Continuous Colorimetric Assay That Enables High-Throughput Screening of N -Acetylamino Acid Racemases

Article abstract of DOI:10.1021/ac5047328

N-Acetyl amino acid racemases (NAAARs) have demonstrated their potential in the enzymatic synthesis of chiral amino acids, molecules of significant biotechnology interest. In order to identify novel activities and to improve these enzymes by engineering approaches, suitable screening methods are necessary. Previous engineering of the NAAAR from Amycolatopsis Ts-1-60 was achieved by relying on an in vivo selection system that linked the viability of an E. coli l-methionine auxotroph to the activity of the improved enzyme. However, this assay was only suitable for the screening of N-acetyl-d-methionine, therefore limiting the potential to evolve this enzyme toward other natural or non-natural acetylated amino acids. Here, we report the optimization and application of a spectrophotometric microtiter-plate-based assay for NAAAR. The assay is based on the detection of the amino acid reaction product formed by hydrolysis of the N-acylated substrate by an l-amino acid acylase and its subsequent oxidation by an FAD-dependent l-amino acid oxidase (l-AAO). Cofactor recycling of the l-AAO leads to the formation of hydrogen peroxide which is easily monitored using horseradish peroxidase (HRP) and o-dianisidine. This method allowed for the determination of the kinetic parameters of NAAAR and led to the identification of N-acetyl-d-naphthylalanine as a novel NAAAR substrate. This robust method is also suitable for the high-throughput screening of NAAAR mutant gene libraries directly from cell lysates. (Chemical Equation Presented)

Full text of DOI:10.1021/ac5047328

Article
pubs.acs.org/ac
Continuous Colorimetric Assay That Enables High-Throughput
Screening of N‑Acetylamino Acid Racemases
Guiomar Sanchez-Carron,† Toni Fleming,‡ Karen E. Holt-Tiffin,‡ and Dominic J. Campopiano*,†
́
́
†The EastChem School of Chemistry, Joseph Black Building, The University of Edinburgh, Edinburgh, EH9 3FJ, U.K.
‡Dr. Reddy’s Laboratories Ltd., Chirotech Technology Centre, Milton Road, 410 Cambridge Science Park, Cambridge CB4 0PE, U.K.
S
* Supporting Information
ABSTRACT: N-Acetyl amino acid racemases (NAAARs) have
demonstrated their potential in the enzymatic synthesis of chiral
amino acids, molecules of significant biotechnology interest. In order
to identify novel activities and to improve these enzymes by
engineering approaches, suitable screening methods are necessary.
Previous engineering of the NAAAR from Amycolatopsis Ts-1-60 was
achieved by relying on an in vivo selection system that linked the
viability of an E. coli L-methionine auxotroph to the activity of the
improved enzyme. However, this assay was only suitable for the
screening of N-acetyl-D-methionine, therefore limiting the potential
to evolve this enzyme toward other natural or non-natural acetylated
amino acids. Here, we report the optimization and application of a
spectrophotometric microtiter-plate-based assay for NAAAR. The
assay is based on the detection of the amino acid reaction product formed by hydrolysis of the N-acylated substrate by an L-
amino acid acylase and its subsequent oxidation by an FAD-dependent L-amino acid oxidase (L-AAO). Cofactor recycling of the
L-AAO leads to the formation of hydrogen peroxide which is easily monitored using horseradish peroxidase (HRP) and o-
dianisidine. This method allowed for the determination of the kinetic parameters of NAAAR and led to the identification of N-
acetyl-D-naphthylalanine as a novel NAAAR substrate. This robust method is also suitable for the high-throughput screening of
NAAAR mutant gene libraries directly from cell lysates.
Q44244) with many desirable features;8,13,14 it requires only a
atural and unnatural α-amino acids have significant
Nindustrial interest as they have been commonly used as divalent metal cation for activity, is active at high temperatures
chiral building blocks for drugs and intermediates.1−7 Hence,
(60 °C), accepts a wide range of amino acid substrates, and
suffers no substrate or product inhibition up to 300 mM.
Unfortunately, the activity of the wild type NAAAR with the
desired hydrophobic N-acetylated substrates (e.g., phenyl-
alanine and naphthyalanine) was too low for industrial use.
A current goal is to improve and expand the application of
natural enzymes as biocatalysts using protein engineering tools
such as random mutagenesis (epPCR) or saturation muta-
genesis in a semirational, directed evolution approach.15−20
This last strategy is aided by detailed knowledge of the
substrate profile of the target enzyme, as well as an
understanding of the enzyme mechanism and identification of
the active site residues revealed by the crystal structure.14 These
combined techniques were applied previously to the NAAAR
enzyme and led to the isolation of a double mutant (NAAAR
G291D/F323Y) with up to 6-fold improvement of activity
toward several N-acetylated amino acids. When the NAAAR
G291D/F323Y mutant was combined with the L-acylase in a
DKR, it allowed the preparation of pure L-allyl-glycine at 99%
development of generally applicable, cost-effective and green
technology for production of single enantiomeric unnatural
amino acids is of significant industrial interest. Natural
proteinogenic L-amino acids are readily available and
inexpensive. In addition, racemic amino acids with a variety
of unnatural side-chains are generally readily prepared syntheti-
cally. Therefore, efficient methods that convert L-amino acids to
their D-enantiomeric forms or resolve racemic amino acid
mixtures into either the L- or D-amino acids provide a simple,
general, and economically viable route to the full range of
enantiopure products. For example, processes that are based on
hydantoinase/carbamoylase have been described and are used
industrially for the production of D-amino acids.1 A coupling
system involving an N-acetylamino acid racemase (NAAAR)
and an enantio-specific acylase (Scheme 1) allows the
preparation of pure L- or D-amino acids from readily
synthesized racemic N-acetylated amino acids. This matched
acylase/racemase combination provides an ideal “dynamic
kinetic resolution (DKR)” which allows the preparation of
products in high yield and enantiomeric excess.8−12
Received: December 19, 2014
Accepted: February 26, 2015
Published: February 26, 2015
Previous work selected Amycolatopsis N-acetyl amino acid
racemase (NAAAR) as a good biocatalyst (UniProt sequence
© 2015 American Chemical Society
3923
DOI: 10.1021/ac5047328
Anal. Chem. 2015, 87, 3923−3928

Analytical Chemistry
Article
in a similar manner to the monoamine oxidase from Aspergillus
niger (MAO-N).22,23 The peroxide can be readily monitored by
using horseradish peroxidase (HRP) to catalyze formation of a
highly colored product (e.g., at 436 nm using the substrate o-
dianisidine, Scheme 1).24,25 The results obtained from kinetic
measurements were verified by a chiral HPLC method. The
colorimetric assay can be used with a range of different
acetylated amino acids and allows the reliable determination of
the specific activity and kinetic constants of NAAARs; it clearly
differentiates between the wild-type NAAAR and the improved
G291D/F323Y mutant. Also, since the aminoacylases and L-
AAO accept a broad variety of natural and non-natural amino
acids, this method is particularly suitable for exploring the
substrate tolerance of NAAARs. Finally, the robustness and
broad utility of the method allows the quantitative screening of
mutant NAAAR libraries.
Scheme 1. Overview of the Coupled NAAAR/Acylase
Dynamic Kinetic Resolution (DKR) and Description of the
L-AAO/HRP-Dependent HTP Colorimetric Assay
EXPERIMENTAL PROCEDURES
■
Chemicals. N-Acetyl-D-naphthylalanine was prepared by Dr.
Reddy’s Laboratories Ltd. (Cambridge, UK) by acylation of the
free amino acid. N-Acetyl-D/L-methionine, D/L-methionine, N-
acetyl-D-phenylalanine, N-acetyl-D-phenylglycine, o-dianisidine,
magnesium chloride, and all other chemicals were obtained
from Sigma-Aldrich unless otherwise stated.
Enzymes. The NAAAR gene (wild type and the G291D/
F323Y double mutant), codon optimized for expression in E.
coli, was cloned in pET26b, and recombinant NAAAR was
prepared as described below. Recombinant L-acylase from
Thermococcus litoralis26 was prepared by Dr. Reddy’s
laboratories (specific activity: 39.7 U/mg). L-Amino acid
oxidase (L-AAO, Cat. No. A5147) from Crotalus atrox and
Horseradish peroxidase (HRP, Cat. No. P8125) were
purchased from Sigma.
NAAAR Expression and Purification. An overnight
culture of E. coli BL21(DE3) cells transformed with pET26b/
NAAAR wild type or pET26b/NAAAR G291D/F323Y was
used to inoculate 500 mL of TB broth (30 μg mL−1
kanamycin) and left to incubate at 37 °C with shaking until
an OD600 of ∼1.2 was reached. The cells were then induced
with 0.4 mM IPTG at 20 °C for 16 h. Cells were harvested by
centrifugation (4,000 rpm, 10 min) and resuspended
immediately in 10 mL of 50 mM sodium phosphate (pH
8.0), 100 mM NaCl, and 1 complete EDTA-free mini-protease
inhibitor tablet (Roche). This homogeneous solution was then
sonicated on ice (30s on/30s off) for 10 min. The supernatant
was then heated to 65 °C for 30 min to precipitate thermolabile
proteins, clarified by centrifugation (12,000 rpm, 60 min, at 4
°C), and filtered (0.45 μm). Purification was carried out by
anion exchange chromatography (HiPrep 16/10 FF Q column,
GE Healthcare) and size exclusion chromatography (HR S300
column, GE Healthcare).
For plate screening, distinct separated transformed colonies
were picked and transferred to individual wells in 48-well
microplates containing 500 μL autoinduction medium with 100
μg/mL kanamycin, followed by 48 h incubation at 30 °C. Two
×48-well plates were then combined in one 96-deep well plate,
and OD600 was normalized to an OD600 of 3 in all wells,
followed by centrifugation to harvest the cells. Cell lysis was
carried out by resuspending the cell pellets with a mixture of
100 mM sodium phosphate buffer pH 8.0, 25% BugBuster
(primary amine-free, Merck) extraction reagent, and 5 U/mL
benzonase (sigma), followed by incubation at room temper-
ature for 1 h with shaking (200 rpm). The clarified cell-free
e.e. from the racemic starting material. This NAAAR mutant
was isolated using an in vivo selection assay constructed so as to
ensure survival of an E. coli host strain only if it harbored the
recombinant NAAAR.21 For this method an L-methionine
auxotrophic E. coli strain (SET21) was generated by making
two specific gene knockouts (dadA and metB), and, as a result,
when grown in L-methionine deficient media, the SET21 host
was now dependent on plasmid-encoded NAAAR activity to
convert the N-acetyl-D-methionine to N-acetyl-L-methionine
which acts as a source of L-methionine.14 This selection assay
was useful for initial screening of NAAAR libraries but relied
heavily on judgment calls on the size/growth of E. coli colonies
and could only be tested using N-acetyl-D-methionine as a
substrate. These limitations prohibit the potential evolution of
NAAARs toward other non-natural and commercially valuable
substrates. Similarly, a chiral HPLC method can be also used to
quantitatively determine the activity of novel NAAAR mutants,
but this method, although simple and accurate, does not allow
the screening of the large number of variants typically produced
in a directed evolution screening campaign.
Therefore, the evaluation of the activity of a library of
enzymes variants toward a range of substrates requires a
suitable method for high-throughput screening in microtiter
plate format. This must be rapid, sensitive, reliable, and widely
applicable. Here, we report the development of a new
continuous colorimetric assay for the high-throughput screen-
ing of NAAAR. The assay is based on the production of free
amino acids from their acetylated counterparts by linking
NAAAR with an appropriate aminoacylase (Scheme 1). The
oxidation of the resultant amino acid by an amino acid oxidase
(L-AAO) can be linked to the production of hydrogen peroxide
3924
DOI: 10.1021/ac5047328
Anal. Chem. 2015, 87, 3923−3928

Analytical Chemistry
Article
extract was obtained after centrifugation (20 min, 4000 rpm).
Autoinduction media: 10 g/L N-Z-amine, 5 g/L yeast extract, 1
mM MgSO4, 1X NPS (20X NPS: 66 g/L (NH4)2SO4, 136 g/L
KH2PO4, 142 g/L Na2HPO4), 1X 5052 (50X 5052:250 g/L
glycerol, 25 g/L glucose, 100 g/L α-lactose), 100 μg/mL
kanamycin.
calculated using the following conditions: 79.4 U/mL L-acylase,
5 mM MgCl2, 70 mU/mL L-AAO, 10 U/mL HRP, 0.1 mg/mL
o-dianisidine (in MeOH), various concentrations of N-acetyl-
amino acids, 12.5 μg/mL NAAAR G291D/F323Y. By varying
the concentration of the substrate, the obtained kinetic data
were fitted using Graph Pad Prism software by nonlinear least-
squares regression using the Michaelis−Menten equation. All
kcat values were defined as per NAAAR monomer. The
concentration of protein was determined by measuring the
absorbance at 280 nm. The extinction coefficient of NAAAR
was calculated using Vector NTI software. Measurements were
obtained in triplicate.
Colorimetric Assay for Detection of NAAAR Activity.
Microtiter plate assay: NAAAR activity of 96-well plate cell
extracts was measured continuously using a Biotek Synergy
plate reader. Each well contained 100 μL cell lysate at an
appropriate dilution and 100 μL assay mixture. Typical assay
mixtures contained the following: 10−30 mM N-acetyl-D-
amino acid, 5 mM MgCl2, 79.4 U/mL L-acylase, 70 mU/mL L-
AAO, 10 U/mL HRP, 0.1 mg/mL o-dianisidine in 50 mM
sodium phosphate buffer pH 8.0. The increase in absorbance
was monitored at 436 nm for 1 h, at 42 °C. Rates were
determined from the slopes of the linear portion of each curve
using an extinction coefficient for o-dianisidine of 8700 M−1
cm−1 and a path length of 0.58 cm for 200 μL reactions.
Relative activity was calculated by subtracting the value
obtained in the negative controls (cells transformed with
empty plasmid) from each well.
RESULTS AND DISCUSSION
■
General Description of the Enzyme-Coupled Assay.
NAAAR and L-acylase together catalyze the conversion of N-
acetylated-D-amino acids to L-amino acids. Rather than make a
chromogenic derivative of the L-amino acid, our approach was
to detect the product using a suitable, broad substrate coupling
enzyme system (Scheme 1). We chose the flavin-dependent L-
AAO/HRP pair in which the L-AAO initially oxidizes the free
amino acid to its corresponding keto-acid.30,31 Oxidation of the
substrate by the flavin cofactor (FAD) generates FADH2 which
is then reoxidized by molecular oxygen, which itself is reduced
to hydrogen peroxide (H2O2) by HRP. The resulting peroxide
can be readily monitored by using o-dianisidine which reacts to
form a colored product that absorbs at 436 nm (ε436 = 8700
M−1 cm−1, Scheme 1).
Determination of the saturating amounts of coupling
enzymes (L-acylase, L-AAO, and HRP) was carried out by
adding increasing concentrations of each enzyme separately to
the reaction. The best NAAAR concentration was also
determined by lowering the NAAAR concentration until
linearity was achieved.
To quantify the reproducibility and uncertainty of the screen,
repeated measurements on single clones were carried out.
Measuring the variability of the screen is important to
determine a lower limit above which an observed improvement
in a targeted property (e.g., enzyme activity) can be considered
significant. The variability or dispersion of data was determined
using the coefficient of variance [CV = (standard deviation/
mean)*100%].27 The CV should be as small as possible
(<10%) to allow the detection of small improvements. The Z′-
factor was also calculated to compare and evaluate the quality
of the assay as described elsewhere.28
Enzyme Linearity. To confirm the linearity of the reaction
and make sure that the observed rates were not limited by the
coupled reaction components, initial velocities were measured
by varying the concentrations of both the NAAAR and the
coupling enzymes. It is worth noting that we used double
mutant NAAAR (G291D/F323Y) to optimize the assay since it
has higher activity than the wild type. When the reaction was
initiated by NAAAR addition, robust initial velocities were
measured, and only low background rates were observed in the
absence of enzyme or substrate (Supporting Information Figure
S-1). The initial velocities were always calculated from the
linear portions of the curves, after a lag phase of approximately
5 min. This lag phase is not unusual in enzyme-coupled
reactions and is likely due to the required accumulation of free
amino acids for the subsequent coupled reaction before a linear
response is achieved. Figure 1 shows a linear relationship
between the hydrogen peroxide production rate and NAAAR
concentration using two different substrates, N-acetyl-D-
methionine and N-acetyl-D-phenylalanine. We were pleased to
3SD positive control + 3SD negative control
Z′ = 1 −
|mean positive control − mean negative control|
HPLC based assay: This was similar to that previously
reported14,29 with the following modifications. For the HPLC
method, reactions containing 0.2 mg/mL NAAAR, 79.4 U/mL
L-acylase, 5 mM MgCl2, and 10−30 mM N-acetyl-D-amino
acids in 50 mM sodium phosphate buffer were incubated at 50
°C before being terminated at different times by diluting 40
fold in 0.025% triethylammonium acetate (TEAA):MeOH
(50:50). Reactions were then analyzed by chiral HPLC using a
Chirobiotic T column (Astec, chiral phase Teicoplanin, 5 μm,
250 mm × 4.6 mm) and the following isocratic conditions:
mobile phase: 0.025% TEAA:MeOH (50:50, v/v), flow rate:
0.5 mL min−1, detection λ: 210 nm, temperature: 25 °C, run
time: 10 min. The column was calibrated with single
enantiomers of the free amino acids and their N-acetyl
derivatives. The conversion of N-acetyl-D-methionine to L-
methionine was used to measure enzyme activity. Chromoleon
software (Dionex) was used to measure the area under
substrate and product peaks.
Figure 1. Activity at different concentrations of the NAAAR double
mutant (G291D/F323Y) toward N-acetyl-D-methionine (red circles)
and N-acetyl-D-phenylalanine (blue diamonds).
Kinetic Measurements. Kinetics parameters were deter-
mined using the microtiter plate assay at 50 °C and were
3925
DOI: 10.1021/ac5047328
Anal. Chem. 2015, 87, 3923−3928

Analytical Chemistry
Article
note the amount of NAAAR required to generate an observable
rate with N-acetyl-D-methionine, the substrate used to generate
the double mutant, was 10−20 μg/mL. Moreover, we also
observed a linear rate with the novel substrate N-acetyl-D-
phenylalanine suggesting that assay would have broad substrate
applicability.
These experiments are evidence that the rates are
independent of the concentrations of L-acylase, L-AAO, or
HRP. To further demonstrate that the components of the
coupled assay do not limit the observed rate, the concentration
of the coupled enzymes (L-acylase, L-AAO, and HRP) was
decreased, and the reactions were repeated (Supporting
Information Figure S-2). Decreasing L-acylase levels 5-fold, L-
AAO levels 4-fold, or HRP levels 10-fold did not significantly
affect the observed rates, thus indicating that the standard
reaction conditions were sufficient and that the coupled
enzymes are not limiting the observed initial rates.
Determination of NAAAR Kinetic Parameters. In order
to evaluate the quality of the data resulting from the
spectrophotometric assay, we determined the kinetic profile
of the NAAAR G291D/F323Y mutant and compared it with
those calculated by the established chiral HPLC method using
N-acetyl-D-methionine as a substrate. A KM value of 11.47
1.36 mM and a kcat value of 48.58
1.64 min−1 were
determined using the spectrophotometric assay, and the HPLC
method gave values of the same order (KM 23.41 2.12 mM,
kcat 153.4
5.4 min−1, Supporting Information Figure S-3).
The higher KM obtained with the HPLC assay is due to the
inherent error in measuring the formation of small amounts of
product at low substrates concentrations with the HPLC
analysis software. In fact, this highlights another limitation of
the HPLC method; the lowest substrate concentration at which
turnover can be confidently measured is only 5 mM N-acetyl
amino acid substrate. In contrast, the spectrophotometric assay
has a lower limit of detection ∼0.25 mM.
Figure 2. Control reaction plates for the NAAAR coupled assay. (A)
Screening with different NAAAR substrates. Each well uses a cell-free
extract from E. coli cells harboring the NAAAR (G291D/F323Y)
expressing plasmid (column 1) or empty pET26b plasmid (column 2).
(B) Comparison of the activities of E. coli cell-free extracts with 10
mM N-acetyl-phenylalanine as substrate from cells harboring (1)
empty pET26b plasmid, (2) wild-type NAAAR, (3) single mutant
NAAAR (G291D), and (4) double mutant NAAAR (G291D/F323Y).
We also note that the measured kcat values are lower
compared to those calculated previously.14 This is due to the
modification of certain parameters in the protocol. We have
replaced the previously used buffer TRIS-HCl with phosphate
buffer to avoid amine contamination. It has been shown that
NAAAR shows greater activity with Co2+ than Mg2+ as the
divalent cation,9 but here we used MgCl2 instead of CoCl2 due
to solubility issues when using phosphate buffer. In this study
the reactions were carried out at 50 °C, which is the maximum
temperature achievable with the plate reader, instead of 60 °C.
We used 49.6 U/mL L-acylase in the spectrophotometric assay,
in contrast to 198.5 U/mL used in previous HPLC assays, as
this was the most appropriate concentration to use based on
the coupling enzyme optimization experiments (see Supporting
Information Figure S-2). All of these combined factors resulted
in a lower specific activity of the enzyme compared to the
previous assays.
Applicability of the Coupled Assay to High-Through-
put Screening of NAAAR Substrates and NAAAR
Variants. With the optimized HTP method in hand, we first
used it to screen the substrate profile of NAAAR and calculated
the kinetic values for those that gave detectable activity (Figure
2A, Table 1). We found that N-acetyl-D-methionine was the
best substrate for NAAAR showing kcat values 3- and 12-fold
higher than those obtained for N-acetyl-D-phenylalanine and N-
acetyl-D-phenylglycine. However, NAAAR shows higher affinity
for N-acetyl-D-phenylalanine than N-acetyl-D-methionine (KM
values of 4.20 0.39 and 11.47 1.36 respectively). Moreover,
the HTP assay was used to identify a novel substrate of
NAAAR, the unnatural acetyl-amino acid N-acetyl-D-naphthy-
lalanine, which had not been described before (Figure 2A,
Table 1). NAAAR displays modest activity for this large
hydrophobic amino acid (KM 0.76 0.06 mM, kcat 3.55 0.06
min−1); however, it displays substrate inhibition above 20 mM
concentrations.
The optimized spectrophotometric assay was evaluated for
its suitability for HTP screening of NAAAR activity in whole
cell lysates/cell free extracts. HTP screening assays are
generally carried out in 96- or 384-well plate formats. Requisites
for a good screening assay are robustness, reproducibility,
accuracy, and adequate sensitivity in order to differentiate
improved activities among a large number of samples. For
proof-of-principle, several replicas of E. coli cells overexpressing
wild type NAAAR, the single mutant NAAAR G291D, the
double mutant NAAAR G291D/F323Y, and empty pET26b
plasmid as a negative control, were used to select the best
culture and lysis conditions (Figure 2B). Two different culture
media were assayed for cell growth, Terrific Broth (TB) and
autoinduction media. Autoinduction media resulted in higher
protein production and avoided the addition of IPTG to induce
NAAAR expression and was therefore selected for further assay
optimization. Cell lysis was tested using 1 mg/mL lysozyme or
BugBuster (primary amine-free) extraction reagent and
benzonase as resuspension buffer. Significantly better cell lysis
was achieved using BugBuster solution. Free extract were tested
3926
DOI: 10.1021/ac5047328
Anal. Chem. 2015, 87, 3923−3928

Analytical Chemistry
Article
Table 1. Kinetic Parameters of Amycolatopsis NAAAR (G291D/F323Y)
N-acetylated amino acid
D-methionine
D-phenylalanine
D-phenylglycine
D-naphthylalanine
KM (mM)
kcat (min−1
11.47 1.36
48.58 1.64
4.24
4.20 0.39
6.23 0.24
1.48
8.39 0.80
4.00 0.21
0.48
0.76 0.06
3.55 0.06
4.67
)
kcat/KM
to select the best enzyme concentration to monitor the activity
in the presence of different substrates. For this purpose, the
optimized spectrophotometric assay conditions were adapted to
96-well plate format, and the assay was performed as described
in Materials and Methods. The best dilution for N-acetyl-D-
methionine was 10-fold, whereas for N-acetyl-D-phenylalanine a
smaller, 2-fold dilution was selected.
For evaluation and validation of the HTP assay, we
determined the coefficient of variance (CV) and the “Z′-factor”
using the standard assay conditions as described in Materials
and Methods. The “screening window coefficient” Z′-factor
provides a useful tool for comparison and evaluation of the
quality of assays since it reflects both the assay signal dynamic
range and the data variation associated with the signal
measurement.28 Z′-values between 0.5 and 1 are considered a
good to excellent assay, whereas Z′-values between 0 and 0.5
relate to double or “yes/no” type assays. To calculate the Z′-
value, a set of control wells was prepared with cells expressing
the wild type NAAAR. These give the minimum activity values
and are defined as the “negative control”. Similarly, wells
containing the double mutant NAAAR (which we know has
increased activity) are defined as “positive controls”.28 Table 2
We note that Isobe and colleagues recently developed an
enzyme coupled system to assay L-amino acylases.32 In their
study they purified an L-AAO from Rhodococcus sp. AIU LAB-3
and used the Aspergillus L-amino acylase (Amano) to optimize
their method. They isolated >10.0 mg of L-AAO from 6 L of
bacterial culture in four chromatographic steps. They chose to
couple hydrogen peroxide formation to react with 4-amino-
antipyrine and detected the product at 555 nm. The L-AAO
displayed broad substrate specificity was most active at ∼50 °C
and stable from pH 6.0−8.0 with an optimum of pH 8.0. Our
study is the first to couple the NAAAR through to a
chromogenic product in a HTP plate-reader format. Isobe
and colleagues work suggests that it should be possible to either
combine or mix and match elements from both our and their
method to cover a broader range of NAAAR substrates and
conditions.
CONCLUSION
■
In this study, we developed a continuous spectrophotometric
assay for the screening of gene libraries to discover improved
NAAAR variants. Moreover, the assay is also applicable for the
identification of new NAAAR substrates. This assay couples
NAAAR and an L-aminoacylase to yield free amino acids from
their N-acetylated substrates. The production of the free amino
acids is easily coupled with L-AAO and HRP, producing a
colored product which can be spectrophotometrically moni-
tored at 436 nm. The development of this assay has allowed us
to overcome the limitations of the previous in vivo auxotrophic
selection and HPLC-based assays.3 We are now able to evolve
NAAAR (by both rational and random methods) toward a large
number of natural and non-natural amino acid substrates with
the proviso that they are substrates for both coupling enzymes
(T. litoralis L-acylase and C. atrox L-AAO). Limitations in the
coupling enzymes used here can be overcome by selecting
alternatives that best meet the requirement of the substrate of
interest (e.g., the L-AAO from Isobe et al., 2013). We appreciate
that the assay can only be used with D-enantiomers. To address
this we are currently developing an equivalent HTP coupled
assay that uses N-acylated L-amino acid substrates.
Table 2. Evaluation of the Spectrophotometric Screening
Method
N-acetyl-D-methionine
N-acetyl-D-phenylalanine
WT
G291D/F323Y
WT
G291D/F323Y
μmol/min/mL
σ
11.93
1.11
9.31
0.5
27.39
1.22
4.47
1.58
0.15
10.08
0.7
20.29
1.89
9.32
CV (%)
Z′-factor
shows that CV values below 10% are obtained for all sample
sets (for both substrates and wild type and double mutant
NAAAR), which suggests good reproducibility of the assay.
Furthermore, Z′-values ≥0.5 are obtained for the two substrates
tested confirming the suitability of the assay for further
applications such as saturation mutagenesis library screening to
generate further improved NAAARs.
The optimization of the assay with purified enzymes, as well
as cell-free extracts, indicates that the HTP assay is rapid,
robust, versatile, and accurate. The kinetic constants calculated
with the new HTP assay were comparable to those obtained by
a previous HPLC method. More importantly, this assay is
compatible with reagents commonly used to prepare cell-free
extracts and is thus amenable to HTP screening of a large
number of samples in a 96-well plate format. The assay allows
rapid determination of activity by direct visualization and also
possesses the advantage of continuously measuring product
formation over time, minimizing the number of false positive
samples. The applicability of the new HTP assay has proved to
be a very useful tool in the identification of further improved
novel NAAAR variants with activities toward new amino acid
substrates with high commercial value.
The data obtained (Figure 2B, Table 2) show substantial
differences between the activities observed for wild-type and
double mutant NAAAR and confirms that this assay could have
been used to detect the double mutant NAAAR (G291D/
F323Y) as a positive hit from our previous mutagenic NAAAR
libraries.14 In fact, the new HTP assay was sensitive enough to
pick out even the single mutant NAAAR variant (G291D)
which displays only a slight increase in activity compared to the
wild type enzyme (Figure 2B). The versatility of the method
relies on the possibility of either detecting obvious color
changes by eye, which saves a lot of time when large number of
variants need to be screened (Figure 2B), but, on the other
hand, it can be continuously monitored with a plate reader,
which avoids many of the artifacts associated with end-point
assays.
3927
DOI: 10.1021/ac5047328
Anal. Chem. 2015, 87, 3923−3928

Analytical Chemistry
Article
(16) Brustad, E. M.; Arnold, F. H. Curr. Opin. Chem. Biol. 2011, 15,
201−210.
ASSOCIATED CONTENT
■
S
* Supporting Information
(17) Tracewell, C. A.; Arnold, F. H. Curr. Opin. Chem. Biol. 2009, 13,
3−9.
Figure S-1. Comparison of NAAAR activity and background
activity Figure S-2. Saturating conditions of coupled enzymes.
Figure S-3. Comparison of KM for N-acetyl-D-methionine
calculated with the chiral HPLC method (A) and HTP
spectrophotometric assay method (B). This material is available
free of charge via the Internet at http://pubs.acs.org.
(18) Bornscheuer, U. T.; Huisman, G. W.; Kazlauskas, R. J.; Lutz, S.;
Moore, J. C.; Robins, K. Nature 2012, 485, 185−194.
(19) Bommarius, A. S.; Paye, M. F. Chem. Soc. Rev. 2013, 42, 6534−
6565.
(20) Reetz, M. T. J. Am. Chem. Soc. 2013, 135, 12480−12496.
(21) Acevedo-Rocha, C. G.; Agudo, R.; Reetz, M. T. J. Biotechnol.
2014, 191, 3−10.
(22) Turner, N. J. Chem. Rev. 2011, 111, 4073−4087.
(23) Carr, R.; Alexeeva, M.; Enright, A.; Eve, T. S. C.; Dawson, M. J.;
Turner, N. J. Angew. Chem., Int. Ed. Engl. 2003, 42, 4807−4810.
(24) Washko, M. E.; Rice, E. W. Clin. Chem. 1961, 7, 542−545.
(25) Claiborne, A.; Fridovich, I. Biochemistry 1979, 18, 2324−2329.
(26) Toogood, H. S.; Hollingsworth, E. J.; Brown, R. C.; Taylor, I.
N.; Taylor, S. J.; McCague, R.; Littlechild, J. A. Extremophiles 2002, 6,
111−22.
AUTHOR INFORMATION
■
Corresponding Author
*Phone: 44 0131 650 4712. E-mail: dominic.campopiano@ed.
ac.uk.
Author Contributions
The manuscript was written through contributions of all
authors. All authors have given approval to the final version of
the manuscript.
(27) Salazar, O.; Sun, L. Methods Mol. Biol. 2003, 230, 85−97.
(28) Zhang, J. H.; Chung, T. D.; Oldenburg, K. R. J. Biomol. Screening
1999, 4, 67−73.
Notes
The authors declare the following competing financial
interest(s): Due to the nature of the research two of the four
authors are employed by Dr. Reddy’s Laboratories Ltd. and
have commercial interests in the use of the enzymes in this
report.
(29) Yu, Y. P.; Wu, S. H. Chirality 2001, 13, 231−235.
(30) Gabler, M.; Hensel, M.; Fischer, L. Enzyme Microb. Technol.
2000, 27, 605−611.
(31) Pollegioni, L.; Motta, P.; Molla, G. Appl. Microbiol. Biotechnol.
2013, 97, 9323−9341.
(32) Isobe, K.; Satou, S.; Matsumoto, E.; Yoshida, S.; Yamada, M.;
Hibi, M.; Ogawa, J. J. Biosci. Bioeng. 2013, 115, 613−617.
ACKNOWLEDGMENTS
■
We thank the BBSRC and Dr. Reddy’s Ltd. for an Industrial
Partnership Award (IPA) (BB/K006487/1) to support G.S.C.
We thank Dr. Ian Taylor from Dr. Reddy’s Laboratories Ltd.
for his help during the development of the high-throughput
screening assay.
REFERENCES
■
(1) Servi, S.; Tessaro, D.; Pedrocchi-Fantoni, G. Coord. Chem. Rev.
2008, 252, 715−726.
(2) Chen, Y.; Goldberg, S. L.; Hanson, R. L.; Parker, W. L.; Gill, I.;
Tully, T. P.; Montana, M. A.; Goswami, A.; Patel, R. N. Org. Process
Res. Dev. 2010, 15, 241−248.
(3) Stueckler, C.; Winkler, C. K.; Hall, M.; Hauer, B.; Bonnekessel,
M.; Zangger, K.; Faber, K. Adv. Synth. Catal. 2011, 353, 1169−1173.
(4) Wohlgemuth, R. Curr. Opin. Microbiol. 2010, 13, 283−292.
(5) Leuchtenberger, W.; Huthmacher, K.; Drauz, K. Appl. Microbiol.
Biotechnol. 2005, 69, 1−8.
(6) Soriano-Maldonado, P.; Rodríguez-Alonso, M. J.; Hernan
Cervantes, C.; Rodríguez-García, I.; Clemente-Jimenez, J. M.;
Rodríguez-Vico, F.; Martínez-Rodríguez, S.; Las Heras-Vazquez, F. J.
Process Biochem. 2014, 49, 1281−1287.
́
dez-
́
́
(7) Wang, W. C.; Chiu, W. C.; Hsu, S. K.; Wu, C. L.; Chen, C. Y.;
Liu, J. S.; Hsu, W. H. J. Mol. Biol. 2004, 342, 155−169.
(8) Takahashi, T.; Hatano, K. Acylamino acid racemase, production
and use thereof. Takeda Chemical Industries, Ltd.: US 4981799 A,
1991.
(9) May, O.; Verseck, S.; Bommarius, A.; Drauz, K. Org. Process Res.
Dev. 2002, 6, 452−457.
(10) Huerta, F. F.; Minidis, A. B. E.; Backvall, J.-E. Chem. Soc. Rev.
2001, 30, 321−331.
(11) Hsu, S. K.; Lo, H. H.; Kao, C. H.; Lee, D. S.; Hsu, W. H.
Biotechnol. Prog. 2006, 22, 1578−1584.
(12) Pamies, O.; Backvall, J. E. Trends Biotechnol. 2004, 22, 130−5.
(13) Verseck, S.; Bommarius, A.; Kula, M. R. Appl. Microbiol.
Biotechnol. 2001, 55, 354−361.
(14) Baxter, S.; Royer, S.; Grogan, G.; Brown, F.; Holt-Tiffin, K. E.;
Taylor, I. N.; Fotheringham, I. G.; Campopiano, D. J. J. Am. Chem. Soc.
2012, 134, 19310−19313.
(15) Turner, N. J. Nat. Chem. Biol. 2009, 5, 567−573.
3928
DOI: 10.1021/ac5047328
Anal. Chem. 2015, 87, 3923−3928