation of a genetically modified
Aspergillus microorganism. The lipase is adsorbed on a macroporous
resin called Lewatit VP OC 1600, according to the information given
by the Novozymes Co. in their website. This macroporous resin is
a polymer of methacrylic acid cross-linked with divinylbenzene
(DVB) [12]. The enantioselectivity of CALB toward R(−)-enantiomer
is well known and has been proven in several reports. In the pro-
cess of enantioselective esterification, racemic ketoprofen acid will
react with an alcohol in the presence of Novozym® 435 to yield the
(R)-ketoprofen ester and water [4]. Esters are generally produced
in organic solvents, such as isooctane [13–20]. Although higher
conversion yields in organic solvent simplify product recovery, sol-
vent toxicity is a problem for many applications. In addition, some
organic solvents employed are too expensive to allow profitable
commercial scale-up [21]. The major advantages of a solvent-free
system are that the absence of solvents facilitates downstream pro-
cessing, since fewer components would be present in the reaction
mixture at the end of the reaction; moreover, the elimination of
solvents from the production step offers significant cost savings
and minimizes environmental impact. In addition, it is possible
to use high substrate concentrations [21,22]. There are few stud-
ies concerning lipase-catalyzed production of esters in solvent-free
system [23].
The amount of enzyme desorbed out of Novozym® 435 was
determined through UV–Vis spectroscopy as reported previously
[25]. In this context, the absorbance of the supernatant solution
remaining after each treatment with 2-propanol was studied in
the 200–400 nm range with an equipment Perkin Elmer Lambda35.
Then, the absorbance at 256 nm was correlated with the concentra-
tion of CALB by means of a calibration curve performed with several
solutions of known concentration of the CALB L enzyme in aque-
ous medium. In turn, the concentration of protein in the starting
solution of CALB L was obtained through a precipitation method as
described previously [25].
2.3. Procedure for the treatment of Novozym® 435 with
2-propanol–4.76% H2O
The effect of 2-propanol on the biocatalyst was investigated by
contacting 1.0000 g of the biocatalyst with 10.00 ml of a mixture of
2-propanol:4.76% (v/v) H2O at 45 ◦C and 200 rpm for 8 days. The
water activity of such mixture was equal to unity according to the
measurement performed with a thermometer and humidity meter
CEM DT-615 MIB instruments.
Then, the beads were dried in a desiccator for 8 days (to dehy-
drate) and further heated at 160 ◦C for 10 min (to desorb the
2-propanol), cooled down and weighed. This last procedure was
repeated six times until constant weight. This procedure allowed
establishing the total weight loss of the biocatalyst and the amount
of adsorbed 2-propanol.
The solvent was allowed to dry and the remaining solid was
dissolved with 1.50 ml of water, centrifuged to separate the
non-soluble substances and recover the enzyme for further quan-
tification through UV–Vis spectroscopy as described in Section
2.2.2.
In this work, the esterification of R/S-ketoprofen using 2-
propanol as reactant and solvent is investigated through several
aspects, such as the effect the alcohol on the physical integrity
and the secondary structure of the active protein of the commer-
cial biocatalyst Novozym® 435. Additionally, the interaction of the
substrates with the active site is theoretically modeled in order to
obtain insights on the reaction at a molecular level.
2. Experimental
2.1. Materials
2.4. Esterification of profens with 2-propanol
C. antarctica lipase (CALB L) (batch LCN02102) and the commer-
cial biocatalyst Novozym® 435 (batch LC200217) were obtained as
a gift from Novozymes Brasil (Paraná, Brazil). The specific surface
area (determined through the BET method) of the biocatalyst is
72 m2/g.
Highly pure C. antarctica lipase (35,500 g/mol) purchased from
Sigma–Aldrich Argentina (10.9 U/mg) was used as a reference for
the infrared analysis purposes.
The reaction of R/S-ketoprofen with 2-propanol was performed
at the optimum operative conditions previously determined for
the esterification of profens with ethanol [26]. In this context,
the reactions were performed in closed 100 ml vials, which were
kept at constant temperature (45 ◦C) and stirring (200 rpm) in a
shaker bath. In all cases 0.500 g of R/S-ketoprofen (1.966 mmol)
was dissolved in the minimum volume of 2-propanol (3.125 ml;
40.95 mmol) with no need of heating. The molar ratio 2-
propanol:R/S-ketoprofen correspond to 21.
Additionally, R/S-ketoprofen (Parafarm, 99.80%, batch 030718
000928/004), 2-propanol (J.T. Baker, 99.93%) and potassium
hydroxide 1 mol/l in ethanol (Riedel-de Haën) were used.
The reaction medium also possesses 4.76% of water added
(150 ␮l) that corresponds to the optimum amount of water for
the biocatalyst performance. The kinetics of the esterification of
R/S-ketoprofen was investigated using Novozym® 435 (160 mg)
as received and in a second set of experiments, the biocatalyst
was pretreated with 2-propanol (according to the methodology
described in Section 2.3) before the esterification of R/S-ketoprofen.
The esterification of R/S-ibuprofen with 2-propanol was also
performed at selected conditions for comparison purposes. In
this context, 0.500 g of R/S-ibuprofen (2.42 mmol) was dis-
solved in 1.00 ml (13.01 mmol); 1.30 ml (17.39 mmol) and 3.20 ml
(42.81 mmol) of 2-propanol corresponding to 2-propanol:R/S-
ibuprofen molar ratios equal to 5.4; 7.2 and 21, respectively.
Chiral analysis of both enantiomers of R/S-ketoprofen and
R/S-ibuprofen was conducted by chiral HPLC analysis using a Nucle-
odex beta-PM (Macherey-Nagel) with an UV detector operated at
230 nm. The mobile phase (metanol/0.1% TEAA pH 4.0) was oper-
ated at a flow rate of 0.7 ml/min. All samples were run in triplicate.
Enantiomeric excess (ee) referred to the form S(+)-ketoprofen was
2.2. Procedures for the quantification of proteins
2.2.1. Determination of the protein loading of Novozym® 435
The total amount of protein of the starting Novozym® 435 was
determined through the method described by Gross and coworkers
[24] and already reported previously by some of us [25]. Novozym®
435 (0.1000 g) was contacted at 37 ◦C for 30 min with 3.00 ml of
dimethylsulfoxide (J.T. Baker, 100%) at 200 rpm in a shaker bath. The
biocatalyst’s beads were separated by filtration and washed with
portions of 5.00 ml of dimethylsulfoxide. Then the recovered beads
were contacted for 30 min with 3.00 ml of 5% Triton X-100 solution
at 37 ◦C, filtered afterwards, washed three times with 5.00 ml por-
tions of the same solution and finally washed with distilled water.
This procedure allowed us the recovering of the support Lewatit VP
OC1600 and to quantify the total amount of enzyme of Novozym®
435 through the bicinchoninic acid assay described previously [25].

110
M.V. Toledo et al. / Journal of Molecular Catalysis B: Enzymatic 83 (2012) 108–119
calculated according to Eq. (1) where [S] and [R] account for the
concentrations of the S(+) and R(−) enantiomers respectively.
dimension values ranging from 2 < D < 2.5 indicate a persistent
smooth surfaces while values 2.5 < D < 3 are an indication of anti-
persistence, being a completely rough surface when D is close to 3.
For many images observed with the SEM, the variogram presents
a fractal behavior at low scale and a behavior that seems to have
an asymptotic tendency at high scale, but, if the vertical axis was
expanded variance maximums and minimums appear. This peri-
odic region was described by two parameters: dmin and dper. The
parameters dmin and D will be used in this work to characterize the
texture of the images corresponding to the different samples. The
dmin parameter corresponds to the inferior end of the periodic scale
region and is representative of the smallest cell size with enough
statistic weight to produce periods. It is worth noticing that the D
parameter determined in this work, is strictly a fractal dimension
estimator since the linear range on the graph (corresponding to
the sample fractal behavior) is not very extensive and moreover,
the linear region includes blended information: corresponding to
the own sample and to the microtome used to cut slices of the
sample.
ꢀ
ꢁ
[S] − [R]
ee =
× 100
(1)
[S] + [R]
The conversion of the profens was also verified by titration of the
final reaction mixture. Being R/S-ketoprofen or R/S-ibuprofen the
only acid compounds present in the reaction mixture, titration of
the samples with a basic solution of KOH in ethanol of known
concentration showed to be an accurate and reliable method to
determine their concentration.
2.5. Determination of the protein’s secondary structure with
diffuse reflectance Fourier infrared spectroscopy (DRIFT)
The secondary structure of the starting Novozym® 435 along
with the biocatalyst exposed to 2-propanol for 8 days and the
evolution of the secondary structure upon reaction was followed
through diffuse reflectance Fourier infrared analysis (DRIFT) cou-
pled with isotopic exchange with deuterium oxide. The isotopic
exchange of water molecules by D2O molecules allows investigat-
ing the Amide I signal (1700–1600 cm−1) without the interference
of the bending vibration of O H species that typically appears at
1640 cm−1. The H–D isotopic exchange of water molecules of all
samples was performed by contacting the sample with 100 ␮l of
deuterium oxide (D > 99%) under a vigorous stirring for 10 min and
further incubation overnight at room temperature.
DRIFT spectra were recorded using a Harrick module with pray-
ing mantis mirrors set-up (Harrick Scientific Co.). A Nicolet 8700
FTIR spectrometer with a MCT-A cryogenic detector was used to
acquire the spectra (4 cm−1 resolution, 100–250 scans). The spec-
trometer and the mirrors that direct the radiation toward the cell
are continuously purged with dry air (from a Parker Balston gener-
ator) in order to eliminate the contribution of CO2 and water vapor
from the spectra.
To estimate the secondary structure, peak fitting of the Amide
I band (1700–1600 cm−1) by Lorentzian-shaped components was
performed on the non-deconvoluted spectra. The software used
with this purpose was a special peak fitting module of Origin 5.0.
The positions and number of the components were determined
from the second derivative analysis of the spectra. The contribution
of each component to the Amide I band was evaluated by integrat-
ing the area under the curve and then normalizing to the total area
of the Amide I band.
2.7. Temperature programmed desorption
Temperature programmed desorption analysis was performed
over Novozym® 435 after being treated with 2-propanol for 8
days. Details of the equipment used in this investigation have
been published before [25,31,32]. The samples (49.9 mg) are
heated up to 400 ◦C at 10 ◦C/min under a flow of pure helium
(35 cm3(NTP) min−1) for the temperature programmed desorption
experiment. The species resulting of desorption and/or reaction
of the surface species are detected in the mass spectrometer
and recorded in a computer. The following m/e ratios were
employed to identify the desorbed species: 2-propanol C3H7OH,
m/e = 45 (100), m/e = 43 (16), m/e = 27 (13) and 29 (10); ace-
tone, m/e = 43 (100) and m/e = 58 (27); propylene, m/e = 41 (100),
m/e = 39 (74) and 42 (70); H2O, m/e = 18; CO2, m/e = 44 and CO,
m/e = 28.
2.8. Molecular modeling
A simple molecular modeling was performed using MM2 from
Chem3D 5.0 Ultra in order to explore the reasons of the enantios-
electivity to the S(+) form and the low catalytic activity. From the
Protein data Bank (PDB) the structure of CALB was obtained and
the portion of the catalytic triad and the medium and large pocket
were included. 2-Propanol and R/S ketoprofen were modeled con-
sidering their steric energetic minima and later 2-propanol was
located near the His and Ser (or not) from the beginning of the
reaction mechanism. The structures were minimized until a cutoff
of 0.1 kcal/mol was obtained between one run and the following
one.
2.6. Scanning electron microscopy analysis and fractal dimension
estimator calculation
The internal texture of Novozym® 435 before and after con-
tacting with 2-propanol for 8 days, and the biocatalyst (with
and without previous exposure to the alcohol) after 72 h of
reaction was investigated through scanning electron microscopy
analysis with an SEM Philips 505. The samples were prepared
as ultra thin specimens by embedding the biocatalyst in a LR
White resin and further sliced with a microtome. These spec-
imens, covered with a conductive gold layer in order to avoid
electrical charges on the surface, were observed in the electron
microscope.
Images of the samples at 8400× magnifications were taken
and analyzed with the FERImage program to calculate the frac-
tal dimension D and the dmin parameter by using the variogram
[27–30]. The variogram used to determine parameters that char-
acterize the surface roughness, consists of a graph of the variance
of variation of heights in a surface for different steps, as a func-
tion of such steps and at logarithmic scale. The slope of the graph
is related to the fractal dimension D as D = 3 − slope/2. The fractal
The reaction mechanism considered has several steps:
1. Adsorption of R/S ketoprofen near serine 105 in the presence or
not of 2-propanol.
2. Intermediate 1 with 2-propanol present or not.
3. Acyl enzyme of R/S ketoprofen without 2-propanol present.
4. Adsorption of 2-propanol near the acyl enzyme carbonyl group
and considering the distance to the N from His 229 in the case
of 2-propanol not present since the beginning; H distance to N
from His 224 important.
5. Intermediate 2, R/S ketoprofen coordinated with 2-propanol.
6. Adsorption of isopropyl ester of ketoprofen and regeneration of
serine.

M.V. Toledo et al. / Journal of Molecular Catalysis B: Enzymatic 83 (2012) 108–119
111
ν(C=O)
Amide I
Amide II
ν(O-H)
ads. alc.
ν(CH )
solid residue
ν(O-H)
biocatalyst/
alcohol
Novozym® 435
Lewatit
VPOC
Candida
antarctica B
4000
3500
3000
2500
2000
1500
1000
500
WAVENUMBER (cm-1)
Fig. 1. Infrared spectra of highly pure CALB, Novozym® 435 before and after being
in contact with 2-propanol for 8 days; the resin Lewatit VP OC1600 and the solid
phase recovered after drying the liquid medium in contact with the biocatalyst.
3. Results
3.1. Degradation of Novozym® 435 upon contact with the
2-propanol–water medium
Previous investigations by some of us demonstrated that
Novozym® 435 deactivates in the fourth cycle of consecutive
use (48 h per cycle that corresponds to a total of 8 days) in
the esterification of R/S-ibuprofen with ethanol as reactant and
solvent [26]. Further investigations provided evidences of the
dissolution–disaggregation of Novozym® 435 in contact with
ethanol which in turn induces the loss of the surface protein (C.
antarctica B lipase) and the cross-linked macroporous polymer
polymethylmethacrylate (PMMA) (known as Lewatit VP OC1600)
that forms the internal core of the biocatalyst’s beads [25,31]. In
this context, the biocatalyst was treated for 8 days at 45 ◦C with
a mixture of 2-propanol containing 4.76% of added water in order
to obtain insights on the effect of the alcohol on the integrity of
Novozym® 435. The methodology is similar to the one applied
before to investigate the effect of ethanol which allows to com-
pare the influence of both alcohols on the physical and chemical
stability of the biocatalyst.
Fig. 1 compares the infrared spectra of highly pure CALB,
Novozym® 435 before and after being in contact with 2-propanol
for 8 days; the resin Lewatit VP OC1600 and the solid phase
recovered after drying the liquid medium in contact with the bio-
catalyst. Additionally, Table 1 summarizes the band’s position and
the assignments of the infrared signals.
The infrared spectra of C. antarctica lipase possesses an intense
signal centered at 3337 cm−1 arising from the stretching vibration
of the intramolecular hydrogen bonded N H species that is super-
imposed with the stretching vibrations of O H species [25,33,34].
Additionally, the lipase possesses the intense bands correspond-
ing to the Amide I and Amide II signals centered at 1653 cm−1 and
1540 cm−1, respectively [33]. The infrared signal known as Amide
I arises from the stretching vibration of the carbonyl bond C O of
the backbone structure of the proteins. This vibration is not affected
by the nature of the side chain but it is influenced by the sec-
ondary structure of the proteins. The infrared signal called Amide
II is attributed to the out-of-plane combination of the in plane

112
M.V. Toledo et al. / Journal of Molecular Catalysis B: Enzymatic 83 (2012) 108–119
550
500
450
400
bending mode of the N H bond and the stretching vibration of the
18
16
14
12
10
8
C
N bond of the protein.
H2O (m/e=18)
The support Lewatit VP OC1600 possesses an intense signal at
1730 cm−1 due to the stretching vibration of the carbonyl groups of
the polymethylmethacrylate (PMMA) (CH2CCH3)n CO
O CH3
[25,31,35,36]. Additionally, the intense signals belonging to asym-
metric and symmetric stretching of the methyl groups (2951 cm−1
and 2874 cm−1, respectively) along with asymmetric and symmet-
ric bending vibrations (of weak intensity) of the methyl groups
(1387 cm−1 and 1361 cm−1, respectively) are observed [25,36,37].
These last two signals are characteristic of the C H vibration of
methyl and methylene groups attached to a tertiary carbon atom
such as the one of the PMMA resin (the expanded molecular struc-
ture of the polymer is available in Ref. [25]). The intense set of
signals observed in the range from 1270 cm−1 to 1090 cm−1 are
characteristic of twisting and wagging vibrations of methylene
groups. The presence of divinylbenzene (used as cross-linker in
PMMA) is evidenced through the doublet of medium intensity at
1511 cm−1 and 1456 cm−1 due to the skeletal vibrations involving
the carbon–carbon stretching within the aromatic ring. Addition-
ally, a set of weak signals is observed from 990 cm−1 to 708 cm−1
that is characteristic of the C H out-of-plane bending bands of the
alkyl-substituted benzene [36,37].
C3H7OH (m/e=45)
Td = 187 oC
6
CH6O (m/e=58)
3
4
CH6 (m/e=41)
3
2
0
50
100
150
200
250
300
TEMPERATURE (OC)
Fig. 2. In situ temperature programmed surface reaction spectra of Novozym® 435
after being in contact with 2-propanol for 8 days at 45 ◦C.
5.00 ml of 2-propanol and the solvent was separated by filtration
(as described before) each time. The liquids remaining after each
wash were added to the liquid phase initially separated after the
8 days treatment of Novozym® 435 and allowed to evaporate.
The solids retained in the filters and the one precipitated from
the liquid phase (after the 2-propanol was completely dried)
were weighed and then suspended in a minimum amount of
water (1.50 ml) and further centrifuged in order to separate the
non-soluble substances. Then the aqueous medium containing the
enzyme was analyzed through UV–Vis spectroscopy to quantify
the amount of desorbed protein.
The biocatalyst’s beads after being dehydrated in a desiccator for
several days were sequentially heated at 160 ◦C for 10 min, cooled
down and weighed (see Section 2.2 for details). The cycle was
repeated until constant weight which ensures that the 2-propanol
was removed from the beads. The temperature chosen for the treat-
ment corresponds to the starting temperature of the 2-propanol
desorption found in the temperature desorption analysis as will be
described in the following section.
The PMMA resin does not have O H species therefore, the broad
signal of medium intensity at ∼3465 cm−1 could be ascribed to
adsorbed water molecules.
The commercial biocatalyst Novozym® 435 possesses both the
signals belonging to CALB (a weak signal belonging to Amide I is
observed at 1654 cm−1) and the macroporous polymer as expected
(see Fig. 1 and Table 1).
The solids recovered from the organic medium show both the
CALB enzyme and the species of the macroporous matrix as can
be concluded from the similarity of the infrared signals of those
three systems (see Table 1). The nature of the solid phase recov-
ered from the 2-propanol that was in contact with Novozym®
435 evidences the disaggregation of the biocatalyst exposed to the
organic medium. In this context, the results are similar to those
reported previously in terms of the detrimental effect of ethanol
on Novozym® 435.
Additionally, the shoulders at 3289 cm−1 and 3541 cm−1
observed in the infrared spectra of the biocatalyst after being in con-
tact with the alcohol and the recovered solids might be an evidence
of 2-propanol that remains molecularly adsorbed even though the
alcohol was allowed to evaporate before the analysis of the sam-
ples. These signals have been ascribed to O H stretching vibrations
of un-dissociated 2-propanol molecules adsorbed on carbon con-
taining samples [38]. This observation was further corroborated
through the temperature programmed surface desorption of the
alcohol as will be shown in the following sections.
The results obtained in that series of experiments are presented
in Table 2 along with the information reported before (in Ref. [25])
concerning the effect of ethanol on Novozym® 435. The experi-
ments demonstrated that the detrimental effect of 2-propanol in
the integrity of the biocatalyst is much less pronounced that the
one observed with ethanol.
In fact, Novozym® 435 loses 1.2% of PMMA support, additives
and protein upon contact with 2-propanol vs. 16.6% when is in
contact with ethanol. The amount of irreversible adsorbed alco-
hol is also an order of magnitude lower (0.0368 g/g biocatalyst vs.
0.3881 g/g biocatalyst) when 2-propanol was used. In this context,
the adsorption phenomenon of this alcohol on the biocatalyst is
discussed in the following section.
3.1.1. Extent of the disaggregation of Novozym® 435 upon
contact with the 2-propanol–water medium
The results discussed above provide evidences of the
dissolution–disaggregation of Novozym® 435 in contact with
2-propanol–4.76% water which in turn induces the loss of the
surface protein (C. antarctica B lipase) and the polymeric matrix
of the biocatalyst’s beads. The amount of protein and PMMA
left in the liquid medium after contacting the biocatalyst with
2-propanol–4.76% water for 8 days was quantified in order to
establish the extent of the disaggregation of Novozym® 435. The
biocatalyst (2.0022 g) was allowed to interact with 20.00 ml of
2-propanol–4.76% H2O for 8 days as described in Section 2.3.
After that period of time, the remaining liquid was separated by
filtration with a GE nylon membrane filter (0.45 ␮m pore size)
in order to retain the non-soluble substances as polymethyl-
methacrylate. The biocatalyst’s beads were washed 3 times with
3.2. The irreversible interaction of 2-propanol over Novozym®
435
The interaction of 2-propanol–4.76% H2O with Novozym®
435 was investigated through temperature programmed
desorption–reaction analysis of the species remaining adsorbed
on the biocatalyst’s surface/bulk after being in contact with such
medium for an extended period of time as described in Section
2.3. Fig. 2 shows the desorption profile of the species with m/e = 45
that belongs to molecular 2-propanol. The alcohol molecularly

M.V. Toledo et al. / Journal of Molecular Catalysis B: Enzymatic 83 (2012) 108–119
113
Table 2
Summary of the amounts of polymethylmethacrylate, additives and enzyme lost and alcohol adsorbed upon contacting Novozym® 435 with ethanol and 2-propanol.
Comparison of the percentage of total weight loss, percentage of protein loss per total amount of protein of the biocatalyst and mass of alcohol adsorbed per gram of
Novozym® 435.
Alcohol
Mass (g)
% weight loss
% protein lossb
Mass alcohol
adsorbed per 1 g
biocatalyst
Starting
biocatalyst
Recovered
biocatalyst
Support/additives Alcohol desorbed Protein dissolved
dissolved in liquid
phase
Ethanola
2-Propanol 2.0022
1.0000
0.8340
1.9778
0.1645
0.0230
0.3881
0.0738
1.4744 × 10−3
1.3591 × 10−3
16.6
1.2
2.76
1.28
0.3881
0.0368
a
The data corresponding to Novozym® 435 in contact with etanol was taken from Ref. [26].
b
Ratio between the amount of enzyme lost due to the alcohol treatment and the total amount of enzyme of Novozym® 435.
desorbs with a maximum at 187 ◦C (ethanol desorbs at 184 ◦C
according to [25]). This fairly high temperature that corresponds
to a high activation energy of desorption is an evidence of a
strong alcohol–biocatalyst interaction. Two desorption events of
water molecules (m/e = 18) are observed. The first one observed
at low temperature (centered at 63 ◦C) corresponds to the loss
on constitutional water of the proteins. A second event happens
when the temperature rises (T > 239 ◦C) and is accompanied with
the desorption of CO and CO2, corresponds to the decomposition
of amino acids and the PMMA support [25].
2-propanol (alcohol:R/S-ibuprofen molar ratio equals to 5.4) after
48 h of reaction at 45 ◦C is presented for comparison.
The results demonstrated that the conversion of R/S-ketoprofen
toward the esters increases slowly from about 10% after 60 min of
reaction to about 19% after 72 h regardless of the previous contact
of the biocatalyst with 2-propanol or not. Although, a 12% enan-
tiomeric excess to the S(+)-ibuprofen was observed after 30 h that
value decreases rapidly. The low values of enantiomeric excess
(∼5%) clearly demonstrate that the biocatalyst is not enantiose-
lective in the esterification of the profen. This observation cannot
be attributed to the nature of the profen since the esterification of
R/S-ibuprofen shows a similar behavior. In this context, the ester-
ification of racemic ibuprofen with an alcohol:profen molar ratio
equal to 5.4 showed a 12% conversion and 1.7% of enantiomeric
excess to S(+)-ibuprofen. Even lower conversion and enantiomeric
excess were observed when the amount of alcohol was increased
providing evidences of the inhibitory effect of 2-propanol. In this
context, alcohol:profen molar ratios equal to 7.2 and 21 provided
conversions of 5.3% (ee = 0%) and 4.7% (ee = −2.1%), respectively.
The absence of acetone (m/e = 58) or propylene (m/e = 41)
demonstrates that the adsorbed alcohol does not further reacts
with the biocatalysts.
3.3. Esterification of R/S-ketoprofen and R/S-ibuprofen with
2-propanol as reactant and solvent
The esterification of R/S-ketoprofen with 2-propanol with 4.76%
of water added was performed with Novozym® 435 with and with-
out previous treatment with the alcohol–water mixture for 8 days.
The objective of such experiments was to elucidate if the detri-
mental effect of the alcohol has an impact on both the conversion
and the enantiomeric excess of the reaction. Fig. 3 shows the per-
formance of the biocatalyst in the esterification of R/S-ketoprofen.
Additionally, the conversion and enantiomeric excess in the ester-
ification of R/S-ibuprofen dissolved in the minimum amount of
3.4. Evolution of the secondary structure of the C. antarctica B
lipase upon contact with 2-propanol–water medium
Novozym® 435 was also analyzed through infrared spec-
troscopy in order to investigate the effect of the 2-propanol–water
media and the reaction conditions on the secondary structure of
the protein that remained adsorbed on the beads. The beads of bio-
catalyst were contacted with deuterium oxide overnight in order to
exchange the water molecules by D2O. The isotopic exchange pos-
sess the advantage that the much lower absorption of D2O solvent
compared with H2O in the 1700–1500 cm−1 allows a better signal-
to-noise ratio along with a higher resolution spectrum [39–41]. The
analysis of the infrared spectra of enzymes in the 1700–1600 cm−1
region (the so-called Amide I region) provides qualitative and also
quantitative information on the secondary structure elements that
compose the protein. The Amide I band originates from the stretch-
ing vibrations of the peptide carbonyl groups in the backbone of the
protein, whose frequency depends on the hydrogen-bonding and
coupling along the protein chain, and is therefore sensitive to the
protein conformation.
2-propanol: 4.76 % H2O
45 oC, 200 rpm
18
16
14
12
10
8
18
15
12
9
6
6
3
4
2
0
The percentage of ␣-helix, ␤-sheet, turns and random to
the structure of the starting Novozym® 435, treated with 2-
propanol–water media, and the evolution of the secondary
structure of the biocatalyst (with and without alcohol treat-
ment) during the esterification of R/S-ketoprofen is shown in
Tables 3 and 4. The percentage contribution of each structure was
obtained through the deconvolution, integration and further nor-
malization of the corresponding signals involved in the Amide
I. In this context, the contribution of the ␣-helix and random
structures correspond to the area of the signals at 1654 cm−1 and
1643 cm−1, respectively. The contribution of the ␤-sheet struc-
ture was obtained by adding the area of the signals appearing at
0
0
10
20
30
40
50
60
70
80
TIME (hours)
Fig. 3. Conversion and enantiomeric excess toward S(+)-ketoprofen in the ester-
ification of R/S-ketoprofen with 2-propanol. Additionally, the conversion and
enantiomeric excess in the esterification of R/S-ibuprofen dissolved in the mini-
mum amount of 2-propanol (alcohol:R/S-ibuprofen molar ratio equals to 5.4) after
48 h of reaction at 45 ◦C is presented for comparison. Symbols: (ꢀ) conversion
of R/S-ketoprofen and (ꢁ ) enantiomeric excess ee % toward S(+)-ketoprofen cat-
alyzed with starting Novozym® 435; (ꢀ) conversion of R/S-ketoprofen and (ꢂ )
enantiomeric excess ee % toward S(+)-ketoprofen catalyzed with Novozym® 435
previously treated with 2-propanol for 8 days; (᭹) conversion of R/S-ibuprofen and
(ꢁ) enantiomeric excess ee % toward S(+)-ibuprofen in 48 h of reaction.

114
M.V. Toledo et al. / Journal of Molecular Catalysis B: Enzymatic 83 (2012) 108–119
Table 3
Contribution of the ␣-helix, ␤-sheet, random structure and ␤-turns to the secondary structure of the starting Novozym® 435, after treated with 2-propanol for 8 days (at
45 ◦C without reaction) and after esterification of R/S-ketoprofen for 1 h toward 72 h using the starting biocatalyst without any previous contact with the alcohol.
Percentage contribution
␣-Helixa
␤-Sheetb
Randomc
␤-Turnsd
Starting Novozym® 435
8 days with 2-propanol
1 h of reaction
3 h
23.6
20.2
22.6
20.3
20.3
20.9
23.2
20.0
20.8
20.1
26.9
10.0
21.3
0.8
7.1
8.2
12.6
7.1
8.5
23.7
1.7
6.5
0.1
1.1
3.6
1.3
17.3
2.0
2.8
25.8
68.1
49.6
78.7
71.5
67.3
62.8
55.4
68.6
55.5
6 h
12 h
24 h
30 h
48 h
72 h
21.6
a
Corresponds to the signal at 1654 cm−1
.
b
Corresponds to the sum of the contribution of the signals at 1631 cm−1, 1637 cm−1 and 1686 cm−1
.
c
Corresponds to the signal at 1643 cm−1
.
d
Corresponds to the sum of the signals at 1664 cm−1, 1666 cm−1 and 1676 cm−1
.
1631 cm−1, 1637 cm−1 and 1686 cm−1 wavenumbers. Similarly, the
contribution of the ␤-turn structure corresponds to the addition
of the areas of the signals located at 1664 cm−1, 1666 cm−1 and
1676 cm−1 [25,33,34,39,41].
reaction was investigated with the fractal dimension estima-
tor D and the dmin parameter, by using the variogram method,
implemented on the FERImage program as described in Section
2.6. This program works with square images therefore, the cen-
tral portion of the corresponding images, squares of typically
6.12 ␮m × 6.12 ␮m, 6.32 ␮m × 6.32 ␮m and 8.69 ␮m × 8.69 ␮m
were taken. If the image is anisotropic, both parameters, D and
dmin, will not necessarily be constant in different directions,
therefore, an average value of those obtained at six different
directions, between 0◦ and 90◦, was used. Figs. 4–7 show the
micrographs and the graphs of the variance vs. step of the cross
section of the starting Novozym® 435, after treatment with 2-
propanol for 8 days, non-pretreated and used in the esterification of
R/S-ketoprofen for 72 h and pretreated and further used in the
esterification of R/S-ketoprofen for 72 h, respectively. Additionally,
Table 5 presents the fractal dimension estimator D and the mini-
mum cell size dmin of the samples described above.
The secondary structure of the protein of the starting Novozym®
435 is composed by 23.6% of ␣-helix, 26.9% of ␤-sheet, 23.7% of ran-
dom structure and 25.8% of ␤-turns. The exposure to 2-propanol
causes a drastic increase in the ␤-turn structure from 25.8% toward
68.1% and diminishes both the contribution of the ␤-sheet and
the random structure (see Tables 3 and 4). A high content of the
␤-turn structure along with a ∼22% (in average) of the ␣-helix
is maintained almost unaltered when the pretreated biocatalyst
(exposed 8 days to 2-propanol) is used in the esterification of R/S-
ketoprofen up to 72 h (see Table 4). This observation was further
confirmed when the secondary structure of the non-treated bio-
catalyst used in the esterification reaction was investigated. In this
context, Novozym® 435 (non previously exposed to 2-propanol)
used in the esterification of R/S-ketoprofen increases the content of
␤-turn from 25.8% (starting sample) toward 49.6% in the first hour
of reaction and reaches 78.7% after 3 h of reaction (see Table 3).
Again, a content of approximately 20% of ␣-helix is observed all
along the 72 h of reaction.
The D values of the starting Novozym® 435 and after treatment
with 2-propanol–4.76% H2O are similar (∼2.74) and indicate an
anti-persistent fractal behavior (D > 2.5). The esterification of R/S-
ketoprofen with 2-propanol–4.76% H2O at 45 ◦C for 72 h affects
the texture of the biocatalyst. The reaction conditions leads to a
smother texture with lower D values than the samples described
above (see Table 5). Although, the extended exposure of the biocat-
alyst to 2-propanol affects the dmin (from 0.0967 ␮m to 0.0651 ␮m)
that parameter goes back to its initial value when this biocatalyst
pretreated is used in the esterification. This observation indicates
that the alcohol diffuses inside the biocatalyst’s beads diminishing
3.5. Effect of 2-propanol on the inner texture of Novozym® 435
The inner texture of the fresh Novozym® 435, after inter-
acting with 2-propanol for 8 days, and the biocatalyst (with
and without being pre-treated with 2-propanol) after 72 h of
Table 4
Contribution of the ␣-helix, ␤-sheet, random structure and ␤-turns to the secondary structure of the starting Novozym® 435, after treated with 2-propanol for 8 days (at
45 ◦C without reaction) and after esterification of R/S-ketoprofen for 1 h toward 72 h using the pretreated biocatalyst.
Percentage contribution
␣-Helixa
␤-Sheetb
Randomc
␤-Turnsd
Starting Novozym® 435
8 days with 2-propanol
1 h of reaction
3 h
23.6
20.2
16.9
20.8
17.4
27.2
25.0
26.9
25.4
17.1
26.9
10.0
0.9
0.2
9.1
3.4
3.3
5.4
0.3
0.9
23.7
1.7
2.86
0.2
3.1
7.2
9.7
2.6
3.6
1.3
25.8
68.1
79.3
78.9
70.3
62.0
61.9
65.1
70.7
80.6
6 h
12 h
24 h
30 h
48 h
72 h
a
Corresponds to the signal at 1654 cm−1
.
b
Corresponds to the sum of the contribution of the signals at 1631 cm−1, 1637 cm−1 and 1686 cm−1
.
c
Corresponds to the signal at 1643 cm−1
.
d
Corresponds to the sum of the signals at 1664 cm−1, 1666 cm−1 and 1676 cm−1
.

M.V. Toledo et al. / Journal of Molecular Catalysis B: Enzymatic 83 (2012) 108–119
115
Table 5
Fractal dimension estimator D and minimum cell size dmin with enough statistic weight to produce periods of starting Novozym® 435, after treated with 2-propanol for 8
days (at 45 ◦C without reaction), used in the esterification of R/S-ketoprofen for 72 h as received and Novozym® 435 treated with 2-propanol and used in the esterification
of R/S-ketoprofen for 72 h.
Novozym® 435
D
dmin [␮m]
Starting sample
Treated with 2-propanol for 8 days
Used as received in the esterification of R/S-ketoprofen for 72 h
Used in the esterification of R/S-ketoprofen for 72 h after treatment with 2-propanol for 8 days
2.744 0.018
0.0967 0.0051
0.0651 0.0032
0.097 0.010
0.096 0.017
2.727 0.022
2.598 0.034
2.698 0.051
Fig. 6. Variogram corresponding to the 8.69 ␮m × 8.69 ␮m central portion of the
image taken at 8400× magnification of a slice of the cross section of Novozym® 435
used in the esterification of R/S-ketoprofen for 72 h. The image appears as an insert
in the graph.
Fig. 4. Variogram corresponding to the 8.69 ␮m × 8.69 ␮m central portion of the
image taken at 8400× magnification of a slice of the cross section of starting
Novozym® 435. The image appears as an insert in the graph.
the size of the SEM image texture pattern. However, the presence
of 2-propanol and the R/S-ketoprofen under reaction conditions
produces a pickling effect over the inner texture that conducts to a
behavior that is similar to the starting biocatalyst.
˚
distances of 2.8 A to hydrogen atoms from these amino acids.
Considering R(−)-ketoprofen, the methyl group interacts with Val
154, Gln 157 and with the phenyl moiety of the benzoyl group – an
interaction that is not present in the case of ibuprofen [26]. Table 6
shows the results of the calculation of the steric energy obtained
for the different steps with R(−) or S(+)-ketoprofen (keto). From
the table is clear that the S(+)-ketoprofen has no hindrance to pro-
duce the intermediate 1 with 2-propanol near the active catalytic
triad, whereas R(−)-ketoprofen shows a barrier of near 7.5 kcal/mol
(see Fig. 8). Later, the barrier to produce the intermediate 2 is near
24 kcal/mol for R(−)-keto vs. near 11 kcal/mol for S(+)-ketoprofen.
3.6. Results of a simple molecular mechanics study of 2-propanol
and R/S ketoprofen esterification mechanism
The ketoprofen is different from ibuprofen considering the volu-
minous substituent in the position 3 of phenyl moiety. When
S(+)-ketoprofen is adsorbed near the serine 105 of CALB, the methyl
group interacts with Thr 40, Val 154 and Ile 285, with average
Fig. 5. Variogram corresponding to the 8.69 ␮m × 8.69 ␮m central portion of the
image taken at 8400× magnification of a slice of the cross section of Novozym® 435
after contacting with 2-propanol for 8 days at 45 ◦C. The image appears as an insert
in the graph.
Fig. 7. Variogram corresponding to the 8.69 ␮m × 8.69 ␮m central portion of the
image taken at 8400× magnification of a slice of the cross section of Novozym® 435
pretreated with 2-propanol and used in the esterification of R/S-ketoprofen for 72 h.
The image appears as an insert in the graph.

116
M.V. Toledo et al. / Journal of Molecular Catalysis B: Enzymatic 83 (2012) 108–119
Table 6
Steric energy (kcal/mol) calculated for the different steps of the mechanism of interaction of R/S-ketoprofen and 2-propanol with the lipase B of Candida antarctica.
Step number
1
Mechanism
R(−)-ketoprofen
S(+)-ketoprofen
Adsorption of ketoprofen near serine 105
With 2-propanol near the ketoprofen
Intermediate 1
Intermediate 1 + 2-propanol
Acyl enzyme without 2-propanol
Adsorption of 2-propanol near acyl enzyme
Intermediate 2
−146.5
−162.3
−152.1
−150.9
−145.6
−168.1
−144.5
−142.3
−143.1
−148.5
−146.8
−152.4
−145.8
−160.5
−149.1
−148.2
2
3
4
5
6
Adsorption of ester of ketoprofen
The comparison is done point to point (changing the R(−) to the
S(+) structure) due to the known problems of the MM2 results when
bonds are broken or charges are included. Therefore, the struc-
tures are similar at each step and comparisons are done between
the conformation when R(−)-ketoprofen is the acid vs. when S(+)-
ketoprofen is the acid. However, they are very useful in qualitative
and comparative ways, especially when hydrogen bonding is con-
sidered. The steric and electronic structure of 2-propanol and its
interactions with ketoprofen and the catalytic triad are completely
different than in the case of ethanol [26].
When 2-propanol is analyzed in the interaction with the acyl-
enzyme intermediate of R/S-ketoprofen, there are at least two
R
O
O
H
C
CH3
H
C
H
S
C
C
H
H
CH3
O
O
H3C
H
H3C
C
C
O
O
CH3
CH3
C
C
Interaction with the lipase structure
Steric impediment for H transfer to histidine
R
O
O
CH3
H
C
C
H
C
S
C
C
H
CH3
H
O
O
H
H
C
O
O
CH3
H3C
CH3
C
C
H3C
Interaction with the lipase structure
No impediments for H transfer
Fig. 8. Representation of the interaction of the S(+)-ketoprofen, R(−)-ketoprofen and 2-propanol with the active catalytic triad of the lipase B of Candida antarctica.

M.V. Toledo et al. / Journal of Molecular Catalysis B: Enzymatic 83 (2012) 108–119
117
His
H
H
224
Ser105 H
H
best interactions to hydrogen
transfer to histidine 224
CH
2
H C
2
OH
H C
3
N
O
O
H
C
4.0
N
O
4.0
H
C
His
224
Ser105
C
CH
CH
2
3
H C
2
H C
3
N
O
3.6
H
O
-166 kcal/mol
H
C
4.5
C
N
O
R
H C
C
3
His
224
Ser105
CH
2
H C
3
H C
2
N
O
-162 kcal/mol
O
C
H
C
R
N 4.9
H C
3.7
O
CH
3
C
H
H
3
CH
3
OH
H
H
H
-157 kcal/mol
R
OH
H
H
H
CH
3
H
Fig. 9. Steric energies involved in the different interactions of 2-propanol with the acyl-enzyme complex.
different forms of adsorption (staggered and eclipsed) near the ser-
ine group (see Fig. 9). The form with two hydrogen atoms side
to side staggered is preferred by closed to −1.5 kcal/mol. But the
adsorption is possible of both forms. This implies that the hydro-
gen transfer from the 2-propanol to the His 224 – a key step in the
mechanism – may be sterically and electronically hindered if the
adsorption form of 2-propanol is staggered but the methyl group
is located near the His and a change of conformation in the alco-
hol is needed to produce the reaction. Here, it is proposed that
the conformational forms of the alcohol and the steric hindrance
to hydrogen transfer to Histidine decreases the reaction rate of
the profen in such a way that the reaction is controlled by the
conformational change of the 2-propanol more than by the con-
formational distribution of substituents around the chiral C of the
profen. The adsorption is however so strong that this step is very
stable in comparison to the intermediate 2, generating a high steric
barrier energy of near 24 kcal/mol for R(−)-keto vs. 11.4 kcal/mol
for S(+)-ketoprofen for the formation of intermediate 2 for this
particular alcohol. It is interesting that R(−)-ketoprofen presents
a lower steric energy in general. The intermediate 2 is particularly
stable and therefore difficult to react to produce the ester for the
R(−) and S(+) species.
Then the problem with 2-propanol is twofold: due to intrin-
sic conformational possibilities in the molecule and exchange
between them and due to the requirement of the hydro-
gen transfer to Histidine and the steric and electronic barriers
that this reaction finds with 2-propanol coordination near the
acyl enzyme. The low 12% enantiomeric excess to the S(+)-
ketoprofen form may be explained by the relative steric interaction
found for the R(−) species in the case of the produced ester
and the low activity to the higher adsorption stability of 2-
propanol near the acyl enzyme of R(−)-ketoprofen in the Ping
Pong Bi Bi mechanism for the R(−) enantiomer. 2-Propanol
adsorption is less stable in the case of the acyl enzyme
of S(−)-ketoprofen than in the case of the acyl enzyme of
R(−)-ketoprofen.

118
M.V. Toledo et al. / Journal of Molecular Catalysis B: Enzymatic 83 (2012) 108–119
3.0x10-2
The R/S enantiomeric structure of ketoprofen is not the key in
this slight preference by the S(+), but the distances of the oxy-
gen atom of the OH species to the carbon atom of carbonyl in acyl
enzyme and the hydrogen from 2-propanol to the N from Histidine
are very important.
The adsorption of 2-propanol near the acyl enzyme in R(−)-
2.5x10-2
˚
ketoprofen produces distances near 4.0 A to carbonyl C and near
˚
4.0 A to N from Histidine (see Fig. 9). The hydrogen must be directed
to the N from Histidine, the approach of the 2-propanol near to
˚
Histidine (distance N
H
3.27 A) generates a distance O C from
1.0x10-2
5.0x10-3
0.0
˚
carbonyl of 4.142 A still too long. The steric energy increases near
9 kcal/mol when an approach is forced; the activation energy is
probably very high to the hydrogen transfer due to the increase of
the repulsive interaction with methyl groups of 2-propanol with
the phenyl group of ketoprofen.
We propose that the 2-propanol increases the activation energy
of the reaction to achieve the intermediate 2 and also has repulsive
interactions that make only some of the conformational species
more able to the reaction to proceed. The S(+)-ketoprofen could be
favored only by subtle differences (less stable intermediate from
the steric point of view) in the case of the intermediate 2.
0
10
20
30
40
50
60
70
80
Time (hs)
Fig. 10. Turnover frequency of the R/S-ketoprofen and R/S-ibuprofen converted to
the R/S-ester per time and amount of remaining protein in the biocatalyst’s beads vs.
time of reaction. Symbols: (ꢁ ) TOF of the esterification of R/S-ibuprofen with ethanol
at 45 ◦C (data from Ref. [27]); (ꢀ , ꢀ) TOF of the esterification of R/S-ketoprofen with
2-propanol at 45 ◦C catalyzed with starting and treated Novozym® 435, respectively.
4. Discussion
The investigation of the secondary structure of the protein
of Novozym® 435 in contact with 2-propanol for 8 days gave
similar results than the biocatalyst used in the esterification of R/S-
ketoprofen with 2-propanol (without being in contact with the
alcohol). This observation indicates that the modification of the
secondary structure is a consequence of the interaction with 2-
propanol rather than the presence of the profen. In this context, the
modification of the protein’s secondary structure produced upon
contacting with 2-propanol–water for extended periods of time (an
increase of the ␤-turn contribution and a decrease in the ␤-sheet
content) happens also after 3 h of reaction using a fresh biocatalyst
not previously in contact with 2-propanol.
The present investigation is a follow up of previous stud-
ies reported by some of us regarding the effect of ethanol
and ethanol–water mixtures on the commercial biocatalyst
Novozym® 435. In this opportunity, the effects of a mixture of
2-propanol–4.76% water on the physical integrity of the biocata-
lyst’s beads, the secondary structure of the active protein and the
catalytic performance of Novozym® 435 are thoroughly screened.
Previous studies demonstrated that ethanol (with or without water
added) dissolves the polymethylmethacrylate that composes the
beads of Novozym® 435. In this context, 16.6% of the biocata-
lyst’s mass was lost and also the 2.76% of the total protein content
of Novozym® 435. In contrast, the 2-propanol does not have a
comparable effect with the ethanol in terms of capability of dis-
solve PMMA. The interaction of Novozym® 435 with 2-propanol
for extended periods of time produces 1.2% weight loss along
with 1.28% of protein loss. The infrared analysis of the residue
left after the treatment with 2-propanol confirmed the presence
of polymethylmethacrylate. The capability of methanol, ethanol,
1-propanol and 2-propanol (with and without water added) of dis-
solving PMMA at room temperature was reported by McEwen and
coworkers [42]. The authors demonstrated a co-solvency effect
that occurs when the components of the mixture can preferen-
tially (and separately) solvate different sites on the polymer chain.
In this context, PMMA possesses hydrophobic side chains (methyl
The results discussed up to here show similarities in terms of
the effects of both ethanol and 2-propanol on the physical integrity
of Novozym® 435, the formation of protein’s dead ends and the
intensity of the alcohol–protein interaction.
However, the catalytic performance in the esterification of R/S-
ibuprofen with ethanol and R/S-ketoprofen with 2-propanol is
significantly different. Fig. 10 compares the turnover frequency of
the R/S-ketoprofen and R/S-ibuprofen converted to the R/S-ester
per time and amount of remaining protein in the biocatalyst’s
beads. The comparison of the turnover frequencies allows a direct
comparison of the performance of the active protein as a func-
tion of the nature of the alcohol (primary or secondary alcohol)
independently of the amount of protein (active sites) and the time
of reaction [44]. The comparison of the TOFs of the esterifica-
tion of R/S-ketoprofen with 2-propanol catalyzed with Novozym®
435 fresh and exposed to 2-propanol, shows a similar behavior in
terms of catalytic performance as discussed before. However, the
TOFs values in the steady state (after 24 h) using ethanol are an
order of magnitude higher (∼4 × 10−3) than the ones in 2-propanol
(∼6 × 10−4). This observation evidences that the nature of the
alcohol that is, primary vs. secondary alcohols is a key in the biocat-
alyst’s performance. Earlier investigations reported by Arroyo and
Sinisterra indicated that the rate of esterification of 2-arilpropionic
acids catalyzed with microbial lipases is influenced by the nature
of the alcohol [45]. The authors demonstrated that the highest rate
of esterification of (R/S)-2-arilpropionic acids is reached with pri-
mary alcohols, diminishes with secondary alcohols and is almost
null with tertiary alcohols (such as tert-butyl alcohol). In this
case, the authors used isooctane as a co-solvent and the reaction
was carried at 50 ◦C with immobilized C. antarctica lipase. Further
groups) and hydrogen bond accepting groups (O
C O CH3 ester
groups) [25,43]. Schubert and coworkers demonstrated that the co-
solvency effect implies the water–hydrogen bonding to the ester
moieties of the polymer forming a hydration-shell around the car-
bonyl groups that somehow facilitates the action of the alcohol
[43]. The modification of the inner texture of the beads evidences
that the 2-propanol–water mixture diffuses into the biocatalyst’s
beads as was already demonstrated with ethanol by some of us
[25,31]. The temperature programmed desorption of 2-propanol
from Novozym® 435 provided evidences of an intense physical
adsorption on the enzyme and/or the polymeric matrix. The alco-
hol molecularly desorbs at 187 ◦C without further reaction similarly
as observed in the temperature programmed desorption of ethanol
(Td = 180 ◦C) from Novozym® 435. This observation leads to the con-
clusion that under reaction conditions (45 ◦C) the alcohol remains
adsorbed as a spectator species most probably forming dead-ends
that inhibits the esterification process [26].

M.V. Toledo et al. / Journal of Molecular Catalysis B: Enzymatic 83 (2012) 108–119
119
studies by Sinisterra and coworkers found no yield toward the ester
References
when 2-propanol was used in the esterification of racemic ibupro-
fen catalyzed with Rhizomucor miehei lipase [46]. More recently,
similar results were reported for the esterification of R/S-ibuprofen
when Candida sp. was used [47]. These investigations described
the enzyme’s behavior attributing the causes of the poor bio-
catalysts’ performance to a steric hindrance of the alcohol. The
molecular modeling studies performed in the present investiga-
tion provides for the first time in the literature a detailed screening
of the different steps of the mechanism and the steric energy
involved in each of them. These and previous results demon-
strated that the poor catalytic performance of the CALB enzyme
cannot be attributed to the profen molecule (being either ibupro-
fen or ketoprofen) but to the conformation arrangements of the
secondary alcohol when adsorbed on the catalytic triad. In this
context, a conformational arrangement of 2-propanol with both
methyl groups away from the OH species favors the transference
of the hydrogen atom to the Histidine and the subsequent forma-
tion of the propyl ester of ketoprofen. In contrast, the adsorption
of the alcohol in such a way that a methyl group is close to His-
tidine increases the activation energy to achieve the intermediate
species.
[1] J.N. Jin, S.H. Lee, S.B. Lee, J. Mol. Catal. B: Enzym. 26 (2003) 209–216.
[2] H. Buschmann, T. Christoph, E. Friderichs, C. Maul, B. Sundermann (Eds.), Anal-
gesics from Chemistry and Pharmacology to Clinical Application, Wiley VCH,
Verlag GmbH & Co. KGaA, Weinheim, 2002.
[3] Ketoprofen capsule, extended release Andrx Pharmaceuticals, Inc., 2008.
http://www.drugsdb.eu/drug.php?d=KetoprofenExtended-release&m=Andrx
Pharmaceuticals,Inc.&id=d471d6b5-2a16-41a0-9058-cbe5dfa3011f.xml.
[4] A.L. Ong, A.H. Kamaruddin, S. Bhatia, Process Biochem. 40 (2005) 3526–3535.
[5] W.W. Xi, J.H. Xu, Process Biochem. 40 (2005) 2161–2166.
[6] M.I. Diaz-Reval, R. Ventura-Martinez, M. Deciga-Campos, J.A. Terron, F. Cabre,
F.J. Lopez-Munoz, Eur. J. Pharmacol. 483 (2004) 241–248.
[7] A.L. Ong, A.H. Kamaruddin, S. Bhatia, W.S. Long, S.T. Lim, R. Kumari, Enzyme
Microb. Technol. 39 (2006) 924–929.
[8] M.H. Ossipov, T.P. Jerussi, K. Ren, H. Sun, F. Porreca, Pain 87 (2000) 193–199.
[9] A.M. Thayer, Biocatalyst, Chemical and Engineering News, Houston, 21st May
2001, pp. 27–31.
[10] R.F. Rossi, M.A. Norton, D.L. Heefner, M.A. Hudson, C.M. Zepp, US Patent
5,273,895 (1993).
[11] M.D. Romero, L. Calvo, C. Alba, A. Daneshfar, H.S. Ghaziaskar, Enzyme Microb.
Technol. 37 (2005) 42–48.
[12] D. Rotticci, Thesis, Royal Institute of Technology, Stockholm, 2000.
[13] G. Langrand, N. Rondot, C. Triantaphylides, J. Baratti, Biotechnol. Lett. 18 (1990)
581–586.
[14] F.W. Welsh, R.E. Williams, K.H. Dawson, J. Food Sci. 55 (1990) 1679–1682.
[15] M. Rizzi, P. Stylos, A. Rich, M. Reuss, Enzyme Microb. Technol. 14 (1992)
709–714.
[16] H. Razafindralambo, C. Blecker, G. Lognay, M. Marlier, P. Wathelet, M. Severin,
Biotechnol. Lett. 16 (1994) 247–250.
[17] H. Vija, A. Telling, V. Tougu, Bioorg. Med. Chem. Lett. 7 (1997) 2592–2662.
[18] L. Gubicza, A. Kabiri-Badr, E. Keoves, K. Belafi-Bako, J. Biotechnol. 84 (2000)
193–196.
5. Conclusions
[19] S.H. Krishna, B. Manohar, S. Divakar, S.G. Prapulla, N.G. Karanth, Enzyme Microb.
Technol. 26 (2000) 131–136.
[20] S.H. Krishna, S. Divakar, S.G. Prapulla, N.G. Karanth, J. Biotechnol. 87 (2001)
193–201.
[21] A.R.M. Yahya, W.A. Anderson, M.E. Moo-Young, Enzyme Microb. Technol. 23
(1998) 438–450.
[22] P. Pepin, R. Lortie, Biotechnol. Bioeng. 63 (4) (1999) 502–505.
[23] A. Güvenc¸ , N. Kapucu, Ü Mehmetog˘lu, Process Biochem. 38 (2002) 379–386.
[24] B. Chen, J. Hu, E.M. Miller, W. Xie, M. Cai, R.A. Gross, Biomacromolecules 9 (2008)
463–471.
[25] C. José, R.D. Bonetto, L.A. Gambaro, M. Del Pilar Guauque Torres, M.L. Foresti,
M.L. Ferreira, L.E. Briand, J. Mol. Catal. B: Enzym. 71 (2011) 95–107.
[26] M.L. Foresti, M. Galle, M.L. Ferreira, L.E. Briand, J. Chem. Technol. Biotechnol. 84
(2009) 1461–1473.
[27] F.D. Bianchi, R.D. Bonetto, Scanning 23 (2001) 193–197.
[28] J.L. Ladaga, R.B. Bonetto, in: P.W. Hawkes (Ed.), Characterization of Texture
in Scanning Electron Microscope Images. Advances in Imaging and Electron
Physics, vol. 120, Academic Press, 2002, pp. 136–189.
[29] A.P. Pentland, IEEE Trans. Pattern Anal. Mach. Intell. PAMI-6 (6) (1984) 661–674.
[30] R.D. Bonetto, J.L. Ladaga, Scanning 20 (1998) 457–463.
[31] C. José, L.E. Briand, React. Kinet. Mech. Catal. 99 (2010) 17–22.
[32] L.A. Gambaro, L.E. Briand, Appl. Catal. A: Gen. 264 (2) (2004) 151–159.
[33] A. Barth, Biochim. Biophys. Acta 1756 (2007) 1073–1101.
[34] M.L. Foresti, G.M. Valle, R.D. Bonetto, M.L. Ferreira, L.E. Briand, Appl. Surf. Sci.
256 (2010) 1624–1635.
[35] Y. Mei, L. Miller, W. Gao, R.A. Gross, Biomacromolecules 4 (2003) 70–74.
[36] R.M. Silverstein, G.C. Bassler, T.S. Morrill, Spectrometric Identification of
Organic Compounds, J. Wiley & Sons, 1991, pp. 102–131.
[37] Infrared spectra database of the National Institute of Standards and Technology
(http://webbook.nist.gov) of methyl acrylate C4H6O2 (CAS: 96-33-3) and m-
divinyl benzene C10H10 (CAS: 108-57-6).
The present investigation is a detailed screening of the effect
of 2-propanol both on the integrity of the commercial biocatalyst
Novozym® 435 and the esterification of a profen at a molecular
level. Moreover, the biocatalyst’s performance in the esterification
of profens (R/S-ibuprofen and R/S-ketoprofen) using both ethanol
and 2-propanol were compared in terms of the turnover frequen-
cies that is a parameter not influenced by the amount of protein and
time of reaction. The TOFs comparison evidences that the low reac-
tion rate of the CALB enzyme (low conversion and enantiomeric
excess toward de S (+)-ketoprofen) when using 2-propanol cannot
be attributed to the nature of the profen itself but to the interac-
tion of the secondary alcohol with the catalytic triad of the CALB
enzyme. The inhibition exerted by the secondary alcohol upon
interaction with the lipase B of C. antarctica is the key to under-
stand the catalytic performance of Novozym® 435. In this context,
the action of 2-propanol on the physical integrity of Novozym®
435 is considerably less intense than the effect of ethanol. In fact,
a 16.6% vs. 1.2% loss of the bead’s mass is produced when the bio-
catalyst was contacted with ethanol and 2-propanol, respectively.
Similarly, the percentage of protein loss (2.73% vs. 1.28%) was also
rather lower when using 2-propanol.
Temperature programmed desorption studies undoubtedly
demonstrated that both ethanol and 2-propanol remain strongly
adsorbed (most probably hydrogen bonded) to the biocatalyst’s
surface. The formation of dead-end complexes between the sec-
ondary alcohol and the protein (similarly to the ethanol-protein
dead-ends) is another drawback in the esterification process using
2-propanol.
[38] J. Zawadzki, M. Wis´niewski, J. Weber, O. Heintz, B. Azambre, Carbon 39 (2001)
187–192.
[39] A. Barth, C. Zscherp, Q. Rev. Biophys. 35 (2002) 369–430.
[40] A. Natalello, D. Ami, S. Brocca, M. Lotti, S.M. Doglia, J. Biochem. 385 (2005)
511–517.
[41] C.R. Llerena Suster, C. José, S.E. Collins, L.E. Briand, S.R. Morcelle, Process
Biochem. 47 (2012) 47–56.
[42] J.M.G. Cowie, M.A. Mohsin, I.J. McEwen, Polymer 28 (1987) 1569–1572.
[43] R. Hoogenboom, C.R. Becer, C. Guerrero-Sanchez, S. Hoeppener, U.S. Schubert,
Aust. J. Chem. 63 (2010) 1173–1178.
Acknowledgments
[44] M. Boudart, Chem. Rev. 95 (1995) 661–666.
The authors acknowledge Prof. Luis A. Gambaro for the TPD
analysis, Lic. Mariela Theiller for the images taken with the SEM
microscope; the financial support of CONICET (project PIP 0083)
and Universidad Nacional de La Plata (project 2012–2015).
[45] M. Arroyo, J.V. Sinisterra, J. Org. Chem. 59 (1994) 4410–4417.
[46] M.T. López-Belmonte, A.R. Alcántara, J.V. Sinisterra, J. Org. Chem. 62 (1997)
1831–1840.
[47] Y. Liu, F. Wang, T. Tan, J. Mol. Catal. B: Enzym. 56 (2009) 126–130.