Abstract
Background: Biocatalysis in high-concentration organic solvents
has been applied to produce various industrial products with many
advantages. However, using enzymes in organic solvents often suffers
from inactivation or decreased catalytic activity and stability. So,
improving the tolerance of enzymes in organic solvents is essential.
Main methods and results: Herein, the method of regional random
mutation combined with combinatorial mutation was used to improve the
resistance of transaminase from Aspergillus terreus(At ATA) in organic
solvents, and the best mutant T23I/T200K/P260S (M3) was acquired. In
different concentrations (ranging from 25% to 45%, v/v ) of
dimethyl sulfoxide (DMSO), the catalytic efficiency
(k cat/K m) toward
1-acetylnaphthalene and the stability (half-lifet 1/2) were higher than the wild-type (WT) ofAt ATA. M3 also showed enhanced stability against six organic
solvents with different oil-water partition coefficients (log P values).
The results of decreased Root Mean Square Fluctuation (RMSF) values via
20-ns molecular dynamics (MD) simulations under 15%, 25%, 35%, and
45% DMSO revealed that mutant M3 had lower flexibility, acquiring a
more stable protein structure and contributing to its organic solvents
stability than WT. Intra- and intermolecular interaction analysis
indicated that the increased hydrogen bonds and hydrophobic interactions
within monomers or at the interface of two monomers also strengthened
the stability of the overall structure against organic solvents.
Furthermore, M3 was applied to convert 1-acetylnaphthalene for
synthesizing (R )-(+)-1(1-naphthyl)-ethylamine ((R )-NEA),
which was a resolving agent for producing L-menthol by resolution of
monomenthyl phthalate, and an intermediate of Cinacalcet Hydrochloride
for the treatment of secondary hyperthyroidism and hypercalcemia.
Moreover, 3~10 mM 1-acetylnaphthalene can be converted
to (R )-NEA with 94.2~38.9% yield and a strictR -stereoselectivity (enantiomeric excess (e.e. ) value
>99.5%) within 10 h under 25% DMSO, which was higher than
WT and expected to be a potential biocatalyst for industrial
application.
Conclusion: The beneficial mutation sites were identified to
tailor At ATA’s organic solvents stability via regional random
mutation. The “best” mutant T23I/T200K/P260S (M3) holds great
potential application for the synthesis of (R )-NEA.
Keywords: transaminase, organic solvent stability, regional random
mutation, MD simulations, (R )-(+)-1(1-naphthyl)-ethylamine
1. Introduction
(R )-1-(1-Naphthyl)-ethylamine ((R )-NEA) is a critical
chiral aromatic amine compound, which is widely used in pharmaceutical
industry, chemical industry, materials and other
fields.[1] For example, (R )-NEA can be used
as a resolving agent for acid/ester enantiomers by exploiting the weakly
basic chemical properties of amino group.[2] Dudas
et al. reported (R )-NEA was a resolving agent for producing
L-menthol by resolution of monomenthyl
phthalate.[2] Morever, (R )-NEA can be used
to prepare the intermediate of Cinacalcet Hydrochloride from
1-acetylnaphthalene, which is a kind of pharmaceutical for the treatment
of secondary hyperthyroidism and hypercalcemia.[3]The synthesis of (R )-NEA via biocatalysts, such as amine
oxidases, imine reductase, amine dehydrogenases, and
transaminases,[4] has attracted extensive
attention of chemists due to its advantages of high selectivity, mild
reaction conditions, and environmental
friendliness.[5] The ω-transaminases are capable
of catalyzing the asymmetric amination from prochiral ketones to chiral
amines with strict stereoselectivity and 100% theoretical
yield,[6] indicating that they are promising
biocatalysts for the production of chiral amines. However, due to the
low solubility of organic substrates in the aqueous phase, it is
difficult for enzyme to convert substrates towards the desired product.
Biocatalysis in organic solvents is high-efficient for converting
organic substrates. Organic solvents can dissolve more nonpolar
substrates[7] and restrain microbial growth and
pollution.[8] Biocatalysis in organic phase is
helpful for the separation and purification of products due to the low
boiling points of organic solvents.[9] Benefiting
from these advantages, biocatalysis in organic solvents has been
employed to produce various high value-added
products.[10] However, the stability of enzymes is
poor, and the activity is decreased or even inactivated in high
concentration organic solvents. Therefore, improving the enzyme
resistance and clarifying the mechanism of enzyme stability in organic
solvents are needed.[7]
The method for improving the stability of enzymes in organic solvents
can be achieved by rational (semi-rational) design and random
mutation.[11,12] Tian et al. reported a
semi-rational method for improving methanol tolerance ofThermomyces lanuginosus lipase,[13] which
aimed for the high B-factor residues[14] to
perform iterative saturation mutagenesis (ISM). In the directed
evolution campaign of metalloprotease PT121 via random mutation, eleven
variants with enhanced stability against 25% (v/v ) acetonitrile
were obtained.[15] The beneficial “wing‐type
gate” for improving organic solvents resistance of ω-ATA fromArthrobacter cumminsii ZJUT212 was identified via a semi‐rational
design and two beneficial variants were
obtained.[16] Strengthening protein surface
hydration via surface charge engineering combined with molecular
dynamics (MD) simulation is an efficient rational strategy for tailoring
enzyme stability in organic solvents.[17] Meng et
al. obtained two optimal ω-transaminase mutants from Pseudomonas
jessenii with improved activity and high concentrations of
isopropylamine and co-solvent tolerance by computational interface
design.[18] Generally, the design principle is
still a considerable challenge for constructing a “small and smart”
mutation library. Combining the advantages of rational design and random
mutation will be helpful in obtaining more positive mutants quickly and
efficiently.[19,20]
In our previous work, wild type (WT) of amine transaminase fromAspergillus terreus (At ATA) has been cloned and expressed
in Escherichia coli , and At ATA showed excellent catalytic
efficiency and high R -enantioselectivity toward various
ketones.[21,22] Herein, the substrate
1-acetylnaphthalene could be catalyzed to produce (R )-NEA byAt ATA. However, engineering of At ATA to develop a robust
biocatalyst for green biomanufacturing of (R )-NEA is in huge
demand due to the poor stability in organic solvents. In this study, we
identified the critical hot spots by a strategy of regional random
mutation, which analyzed the structure of At ATA, performed the
error-prone PCR (epPCR) and combinatorial mutation, to evolve the
organic solvents resistance of At ATA. For the “best” variant,
its enzymatic properties towards 1-acetylnaphthalene in different
organic solvents were performed in detail. Moreover, the underlying
molecular mechanism was clarified by docking analysis and MD
simulations.
2 Materials and methods
2.1 Materials
E. coli BL21(DE3) used for cloning and recombinant protein
expression was purchased from TransGen Biotech Co. Ltd. (Beijing,
China). The strain E. coli BL21(DE3)/pET28a(+)-At ATA was
constructed in previous work and preserved in our
laboratory.[22] Prime STAR® Max
DNA polymerase was purchased from Takara Biotechnology (Dalian, China).
The restriction enzyme Dpn I was purchased from Thermo Scientific
(Shanghai, China). All polymerase chain reaction (PCR) primers were
synthesized by Tsingke Biological Technology (Hangzhou, China) and
listed in Table S1. Other chemicals were of analytical grade and
obtained from standard commercial sources.
2.2 Methods
2.2.1 Selection of organic solvent
According to the log P values, seven organic solvents containing
acetonitrile (ACN), acetone (AC),
methanol (MeOH), dimethyl sulfoxide (DMSO), N,N -dimethylformamide
(DMF), isopropanol (IPA), and ethanol (EtOH) were selected for
co-solvent of At ATA in this study. The reaction mixture contained
1-(R )-phenylethylamine (1-(R )-PEA) (5 mM),
1-acetylnaphthalene (5 mM), PLP (0.1 mM), the purified enzyme ofAt ATA-WT, and different organic solvents (the concentration
ranging from 5~50%, v/v ). The reactions were
performed at 30 °C, 500 rpm for 30 minutes and the product
(R )-NEA was analyzed by high performance liquid chromatography
(HPLC). One unit of enzyme activity (U) was defined as the amount of
enzyme required for 1 μmol of (R )-NEA formed per minute under
optimum conditions. The specific activity was expressed as the units of
activity per gram purified enzyme (U/g). All experiments were conducted
in triplicates.
2.2.2 Mutagenesis and screening
Three regions of amino acid sequence fragments were selected to
construct mutagenesis libraries via in situ epPCR based on WT as
template. MnCl2 (0.1 mM) was added to the PCR mixture to
control the mutagenesis rate (1 to 2 mutation sites per gene). The PCR
product was digested by Eco R I and Xho I, and ligated to
the vector pET-28a. The PCR reaction conditions were supplemented in
Table S2 to Table S5, and the template was digested by Dpn I at
37 °C for 0.5 h (refer to Table S6 for details). The PCR products were
transformed into E. coli BL21 (DE3) competent cells for enzyme
expression.
Colonies were picked up in 96 deep-well plates containing 1 mL of LB
medium with 50 μg/mL kanamycin and cultured for 8 h at 37 °C.
Subsequently, isopropyl β-D-1-thiogalactopyranoside (IPTG) was added to
96 deep-well plates with a final concentration of 0.1 mM to induce
protein expression. After incubation for 20 h at 25oC, the cells were harvested by centrifugation for 15
min at 3900 × g. The cells in 96 deep-well plates were resuspended in
250 μL PBS buffer (100 mM, pH 8.0) with 5 mg/mL lysozyme at 37 °C for 30
min. The crude enzyme solution was centrifuged for 30 min at 3500 × g.
Next, 50 μL crude enzyme solution and 50 μL of 80% (v/v ) DMSO
was added to a new 96-well plate and incubated at 30 °C for 30 min.
A chromogenic reaction-based screening method was developed. In this
section, 10 mM 1-(R )-PEA, 40% (v/v ) DMSO, 25 mM
4-nitrophenylethylamine, 50 mM sodium phosphate (pH 8.0), 0.1 mM PLP,
and enzymes incubating by 40% (v/v ) DMSO, were added to 96-well
plate at 30 °C, 600 rpm for 30 min.[23] The
absorbance values of the reaction solution were measured in 96-well
microtiter plates at 500 nm using SpectraMax 190 Microplate Reader
(Molecular Devices, USA), and Optical Density at 500 nm
(OD500) was calculated.
2.2.3 Protein expression and purification
All enzymes that we constructed were expressed and purified as described
in our previous work.[24] The mutants and WT with
His6-Tag were purified by nickel affinity chromatography
and analyzed by SDS-PAGE.
2.2.4 Determination of kinetic parameters and half-lives in different
concentration DMSO
The purified At ATAs were incubated in different concentrations of
DMSO (the concentration ranging from 15~45%,v/v ) for 15 min, and the temperature was maintained at 30 °C.
Then, enzyme activity assay was performed. Each sample of 1 mL mixture
was centrifuged at 13800 ×g for 2 min, and the supernatant was subjected
to microfiltration with 0.22 μm PTFE organic membranes. The product and
remaining substrate were analyzed by HPLC to measure the
1-acetylnaphthalene conversion, (R )-NEA yield, and enantiomeric
excess (e.e .). All the assays were conducted in triplicate.
The half-lives (t 1/2) of At ATAs were
determined by incubating each purified enzyme (1 mg/mL) for appropriate
times in different concentrations of DMSO (25%, 35%, and 45%,v/v ) at 30 °C, respectively, followed by measuring the residual
activity. The half-lives were calculated according to the first order
deactivation equation 1 and 2:
ln(A/A0 )=-k dt(equation 1)
t 1/2=ln2/k d (equation 2)
where A 0 is the initial activity, A is the
residual activity at time t during the thermal deactivation,k d is the deactivation rate constant
(min-1)
2.2.5 Enzyme kinetics characterization under different concentrations of
DMSO
The reaction solution was added to DMSO (the final concentration of DMSO
to 25%, 35%, or 45%, v/v ). The initial rates were measured at
various concentrations of 1-acetylnaphthalene in a range of from 0.1 to
15 mM with a fixed 1-(R )-PEA concentration (5 mM) and PLP (0.1
mM). For 1-(R )-PEA, 1-acetylnaphthalene (5 mM) and
1-(R )-PEA in the range of 0.1 to 1.5 mM were used for the enzyme
kinetic assays. The kinetic parameters were obtained by nonlinear
fitting the experimental data to Michaelis-Menten
equation.[25,26]
2.2.6 MD simulation
The starting model was generated from the crystallography structure (PDB
ID: 4CE5).[22] The models of the variants were
generated by FoldX software. MD simulations of At ATAs were
performed via software YASARA at 308 K for 20 ns under AMBER14 force
field.[24] A cubic box was constructed that filled
with 15~45% (v/v) DMSO. The simulation system
was constructed with 15818~36056 water molecules, the
number of which varied depending on the DMSO concentration. Counterions
Na+ and Cl− were used to neutralize
the total net charge of the systems, and all resulting systems had a net
charge of zero. During the simulation, coordinates, energies, and
velocities were stored per 0.5 ns for further analysis.
2.2.7 Asymmetric synthesis of (R )-NEA by WT and mutant M3
The bio-asymmetric ammoniation 1-acetylnaphthalene to produce
(R )-NEA was performed in a 20-mL scale reaction including 6.0 gdry cell weight/L recombinant E.coli whole cells
expressing WT or M3, 3~20 mM 1-(R )-PEA,
3~20 mM 1-acetylnaphthalene, and 25% (v/v ) DMSO
at 30 °C, 600 rpm. The yield and e.e . value of (R )-NEA
were assayed via HPLC.
2.2.8 Analytical methods
The e.e. value of the product was determined after derivatization
with Marfey’s reagent (1-fluoro-2,4-dinitrophenyl-5-L-alaninamide, FDAA)
as follows: 50 μL of the reaction solution was mixed with 100 μL of 1%
(m/v ) FDAA solution in acetone and 20 μL of
NaHCO3 solution. After incubation at 40 °C for 2 h, 20
μL of 2 M HCl was added to terminate the reaction. The samples were
extracted with 3 times volume of dichloromethane, and evaporated at room
temperature. Then, samples were dissolved in 50% (v/v )
acetonitrile aqueous solution for HPLC analysis. An HPLC 1220 Infinity
II system (Agilent Technologies) with an EC-C18 column (4.6 × 150 mm, 4
μm) was used at 30 °C. Detector wavelength was set at 340 nm. The mobile
phase was composed of acetonitrile and ultra-pure water at a volumetric
ratio of 60: 40 (v/v ), and ran at a flow rate of 1 mL/min for 8
min.
The contents of 1-acetylnaphthalene and (R )-NEA were
quantitatively analyzed by using an HPLC 1220 Infinity II system
(Agilent Technologies) with a C-18 column (4.6 × 150 mm, 4 μm). Detector
wavelength was set at 210 nm. The mobile phase was composed of 0.15%
ethanolamine, acetonitrile and ultra-pure water at a volumetric ratio of
1 : 39 : 60 (v/v/v ), and ran at a flow rate of 1.0 mL/min. The
column temperature was maintained at 30 °C. Each measurement was
conducted at least 3 times with a standard deviation of less than 5%.
3. Results and discussion
3.1 Determination of enzyme activity under different organic solvents
ACN, AC, MeOH, DMSO, DMF, IPA, and EtOH are the representatively and
commonly used organic co-solvents with different log P values. The
enzyme activities of At ATA in different concentrations of organic
solvents were shown in Figure 1. All seven kinds of organic solvents had
good solubility for 1-acetylnaphthalene (5 mM), and the activity ofAt ATA in DMSO was higher than other six organic solvents. Enzyme
in 25% (v/v ) DMSO displayed the highest activity, which the
specific activity reached 8.86 U/g and the activity showed a bell-shape
trend with the DMSO concentration increasing. Instead, the enzyme
activities were decreased when the concentrations of other six organic
solvents increased. Many reports indicated that DMSO was extensively
used in the fields of organic synthesis and biocatalysis due to its
favorable properties (amphiphilicity, dissolving ability, low chemical
reactivity).[27,28] In the following experiments,
DMSO was selected as the organic solvent in this study.