A novel phenylalanine ammonia-lyase from Pseudozyma antarctica for stereoselective biotransformations of unnatural amino acids

Article abstract of DOI:10.1016/j.cattod.2020.04.002

A novel phenylalanine ammonia-lyase of the psychrophilic yeast Pseudozyma antarctica (PzaPAL) was identified by screening microbial genomes against known PAL sequences. PzaPAL has a significantly different substrate binding pocket with an extended loop (26 aa long) connected to the aromatic ring binding region of the active site as compared to the known PALs from eukaryotes. The general properties of recombinant PzaPAL expressed in E. coli were characterized including kinetic features of this novel PAL with L-phenylalanine (S)-1a and further racemic substituted phenylalanines rac-1b-g,k. In most cases, PzaPAL revealed significantly higher turnover numbers than the PAL from Petroselinum crispum (PcPAL). Finally, the biocatalytic performance of PzaPAL and PcPAL was compared in the kinetic resolutions of racemic phenylalanine derivatives (rac-1a-s) by enzymatic ammonia elimination and also in the enantiotope selective ammonia addition reactions to cinnamic acid derivatives (2a-s). The enantiotope selectivity of PzaPAL with o-, m-, p-fluoro-, o-, p-chloro- and o-, m-bromo-substituted cinnamic acids proved to be higher than that of PcPAL.

Full text of DOI:10.1016/j.cattod.2020.04.002

Journal Pre-proof
A novel phenylalanine ammonia-lyase from Pseudozyma antarctica for
stereoselective biotransformations of unnatural amino acids
Andrea Varga (Investigation) (Methodology) (Writing - review and
´
editing) (Writing - original draft), Pal Csuka (Investigation) (Writing -
review and editing), Orlavanah Sonesouphap (Investigation),
´ ´
Gergely Banoczi (Software) (Visualization), Monica Ioana Tos¸a
(Methodology) (Project administration) (Formal analysis), Gabriel
´
´
Katona (Validation) (Resources), Zsofia Molnar (Investigation),
´
´
Laszlo Csaba Bencze (Writing - review and editing) (Supervision),
´
´
Laszlo Poppe (Conceptualization) (Writing - review and editing)
(Supervision) (Project administration) (Funding acquisition), Csaba
Paizs (Conceptualization) (Methodology) (Project administration)
(Supervision) (Writing - review and editing)
PII:
S0920-5861(20)30190-5
DOI:
https://doi.org/10.1016/j.cattod.2020.04.002
CATTOD 12781
Reference:
To appear in:
Catalysis Today
Received Date:
Revised Date:
Accepted Date:
2 December 2019
30 March 2020
1 April 2020
´ ´
Please cite this article as: Varga A, Csuka P, Sonesouphap O, Banoczi G, Tos¸a MI, Katona G,
´
Molnar Z, Bencze LC, Poppe L, Paizs C, A novel phenylalanine ammonia-lyase from
Pseudozyma antarctica for stereoselective biotransformations of unnatural amino acids,
Catalysis Today (2020), doi: https://doi.org/10.1016/j.cattod.2020.04.002

This is a PDF file of an article that has undergone enhancements after acceptance, such as
the addition of a cover page and metadata, and formatting for readability, but it is not yet the
definitive version of record. This version will undergo additional copyediting, typesetting and
review before it is published in its final form, but we are providing this version to give early
visibility of the article. Please note that, during the production process, errors may be
discovered which could affect the content, and all legal disclaimers that apply to the journal
pertain.
© 2020 Published by Elsevier.

A novel phenylalanine ammonia-lyase from Pseudozyma antarctica for
stereoselective biotransformations of unnatural amino acids
Andrea Vargaa, Pál Csukab, Orlavanah Sonesouphapa,c, Gergely Bánóczia, Monica Ioana Toșaa,
Gabriel Katonaa, Zsófia Molnárb, László Csaba Bencze*a, László Poppe*a,b,d, Csaba Paizs*a
aBiocatalysis and Biotransformation Research Center, Faculty of Chemistry and Chemical
Engineering, Babeş-Bolyai University of Cluj-Napoca, Arany János str. 11, 400028, Cluj-
Napoca, Romania,
bDepartment of Organic Chemistry and Technology, Budapest University of Technology and
Economics, Műegyetem rkp. 3, 1111, Budapest, Hungary
cDepartment of Chemical Engineering, Faculty of Engineering, National University of Laos,
P.O.Box: 7322, Dongdok, Laos
dSynBiocat Ltd, Szilasliget u. 3, 1172, Budapest, Hungary
Corresponding authors:
Csaba Paizs
E-mail:paizs@chem.ubbcluj.ro
László Poppe
E-mail:poppe@mail.bme.hu
László Csaba Bencze
E-mail:cslbencze@chem.ubbcluj.ro
Graphical abstract
Highlights
● Novel phenylalanine ammonia-lyase for the production of unnatural amino acids;
● Superior turnover numbers compared to known phenylalanine ammonia-lyases;
● Superior enantioselectivity in the synthesis of unnatural phenylalanines bearing electron
withdrawing substituents;
● High thermal stability for an enzyme of psychrophilic origin.
1

Abstract
A novel phenylalanine ammonia-lyase of the psychrophilic yeast Pseudozyma antarctica
(PzaPAL) was identified by screening microbial genomes against known PAL sequences. PzaPAL
has a significantly different substrate binding pocket with an extended loop (26 aa long) connected
to the aromatic ring binding region of the active site as compared to the known PALs from
eukaryotes. The general properties of recombinant PzaPAL expressed in E. coli were characterized
including kinetic features of this novel PAL with L-phenylalanine (S)-1a and further racemic
substituted phenylalanines rac-1b-g,k. In most cases, PzaPAL revealed significantly higher
turnover numbers than the PAL from Petroselinum crispum (PcPAL). Finally, the biocatalytic
performance of PzaPAL and PcPAL was compared in the kinetic resolutions of racemic
phenylalanine derivatives (rac-1a-s) by enzymatic ammonia elimination and also in the
enantiotope selective ammonia addition reactions to cinnamic acid derivatives (2a-s). The
enantiotope selectivity of PzaPAL with o-, m-, p-fluoro-, o-, p-chloro- and o-, m-bromo-substituted
cinnamic acids proved to be higher than that of PcPAL.
Keywords
biocatalysis; phenylalanine ammonia-lyase; yeast; Pseudozyma antarctica; unnatural amino acid
2

1. Introduction
The pharmaceutical industry is in an ever-growing need for enantiopure compounds and
more effective synthetic routes.[1] Biocatalysts being related almost exclusively to enzymes are
eco-friendly biomaterials with high application potential in synthetic chemistry.[2] Enzymes can
catalyze stereoselective reactions resulting in various building blocks, intermediates or active
pharmaceutical ingredients (APIs) of high value.[3] It is an excellent choice to utilize the benefits
of biocatalysis in production of enantiopure API molecules.
Enantiopure aromatic α- and β-amino acids are frequently occurring building blocks of natural
secondary metabolites such as andrimide,[4] enedyine C-1027,[5] taxol,[6,7] chondramide C, L-
DOPA[8] or L-thyroxine. Thus, enantiopure α- and β-amino acids and their unnatural analogues
are well-applicable precursors of nuclear hormone inhibitors, hepatitis C antiviral agents or
cholecystokinin B antagonist and are frequently used as building blocks to create novel peptide or
cyclopeptide APIs.[9,10]
Aromatic ammonia-lyases and 2,3-aminomutases comprise a useful and rapidly expanding
toolbox of organic synthesis to create enantiopure unnatural α- and β-amino acids.[11]
All these enzymes contain the 3,5-dihydro-5-methylidene-4H-imidazol-4-one (MIO) prosthetic
group[12,13,14] and play important roles in the metabolism of aromatic amino acids in bacteria,
fungi, plants, and animals. The aromatic 2,3-aminomutases mediate the essential step of the
enantiopure β-amino acid formation from α-amino acid in the secondary metabolite
production.[15] Aromatic ammonia-lyases play essential roles in the non-oxidative ammonia
elimination of histidine,[16] phenylalanine, and tyrosine,[17] to create the α,β-unsaturated
arylacrylates for the biosynthesis of phenylpropanoids or the L-histidine elimination from the
blood by animals.
Under natural conditions, the enantiomer selective ammonia elimination is the preferable
direction of the aromatic ammonia-lyases. These enzymes were applied to produce enantiopure D-
enantiomers of arylalanines by enantiomer selective destruction of the L-enantiomers from their
racemates. Contrary, under ammonia rich reaction conditions these enzymes can catalyze the
reaction in a reversible manner to create the L-α-amino acid in high enantiomeric purity by the
enantiotope selective ammonia addition onto α,β-unsaturated arylacrylates. Due to their broad
substrate tolerance, the most often used biocatalysts are the phenylalanine ammonia-lyases from
Rhodotorula glutinis (RgPAL)[18,19] Petroselinum crispum (PcPAL)[20] and Anabaena
variabilis (AvPAL).[21,22,23]
DSM Pharma Chemicals (Netherlands) developed a multi-ton scale two-step chemo-enzymatic
process (thermal cyclisation of L-2-chlorophenylalanine produced by a PAL mediated process),
replacing a seven-step chemical procedure for production of (S)-2,3-dihydro-1H-indole-2-
carboxylic acid, useful in the synthesis of hypertension drugs such as indolapril and
perindopril.[24]
The histidine ammonia-lyase (HAL, EC 4.3.1.3), phenylalanine ammonia-lyase (PAL, EC
4.3.1.24), tyrosine ammonia-lyase (TAL, 4.3.1.23), phenylalanine/tyrosine ammonia-lyase
(PTAL, EC 4.3.1.25), and the recently discovered phenylalanine/tyrosine/histidine ammonia-lyase
(XAL)[25] have high sequence similarity in the conserved domains containing the tyrosine as
catalytic base, the arginine interacting with the carboxylic group of substrate and the catalytically
essential MIO group which is formed usually from an alanine-serine-glycine triad under the
assistance of a neighboring tyrosine and asparagine. These two residues are also involved in
stabilization of the electrophilic MIO within the catalytic site.[26] While the catalytic machinery
performing the substrate binding and deamination within the MIO-enzymes is highly conserved,
the aromatic binding pocket of these enzymes is more variable. This part of the catalytic site could
be attributed to the differences in the type of enzyme action (lyase vs. mutase activity) and to the
3

substrate selectivity of natural and unnatural substrates in the MIO-containing aromatic ammonia-
lyases and 2,3-aminomutases.[25,27]
Site directed mutational studies revealed that single point mutations in the aromatic binding
pocket could increase or decrease the affinity towards the different substrates. The H89F mutation
in Rhodobacter sphareoides TAL resulted in an enzyme with high PAL activity and decreased
TAL activity, whereas the F144H mutant of Arabidopsis thaliana PAL achieved TAL activity and
have decreased PAL activity.[8,28] Moreover, the steric hindrance within the aromatic binding
pocket of AvPAL or PcPAL was considerably reduced by rational mutations, providing highly
increased enzyme activity towards bulky substrates.[29]
Bioinformatics tools using sequence similarity searches focusing on the highly conserved and
the more variable sequence motifs of the MIO-enzyme active site regions allowed identification
of novel enzymes with potential unique properties. The motif-based sequence similarity searches
and homology modeling were useful tools in identification and characterization of various MIO-
enzymes from extremophiles such as the PAL from the thermophilic Rubrobacter xylanophilus
(RxPAL),[30] PAL from the halophilic marine bacterium Kangiella koreensis (KkPAL)[31] and
three MIO enzymes from a unique Pseudomonas fluorescens strain isolated from a nutrient limited
sandstone cave (PfXAL, PfHAL, and PfPAM).[25]
Permanently cold environments on Earth host microorganisms that evolved enzymes with
impressive catalytic properties at low temperatures.[32] These enzymes are characterized by high
turnover numbers at low to moderate temperatures, by high thermolability at elevated
temperatures, and by increased ability to function in organic solvents.[33,34,35,36,37] Cold-
adapted enzymes can afford several economic benefits: i) as being more productive than
mesophilic or thermophilic homologues at room temperature, they can contribute to energy saving,
ii) as being more active at lower temperatures they can reduce the danger of undesirable chemical
reactions occurring at higher temperatures, iii) at lower temperatures they are able to convert heat-
sensitive substrates, and iv) enzymatic systems can be modulated by rapidly inactivating
psychrophilic enzymes with heat.[38]
Pseudozyma antarctica (also known as Candida antarctica), first isolated as Sporobolomyces
antarctica from 9 m depth of Lake Vanda, Antarctica,[39] is a well-known yeast species of the
Pseudozyma genus. This yeast provided well-studied enzymes from the hydrolase family, most
notably lipase A (CaL A) and lipase B (CaL B), with massive industrial use as biocatalysts.[40]
These biocatalysts are stable, stereoselective, and well-applicable in hydrolase-catalyzed
biotransformations such as hydrolysis, esterification, and transesterification reactions.[41,42,43]
Based on the above considerations, our aim in this study was to find enzymes of PAL or PAM
activity from psychrophilic organisms with a special focus on the industrially relevant Pseudozyma
antarctica yeast species. The heterologous expression of the potential candidate(s) would enable
the characterization of the new enzyme(s) (their ammonia-lyase and/or 2,3-aminomutases activity,
kinetic parameters, stability issues, etc.), but also the investigation of their biocatalytic ability in
the synthesis of variously substituted L- and D-/-arylalanines. Moreover, homology models of
the new candidates should provide a valuable assistance in establishing the correlation between
structure and catalytic activity of the novel biocatalyst(s).
2. Material and methods
2.1 Identification of the PzaPAL gene
The Pseudozyma antarctica PAL gene (PzaPAL, M9M0D4) was identified in the NCBI
database with blastp searches with specified MIO sequence fragments. The UniProt Align tool was
used for sequence analysis and comparison to other aromatic ammonia-lyase and 2,3-aminomutase
4

sequences [Ustilago maydis PAL (UmPAL, Q96V77), Rubrobacter xylanophilus PAL (RxPAL,
Q1AV79), Anabaena variabilis PAL (AvPAL, Q3M5Z3), Petroselinum crispum PAL (PcPAL,
P24481), Aradopsis thaliana PAL (AtPAL, P35510), Sorghum bicolor PAL (SbPAL, C5XXT9),
Rhodobacter sphaeroides TAL (RsTAL, Q3IWB0), Rhodosporidium toruloides TAL (RtTAL,
P11544), Pseudomonas fluorescens XAL (PfXAL, K0W9Y7), Pseudomonas fluorescens HAL
(PfHAL, K0WDT6), Pseudomonas putida HAL (PpHAL, P21310), Pseudomonas fluorescens
PAM (PfPAM, K0WLD2), Pantoea agglomerans PAM (PaPAM, Q84FL5), Taxus canadensis
PAM (TcPAM, Q6GZ04), Chondromyces crocatus TAM (CcTAM, Q0VZ68) and Streptomyces
globisporus TAM (SgTAM, Q8GMG0)].
2.2. Cloning and amplification of PzaPAL gene
See in ESI Section 3.
2.3. Expression and purification of PzaPAL
See in ESI Section 4 and 5.
2.4. pH optimum of the PzaPAL-catalyzed reaction
Initial activity of the deamination of L-phenylalanine (5 mM) was monitored at 290 nm for 10
min at 30 °C and different pH-values. For pH 7.0 to 9.0 TRIS-buffer (100 mM) was used;
measurements at pH 9.5 to 10.5 were conducted in (NH4)2CO3 buffer (100 mM). Measurements
at pH 6.0, 6.5, 11.0, 11.5, 12.0 were carried out in PB (Phosphate buffer) (100 mM). Buffers were
prepared at room temperature.
2.5. Thermal stability of PzaPAL
The thermal stability tests were investigated i) between pH 7.0 and 10.5 (pH 7.0-9.0 in 100 mM
TRIS buffer; pH 9.5-10.5 in 100 mM ammonium carbonate buffer); ii) after 48 h incubation at 10,
20, 30, 40, 50 °C in 100 mM TRIS buffer, pH 8.5; iii) in 100 mM TRIS buffer (pH 8.5)
implemented with sodium chloride between 50 and 500 mM. The thermal stability of the enzyme
was determined by nano-DSF (Prometheus NT.48, NanoTemper Technologies; München,
Germany). The capillaries were filled with 0.1 mg mL-1 PzaPAL solution diluted with the required
buffer solutions and placed into the capillary tray of the nano-DSF equipment. Melting curves
were measured by heating the samples from 20 °C to 95 °C with 1 °C min-1. The evaluation was
executed on the base of the first derivative of the fluorescent signal.
2.6. Kinetic characterization of PzaPAL-catalyzed ammonia elimination reactions
The reaction mixtures contained 100 mM TRIS buffer, pH 8.5, and different concentrations of
substrates ((S)-1a, rac-1b-g, k) and were incubated at 30 or 42°C (in case of (S)-1a) for 3 min. The
reactions were started by the addition of purified PzaPAL (10 μg mL-1 for rac-1a-d,f,k and 50 μg
mL-1 for rac-1e,g in final concentration) in a final volume of 0.2 mL. The measurements were
performed in triplicates.
The values of Vmax and KM were obtained from non-linear regression fitting of the Michaelis-
Menten curves by Statistica 13.4 using the equation V= Vmax×[S]/(KM + [S]) (Figures S9-S25 in
ESI). The kcat values were calculated according to equation kcat= Vmax/[Eo], where [Eo] represents
the enzyme concentration used in the kinetic assay. The velocity for each substrate concentration
was measured in triplicate.
5

2.7. Biotransformation of racemic phenylalanine derivatives (rac-1a-s) catalyzed by PzaPAL and
PcPAL
The ammonia elimination reaction mixtures containing 5 mM of the racemic phenylalanines
(rac-1a-s) in 100 mM TRIS buffer (pH 8.5) and purified PzaPAL or PcPAL (50 µg, either) in 1
mL reaction volume were incubated at 30 °C. Samples (50 μL) taken at different time points (17,
40, 64, and 168 h) from the reaction mixtures were analyzed by HPLC for conversion and ee
values, using previously developed methods.[44]
The ammonia addition and elimination reactions were followed by HPLC measurements.
Conversions were determined on Agilent 1200 HPLC instrument using Phenomenex Gemini NX-
C-18 column and a mixture of NH4OH buffer (0.1 M, pH 8.5) and MeOH at a flow rate of
1 mL min-1. Conversions were calculated from peak area integrations with use of appropriate
response factor (Table S2). Enantiomer separations were carried out on Agilent 1100 HPLC
instrument using Crownpak CR-I (+) column and a mixture of aqueous HClO4, pH 1.5 and
acetonitrile as eluent at a flow rate of 0.4 mL min-1. (Table S3).
2.8. Biotransformation of (E)-cinnamic acid derivatives (2a-s) catalyzed by PzaPAL and PcPAL
The reaction mixtures containing substituted (E)-cinnamic acids (2a-s, 5 mM) in NH2CO2NH4
(3M, pH 9.1) supplemented with PzaPAL (50 µg mL-1) or in NH3 (6M, pH 10) supplemented with
PcPAL (50 µg mL-1) were incubated at 30 °C. Samples (50 μL) taken at different time points (17,
40, 64, and 168 h) from the reaction mixtures were quenched by adding an equal volume of MeOH,
vortexed and centrifuged (13000 rpm, 2 min). The supernatant was transferred to a 0.22 μm filter
and used directly for HPLC analysis to determine conversion and ee values.
2.9. Homology modeling and simulation of enzyme-substrate interactions
PzaPAL initial homology models were created according to our previously published
protocol.[45] The initial 8 homology models were subjected to structure superposition, clustering,
and a brief CG minimization to yield an average cluster model with MODELLER.[46]This model
was completed and adjusted corresponding to the assay conditions of the kinetics characterization
of ammonia elimination (100 mM TRIS buffer, pH 8.5). Protein Preparation Wizard[47,48] of
Maestro[49] was used in two steps: i) hydrogen atoms were added and bond orders were assigned,
and ii) the hydrogen bond network, tautomeric states, side chain conformations of selected amino
acids, and ionization states were determined and optimized (pH=8.5). In all four active centers,
Tyr91 was set deprotonated and Tyr361 protonated. The L-phenylalanine was docked in the active
center of chain A similarly to our previously published induced-fit, covalent docking protocol.[50]
Leu115, Leu262, Leu265, Met118, Gln119 of chain A and Ile472, and Lys468 of chain C were
mutated to alanine, the MIO group was mutated to alanine and glycine. The L-phenylalanine was
docked in a 272727 Å3 grid, centered between the phenolic oxygen of Tyr91 and the exocyclic
methylene carbon of the MIO prosthetic group using Glide.[51] Then, all side chains and the MIO
group were restored. A covalent bond was created between the amino group of the substrate L-
phenylalanine and the exocyclic carbon of MIO. The ligands and the residues in close proximity
were minimized, and finally, redundant conformations were eliminated with MacroModel.[52]
After this step, side chain conformations of the previously restored residues (except for Lys468 of
chain C and the MIO group) were predicted with Prime[53] (with Cα-Cβ vectoring, vacuum, 2
steps).
Next, water molecules in the active center and its immediate vicinity were predicted and placed
with grand canonical Monte Carlo simulation with the following parameters using Desmond,[54]
6

[T=303K; 202020 Å3 box size] Simulations were carried out 15 times in parallel (with different
random seeds), 500.000 number of passes were used for each simulation, and the final waters were
determined upon the most frequent number and configuration of water molecules of the last 10.000
passes.
The final model was minimized with Prime[53] [RMSG: 0.5 kcal mol-1 Å-1, algorithm: BFGS,
implicit water solvation model: VSGB).
This structure was simulated with Desmond in an orthorhombic cell. The box was solvated with
TIP3P waters, neutralized with protonated Tris and augmented with 0.072 M Tris, 0.028 M
protonated Tris and chloride anion (calculated with a pKa of 8.1). The system was then equilibrated
for i) 100 ps with Brownian dynamics at 10K, NVT ensemble; ii) 12 ps with molecular dynamics
at 10 K, NVT ensemble; iii) 12 ps with molecular dynamics at 10 K, NPT ensemble, and iv) 12
and 24 ps with molecular dynamics at 303K, NPT ensemble. Harmonic restraints (50 kcal mol-1Å-
1) were applied to protein heavy atom positions in all steps. The OPLS3e force field was applied
in all molecular mechanics calculations and simulations.
Panels of Figure 3 were created with Maestro[49] and Pymol[55].
3. Results and discussion
To our satisfaction, the sequence similarity search allowed the identification of a sequence in
the Pseudozyma antarctica genome[56] using the highly conserved and the more variable
sequence motifs of the MIO-enzyme active site regions [sequence portions of the phenylalanine
2,3-aminomutase of Taxus canadensis (TcPAM) and parsley PAL (PcPAL)] against translated
nucleotide sequences of yeasts resembling to the MIO-enzymes (UniProt ID: M9M0D4). Since P.
antarctica is a well-known yeast isolated from cold environment and producing industrially highly
important lipases, the PAL of this yeast has been selected as target of the present study. After its
identification, the nucleotide sequence of the phenylalanine ammonia-lyase Pseudozyma
antarctica (PzaPAL) gene was optimized for the E. coli codon usage. The reoptimized synthetic
gene of PzaPAL was overexpressed in E. coli using a proper production vector coding the His-
tagged PzaPAL protein. In this study the enzymatic properties and biocatalytic applicability of
PzaPAL (Scheme 1) were investigated. First, kinetic resolution of racemic substituted
phenylalanines 1a-s by the enzymatic ammonia elimination was studied leading to substituted
cinnamic acids and to the residual (R)-phenylalanines. Then, the PzaPAL-catalyzed enantiotope
selective ammonia addition to substituted cinnamic acids 2a-s was studied leading to the
corresponding L-amino acids in high enantiopurity.
7

Scheme 1. Substrates and products in PzaPAL-catalyzed kinetic resolution of substituted racemic arylalanines rac-
1a-s and enantiotope selective ammonia addition onto (E)-arylacrylates 2a-s
3.1. Identification of the MIO-containing phenylalanine ammonia-lyase from Pseudozyma
antarctica (PzaPAL)
The PzaPAL gene was identified by BLAST screening yeast genes in the UniProt[57] and
GeneBank[58] databases against sequence portions of the aromatic binding pocket of TcPAM and
PcPAL, since variations in the substrate affinity of MIO enzymes is modulated mostly by the
interactions of the substrate molecule with the amino acid side chains of the aromatic binding
region (Figure 1).
The PzaPAL has an extra loop with 26 amino acids connected to the aromatic binding pocket
containing a -QMQQ- motif of in place of the -xFLx- motif usually present in other PALs which
is not typical for the most of MIO enzymes (Figure 1). This -QMQQ- motif and the extra loop
region of similar length could be identified only in sequences of closely related fungal PALs of
Ustilago maydis (UmPAL)[59] or Ustilago hordei (UhPAL).[60] The three glutamines within the
aromatic binding pocket of PzaPAL can establish more polar interactions with substituted
arylalanines than the conventional PALs with less polar motifs such as -RFLG- in RxPAL, -
WFLK- in AvPAL, -RFLN- in PcPAL and AtPAL. The -QMQQ- motif differs significantly from
the moderately polar motifs of -HHLA- in RsTAL, -RHLN- in SbPAL, -EHQL- in RtTAL; also,
from the -TFHG- motif in PfXAL or from the -LSHA- motif in PfHAL and PpHAL. The extra
loop with 26 amino acids following the aromatic binding pocket motif in PzaPAL is not typical
for most of the MIO enzymes. The RtTAL contains also an extra loop but only with 14 amino
8

acids, while PcPAL, AtPAL and TcPAM contains even shorter extra loops. The extra loop is not
present in MIO-enzymes of prokaryotes.
The presence of the highly conserved catalytic tyrosine base at the position 91 in PzaPAL is
consistent with the other MIO enzymes (Figure 2). The highly conserved amino acid triad ASG
forming MIO is also present in PzaPAL and most of the MIO-enzymes (except the bacterial PAMs
having a TSG triad as precursor of the MIO catalytic group). The MIO group stabilizing residues
(N266 and Y361) and the substrate carboxylic acid binding arginine (R364) is also highly
conserved in PzaPAL and in other MIO enzymes (Figure 2).
Figure 1. Sequence alignment of the aromatic binding pocket region including an extended loop in various MIO-
enzymes [Pseudozyma antarctica PAL (PzaPAL), Ustilago maydis PAL (UmPAL), Rubrobacter xilanophilus PAL
(RxPAL), Anabaena variabilis PAL (AvPAL), Petroselinum crispum PAL (PcPAL), Arbidopsis thaliana PAL
(AtPAL), Sorghum bicolor PAL (SbPAL), Rhodobacter spheroides TAL (RsTAL), Rhodosporidum toruloides TAL
(RtTAL), Pseudomonas fluorescens XAL (PfXAL), Pseudomonas fluorescens HAL (PfHAL), Pseudomonas putida
HAL (PpHAL), Pseudomonas fluorescens PAM (PfPAM), Pantenoa agglomerans PAM (PaPAM), Taxus canadensis
PAM (TcPAM), Chondromyces crocatus TAM (CcTAM) and Streptomyces globisporus TAM (SgTAM)] – color
codes: pink - the aromatic pocket motif of PzaPAL; blue: important amino acids in the aromatic binding pocket of
other MIO-enzymes.
Figure 2. Sequence alignment of the highly conserved catalytic domains in various MIO-enzymes. Sequence
alignment from by the presented MIO-enzymes. Color codes: red – catalytically essential tyrosine; orange – the MIO
forming amino acid triad; green – amino acids stabilizing MIO group; yellow – arginine responsible for binding the
carboxylic group of substrate; purple – amino acids related to the (S)-PAM activity.
9

3.2. Homology model of PzaPAL
The detailed homology model of PzaPAL was created with our previously published[45] and
updated protocol. Since the point mutation of a surface exposed cysteine into serine in the C-
terminal part did not decrease the catalytic activity of PcPAL, but decreased the aggregation ability
of the enzyme expressed in E. coli,[61] we followed a similar strategy in this study as well. The
homology model of PzaPAL showed a surface exposed cysteine at the position 655 which
presumably does not have influence on the catalytic activity but contributes to the aggregation
ability of the homotetrameric enzyme. Thus, the point mutation of this cysteine into serine (C655S)
was introduced to the synthetic gene of PzaPAL to reduce the opportunity of enzyme aggregation.
The homology model including simulated structural waters revealed the three conserved water
molecules in close proximity to the MIO which were also present in the structures of other PALs
(Figure 3, panel A, HOH1-3). However, the PzaPAL homology model predicted an additional
water molecule in the active center, being atypical for PALs (Figure 3, panel A, HOH4). This
water establishes four hydrogen bonds with neighboring residues HOH3, Ser201, Gly467, and
with the also atypical Gln119 which belongs to the -QMQQ- motif. This water can play an
important role in the orientation of Gln119 pointing the amide carbonyl oxygen towards the
aromatic ring of the substrate. The excess negative charge on this carbonyl oxygen can decrease
the electron withdrawing (EW) effect of EW substituents at position 4 of the aromatic ring in
substituted phenylalanines.
The homology model predicts a possible function of the extended loop with the 26 extra amino
acids [loop1, residues 121-150]. The loop1 in two of the four of peptide chains of the
homotetrameric PzaPAL establishes interactions with each other (Figure 3, panel C). Moreover,
another shorter loop [loop2, residues 226-237] is involved in the local loop network (Figure 3,
panel B). Since loop1 is directly connected to the -QMQQ- motif of the aromatic binding pocket
(Figure 3, panel B and C), it can be assumed that this interacting loop network is responsible for
allosteric modulation of adjacent active sites within the tetrameric PzaPAL during substrate
transformation.
3.3. Cloning and expression of PzaPAL
The PzaPAL gene was codon optimized for the E. coli codon usage and synthetized in
pUC57 vector. The gene was inserted in to pET-19b expression vector with NdeI and BamHI
restriction endonucleases to encode the N-terminal His10-tagged form of the PzaPAL. The
transformed E. coli Rosetta (DE3)pLysS cells with pET-19b_PzaPAL expression vector were
grown in LB medium supplemented with carbenicillin and chloramphenicol at 37 °C, and PzaPAL
expression was induced with addition of IPTG at 25 °C. Purification of the His-tagged PzaPAL
was performed with Ni-NTA agarose affinity chromatography.
A
10

B
C
Figure 3. The catalytic site (A) and the extended loops (B, C) in the structure of tetrameric PzaPAL (A) predicted by
homology modeling. [Panel A: overlay of the active centers of PzaPAL and PcPAL displaying identical residues next
to the substrate and three identically positioned waters HOH1-3 by sticks representation and the main differences in
the two active sites M118/F137, Q119/L138, HOH4 (PzaPAL/PcPAL, respectively) by ball and stick representation.
HOH4 interactions are displayed with S201 and G467. Panel B: chains A and C of the homology model displaying
the extended loops of PzaPAL (loop 121-150 in dark green, loop 226-237 in light green) compared to the
corresponding shorter loops of PcPAL (loop 141-150 in dark red, loop 223-232 in light red) and even shorter loops of
AvPAL (loop 189-198 in dark blue, loop 109-115 in light blue). Panel C: the extensive interactions of the extended
loops 121-150 and 226-237 within the tetramer of PzaPAL].
3.4. pH optimum of the PzaPAL-catalyzed elimination reaction
The effect of the pH on the ammonia-lyase activity of PzaPAL with L-phenylalanine was
investigated between 6.0 and 12.0 (Figure 4). The maximum activity was observed at pH 8.5 which
is similar to UhPAL (pH 8.8)[60] and UmPAL (pH 8.8-9.2).[59] Differential thermal calorimetric
measurements indicated that the PzaPAL enzyme retained its thermal stability in a broad pH range
from 7.0 to 10.5, while at higher pH values the structural stability of the enzyme decreased
considerably (Figure 5).
11

100
80
60
40
20
0
6
6.5
7
7.5
8
8.5
9
9.5 10 10.5 11 11.5 12
pH
Figure 4. Effect of the pH upon PzaPAL activity in presence of L-Phe.
pH 7.0
pH 8.5
pH 10.0
pH 11.5
pH 7.5
pH 9.0
pH 10.5
pH 12.0
pH 8.0
pH 9.5
pH 11.0
0.040
0.035
0.030
0.025
0.020
0.015
0.010
0.005
0.000
20 25 30 35 40 45 50 55 60 65 70 75 80 85 90
Temperature [°C]
Figure 5. Thermal unfolding curves of PzaPAL at different pH values by nano-DSF measurements.
3.5. Thermal behavior of PzaPAL
The effect of temperature on enzyme activity of PzaPAL (without prior incubation) was assayed
at different temperatures ranging from 5-60 °C at the optimal pH value. A strong temperature
dependence of the enzyme activity with temperature was detected, displaying highest value at
45 °C (Figure 6). However, it is worth mentioning that despite of the psychrophilic origin of
PzaPAL, the high melting temperature (77.5 °C) of the enzyme was found to be superior compared
to most of other known PALs (see ESI 1.12 and 1.14).
12

100
80
60
40
20
0
5
15 20 25 30 35 40 45 50 55 60
Temperature [°C]
Figure 6. Temperature effect upon PzaPAL activity using L-Phe as substrate (without preincubation).
PzaPAL was also pre-incubated at different temperatures in the range of 10-50 °C, at the optimal
pH value for 24 h and assayed afterwards. The results (Figure S3 in ESI) showed that in the range
of 10-40 °C after 8 h preincubation the enzyme retained more than 90% of its original activity.
Considerable activity decrease was observed after 24 h preincubation at 40 and 50 °C. This was in
good agreement with the behavior of UhPAL remaining active after 10 min incubation at 50 °C
but inactivating at 60 °C and 70 °C.[60]
The enzyme was pre-incubated also in 6M NH3, pH 10.0 (adjusted by bubbling CO2)at different
temperatures ranging from 10-50 °C for 24 h and assayed afterwards. The PzaPAL inactivated
rapidly in 6M NH3, after 24 h only the enzyme incubated at 10 °C exhibited some residual activity,
while in 0.1M TRIS-buffer the enzyme activity was well preserved (Figure S5 in ESI).
3.6. Effect of salt concentration on the activity of PzaPAL in the elimination from L-Phe
Various NaCl concentrations were studied for their impact on PzaPAL activity at pH 8.5. As
NaCl concentration was increased from 0 to 500 mM, the activity decreased monotonously (Figure
S6 in ESI). According to this analysis, the enzyme presented the maximal thermal stability in the
absence of NaCl, therefore no salt was used in the buffer solutions (Figure S7 and Table S9 in
ESI).
3.7. Kinetic characterization and substrate scope of the PzaPAL-catalyzed ammonia elimination
Besides with L-Phe [(S)-1a], the kinetics of ammonia elimination by PzaPAL was characterized
with several racemic phenylalanine derivatives rac-1a-g,k (Table 1). The non-conventional
aromatic binding pocket of the PzaPAL enzyme, compared to other PALs, resulted in significant
differences in the catalytic activity and substrate affinity (Table 1).
The KM value of L-phenylalanine (S)-1a with PzaPAL (KM= 0.83 mM) was similar to those
found with the closely related UhPAL (KM= 0.45 mM)[60] and UmPAL (KM= 1.05 mM),[59] one
order of magnitude higher than with PcPAL (KM= 0.083 mM),[50] AvPAL (KM= 0.06 mM),[62]
AtPAL1 (KM = 0.068 mM),[63] but smaller than with PfXAL (KM= 2.6 mM).[25] PzaPAL had an
about two times higher turnover number (kcat= 2.97 s-1) than PcPAL (kcat= 1.38 s-1). This turnover
value was similar to that of PfXAL (kcat= 2.38 s-1)[25] and somewhat lower than of AvPAL (kcat=
4.3 s-1).[62] Temperature-dependent kinetics of PzaPAL were studied with L-phenylalanine (S)-
1a at 30°C (Figure S8 in ESI) and 42°C (Figure S9 in ESI) showing weakly decreasing KM values
(KM(30°C)= 0.8 mM; KM(42°C)= 1.1 mM) with significantly increasing turnover numbers (kcat(30°C)
=
2.97 s-1; kcat(42°C)= 8.2 s-1).
13

Table 1. Kinetic parameters of PzaPAL and PcPAL in ammonia
elimination reactions of several phenylalanine derivatives.
Substrate
PzaPAL
PcPAL
KM
[mM]
kcat
KM
[mM]
kcat
[s-1]
[s-1]
(S)-1a (Ph)
rac-1b (o-F)
rac-1c (m-F)
rac-1d (p-F)
rac-1e (o-Cl)
rac-1f (m-Cl)
rac-1g (p-Cl)
rac-1k (o-Br)
0.83
2.33
1.13
1.69
2.94
4.48
0.055
0.079
0.147
0.024
0.083
0.041
0.033
0.068
1.38
1.01
0.96
0.41
0.95
0.96
0.55
0.88
8.00
4.22
>2.82[a]
1.45
1.03[b]
1.92
>10.8[a]
>2.62[a]
0.26[b]
0.69[b]
[a] The KM value could not be determined exactly because at high
substrate concentration the PzaPAL-catalyzed reaction did not approach
Vmax. [b] The kcat parameter was determined at the highest achievable
substrate concentration of the kinetics study with different enzyme
concentrations.
Kinetics investigations of AvPAL with rac-1b (KM= 0.75 mM, kcat= 1.1 s-1) and with rac-1d
(KM= 0.56 mM, kcat= 1.06 s-1) were also performed.[23] The KM values AvPAL for the o- and p-
fluorophenylalanines rac-1b,d were 2-3 times smaller than for PzaPAL, but PzaPAL had 3-4 times
higher turnover values with these substrates as AvPAL.
The comparative kinetic studies with PzaPAL and PcPAL using racemic substituted
phenylalanines rac-1b-g,k revealed that the saturated rate (Vmax) was not achievable with rac-
1e,g,k in the investigated substrate concentration range due to the limited solubility of the
unnatural amino acids. Therefore, only the kcat values were determined at the highest achievable
substrate concentration with different enzyme concentrations (Figures S13, S15, S16 in ESI). The
fluoro- (rac-1b-d), chloro- (rac-1e-g) and o-bromo- (rac-1k) phenylalanines had also higher KM
values with PzaPAL than with PcPAL, but again the turnover values with PzaPAL are higher than
with PcPAL. Presumably, the completely different and more polar amino acids in the aromatic
binding pocket of PzaPAL could cause this increase of turnover values compared to the
“conventional” PALs.
In addition to L-phenylalanine (S)-1a, PzaPAL was tested with several further natural amino
acids for ammonia elimination reaction. L-Tyr was found to be a quite poor substrate for PzaPAL
(1 % relative activity compared to that with L-Phe), similar to UhPAL[60] and UmPAL[59] which
were reported having no reaction with L-Tyr. Furthermore, PzaPAL did not catalyze the ammonia
elimination from L-His, L-Trp, L-Asp, L-Asn or L-Gln.
3.8. Kinetic resolution of rac-phenylalanine derivatives (rac-1a-s) mediated by PzaPAL
The biocatalytic properties of PzaPAL were investigated first by kinetic resolution of rac-1a-s
using the ammonia elimination reactions. The biocatalytic performance of PzaPAL was compared
to the already well-characterized PAL from parsley (PcPAL) (Table 2).
High enantiomer selectivity was observed in most of the PzaPAL-catalyzed kinetic resolutions
(plain typed values in Table 2, with conversion values within 3% deviation for the calculated full
selectivity case) except the ammonia elimination from the racemic o-nitro- (rac-1h), p-nitro- (rac-
14

1j), p-bromo- (rac-1m) phenylalanines. In the case of attempted kinetic resolutions with PzaPAL
using phenylalanines substituted with electron-donating groups (o-, m-, or p-methyl: rac-1n-o; o-,
m-, or p-methoxy: rac-1q-s) reactions proceeded with moderate, low or even negligible conversion
(no product was detected with o- and p-methoxyphenylalanines rac-1q,s) but with perfect
enantiomer selectivity (for rac-1n-p,r).
Table 2. Kinetic resolution of racemic phenylalanine derivatives (rac-1a-s) catalyzed
by PzaPAL (as compared to by PcPAL).
after 17 h
after 168 h
Subst.
PzaPAL
Conv.[a]
PcPAL
PzaPAL
Conv.[a]
PcPAL
Conv.[a]
Conv.[a]
(ee(R)-1a-s
)
(ee(R)-1a-s
)
(ee(R)-1a-s
)
(ee(R)-1a-s)
[%]
[%]
[%]
[%]
rac-1a (Ph)
rac-1b (o-F)
49 (>99)
49 (>99)
49 (>99)
51 (>99)
48 (98)
47 (88)
44 (82)
35 (43)
21 (26)
56 (88)
43 (79)
29 (39)
18 (17)
50 (>99)
10 (12)
49 (>99)
52 (>99)
52 (>99)
54 (>99)
49 (>99)
51 (>99)
53 (>99)
67 (>99)
63 (>99)
96 (>99)
49 (>99)
50 (>99)
50 (>99)
49 (>99)
29 (45)
49 (>99)
53 (>99)
52 (>99)
51 (>99)
50 (>99)
50 (>99)
50 (>99)
62 (>99)
40 (64)
78 (94)
49 (99)
54 (>99)
69 (>99)
63 (>99)
59 (>99)
64 (>99)
61 (>99)
64 (>99)
82 (>99)
rac-1c (m-F)
rac-1d (p-F)
rac-1e (o-Cl)
rac-1f (m-Cl)
rac-1g (p-Cl)
rac-1h (o-NO2)
rac-1i (m-NO2)
rac-1j (p-NO2)
rac-1k (o-Br)
rac-1l (m-Br)
rac-1m (p-Br)
rac-1n (o-CH3)
rac-1o (m-CH3)
rac-1p (p-CH3)
rac-1q (o-CH3O)
rac-1r (m-CH3O)
rac-1s (p-CH3O)
91 (>99)
[c]
99[b]
( )
58 (>99)
60 (>99)
53 (>99)
52 (>99)
48 (>99)
48 (>99)
25 (30)
47 (87)
38 (58)
50 (>99)
18 (23)
1 ([c]
0 ([c]
)
)
47 (94)
6 (7)
7 (6)
0 ([c]
35 (53)
0 ([c]
)
20 (25)
0 ([c]
48 (>99)
22 (25)
51 (>99)
37 (57)
)
)
[a] Conversions were determined by HPLC on C-18 reversed phase.
Conversion values in plain: high enantiomer selectivity, conversion values
in underscored bold: low enantiomer selectivity due to partial consumption
of the (R)-enantiomer. [b] Both enantiomers were consumed. [c] Not
determined.
3.9. The PzaPAL-catalyzed ammonia addition reaction onto cinnamic acid derivatives 2a-s
Addition reactions with various PALs were typically performed in 5-6M NH3, pH 10 (adjusted
by bubbling CO2) at 30 °C or 37 °C.[11] Herein, we investigated the optimal reaction conditions
for PzaPAL in the enantiotope selective ammonia addition reaction, including the temperature, the
optimal pH, the effect of the ammonia source and concentration.
15

3.10. Temperature effect on the ammonia addition reaction onto 2a by PzaPAL
Because PzaPAL originated from Pseudozyma antarctica, a yeast first isolated from cold
Antarctic environment,[39] we also investigated the effect of the temperature on the amination
reaction of (E)- cinnamic acid. The activity of PzaPAL pre-incubated in 6M NH3 at different
temperatures (10-50 °C: Figure S5 in ESI) decreased rapidly with increasing temperature
(negligible residual activity of PzaPAL was observed after 1 h preincubation at 40 or 50 °C).
Therefore, the detailed temperature effect study on the addition reaction was performed in the
range of 10-30 °C (Table S10 in ESI) indicating maximal activity at 30 °C. Consequently, all
further addition reactions were performed at 30 °C.
3.11. Effect of pH on the ammonia addition reaction onto 2a by PzaPAL
One of the most important factors influencing enzyme activity is the pH. Therefore, the effect
of pH on the ammonia addition reaction was also investigated using 6M NH3 solution in the range
pH 9.1–10.0 (adjusted by bubbling CO2). Buffering the reaction to lower pH values of the NH3-
containing medium was beneficial and the PzaPAL-catalyzed conversion of (E)-cinnamic acid 2a
to L-phenylalanine (S)-1a was maximal at pH 9.1 (Table S11 in ESI). It was not possible to test
NH3 buffer solutions of 6M at pH values under 9.1.
3.12. Effect of the ammonia source and concentration on the ammonia addition reaction onto 2a
by PzaPAL
The concentration of ammonia has a significant effect on the rate of reaction and position of
equilibrium. In order to achieve good conversions, 5-6M ammonia is typically used in the PAL-
catalyzed ammonia addition reactions.[11] The effect of different ammonia sources on PzaPAL-
catalyzed (E)-cinnamic acid 2a amination was investigated using purified enzyme at the optimal
pH (9.1) in 0.5-5M NH3 and in 0.5-4M NH2CO2NH4 buffers, respectively. It was shown that in
both buffers the reaction was fully enantiotope selective providing the product (S)-1a in
enantiomeric excess exceeding 99%. The conversion (and its rate) was significantly lower at 0.5M
concentration of ammonia than at 3-5M (Table 3 and S12 in ESI). The highest conversion of
amination could be achieved using ammonium carbamate as an ammonia source at a concentration
of 3M. Therefore, 3M NH2CO2NH4 was selected as the ammonia source for the further
investigations on ammonia addition onto substituted (E)-cinnamic acids 2a-s (Table 4).
Table 3. Biotransformation of (E)-cinnamic acid (2a) catalyzed by PzaPAL at different concentrations of NH2CO2NH4 buffer.
Buffer conc.
[M]
Conversion in NH2CO2NH4 [%]
4 h
18
20 h
34
43 h
33
67 h
35
0.5
1
20
17
16
11
48
52
52
32
54
70
79
68
53
71
83
82
2
3
4
3.13. The substrate scope of PzaPAL-catalyzed enantiotope selective ammonia addition reaction
onto cinnamate derivatives 2a-s
16

Most of the o-substituted (E)-cinnamic acids (2b,e,h,k,n) were proper substrates in the
PzaPAL-catalyzed enantiotope selective ammonia addition, with the exception of (E)-o-methoxy-
cinnamic acid 2q which was not transformed at all.
Table 4. Enantiotope selective ammonia addition onto (E)-arylacrylates (2a-s)
catalyzed by PzaPAL (as compared to by PcPAL).
after 17 h
after 168 h
Substrate
PzaPAL
Conv.
PcPAL
PzaPAL
Conv.
PcPAL
Conv.
Conv.
[a]
[a]
[a]
[a]
(ee(S)-1a-s
)
(ee(S)-1a-s
)
(ee(S)-1a-s
)
(ee(S)-1a-s)
[%]
[%]
[%]
[%]
2a (Ph)
2b (o-F)
41 (>99)
94 (>99)
93 (>99)
11 (>99)
98 (>99)
5 (91)
81 (>99)
96 (97)
80 (>99)
96 (>99)
93 (>99)
51 (>99)
98 (>99)
20 (96)
9 (>99)
45 (92)
4 (89)
81 (98)
95 (92)
89 (97)
78 (98)
98 (70)
93 (93)
91 (96)
98 (82)
89 (78)
97 (62)
97 (95)
93 (95)
90 (>99)
84 (99)
74 (97)
55 (>99)
49 (>99)
71 (>99)
2c (m-F)
89 (97)
2d (p-F)
78 (>99)
97 (94)
2e (o-Cl)
2f (m-Cl)
93 (97)
2g (p-Cl)
3 (>99)
32 (92)
2 (78)
92 (99)
2h (o-NO2)
2i (m-NO2)
2j (p-NO2)
2k (o-Br)
98 (89)
80 (89)
78 (92)
41 (>99)
97 (75)
91 (92)
95 (>99)
5 (>99)
97 (>99)
92 (>99)
88 (>99)
84 (>99)
44 (>99)
34 (>99)
21 (>99)
58 (>99)
2l (m-Br)
1 ([b]
0 ([b]
)
)
2m (p-Br)
2n (o-CH3)
2o (m-CH3)
2p (p-CH3)
2q (o-CH3O)
2r (m-CH3O)
2s (p-CH3O)
1 ([b]
)
27 (>99)
74 (>99)
0 ([b]
0 ([b]
0 ([b]
0 ([b]
0 ([b]
)
)
)
)
)
0 ([b]
0 ([b]
0 ([b]
0 ([b]
0 ([b]
)
)
)
)
)
0 ([b]
)
0 ([b]
)
[a] Enantiomeric excess values were determined by HPLC on Crownpak CR-I (+)
chiral column. Ee values in plain: perfect enantiotope selectivity (no (R)-enantiomer
could be detected); Ee values in underscored bold: non-perfect enantiotope
selectivity due to formation of the (R)-enantiomer. [b] Not determined.
The PzaPAL-catalyzed ammonia addition onto substituted (E)-cinnamic acids 2a-e,g,k,l,n
yielded the corresponding enantiopure L-phenylalanines (S)-1a-e,g,k,l,n with variable
conversions. Excellent conversions (>90%) were achieved with the o-F, m-F and o-Cl substituted
cinnamates 2b,c,e within 17 h, and with the o-Br substituted cinnamate after 168 h. The absent or
negligible conversion (<5% within 168 h) observed with substituted (E)-cinnamic acids 2i,m,o-s
could be rationalized by the increased negative charge of the aromatic binging pocket due to the
partial negative charge on Q119 carbonyl oxygen. The excess negative charge in the aromatic
17

binding pocket in the neighborhood of the p- and m-position of the phenyl ring of the substrate
(Figure 3A) could enhance the electron donating effect of the methyl (2n-p) or methoxy (2q-s)
substituents present in substituted (E)-cinnamic acids.
As the reaction should involve an interaction of the forming amino-MIO intermediate as
nucleophile with an electrophilic intermediate forming from the substrate cinnamate, the
increasing electron donating effect of the aromatic ring decreases the reactivity in the ammonia
addition reaction.
In case of the nitro-substituted (E)-cinnamic acids (2h-j) more rigid aromatic binding pocket
partially hindered the reaction due to steric effects but the more electron rich environment could
contribute to the higher degree of enantiotope selectivity of the PzaPAL-catalyzed ammonia
addition onto nitro-substituted (E)-cinnamic acids (2h-j) than found with the PcPAL indicated by
the higher enantiomeric excess of nitrophenylalanines (S)-1h-j from the PzaPAL-catalyzed
reaction after 168 h as compared to the corresponding values from the PcPAL-catalyzed process.
Interestingly, while in many instances the PzaPAL-catalyzed process after 168 h reaction time
resulted in enantiopure products (ee(S)-1a-e,g,k,l,n >99 %; with the exception of the m-chloro-, o-, m-
and p-nitro-substituted cinnamates 2f,h,i,j), the PcPAL-catalyzed ammonia addition provided
incomplete enantiotope selectivity onto cinnamates 2b-l and 2o.
4. Conclusions
In this study, the properties and biocatalytic applicability of a novel ammonia-lyase from
the yeast Pseudozyma antarctica were investigated. The PzaPAL gene was successfully identified,
cloned and expressed in E. coli and the forming enzyme was characterized. The PzaPAL exhibited
maximal activity at pH 8.5, similarly to other characterized PALs. The highest enzyme activity
was detected at 45 °C (without prior incubation) while thermal unfolding of the enzyme was
observed at 77.5 °C. Long-term stability tests indicated that in the range of 10-40 °C the enzyme
retained more than 90% of its initial activity within 8 h. Despite of the lower affinity of
phenylalanines (S)-1a and rac-1b-f to PzaPAL than to PAL from parsley (PcPAL), with PzaPAL
higher turnover number/catalytic efficiency were observed than with PcPAL.
The significantly different catalytic activity of PzaPAL could be attributed to the completely
different amino acids in the aromatic binding pocket as compared to PcPAL. The structure of
PzaPAL predicted by homology modeling showed a local network of extended loops and predicted
the presence of an atypical, bound water molecule partly responsible for the orientation of Gln119
enhancing the local negative charge in the aromatic binding region. The PzaPAL-catalyzed
elimination reactions from racemic phenylalanines with moderately bulky substituents rac-1a-
g,k,n proceeded with high enantiomer selectivity providing at near to 50 % conversion the residual
enantiopure phenylalanine (R)-1a-g,k,n. Elimination from rac-1i,m,o,p,r bearing larger but not
extremely charged substituents proceeded with moderate to low conversion, while no product was
detected from phenylalanines rac-1q,s bearing o- or p-methoxy substituents. While the PcPAL-
catalyzed ammonia addition onto 2b-l and 2o proceeded with incomplete enantiotopic selectivity,
PzaPAL yielded from a number of substituted cinnamates 2a-e,g,k,l,n the enantiopure products
(ee(S)-1a-e,g,k,l,n >99%) These properties render PzaPAL as catalyst of choice for preparation of a
number of substituted (R)- or (S)-phenylalanines in enantiopure form. Moreover, due to its
remarkably different aromatic binding site features, PzaPAL is a good initial candidate for
biocatalyst engineering with targeted mutations for altered substrate scope leading to further
unnatural α-amino acids in high enantiopurity.
Credit author statement
18

Andrea Varga: Investigation, Methodology, Editing, Writing - Original Draft; Pál Csuka:
Investigation, Editing; Orlavanah Sonesouphap: Investigation; Gergely Bánóczi: Software,
Visualization; Monica Ioana Toșa: Methodology, Project administration, Formal analysis; Gabriel
Katona: Validation, Resources; Zsófia Molnár: Investigation; László Csaba Bencze: Writing -
Review & Editing, Supervision; László Poppe: Conceptualization, Writing - Review & Editing,
Supervision, Project administration, Funding acquisition; Csaba Paizs: Conceptualization,
Methodology, Project administration, Supervision, Writing - Review & Editing.
Declaration of interests
☒ The authors declare that they have no known competing financial interests or personal relationships
that could have appeared to influence the work reported in this paper.
Acknowledgements
This work was supported by the project “Entrepreneurship for innovation through doctoral and
postdoctoral research”, POCU/360/6/13/123886 co-financed by the European Social Fund,
through the Operational Program for Human Capital 2014-2020, by project NEMSyB [ID
P37_273, Cod MySMIS 103413; funded by National Authority for Scientific Research and
Innovation (ANCSI) and co-funded by the European Regional Development Fund,
Competitiveness Operational Program 2014-2020 (POC) and by the Swiss National Science
Foundation (SNSF) - project PROMYS, grant nr. IZ11Z0_166543. This work was also supported
by the Higher Education Excellence Program of the Ministry of Human Capacities (Budapest,
Hungary) in the frame of Biotechnology research area of Budapest University of Technology and
Economics (BME FIKP-BIO), the National Research, Development and Innovation Fund of
Hungary (Budapest, Hungary; project SNN-125637) and by CELSA (BME-KU Leuven, ConSolid
project).
References
[1] F. Agostini, J. S. Völler, B. Koksch, G. C. Acevedo-Rocha, V. Kubyshkin, N. Budisa, Angew. Chem. Int. Ed.
56 (2017) 9680–9703.
[2] Biocatalysis: An Industrial Perspective (G. de Gonzalo Calvo, P. Domínguez de Maria, Eds.), RSC Publishing:
Cambridge, 2018.
[3] K. Faber, Biotransformations in Organic Chemistry, 7th Ed., Springer: Heildelberg, 2018.
[4] M. Jin, M. A. Fischbach, J. Clardy, J. Am. Chem. Soc. 128 (2006) 10660–10661.
[5] S. D. Christenson, W. Liu, M. D. Toney, B. Shen, J. Am. Chem. Soc. 125 (2003) 6062–6063.
[6] C. L. Steele, Y. Chen, B. A. Dougherty, W. Li, S. Hofstead, K. S. Lam, Z. Xing, S. J. Chiang, Arch. Biochem.
Biophys. 438 (2005) 1–10.
[7] K. D. Walker, K. Klettke, T. Akiyama, R. Croteau, J. Biol. Chem. 279 (2004) 53947–53954.
[8] G. V. Louie, M. E. Bowman, M. C. Moffitt, T. J. Baiga, B. S. Moore, J. P. Noel, ChemBiol 13 (2006) 1327–
1338.
[9] F. Kudo, A. Miyanaga, T. Eguchi, Nat. Prod. Rep. 31 (2014) 1056–1073.
[10] A. Stevenazzi, M. Marchini, G. Sandrone, B. Vergani, M. Lattanzio, Bioorg. Med. Chem. Lett. 24 (2014) 5349–
5353.
[11] F. Parmeggiani, N. J. Weise, S. Ahmed, N. J. Turner, Chem. Rev. 118 (1) (2018) 73–118.
[12] a) L. Poppe, Curr. Opin. Chem. Biol. 2001, 5, 512–524; b) L. Poppe, J. Rétey, Curr. Org. Chem. 7 (2003) 1297–
1315.
[13] N. J. Turner, Curr. Opin. Chem. Biol. 15 (2011) 234–240.
19

[14] L. Poppe, C. Paizs, K. Kovács, F. D. Irimie, B. G. Vértessy, Meth. Mol. Biol. 794 (2012) 3–19.
[15] B. Wu, W. Szymański, M. M. Heberling, B. L. Feringa, D. B. Janssen, Trends Biotechnol. 29 (2011) 352–362.
[16] H. Morrison, C. Bernasconi, G. Pandey, Photochem. Photobiol. 40 (1984) 549–550.
[17] X. Zhang, C. Liu, Mol. Plant 8 (2015) 17–27.
[18] S. Yamada, K. Nabe, N. Izuo, K. Nakamichi, I. Chibata, Appl. Environ. Microbiol. 42 (1981) 773−778.
[19] G. Renard, J.-C. Guilleux, C. Bore, V. Malta-Valette, D. A. Lerner, Biotechnol. Lett. 14 (1992) 673−678.
[20] C. Paizs, A. Katona, J. Rꢀtey, Chem. Eur. J. 12 (2006) 2739−2744.
[21] N. J. Weise, F. Permeggiani, S. T. Ahmed, N. J. Turner, Top. Catal. 61 (2018) 288–295.
[22] S. T. Ahmed, F. Permeggiani, N. J. Weise, S. L. Flitsch, N. J. Turner, Org. Lett. 18 (2016) 5468–5471.
[23] S. L. Lovelock, N. J. Turner, Bioorg. Med. Chem. 22 (2014) 5555−5557.
[24] F. B. J. van Assema, N. Sereinig (DSM): Method for producing optically active phenylalanine compounds from
cinnamic acid derivatives employing a phenylalanine ammonia lyase derived from Idomarina loihiensis. WO
2008/031578 (2008) and PCT/EP 2007/007945 (2007).
[25] P. Csuka, V. Juhász, S. Kohári, A. Filip, A. Varga, P. Sátorhelyi, L. C. Bencze, H. Barton, C. Paizs, L. Poppe,
ChemBioChem, 19 (2017) 1–9.
[26] G. G. Wybenga, W. Szymanski, B. Wu, B. L. Feringa, D. B. Janssen, B. W. Dijkstra, Biochemistry 53 (2014)
3187−3198.
[27] B. Wu, W. Szymanski, G. G. Wybenga, M. M. Heberling, S. Bartsch, S. Wildeman, G. J. Poelarends, B. L.
Feringa, B. W. Dijkstra, D. B. Janssen, Angew. Chem. Int. Ed. 51 (2012) 482–486.
[28] K. T. Watts, B. N. Mijts, P. C. Lee, A. J. Manning, C. Schmidt-Dannert, Chem. Biol. 13 (2006) 1317–1326.
[29] a) S. T. Ahmed, F. Parmeggiani, N. J. Weise, S. L. Flitsch and N. J. Turner, ACS Catal. 5 (2015) 5410−5413; b)
A. Filip, E. Z. A. Nagy, S. D. Tork, G. Bánóczi, M. I. Toşa, F. D. Irimie, L. Poppe, C. Paizs and L. C. Bencze,
ChemCatChem, 10 (2018) 2627-2633; c) E. Z. A. Nagy, S. D. Tork, P. A. Lang, A. Filip, F. D. Irimie, L. Poppe, M. I.
Toşa, C. J. Schofield, J. Brem, C. Paizs and L. C. Bencze, ACS Catal. 9 (2019) 8825-8834.
[30] K. Kovács, G. Bánóczi, A. Varga, I. Szabó, A. Holczinger, G. Hornyánszky, I. Zagyva, C. Paizs, B. G. Vértessy,
L. Poppe, PLoS ONE 9 (1) (2014) e85943.
[31] A. Varga, Z. Bata, P. Csuka, D. M. Bordea, B. G. Vértessy, A. Marcovici, F. D. Irimie, L. Poppe, L. C. Bencze,
Studia Univ. Babes Bolyai, Ser. Chem. 63 (2017) 293−308.
[32] R. Cavicchioli, T. Charlton, H. Ertan, S. Mohd Omar, K. S. Siddiqui, T. J. Williams, Microb. Biotechnol. 4 (4)
(2011) 449-460.
[33] C. Gerday, M. Aittaleb, M. Bentahir, J. P. Chessa, P. Claverie, T. Collins, Trends Biotechnol. 18 (2000) 103–
107.
[34] R. Cavicchioli, K. S. Siddiqui, Enzyme Technology, Springer Science, New York, 2006, pp. 615–638.
[35] R. Cavicchioli, K. S. Siddiqui, D. Andrews, K. R. Sowers, Curr. Opin. Biotechnol. 13 (2002) 253–261.
[36] J. C. Marx, T. Collins, S. D’Amico, G. Feller, C. Gerday, Mar. Biotechnol. 9 (2007) 293–304.
[37] R. Margesin, G. Feller, Environ. Technol. 31 (2010) 835–844.
[38] J. Jeon, J. T. Kim, S. Kang, J. H. Lee, S. J. Kim, Mar. Biotechnol. 11 (2009) 307–316.
[39] S. Goto, J. Sugiyama, H. Iizuka, Mycologia, 1969, 61, 748–774.
[40] a) U.T. Bornscheuer, R.J. Kazlauskas, Hydrolases in Organic Synthesis, Regio- and Stereoselective
Biotransformations, 2nd ed., Wiley-VCH, Weinheim, 2006; b) V. Gotor-Fernández, E. Busto, V. Gotor, Adv.
Synth. Catal. 348 (2006) 797–812.
[41] E. Abaházi, D. Lestál, Z. Boros, L. Poppe, Molecules 21 (2016) 767.
[42] P. Csuka, Z. Boros, L. Őrfi, J. Dobos, L. Poppe, G. Hornyánszky, Tetrahedron: Asymmetry 26 (2015) 644–649.
[43] Z. Boros, P. Falus, M. Márkus, D. Weiser, M. Oláh, G. Hornyánszky, J. Nagy, L. Poppe, J. Mol. Catal. B Enzym.
85-86 (2013) 119–125.
[44] a) A. Varga, G. Bánoczi, B. Nagy, L. C. Bencze, M. I. Toşa, Á. Gellꢀrt, F. D. Irimie, J. Rꢀtey, L. Poppe and C.
Paizs, RSC: Adv. 6 (2016) 56412-56420; b) S. D. Tork, E. Z. A. Nagy, L. Cserepes, D. M. Bordea, B. Nagy, M. I.
Toşa, C. Paizs, L. C. Bencze, Sci. Rep. 9 (2019) 20123.
[45] G. Bánóczi, C. Szabó, Z. Bata, G. Hornyánszky, L. Poppe, Studia Univ. Babes Bolyai, Ser. Chem. 60(4) (2015)
213–228.
[46] A. Šali, T. L. Blundell, J. Mol. Biol. 234 (1993) 779–815.
[47] G. M. Satry, M. Adzhigirey, T. Day, R. Annabhimoju, W. Sherman, J. Comput. Aid. Mol. Des. 27 (2013) 221–
234.
[48] M. H. M. Olsson, C. R. Søndergard, M. Rostkowski, J. H. Jensen, J. Chem. Theor. Comput. 7 (2011) 525–537.
[49] Schrödinger Release 2018-4: Maestro, Schrödinger, LLC, New York, NY, 2018
[50] L. C. Bencze, A. Filip, G. Bánóczi, M. I. Toşa, F. D. Irimie, Á. Gellꢀrt, L. Poppe, C. Paizs, Org. Biomol. Chem.
15 (2017) 3717−3727.
[51] a) Schrödinger Release 2018-4: Glide, Schrödinger, LLC, New York, NY, 2018; b) R. A. Friesner, J. L. Banks,
R. B. Murphy, T. A. Halgren, J. L. Klicic, S. T. Mainz, M. T. Repasky, E. H. Knoll, D. E. Shaw, M. Shelley, J.
K. Perry, P. Francis, P. S. Shenkin, J. Med. Chem. 47 (2004), 1739-1749; c) T. A. Halgren, R. B. Murphy, R. A.
Friesner, H. S. Beard, L. L. Frye, W. T. Pollard, J. L. Banks, J. Med. Chem. 47 (2004), 1750-1759.
[52] Schrödinger Release 2018-4: MacroModel, Schrödinger, LLC, New York, NY, 2018.
[53] a) Schrödinger Release 2018-4: Prime, Schrödinger, LLC, New York, NY, 2018. b) M. P Jacobson, D. L. Pincus,
C.S. Rapp, T. J. F. Day, B. Honig, D. E. Shaw, R. A. Friesner, Proteins Struct. Funct. Bioinf. 55 (2004), 351-
367. c) M. P. Jacobson, R. A. Friesner, Z. Xiang, B. Honig, J. Mol. Biol., 320 (2002), 597-608
[54] a) Schrödinger Release 2018-4: Desmond Molecular Dynamics System, D. E. Shaw Research, New York, NY,
2018. Maestro-Desmond Interoperability Tools, Schrödinger, New York, NY, 2018 b) K. J. Bowers, E. Chow,
H. Xu, R. O. Dror, M. P. Eastwood, B. A. Gregersen, J. L. Klepeis, I. Kolossvary, M. A. Moraes, F. D. Sacerdoti,
20

J. K. Salmon, Y. Shan, D. E. Shaw, Proceedings of the ACM/IEEE Conference on Super-computing (SC06),
Tampa, FL, 2006.
[55] The PyMOL Molecular Graphics System, Open source version 1.8.2.0; Schrꢁdinger LLC
[56] T. Morita, H. Koike, Y. Koyama, H. Hagiwara, E. Ito, T. Fukuoka, T. Imura, M. Machida, D. Kitamoto, Genome
Announc. 1(2) (2013) e00064-13.
[57] The UniProt Consortium, Nucleic Acids Res. 47 (2019) D506–D515.
[58] GenBank, Nucleic Acids Res. 47 (2019) D94–D99.
[59] S. H. Kim, J. W. Kronstad, B. E. Ellis, Phytochemistry, 43(2) (1996) 351−357.
[60] P. V. Subba Rao, K. Moore, G. H. N. Towers, Can. J. Biochem. 45 (1967) 1863−1872.
[61] N. A. Dima, A. Filip, L. C. Bencze, M. Oláh, P. Sátorhelyi, B. G. Vértessy, L. Poppe, C. Paizs, Studia Univ.
Babes Bolyai, Ser. Chem. 61(2) (2016) 21−34.
[62] M. C. Moffitt, G. V. Louie, M. E. Bowman, J. Pence, J. P. Noel, B. S. Moore, Biochemistry 46 (2007) 1004–
1012.
[63] F. C. Cochrane, L. B. Davin, N. G. Lewis, Phytochemistry 65 (2004) 1557–1564.
21