Second Generation Sol-Gel Encapsulated Lipases: Robust Heterogeneous Biocatalysts

Article abstract of DOI:10.1002/adsc.200303016

The original procedure for the encapsulation of lipases in sol-gel materials produced by the fluoride-catalyzed hydrolysis of mixtures of RSi(OCH₃)₃ and Si(OCH₃)₄ has been improved considerably. This involves higher enzyme loading, variation of the alkylsilane precursor, and the use of additives such as isopropyl alcohol, 18-crown-6, Tween 80, methyl-β-cyclodextrin and/or KCl. A dramatic increase in enzyme activity is observed. The sol-gel lipase immobilizates are also excellent catalysts in the kinetic resolution of chiral alcohols and amines, recycling without any substantial loss in enantioselectivity and a residual activity of 70% being possible even after 20 reaction cycles.

Full text of DOI:10.1002/adsc.200303016

FULL PAPERS
Second Generation Sol-Gel Encapsulated Lipases: Robust
Heterogeneous Biocatalysts
Manfred T. Reetz,* Patrick Tielmann, Wolfgang Wiesenhˆfer, Werner Kˆnen,
Albin Zonta
Max-Planck-Institut f¸r Kohlenforschung, Kaiser-Wilhelm-Platz 1, 45470 M¸lheim/Ruhr, Germany
Fax: ( ‡ 49)-208-306-2985, e-mail: reetz@mpi-muelheim.mpg.de
Received: January 21, 2003; Accepted: March 21, 2003
Abstract: The original procedure for the encapsula- The sol-gel lipase immobilizates are also excellent
tion of lipases in sol-gel materials produced by the catalysts in the kinetic resolution of chiral alcohols
fluoride-catalyzed hydrolysis of mixtures of and amines, recycling without any substantial loss in
RSi(OCH3)3 and Si(OCH3)4 has been improved enantioselectivity and a residual activity of 70% being
considerably. This involves higher enzyme loading, possible even after 20 reaction cycles.
variation of the alkylsilane precursor, and the use of
additives such as isopropyl alcohol, 18-crown-6,
¾
Tween 80 , methyl-b-cyclodextrin and/or KCl. A Keywords: asymmetric catalysis; enzyme catalysis;
dramatic increase in enzyme activity is observed. immobilization; kinetic resolution; sol-gel processes
Introduction
Sol-gel encapsulation has proven to be a particularly
easy and effective way to immobilize enzymes.[7] Fol-
Lipases (EC 3.1.1.3) are the most used enzymes in lowing isolated reports describing specific examples, it
synthetic organic chemistry, catalyzing the hydrolysis of was the seminal work of Avnir and co-workers which led
carboxylic acid esters in aqueous medium or the reverse to the generalization of this technique.[7,8] It involves the
reaction in organic solvents.[1,2] They are structurally acid-or base-catalyzed hydrolysis of tetraalkoxysilanes
[9]
characterized by a so-called lid. When hydrophobic Si(OR)4 in the presence of an enzyme. Mechanisti-
substrates interact with the lipase, the lid opens and thus cally, the silane-precursor undergoes hydrolysis and
exposes the active site (serine) in a process called cross-linking condensation with formation of an SiO2
interfacial activation.[2] The catalytic triad, composed of matrix in which the enzyme is encapsulated. This type of
aspartate, histidine and serine, then induces the forma- encapsulation works well for a number of enzymes.[7,8]
tion of a short-lived acyl-enzyme intermediate which However, when we attempted to apply the method to
reacts with water or an alcohol liberating the acid or the lipases, materials were obtained which showed disap-
ester, respectively. Other nucleophiles such as hydrogen pointingly low enzyme activities, as measured by the
peroxide or amines can also be used, which results in the rate of the model reaction involving the esterification of
formation of acid peroxides or amides, respectively. lauric acid (1) by n-octanol (2) in non-dried isooctane as
Numerous examples involving enantioselectivity have solvent.[10] Only 5 10% activity relative to the tradi-
been reported, either in the desymmetrization of tional use of the respective lipase powder was observed,
prochiral (meso-type) substrates or in the kinetic equivalent to relative rates of 0.05 to 0.1.
resolution of chiral compounds.[1,2]
Reasoning that the micro-environment in SiO2 may be
In order for lipases or other enzymes to be industrially too polar, we tested the use of mixtures of Si(OCH3)4
viable, they need to be recycled and reused.[3] This can be
accomplished by employing membrane reactors, special
solvent systems or immobilization techniques.[4] More-
over, cross-linking of lipases (CLECS)[5] and cross-
linking of lipase aggregates[6] have also been reported.
So far most efforts have focused on immobilization, of
which many approaches are known including adsorption
on appropriate supports or covalent attachment to such
materials as well as encapsulation in polymers or in sol-
gel materials.
Scheme 1.
Adv. Synth. Catal. 2003, 345, 717 728
DOI: 10.1002/adsc.200303016
¹ 2003 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim
717

Manfred T. Reetz et al.
FULL PAPERS
and alkylsilanes of the type RSi(OCH3)3 having non- loading, variation of the alkyl group in the silane
hydrolyzable alkyl groups (R).[10] We speculated that the precursor and the use of appropriate additives.
silicon oxide matrix, which is now hydrophobic, could
facilitate or simulate interfacial activation of the en-
trapped lipase. Indeed, dramatically improved relative Results and Discussion
lipase activities typically amounting to 200 800% were
observed in the model reaction, which corresponds to an
Optimization of the Sol-Gel Lipase-Immobilizates
enhancement of relative enzyme activity by a factor
ranging from 2 to 8 with respect to the traditional use of
the corresponding lipase powder (lyophilizate).[10,11] In our original work we had observed an ™alkyl effect∫,
Relative activity is defined as [n(immobilized lipase)/ i.e., enhancement of lipase activity upon using
n(commercial lipase)], where n is the initial rate of the RSi(OCH3)3 in the series methyl < ethyl < n-propyl <
reaction in each case. A pronounced increase in thermal n-butyl.[11] It appeared that enhanced hydrophobicity
stability was also observed. In most cases the optimal in the silicon oxide matrix correlates with increased
ratio of RSi(OCH3)3 to Si(OCH3)4 turned out to be enzyme activity. Higher thermal stability and activity
about 5:1, although it was not possible to present an were postulated to result from multipoint interaction
experimental protocol which is completely general. through hydrogen bonding as well as ionic and hydro-
Usually CH3Si(OCH3)3 was used as the precursor and phobic interactions (van der Waals), which can be
polyvinyl alcohol (PVA) as an additive, the latter schematized as shown in Figure 1. Hydrophobic inter-
possibly acting as a stabilizer of the lipase.[10,11] Other actions could result in a type of interfacial activation. It
research groups have employed these lipase immobili- may be that the lipase is conformationally arrested in the
zates and/or developed similar materials which are also matrix in a ™lid-opened∫ and therefore active form.
based on hydrophobic alkylsilanes.[12,13] An interesting
Considerably larger alkyl groups at silicon cause
extension of our method pertains to the use of additional problems because this slows down the rate of gelation
porous solid supports during the sol-gel process.[14] This to a point where the process becomes unpractical.
type of ™double immobilization∫ involves binding of the However, small branched alkyl groups at silicon had
lipase-containing gels in the pores of the solid support been neglected. In the present study we therefore
¾
¾
(e.g., silicates of the type SIRAN or Celite ) as gelation concentrated on a comparison of the n-butyl- and
occurs and results in higher thermal stability and activity isobutylsilanes (4d and 4e, respectively) in the sol-gel
(up to a factor of 88). It should be noted that sol-gel process, hoping to learn whether branching in the alkyl
encapsulation is crucial in both variants since conven- group of the silane has any effect on the activity of the
tional adsorption on hydrophobic silicates or on SIR- lipase. Moreover, the potential effects of additional
¾
AN alone affords poor catalysts.[10]
additives such as crown ethers (e.g., 18-crown-6), b-
Successful recycling of first generation sol-gel lipase cyclodextrin derivatives (e.g., methyl-b-cyclodextrin),
¾
immobilizates was demonstrated in a series of experi- salts (e.g., KCl) and the surfactant Tween 80 were
ments using the model reaction, 30 consecutive runs
resulting in only 12 15% decrease in activity.[11,15]
Although these early immobilizates are commercially
available,[16] they have not been utilized often in
enantioselective processes.[10a,13] Immobilization was
found to increase enantioselectivity in a few examples
of kinetic resolution, but general conclusions were not
possible. In the present study we describe a highly
improved version of sol-gel encapsulation of lipases, as
well as the use of these new immobilizates as heteroge-
neous biocatalysts in a select number of enantioselective
reactions. Optimization was achieved by higher enzyme
Figure 1. Schematic view of non-covalent interactions be-
tween the gel matrix and the lipase.
Scheme 2.
718
¹ 2003 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim
asc.wiley-vch.de
Adv. Synth. Catal. 2003, 345, 717 728

Second Generation Sol-Gel Encapsulated Lipases
FULL PAPERS
studied. Previous studies by Reinhoudt,[17] Griebe- shows the results of a series of experiments using iBTMS
now,[18] Dordick[19] and Xu,[20] respectively, had shown (4e) which demonstrate that the amount of 18-crown-6
that lipases show higher activities and occasionally is crucial to success. Enzyme activity in the model
enhanced stereoselectivities when used in the presence reaction is more than doubled when employing
of such additives. Moreover, in yet another study 0.5 mmol of this additive. When calculating the specific
Kazlauskas demonstrated that the presence of a small activity, which considers the degree of loading, an even
amount of isopropyl alcohol is beneficial.[21] Usually greater value results, amounting to a factor of 3.4.
interactions of the additive and the lipase on the basis of Interestingly, when using greater amounts of the addi-
hydrogen bonds were postulated, although in some tive, activity decreases. Similar results were obtained
cases ambiguities as to the origin of the activating effect when using BTMS (4d)/TMOS (5) as the silane pre-
remain.
cursors (Table 1). Thus, in this particular case it makes
We have performed extensive optimization,[22] but only little difference whether the isobutyl-or n-butyl-derived
the most important results are presented here. In silicate is employed, although the former seems to lead
preliminary experiments it was shown that the presence to slightly better results.
of isopropyl alcohol during the sol-gel process does
In further experiments the standard protocols were
¾
indeed exert a beneficial effect. Therefore, isopropyl extended to include the presence of 50 mg of Celite as a
alcohol and PVA were included in all experiments. porous solid support, which led to two interesting
Moreover, considerably higher loadings were strived for. observations (Table 1). In the case of the iBTMS-
The degree of enzyme loading can in principle be derived sol-gel, the ™double immobilization∫ results in
controlled by adjusting the amount of enzyme per unit of increased enzyme activity, both in the absence or
silane used in the sol-gel process, maximum loading presence of 18-crown-6. Thus, the best CaLB-immobi-
¾
being the goal because this results in the least amount of lizate makes use of 18-crown-6 and Celite , this
the heterogeneous catalyst weight-wise subsequently combination increasing the relative activity to a factor
needed in catalysis. However, this feature had not been of 30.5. Table 1 also reveals that KCl (optimally
optimized in the first generation sol-gel lipase immobi- 0.75 mmol) as an additive likewise increases enzyme
lizates. Indeed, rather low loading was used (1 5 mg of activity, although the effect is not as pronounced. In
¾
protein per gram of silicate matrix), and early attempts contrast, methyl-b-cyclodextrin or Tween 80 showed
to double the loading led to lower enzyme activities.[10b] no activating effect. Most of the experiments were
The disadvantage of having to use large amounts of carried out several times, and reproducibility turned out
heterogeneous catalysts becomes accentuated when to be remarkably good ( Æ 20% of the values for
performing the sol-gel encapsulation in the presence activity).
¾
ofa solid support such asglassbeads ofthe type SIRAN
or other porous silicates such as Celite .
The same types of experiments were then performed
with the lipases from Pseudomonas fluorescens (PfL,
¾
In the present study initial experiments were per- Amano AK), Burkholderia cepacia (BcL, Amano PS),
¾
formed with the lipase B from Candida antarctica Mucor miehei (MmL, Chirazyme L-9 /Roche), Asper-
¾
(CaLB; Chirazyme L-2 /Roche). Using either BTMS gillus niger (AnL, Amano AS), Candida rugosa (CrL,
¾
(4d) or iBTMS (4e) as the silane precursors in the Chirazyme L-3 ), Candida rugosa type VII (CrL type
presence of isopropyl alcohol and PVA, it was possible
to increase loading considerably without any loss in
activity or specific activity. These are defined as mmol
product per minute per gram of gel [mmol/min ¥ g gel]
and mmol product per minute per gram of protein [mmol/
min ¥ g protein], respectively. In fact, intrinsic lipase
activity as measured by the specific activity turned out to
be higher at such high loadings, the latter typically
ranging between 50 and 600 mg of protein per gram of
silicate matrix. The reason for this positive effect is
currently unclear. Each additional additive was then
studied separately at high loading and in the presence of
isopropyl alcohol and PVA. Optimization was restricted
to the variation of the amount of additional additive. In
typical experiments the possible influence of 18-crown-6
was first investigated. Accordingly, in a standard sol-gel
reaction 3 mmol of silane (4 ‡ 5 in a 5:1 ratio), isopropyl
alcohol, PVA, buffer (0.1 M TRIS/HCl; pH 7.5) and
different amounts of 18-crown-6 were shaken in the
presence of CaLB and a basic catalyst (NaF). Figure 2 additive (see text).
Figure 2. Influence of different amounts of 18-crown-6 as an
Adv. Synth. Catal. 2003, 345, 717 728
asc.wiley-vch.de
¹ 2003 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim
719

Manfred T. Reetz et al.
FULL PAPERS
Table 1. Activity of sol-gel encapsulated CaLB in the model reaction involving 1/2 compared to the use of non-immobilized
lipase powder (activity ˆ 244 mmol/min ¥ g; specific activity ˆ 458 mmol/min ¥ g protein). In all cases a 4/5 ratio of 5: 1 (total
3 mmol) was used in the presence of PVA, isopropyl alcohol and CaLB (125 mg of lyophilizate).
Entry
Silane
Additives
Activity
Specific Activity
Relative Activity
[mmol/min ¥ g gel]
[mmol/min ¥ g protein]
1
2
3
4
5
6
7
8
9
4d
4d
4d
4d
4d
4d
4e
4e
4e
4e
4e
4e
none
767
711
1088
1080
751
2584
1684
3912
3313
1975
2403
1259
2631
4252
13997
2927
2807
5.6
3.7
8.5
7.2
4.3
5.2
2.7
5.7
9.3
30.5
6.4
6.1
¾
[a]
Celite
18-crown-6[b]
¾
Celite [a]/18-crown-6[b]
KCl[c]
¾
Celite [a]/KCl[c]
490
843
none
¾
[a]
Celite
1076
1871
2729
1639
1288
18-crown-6[b]
¾
10
11
12
Celite [a]/18-crown-6[b]
KCl[c]
¾
Celite [a]/KCl[c]
[a]
[b]
[c]
50 mg.
0.5 mmol.
0.75 mmol.
VII, Sigma), Thermomyces lanuginosa (TIL, Novo Enantioselective Reactions
Nordisk), lipase from pig pancreas (PpL, Chirazyme
¾
L-7 ) and Penicillium roqueforti (PrL, Fluka). The It cannot be safely assumed that the result of an
results are summarized in Table 2. Although all of the optimization procedure based on the model reaction
above additives were tested in every case, positive involving the esterification of lauric acid (1) by n-
(activating) effects were not always found (which are octanol (2) with formation of the ester 3 also represents
not listed). In two extreme cases none of the additives the best catalyst for the lipase-catalyzed transformation
resulted in increased lipase activity (namely MmL and of other substrates. Thus, in a real industrial application
CrL type VII), although the advantage of high loading involving a given substrate of interest, the first step
was in fact achieved here (Table 2, entries 33 36 and should make use of the sol-gel catalysts already avail-
45 48) as in all other cases.
able, but in a second step it may be useful to perform the
Some of the results deserve additional comment, type of optimization described above. In the present
especially in view of the fact that a general protocol for case such an additional search was not performed.
all lipases cannot be expected. The difference between Rather, we simply tested some of the above optimized
BTMS (4d) and iBTMS (4e) is often small (e.g., entries second generation sol-gel lipase immobilizates in the
33/35), but in some cases substantial (e.g., entries 17/25). kinetic resolution of six different chiral substrates.
With respect to additives, the use of 18-crown-6 or
From an industrial point of view the quality of a given
¾
Tween 80 leads most often to positive effects, whereas kinetic resolution not only depends upon the degree of
KCl or methyl-b-cyclodextrin are only useful in isolated enantioselectivity (which is the primary focus in aca-
cases. At this stage we prefer not to attempt to explain demic publications), but also on the activity and the
the various trends on a molecular basis. Empirically, in possibility of recycling and reusing the lipase (which are
going from the first generation sol-gel lipase immobili- usually neglected in such investigations). We therefore
zates to the second generation variants, the specific and studied all of these factors in a test reaction involving the
relative activities increase by an additional factor acylating kinetic resolution of rac-2-octanol (6), al-
ranging between 2 and 8. The numbers are even more though complete optimization was not strived for. All
revealing when comparing the new gels with the tradi- reactions were carried out on a small scale (0.5 mmol)
tional use of non-immobilized lyophilizates (Table 2), and stereoselectivity was ascertained by measuring the
¾
the most dramatic case pertaining to Tween 80 - selectivity factor E on the basis of the formula of Sih
¾
activated TIL in a BTMS-derived gel on Celite which et al.[23] All kinetic resolutions were carried out in non-
results in a relative activity factor of 1391 (entry 64). dried isooctane using vinyl acetate (7) as the acylating
These observations set the stage for potentially broad agent and CaLB, BcL and PfL as the lipases encapsu-
application in stereoselective lipase-catalyzed reactions, lated in gels derived from iBTMS/TMOS (5:1).
especially with the prospect of multiple recycling.
Table 3 reveals some remarkable trends. Of the three
lipases tested, CaLB is by far the most active and
enantioselective. The commercial CaLB powder leads
to a respectable E-value of 70 and a specific activity of
720
¹ 2003 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim
asc.wiley-vch.de
Adv. Synth. Catal. 2003, 345, 717 728

Second Generation Sol-Gel Encapsulated Lipases
FULL PAPERS
Table 2. Activity of sol-gel encapsulated lipases in the model reaction involving 1/2 compared to the use of the respective non-
immobilized lipase powder. Reaction scale and conditions are the same as in Table 1. Amounts of lipase lyophilizates used in
the sol-gel process at 3 mmol scale: PfL (150 mg; spec. activity ˆ 91 mmol/min ¥ g protein); BcL (150 mg; spec. activity ˆ
193 mmol/min ¥ g protein); MmL (150 mg; spec. activity ˆ 236 mmol/min ¥ g protein); AnL (150 mg; spec. activity ˆ 12 mmol/
min ¥ g protein); CrL type VII (60 mg; spec. activity ˆ 134 mmol/min ¥ g protein); CrL (150 mg; spec. activity ˆ 86 mmol/min ¥ g
protein); TIL (70 mg; spec. activity ˆ 6 mmol/min ¥ g protein); PpL (150 mg; spec. activity ˆ 40 mmol/min ¥ g protein); PrL
(150 mg; spec. activity ˆ 24 mmol/min ¥ g protein).
Entry
Lipase
Silane
Additives
Activity
[mmol/min ¥ g gel]
Specific Activity
[mmol/min ¥ g protein]
Relative
Activity
1
2
3
4
5
6
7
8
9
PfL
PfL
PfL
PfL
PfL
PfL
PfL
PfL
PfL
PfL
PfL
PfL
PfL
PfL
PfL
PfL
BcL
BcL
BcL
BcL
BcL
BcL
BcL
BcL
BcL
BcL
BcL
BcL
BcL
BcL
BcL
BcL
4d
4d
4d
4d
4d
4d
4d
4d
4e
4e
4e
4e
4e
4e
4e
4e
4d
4d
4d
4d
4d
4d
4d
4d
4e
4e
4e
4e
4e
4e
4e
4e
4d
4d
4e
4e
4d
4d
4d
4d
4e
4e
4e
4e
4d
4d
4e
4e
4d
4d
4d
4d
none
107
123
119
97
127
103
192
105
107
60
210
36
207
83
162
46
442
396
371
309
438
379
358
348
104
147
277
167
200
57
203
201
3224
1015
3953
655
34
657
1319
1504
1546
883
974
1017
846
412
621
3275
355
1071
7.2
14.4
16.5
16.9
9.7
10.7
11.1
9.3
4.5
6.8
35.8
3.9
11.7
¾
[a]
Celite
18-crown-6[b]
¾
Celite [a]/18-crown-6[b]
methyl-b-CD[c]
¾
Celite [a]/methyl-b-CD[c]
¾
[d]
Tween 80
¾
¾
Celite [a]/Tween 80
[d]
none
¾
[a]
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
Celite
18-crown-6[b]
¾
Celite [a]/18-crown-6[b]
methyl-b-CD[c]
¾
Celite [a]/methyl-b-CD[c]
739
918
348
8.1
10.0
3.8
¾
[d]
Tween 80
¾
¾
Celite [a]/Tween 80
[d]
none
10771
19775
9280
15435
24333
27100
9421
24864
2088
3984
5896
6958
6663
2850
7526
11161
7394
3679
6816
2039
129
116
114
184
50
42
76
308
558
840
584
561
2122
226
55.7
102.0
48.0
79.8
126.0
140.0
48.7
129.0
10.8
20.6
30.5
36.0
34.5
14.7
38.9
57.7
31.3
15.6
28.9
8.7
11.1
10.0
9.8
15.9
4.3
3.6
6.6
8.4
4.2
¾
[a]
Celite
18-crown-6[e]
¾
Celite [a]/18-crown-6[e]
methyl-b-CD[f]
¾
Celite [a]/methyl-b-CD[f]
¾
[g]
Tween 80
¾
¾
Celite [a]/Tween 80
[g]
none
¾
[a]
Celite
18-crown-6[e]
¾
Celite [a]/18-crown-6[e]
methyl-b-CD[f]
¾
Celite [a]/methyl-b-CD[f]
¾
[g]
Tween 80
¾
¾
Celite [a]/Tween 80
[g]
MmL
MmL
MmL
MmL
AnL
AnL
AnL
AnL
AnL
none
Celite
none
Celite
none
Celite
¾
[a]
¾
[a]
¾
[a]
24
34
41
22
13
28
30
44
44
49
29
297
58
¾
[g]
Tween 80
¾
¾
¾
Celite [a]/Tween 80
[g]
[g]
none
Celite
¾
[a]
AnL
AnL
AnL
¾
[g]
Tween 80
¾
Celite [a]/Tween 80
CrL (type VII)
CrL (type VII)
CrL (type VII)
CrL (type VII)
CrL (L-3)
CrL (L-3)
CrL (L-3)
CrL (L-3)
none
Celite
none
Celite
none
¾
[a]
6.3
4.4
4.2
24.6
2.6
¾
[a]
¾
[a]
Celite
¾
[h]
Tween 80
245
192
1664
1861
19.3
21.6
¾
¾
Celite [a]/Tween 80
[h]
Adv. Synth. Catal. 2003, 345, 717 728
asc.wiley-vch.de
¹ 2003 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim
721

Manfred T. Reetz et al.
FULL PAPERS
Table 2 (cont.)
Entry
Lipase
Silane
Additives
none
Activity
[mmol/min ¥ g gel]
Specific Activity
[mmol/min ¥ g protein]
Relative
Activity
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
CrL (L-3)
CrL (L-3)
CrL (L-3)
CrL (L-3)
TIL
TIL
TIL
TIL
TIL
TIL
TIL
TIL
TIL
TIL
TIL
TIL
TIL
TIL
TIL
TIL
PpL
PpL
PpL
PpL
PpL
PpL
PpL
PpL
PrL
PrL
PrL
PrL
PrL
4e
4e
4e
4e
4d
4d
4d
4d
4d
4d
4d
4d
4e
4e
4e
4e
4e
4e
4e
4e
4d
4d
4d
4d
4e
4e
4e
4e
4d
4d
4d
4d
4e
4e
4e
4e
256
39
625
40
1028
134
2225
232
3947
5909
2968
4862
2485
4816
6194
8899
1007
2290
2108
3856
1901
2801
2762
4705
302
407
134
223
67
11.9
1.6
25.8
2.7
¾
[a]
Celite
¾
[h]
Tween 80
¾
¾
Celite [a]/Tween 80
none
[h]
710
857
848
977
646
848
1208
1237
405
577
742
899
817
947
644
823
126
125
55
63
40
47
142
35
15
10
8
617
923
465
760
388
753
968
1391
157
358
329
603
297
438
432
735
7.5
10.1
3.3
5.5
1.7
3.0
6.4
2.2
3.2
3.1
¾
[a]
Celite
18-crown-6[i]
¾
Celite [a]/18-crown-6[i]
KCl[e]
¾
Celite [a]/KCl[e]
¾
[j]
Tween 80
¾
¾
Celite [a]/Tween 80
[j]
none
¾
[a]
Celite
18-crown-6[i]
¾
Celite [a]/18-crown-6[i]
KCl[e]
¾
Celite [a]/KCl[e]
¾
[j]
Tween 80
¾
¾
¾
¾
Celite [a]/Tween 80
[j]
none
Celite
¾
[a]
¾
[k]
Tween 80
¾
Celite [a]/Tween 80
[k]
[k]
none
Celite
¾
[a]
122
257
88
78
74
50
94
57
47
¾
[k]
Tween 80
¾
Celite [a]/Tween 80
none
¾
[a]
Celite
18-crown-6[l]
2.1
3.9
2.4
2.0
26.7
2.3
¾
Celite [a]/18-crown-6[l]
none
10
16
8
74
¾
[a]
PrL
PrL
PrL
Celite
18-crown-6[l]
640
56
¾
Celite [a]/18-crown-6[l]
9
[a]
[b]
[c]
[d]
[e]
[f]
[g]
[h]
[i]
[j]
50 mg.
1 mmol.
140 mg.
30 mg.
2 mmol.
40 mg.
60 mg.
180 mg.
0.5 mmol.
120 mg.
[k]
[l]
90 mg. 1.5 mmol.
188 mmol minÀ1 gÀ1 protein, but upon immobilization at entrapment causes a significant enhancement in enan-
high loading in the absence of an additional additive tioselectivity, while also increasing the activity by a
these increase to E > 100[24] and 790 mmol minÀ1 gÀ1 factor of 4.2. The additive 18-crown-6 has only a small
protein, respectively (Table 3, entry 1/2). Thus, lipase- effect (entry 3). In the case of the other two lipases (BcL
and PfL) the substrate shows poor enantioselectivity
irrespective of immobilization, although rate enhance-
ment is observed.
The CaLB-immobilizate was then subjected to a series
of recycling experiments. The gel was separated, washed
and reused, the total number of reactions amounting to
20. The selectivity factor remained constant at E > 100
over all cycles. Moreover, at least 70% of enzyme
activity is retained after 20 reactions, whereas only 40%
is present when using the commercial CaLB powder
(Figure 3).
Scheme 3.
722
¹ 2003 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim
asc.wiley-vch.de
Adv. Synth. Catal. 2003, 345, 717 728

Second Generation Sol-Gel Encapsulated Lipases
FULL PAPERS
Table 3. Kinetic resolution of alcohol 6 using free and sol-gel encapsulated lipases (Table 1) as catalysts, vinyl acetate as the
acylating agent and isooctane as solvent.
Entry
Lipase
Additives used
in Sol-Gel Process
Activity
[mmol/min ¥ g gel]
Specific Activity
[mmol/min ¥ g protein]
Relative
Activity
E
1
2
3
4
5
6
7
8
9
CaLB (powder)
CaLB (sol-gel)
CaLB (sol-gel)
BcL (powder)
BcL (sol-gel)
BcL (sol-gel)
BcL (sol-gel)
PfL (powder)
PfL (sol-gel)
PfL (sol-gel)
none
none
18-crown-6
none
none
18-crown-6
Tween 80
none
none
100
529
425
6
14
45
28
6
188
790
966
51
288
964
937
34
70.1
> 100
> 100
4.2
5.1
4.4
5.6
18.8
18.3
4.1
4.0
4.5
2.2
3.3
6.0
¾
28
40
108
625
3.2
18.4
10
18-crown-6
Similar (non-optimized) experiments were then per- (15) was obtained having an enantiomeric purity of ee ˆ
formed with the chiral alcohols 9 12. In almost all cases 93% at 52% conversion, while the (S)-amine (13) was
sol-gel entrapment has a strongly beneficial effect as isolated enantiomerically pure (ee ˆ 99%).[27] The E-
measured in the first cycle (Table 4). An exception is the value was reported to be > 100. Upon immobilizing the
kinetic resolution of 1-phenylethanol (rac-9) catalyzed
by CaLB. Although the selectivity factor (E > 100) is not
effected by sol-gel encapsulation, the lipase activity is
reduced slightly (Table 4), which is currently difficult to
explain. Of course, the advantage regarding the possi-
bility of recycling the heterogeneous catalyst remains. In
several other cases involving other substrates recycling
experiments were also performed, again illustrating
industrial viability. These as well as our original
recycling experiments[10,11] involving the first generation
immobilizates show that sol-gel encapsulated lipases are
extremely robust heterogeneous biocatalysts.
A disadvantage of using enzymes in organic synthesis
is the fact that in many cases the amount of biocatalyst
used weight-wise equals or even exceeds the weight of
substrate used in the reaction.[25] In the case of the
present immobilizates this is not the case, in spite of the
fact that the major part of the catalyst weight-wise
originates from the silicate matrix. For example, in a
preparative scale reaction of rac-10, only 250 mg of sol-
gel CaLB-immobilizate was used per 10 g of substrate
(see Experimental Section).
Finally, the acylating kinetic resolution of 1-phenyl-
ethylamine (rac-13) was studied. Whereas the reaction
is extremely slow when employing ethyl acetate as the
acylating agent (reaction time: 7 days),[26] scientists at
BASF discovered that methoxyacetic acid ethyl ester
(14) results in dramatic rate acceleration.[27] Upon using
the lipase from Burkholderia plantarii, the (R)-amide
Figure 3. Recycling experiments in the kinetic resolution of
&
rac-6 using sol-gel CaLB-immobilizate ( ) and commercial
^
CaLB powder ( ). a) Specific activity. b) Decrease in
activity (%).
Scheme 4.
Adv. Synth. Catal. 2003, 345, 717 728
asc.wiley-vch.de
¹ 2003 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim
723

Manfred T. Reetz et al.
FULL PAPERS
Table 4. Kinetic resolution of alcohols 9 11 using free and sol-gel encapsulated lipases (Table 1) as catalysts, vinyl acetate as
the acylating agent and isooctane as solvent.
Entry
Substrate
Lipase
Additives
Activity
[mmol/min ¥ g gel]
Specific Activity
[mmol/min ¥ g protein]
Relative
Activity
E
1
2
3
4
5
6
7
8
9
9
9
9
9
9
9
9
9
CaLB (powder)
CaLB (sol-gel)
CaLB (sol-gel)
BcL (powder)
BcL (sol-gel)
BcL (sol-gel)
BcL (sol-gel)
PfL (powder)
PfL (sol-gel)
PfL (sol-gel)
CaLB (powder)
CaLB (sol-gel)
CaLB (sol-gel)
BcL (powder)
BcL (sol-gel)
BcL (sol-gel)
BcL (sol-gel)
PfL (powder)
PfL (sol-gel)
PfL (sol-gel)
CaLB (powder)
CaLB (sol-gel)
CaLB (sol-gel)
BcL (powder)
BcL (sol-gel)
BcL (sol-gel)
BcL (sol-gel)
PfL (powder)
PfL (sol-gel)
PfL (sol-gel)
BcL (powder)
BcL (sol-gel)
BcL (sol-gel)
BcL (sol-gel)
none
none
18-crown-6
none
none
18-crown-6
Tween 80
none
none
18-crown-6
none
none
18-crown-6
none
none
18-crown-6
Tween 80
none
none
18-crown-6
none
none
18-crown-6
none
none
18-crown-6
Tween 80
none
none
18-crown-6
none
none
18-crown-6
Tween 80
467
292
332
8
35
50
42
3
21
25
214
214
253
6
59
36
23
6
23
14
876
436
755
63
700
1057
1393
15
>100
>100
>100
>100
>100
>100
>100
>100
>100
>100
>100
>100
>100
>100
>100
>100
>100
>100
>100
>100
0.5
0.9
11.1
16.8
22.2
¾
9
9
80
5.3
26.1
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
392
401
320
575
51
1174
774
777
32
10[a]
10[a]
10[a]
10[a]
10[a]
10[a]
10[a]
10[a]
10[a]
10[a]
11
0.8
1.4
22.9
15.1
15.2
¾
87
2.8
7.1
225
1726
664
1695
419
1956
4662
4843
367
713
1311
0.33
5.2
920
444
746
51
1.9
2.0
1.2
1.1
2.7
2.5
2.4
1.2
1.6
11
11
11
11
11
11
11
11
0.4
1.0
98
4.7
11.1
11.6
219
145
68
186
84
0.040
0.262
0.262
0.323
¾
1.9
3.6
11
1.5
12[b]
12[b]
12[b]
12[b]
29.1
31.6
40.2
32.6
15.9
16.9
36.2
5.6
12.0
¾
[a]
[b]
Solvent: toluene.
Solvent: tetrahydrofuran.
lipase on Amberlite XAD7 and carrying out the commercial powder as well as the sol-gel (from 4e/5)
reaction 10 times, the remaining activity was measured encapsulated forms with or without 18-crown-6 are
to be 80%. The BASF plant has started to produce the excellent catalysts, the desired 50% conversion being
amine at a scale of 1000 tons/year.[28]
reached within 29 hours and the enantioselectivity being
Since the lipase from Burkholderia plantarii is not very high (E > 100). This means that the reaction
commercially available, we began to test the lipases essentially stops when the (R)-enantiomer of 13 has
accessible in the present study. Since we already knew been consumed, which is ideal. Of the three catalysts
that CaLB is very enantioselective in the acylation of 13 studied, the sol-gel encapsulated variant prepared in the
using ethyl acetate,[26] the reaction was repeated with 14 presence of 18-crown-6 turned out to be most active.
as the acylating agent. It was observed that the Therefore, recycling experiments were performed with
Scheme 5.
724
¹ 2003 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim
asc.wiley-vch.de
Adv. Synth. Catal. 2003, 345, 717 728

Second Generation Sol-Gel Encapsulated Lipases
FULL PAPERS
Experimental Section
Materials and Instruments
¾
The lipases from Candida antarctica (CaLB; Chirazyme L-2 ),
¾
Candida rugosa (CrL; Chirazyme L-3 ) and Mucor miehei
¾
(MmL; ChirazymeL-9 ) were obtained from Roche; Aspergil-
lus niger (AnL; Amano AS), Burkholderia cepacia (BcL;
Amano PS) and Pseudomonas fluorescens (PfL; Amano AK
™20∫) from Amano Pharmaceutical Co.; Candida rugosa type
VII (CrL type VII) from Sigma; Penicillium roqueforti (PrL)
from Fluka; Thermomyces lanuginosa (TIL; Novozym SP523)
from Novo Nordisk. The chiral alcohols were purchased from
Aldrich; isobutyltrimethoxysilane (98%) from Lancaster; n-
¾
butyltrimethoxysilane (97%) from ABCR; Tween 80 [poly-
oxyethylene(20)sorbitan monooleate] from Fluka; polyvinyl
Figure 4. Recycling experiments in the kinetic resolution of
rac-13 using the sol-gel CaLB-immobilizate prepared with 18-
crown-6 as an additive.
ˆ
alcohol (PVA; MW 15000) from Merck; sodium fluoride
¾
(99%) from Fluka; vinyl acetate (99%) from Acros; Celite 577
from Fluka; isooctane (99%) from Fluka; methyl-b-cyclodex-
trin from Aldrich; 18-crown-6 (99%) from Fluka; tetramethoxy-
silane (98%) from Fluka; lauric acid (99%) from Fluka; 1-
phenylethylamine (98%) from Fluka; methoxyacetic acid ethyl
ester (97%) and n-octanol (99.5%) from Fluka.
Gas chromatographic analyses were performed in the
Department of Chromatography, Max-Planck-Institut f¸r
Kohlenforschung, M¸lheim/Ruhr, Germany.
this material. Figure 4 shows the results of five cycles,
which demonstrate constant performance with respect
to activity. Moreover, no reduction in enantioselectivity
was observed in any of the runs.[22a]
Conclusions
General Procedure for Sol-Gel Entrapment of Lipases
The present study reveals that the second generation
sol-gel lipase immobilizates are extremely active, robust
and recyclable heterogeneous biocatalysts in esterifica-
tion reactions. The preparation involves high lipase
loading as well as the use of additives such as 18-crown-6
A commercial lipase powder (lyophilizate) such as AnL
(150 mg), BcL (150 mg), CaLB (125 mg), CrL (150 mg), CrL
type VII (60 mg), MmL (150 mg), PfL (150 mg), PpL (150 mg),
PrL (150 mg) or TIL (70 mg) was placed in a 50-mL Falcon
tube (Corning) together with TRIS/HCl-buffer (390 mL; 0.1 M;
pH 7.5) and the mixture was vigorously shaken with a Vortex-
¾
or Tween 80 . Ten different lipases were tested, and all
¾
of them showed significantly improved performance,
although in two cases additive effects were not observed.
Mixer. In the case of an additive or Celite , these materials
were included, the amounts of which are listed in the tables.
Upon going from n-butyl- to isobutyltrimethoxysilane Then 100 mL of aqueous PVA (4% W/V), aqueous sodium
fluoride (50 mL of a 1 M solution) and isopropyl alcohol
(100 mL) were added, and the mixture homogenized using a
Vortex-Mixer. Then the alkylsilane (2.5 mmol) and TMOS
(0.5 mmol; 74 mL; 76 mg) were added and the mixture agitated
once more for 10 to 15 seconds. Gelation was usually observed
within seconds or minutes while gently shaking the reaction
vessel. Following drying overnight in the opened Falcon tube,
isopropyl alcohol (10 15 mL) was added in order to facilitate
removal of the white solid material (filtration). The gel was
successively washed with distilled water (10 mL), isopropyl
alcohol (10 mL) and n-pentane (10 mL). During this process a
spatula was used to crush the gel. Thereafter the lipase
immobilizate was placed in an open 2-mL plastic vessel and
dried in the air at room temperature.
(4d and 4e, respectively) as the precursor in the sol-gel
process, differences in activity and enantioselectivity are
usually small, although notable exceptions were ob-
served depending upon the particular substrate used. In
typical preparative scale reactions only about 250 mg of
a lipase-containing gel are needed for 10 g of substrate.
In future applications involving new substrates it is
advisable to test several variants of the sol-gel encapsu-
lated lipases described herein. In industrial applications
this should include those materials prepared by ™double
immobilization∫ based on the use of an additional solid
¾
porous material such as Celite which bonds in its pores
the actual lipase-containing gels and imparts enhanced
mechanical stability as well as activity. Future studies in
our laboratories will focus on the origin of improved
catalyst performance.
Determination of Protein Content
In order to determine the protein content of the commercial
lipases as well as the degree of loading, the BCA Protein Assay
Kit of Sigma was used. Accordingly, solutions of the commer-
cial lipase or the wash-solutions following the sol-gel process
were incubated for 30 min and measured at 378C using a UV/
Adv. Synth. Catal. 2003, 345, 717 728
asc.wiley-vch.de
¹ 2003 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim
725

Manfred T. Reetz et al.
FULL PAPERS
Vis-spectrometer at 562 nm according to the Technical Bulletin were added to H2O-saturated toluene (10 mL). The reaction
of Sigma. Distilled water was used as a reference and BSA as
standards. The degree of immobilization ranges between 0.3
and 0.9.
mixture was shaken at ambient temperature (230 minÀ1) for
several hours. After defined intervals samples of 50 mL were
taken, after which they were diluted with 400 mL toluene and
centrifuged for 2 min (13000 min-1). 200 mL of the supernatant
were analyzed via gas chromatography. For recycling of the
immobilizates the reaction mixture was filtered and the
immobilizates were washed three times with 10 15 mL
toluene and once with 10 15 mL pentane. After drying the
entrapped lipase for 24 h at room temperature under air, they
were reused in the kinetic resolution described above. For this
reason the amounts of reactants were adjusted to the amountof
recycled immobilizate.
Determination of Lipase Activity
The sol-gel lipase immobilizate (1 to 50 mg, depending on the
activity) was placed in a Falcon tube together with a solution of
lauric acid (100 mg; 0.5 mmol) and n-octanol (158 mL; 130 mg;
1.0 mmol) in non-dried (H2O-saturated) isooctane. The mix-
ture was then shaken (180 minÀ1) at 308C. At defined intervals
(usually after 15, 30, 45 and 60 min) samples (300 mL) were
taken and centrifuged (13000 minÀ1) before the gas chromato-
graphic determination of lauric acid and lauric acid n-octyl
ester was carried out using 150 mL of the supernatant. By
applying a linear regression of the measured values the initial
reaction rate in mmol/min and therefore the activity relative to
1 g of immobilizate can be determined. By considering the
degree of loading, the specific activity is calculated. The
relative activity is determined by dividing the specific activity
of the immobilizate by the specific activity of the commercial
lipase powder.
Acknowledgements
Support by the Fonds der Chemischen Industrie is gratefully
acknowledged.
References and Notes
[1] a) H. G. Davies, R. H. Green, D. R. Kelly, S. M. Roberts,
Biotransformations in Preparative Organic Chemistry:
The Use of Isolated Enzymes and Whole Cell Systems in
Synthesis, Academic Press, London, 1989; b) C.- H.
Wong, G. M. Whitesides, Enzymes in Synthetic Organic
Chemistry, (Tetrahedron Organic Chemistry Series, Vol.
12), Pergamon, Oxford, 1994; c) Enzyme Catalysis in
Organic Synthesis: A Comprehensive Handbook, Vol. I-
II, (Eds.: K. Drauz, H. Waldmann), VCH, Weinheim,
1995; d) K. Faber, Biotransformations in Organic Chem-
istry, 3rd edn., Springer, Berlin, 1997.
[2] a) A. M. Brzozowski, U. Derewenda, Z. S. Derewenda,
G. G. Dodson, D. M. Lawson, J. P. Turkenburg, F. Bjor-
kling, B. Huge-Jensen, S. A. Patkar, L. Thim, Nature
1991, 351, 491 494; b) K.-E. Jaeger, S. Ransac, B. W.
Dijkstra, C. Colson, M. van Heuvel, O. Misset, FEMS
Microbiol. Rev. 1994, 15, 29 63; c) H. van Tilbeurgh, M.-
P. Egloff, C. Martinez, N. Rugani, R. Verger, C.
Cambillau, Nature 1993, 362, 814 820; d) R. D. Schmid,
R. Verger, Angew. Chem. 1998, 110, 1694 1720; Angew.
Chem. Int. Ed. 1998, 37, 1608 1633; e) P. Villeneuve,
J. M. Muderhwa, J. Graille, M. J. Haas, J. Mol. Catal. B:
Enzym. 2000, 9, 113 148; f) A. Schmid, J. S. Dordick, B.
Hauer, A. Kiener, M. Wubbolts, B. Witholt, Nature 2001,
409, 258 268; g) A. M. Klibanov, Nature 2001, 409, 241
246; h) M. T. Reetz, Curr. Opin. Chem. Biol. 2002, 6,
145 150.
General Procedure for Evaluating Lipase-
Immobilizates as Catalysts in the Kinetic Resolution of
Chiral Alcohols
A Falcon tube (Corning) was charged with a mixture of a
racemic alcohol (0.5 mmol), vinyl acetate (93 mL; 86 mg;
1 mmol) and free or sol-gel immobilized lipase [typically
5 mg in 10 mL solvent (typically isooctane)]. The mixture was
shaken at 308C, samples (150 mL) being taken for GC analysis
at 15 minutes intervals. The formula of Sih et al.[23] was used to
calculate the E-value. For recycling experiments the reaction
mixture was centrifuged, the supernatant was decanted and the
lipase or the sol-gel immobilizate washed 4 times with
isooctane and reused without any further treatment.
Typical Kinetic Resolution on a Preparative Scale
Similar to the above protocol a mixture of rac-1-(2-naphthyl)-
ethanol (10 g; 58 mmol), vinyl acetate (8.1 mL; 7.5 g; 87 mmol)
and 250 mg of a sol-gel CaLB-immobilizate (prepared in the
presence of 18-crown-6 as additive) in toluene (300 mL) was
shaken at 358C for 48 h. Following filtration GC analysis
showed a conversion of 50.0%, the ee of non-reacted (S)-1-(2-
naphthyl)ethanol and product ((R)-acetate) each being >
99.9%. The immobilizate was removed by filtration, washed
with toluene and pentane and can be reused without significant
loss in activity or enantioselectivity.
[3] D. Cowan, Trends Biotechnol. 1996, 14, 177 178.
[4] a) Methods in Enzymology, Vol. 135 137: Immobilized
Enzymes and Cells, Parts B-D, (Ed.: K. Mosbach)
Academic Press, San Diego, 1987/1988; b) Bioprocess
Technology, Vol. 16: Industrial Application of Immobi-
lized Biocatalysts, (Eds.: A. Tanaka, T. Tosa, T. Kobaya-
shi), Marcel Dekker, New York, 1993; c) J. F. Kennedy,
E. H. M. Melo, K. Jumel, Chem. Eng. Prog. 1990, 7, 81
89; d) Y. Okahata, T. Mori, Trends Biotechnol. 1997, 15,
50 54; e) R. V. Parthasarathy, C. R. Martin, Nature
General Procedure for the Kinetic Resolution of
1-Phenylethylamine
rac-1-Phenylethylamine (127.4 mL; 121.3 mg; 1.0 mmol), me-
thoxyacetic acid ethyl ester (352.2 mL; 354.3 mg; 3.0 mmol) and
75 mg of free or immobilized lipase B from Candida antarctica
726
¹ 2003 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim
asc.wiley-vch.de
Adv. Synth. Catal. 2003, 345, 717 728

Second Generation Sol-Gel Encapsulated Lipases
FULL PAPERS
1994, 369, 298 301; f) W. Tischer, F. Wedekind, Top.
Curr. Chem. 1999, 200, 95 126; g) D. E. Stevenson,
R. A. Stanley, R. H. Furneaux, Enzyme Microb. Technol.
1994, 16, 478 484; h) J.-T. Lai, S.-C. Wu, H.-S. Liu,
Bioprocess Eng. 1998, 18, 155 161; i) A. Corma, V.
Fornes, F. Rey, Adv. Mater. (Weinheim, Ger.) 2002, 14,
71 74; j) A. R. Rodrigues, J. M. S. Cabral, M. A. Taipa,
Enzyme Microb. Technol. 2002, 31, 133 141; k) N.
Qureshi, D. V. Tamhane, Appl. Microbiol. Biotechnol.
1985, 21, 280 281; l) I. Umemura, S. Takamatsu, T. Sato,
T. Tosa, I. Chibata, Appl. Microbiol. Biotechnol. 1984, 20,
291 295; m) N. W. Fadnavis, R. L. Babu, G. Sheelu, A.
Deshpande, Tetrahedron Asymmetry 2001, 12, 1695
1699; n) S. Fukui, A. Tanaka, Adv. Biochem. Eng./
Biotechnol. 1984, 29, 1 33; o) M. Bakker, F. van de
Velde, F. van Rantwijk, R. A. Sheldon, Biotechnol. Bio-
eng. 2000, 70, 342 348; p) M. Yasuda, M. Kobayashi, S.
Watanabe, H. Ogino, K. Ishimi, H. Ishikawa, J. Polym.
Sci., Part A: Polym. Chem. 2002, 40, 874 884; q) L.
Girono, E. Drioli, Trends Biotechnol. 2000, 18, 339 349.
[5] A. L. Margolin, Trends Biotechnol. 1996, 14, 223 230.
Physics and Chemistry of Sol-Gel Processing, Academic
Press, Boston, 1990.
[10] a) M. T. Reetz, A. Zonta, J. Simpelkamp, Angew. Chem.
1995, 107, 373 376; Angew. Chem. Int. Ed. Engl. 1995,
34, 301 303; b) A. Zonta, Dissertation, Ruhr-Universit‰t
Bochum, Germany, 1994; c) M. T. Reetz, A. Zonta, J.
¬
Simpelkamp, A. Rufinska, B. Tesche, J. Sol-Gel Sci.
Technol. 1996, 7, 35 43; d) M. T. Reetz, A. Zonta, V.
Vijayakrishnan, K. Schimossek, J. Mol. Catal. A: Chem.
1998, 134, 251 258; e) M. T. Reetz, R. Wenkel, D. Avnir,
Synthesis 2000, 781 783; f) M. T. Reetz, A. Zonta, J.
Simpelkamp, Patents DE-A-4408152 A1 and US-PS
5,817,493.
[11] Review of enzymes entrapment in hydrophobic silicates:
M. T. Reetz, Adv. Mater. (Weinheim, Ger.) 1997, 9, 943
954.
[12] a) G. Kuncova, M. Guglielmi, P. Dubina, B. Safar,
Collect. Czech. Chem. Commun. 1995, 60, 1573 1577;
b) M. Heidt, U. T. Bornscheuer, R. D. Schmid, Biotech-
nol. Tech. 1996, 10, 25 30; c) K. Kawakami, S. Yoshida,
J. Ferment. Bioeng. 1996, 82, 239 245; d) R. Pfau, H.
Kunz, Synlett 1999, 1817 1819; e) S.-Y. Furukawa, K.
Kawakami, J. Ferment. Bioeng. 1998, 85, 240 242; f) S.-
Y. Furukawa, T. Ono, H. Ijima, K. Kawakami, J. Mol.
Catal. B: Enzym. 2002, 17, 23 28; g) T. Keeling-Tucker,
M. Rakic, C. Spong, J. D. Brennan, Chem. Mater. 2000,
12, 3695 3704; h) M. Schuleit, P. L. Luisi, Biotechnol.
Bioeng. 2001, 72, 249 253; i) K. Kato, Y. F. Gong, T.
Saito, Y. Yokogawa, Biosci. Biotechnol, Biochem. 2002,
66, 221 223; j) M. Pierre, P. Buisson, F. Fache, A. Pierre,
Biocatal. Biotransform. 2000, 18, 237 251; k) P. Buisson,
C. Hernandez, M. Pierre, A. C. Pierre, J. Non-Cryst.
Solids 2001, 285, 295 302; l) D.-J. van Unen, J. F. J.
Engbersen, D. N. Reinhoudt, Biotechnol. Bioeng. 2001,
75, 154 158; m) S.-Y. Furukawa, T. Ono, H. Ijima, K.
Kawakami, Biotechnol. Lett. 2002, 24, 13 16.
¬
[6] P. Lopez-Serrano, L. Cao, F. van Rantwijk, R. A. Shel-
don, Biotechnol. Lett. 2002, 24, 1379 1383.
[7] Reviews of enzyme entrapment in SiO2: a) D. Avnir, S.
Braun, O. Lev, M. Ottolenghi, Chem. Mater. 1994, 6,
1605 1614; b) D. Avnir, Acc. Chem. Res. 1995, 28, 328
334; c) J. Livage, C. R. Acad. Sci., Ser. IIb: Mec., Phys.
Chim. Astron. 1996, 322, 417 427; d) I. Gill, Chem.
Mater. 2001, 13, 3404 3421.
[8] See for example: a) L. M. Ellerby, C. R. Nishida, F.
Nishida, S. A. Yamanaka, B. Dunn, J. S. Valentine, J. I.
Zink, Science 1992, 255, 1113 1115; b) H. H. Weetall, B.
Robertson, D. Cullin, J. Brown, M. Walch, Biochim.
Biophys. Acta 1993, 1142, 211 233; c) Y. Tatsu, K.
Yamashita, M. Yamaguchi, S. Yamamura, H. Yamamoto,
S. Yoshikawa, Chem. Lett. 1992, 1615 1618; d) P.
Audebert, C. Demaille, C. Sanchez, Chem. Mater. 1993,
5, 911 913; e) V. Glezer, O. Lev, J. Am. Chem. Soc. 1993,
115, 2533 2534; f) R. Zusman, D. A. Beckman, I. Zus-
man, R. L. Brent, Anal. Biochem. 1992, 201, 103 106;
g) C. Dosoretz, R. Armon, J. Starosvetzky, N. Roths-
child, J. Sol-Gel Sci. Technol. 1996, 7, 7 11; h) C. T. Lin,
C. M. Catuara, J. E. Erman, K. C. Chen, S. F. Huang,
W. J. Wang, H. H. Wei, J. Sol-Gel Sci. Technol. 1996, 7,
19 26; i) M. M. Pereira, L. L. Hench, J. Sol-Gel Sci.
Technol. 1996, 7, 59 68; j) M. Rietti-Shati, D. Ronen,
R. T. Mandelbaum, J. Sol-Gel Sci. Technol. 1996, 7, 77
79; k) Z. Chen, D. L. Kaplan, K. Yang, J. Kumar, K. A.
Marx, S. K. Tripathy, J. Sol-Gel Sci. Technol. 1996, 7, 99
108; l) S. Yamanaka, N. P. Nguyen, B. Dunn, J. S.
Valentine, J. I. Zink, J. Sol-Gel Sci. Technol. 1996, 7,
117 121; m) A. Turniansky, D. Avnir, A. Bronshtein, N.
Aharonson, M. Altstein, J. Sol-Gel Sci. Technol. 1996, 7,
135 143; n) R. Armon, C. Dosoretz, J. Starosvetsky, F.
Orshansky, I. Saadi, J. Biotechnol. 1996, 51, 279 285;
o) Y. Wei, J. Xu, Q. Feng, M. Lin, H. Dong, W.-J. Zhang,
C. Wang, J. Nanosci. Nanotechnol. 2001, 1, 83 93.
[9] a) L. L. Hench, J. K. West, Chem. Rev. 1990, 90, 33 72;
b) C. J. Brinker, G. W. Scherer, in Sol-Gel Science: The
¬
¬
[13] J. D. Badjic, E. N. Kadnikova, N. M. Kostic, Org. Lett.
2001, 3, 2025 2028.
[14] M. T. Reetz, A. Zonta, J. Simpelkamp, W. Kˆnen, Chem.
Commun. (Cambridge, U. K.) 1996, 1397 1398.
[15] M. T. Reetz, A. Zonta, J. Simpelkamp, Biotechnol.
Bioeng. 1996, 49, 527 534.
[16] Several first generation sol-gel lipase immobilizates are
commercially available from Fluka (Switzerland), al-
though enzyme loading is very low leading to low
activities.
[17] a) D. N. Reinhoudt, A. M. Eendebak, W. F. Nijenhuis, W.
Verboom, M. Kloosterman, H. E. Schoemaker, J. Chem.
Soc. Chem. Commun. 1989, 399 400; b) J. F. J. Engbers-
en, J. Broos, W. Verboom, D. N. Reinhoudt, Pure Appl.
Chem. 1996, 68, 2171 2178; c) D.-J. van Unen, J. F. J.
Engbersen, D. N. Reinhoudt, Biotechnol. Bioeng. 2002,
77, 248 255.
[18] a) K. Griebenow, Y. D. Laureano, A. M. Santos, I. M.
Clemente, L. RodrÌguez, M. W. Vidal, G. Barletta, J. Am.
Chem. Soc. 1999, 121, 8157 8163; b) A. M. Santos, I. M.
Clemente, G. Barletta, K. Griebenow, Biotechnol. Lett.
1999, 21, 1113 1118.
Adv. Synth. Catal. 2003, 345, 717 728
asc.wiley-vch.de
¹ 2003 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim
727

Manfred T. Reetz et al.
FULL PAPERS
[19] a) Y. L. Khmelnitsky, S. H. Welch, D. S. Clark, J. S.
Dordick, J. Am. Chem. Soc. 1994, 116, 2647 2648;
b) D. H. Altreuter, J. S. Dordick, D. S. Clark, J. Am.
Chem. Soc. 2002, 124, 1871 1876.
[20] Y.-Y. Liu, J.-H. Xu, Y. Hu, J. Mol. Catal. B: Enzym. 2000,
10, 523 529.
[24] Since at very high enantioselectivity precision in the E-
value decreases,[23] we prefer not to rely on specific
numbers in such cases but rather report E > 100 which
for all practical purposes means that the reaction stops at
50% conversion with formation of enantiopure com-
pounds.
[21] a) I. J. Colton, S. N. Ahmed, R. J. Kazlauskas, J. Org.
Chem. 1995, 60, 212 217; b) K. Zhu, A. Jutila, E. K. J.
Tuominen, P. K. J. Kinnunen, Protein Sci. 2001, 10, 339
351; c) A. Cipiciani, F. Bellezza, J. Mol. Catal. B: Enzym.
2002, 17, 261 266.
[25] Margolin has stated that ™it is not unusual to see
processes employing more enzyme than substrate by
weight∫: N. Khalaf, C. P. Govardhan, J. J. Lalonde, R. A.
Persichetti, Y.-F. Wang, A. L. Margolin, J. Am. Chem.
Soc. 1996, 118, 5494 5495.
[22] a) P. Tielmann, Dissertation, Ruhr-Universit‰t Bochum,
Germany, 2002; b) W. Wiesenhˆfer, Diplomarbeit, Ger-
hard-Mercator-Universit‰t GH Duisburg, Germany,
2001.
[26] M. T. Reetz, C. Dreisbach, Chimia 1994, 48, 570.
[27] F. Balkenhohl, K. Ditrich, B. Hauer, W. Ladner, J. Prakt.
Chem./Chem.-Ztg. 1997, 339, 381 384.
[28] V. Gotor, Biocatal. Biotransform. 2000, 18, 87 103.
[23] C.-S. Chen, Y. Fujimoto, G. Girdauskas, C. J. Sih, J. Am.
Chem. Soc. 1982, 104, 7294 7299.
728
¹ 2003 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim
asc.wiley-vch.de
Adv. Synth. Catal. 2003, 345, 717 728