Thermal and Mechanical Stability of Immobilized Candida antarctica Lipase B: an Approximation to Mechanochemical Energetics in Enzyme Catalysis.

Article abstract of DOI:10.1002/cctc.201901714

Very recently, several successful enzymatic processes performed with mechanical activation have been disclosed; that is, despite the mechanical stress caused by High-Speed Ball-Milling, immobilized enzymes can retain activity. In the present study, the effect of thermal and mechanical stress was examined as potential inducers of enzymatic denaturation, when using either free, immobilized, or ground immobilized enzyme. The recorded observations show a remarkable stability of ground immobilized enzyme. Moreover, ground biocatalyst turns out to exhibit an increase of one order of magnitude in the efficiency of the catalytic process, maintaining excellent enantiodiscrimination, without significant activity loss even after four milling cycles. These observations rule out enzyme inactivation as direct consequence of the milling process. Additionally, boosted enzyme efficiency was used to optimize a relatively inefficient chiral amine resolution reaction, achieving a 25 % faster biotransformation (in 45 min) and yielding essentially enantiopure products (ee>99%, E>500).

Full text of DOI:10.1002/cctc.201901714

Accepted Article
Title: Thermal and mechanical stability of immobilized Candida
antarctica Lipase B: an approximation to mechanochemical
energetics in enzyme catalysis.
Authors: Mario Pérez-Venegas, Miriam M Tellez-Cruz, Omar Solorza-
Feria, Agustín Lopéz-Munguía, Edmundo Castillo, and
Eusebio Juaristi
This manuscript has been accepted after peer review and appears as an
Accepted Article online prior to editing, proofing, and formal publication
of the final Version of Record (VoR). This work is currently citable by
using the Digital Object Identifier (DOI) given below. The VoR will be
published online in Early View as soon as possible and may be different
to this Accepted Article as a result of editing. Readers should obtain
the VoR from the journal website shown below when it is published
to ensure accuracy of information. The authors are responsible for the
content of this Accepted Article.
To be cited as: ChemCatChem 10.1002/cctc.201901714
Link to VoR: http://dx.doi.org/10.1002/cctc.201901714
A Journal of
www.chemcatchem.org

10.1002/cctc.201901714
ChemCatChem
FULL PAPER
Thermal and mechanical stability of immobilized Candida
antarctica Lipase B: an approximation to mechanochemical
energetics in enzyme catalysis.
Mario Pérez-Venegas,*[a] Miriam M. Tellez-Cruz,[a] Omar Solorza-Feria,[a] Agustín López-Munguía,[b]
Edmundo Castillo,*[b] Eusebio Juaristi,*[a, c]
Abstract: Very recently, several successful enzymatic processes
carried out with mechanical activation have been disclosed; that is,
despite the mechanical stress caused by High-Speed Ball-Milling,
immobilized enzymes can retain activity. In the present study, the
effect of thermal and mechanical stress was examined as potential
inducers of enzymatic denaturation, when using either free,
immobilized, or ground immobilized enzyme. The recorded
observations show a remarkable stability of ground immobilized
enzyme. Moreover, ground biocatalyst turns out to exhibit an increase
of one order of magnitude in the efficiency of the catalytic process,
maintaining excellent enantiodiscrimination, without significant activity
loss even after four milling cycles. These observations rule out
enzyme inactivation as direct consequence of the milling process.
Additionally, boosted enzyme efficiency was used to optimize a
relatively inefficient chiral amine resolution reaction, achieving a 25%
faster biotransformation (in 45 min) and yielding essentially
enantiopure products (ee > 99%, E > 500).
issues to overcome in order to generalize the use of enzymes
on an industrial scale, especially in the area of organic
synthesis. Salient obstacles include the normally high cost
and limited commercial availability of enzymes. In some
cases this is directly related to a limited industrial production
of bioactive pure proteins.[13] Furthermore, the recent
10]
developments in enzyme engineering[9,
and in novel
methods for protein production and purification have
contributed to increase the availability of biocatalysts at more
affordable prices.[11, 12] Additional limitations in the use of
biocatalysts in organic synthesis come from severe
operational requirements of enzymatic processes such as
the use of non-aqueous solvent media and prevention of high
temperatures in order to avoid denaturation of the protein
and, as a consequence, a reduction of reaction efficiency.[14-
16]
Indeed, biocatalyst inactivation provoked by harsh chemical
reaction conditions that contrast with the mild natural media
of enzymes,[16] is the most common challenge when sensible
proteins are used in synthetic chemistry. Normally, enzyme
inactivation arises from partial or complete protein unfolding,
that is physical loss of the native structure. This process
could lead to changes in the tertiary or quaternary structure
of the enzyme’s active site that hamper the predicted
bioconversion.[17] In this regard, several attempts to stabilize
the enzymes have been described.[18] These efforts include
the use of non-conventional reaction media,[19, 20] appropriate
chemical modification of the enzyme,[21] and the design of
biocatalysts through enzyme immobilization on inert
supports, which offer substantial benefits when compared
with the use of free enzymes in solution.[22] The easy
Introduction
Enzymes are potent biocatalysts that are able to carry out
numerous chemical transformations with remarkable chemo-
, regio- and stereo-selectivity.[1-5] The intrinsic sustainability
of enzymatic processes along with the use of green reaction
conditions such as absence of potentially toxic solvents, and
activation with non-conventional sources of energy has
expanded the use of biocatalysis in both chemical industries
and in academic institutions, focusing in drug discovery and
in the synthesis of molecules with relevant biological
activity.[6-8] Recent developments in protein engineering and
enzyme directed evolution,[9, 10] together with the availability
of efficient methods for the production of recombinant
molecules, have led to rapidly expanding new fields in
biocatalysis.[11, 12] Nevertheless, there are still several major
handling and convenient
recovery, reduced protein
contamination as well as compatibility with continuous flow
procedures have intensified the application of immobilized
biocatalysts, especially those produced with broadly used
enzymes in industry such as isomerases and lipases.[23, 24]
In this context, Candida antarctica Lipase B (CALB,
commercially available as Novozym 435® [N435]) is a widely
used enzyme, supported on acrylic resin, that has exhibit
multiple pharmaceutical applications.[25] Particularly relevant
is the extensive use of CALB to promote highly efficient
bioconversions using non-aqueous media at relatively high
temperatures.[26] Furthermore, CALB as N435 has shown
remarkable stability and recyclability under rough chemical
conditions.[27, 28]
[a]
[b]
[c]
M. Pérez-Venegas, M. M. Tellez-Cruz, O. Solorza-Feria and E.
Juaristi
Department of Chemistry
Centro de Investigación y de Estudios Avanzados
Av. IPN 2508, 07360 Ciudad de México, Mexico
E-mail: mperezv@cinvestav.mx; juaristi@relaq.mx
A. López-Munguía, E. Castillo
Department of cellular engineering and biocatalysis
Universidad Nacional Autónoma de México
Av. Universidad 2001, Col. Chamilpa, 62210 Cuernavaca, Morelos,
Mexico
E-mail: edmundo@ibt.unam.mx
E. Juaristi
El Colegio Nacional
Luis Gonzáles Obregón 23, Centro Histórico, 06020 Ciudad de
México, Mexico.
E-mail: juaristi@relaq.mx
Supporting information for this article is given via a link at the end of
the document.
This article is protected by copyright. All rights reserved.

10.1002/cctc.201901714
ChemCatChem
FULL PAPER
Recently, as part of our interest in the development of more
sustainable synthetic methodologies,[29-32] and considering
previous reports using proteins mechanically activated
through the High-Speed Ball-Milling technique,[33, 34] we were
nitrophenyl butyrate as substrate, in order to establish the
appropriate conditions to carry out the experimental
measurements of thermal stability (for extensive detailed
information see Experimental Section and Supporting
Information [SI], Section 1), which were then conducted at
temperatures above the optimal stable range of CALB. As
shown in Figure 1, when free CALB was incubated at 65 °C,
a typical first order decay was observed, the residual activity
rapidly decreasing during the first part of the essay.
Nevertheless, it is apparent that inactivation of the enzyme
significantly slowed down at longer periods of time at high
temperature exposure, probably due to an apparent
stabilizing effect of CALB forms (labile [L] and resistant [R]). It
is also likely that, aggregation of the free enzyme at high
temperatures may also contributed to the loss of activity,
indicating a most complex inactivation mechanism. A similar
behavior was also observed when CALB was incubated at
75 °C and 85 °C (Figure 1).
able
to
carry
out
various
stereoselective
biotransformations.[35, 36] Gratifyingly, CALB partially retains
activity after the harsh mechanochemical conditions
provoked by the collisions inside the milling-jar, and proved
to be stable enough to carry out its catalytic function,
36]
affording highly enantioenriched products.[35,
In this
regard, the stability of enzymes under mechanical stress has
been confirmed using continuous grinding processes or
mechanochemical activation followed by static incubation
38]
(RAging) with excellent results.[37,
Nevertheless, no
detailed information concerning the effect of milling on the
degree of denaturation, the parameters and conditions
affecting enzyme efficiency, or changes in the structure of
CALB have been disclosed.
On the other hand, besides the gradual protein denaturation
reported in experiments carried out at high temperatures or
36]
aggressive reaction conditions,[35,
one might also
anticipate loss of the enzyme due to leakage from the
commercial support, that could arise from the stressful
mechanical methods employed in the ball milling process.
Nevertheless, the studies reported so far using CALB
together with mechanical force do not identify explicitly the
possible denaturation effect of milling on enzymatic
reactivity. This makes difficult to assign the loss in activity to
either mechanical, thermal, or alternative inactivation
mechanisms. Additionally, leaking of the enzyme from the
supporting matrix could modify the protein’s activity.
Considering eventual physical changes of the biocatalyst
exposed to the ball-mill, the assays were also performed with
previously ground N435, aiming to understand the
inactivation mechanism of CALB on both N435 and in the
ground biocatalyst.
Figure 1. Residual activity of free CALB incubated at 65°C, 75°C and 85°C,
showing a typical first order decay.
Complementary thermal inactivation studies were then
carried out using N435 and previously ground N435,
following a similar protocol than for the free CALB (see SI,
Sections 2 and 3). Nevertheless, a key observation was
made. The initial reaction rate of ground N435 substantially
increased, reaching the highest value (up to one order of
magnitude) when N435 was ground for 90 minutes (Figure
Results and discussion
The overall objective of the present work was to investigate
potential inactivation mechanisms of N435 under grinding
conditions in a ball mill. In this regard, it was deemed
essential to understand the corresponding inactivation
process on soluble (free) CALB enzyme in order to determine
the consequences of immobilization on CALB stability, and
afterwards, the consequences that milling might have on the
performance of the biocatalyst. Additionally, considering
current theories formulated for heat distribution in
mechanochemical methods (i.e., the hot spot theory),[39] it
was decided to confirm or discard thermally induced
denaturation as a source of enzyme inactivation.
2
). The increase of enzyme activity due to the milling process
may be associated to diffusional effects in N435 that are
overcome when the particle size of the biocatalyst is
reduced. This leads to a higher effective concentration of
available enzyme that increases the efficiency factor of the
enzymatic reaction by approximately one order of
magnitude. Thus, a thermal inactivation study was carried
out using immobilized CALB; we found that N435 also
exhibits a complex thermal inactivation mechanism following
the same two-stage pattern exhibit by the free CALB that
might be also affected by the diffusional effects present in the
immobilized CALB (Figure 3).
Free CALB (Lipozyme, Novozymes A/S Denmark) activity
was assayed by means of a kinetic protocol using p-
This article is protected by copyright. All rights reserved.

10.1002/cctc.201901714
ChemCatChem
FULL PAPER
denaturation rate. Thus, N435 was ground in an agate
milling-jar using one agate ball (For additional details see the
Experimental Section). Residual activity of the ground
supported biocatalyst (Figure 4) shows an apparently higher
stability of the enzyme relative to the previous studies carried
out employing free CALB and N435 (except for experiments
carried out at 85°C), suggesting a high stability of N435 to
the harsh conditions of the mechanical ball milling procedure,
and indicating that the difference in particle size and the
concomitant increase in enzyme activity due to the grinding
process resulted in an apparent higher thermal stability of
CALB in ground N435.
Figure 2. Enzyme activity as function of milling time [0.1mg of N435 or ground
N435 / mL], using agate milling-jar and ball. Bar at 90 minutes of milling (Grey)
shows the increase in activity of one order of magnitude due to the milling
process
Nevertheless, it should be noted that CALB in N435 is more
stable than in its free form as demonstrated by the higher
residual activities found in the 65 °C to 75 °C temperature
range (compare Figure 3 with Figure 1). Indeed, thermal
stability presented by the immobilized biocatalyst can be
clearly appreciated at moderate temperatures (60 °C and 65
°C). It can also be observed that thermal stability provided by
the enzyme’s support becomes insufficient when the
experiment is carried out at higher temperatures, as shown
in the inactivation curve at 85 °C in Figure 3. This
observation suggests that an abrupt inactivation of
immobilized enzyme can take place upon exposure to high
temperatures.
Figure 4. Thermal inactivation assay of previously ground N435 [0.1mg of
ground N435 / mL] at 60°C, 65°C, 75°C and 85°C, showing a typical first order
decay.
Moreover, potential leaking of CALB from the milled support
was evaluated; nevertheless, no leaked protein was
detected. Thus, it can be pointed out that the enzyme´s
inactivation can be induced by thermal energy as indicated
in Figures 1, 3 and 4; however, immobilization provides
substantial resistance to the harmful thermal effects;
additionally, enzyme stability seems not to be affected by the
mechanical process as indicated in previous reports.[33]
Ground biocatalyst demonstrates an important diffusional
limitation of N435 for this reaction which is demonstrated by
the increase in the observed initial reaction rate. Indeed, it
appears that mechanical stress overcomes this limitation and
increases the biocatalyst efficiency. As it can be appreciated
in Figure 2, the apparent enzyme activity of the ground N435
milled for 90 minutes is one order of magnitude higher than
the apparent enzyme activity of N435 (compare the red and
grey bars in Figure 2). In this regard, grinding apparently
leads to an increase in the apparent activity of N435 (0.1mg
N435/mL) from 7.4 IU/mL to 98.4 IU/mL on the previously
ground biocatalyst (0.1mg of ground N435 / mL).
Figure 3. Thermal inactivation assay of N435 [0.1mg of N435 / mL] at 60°C,
65°C, 75°C and 85°C, showing a typical first order decay.
Once the pattern of thermal inactivation on both free and in
N435 was established, we proceeded to evaluate the
enzyme’s inactivation using the previously ground
The recorded progressive increase in enzyme activity and in
the apparent available enzyme associated to the milling
process are in line with observations (higher enzyme activity)
made in previous mechanoenzymatic protocols, in which the
biocatalyst, knowing that diffusional limitations (Figure 2
)
may interfere in the measurement of the observed
This article is protected by copyright. All rights reserved.

10.1002/cctc.201901714
ChemCatChem
FULL PAPER
a1)
a2)
a3)
Mag = 139
b1)
Mag = 500
b2)
Mag = 10K
b3)
Mag = 237
Mag = 61
Mag = 5K
Figure 5. SEM images of commercial supported (a, Novozym 435) and ground CALB (b, 10 Hz, 90 min), showing the irregular distribution of channels in N435
[a2)]. This irregular distribution, as well as variations in particle size, decrease as the biocatalyst is ground [b3)]. A complete gallery of SEM images is presented in
SI, Section 4.
biocatalytic reaction was faster relative to procedures using
conventional heat and stirring conditions.[35-36, 38] In fact, this
explains the apparent higher stability showed by the
previously ground biocatalyst (Figure 3) when compared
with N435, as the activity of the un-milled biocatalyst is
controlled by a low diffusional rate. In this regard, it appears
that the increase of available enzyme may be the
consequence of a decrease in the diffusional barrier due to
the reduction of N435 particle size (Figure 2) as a result of
mechanical beating.[40]
Section 4) with a rather polished continuous surface where
no channels are observed (Figure 5a3
In contrast, the ground N435 shows a distribution of different
sizes and shapes after the mechanical process (Figure 5b1
)
)
varying from 0.5 µm to 20 µm diameter, in the most
homogenous milling process, presenting a higher channel
distribution (Figure 5b3). These observations demonstrate
that for this particular reaction system diffusional limitations
control the observed reaction rate in N435 as also found in
[40]
previous reports.
Nevertheless, this limitation may be
To gather information that could shed light on this proposal,
a Brunauer-Emmett-Teller (BET) analysis of the total specific
area and pore volume was carried out. As expected, no
considerable increase in specific surface area was found
after the milling process (81m2g-1 and 83m2g-1 for N435 and
ground N435 respectively, see SI, Section 4). However, it is
likely that the reduction in particle size (from 600 μm in N435
to 0.5 μm in the ground biocatalyst) induced by the milling
process is sufficient to explain the higher activity observed in
the treated biocatalyst. This phenomenon clearly facilitates
the substrate diffusion and in consequence increases the
apparent activity of the N435 after the grinding process.
Additionally, N435 and the previously ground biocatalyst
were examined through scanning electron microscopy
(SEM) as shown in Figure 5. N435 is constituted by 330 µm
to 580 µm diameter spheres (Figure 5a1, see also SI,
overcome with the ground biocatalyst.
On the other hand, even when Figures 1, 3, and 4 present
similarities, an appropriate evaluation of several inactivation
parameters such as the inactivation energy (Ea), the time
required to reduce the activity of the biocatalyst by 10% (D),
and the temperature needed to diminish D in one logarithmic
unit (z) for both labile (L) and resistant (R) forms of the free
CALB, N435 and ground N435 was necessary for a better
understanding of this process, especially in the CALB and
ground N435. The above parameters were calculated
employing the equations for nonlinear Arrhenius behavior
and are summarized in Table 1. (For details of the
calculations see SI, Section 5).[41, 42]
The dependence of the enzymatic stability with respect to the
temperature of the reaction is shown in Table 1 (cf. decrease
This article is protected by copyright. All rights reserved.

10.1002/cctc.201901714
ChemCatChem
FULL PAPER
Table 1. Inactivation parameters for free, Novozym 435® and ground Novozym 435®.
T
kR
kL
EaR
EaL
DR-value
DL-value
zR
zL
(°C)
(1/sec)
(1/sec)
(Kcal/mol)
(Kcal/mol)
(min)
(min)
(°C)
(°C)
Free CALB
65
75
85
0.414±0.014
0.258±0.048
0.158±0.048
0.063±0.000
0.078±0.003
0.092±0.004
-11.6±0.3
4.46±0.18
5.555
8.920
36.380
28.403
25.125
47.72±2.5
119.87±6.5
14.580
Novozym 435®
60
65
75
85
0.517±0.048
0.630±0.035
0.862±0.117
1.046±0.114
0.517±0.014
0.630±0.016
0.862±0.132
1.046±0.186
6.68±0.36
2.75±0.09
1.932
1.588
1.160
0.955
1.623
0.870
0.723
0.692
31.64±2.6
113.63±3.3
Ground Novozym 435®
60
65
75
85
0.138±0.014
0.217±0.037
0.486±0.043
1.058±0.208
0.244±0.020
0.268±0.027
0.633±0.036
1.567±0.274
19.25±1.22
18.33±0.12
7.249
4.609
2.056
0.945
4.104
3.730
1.580
0.640
4.11±0.3
6.72±0.6
kR and kL = rate constant for the resistant and labile component in CALB. Ea = activation energy. D = time required to decreased 10% the enzyme’s activity. z =
temperature needed to diminish D in one logarithmic unit
in D value, expect for free enzyme). Additionally, high
thermal stability (high z value) and exceptional resilience to
temperature changes (low Ea value) confirm the high stability
for N435 in contrast with the data for free CALB. On the other
hand, the thermal parameters of the previously ground
biocatalyst (Table 1) suggest a sensitizing process that could
arise from the grinding procedure that gives rise to
amorphous biocatalyst particles.
process.[46] Thus, an adequate assessment of
a
stereoselective reaction carried out under conventional
heating and stirring conditions vis-a-vis mechanical
activation could lead to an estimation of the approximate
temperature range for the process and could help to clarify
the extend of inactivation due to mechanical stress. In this
regard, we examined the efficiency of a kinetic resolution
reaction employing α-methylbenzyl alcohol as racemic
secondary alcohol and both, Novozym 435 and the ground
biocatalyst (Scheme 1).
Thermal parameters of N435 and previously ground N435
can be used to estimate the relative thermal stability of the
biocatalyst in the two different presentations. Nevertheless,
relevant experimental parameters, such as the actual
temperature during the mechanochemical process, as well
as relative biocatalyst enantioselectivity, which would
provide complementary information about the extent of
enzyme inactivation due to mechanical energy, could not be
evaluated in the present thermal study.
Indeed, temperature analysis during mechanochemical
reactions as well as the energy transfer mechanisms from
the milling-jar and balls to the reactants are cutting-edge
topics in the study of mechanochemical transformations.[43-
45] In this context, the enantiopreference of a biocatalyst (E)
in an enzymatic resolution is a temperature dependent
Scheme 1. Enzymatic kinetic resolution of -methylbenzyl alcohol.
In particular, the extent of conversion and enantioselectivity
under ball milling settings at various reaction temperatures
This article is protected by copyright. All rights reserved.

10.1002/cctc.201901714
ChemCatChem
FULL PAPER
Figure 6. Enzyme kinetics of N435 for the resolution of -methylbenzyl alcohol under conventional heat and stirring conditions.
Figure 7. Enzyme kinetics of N435 for the resolution of -methylbenzyl alcohol under mechanical activation.
was estimated. To this end, using conventional heat and
stirring conditions in hexane solution,[47] the enzymatic kinetic
resolution using N435 proceeded with high enantioselectivity
and enantiopreference (Figure 6a). It is noteworthy that, a
slight decrease in enantiomeric excess (as little as 1%)
provokes a rapid decease in enantiopreference (E) as
indicated in Figure 6b and Figure 6c. A similar behavior was
observed with previously ground N435 in the kinetic
resolution process under thermal activation (see SI, Section
6).
ball milling process, this is in accord to the aforementioned
low effect of milling over the biocatalyst´s stability. Finally,
ground biocatalyst was evaluated along with mechanical
activation in the mechanoenzymatic procedure. No
significant differences were observed relative to the same
process started with non-milled N435 (see SI, Section 6),
confirming the null effect of milling over the stability of the
enzyme.
The increasing enantiopreference observed during the
enzymatic resolution of the α-methylbenzyl alcohol under
mechanical activation (Figure 7c) at high frequencies and
using conventional heat and stirring conditions (Figure 6a),
indicates that the transferred kinetic energy in the ball milling
procedure is necessarily lower than the energy provided by
the system using thermal activation at 35°C. Therefore,
With the aim to compare the energetics of the process using
conventional heat and stirring conditions with the process
under mechanochemical activation, a similar protocol was
carried out in a ball mill, employing agate components
(milling-jars and balls) and N435, evaluating the effect of
milling frequency in the process. Surprisingly, the kinetic
resolution proceeds smoothly, even at high frequencies
comparing the kinetics in Figure 6 and 7, we can conclude
that the temperature inside the ball milling reactor is similar
to the thermal energy provided by the reaction under
conventional heat and stirring conditions between 25°C
(room temperature) and 35°C, when agate components were
(
Figure 7c), reaching a high E value (E >> 200) at long
reaction times (90 min). High enantiomeric excess for the
main product (R)- (ee > 99%) as well as the gradual
increase of enantiopurity of the non-transformed alcohol (S)-
, the extent of conversion (c) and the high E value
(enantiopreference) along the mechanoenzymatic
procedure, indicate the high stability of N435 exposed to the
2
used. With the aim to obtain
a
more efficient
1
mechanoenzymatic biocatalytic process, several reactor and
balls materials were used, in order to evaluate the effect of
the material of the milling-jar employed in the process.
This article is protected by copyright. All rights reserved.

10.1002/cctc.201901714
ChemCatChem
FULL PAPER
Surprisingly, when Teflon® milling-jar and ball were applied
in the mechanoenzymatic procedure a faster and complete
resolution was reached after 30 min of reaction without loss
in enantiopreference (Figure 8), even at long mechanical
exposure times (90 min). This is in contrast with processes
using thermal activation (see SI, Section 6)[48] in which the
enantiopreference diminished at long time exposures.
Moreover, to the best of our knowledge, this
mechanoenzymatic resolution using Teflon® components is
the most enantioselective and fastest resolution reported so
far.[33]
Figure 9. Eyring equation plot for the resolution of -Methylbenzyl alcohol
used to estimate the ∆∆퐺‡ for both process.
The results obtained so far (vide supra) demonstrate the high
stability of the N435 to milling conditions. However, as
pointed out in previous reports,[35-36] sequential milling cycles
usually reduce the activity and the enantiopreference of
mechanoenzymatic reactions. To examine this potential
drawback, commercial immobilized CALB was milled up to
four times in an agate reactor as indicate in the Experimental
section, and the resulting fine powder was tested in the
mechanoenzymatic resolution of α-methylbenzyl alcohol
(
Scheme 1).
Figure 8. Enzyme kinetics of N435 under mechanochemical conditions
using Teflon® milling-jar and ball, showing
a swift (30 min) and
enantioselective (ee > 99%, E >> 500) process.
Therefore, the kinetic resolution under thermal activation at
several temperatures in the range of 35°C to 85°C were used
to calculate the change in Gibbs free energy (∆∆퐺‡ ) for the
process using N435 and ground biocatalyst. In this regard,
Eyring transition state theory was applied by considering the
rate constants of the enzymatic process to derive a modified
equation (eq. 1) implicating a temperature dependence of
the system throughout the kinetic resolution process.[46]
ln 퐸 = −(∆∆퐺‡/푅푇) … (퐞퐪. ퟏ)
In this equation R is the ideal gas constant and ΔΔG‡
corresponds to the Gibbs free energy difference for the
activation of the two enantiomers. Such dependence on
reaction temperature can be evaluated by means of a plot of
RlnE vs 1/T that for the kinetic resolution under thermal
activation after 90 minutes of incubation allowed the
estimation of the Gibbs free energy (Figure 9).
For the kinetic resolution using N435 and ground N435
(GN435), ∆∆퐺푁‡435 = 8.41 ± 0.79 kcal/mol and ∆∆퐺퐺‡푁435
=
Figure 10. Enzyme kinetics for the resolution of -Methylbenzyl alcohol
12.20 ± 0.91 kcal/mol, respectively. These results indicate a
rather large energy difference, and consequently
after 4 grinding cycles, indicating the insignificant effect of milling on
enzyme stability.
a
substantial preference for the transitions state involving the
(R)-isomer of the alcohol when ground N435 was used. This
result agrees with previous observations and supported the
positive effect of milling over the biocatalyst,[49, 50] moreover,
helps to understand the high enantiopreference during the
mechanoenzymatic process (Figure 8).
No significant differences (P << 0.05) in the kinetics were
detected after each catalytic cycle (Figure 10, and see SI,
Section 9).
Enzyme kinetics clearly indicate no damage in the structure
of the biocatalyst as a consequence of mechanical stress, at
least at moderate frequencies (10 Hz to 25 Hz), short
This article is protected by copyright. All rights reserved.

10.1002/cctc.201901714
ChemCatChem
FULL PAPER
reaction times (up to 90 minutes) and material hardness.
That is, decreased enzyme activity and loss in
enantiopreference after the milling process in N435, must be
associated to a different kind of inactivation, probably caused
by contact with organic solvents (alcohols and ketones)
during enzyme recovery following the mechanoenzymatic
process. Indeed, enzyme inactivation due to contact with
organic solvents has been well established.[51-53] Moreover,
calculations also suggest a loss in the secondary structure of
CALB due to the use of organic solvents,[49] which can
perturb the enzyme’s active site. With the aim to study the
effect of mechanical force in other lipases, Candida antarctica
lipase A (CALA) immobilized on immobead 150, and soluble
porcine pancreatic lipase type II (PPL) were used in the kinetic
resolution process under the mechanochemical conditions
described in the general procedure. In the event, CALA shows
a decrease in enantioselectivity and enzyme activity after the
mechanochemical process, indicating a denaturation process
that is not detected when N435 is used. This result might be
the consequence of the different support matrix. Interestingly,
while PPL shows no activity in hexane solution, when the
reaction is carried out under mechanochemical conditions a
significant but not enantioselective conversion to the acylated
product was detected, indicating that the mechanical force
can induce the desired bioconversion (see SI, Section 10). This
observation paves the road for future studies of biocatalytic
processes under mechanochemical conditions.
Additionally, enantiopreference during resolution of rac-3
using ground biocatalyst is substantially higher than the
corresponding process using N435 under mechanical
activation or conventional heat and stirring conditions (EGN435
>> 500, EN435 = 595 and E80 °C = 317), leading to highly
enantiopure compounds, higher than 99% for both (S)-
3 and
(R)- ), and corroborating the aforementioned increase in
4
ΔΔG‡ when ground N435 is used (Figure 9).
Conclusions
The potential effect of mechanochemical activation on the
efficiency of biocatalytic processes was evaluated. In
particular, enzyme kinetics for commercial immobilized and
previously ground N435 exhibited a very high activity of the
biocatalyst during the milling process with no visible
detrimental effect in enzyme activity due to mechanical
stress as indicated by the high E value achieved in the
mechanoenzymatic resolutions. The calculated thermal
parameters also confirm a high mechanical stability for
Novozym 435®. The observed increase of one order of
magnitude in the enzyme activity in water as a result of the
milling process can be used to accelerate slow enzymatic
organic reactions as illustrated by the mechanoenzymatic
resolution of racemic amines. The finding of more efficient
enzymatic reactions under mechanochemical conditions can
be explained in terms of a significant decreased of diffusional
effects in the N435 that are eliminated when the supported
material is broken, inducing a high value of apparent
available enzyme of the biocatalyst without loss in
enantiodiscrimination or stability. The enhanced activity of
Finally, the finding of an increased efficiency of ground N435
as a result of the milling process can be used to improve
inefficient enzymatic organic reactions. In this regard we
decided to re-examine
a well-known slow enzymatic
procedure used in the resolution of racemic amines using
ground biocatalyst (Scheme 2).[36] In this regard, CALB
shows an extraordinary preference for the acylation of the
ground N435 is also accompanied by
a
more
enantiopreferent process as determined from the kinetics of
the resolution of α-methylbenzyl alcohol using ground
biocatalyst, affording a high enantiopure process (ee > 99%,
E >> 500) even over extended periods of time. This is in
contrast with the same process under conventional heat and
stirring conditions. Accordingly, previous reports of
enzymatic inactivation under milling conditions cannot be
ascribed to the mechanical process itself and must be
reexamined considering the specific reaction in each case.
(R)-enantiomer of racemic amine
3
,
even at high
temperatures (%eeR,80 °C = 98%);[36] nevertheless, this kinetic
resolution process is much slower than the corresponding
alcohol deracemization.
Experimental Section
1H and 13C spectra were recorded on a BRUKER DP300 (300MHz).
High resolution mass spectra were recorded on a HPLC 1100
coupled to a MSD-TOF Agilent series HR-MSTOF model 1969 A.
Chromatograms were acquired in a Dionex HPLC Ultimate 3000 with
a UV/VIS detector, with a diode array, at 210 and 254 nm. High-
Speed Ball-Milling reactions were carried out in a Retsch, Mixer Mill
(MM200). Enzyme activity was measured in a Biotek Synergy H4
Hybrid Reader. All reagents were purchased from Sigma-Aldrich
(Merck) and used as received. Free CALB was purchased from
Novozymes, Lipozyme. Immobilized CALB was purchased from
Scheme 2. Enzyme kinetics resolution of racemic chiral amine rac-3.
The positive effect in the use or ground supported CALB was
evidenced when comparing the conversion rate for the
mechanochemical procedure using Novozym 435 in solution,
achieving ca. 25% faster biocatalysis when ground
supported lipase was used in the resolution process (see SI,
Section 11).
This article is protected by copyright. All rights reserved.

10.1002/cctc.201901714
ChemCatChem
FULL PAPER
Novozymes, Novozym 435® (Immobilized on acrylic resin, IU/g >
and enantiopreference (E = ln[1-c(1+%eep)]/ln[1-c(1-%eep)]) were
calculated from the integration of peaks for each product (alcohol (R)
16.59 min, (S) 20.29 min, ester (R) 8.84 min, (S) 9.30 min). The peak
for isopropenyl acetate appears at 6.57 min.
10000), Candida antarctica lipase
A (CALA) immobilized on
immobead 150 was purchased from Sigma-Aldrich (Merck) and
soluble porcine pancreatic lipase type II was purchased from Sigma-
Aldrich (Merck). For all the activity analysis, a 200 mM solution of p-
nitrophenyl butyrate in 2-methyl-2-butanol was used as a substrate.
Specific surface area of the immobilized and milled immobilized
CALB was measured using Brunauer-Emmett-Teller (BET) analysis
(Gemini equipment II-2370, Micrometrics). Samples were degassed
at room temperature overnight in vacuum to prevent denaturation.
General method for the kinetic resolution of α-methylbenzyl
alcohol under mechanical activation.
To a ball mill equipped with one agate milling-jar (12 mm of diameter,
4.6 mL capacity) and the corresponding ball (6 mm of diameter, 480
mg of weight); or stainless-steel milling-jar (15 mm of diameter, 4.6
mL of capacity) and the corresponding ball (8 mm of diameter, 1.5 g
of weight); or Teflon® milling-jar (10 mm of diameter, 6.5 mL of
capacity) and the corresponding ball (8 mm of diameter, 1.2 g of
weight) were added 50 μL of α-Methylbenzyl alcohol, hexane (0.2
mL, HPLC grade), 41 μL of isopropenyl acetate, and 30 mg of
commercial supported CALB. Reaction samples were collected at
different times (2.5, 5, 7.5, 10, 15, 30, 45, 60 and 90 min), diluted with
1.3 mL of hexane HPLC grade, filtered and analysed by HPLC
equipped with a chiral column (AD-H, hexane: IPA 95:5, Flow 0.5
mL/min, runtime 25 min).
General method for the measurement of free enzyme activity.
The substrate (230 μL, see Experimental section) was placed in a 96
well-plate previously incubated at 45 °C (average stable
temperature), before the addition of 20 μL of the free enzyme (as
received or diluted with Tris buffer [10mM Tris HCl, pH 7, 2% Triton]).
The plate progress was monitored at 405 nm in a Synergy H4 Hybrid
reader for 20 minutes with readings in short time periods. Data were
collected and plotted in order to determine the maximum conversion
rate.
General method for the measurement of free CALB thermal
inactivation.
Acknowledgements
Free enzyme (or diluted enzyme [10 mM Tris HCl, pH 7, 2% Triton])
was placed in a 1.5 mL Eppendorf tube and incubated in a
Thermomixer at 300 rpm at various temperatures. Enzyme samples
of 20 μL were taken at different periods of time. Enzyme activity was
examined according to the General method for the measurement of
free enzyme activity.
We are indebted to fund SEP-CINVESTAV via grant 126. M.
P.-V., thanks CONACYT for PhD scholarship 70766.
Conflicts of interest
General method for the measurement of Novozym 435® and
ground Novozym 435® thermal inactivation.
There are no conflicts to declare.
To a series of Eppendorf tubes of 1.5 mL containing 300 μg of N435
or ground N435 (Agate milling-jar and ball, 10 Hz for 1.5 hours) was
added 1.4 mL of Tris buffer (10 mM Tris HCl, pH 7, 2% Triton) and
the resulting mixture was incubated in a Thermomixer at 300 rpm at
different temperatures. At the end of each incubation time, the
content of the Eppendorf tube was transferred to a well of a 6 well-
plate and treated with 1.6 mL of cold buffer. The plate was incubated
at 45 °C for 5 minutes in a Synergy H4 Hybrid reader and then 20 μL
of substrate was added. The plate was monitored at 405 nm for 20
minutes with short interval readings. Data were collected and plotted
in order to determine the maximum rate of conversion.
Keywords: Mechanoenzymatic • Biocatalyst • CALB •
Mechanostability • Sustainable
[1] U. Weißbach, S. Dabral, L. Konnert, C. Bolm, J. G. Hernández, Beilstein
J. Org. Chem. 2017, 13, 1788-1795.
[2] E. E. Schultz, N. R. Braffman, M. U. Luescher, H. H. Hager, E. P.
Balskus, Angew. Chem., Int. Ed. 2019, 58, 3151-3155.
[3] J. Manning, M. Tavanti, J. L. Porter, N. Kress, S. P. De Visser, N. J.
Turner, S. L. Flitsch, Angew. Chem., Int. Ed. 2019, 58, 5668-5671.
[4] C.-H. Wong, G. M. Whitesides, Enzymes in Synthetic Organic Chemistry,
Pergamon, Oxford, 1994.
[5] R. A. Sheldon; D. Brady. ChemSusChem. 2019, 12, 2859-2881.
[6] R. A. Sheldon, P. C. Pereira, Chem. Soc. Rev. 2017, 46, 2678-2691.
[7] K. Rosenthal, S. Lütz, Curr. Opin. Green Sustain. Chem. 2018, 11, 58-
64.
General method for the kinetic resolution of α-methylbenzyl
alcohol in solution.
[8] D. C. Blakemore, L. Castro, I. Churcher, D. C. Rees, A. W. Thomas, D.
M. Wilson, A. Wood, Nat. Chem. 2018, 10, 383-394.
To a solution of 500 μL of α-Methylbenzyl alcohol, hexane (10 mL)
and isopropenyl acetate (450 μL), previously incubated at the desired
temperature was added 300 mg of commercial supported CALB.
Samples of 40 μL were taken in the initial time and after 2.5, 5, 7.5,
10, 15, 30, 45, 60 and 90 min, and were diluted with 1.5 mL of hexane
HPLC grade. The resulting solution was filtered and analyzed by
HPLC equipped with a chiral column. Both products (enantioenriched
alcohol and acylated product) were analyzed simultaneously in an
AD-H chiral column (hexane: IPA 95:5, Flow 0.5 mL/min, runtime 25
min). Enantiomeric excess (ee), conversion (c = %ees/(%ees+%eep))
[9] R. A. Sheldon, D. Brady, Chem. Commun. 2018, 54, 6088-6104.
[10] F. H. Arnold, Angew. Chem., Int. Ed. 2018, 57, 4143-4148.
[11] S. Jemli, D. Ayadi-Zouari, H. B. Hlima, S. Bejar, Crit. Rev. Biotechnol.
2016, 36, 246-258.
[12] F. Parmeggiani, N. J. Weise, S. T. Ahmed, N. J. Turner, Chem. Rev.
2018, 118, 73-118.
[13] F. Rudroff, M. D. Mihovilovic, H. Gröger, R. Snajdrova, H. Iding, U. T.
Bornscheuer, Nat. Catal. 2018, 1, 12-22.
[14] N. Doukyu, H. Ogino, Biochem. Eng. J. 2010, 48, 270-282.
[15] A. Kumar, R. A. Gross, Biomacromolecules 2000, 1, 133-138.
[16] M. T. Reetz, J. Am. Chem. Soc. 2013, 135, 12480-12496.
This article is protected by copyright. All rights reserved.

10.1002/cctc.201901714
ChemCatChem
FULL PAPER
[17] P. V. Iyer, L. Ananthanarayan, Process Biochem. 2008, 43, 1019-1032.
[18] V. Stepankova, S. Bidmanova, T. Koudelakova, Z. Prokop, R.
Chaloupkova, J. Damborsky, ACS Catal. 2013, 3, 2823-2836.
[19] M. Yoshimoto, Y. Miyazaki, Y. Kudo, K. Fukunaga, K. Nakao,
Biotechnol. Prog. 2006, 22, 704-709.
[51] C. Laane, S. Boeren, K. Vos, C. Veeger, Biotechnol. Bioeng. 1987, 30,
81-87.
[52] Ö. Köse, M. Tüter, H. A. Aksoy, Bioresour. Technol. 2002, 83, 125-129.
[53] A. Kumar, K. Dhar, S. S. Kanwar, P. K. Arora, Biol. Proced. Online 2016
,
18, DOI: 10.1186/s12575-0.
[20] E. M. Nordwald, J. L. Kaar, Biotechnol. Bioeng. 2013, 110, 2352-2360.
[21] F. M. Veronese, R. Largajolli, E. Boccu, C. A. Benassi, O. Schiavon,
Appl. Biochem. Biotechnol. 1985, 11, 141-152.
[22] R. A. Sheldon, Adv. Synth. Catal. 2007, 349, 1289-1307.
[23] R. C. Rodrigues, C. Ortiz, Á. Berenguer-Murcia, R. Torres, R.
Fernández-Lafuente, Chem. Soc. Rev. 2013, 42, 6290-6307.
[24] M. P. Thompson, I. Peñafiel, S. C. Cosgrove, N. J. Turner, Org. Process
Res. Dev. 2019, 23, 9-18.
[25] R. DiCosimo, J. McAuliffe, A. J. Poulose, G. Bohlmann, Chem. Soc.
Rev. 2013, 42, 6437-6474.
[26] S. S. Ribeiro, C. Raminelli, A. L. M. Porto, J. Fluorine Chem. 2013, 154,
53-59.
[27] N. Zhang, W. C. Suen, W. Windsor, L. Xiao, V. Madison, A. Zaks,
Protein Eng. 2003, 16, 599-605.
[28] B. Réjasse, T. Besson, M.-D. Legoy, S. Lamare, Org. Biomol. Chem.
2006, 4, 3703-3707.
[29] J. G. Hernández, E. Juaristi, J. Org. Chem. 2010, 75, 7107-7111.
[30] J. G. Hernández, E. Juaristi, Chem. Commun. 2012, 48, 5396-5409.
[31] L. A. Polindara-García, E. Juaristi, Eur. J. Org. Chem. 2016, 17, 1095-
1102.
[32] J. M. Landeros, E. Juaristi, Eur. J. Org. Chem. 2017, 18, 687-694.
[33] J. G. Hernández, M. Frings, C. Bolm, ChemCatChem. 2016, 8, 1769-
1772.
[34] See also: K. J. Ardila-Fierro, D. E. Crawford, A. Körner, S. L. James, C.
Bolm, J. G. Hernández, Green Chem. 2018, 20, 1262-1269.
[35] M. Pérez-Venegas, G. Reyes-Rangel, A. Neri, J. Escalante, E. Juaristi,
Beilstein J. Org. Chem. 2017, 13, 1728-1734. [35] M. Pérez-Venegas, G.
Reyes-Rangel, A. Neri, J. Escalante, E. Juaristi, Beilstein J. Org. Chem.
2017, 13, 2128-2130. (Corrigendum).
[36] M. Pérez-Venegas, E. Juaristi, Tetrahedron 2018, 74, 6453-6458.
[37] F. Hammerer, L. Loots, J.-L. Do, J. P. D. Therien, C. W. Nickels, T.
Friščić, K. Auclair, Angew. Chem. 2018, 130, 2651-2654.
[38] J. P. D. Therien, F. Hammerer, T. Friščić, K. Auclair, ChemSusChem
2019, 12, 3481-3490.
[39] J. Andersen, J. Mack, Green Chem. 2018, 20, 1435-1443, and
references therein.
[40] M. J. J. Litjens, K, Q. Le, A. J. J. Straathof, J. A. Jongejan, J. J. Heijnen,
Biocatal Biotransformation 2000, 19, 1-19.
[41] H. Y. Yamamoto, M. P. Steinberg, A. I. Nelson, J. Food. Sci. 1962, 27,
113-119.
[42] A. C. Ling, D. B. Lund, J. Food. Sci. 1978, 43, 1307-1310.
[43] K. C. McKissic, J. T. Caruso, R. G. Blair, J. Mack, Green Chem. 2014
16, 1628-1632.
,
[44] J. M. Andersen, J. Mack, Chem. Sci. 2017, 8, 5447-5453.
[45] J. Andersen, J. Mack, Angew. Chem. Int. Ed. 2018, 57, 13062-13065.
[46] P. L. A. Overbeeke, J. Ottosson, K. Hult, J. A. Jongejan, J. A. Duine,
Biocatal. Biotransformation 1999, 17, 61-79.
[47] H. V. Ferreira, L. C. Rocha, R. P. Severino, A. L. M. Porto, Molecules
2012, 8, 8955-8967.
[48] A more efficient process can be obtained when different component’s
materials were used. For example, stainless steel and Teflon® (PTFE)
milling-jars and balls led to considerably improved results. For details and
graphs see SI, Section 7.
[49] C. Li, T. Tan, H. Zhang, W. Feng, Biol. Chem. 2010, 285, 28434-28441.
[50] A similar procedure can be used to establish the inversion temperature
(Tinv), that is the temperature at which the E value has an inflection point in
the enzymatic process. In this way, Tinv for the N435 (Tinv = 65°C) and the
previously ground N435 (Tinv = 55°C) were calculated (see SI, Section 6).
Additionally, a different substrate was tested in order to obtain a better
enzyme kinetics and evaluate free Gibbs energy in the mechanoenzymatic
resolution (see SI, Section 8).
This article is protected by copyright. All rights reserved.

10.1002/cctc.201901714
ChemCatChem
FULL PAPER
FULL PAPER
M. Pérez-Venegas,* M. M. Tellez-Cruz,
O. Solorza-Feria, A. López-Munguía, E.
Castillo,* E. Juaristi.*
Page No. – Page No.
Thermal and mechanical stability of
immobilized Candida antarctica
Lipase B: an approximation to
mechanochemical energetics in
enzyme catalysis.
For the first time, the effect of grinding over enzyme activity and stability was
systematically studied. The results indicate a high enzyme stability to the
mechanochemical conditions and, unexpectedly, and increment in enzyme activity of
one order of magnitude after the mechanochemical process without loss in
enantiopreference of the enzyme.
This article is protected by copyright. All rights reserved.