Lipases from Candida antarctica: Unique biocatalysts from a unique origin

Article abstract of DOI:10.1021/op0200165

The specificity of the A-lipase from Candida antarctica (CALA) has been characterized to further clarify the scope of the biocatalyst. The lipase was found to exhibit an almost uniform activity towards various straight-chained primary alcohols and carboxylic acids, only exhibiting a low activity towards very short-chained acids. More interestingly, the enzyme was found to exhibit a high activity towards a surprising diversity of sterically hindered alcohols, including both secondary and tertiary alcohols. These results indicate that CALA can have a unique applicability for the conversion of highly branched substrates where most other lipases fail to display any activity. A new, potentially highly cost-effective, immobilization technology using silica-based granulation has been applied in the immobilization of the B-lipase from the same yeast (CALB). Highly stable particles were obtained with an activity comparable to that of the commercially available immobilized preparations of this enzyme.

Full text of DOI:10.1021/op0200165

Organic Process Research & Development 2002, 6, 446−451
Lipases from Candida antarctica: Unique Biocatalysts from a Unique Origin
Ole Kirk and Morten Wu¨rtz Christensen*
Research and DeVelopment, NoVozymes A/S, KrogshoejVej 36, DK-2880, Denmark
Table 1. Characteristics of CALA and CALB
CALA
Abstract:
The specificity of the A-lipase from Candida antarctica (CALA)
has been characterized to further clarify the scope of the
biocatalyst. The lipase was found to exhibit an almost uniform
activity towards various straight-chained primary alcohols and
carboxylic acids, only exhibiting a low activity towards very
short-chained acids. More interestingly, the enzyme was found
to exhibit a high activity towards a surprising diversity of
sterically hindered alcohols, including both secondary and
tertiary alcohols. These results indicate that CALA can have a
unique applicability for the conversion of highly branched
substrates where most other lipases fail to display any activity.
A new, potentially highly cost-effective, immobilization technol-
ogy using silica-based granulation has been applied in the
immobilization of the B-lipase from the same yeast (CALB).
Highly stable particles were obtained with an activity compa-
rable to that of the commercially available immobilized prepa-
rations of this enzyme.
CALB
molecular weight (kD)3
isoelectric point (pI)3
pH optimum4
45
7.5
7
33
6.0
7
435
15 [0]
7-10
specific activity (LU/mg)4
thermostability at 70 °C3,a
pH stability3,b
420
100 [100]
6-9
yes (but low) No
interfacial activation4
positional specificity toward triglycerides5 Sn-2
Sn-3
a Residual activity after incubation at 60 oC in 0.1 M tris buffer (pH 7.0) for
20 min and [120 min]. b pH at which more than 75% activity is retained following
incubation for 20 h at room temperature.
form and can be used in operation at elevated temperature
for thousands of hours without any significant loss in
activity.7 Activity and specificity are, of course, other critical
factors defining the usability of an enzyme catalyst. As
outlined below, perhaps the most unique features of CALA
and CALB are their substrate specificities that in many
applications have been found to be very different from most
other known lipases, thereby opening up for unique applica-
tions.
Activity and Specificity of CALA. Relatively few studies
have been published on this lipase to date. Initial investiga-
tions did not reveal any unusual specificity of the enzyme.
Even though it has a preference for the sn-2 position in
triglycerides, this selectivity is not pronounced enough to
enable selective synthesis of 1,3-diglycerides or 2-monoglyc-
erides. In practical interesterification of triglycerides, the
lipase is rather nonselective.7 Furthermore, early work on
the lipase did not reveal any unusual specificity in the
esterification of simple alcohols7 and as it, moreover, was
found to have a relatively low specific activity compared to
many other lipases, there were no indications that the enzyme
possessed any unique features apart from the extremely high
thermostability. This feature did, however, open up for some
applications in which operation in solution at very high
temperature was required. Thus, the enzyme has found use
in the paper and pulp and textile industries where it is used
to hydrolyze very high-melting fatty acid glycerides and resin
esters.8 Some more recent studies focusing on organic
synthesis have, however, indicated that the enzyme can
exhibit a highly unique specificity towards some selected
substrates. CALA has been tested among other lipases in
Introduction
The basidiomyceteous yeast Candida antarctica produces
two different lipases, named A and B.1,2 Throughout this
contribution, CALA and CALB will be used as abbreviations
for the A- and B-lipase, respectively. As the name suggests,
the strain was originally isolated in Antarctica, with the aim
of finding enzymes with extreme properties, and both CALA
and CALB do, indeed, exhibit unusual properties. As
indicated in Table 1 the two lipases are very different.3-5
The stability of any enzyme is a critical factor determining
its applicability. Both these lipases are stable over a relatively
broad pH range with CALA being most stable at acidic pH
and CALB being most stable at alkaline pH. CALA is an
extremely thermostable protein, and the enzyme has its
temperature optimum above 90 °C.6 The lipase is, in fact,
probably the most thermostable lipase described to date, a
surprising feature given its origin from a cold-adapted
organism. Even though CALB is not as stable as CALA in
solution, both lipases are highly stable in an immobilized
* To whom correspondence should be addressed. E-mail: mwc@
novozymes.com.
(1) Michiyo, I. (Novo Nordisk A/S). EP 287,634, October 16, 1987; Chem.
Abstr. 1988, 110, 20529.
(2) Hoegh, I.; Patkar, S.; Halkier, T; Hansen, M. T. Can. J. Bot. 1995, 73,
S869-S875.
(3) Patkar, S. A.; Bjorkling, F.; Zundel, M.; Schulein, M.; Svendsen, A.; Heldt-
Hansen, H. P.; Gormsen, E. Indian. J. Chem. 1993, 32B, 76-80.
(4) Martinelle, M.; Holmquist, M.; Hult, K. Biochim. Biophys. Acta 1995, 1258,
272-276.
(5) Rogalska, E.; Cudrey, C.; Ferrato, F.; Verger, R. Chirality 1993, 5, 24-30.
(6) Zamost, B. L.; Nielsen, H. K.; Starnes, R. L. J. Ind. Microbiol. 1991, 8,
71-82.
(7) Heldt-Hansen, H. P.; Ishii, M.; Patkar, S.; Hansen, T. T.; Eigtved, P. In
Biocatalysis in Agricultural Biotechnology; Whitaker, J. R., Sonnet, P. E.,
Eds.; American Chemical Society: Washington, DC, 1989; pp 158-172.
(8) Nielsen, T. B.; Ishii, M.; Kirk, O. In Cold Adapted Organisms; Margesin,
R., Schinner, S., Eds.; Landes Bioscience: Austin, Texas, 1999; pp 49-
61.
446
•
Vol. 6, No. 4, 2002 / Organic Process Research & Development
10.1021/op0200165 CCC: $22.00 © 2002 American Chemical Society
Published on Web 06/27/2002

Figure 2. Examples of substrates successfully synthesized using
CALB as catalyst.19
Figure 1. Examples of unique substrates accepted by CALA.
the search for enantioselective catalysts in several studies,
but has in most cases failed to provide any interesting results.
However, in the esterification of highly sterically hindered
substrates, CALA has in several studies been found to be
the only lipase able to catalyze the desired reactions. The
enzyme has been found to be able to accept highly branched
acyl groups as well as sterically hindered alcohols and
amines. Some examples of substrates accepted by CALA
are shown in Figure 1. Towards these substrates, CALA
exhibits a highly useful selectivity. Thus, in the preparation
of 1, 2, 3, and 4 CALA-catalyzed synthesis provided an
enantiomeric excess of 99,9,10 99,11 98,12 and 87%,13 respec-
tively, and in the preparation of 5, CALA exhibited a Z/E
selectivity of 100/0 at 94% conversion.14 CALA has in the
patent literature been claimed to be able to esterify even
tertiary alcohols such as 6.15 This is a very unique feature
as tertiary alcohols, such as t-butyl alcohol, are often used
as more polar solvent when this is required in reactions using
other lipases as catalysts as these enzymes exhibit no activity
at all towards tertiary alcohols. Finally, also in the esterifi-
cation of cis/trans-isomers of unsaturated fatty acids such
as 7, CALA has shown a unique selectivity. While other
lipases tend to favor the esterification of cis-fatty acids,
CALA has in several studies been found to have a clear
preference for the trans-isomer.16-18
tions of CALA, accordingly, have been identified, the
potential of the enzyme is, most likely, still far from being
fully explored. In this report we present a more fundamental
characterization of the selectivity of CALA that can, hope-
fully, further help identifying new applications for this unique
catalyst.
Activity and Specificity of CALB. In contrast to CALA,
CALB is a very well characterized catalyst and its highly
diversified use has recently been reviewed.19 In immobilized
form, the catalyst has been found to tolerate a great variation
in experimental conditions and it has in numerous publica-
tions been shown to be a particularly efficient enzyme
catalyzing a great number of different organic reactions
including many that have been scaled up to commercial scale.
CALB exhibits a very high degree of substrate selectivity
both with respect to regio-selectivity and enantioselectivity.
CALB has been used intensively as a region-selective
catalyst, first of all to selectively acylate different carbohy-
drates.19 For CALB, however, the most extensive area of
use is in the resolution of racemic alcohols, amines, and
acids, or the preparation of optically active compounds from
meso reactants. The resulting optically pure compounds are
highly difficult to obtain by alternative routes and can be of
great synthetic value. The diversity of products successfully
synthesized using CALB as a regio- or enantioselective
catalyst is illustrated in Figure 2.
CALB has a very broad substrate specificity,20 and the
lipase can potentially be used in a number of other synthetic
reactions, leading to the formation of less complex structures.
The primary factor limiting the present industrial use of
CALB is, most likely, the relatively high price of the
immobilized products available today, Novozym 435 (No-
vozymes A/S) and Chirazyme L-2 (Roche Molecular Bio-
chemicals). These products are primarily targeted for the
resolution of chiral intermediates and other high-priced
specialty chemicals where performance and purity of the
catalyst is more critical than the price. If a less expensive
immobilized preparation could be made available, it could,
In summary, while early studies did not indicate that
CALA possessed any unique specificity, more recent studies
have, indeed, indicated that this is not the case. First of all,
it seems that CALA may have a unique ability to accept
very bulky substrates. Even though several unique applica-
(9) Holla, E. W.; Rebenstock, H. P.; Napierski, B.; Beck, G. Synthesis 1996,
823-825.
(10) Kingery-Wood, J.; Johnson, J. S. Tetrahedron Lett. 1996, 37, 3975-3976.
(11) Csomos, P.; Kanerva, L. T.; Sundholm, O.; Berhath, G.; Fu¨lo¨p, F. Magayar
Kemiaia Folyoirat 2000, 106, 71-84.
(12) Gedey, S.; Liljeblad, A.; Lazar, L.; Fu¨lo¨p, F.; Kanerva, L. T. Tetrahedron:
Asymmetry 2001, 12, 105-110.
(13) Ayers, T. A.; Schnettler, R. A.; Marciniak, G.; Stewart, K. T.; Mishra, R.
K.; Krysan, J. D.; Bernas, B. R.; Bardwaj, P.; Fevig, T. L. Tetrahedron:
Asymmetry 1997, 8, 45-55.
(14) Henkel, B.; Kunath, A.; Schick, H. J. Prakt. Chem. 1997, 339, 434-40.
(15) Bosley, J. A.; Casey, J.; Macrae, A. R.; Mycock, G. (Unichema Chemie
B.V.). WO 95/01450, July 2, 1993; Chem. Abstr. 1995, 122, 185512.
(16) Borgdorf, R.; Warwel, S. Appl. Microbiol. Biotechnol. 1999, 51, 480-485.
(17) Warwel, S.; Borgdorf, R.; Bru¨hl, L. Biotechnol. Lett. 1999, 21, 431-436.
(18) Warwel, S.; Borgdorf, R. Biotechnol. Lett. 2000, 22, 1151-1155.
(19) Anderson, E. M.; Larsson, K. M.; Kirk, O. Biocatal. Biotransform. 1998,
16, 181-204.
(20) Kirk, O.; Bjo¨rkling, F.; Godtfredsen, S. E.; Larsen, T. O. Biocatalysis 1992,
6, 127-134.
Vol. 6, No. 4, 2002 / Organic Process Research & Development
•
447

Table 2. Specificity of CALA towards straight-chained
primary alcohols relative to 1-hexanol using dodecanoic acid
as acyl donor (100 mg of immobilized CALA, 6 mL of
reaction mixture, 0.2 M acid and 0.2 M total alcohol, room
temperature, speed of stirring 30 rpm)
most likely, open up for new applications focusing on more
commodity-type products such as aliphatic polyesters, bio-
diesel, and other simple esters used in bulk quantities.
Lipase Immobilization. Immobilization technology is a
key technology for the applied usages of biocatalysts, which
ensure that the biocatalyst can easily be removed from
reaction mixture and, preferably, be reused. Furthermore,
immobilization does also, in general, increase the thermo-
stability of the enzyme by stabilizing the tertiary structure
of the protein. Several immobilization technologies can be
used for the immobilization of lipases,21-23 for example: (1)
adsorption onto polymer based carriers, (2) covalent linkage
to carriers, for example, epoxy-based chemistry, (3) cross-
linking of enzyme protein using, for example, glutaraldehyde,
4) encapsulation (e.g., in alginate beads or siloxane matrices),
and (5) deposition onto hydrophilic inorganic material, for
example, Celite types.
competitive
specificity
substrate
methanol
ethanol
1-propanol
1-butanol
1-pentanol
1-hexanol
1-octanol
factor, R
constant, 1/R
1.23
1.03
1.00
0.92
0.97
1
0.93
0.74
0.74
0.54
0.52
0.81
0.97
1.00
1.08
1.03
1
1.08
1.35
1.35
1.87
1.91
1-decanol
1-dodecanol
1-hexadecanol
1-octadecanol
From a large-scale production standpoint several factors
become important when considering possible immobilization
technologies: (1) The procedures should preferably be robust
and re-producible; (2) the production logistics should entail
no process steps with a high-holding time to ensure a
maximum capacity; (3) enzyme stability during each process
steps has to be robust; (4) handling of, for example, cross-
linking chemicals and dust-producing materials should be
taken into account; and (5) correlation with pilot-scale
immobilization will be preferable when considering future
process developments. In this report we present a new
approach for immobilizing CALB based on silica granulation,
a technology addressing most of the above factors, and the
characterization of these products.
Table 3. Activity of CALA towards various substituted
alcohols using dodecanoic acid as acyl donor (100 mg of
immobilized CALA, 6 mL of reaction mixture, 0.2 M acid
and 0.2 M alcohol, room temperature, speed of stirring
30 rpm)
initial rate
(µmol/mg/min)
substrate
1-hexanol
benzyl alohol
2-phenyl-ethanol
2-phenyl-1-propanol
2-ethyl-1-hexanol
2-propanol
2-butanol
R,S-2-octanol
S-2-octanol
3-pentanol
0.120
0.082
0.198
0.222
0.111
0.031
0.056
0.052
0.054
0.082
0.073
0.017
0.075
0.113
Results and Discussion
2-methyl-2-propanol (tert-butyl alcohol)
2-methyl-2-butanol (tert-pentyl alcohol)
cholesterol
Substrate Specificity of CALA. For the use in the
specificity studies CALA was immobilized onto porous
polypropylene following standard procedures.24 The specific-
ity of CALA towards various straight-chain carboxylic acids
has previously been investigated. The observation previously
reported that CALA tends to prefer more long-chained
straight-chained carboxylic acids7 was confirmed using
CALA immobilized on polypropylene as no reaction was
observed with acetic and butyric acid, while the highest
reaction rates were achieved using more long-chained
carboxylic acids (results not shown). As the specificity of
CALA towards highly branched carboxylic acids and towards
various unsaturated acids, as indicated above, is well
investigated, the studies were concentrated on the activity
of the enzyme towards various alcohols.
â-sitosterol
experiments performed using 1-hexanol as reference (Table
2) and dodecanoic acid as acyl donor. The specificity
constants indicate the enzyme to exhibit an almost identical
preference for all the lower alcohols, while a lower preference
for linear primary alcohols with more than 10 carbon atoms.
Interestingly, the enzyme also seems to have a preference
for the shortest alcohol tested, methanol.
The activity of the enzyme towards various substituted
primary alcohols as well as secondary and tertiary alcohols
was then investigated. In this study initial rates were
measured in individual experiments performed, again using
dodecanoic acid as acyl donor and 1-hexanol as reference.
As indicated in Table 3, CALA exhibited a surprisingly high
activity towards all the sterically hindered alcohols tested
as initial rates comparable to the one obtained using
1-hexanol as substrate was observed in all experiments. These
results further confirm the indications in the literature that
CALA has a unique activity towards more bulky substrates
The lipase exhibited even an almost 2-fold higher activity
towards phenyl-substituted alcohols with the phenyl group
in the 2-position, while a somewhat lower activity was seen
Initially, the specificity of CALA towards primary,
straight-chained aliphatic alcohols of varying chain length
was investigated. Thus, specificity constants (1/R) were
determined for each alcohol in a series of competition
(21) Buchholz, K.; Kasche, V. Biokatalysatoren und Enzymetechnologie; VCH:
Weiheim, 1997; pp 143-190.
(22) Pedersen, S.; Christensen, M. W. In Applied Biocatalysis; Straathof, A. J.
J., Adlercreutz, P, Eds.; Harwood Academic Publishers: Amsterdam, 2000;
pp 213-226.
(23) Reetz, M. T.; Simplekamp, J.; Zonta, A. (Studiengesellschaft Kohle GmbH,
Germany). DE 94/4408152, March 11, 1994; Chem. Abstr. 1995, 124, 49503.
(24) Bosley, J. A.; Peilow, A. D. J. Am. Oil Chem. Soc. 1997, 74, 107-111.
448
•
Vol. 6, No. 4, 2002 / Organic Process Research & Development

using benzyl alcohol as substrate. The reports in the patent
literature15 that CALA is capable of esterifying tertiary
alcohol was confirmed because an excellent activity was
observed towards tert-butyl alcohol, while a somewhat lower
activity was observed using tert-pentyl alcohol as substrate.
To the best of our knowledge no other lipase has to date
been reported as being able to accept tertiary alcohols as
substrate. Finally, the poor enantioselectivities observed in
several previous studies using various simple secondary
alcohols as substrates were confirmed as almost identical
initial rates were observed in experiments using racemic
2-octanol and the pure S-enantiomer as substrates.
Esters of various phytosterols are interesting products as
they have been reported to exhibit beneficial nutritional
effects when added to fat blends such as margarine.25 As
these alcohols are highly sterically hindered, they are poor
substrates of most other lipases tested. As CALA is able to
esterify other sterically hindered alcohols, the activity of the
enzyme was tested towards the cholesterol and â-sitosterol
alcohols. As indicated in Table 3, CALA was able to esterify
both steroid alcohols with an activity comparable to most
other alcohols tested.
Figure 3. Particle size distribution of CALB granulates. (4)
NS 40013 5 kLU CALB/g. (0) NS 40013 10 kLU CALB/g. (O)
NS 40013 15 kLU CALB/g. (]) NS 40013 20 kLU CALB/g.
Data not shown for NS 40013, 41 kLU CALB/g.
Table 4. Physical characteristics of CALB granulate
compared to Novozym 435
Immobilization of CALB by Silica Granulation. Granu-
lation is an important formulation technology for both
pharmaceutical applications as well as for enzyme applica-
tions. A well-known example for the latter application is
enzyme granules for detergent powder where the enzyme is
released into the washing liquor during the washing. Im-
mobilization onto inorganic type of carriers or powder has
been reported to a great extent. Using granulation technology
to build up the carrier with lipase adsorbed on silica powder
is, on the contrary, a new development within immobilization
technology.26,27 Due to the composition of the granulates,
they are intended for micro-aqueous mixtures only. Lipase
will desorb and the particle will slowly disintegrate if taken
into aqueous systems. However, the CALB silica granules
are applicable in a direct ester synthesis reaction where water
is formed, provided that the formed water is removed from
the reaction mixture. Either vacuum, azeotrope distillation,
or prevaporization can be applied to remove water produced.
This also helps forcing the reaction to completion. Applying
the granules in packed bed reactors also minimizes the
contact time with higher water concentrations.
The physical characteristics of the pilot plant preparations
(NS 40013) were characterized with respect to particle size
distribution, surface area measurement, and bulk density. A
series of granulations were performed by varying the lipase
activity dosage (5, 10, 15, 20, and 41 kLU/g of dry matter
of total granulation mixture, respectively). The physical yield
in those experiments was in the range 68-78% of the
theoretical yield. The obtained particle size distribution,
measured by sieve analysis, is shown in Figure 3. Table 4
summarizes the physical characteristics of the silica granules.
surface area bulk density porosity
preparation
(m2/g)
(g/mL)
(mL/g)
NS 40013, 20 kLU/g
Novozym 435
48-55
80
0.58
0.42
0.72
0.34
We were able to produce silica granules with a particle size
distribution comparable to many other polymer-based en-
zyme carriers and with a medium surface area (48-55 m2/
g). However, the porosity of NS 40013 20 kLU CALB/g
(determined as bed porosity) was high (0.72) compared to
that of Novozym 435 (0.34). As for normal heterogeneous
catalysis, surface areas as well as porosity and particle size
are very important factors when assessing the catalytic
performance of, in this case, an immobilized biocatalyst. The
granules obtained were incompressible, exhibited minimum
swelling characteristics and were mechanically stable and
therefore suitable for applications in packed bed reactors as
well as stirred batch operations.
Performance of Silica-Granulated CALB in Ester
Synthesis. The catalytic activity performance of the granu-
lates were compared to Novozym 435 in an industrial
relevant application, the synthesis of iso-propyl myristate,
an oleochemical used as a skin emollient. Two different silica
preparations were tested (lipase dosage 20 and 41 kLU/g
respectively, sieve fraction 300-1000 µm) and the time
course of the reaction is shown in Figure 4. The estimated
relative initial activities were calculated to: Novozym 435
(normalized to 100%), NS 40013 batch 41 kLU CALB/g
(47%) and NS 40013 batch 20 kLU CALB/g (22%). In this
particular case, we observed approximately 100% increase
in initial activity when applying double the amount of lipase
in the granulation. However, we do not expect to find a linear
correlation between lipase dosage in the granulation and the
observed initial activity in future preparations. Since the
lipase is first adsorbed onto the silica powder and then
agglomerated, lipase protein does occur throughout the core
(25) Weststrate, J. A.; Ayesh, R.; Bauer-Planck, C.; Drewitt, P. N. Food Chem.
Toxicol. 1999, 37, 1063-1071.
(26) Pedersen, S.; Larsen, A. M.; Aasmul, P. (Novo Nordisk A/S, Denmark).
WO 95/22606, February 21, 1994,;Chem. Abstr. 1995, 124, 250119.
(27) Christensen, M. W.; Andersen, L.; Kirk, O.; Holm, H. C. Lipid Technol.
Newletters 2001, 7, 33-37.
Vol. 6, No. 4, 2002 / Organic Process Research & Development
•
449

cal methods. Precipitated hydrophilic silica powder (Sipernat
22), cellulose (Arbocel BFC-200), dextrin (Glucidex DE21),
and Accurel EP 100 were obtained from Degussa, Retten-
maier and Roquette, and Akzo, respectively. Purified CALA
and CALB concentrate was from Novozymes A/S and had
enzymatic activities of 37.1 and 36.6 kLU/mL, respectively
Immobilization of CALA. CALA was immobilized on
Accurel EP 100, a porous polypropylene matrix, as previ-
ously described.24A lipase load of 52 kLU/g was obtained.
The final, dried immobilized preparation had a water content
of 1% (w/w).
Esterification Using CALA as Catalyst. Solutions of
the alcohol and the acid (0.4 M) were prepared in n-heptane.
In the case of competition experiments, six different alcohols
(always including 1-hexanol) were dissolved in individual
concentrations of 66.7 mM each to yield a total alcohol-
concentration of 0.4 M. Then 3.0 mL of the acid solution
and 3.0 mL of the carboxylic acid solution were mixed, and
20 µL of cyclohexylbenzene was added as the internal
standard. The experiments were initiated by the addition of
100 mg of immobilized CALA, and the reaction mixtures
were stirred (30 rpm) at room temperature. Every 5 min 40
µL was withdrawn and mixed with 500 µL of n-heptane,
and the progress of the reaction was analyzed by capillary
GC analysis.
Figure 4. CALB-catalyzed synthesis of iso-propyl myristate
(solvent-free reaction: 21 mmol acid dissolved in 83 mmol
alcohol, 10% catalyst (w/w), 40 °C, speed of stirring 150 rpm).
(9) NS 40013 20 kLU CALB/g. (4) NS 40013 41 kLU CALB/g.
(b) Novozym 435.
of the particle and not only on the surface of it. The apparent
activity is therefore dependent not only on the amount of
adsorbed lipase but also on the physical access to the lipase
protein, in terms of pore size, surface area, and pore volume
in connection with mass transfer and film diffusion problems.
Conclusions
It is well established that both CALA and CALB are
highly unique lipases. Even though both enzymes have found
various uses their full potential is, most likely, far from being
fully explored. The present studies address some of the
factors that, hopefully, can help further expand their use.
Our studies demonstrate that CALA exhibits a unique activity
towards various sterically hindered alcohols even including
tertiary alcohols. These results indicate that CALA may open
up for the use of lipase catalysis in reactions that are normally
not considered compatible with the use of enzymes due to
steric hindrance. With respect to CALB, a presently much
more established catalyst within organic synthesis, we have
shown that a new, potentially highly cost-effective, im-
mobilization technology based on silica-granulation can
provide a product with an activity and stability comparable
to the commercially available immobilized preparations of
this enzyme. This new product could, potentially, help further
expand the large-scale use of this catalyst into areas where
the price of the presently available products is prohibitive.
Measurement of Competitive Factors. The competitive
factor R has been defined by for lipase-catalyzed esterifi-
cations.28 Thus, if two substrates R1OH and R2OH compete
for the enzyme, the ratio of the reaction rates for each
substrate (V1 and V2) is at a given point in time related to
the concentration of the substrates by:
[R1OH]
[R2OH]
V1
V2
) R
The competitive factors were determined using 1-hexanol
as reference based on the initial rates observed essentially
as previously described.20
Granulation of CALB. The granulation was performed
using a mechanical granulator (Lo¨dige Ploughshare Mixer
FM50, Paderborn, Germany). The operation was batch-wise.
Silica-powder (5.00 kg) and cellulose (1.27 kg) were fed into
the container manually and blended. Dextrin (1.32 kg) was
dissolved in the CALB liquid concentrate (4.40 kg) and
pumped into the granulation chamber through an atomization
nozzle, and the granulation was started by applying mechan-
ical shear forces with ploughs inside the granulator. After
15 min granulation time additional water (3.00 kg) was
atomized, and agglomeration started to occur. During the
following 10 min, agglomeration was increased, and final
granules were building. After the granulation had occurred,
the process was terminated, and the granulates were dried
in a fluidized bed system (Uni Glatt, product temperature
<40 °C), after which the granulates were dry (water content
2.5%). The final product (6.76 kg, theoretical yield: 8.28
kg) was sieved in the fraction 300-1000 µm.
Experimental Section
Analytical Methods. 1H NMR spectra were recorded on
a Varian Mercury VX 400 MHz NMR spectrometer equipped
with a 4N-ASW probe. Spectra were recorded in CDCl3 at
30 °C using the CHCl3 signal as internal standard. GC
analysis was performed on a Varian Star 3400 CX equipped
with a FID detector, helium as carrier gas and a Stabilwax-
DA capillary column (30 m, 0.025 mm ID, Resteck). A
starting temperature of 60 °C was used for 1 min and them
an increase in 50 °C/min until reaching the final temperature
of 250 °C that was kept for 3 min. In the case of very short-
chained substrates, a starting temperature of 30 °C, a gradient
of 20 °C/min and a final temperature of 200 °C were used.
Materials. All unspecified reagents were from com-
mercial sources and of analytical grade. A reference of iso-
propyl myristate was made in-house using traditional chemi-
Physical Characterization of Granulates. Surface area
measurements (BET isotherm) were determined using a
(28) Rangheard, M.-S.; Langrand, G.; Triantaphylides, C.; Baratti, J. Biochim.
Biophys. Acta 1989, 1004, 20-28.
450
•
Vol. 6, No. 4, 2002 / Organic Process Research & Development

Micromeritics Gemini nitrogen adsorption instrument. The
sieve fraction 425-710 µm was used for surface area
determination. Porosity measurement (bed porosity) was
performed by slowly immersing a known volume of the silica
granules (50.0 mL, sieve fraction 300-1000 µm) into a
known volume of n-heptane, ensuring that air bubbles can
freely diffuse from the porous core of the particles. The
change in total volume (n-heptane and granules) was
determined and divided by the volume of the silica granules.
The bed porosity includes both the internal pore volume and
the space volume between the granules packed in a fix bed.
Assay for Free Lipase Activity (LU). Tributyrin was
hydrolyzed under standard conditions at 30 °C, pH 7.0, and
the activity was determined from the alkali consumption
using a pH-stat (Radiometer, Denmark). The activity was
given as LU (lipase unit), where 1 LU corresponds to the
amount of lipase, which liberates 1 mmol butyric acid min-1;
1 kLU ) 1000 LU.
83.3 mmol) in a shaking water bath at 40 °C (150 rpm).
Molecular sieves (1 g, 4 Å) and 1.00 g immobilized CALB
(silica granulated or Novozym 435) were added to the
reaction mixture. Samples of the reaction mixture were
withdrawn after 0, 15, 30, 60, 120, 180, and 240 min reaction
time, respectively, diluted directly in CDCl3 and analyzed
by 1H NMR. The conversion of the reaction could be
determined directly from the synthesis mixture by comparing
integrals of the carboxylic acid and its iso-propyl ester,
respectively. No conversion was observed in blind test
without enzyme.
Acknowledgment
The excellent technical assistance of Lotte Andersen,
Thomas Winther Christensen, and Anders Nørbygaard is
gratefully acknowledged.
Received for review January 28, 2002.
OP0200165
Synthesis of iso-Propyl Myristate (IPM). Myristic acid
(4.75 g, 20.8 mmol) was dissolved in 2-propanol (5.01 g,
Vol. 6, No. 4, 2002 / Organic Process Research & Development
•
451