bubble column enables higher reaction rate for
TRANSCRIPT
General rights Copyright and moral rights for the publications made accessible in the public portal are retained by the authors and/or other copyright owners and it is a condition of accessing publications that users recognise and abide by the legal requirements associated with these rights.
Users may download and print one copy of any publication from the public portal for the purpose of private study or research.
You may not further distribute the material or use it for any profit-making activity or commercial gain
You may freely distribute the URL identifying the publication in the public portal If you believe that this document breaches copyright please contact us providing details, and we will remove access to the work immediately and investigate your claim.
Downloaded from orbit.dtu.dk on: Jan 07, 2022
Bubble Column Enables Higher Reaction Rate for Deracemization of (R,S)1Phenylethanol with Coupled Alcohol Dehydrogenase/NADH Oxidase System
Dias Gomes, Mafalda; Bommarius, Bettina; Anderson, Shelby; Feske, Brent D.; Woodley, John;Bommarius, Andreas
Published in:Advanced Synthesis and Catalysis
Link to article, DOI:10.1002/adsc.201900213
Publication date:2019
Document VersionPeer reviewed version
Link back to DTU Orbit
Citation (APA):Dias Gomes, M., Bommarius, B., Anderson, S., Feske, B. D., Woodley, J., & Bommarius, A. (2019). BubbleColumn Enables Higher Reaction Rate for Deracemization of (R,S)1Phenylethanol with Coupled AlcoholDehydrogenase/NADH Oxidase System. Advanced Synthesis and Catalysis, 361(11), 2574-2581.https://doi.org/10.1002/adsc.201900213
Title: Bubble Column Enables Higher Reaction Rate forDeracemization of (R,S)-1-Phenylethanol with Coupled AlcoholDehydrogenase/NADH Oxidase System
Authors: Mafalda Dias Gomes, Bettina Bommarius, Shelby Anderson,Brent D. Feske, John Woodley, and Andreas Bommarius
This manuscript has been accepted after peer review and appears as anAccepted Article online prior to editing, proofing, and formal publicationof the final Version of Record (VoR). This work is currently citable byusing the Digital Object Identifier (DOI) given below. The VoR will bepublished online in Early View as soon as possible and may be differentto this Accepted Article as a result of editing. Readers should obtainthe VoR from the journal website shown below when it is publishedto ensure accuracy of information. The authors are responsible for thecontent of this Accepted Article.
To be cited as: Adv. Synth. Catal. 10.1002/adsc.201900213
Link to VoR: http://dx.doi.org/10.1002/adsc.201900213
1
VERY IMPORTANT PUBLICATION FULL PAPER
DOI: 10.1002/adsc.201((will be filled in by the editorial staff))
Bubble Column Enables Higher Reaction Rate for
Deracemization of (R,S)-1-Phenylethanol with Coupled Alcohol
Dehydrogenase/NADH Oxidase System
Mafalda Dias Gomes†,a, Bettina R. Bommarius†,b, Shelby R. Anderson b, Brent D.
Feskec John M. Woodley a* and Andreas S. Bommarius b*
a Department of Chemical and Biochemical Engineering, Technical University of Denmark, Building 229, Søltofts Plads,
DK-2800 Kgs. Lyngby, Denmark
+45 45 25 28 85
b School of Chemical and Biomolecular Engineering, Krone Engineered Biosystems Building, Georgia Institute of
Technology, 950 Atlantic Drive N.W., Atlanta, GA 30332, USA
+1 (404) 385-1334
c Department of Chemistry and Biochemistry, Georgia Southern University, Science Center Suite 1505, 11935 Abercorn
St., Savannah, GA 31419
+1 (912) 344-3210
†: contributed equally
Received: ((will be filled in by the editorial staff))
Supporting information for this article is available on the WWW under http://dx.doi.org/10.1002/adsc.201######.
Abstract: One of the major drawbacks for many biocatalysts is their poor stability under industrial process conditions. A particularly interesting example is the supply of oxygen to biooxidation reactions, catalyzed by oxidases, oxygenases or alcohol dehydrogenases coupled with NAD(P)H (reduced nicotinamide adenine dinucleotide phosphate) oxidases, which all require the continuous supply of molecular oxygen as an oxidant or electron acceptor. Commonly, oxygen is supplied to the bioreactor by air sparging. To ensure sufficient oxygen transfer from the gas to the liquid phase, stirring is essential to disperse the gas bubbles and create high gas-liquid interfacial area. Studies indicate that the presence of gas-liquid interface induces enzyme deactivation by protein unfolding which then readily aggregates and can subsequently precipitate. This contribution has examined the effects of stirring and the presence of gas-liquid interface on the kinetic stability of water-forming NAD(P)H oxidase (NOX) (EC1.6.3.2).
These effects were studied separately and a bubble column apparatus was successfully employed to investigate the influence of gas-liquid interfaces on enzyme stability. Results showed that NOX deactivation increases in proportion to the gas-liquid interfacial area. While air enhances the rate of stability loss compared to nitrogen, stirring causes faster loss of activity in comparison to a bubble column. Finally, deracemization of 1-phenylethanol, using a coupled alcohol dehydrogenase /NADH oxidase system (ADH/NOX), proceeded with a higher rate in the bubble column than in quiescent or in a stirred solution, although, inactivation was also accelerated in the bubble column over a quiescent solution.
Keywords: Bubble column; NADH oxidase; deracemization; gas-liquid interface; biocatalysis
10.1002/adsc.201900213
Acc
epte
d M
anus
crip
t
Advanced Synthesis & Catalysis
This article is protected by copyright. All rights reserved.
2
Introduction
The integration of biocatalysis into synthetic
production schemes for pharmaceuticals and fine
chemicals continues to make significant progress.
Whilst an enzyme’s ability to catalyze useful
reactions under mild conditions is of great importance,
it is their excellent selectivity that draws attention for
wider applications in chemical synthesis. [1] [1a].
Additionally, biocatalysis offers economic and
environmental advantages over chemical routes,
contributing to the development of more sustainable
industrial processes. [2] [3]. However, despite the many
benefits of using biocatalysts, their stability under
relevant process conditions is frequently cited as a
drawback to the wider usage of enzymatic catalysis [4].
The oxidoreductases are one class of enzymes that
has become increasingly relevant in recent years. [5] [6] These enzymes can catalyze oxidation reactions to
produce molecules of considerable commercial
interest, that are otherwise difficult to obtain in a
single step using non-enzymatic methods. [7] [8]. [9] In
particular, oxidation of alcohols to ketones or
aldehydes, essential to produce many building blocks
for valuable chemicals, can be catalyzed using either
alcohol oxidases (AOXs) or alcohol dehydrogenases
(ADHs). The latter are better known operating in the
reductive direction [10]. Nevertheless, in the oxidative
mode, the ADHs have been more commonly used
than AOXs, since there are more discovered enzymes
with activity and selectivity towards molecules of
interest.[11] [12] Nevertheless, in the oxidative direction,
ADHs require NAD(P)+ as an electron acceptor,
which is an expensive cofactor and therefore needs to
be regenerated to achieve an economically feasible
process. [13]. One of the most interesting solutions to
this problem is the use of water-forming NAD(P)H
oxidases, NOX2 (herein termed NOX). Such
enzymes oxidize NAD(P)H to NAD(P)+ while
concomitantly reducing molecular oxygen to water
(Scheme 1). NOX2 enzymes from Lactobacillus
sanfranciscensis, [14] [15] Lactococcus lactis [15],
Lactobacillus pentosus, [16] Streptococcus mutans [17]
and Lactobacillus plantarum [18] have been
characterized (for a review, see [13]). Ultimately,
therefore both types of oxidative system, AOX and
ADH/NOX, need molecular oxygen as an oxidant in
stoichiometric amounts (0.5 mol O2/ mol product).
This is highly attractive for industrial manufacturing
processes since it avoids the use of harmful metal
oxidants. However, it also comes with some
limitations, especially with respect to enzyme
stability under the target process conditions.
Scheme 1. Enzymatic oxidation of racemic 1-
phenylethanol using a coupled ADH/NOX system.
Since oxygen has a particularly low solubility in
water (where the reaction occurs) it cannot be
dissolved in solution prior to the reaction as would be
the case with other reactants, but instead it is
necessary to continuously supply it to the reaction
medium. Usually, this is done by bubbling (sparging)
air into the liquid, in a well agitated reactor. Stirring
is essential to break the gas into small bubbles,
increasing their surface area and thereby their
residence time in the reactor in order to ensure
sufficient oxygen transfer from the gas to the liquid
phase, where reaction occurs. To achieve industrially
relevant productivities [2], [19] fast oxygen transfer
from the gas to the liquid phase is needed. Therefore,
to obtain a high mass transfer coefficient (kla) a large
gas-liquid interfacial area is needed. However, it has
been shown that the presence of gas-liquid interface
can deactivate some enzymes. For instance,
Donaldson et al. (1980) [20] observed that acid
phosphatase deactivates in the presence of air-liquid
interfaces and that the rate of deactivation is
dependent upon the air-liquid contact time in a
laminar flow reactor. Later, it was found that
lysozyme deactivates when in contact with air-liquid
interfaces and also in the presence of gaseous
nitrogen. [21]. Around the same time, Patil et al.
(2000) and Mohanty et al.(2001) [22] [23] have also
shown that lipases deactivate in the presence of
surface aeration. Likewise, Bommarius & Karau
(2005) [24] found that formate dehydrogenase
deactivates at the gas-liquid interface (using both air
and gaseous oxygen). More recently, Findrik et al.
(2014) [25] revealed that the increase of aeration in a
stirred tank reactor decreased the stability of D-amino
acid oxidases. Finally, recent work developed in
shake flasks demonstrates that cellulases also
deactivate at the air-liquid interface. [26]. Nevertheless,
not all enzymes appear to be affected by the interface.
For example, Toftgaard Pedersen et al. (2015) [27]
showed that galactose oxidase did not lose any
activity when exposed to an air-liquid interface in a
stirred tank reactor.
Since sparging air into a stirred tank reactor is still
the most efficient and economical way of supplying
oxygen to biooxidations, the presence of a gas-liquid
10.1002/adsc.201900213
Acc
epte
d M
anus
crip
t
Advanced Synthesis & Catalysis
This article is protected by copyright. All rights reserved.
3
interface is unavoidable. As mentioned previously,
the stability of the biocatalyst is crucial for successful
application of biocatalytic oxidations in industry.
Therefore, the reasons for protein inactivation in the
presence of such interfaces needs to be better
understood. This is especially important for oxidases,
since oxygen is also a co-substrate in the reaction.
Additionally, methods that allow the study of the
deactivation of these proteins in the presence of gas-
liquid interfaces also need to be developed.
In order to explore the effect of gas-liquid
interfacial area on the stability of oxygen-dependent
enzymes, a bubble column was built (Figure 1) based
on previously published work by Bommarius &
Karau (2005) [24] where, inter alia, the deactivation of
formate dehydrogenase (FDH) at gas-liquid interfaces
was studied. The bubble column design was
reproduced and used specifically for NOXs. Aside
from using such a device with a controlled interfacial
area for investigations of the effect of the gas-liquid
interface on enzymes, we also reasoned that the
gentler conditions (in the absence of stirring) might
allow enzymes which would otherwise be sensitive to
such interfaces to be used effectively. Such an
equipment therefore represents a compromise, where
bubbles are formed to allow sufficient mass transfer,
whilst still avoiding mixing and therefore an excess
of interface which might otherwise deactivate the
enzymes. To test this concept, in the current
contribution we have tested an ADH/NOX system in
a bubble column.
Figure 1. Diagram of the bubble column setup. Inset:
image of bubbles during representative experiments.
The present work investigates the influence of air-
liquid interface on the stability of water-forming
NOXs from Lactobacillus plantarum (NoxV), [18]
both for NOX alone as well as in the context of
deracemization of a racemic secondary alcohol
catalyzed by ADH/NOX. The ADHs chosen belong
to the SDR (Short-chain Dehydrogenases/Reductase)
superfamily [28] and catalyze the respective
enantiomers of the chosen substrate 1-phenylethanol,
(R)-ADH from Lactobacillus brevis and (S)-ADH
from Bacillus subtilis. The (R)-ADH has been
employed frequently for enantioselective alcohol
production [29] including by some of us for oxidative
deracemization. [14] Aqueous solutions of both
enzymes were sparged with air (21% oxygen)
through a bubble column. As the presence of oxygen
is known to over-oxidize some amino acid residues in
oxidases, [30] the stability of these enzymes was also
tested in the presence of nitrogen gas (i.e. in the
absence of oxygen). Control experiments in quiescent
and in stirred solutions were also carried out to
account for time-dependent protein deactivation.
Furthermore, the causes of enzyme deactivation in
the presence of a gas-liquid interface were addressed
and discussed.
Results and Discussion
Influence of reaction environment on the
stability of NOX and ADH
With the purpose of achieving enzymatic
deracemization of (R/S)-phenylethanol employing the
ADH/NOX system, and since the supply of molecular
oxygen is required for the reaction to occur, we tested
the influence of gas-liquid interface on the kinetic
stability of NOX. Several control experiments were
performed to compare three distinct reaction
environments: (1) quiescent liquid (2) agitated/stirred
liquid and (3) sparged liquid. As observed in Figure 2,
the NOX deactivation kinetics were close to first-
order. Accordingly, a deactivation rate constant (kd)
was calculated from the slope of the natural logarithm
of the residual activity over time. [31]. Deactivation
rate constants kd and enzyme half-lives t1/2 are
presented in Table 1.
Figure 2. Residual activity of NOX as a function of time.
10.1002/adsc.201900213
Acc
epte
d M
anus
crip
t
Advanced Synthesis & Catalysis
This article is protected by copyright. All rights reserved.
4
NOX was incubated in quiescent conditions in the
presence of low interfacial areas of pure nitrogen and
air (21 % O2; 79 % N2), by filling the headspace of
the vials independently with each of the two gases.
NOX was significantly more stable (i.e. longer half-
life) in the absence than in the presence of oxygen,
with half-lives of 53 and 21 h, respectively (Table 1
and Supporting Information, Table S1). Furthermore,
NOX kinetic stability was tested in stirred solution in
the presence of pure nitrogen and air. Under nitrogen,
the NOX half-life was found to be shorter compared
to quiescent conditions, 40 vs 53 h, indicating that
mixing increases NOX deactivation (Supporting
Information, Table S1). In aerated, stirred solution,
NOX deactivation was almost doubled compared to
quiescent solution (t1/2 of 12 vs 21 h, Table 1).
Table 1 Deactivation of NOX in the presence of air under
three reaction environments, with and without
(R,S)-1-phenylethanol (substrate) present.
Reaction a) kd b) (h-1) t 1/2 (h) c)
Quiescent 0.033 21
Stirred 0.06 12
Sparged
Quiescent, substrate
Stirred, substrate
Sparged, substrate
0.031
0.032
0.021
0.05
22
22
32
13 a) reaction conditions for NOX, pH 7, 25 oC. b) rate constant of
enzyme deactivation c) enzyme half-life (t1/2 = ln 2/kd).
Figure 3: A Deactivation of NOX in the presence of either (R) or (S)-ADH during conversion of 1-phenylethanol. B
Corresponding HPLC analysis of consumed (R,S)-1-phenylethanol. C Deactivation of NOX in the presence of both (R)
and (S)-ADH during conversion of 1-phenylethanol. D Corresponding HPLC analysis of consumed 1-phenylethanol. All
experiments were performed with sparged air. All measurements for the experiments were done in duplicates with a
standard deviation of 1.5 %. The experiment with both ADH in C,D was repeated twice with similar results within the time
frame of 6 hours.
10.1002/adsc.201900213
Acc
epte
d M
anus
crip
t
Advanced Synthesis & Catalysis
This article is protected by copyright. All rights reserved.
5
Therefore, to eliminate the effect of mixing and to
investigate the effect of the gas-liquid interface alone,
we incubated NOX in the bubble column sparged
with air or N2. The enzyme was exposed to larger
interfacial areas with increasing incubation times.
Sparging with air, we observed that NOX was
significantly more stable in the bubble column than in
stirred conditions, indeed, as stable as in quiescent
solution (Table 1). Sparging with nitrogen, we
observed a slightly reduced half-life compared to
quiescent solution but enhanced stability compared to
stirred solution (Supporting Information, Table S1).
During substrate deracemization in the bubble
column, the low solubility of the product
acetophenone resulted in its precipitation, which first
decreased bubble size and then clogged the needle,
both of which decreased NOX stability (last data
point in Table 1). In contrast, incubation of NOX
with the ADH substrate 1-phenylethanol under
quiescent conditions with air was found to have no
effect on the kinetic stability of the enzyme (Table 1).
NOX was found to be most stable in a quiescent
liquid followed by a sparged liquid and, finally, in an
agitated liquid. While not relevant for the ADH/NOX
system which requires the presence of oxygen to
function, NOX was more stable in the presence of
nitrogen than in air (Table 1 and Supporting
Information, Table S1). Importantly, these results
therefore support the idea of using a bubble column
for the ADH/NOX system, since it avoids severe
agitation.
In comparing NOX stability in the bubble column
with quiescent conditions over the same time scale,
we see that the deactivation of NOX in the bubble
column is dependent on the gas-liquid interfacial area
(Supporting information, Figure S2), and not on the
incubation time, in accordance with previous
observations with formate dehydrogenase (FDH). [24].
To account for ADH stability, both enzymes were
incubated in sparged, stirred, and quiescent
conditions; similar deactivation behavior was
observed (Supporting Information, Table S1). All
experiments were performed at pH 7 since the
coupled reaction with ADH/NOX for deracemization
of 1-phenylethanol is most favorable at neutral pH 7. [14] [18]
ADH/NOX reactions in the bubble column
To demonstrate suitability of NOX for
deracemization reactions we chose to compare
phenylethanol oxidation catalyzed by ADH/NOX in a
bubble column, quiescent solution, and in an agitated/
stirred solution. ADH was chosen as the model
enzyme since both (R)-ADH and (S)-ADH were
readily available. The first two deracemization
experiments were performed using each ADH in
combination with NOX to achieve 50 % conversion
of the racemic substrate, 1-phenylethanol, at 50 mM
racemate.
Results depicted in Figure 3 reveal that use of the
bubble column enabled a reaction rate of 0.12
mM/min for (R)-ADH (Panel 3B), almost 10-fold the
rate in quiescent solution for the first 3 hours. It is
important to note that whereas the (R)-ADH is
enantioselective and the conversion stopped at 50 %
when using this ADH alone, the not fully
characterized (S)-ADH seems to be less selective, as
the conversion continued to about 40 % for the
sparged reaction and 22 % for quiescent conditions
(Panel 3B). Enantiomeric excess will be evaluated in
the future. When both (R)-ADH and (S)-ADH were
combined with NOX in the bubble column (Panel
3D), the reaction in the bubble column was almost
complete after 6 hours with a reaction rate of 0.19
mM/min for the first 2 hours, 6-fold higher than the
quiescent and twice as high as the stirred controls
(0.03 and 0.1 mM/min, respectively).
ADH/NOX-catalyzed deracemization of (R/S)-
phenylethanol proceeded 6-10 times faster in a
bubble column sparged with air than in air-saturated
quiescent solution; in a gently stirred solution
(Reynolds number Re 33; Re = Nd2/, with
density = 1015 kg/m3, rotational velocity of the
stirrer N = 2 s-1, stirrer diameter d of 5 mm, and
viscosity = 1.0510-3 kg/(ms)), the reaction
proceeded about twice as fast as in quiescent solution.
These advantages of sparged and stirred systems have
to be weighed against enhanced enzyme deactivation.
Incubation experiments of NOX, carried out in three
different reaction environments, showed that NOX is
indeed sensitive towards gas-liquid interfaces (Panels
3A and 3C). Results also reveal that the presence of
oxygen in the gas phase enhances the deactivation
rate of the enzyme (compared to solely nitrogen).
Furthermore, it was observed that NOX is more
stable in quiescent solution than in a sparged liquid,
which in turn was found to have a longer half-life
than NOX in stirred solution. Thus mixing results in a
higher deactivation rate constant than the presence of
gas-liquid interfaces. The ADH kinetic stability was
also tested and was found to follow the same trend as
NOX.
While both NOX and ADHs are most stable in
quiescent solution, the air-liquid interfacial area is
smaller in such a system than in sparged solution in
the bubble column. However, the smaller interfacial
area results in a lower oxygen transfer rate (OTR)
from the gas to the liquid phase. Then, under
quiescent conditions, oxygen transfer can limit the
reaction rate and thus volumetric productivity (space-
time yield). Although the enzyme stability may be
compromised in the bubble column, this set-up
10.1002/adsc.201900213
Acc
epte
d M
anus
crip
t
Advanced Synthesis & Catalysis
This article is protected by copyright. All rights reserved.
6
facilitates oxygen transfer, resulting in higher
reaction rates for oxygen-dependent reactions.
To demonstrate the positive effect of continuously
supplied air-liquid interface on the reaction rate of
ADH/NOX catalyzed deracemization of 1-
phenylethanol this reaction was performed in the
bubble column and in a quiescent solution.
Conclusion
For ADH/NOX-catalyzed deracemization of alcohols,
a bubble column with sparged air yields a higher
reaction rate than a quiescent solution but also
encounters a higher rate of enzyme deactivation.
Nevertheless, the half-life of NOX in the bubble
column was sufficient to achieve complete
conversion of 50 mM phenylethanol for the coupled
reaction with ADH. Therefore, the bubble column is
an attractive apparatus to run enzymatic oxidation
reactions using coupled ADH/NOX systems.
As had been observed for formate
dehydrogenase (FDH), deactivation of NOX in a
bubble column depends on interfacial area, not on
time. Sparging results in slower deactivation than in
stirred solution. Sparging or stirring in presence of
air rather than nitrogen enhances deactivation. These
trends were also followed by the alcohol
dehydrogenases employed in this study (details in the
supporting information).
Experimental Section
Materials
Potassium phosphate buffer was used for all pH 7
solutions. All chemicals used for protein purification
were of ACS chemical grade. DTT (Goldbio, St.
Louis, MO) was used for NOX purification. Lysates
and pure NOX from L. plantarum (NoxV) were
prepared and modified according to Park et al. (2011). [18] NADH was purchased from VWR (Radnor, PA,
USA).
Enzyme purification
NoxV in pET28a was expressed using BL21 plys
cells and overnight express media (VWR, Radnor,
PA, USA). Cells were harvested at 4000 g (Beckman,
Brea, CA) for 20 min and then lysed in 50 mM
sodium acetate pH 5.0 + 5 mM DTT. This resulted in
much improved lysate which then was further
purified to 90 % homogeneity using a
Phenylsepharose FF on an AKTA system and a linear
gradient from 2 M to 0 M ammonium sulfate, 50 mM
Tris pH 7.5 and 5 mM DTT.
(R)-ADH (Lactobacillus brevis) [29a] and (S)-ADH
(Bacillus subtilis) were purified using IMAC-Ni-
NTA technology, for the (R)-ADH 5 mM MgCl2 was
added to all solutions including reactions.
Control Experiments
Control experiments were performed in glass vials
(4 and 25 mL). A magnetic stirrer and a stirring plate
were used when agitation was desired. Pure NOX and
ADH were incubated for 72 hours, at room
temperature and pH 7. All experiments were carried
out with a pure enzyme solution of 10 % (v/v) in
phosphate buffer in a total volume of 4 mL. For each
experiment, two different control conditions were
tested: (1) incubation under a quiescent solution in 4
mL closed vials (2) incubation in the presence of
gentle agitation/stirring in a closed container (4 mL
vials). Samples were taken over time, stored on ice to
stop further the enzyme deactivation and the residual
activity was measured following the NOX/ADH
activity assay procedure described below. Control
experiments for the biocatalytic conversion were set
up accordingly and the corresponding substrate was
added in the concentration described in the results
section.
NOX activity assay
Initial activity of NOX was determined by
following absorption changes of NADH consumption
in a spectrophotometer at 340 nm (ɛ = 6.22 mM-1 cm-
1), at 25 °C with a light path of 1 cm. Absorbance was
measured for 2 minutes using 0.2 mM of NADH, 1 %
(v/v) of clarified lysate and phosphate buffer at pH 7
(concentrations in the cuvette). The cuvettes were
well mixed to ensure saturation of oxygen in the
solution with air at atmospheric pressure. One unit
(U) corresponds to 1 µmol of NAD+ produced per
minute at 25°C, pH 7 using phosphate buffer. NoxV
from L. plantarum had an average activity of 50
U/mL lysate and between 60-80 U/ml for the pure
NOX.
ADH activity assay
ADH activity after purification was assessed using
absorption changes of NADH consumption with 20
mM acetophenone as a substrate. Same NADH
conditions as above were applied. For monitoring
enzyme deactivation during the biocatalytic
conversions 50 mM phenylethanol and 0.4 mM
NAD+ were used to ensure separation from NOX
activity in the same experiment.
10.1002/adsc.201900213
Acc
epte
d M
anus
crip
t
Advanced Synthesis & Catalysis
This article is protected by copyright. All rights reserved.
7
Bubble column experiments
To investigate the deactivation of enzymes in the
presence of a gas-liquid interface, a bench scale
bubble column was used. In this defined and
controlled environment, gas bubbles rise in a stagnant
liquid solution under the influence of gravitational
force. The gas flow rate was defined so the bubbles
did not collide with each other and remained
spherical throughout the experiments. The bubbles
rose in a zig-zag and spiral pattern, hitting the column
wall in the same places. Even so, from the recorded
videos, the bubbles appear to remain nearly spherical
for the whole course of the experiment. The diameter
of the bubbles was observed to be 3 mm.
Bubble column setup
The bubble column was a glass tube with a length
of 48 cm and an inner diameter of 6.2 mm
(Supporting Information, Figure S1). A needle with
an inner diameter of approximately 0.6 mm was
attached to the bottom of the column and operated as
a nozzle (Supporting Information, Figure S1). Air and
nitrogen (99.9 % pure) were sparged through the
nozzle at a given flow rate (Q) of 14 mL min-1 (2.3
x10-7 m3 s-1). The gas bottles were connected to a
mass flow controller to measure the flow rate and
assure a constant value. The column had a liquid
height (L) of 25 cm and the residence time of a
bubble (θ) was approximately 1.7 seconds. A GoPro
Hero 6 camera (frame rate of 240 fps, San Mateo, CA,
USA) photographed and filmed the set-up to verify
the rising bubbles regime and that the bubbles did not
touch each other (for video URL, see Supplemental
Information). The bubble diameter (db) was
established by a photograph of the column that
contained a graduated scale. A tube connected to a
syringe was installed at the top of the column to fill in
the column with the enzyme solution and to collect
samples (sample port).
Procedure
First, the gas flow rate was set to 14 mL min-1.
When it was constant, 10 mL NOX solution of 10 %
(v/v) in buffer was slowly injected from the sample
port. Samples were collected at regular time intervals,
stored on ice to stop further enzyme deactivation and
tested for residual activity following the NOX
activity assay described above. Before starting an
experiment, the column was cleaned with acetone so
the path of the bubbles was not interrupted. The
concentration of enzyme solution was kept the same
to ensure constant surface tension. For conversion
experiments, 50 mM racemic 1-phenylethanol was
employed as the substrate in the ADH/NOX
experiments.
HPLC analysis
50 µl samples were taken of each tested conditions
((1) quiescent liquid (2) agitated/stirred liquid and (3)
sparged liquid.) that underwent biocatalytic
conversion of 50 mM 1-phenylethanol at several time
points. These samples were diluted 1:5 in acetonitrile
and filtered to remove protein matter. 5 µl of these
samples were separated on a C18 YMC ODS-AQ
column (5.5 µm, 3x100 mm, YMC, America) using a
Shimadzu ULPC 20A system with a water-
acetonitrile mobile phase. A step gradient from 5% to
15% in 2 min and then up to 40 % acetonitrile in 20
min was used to separate alcohol from ketone. The
alcohol eluted at 10.5 min and the ketone at 14 min
(Figure S3).
Product characterization
For Mass spectrometry analysis, the product peak
observed in HPLC was fractionated (see Figure S3),
fractions collected, extracted with 100 % chloroform,
dried and analyzed via Mass Spec. High
resolution mass spectra (HRMS) were obtained on a
Thermo Scientific Q Exactive Plus Orbitrap Mass
Spectrometer. The sample was dissolved in 50 %
acetonitrile/ 50 % formic acid (0.1%) for analysis in
positive ion mode (Figure S4, panel A).
For NMR analysis, after 6h conversion with
ADH/NOX the reactor volume was collected,
extracted with chloroform and evaporated affording a
colorless crude oil.
The resulting crude oil was then purified by flash
chromatography (solvent 8 Hex:1 EtOAc) to afford 9
mg of acetophenone in a 15 % yield as a colorless oil. 1H NMR (300 MHz CDCl3) δ: 7.2-7.7 (m, 5H), 2.36
(s, 3H); 13C NMR (300 MHz CDCl3) δ: 198, 136.9,
132.9, 128.4, 128.1, 26.5; mass (EI) 51, 77, 105, 120.
Acknowledgements
MDG acknowledges financial support from the Technical University of Denmark (DTU). BB and ASB gratefully acknowledge financial support from National Science Foundation of the United States (NSF) grant IIP-1540017, BDF and ASB gratefully acknowledge financial support from NSF grant CBET-1512848. We wish to acknowledge the Systems Mass Spectrometry Core Facility at the Parker H. Petit Institute for Bioengineering and Bioscience at the Georgia Institute of Technology for the use of their shared equipment, services and expertise.
References
10.1002/adsc.201900213
Acc
epte
d M
anus
crip
t
Advanced Synthesis & Catalysis
This article is protected by copyright. All rights reserved.
8
[1] a) D. J. Pollard, J. M. Woodley, Trends
Biotechnol. 2007, 25, 66-73; b) J. M.
Woodley, M. Breuer, D. Mink, Chem. Eng.
Res. Des. 2013, 91, 2029-2036.
[2] J. M. Woodley, Comput. Chem. Eng. 2017,
105, 297-307.
[3] R. A. Sheldon, J. M. Woodley, Chem. Rev.
2018, 118, 801-838.
[4] A. S. Bommarius, Annu. Rev. Chem. Biomol.
Eng. 2015, 6, 319-345.
[5] J. Dong, E. Fernandez-Fueyo, F. Hollmann,
C. E. Paul, M. Pesic, S. Schmidt, Y. Wang, S.
Younes, W. Zhang, Angew. Chem. Int. Ed.
Engl. 2018, 57, 9238-9261.
[6] N. J. Turner, Chem. Rev. 2011, 111, 4073-
4087.
[7] N. Turner, in Enzyme Catalysis and Organic
Synthesis 2012.
[8] G. Grogan, Curr. Opin. Chem. Biol. 2018, 43,
15-22.
[9] W. R. Birmingham, N. J. Turner, ACS Catal.
2018, 8, 4025-4032.
[10] S. M. A. De Wildeman, T. Sonke, H. E.
Schoemaker, O. May, Accounts Chem. Res.
2007, 40, 1260-1266.
[11] J. Brummund, T. Sonke, M. Müller, Org.
Process Res. Dev. 2015, 19, 1590-1595.
[12] J. Liu, S. Wu, Z. Li, Curr. Opin. Chem. Biol.
2018, 43, 77-86.
[13] G. Rehn, A. T. Pedersen, J. M. Woodley, J.
Mol. Catal. B: Enzymatic 2016, 134, 331-339.
[14] B. R. Riebel, P. R. Gibbs, W. B. Wellborn, A.
S. Bommarius, Adv. Syn. Catal. 2003, 345,
707-712.
[15] G. T. Lountos, R. Jiang, W. B. Wellborn, T.
L. Thaler, A. S. Bommarius, A. M. Orville,
Biochemistry 2006, 45, 9648-9659.
[16] C. Nowak, B. Beer, A. Pick, T. Roth, P.
Lommes, V. Sieber, Front. Microbiol. 2015,
6, 957.
[17] B. Petschacher, N. Staunig, M. Muller, M.
Schurmann, D. Mink, S. De Wildeman, K.
Gruber, A. Glieder, Comput. Struct.
Biotechnol. J. 2014, 9, e201402005.
[18] J. T. Park, J.-I. Hirano, V. Thangavel, B. R.
Riebel, A. S. Bommarius, J. Mol. Catal. B:
Enzymatic 2011, 71, 159-165.
[19] P. Tufvesson, J. Lima-Ramos, J. S. Jensen, N.
Al-Haque, W. Neto, J. M. Woodley,
Biotechnol. Bioeng. 2011, 108, 1479-1493.
[20] T. L. Donaldson, E. F. Boonstra, J. M.
Hammond, J. Colloid Interface Sci. 1980, 74,
441-450.
[21] a) M. Caussette, A. Gaunand, H. Planche, S.
Colombié, P. Monsan, B. Lindet, Enzyme
Microb. Technol. 1999, 24, 412-418; b) S.
Colombié, A. Gaunand, B. Lindet, Enzyme
Microb. Technol. 2001, 28, 820-826.
[22] N. S. Patil, R. S. Ghadge, S. B. Sawant, J. B.
Joshi, AIChE J. 2000, 46, 1280-1283.
[23] M. Mohanty, R. S. Ghadge, N. S. Patil, S. B.
Sawant, J. B. Joshi, A. V. Deshpande, Chem.
Eng. Sci. 2001, 56, 3401-3408.
[24] A. S. Bommarius, A. Karau, Biotechnol.
Prog. 2005, 21, 1663-1672.
[25] Z. Findrik, I. Valentović, Đ. Vasić-Rački,
Appl. Biochem. Biotechnol. 2014, 172, 3092-
3105.
[26] S. Bhagia, R. Dhir, R. Kumar, C. E. Wyman,
Sci. Rep. 2018, 8, 1350.
[27] A. Toftgaard Pedersen, W. R. Birmingham,
G. Rehn, S. J. Charnock, N. J. Turner, J. M.
Woodley, Org. Proc. Res. Dev. 2015, 19,
1580-1589.
[28] M. Gräff, P. C. F. Buchholz, P. Stockinger,
B. Bommarius, A. S. Bommarius, J. Pleiss,
Proteins 2019, 1-9; 10.1002/prot.25666. [29] a) N. H. Schlieben, K. Niefind, J. Muller, B.
Riebel, W. Hummel, D. Schomburg, J. Mol.
Biol. 2005, 349, 801-813; b) J. Schumacher,
M. Eckstein, U. Kragl, Biotechnol. J. 2006, 1,
574-581.
[30] A. Slavica, I. Dib, B. Nidetzky, Appl.
Environ. Microbiol. 2005, 71, 8061.
[31] A. Sadana, Biocatalysis 1989, 2, 175-216. ä
10.1002/adsc.201900213
Acc
epte
d M
anus
crip
t
Advanced Synthesis & Catalysis
This article is protected by copyright. All rights reserved.
9
FULL PAPER
Bubble Column Enables Higher Reaction Rate for Deracemization of 1-Phenylethanol with Coupled Alcohol Dehydrogenase/NADH Oxidase System
Adv. Synth. Catal. Year, Volume, Page – Page
Mafalda Dias Gomes†, Bettina R. Bommarius†, Shelby R. Anderson, Brent D. Feske, John M. Woodley* and Andreas S. Bommarius*
10.1002/adsc.201900213
Acc
epte
d M
anus
crip
t
Advanced Synthesis & Catalysis
This article is protected by copyright. All rights reserved.