Large α-aminonitrilase activity screening of nitrilase superfamily members: Access to conversion and enantiospecificity by LC-MS

Article abstract of DOI:10.1016/j.molcatb.2014.05.019

A high-throughput screening for the identification of nitrilases demonstrating activity towards alpha-aminonitriles is reported. A LC-MS assay giving access to both conversion and enantiospecificity was developed. 588 candidate enzymes were screened as cell lysates against six alpha-aminonitriles in 96-well microplates. The candidate enzymes were selected following two criteria, their sequence identity with a set of known nitrilases or their phylogenetic position among the nitrilase superfamily. Five enzymes were identified and found to hydrolyse alpha-aminonitrile into the corresponding alpha-aminoacid. The substrate range was found to be very narrow as only two different alpha-aminonitriles, 2-aminovaleronitrile and 2-amino-2- phenylacetonitrile, were found to be substrates. The biocatalytic capabilities of three enzymes were further investigated and the best result was obtained with an enzyme from Burkholderia xenovorans catalysing the enantiospecific hydrolysis of 2-aminovaleronitrile into (S)-norvaline with excellent conversion and enantiomeric excess.

Full text of DOI:10.1016/j.molcatb.2014.05.019

Accepted Manuscript
Title: Large ␣-aminonitrilase activity screening of nitrilase
superfamily members: Access to conversion and
enantiospecificity by LC-MS
Author: Franck Bordier Mark Stam Ekaterina Dariy Sabine
´
Tricot Aurelie Fossey Johanna Rohault Adrien Debard Aline
Mariage Virginie Pellouin Jean-Louis Petit Alain Perret David
Vallenet Marcel Salanoubat Jean Weissenbach Carine
´
Vergne-Vaxelaire Veronique de Berardinis Anne Zaparucha.
PII:
S1381-1177(14)00169-6
DOI:
Reference:
http://dx.doi.org/doi:10.1016/j.molcatb.2014.05.019
MOLCAB 2967
To appear in:
Journal of Molecular Catalysis B: Enzymatic
Received date:
Revised date:
Accepted date:
25-3-2014
23-5-2014
24-5-2014
Please cite this article as: F. Bordier, M. Stam, E. Dariy, S. Tricot, A. Fossey, J. Rohault,
A. Debard, A. Mariage, V. Pellouin, J.-L. Petit, A. Perret, D. Vallenet, M. Salanoubat,
J. Weissenbach, C. Vergne-Vaxelaire, V. de Berardinis, A. Zaparucha., Large rmalpha-
aminonitrilase activity screening of nitrilase superfamily members: access to conversion
and enantiospecificity by LC-MS., Journal of Molecular Catalysis B: Enzymatic (2014),
http://dx.doi.org/10.1016/j.molcatb.2014.05.019
This is a PDF file of an unedited manuscript that has been accepted for publication.
As a service to our customers we are providing this early version of the manuscript.
The manuscript will undergo copyediting, typesetting, and review of the resulting proof
before it is published in its final form. 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.

*Manuscript
1
2
Large -aminonitrilase activity screening of nitrilase superfamily members: access to
conversion and enantiospecificity by LC-MS.
3
4
5
6
7
8
Franck Bordier,a,b,c,d Mark Stam,a,b,c Ekaterina Dariy,a,b,c Sabine Tricot,a,b,c Aurélie Fossey,a,b,c
Johanna Rohault,a,b,c Adrien Debard,a,b,c Aline Mariage,a,b,c Virginie Pellouin,a,b,c Jean-Louis
Petit,a,b,c Alain Perret,a,b,c David Vallenet,a,b,c Marcel Salanoubat,a,b,c Jean Weissenbach,a,b,c
9
Carine Vergne-Vaxelaire,a,b,c,* Véronique de Berardinis,a,b,c and Anne Zaparucha.a,b,c,
*
10
a
11
12
13
14
15
CEA, DSV, IG, Genoscope, 2 rue Gaston Crémieux, 91057 Evry, France
Fax: (+33)-1-6087-2514; phone: + (33)-1-6087-4578; e-mail: azaparuc@genoscope.cns.fr; cvergne@genoscope.cns.fr
b
CNRS-UMR8030, 2 rue Gaston Crémieux, 91057 Evry, France
Université d’Evry Val d’Essonne, boulevard François Mitterrand, 91057 Evry, France
c
d Present address: Department of Biocatalysis and Isotope Chemistry, Almac Sciences, Craigavon, BT63 5QD, UK.
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
1. Introduction.
Natural and non-natural -aminoacids (-AA) are key chiral building blocks for the
synthesis of fine chemicals.[1] Common proteinogenic -AAcan be obtained directly from
natural sources while others are mainly obtained by asymmetric synthetic routes often
requiring non environmentally benign conditions.[2] In that context, biocatalysis appears to
be a favorable method for chiral synthesis purpose.[3] Hydrolytic enzymes such as acylases,
hydantoinases, carbamoylases have been largely used for resolution and dynamic kinetic
resolution processes on protected form of racemic -AA.[4] Starting from -aminonitriles (-
AN), -AAhave been obtained by the enzyme duo nitrile hydratase and amidase [5-8] or by
the sole action of a nitrilase. Example of asymmetric synthesis of -AAthrough the combined
action of three enzymes (nitrile hydratase, aminopeptidase and ACL racemase) has been
recently published.[9]
Most examples reported the synthesis of chiral -AAfrom -AN via whole cells, only
a few dealt with the production with isolated nitrilases. In their pioneer work, Choi et al
reported the synthesis of (S)-phenylglycine by free cell extracts of Asperigillus
furmigatus.[10] Few years later, various -ANwere found to be substrates of a nitrilase
isolated from Rhodococcus rhodochrous PA-34 leading to (S)- or (R)--aminoacids.[11]
However, even sounding promising, the biocatalytic synthesis of -AA from -AN with
nitrilases was rarely exploited.[12-15]
We recently reported the creation of a large collection of nitrilases selected by sequence
comparison.[16] In order to discover new -aminonitrilases that catalyse the formation of
chiral -AA, we decided to extend this selection to better cover the whole biodiversity of the
nitrilase superfamily. The enzymes were so also selected by shared protein domains (protein
signature, Pfam, InterPro). We screened this larger collection of 588 enzymes as cell lysates
against various -AN. The six -AN selected substrates(1-6) are precursors of natural and
non-natural -AAwith diversified side chains (alkyl, polar, aromatic) (Figure 1). The HTS
nitrilase assay is generally based on the ammonia detection liberated during the reaction.[17]
With -ANsubstrates, this assay is not feasible due to the loss of NH3 in the reverse
Strecker reaction (Scheme 1). Moreover, it doesn’t give access to the enantiospecificity of
Page 1 of 20

1
2
3
4
5
6
7
8
9
the reaction. Numerous methods have been developed to monitor the conversion rate and
the stereochemical outcome of biocatalytic reactions. Among them, chiral HPLC fitted with
UV detector is one of the most efficient and reliable methods, but it is mainly suitable for the
detection of conjugated derivatives and require expensive and specific chiral column. In this
study, we present a liquid chromatography-electrospray ionization tandem mass
spectrometry (ESI-MS/MS) method for reliable measurement of six compounds. The LC
approach uses achiral conventional C18 column with prior derivatization of the reaction
mixture with a chiral UV chromophore. MS/MS detection is achieved by scanning through
Multiple Reaction Monitoring (MRM)[18] events on a triple quadrupole instrument. MRM
involves selecting for ions of a specified parent molecular weight (m/z ratio), fragmenting the
parent ion at an optimal collision energy for producing particular daughter ions, and then
quantifying the production of the ions of interest. This simple method affords a quick and
reliable assay for nitrilase activity, giving access to both conversion rate and
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
enantiospecificity.
From our activity screening, we obtained five nitrilases exhibiting an -aminonitrilase
activity. On the basis of their different catalytic capabilities, three were purified and further
studied. From the six -AN, only 2-aminovaleronitrile and 2-amino-2-phenylacetonitrile
(norvaline and phenylglycine precursors, respectively) were found to be substrates. To
extend the substrate scope, three 2-amino-2-phenylacetonitrile derivatives 7-9 (Figure 2),
precursors of valuable synthons, were tested on discovered -aminonitrilases and two of
them were hydrolyzed into their corresponding -AA by two different enzymes.
2. Material and methods.
2.1 Chemicals
All reagents were purchased from commercial sources and used without purification.
Racemic 2-amino-2-phenylacetonitrile (5) was purchased from Sigma Aldrich. Racemic 2-
amino-3-methylbutanenitrile (1) was prepared according to the procedure previously
described by López-Serrano et al,[19] racemic aminophenylacetonitrile derivatives 7, 8 and
9 following the one reported by Wang et al.[20] NMR spectra were recorded on a Bruker 300
MHz spectrometer (Evry University, France). Chemical shifts (expressed in ppm) of 1H and
13C NMR spectra were referenced to the solvent peaks δ H 3.31 and δ C 49.2 for MeOD-d4,
δ H 2.50 and δ C 39.5 for DMSO-d6. Thin-layer chromatography was performed with
aluminium-backed sheets with silica gel 60 F254 (Merck). Column chromatography was
performed on a CombiFlash® Companion using RediSep® Rf or GraceResolv Silica
cartridges.
2.2 Synthesis
2.2.1 Synthesis of 2-aminopentanenitrile as hydrochloride salt (2) (AN-norvaline)
To a mixture of butyraldehyde (1.25 ml, 13.9 mmol, 1 equiv.) and zinc iodide (177 mg, 0.56
mmol, 0.04 equiv.) cooled to 0°C was added trimethylsilyl cyanide (2.18 ml, 17.2 mmol, 1.25
equiv.). The reaction mixture was stirred for 15 min at room temperature then a 7M solution
of ammonia in methanol (10.5 ml) was added. The resulting solution was stirred for 2h at
40°C in closed system and concentrated to dryness under reduced pressure. Diethylether
(30 ml) was added to the residue and the supernatant was dried over magnesium sulfate,
Page 2 of 20

1
2
3
4
5
filtered and concentrated to dryness under reduced pressure. A solution of hydrochloric acid
2M in diethylether (7.5 ml) was added to the residue and the resulting solution was stirred for
30 min at room temperature and then concentrated to dryness under reduced pressure to
afford the desired compound 2 as a slight yellow solid; yield: 1.31 g (70 %). 1H NMR (DMSO-
d6, 300 MHz) δ 0,90 (t, J = 7.2Hz, 3H, H5), 1.43 (m, 2H, H4), 1.86 (m, 2H, H3), 4.54 (t, J =
7.6 Hz, 1H, H2); 13C NMR (DMSO-d6, 75 MHz) δ 13.2 (C5), 18.1 (C4), 32.1 (C3), 39.5 (C2),
116.9 (C1) (see SI, Figure S1 and S2).
6
7
8
9
2.2.2 Synthesis of 2,6-diaminohexanenitrile as dihydrochloride salt (3) (AN-lysine)
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
See SI Scheme S1 for the synthetic scheme
Benzyl (5-amino-5-cyanopentyl)carbamate
To a solution of oxalyl chloride (814 µL, 9.62 mmol, 2.3 equiv.) in dried dichloromethane (25
ml) under an inert atmosphere was slowly added a solution of dimethylsulfoxyde (1.3 ml,
18.2 mmol, 4.3 equiv.) in dried dichloromethane (10 ml). The solution was stirred for 5 min at
-70°C then a suspension of benzyl (5-hydroxypentyl)carbamate)[21] (1 g, 4.22 mmol, 1
equiv.) in dried dichloromethane (25 ml) was transferred via canula. The mixture was stirred
for 25 min at -70°C then triethylamine (11.7 ml, 8.44 mmol, 20 equiv.) was slowly added. The
reaction mixture was stirred for 20 min at -70°C and 15 min at room temperature and then
quenched by addition to 600 ml of water. The aqueous phase was extracted with diethylether
(3 x 100 ml) and the combined organic layers were dried over magnesium sulfate, filtered
and concentrated to dryness under reduced pressure to obtain the aldehyde (1.1 g) used in
the next step without purification.
To a suspension of potassium cyanide (249 mg, 3.83 mmol, 1 equiv.) in water (520 µL) were
added ammonium chloride (203 mg, 3.83 mmol, 1 equiv.) and ammonium hydroxide 33%
(1.32 ml; 20.3 mmol, 5.3 equiv.). The reaction mixture was stirred for 15 min at 0°C then a
solution of the aldehyde (900 mg, 3.83 mmol, 1 equiv.) in dioxane (1.35 ml) was added
dropwise over 20 min. The mixture was stirred for 20 h and then quenched by addition of
water (10 ml) and diethylether (30 ml). The aqueous phase was acidified with HCl 1M (50
ml), washed with diethylether (3 x 20 ml), basified to pH = 12 with NaOH 1M (20 ml) and then
extracted with dichloromethane (4 x 10 ml). The combined organic layers were dried over
magnesium sulfate, filtered and concentrated to dryness under reduced pressure to obtain
the protected aminonitrile; yield: 905 mg (91 %). 1H NMR (MeOD-d4, 300 MHz) δ 1.52 (m,
4H, H4, H5), 1.71 (m, 2H, H3), 3.14 (t, 2H, H6), 3.71 (t, 2H, H2), 5.06 (s, 2H, H8), 7.34 (m,
5H, H10-H12); 13C NMR (MeOD-d4, 75 MHz) δ 23.9 (C4), 30.4 (C5), 35.9 (C3), 41.5 (C6),
44.2 (C2), 67.4 (C8), 123.1 (C1), 128.8-129.6 (C10-C12), 138.6 (C9), 159.0 (C7) (see SI,
Figure S3 and S4).
2,6-diaminohexanenitrile as dihydrochloride salt (3)
To a suspension of dried sodium iodide (1.2 g, 8.03 mmol, 7.5 equiv.) in dried acetonitrile (6
ml) under inert atmosphere was added trimethylsilyl chloride (680 µL, 5.35 mmol, 5 equiv.).
The mixture was stirred for 45 min at room temperature then the amine (279 mg, 1.07 mmol,
1 equiv.) was added dropwise. The reaction mixture was stirred for 4 h at room temperature
and then concentrated to dryness under reduced pressure. Water (10 ml) and
dichloromethane (10 ml) were added to the obtained crude oil. The two layers were
separated, the aqueous layer was washed with dichloromethane (7 x 10 ml), neutralized with
a saturated solution of sodium hydrogen carbonate and then lyophilized. Purification of the
Page 3 of 20

1
2
3
4
5
6
7
8
9
product by flash chromatography on silicagel (MeOH + 0.1 % NH4OH) afforded the 2,6-
diaminohexanenitrile (112 mg). To a solution of this compound in methanol (10 ml) at 0°C
was added dropwise a solution of hydrochloric acid 2M in diethylether (1.1 ml, 2.5 equiv.).
The mixture was stirred for 15 min at room temperature and then concentrated to dryness
under reduced pressure to obtain 3 as a very hygroscopic yellow solid; yield:157 mg, (75 %).
1H NMR (MeOD-d4, 300 MHz) δ 1.59-1.81 (m, 4H, H4-H5), 2.05 (m, 2H, H3), 2.99 (m, 2H,
H6), 4.57 (t, J = 7.4 Hz, 1H, H2); 13C NMR (MeOD-d4, 75 MHz) δ 23.4 (C4), 27.7 (C5), 31.7
(C3), 40.3 (C6), 42.6 (C2), 116.8 (C1) (see SI, Figure S5 and S6). HRMS (ESI): m/z calcd.
For C6H14N3 (M + H)+ 128.1182, found 128.1179.
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
2.2.3 Synthesis of 4-amino-4-cyanobutanoic acid (4) (AN-glutamate)
To a suspension of ethyl 4-amino-4-cyanobutanoate[22] (2.13 g, 7.29 mmol, 1 equiv.) in
tetrahydrofuran (30 ml) was added a solution of hydrochloric acid 1M (7.7 ml). The reaction
mixture was stirred for 30 min at room temperature then neutralized with a saturated solution
of sodium hydrogen carbonate and concentrated to dryness under reduced pressure. Water
(15 ml) and dichloromethane (15 ml) were added to the crude product. The two layers were
separated, the aqueous layer was washed with dichloromethane (3 x 15 ml), acidified with a
cooled solution of hydrochloric acid 5M until pH 5 and concentrated to a final volume of
approximately 5 ml. The resulting suspension was centrifuged (12000 rpm, 4°C, 10 min) and
the supernatant removed. Ethanol (10 ml) was added to the resulting solid, the suspension
centrifuged (12000 rpm, 4°C, 10 min) and the supernatant removed. This procedure was
repeated three times to obtain the desired aminonitrile 4 as a white powder; yield: 571 mg
(61 %). 1H NMR (DMSO-d6, 300 MHz) δ 1.82 (q, J = 7.4 Hz, 2H, H4), 2.35 (dt, J = 7.4 et 2,3
Hz , 2H, H3), 3.71 (t, J = 7.4 Hz, 1H, H2), 4.51 (s, 6H, NH2); 13C NMR (DMSO-d6, 75 MHz) δ
29.9 (C4), 30.1 (C3), 42.1 (C2), 122.8 (C1), 173.7 (C5) (see SI, Figure S7 and S8). HRMS
(ESI): m/z calcd. For C5H9N2O2 (M + H)+ 129.0659, found 129.0658.
2.2.3 Synthesis of 2-amino-3-(4-hydroxyphenyl)propanenitrile as hydrochloride salt (6)
See SI Scheme S2 for the synthetic scheme
2-amino-3-(4-(benzyloxy)phenyl)propanenitrile
To a solution of 2-((4-benzyloxy)phenyl)acetaldehyde (200 mg, 0.89 mmol, 1 equiv.)
(synthesized according to Frigerio et al.[23] from the corresponding alcohol[24]) in
dichloromethane (4.5 ml) were added successively zinc iodide (1.1 mg, 0.035 mmol, 0.04
equiv.) and trimethylsilyl cyanide (122 µL, 0.97 mmol, 1.1 equiv.). The reaction mixture was
stirred for 20 h at room temperature then concentrated to dryness under reduced pressure.
The residue was dissolved in a solution of ammonia in methanol (7M, 9 ml) and the resulting
solution was stirred for 4 h at room temperature. The residue obtained after concentration
under reduced pressure was dissolved in a solution of hydrochloric acid 1M (5 ml). The
aqueous layer was washed with diethyl ether (2 x 4 ml), basified to pH = 12 with a solution of
sodium hydroxide 6M and then extracted with dichloromethane (3 x 5 ml). The organic layers
were combined, washed with a solution of brine (5 ml), dried over magnesium sulfate, filtered
and concentrated to dryness under reduced pressure to afford the desired protected
aminonitrile as a yellow oil; yield: 214 mg (96 %).1H NMR (MeOD-d4, 300 MHz) δ 3.10-3.28
(m, 2H, H3), 4.69 (m, 1H, H2), 5.07 (s, 2H, H8), 7.01 (m, 2H, H5), 7.29-7.43 (m, H5 and H11-
Page 4 of 20

1
2
H13); 13C NMR (MeOD-d4, 75 MHz) δ 37.1 (C3), 44.3 (C2), 71.1 (C8), 116.6 (C1), 126.3 (C6)
128.7-131.9 (C4-C5, C11-C13), 138.6 (C9), 160.3 (7) (see SI, Figure S9 and S10).
3
4
2-amino-3-(4-hydroxyphenyl)propanenitrile as hydrochloride salt (6)
5
6
7
8
To a suspension of protected aminonitrile (1.06 g, 4.93 mmol, 1 equiv.) in dichloromethane
(90 ml) cooled to 0°C was added dropwise a 1M solution of boron tribromide in heptane (4.22
ml, 4.22 mmol, 0.85 equiv.). The reaction mixture was stirred for 2 h at room temperature
and then concentrated to dryness under reduced pressure. Water (100 ml) and AcOEt (100
ml) were added to the residue, the two layers were separated and the aqueous layer was
washed with AcOEt (4 x 50 ml), basified to pH = 11 with a solution of sodium hydroxide 2.5 M
(10 ml) and then extracted with AcOEt (6 x 20 ml). The organic layers were combined, dried
over magnesium sulfate, filtered and concentrated to dryness under reduced pressure. A
solution of hydrochloric acid 2M in diethylether (20 ml) was added slowly to a precooled
solution of the residue in AcOEt (2 ml). The mixture was stirred for 30 min at room
temperature and then concentrated to dryness under reduced pressure. The crude product
was triturated with AcOEt (5 x 10 ml) to afford the desired compound 6 as a brown powder;
yield: 400 mg (41 %). 1H NMR (MeOD-d4, 300 MHz) δ 3.15 (m, 2H, H3), 4.65 (dd, J = 8.3 et
6.0 Hz, 2H, H1), 6.81 (m, 2H, H5), 7.19 (m, 2H, H6); 13C NMR (MeOD-d4, 75 MHz) δ 37.3
(C3), 44.4 (C1), 116.8 (C2), 117.1 (C5), 124.7 (C4), 131.9 (C6), 158.9 (C7) (see SI, Figure
S11 and S12). HRMS (ESI): m/z calcd. For C9H11N2 (M + H)+ 163.0866, found 163.0800.
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
2.2 Protein selection
α-aminonitrilase activity is part of carbon-nitrogen hydrolysis activities, which are
performing by enzymes belonging to the nitrilase superfamily[25] containing more than
10,000 sequences. To select candidate proteins in prokaryotic organisms, the strategy was
based not only on the sequence identity with known nitrilases[16] but also on shared protein
signatures from several resources: PFAM based on protein domains,[26] Gene 3D based on
3D structures[27] and Prosite based on sequence motifs.[28] Thus, sequences from
UniprotKB,[29] having the PFAM identifier PF0079, the Gene 3D identifier 3.60.110.10 and
the Prosite identifier PS50263, and coming from prokaryotic organisms were selected. We
obtained 7,652 sequences which were divided into three main groups using different
clustering methods (Kmeans, single linkage clustering): glutamine-dependent NAD(+)
synthetase (1,131 sequences), apolipoprotein N-acyltransferase (1,279 sequences) and the
rest of the nitrilase superfamily (5,242 sequences). Proteins from each set were compared by
BLASTp[30] and from the alignment score, a distance was calculated. The sequences were
clusterized with R hclust method and one sequence per cluster was selected in agreement
with the availability of the corresponding DNA in the genomic Genoscope collection giving a
total of 599 sequences (see SI Figure S13).
40
41
42
43
44
45
46
47
2.3 Cloning, candidate enzyme production and enzyme purification
All steps from primers purchase to biochemical assays were performed in 96-well
microplates. The procedures used for gene cloning, protein production in E. coli BL21(DE3)
plysE strains (Invitrogen), cell lysate preparations and enzyme purification of NIT28, NIT158
and NIT191 were carried out as previously described.[16] 425 candidate enzymes were
obtained and added to the 163 of the previously published candidate nitrilase selection [16]
giving a set of 588 candidate enzymes to be screened.
Page 5 of 20

1
2
2.4 Enzyme assay and sample preparation for general activity screening
3
4
5
6
7
8
9
The activity screening was performed in 96-well microplates. The enzyme reactions
were performed in a final volume of 100 µl containing 1 mM DTT, 3 µl cell lysate (0.05 to 0.1
mg/ml of total proteins), 10 mM substrate in 10% methanol at pH 7.3 (50 mM potassium
phosphate buffer), at 30°C for 4h. The reactions were stopped with 3.5 µl of 25% TFA and 40
µl of each well of acidified reaction media were transferred to a 96-well daughter microplate
and subjected to derivatization as described in paragraph 2.6. To minimize the number of
microplates to be analyzed (42 microplates), two pools (pool I and pool II) of derivatized
reaction media were generated according to MS detection criteria (see paragraph 2.7.1). 2 µl
of each well of microplates containing derivatized reactions with substrates 2, 3 and 6 (pool I)
were combined in a 96-well daughter microplate and diluted with 194 µl of TFA 0.01% before
LC-MS analyses (volume of injection: 10 µl). Similarly, another 96-well daughter microplate
was generated for pool II (substrates 1, 4 and 5). Conversion rates and enantiomeric excess
were determined by LC-MS analysis comparing derivatized substrate and product
concentrations to standard curves of their racemic forms. The standards were prepared as
described above replacing enzyme cell lysate by E. coli BL-21 blank cell lysate (see SI
Figure S14).
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
2.5 Activity of the purified enzymes
2.5.1 Activity of NIT191 on 2-aminovaleronitrile (2)
10 mM of 2-aminovaleronitrile 2 were incubated at 30°C in 100 mM potassium phosphate
buffer pH 7.3 with 0.1 mg of purified NIT191 in a final volume of 1.0 mL. An aliquot (100 µl) of
the reaction mixture was quenched by addition of 1 µl of TFA and centrifugated (13000 rpm,
10 min) to pellet the precipitated protein. 8 µl of a 60 mM solution of L-alanine were added as
internal standard to 40 µl of the supernatant and 40 µl of the resulting solution were
subjected to (S)-NIFE derivatization as described in paragraph 2.6. The resulting solution
was subjected to HPLC-UV analysis (injection 10 µl; conditions 1, tR [(S)-alanine] = 6.72 min,
tR [(S)-norvaline] = 8.34 min, tR [(R)-norvaline] = 9.16 min). Conversion rates and
enantiomeric excess were determined by comparing enzyme derivatized product
concentration to standard curves of racemic products (standardized with L-alanine).
2.5.2 Activity of NIT28 on 2-amino-2-phenylacetonitrile (5)
10 mM of 2-amino-2-phenylacetonitrile (5) were incubated at 30°C in 100 mM potassium
phosphate buffer pH 7.3 with 0.008 mg of purified NIT28 in a final volume of 1.0 mL. An
aliquot (100 µl) of the reaction mixture was quenched after a specified period of time by
addition of 1 µl of TFA, centrifugated (13000 rpm, 10 min) to pellet the precipitated protein
and 5 µl of the supernatant were directly subjected to HPLC analysis (conditions 2, tR
(PheGly) = 3.78 min, tR (PheGlyCN) = 7.55 min). Conversion rates were determined by
comparing enzymatic product concentration to standard curves of synthesized racemic
product. For determination of the enantiomeric excess, 50 µl of the supernatant were mixed
with 20 µl of 1M K2CO3 solution and 100 µl of a solution of (S)-2-((5-fluoro-2,4-
dinitrophenyl)propanamide) (FDAA) (6.4 mM in acetone). The resulting yellow solution was
heated at 40°C for 1h30 min under stirring (1000 rpm) then quenched with HCl 1M (20 µl) ,
Page 6 of 20

1
2
diluted with 100 µl of MeCN and filtered off before HPLC analysis (injection 10 µl; conditions
3, tR [(S)-PheGly] = 9.30 min, tR [(R)-PheGly] = 12.51 min).
3
4
5
2.5.3 Screening of NIT28, NIT158 and NIT191 on 2-amino-2-phenylacetonitrile derivatives 7,
8, 9.
6
7
8
9
10 mM of aminonitrile were incubated at 30°C in 100 mM potassium phosphate buffer pH 7.3
with 0.005 mg of purified enzyme for 24h in a final volume of 50 µl. The reaction mixture was
quenched by addition of 1 µl of TFA, centrifugated (13000 rpm, 10 min) to remove the
protein and 40 µl of the supernatant mixed with 16 µl of 1M NaHCO3 solution and 80 µl of a
solution of FDAA (6.4 mM in acetone). The resulting yellow solution was heated at 40°C for
1h30 min under stirring (1000 rpm) then quenched with HCl 1M (16 µl), diluted with 170 µl of
MeCN and filtered off before HPLC analysis (conditions 5, tR (7b-(S)) = 27.04 and tR (7b-
(R)) 27.41 min; conditions 3, tR (9b-(S)) = 9.93 and tR (9b-(R)) = 12.52 min).
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
2.5.4 Activity of NIT28 on 2-amino-2-(4-methoxyphenyl)acetonitrile (9a)
Same protocol as described in paragraph 2.7.2 with 2-amino-2-(4-methoxyphenyl)acetonitrile
(9a) as substrate and 0.11 mg of enzyme in a final volume of 1.0 mL. Analytical conditions:
conditions 4 for determination of the conversion (tR (9) = 22.30 min) and conditions 3 for
determination of the enantiomeric excess (tR (9b-(S)) = 9.93 and tR (9b-(R)) = 12.52 min).
2.6 Derivatization procedure with N-(4-Nitrophenoxycarbonyl)-L-phenylalanine 2-
methoxyethyl ester ((S)-NIFE)
40 µl of the reaction mixture (acidified as described for deactivation of the enzyme)
were basified to pH 7.5 with 3.5 µl of a 2.5 M K2CO3 stock solution. 16 µl of a 20 mg/ml
solution of (S)-NIFE in acetonitrile were then added and the resulting yellow solutions were
shaken for 20 min at room temperature. The derivatized media were subjected directly to
HPLC-UV analysis or diluted (1/100) and pooled with 0.01 % TFA solution for LC-MS
analysis to raise a pH around 6.5-7 (see paragraph 2.7).
2.7 Analytical methods
2.7.1 LC-MS analyses
LC/MS/MS analyses were carried out on a Dionex UltiMate 3000 RS LC system
(Thermo Scientific Dionex Corporation, Sunnyvale, CA, USA) coupled to a Hybrid triple
quadrupole-linear ion trap mass spectrometer (QTRAP 5500 from ABSciex, Toronto,
Canada). MS/MS experiments were conducted in the triple quadripole mode.
Preliminary MS experiments allowed pooling the derivatized enzymatic media in two groups
before the LC-MS analysis, providing that pooled compounds do not have identical m/z ratio,
leading to a shorten analysis time. (S)-NIFE derivatized -AN 2, 3 and 6, synthetic
precursors of nor-Val, Lys and Tyr respectively, were grouped in pool I, and (S)-NIFE
derivatized -AN 1, 4 and 5, synthetic precursors of Val, Glu and PheGly respectively, in pool
II.
For HPLC separations before mass spectrometry analyses, an ACQUITY UPLC BEN C18
column (1.7 µm, 2.1 x 150 mm) from Waters (Milford, USA) at a flow rate of 400 µl / min was
Page 7 of 20

1
2
3
4
5
used. The column was maintained at 50°C with a column heater. Mobile phases were (A)
0.1% TFA in water and (B) acetonitrile with the following gradient mode: ratio A/B 100/0 to
50/50 in 16 min and 50/50 to 0/100 in 5 min. The separation conditions were developed with
standard mixtures of nitriles and corresponding aminoacids (stock solution at 10 mM in 50
mM potassium phosphate buffer pH 7.3 with 10% methanol diluted 100 times in water)
derivatized as described in paragraph 2.6.
6
7
8
9
Mass spectrometry analyses were conducted in the positive ESI ion mode with the following
parameters: ion source (IS) 5500 V, curtain gas (CUR) 20 a.u., temperature (TEM) 450°C,
gas 1 (GS1) 45 a.u., gas 2 (GS2) 60 a.u., CAD – medium.
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
MS/MS experiments were carried out using MRM (multiple reaction monitoring) scan type
allowing simultaneous monitoring of several compounds using multiple ion transitions and
their subsequent quantitation in complex mixtures. For each compound of interest presented
in the pools I and II the optimization of following MS parameters (declustering potential (DP),
collision energy (CE) and cell exit potential (CXP)) was performed in order to establish the
best intensity transitions. Two transitions for each compound were chosen to build the final
MRM methods for pool I and pool II. In fact, each method includes a set of transitions for
detection and quantitation of derivatized amino acids and corresponding amides and nitriles.
(for more details see SI Table S1). Figure 3 illustrates separation and MS detection of
derivatized -aminonitriles and corresponding -aminoacids in standard mixtures.
2.7.2 HPLC-UV analyses
For HPLC analyses, an Atlantis® T3 column (3 µm, 4.6 x 150 mm) from Waters (Milford,
USA) at a flow rate of 1 ml/min was used. Mobile phases were (A) 0.1% TFA in water and (B)
0.1% TFA in acetonitrile in the specified A/B ratio. Conditions 1: gradient mode (ratio A/B
70/30 to 50/50 in 20 min) and UV detection at = 210 nm. Conditions 2: ratio A/B 90/10 and
diode array detection. Conditions 3: gradient mode (ratio A/B 65/35 to 45/55 in 20 min) and
UV detection at =338 nm. Conditions 4: ratio A/B 99/1 and UV detection at =229 nm.
Conditions 5: gradient mode (ratio A/B 80/20 to 50/50 in 20 min then isocratic mode 50/50)
and UV detection at =338 nm.
3. Results and discussion
3.1 Choice of derivatization procedure
We set up a valuable high-throughput assay giving access to both conversion and
enantiospecificity of the nitrilase reaction for a broad substrate scope. In place of Marfey’s
reagent (FDAA), commonly used for aminoacid analysis, which requires elevated
temperature not compatible with -aminonitrile stability, we chose the (S)-NIFE as the most
suitable derivatization reagent. Already developed for aminoacid[31, 32] this carbamate
revealed to be appropriate for -aminonitrile derivatization as the reaction can occur at pH
near neutrality (pH = 7.5) and room temperature. However, an exception was noted for 2-
amino-2-phenylacetonitrile (5) and its corresponding derivatized -aminoacid which are
much more unstable.[12] Nevertheless, for this substrate, our procedure gave reliable
qualitative information concerning the monitoring activity.
Page 8 of 20

1
2
3
3.2 LCMS-screening method
We developed a procedure to perform a general screening of the candidate enzymes
4
5
6
7
8
(588) against the various -aminonitriles (6) in an “all against all” strategy, leading to a good
quantification of both conversion rate and enantiomeric excess. Considering the large
number of analyses to be performed, we set up a sensitive method with a compromise
between quality and duration. Each targeted reaction was monitored through three
parameters: retention time, abundance of two daughter ions produced by optimized
fragmentation of the parent ion for each substrate and corresponding product (see paragraph
2.7.1 and SI Table S1). To minimize time required for analysis, the derivatized reactions
were grouped into two pools. Two derivatized -AN and -AA (1 and 2 and their
corresponding -AA Val and Nor-Val) which had the same parent mass (m/z 348.0 and
367.1 respectively) were consequently separated in different pools. After optimization, we
were able to test 3528 reactions. As shown in Figure 3, baseline resolution was not perfectly
achieved in all cases. Nevertheless, the aim of the screening was first to detect nitrilase
activity and secondly to discriminate rapidly high enantiospecific nitrilases from very poor
ones. This goal was attained in these conditions. The stereochemical configuration of
produced -aminoacids was assigned by comparison with enantiopure derivatized standards
(S) and (R)-aminoacid LCMS chromatograms. The method provides high sensitivity as the
smallest conversion found (5%, vide infra) gave a signal significantly greater than the
background (see SI Table S2 for peak area data). Keeping in mind that the stereospecificity
is a critical parameter, this screening method allows to directly assessing the viability of a
nitrilase in a biocatalytic process. It offers a significant advantage over existing methods that
just give qualitative information of conversion, e.g. spectrophotometry,[33, 34] microorganism
cultures with nitriles as the sole source of nitrogen[11, 35, 36]. In these reported methods,
only positive hits are then studied by chiral HPLC to validate or discredit them, thus
consuming time and using expensive columns.[37]
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
3.3 High-throughput screening results
The candidate enzymes were screened under standard conditions (see paragraph
2.4). The results are summarized in Table 1. Out of the very large number of candidate
nitrilases assayed, only five exhibited an -aminonitrilase activity and only compounds 2 and
5 were hydrolyzed into their corresponding -AA (Figure 4). Peak areas and conversions
deduced are detailed in SI Table S2. These compounds were already known to be
substrates of few nitrilases, giving the corresponding acids with good stereopreference for
the (S)-norvaline (NitA from Rhodococcus rhodochrous PA-34)[11] or (R)-phenylglycine
(Nit5086 from Diversa,[12] nitrilase from Pseudomonas fluorescens DSM7155 and NIT-106
from Codexis[37]). Nevertheless the conversion of -AN 5 was obtained on whole cells[5, 10,
35, 36, 38], i.e. without knowing which enzyme catalyses the reaction, nitrilase alone or nitrile
hydratase / amidase couple. No candidate was found for -aminonitriles bearing basic (2,6-
diaminohexanenitrile (3)), acidic (4-amino-4-cyanobutanoic acid (4)), benzylic (2-amino-3-(4-
hydroxyphenyl)propanenitrile (6)) groups or ramified alkane chain (2-amino-3-
methylbutanenitrile (1)). It is worthy to note that for the six substrates, none amide formation
was observed. This undesired by-product is often a brake for the development of the
biocatalytic nitrile hydrolysis. Therefore, our new enzymes are suitable candidates for
hydrolysis of 2-aminovaleronitrile (2) and 2-amino-2-phenylacetonitrile (5). From our results,
Page 9 of 20

1
2
3
4
5
6
7
8
9
one nitrilase (NIT191, Table 1 entry 1) exhibited a very good activity towards 2-
aminovaleronitrile (2), together with a very high ee for the (S)-enantiomer, while this nitrilase
plus four others (NIT09, NIT28, NIT29, NIT158) were found to be active towards 2-amino-2-
phenylacetonitrile (5) (Table 1, entries 2-6). The (R) enantiomer of this -aminoacid is of
commercial interest as a raw material for cephalosporin synthesis.[39] Interestingly, four
nitrilases were active towards the (R) -aminonitrile enantiomer, the fifth one, NIT191, being
active towards the (S) enantiomer although with a low specificity (entry 6). It is worthy to note
that NIT191 was the only enzyme active towards both substrates. This enzyme, from
Burkholderia xenovorans LB400, was already described[16, 40] as a nitrilase but -AN were
not yet reported as substrates. The determination of sequence identity between our five
nitrilases and already published nitrilases[12, 41-44] catalysing the formation of (S)-norvaline
and (R)-phenylglycine showed that enzymes annotated as nitrilases (NIT28 (A5V4Y8),
NIT158 (A5VE02) and NIT191 (Q13VV3)) have a moderate (50%) to quite significant (75%)
sequence identity (Table 2). The two others hits (NIT09 (B1MWQ8) and NIT29 (Q9Z5U5)),
annotated as predicted amidohydrolase and hydrolase respectively, exhibit very low
sequence identities with known ones (<30%, Table 2).
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
NIT191, together with NIT09, NIT28 and NIT158, has already been found in our laboratory to
hydrolyse a broad range of substrates including -hydroxynitriles. This illustrates the
promiscuity found in the nitrilase family. These results suggest also that -aminonitrilases are
probably only very weakly present in nature as only a few were depicted, in these study and
others, despite the different implemented protein selection methodologies covering the whole
prokaryotic biodiversity.
3.4 Enantiospecificy profiles of selected hits with purified enzymes
NIT28, NIT158 and NIT191, exhibiting promising results according to conversion and
enantiospecificity on cell lysates, were purified for further studies. One representative for
both substrate 2 and 5 were chosen to check the reliability of our method as compared to
existing ones. Conversion rates and enantiomeric excess over time for NIT191 from
Burkholderia xenovorans on aliphatic aminonitrile 2 and NIT28 from Sphingomonas wittichii
on aromatic aminonitrile are presented below.
The conversion of 2-aminovaleronitrile (2) by NIT191 and the enantiomeric excess of the
obtained norvaline were monitored over 24 hours by HPLC-UV after (S)-NIFE derivatization
(Figure 5). NIT191 appeared to be very stereospecific for the (S) enantiomer (ee 96%) as
only 2% of derivatized (R)-norvaline can be detected after 24h and the final conversion
attained 52% (RHEA: 40780). During the LCMS- screening, the deductive conversion from
calibration curve was 53% and the corresponding enantiomeric excess 94% ee (entry 1,
Table 1). These screening results are in very good accordance with the ones obtained with
the purified NIT191. This high stereospecificity makes this enzyme a very good candidate for
the formation of (S)-norvaline from the corresponding -aminonitrile. To our knowledge, this
result is the best reported for this biocatalytic transformation. The only other described
example was the nitrilase from Rhodococcus rhodochrous, also (S) stereospecific but with a
lower stereopreference.[11]
For the hydrolysis of 2-amino-2-phenylacetonitrile (5), the conversion was monitored by
HPLC (conditions 2) without any derivatization process as phenylglycine presents a sufficient
UV response to enable its direct UV-quantitation. The determination of the enantiomeric
excess was deduced after chiral derivatization with the Marfey’s reagent. Evolution of
Page 10 of 20

1
2
3
4
5
6
7
8
9
conversion and enantiomeric excess over time are presented in Figure 6. NIT28 presented a
high stereopreference for the (R) enantiomer but not total as 8% of the (S)-phenylglycine (ee
84%) was detected by HPLC-UV for a conversion of 49% (RHEA: 40776). Thus, this enzyme
was found active towards this substrate as it was the case in the activity screening, but with a
relative good enantiospecificity. That point is in discrepancy with the one obtained during the
screening where NIT28 appeared very poorly stereospecific (entry 6, Table 1). In this
particular case, our screening monitoring assay gave inconsistent results concerning
enantiomeric excess, mainly due to medium instability during the acidic quench and weak re-
basification to pH 7.5 for (S)-NIFE derivatization. This observation was not made for the
others substrates, making our screening method still relevant for all the others tested -
aminonitriles. For this transformation, NIT28 appeared to be an alternative to the nitrilase
5086 from Diversa,[12] however with a slight lower enantiospecificity. Its good relative activity
at temperature around 50°C on other substrates[16] makes it a potential promising stable
biocatalyst. Thermal and pH stabilities / activities are under investigations in the laboratory.
With the aim of broadening the substrate scope, the three purified enzymes were assayed on
aminophenylacetonitrile derivatives 7a-9a (Figure 2), precursors of valuable synthons.
Diphenol 7a and cyano compound 8a are respectively direct precursors of 3,5-
DiHydroxyPhenylGlycine 7b (3,5-DHPG) and 4-MethylCarboxyPhenylGlycine (4-MCPG), two
non-natural very expensive -aminoacids with pharmacological interest in the treatment of
Alzeihmer disease.[45, 46] The -AA 9b obtained by hydrolysis of -AN 9a is an
intermediate in the synthesis of various active compounds. The results are summarized in
Table 3. Unexpectedly, despite its structural similarity with substrate 5, no activity was
detected on derivative 8a. NIT28 and NIT158 were found to be active towards both
derivatives 7a (entries 1-2) and 9a (entries 5-6) giving predominantly the corresponding (R)-
aminoacids; NIT191 gave a good conversion of the methoxy-derivative 9a but with (S)-
stereopreference (entry 7) as already depicted for 5. NIT28, exhibiting the best
enantiospecificity and activity towards -aminonitrile 9a was further studied (Figure 7). A
satisfying enantiospecificity for the (R)-stereoisomer was observed as only 4.5% of the other
enantiomer could be detected after a conversion of 26% over 20 h. However this enzyme
was found less active on this methoxy derivative 9a than on the former 2-amino-2-
phenylacetonitrile (5).
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
4. Conclusion.
The high-throughput screening based on LC-MS analysis of derivatized substrates
and compounds is a very efficient method to identify new enantiospecific activities leading to
-aminoacids. In our work, this method was applied to the identification of enzymes able to
convert -aminonitriles into their corresponding -aminoacids. Five hundred eighty eight
candidate enzymes representative of the prokaryotic biodiversity were selected following two
criteria, their sequence identity with a set of known nitrilases or their phylogenetic position
among the nitrilase superfamily. Despite the high number of candidate enzymes screened
against a representative substrate set (alkyl, polar, aromatic), only five enzymes exhibiting an
-aminonitrilase activity were depicted. Interestingly, no nitrile hydratase activity was
observed. Only a few -aminonitrilases were already known and our large screening allows
us to conclude that, probably, this activity is very rare among the nitrilase superfamily.
Moreover, the substrate range was found to be very narrow with four -aminonitriles
Page 11 of 20

1
2
3
4
5
6
hydrolyzed by the enzymes of this study, 2-aminovaleronitrile, 2-amino-2-phenylacetonitrile
and two derivatives thereof. Consequently the general access of non-natural -aminoacids
from -aminonitrile precursors by biocatalytic hydrolysis with nitrilases is questioned.
Nevertheless, the new -aminonitrilases displayed interesting activities suggesting the
potential use of these enzymes in biocatalytic synthesis of (S)-norvaline and chiral
phenylglycine derivatives.
7
8
9
10
11
12
Page 12 of 20

1
2
3
4
5
Fig. 1. -aminonitrile structures.
6
7
8
9
10
11
12
Scheme 1. NH3 formation in the nitrilase reaction with -aminonitrile.
Table 1. Activity screening results: conversion and enantiomeric excess
Entry NIT Uniprot KB
number
organism
substrate conversion
(%)a
ee
(%)a
94 (S)
1
2
3
4
5
6
191 Q13VV3
Burkholderia
xenovorans
Leuconostoc
citreum
Sphingomonas
wittichii
Zymomonas
mobilis
Sphingomonas
wittichii
2
5
5
5
5
5
53
09
28
29
B1MWQ8
A5V4Y8
Q9Z5U5
7
95 (R)
9 (R)
63
5
95 (R)
95 (R)
≤5 (S)
158 A5VE02
191 Q13VV3
11
64
Burkholderia
xenovorans
13
14
15
16
17
18
19
20
21
See paragraph 2.4 for experimental details.
No conversion detected for substrates 1, 3, 4 and 6.
a Determined by LC-MS after reaction medium derivatization by (S)-NIFE (see paragraph
2.6).
Page 13 of 20

1
2
3
4
5
Table 2. Activity screening results: -aminonitrilase hits and percent sequence identity with
published nitrilases catalysing the formation of (S)-norvaline and (R)-phenylglycine
(R)-phenylglycine producer
NIT09 NIT28 NIT29
(S)-norvaline
producer
NIT191
NIT158
(B1MWQ8) (A5V4Y8) (Q9Z5U5) (A5VE02) (Q13VV3)
K4U7B1 (R. rhodochrous)
50
Nit 5086 Diversa/ Q6RWK5 <30
(Uncultured organism)
61
59
59
<30
<30
<30
42
48
46
65
75
60
Q5EG61
<30
(P. fluorescens)
NIT106 Codexis/Mutant of
Q89GE3
<30
(B.diazoefficiens)
6
7
The determination of optimal alignment and percent sequence identity was performed using
the BLAST software publicly available through the National Center for Biotechnology
Information website. Indicated values are the percent sequence identity.
For phylogenetic tree of these sequences, see SI Figure S15
8
9
10
11
12
13
14
Table 3. Activity results on derivatives 7a-9a: conversion and enantiomeric excess
15
Entry
substrate
NIT
28
158
191
28,158,191
28
158
conversion (%)a
2
1
2
ee (%)a
96 (R)
74 (R)
>99(S)
1
2
3
4
5
6
7
7a
7a
7a
8a
9a
9a
9a
no product detection
51
93
64
84 (R)
42 (R)
34 (S)
191
16
17
18
a see section 2.5.3 for experimental details
19
20
21
22
Fig. 2. 2-amino-2-phenylacetonitrile derivatives
Page 14 of 20

1
2
3
4
5
6
7
Fig. 3. LC-MS analysis of standards (S)-NIFE derivatized α-aminonitriles and corresponding
α-amino acids. A: pool I. 1, (S)-lysine; 2, (R)-lysine; 3, AN-lysine; 4, (S)-tyrosine; 5, (R)-
tyrosine; 6, (S)-norvaline; 7, (R)-norvaline; 8, AN-tyrosine; 9, AN-norvaline; 10, (S) and (R)
doubly derivatized lysine; 11, AN-lysine. B: pool II. 1, (S)-glutamate; 2, (R)-glutamate; 3, AN-
glutamate; 4, (S)-valine; 5, (R)-valine; 6, (S)-phenylglycine; 7, (R)-phenylglycine; 8, AN-
valine; 9, AN-phenylglycine. For clarity, a single MRM transition is represented.
8
9
10
11
12
Page 15 of 20

1
2
3
4
5
Fig. 4. A. MRM analysis of derivatized reaction medium NIT191 on pool I: detection of (S)-
norvaline (94% ee) B. MRM analysis of derivatized reaction medium NIT28 on pool II:
detection of (R)-Phenylglycine (9% ee).
100
90
80
70
60
50
40
30
20
10
0
0
5
10
15
20
25
30
time (h)
6
7
8
9
Fig. 5. Evolution of the conversion of 10 mM of 2-aminovaleronitrile (2) by NIT191 and
enantiomeric excess  of (S)-norvaline formed after derivatization of the reaction medium
with (S)-NIFE (see paragraph 2.5.1 for experimental conditions).
10
100
90
80
70
60
50
40
30
20
10
0
0
2
4
6
8
10
time (h)
11
Page 16 of 20

1
2
3
4
Fig. 6. Evolution of the conversion of 10 mM of 2-amino-2-phenylacetonitrile (5) by NIT28
and enantiomeric excess  of (R)-phenylglycine formed (see paragraph 2.5.2 for
experimental conditions).
100
90
80
70
60
50
40
30
20
10
0
0
5
10
15
20
25
time (h)
5
6
7
8
Fig. 7. Evolution of the conversion  of 10 mM of 2-amino-2-(4-methoxyphenyl)acetonitrile
(9a) by NIT28 and enantiomeric excess  of (R)-2-amino-2-(4-methoxyphenyl)acetic acid
(9b) formed (see paragraph 2.5.4 for experimental conditions).
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
[1] M. Breuer, K. Ditrich, T. Habicher, B. Hauer, M. Kesseler, R. Sturmer, T. Zelinski, Angew. Chem. Int.
Ed. Engl. 43 (2004) 788-824.
[2] C. Nájera, J.M. Sansano, Chem. Rev. 107 (2007) 4584-4671.
[3] O. May, S. Verseck, A. Bommarius, K. Drauz, Org. Process Res. Dev. 6 (2002) 452-457.
[4] R. Olivieri, E. Fascetti, L. Angelini, L. Degen, Biotechnol. Bioeng. 23 (1981) 2173-2183.
[5] M.X. Wang, S.J. Lin, J. Org. Chem. 67 (2002) 6542-6545.
[6] M.-X. Wang, S.-J. Lin, Tetrahedron Lett. 42 (2001) 6925-6927.
[7] M.-X. Wang, S.-J. Lin, J. Liu, Q.-Y. Zheng, Adv. Synth. Catal. 346 (2004) 439-445.
[8] M.A. Wegman, U. Heinemann, F. van Rantwijk, A. Stolz, R.A. Sheldon, J. Mol. Catal. B: Enzym. 11
(2001) 249-253.
[9] K. Yasukawa, R. Hasemi, Y. Asano, Adv. Synth. Catal. 353 (2011) 2328-2332.
[10] S.Y. Choi, Y.M. Goo, Arch. Pharm. Res. 9 (1986) 45-47.
[11] T. Bhalla, A. Miura, A. Wakamoto, Y. Ohba, K. Furuhashi, , Appl. Microbiol. Biotechnol. 37 (1992)
184-190.
[12] J.A. Chaplin, M.D. Levin, B. Morgan, N. Farid, J. Li, Z. Zhu, J. McQuaid, L.W. Nicholson, C.A. Rand,
M.J. Burk, Tetrahedron: Asymmetry 15 (2004) 2793-2796.
[13] J. Qiu, E. Su, W. Wang, D. Wei, Tetrahedron Lett. 55 (2014) 1448-1451.
[14] J. Qiu, E.Z. Su, H.L. Wang, W.W. Cai, W. Wang, D.Z. Wei, Appl. Biochem. Biotechnol. 173 (2014)
365-377.
[15] G. DeSantis, E. Chi, J.A. Chaplin, A. Milan, J. Short, D. Weiner, M. Madden, M. Burk, D. Robertson,
in: US 7993901, 2011.
[16] C. Vergne-Vaxelaire, F. Bordier, A. Fossey, M. Besnard-Gonnet, A. Debard, A. Mariage, V.
Pellouin, A. Perret, J.-L. Petit, M. Stam, M. Salanoubat, J. Weissenbach, V. De Berardinis, A.
Zaparucha, Adv. Synth. Catal. 355 (2013) 1763-1779.
[17] L. Martínková, V. Vejvoda, V. Křen, J. Biotechnol. 133 (2008) 318-326.
[18] N.R. Kitteringham, R.E. Jenkins, C.S. Lane, V.L. Elliott, B.K. Park, J. Chromatogr. B 877 (2009)
1229-1239.
Page 17 of 20

1
2
3
[19] P. López-Serrano, J.A. Jongejan, F. van Rantwijk, R.A. Sheldon, Tetrahedron: Asymmetry 12
(2001) 219-228.
[20] H. Wang, Y. Byun, C. Barinka, M. Pullambhatla, H.E. Bhang, J.J. Fox, J. Lubkowski, R.C. Mease,
M.G. Pomper, Bioorg. Med. Chem. Lett. 20 (2010) 392-397.
4
5
6
7
8
[21] M.N. Chatterjee, E.R. Kay, D.A. Leigh, J. Am. Chem. Soc. 128 (2006) 4058-4073.
[22] M. McLaughlin, R.M. Mohareb, H. Rapoport, J. Org. Chem. 68 (2003) 50-54.
[23] M. Frigerio, M. Santagostino, Tetrahedron Lett. 35 (1994) 8019-8022.
[24] G.-H. Chu, S. Milano, L. Kluth, M. Rhodes, R. Boni, D.A. Johnson, P.-K. Li, Steroids 62 (1997) 530-
535.
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
[25] H.C. Pace, C. Brenner, Genome Biol. 2 (2001) REVIEWS0001.
[26] M. Punta, P.C. Coggill, R.Y. Eberhardt, J. Mistry, J. Tate, C. Boursnell, N. Pang, K. Forslund, G.
Ceric, J. Clements, A. Heger, L. Holm, E.L. Sonnhammer, S.R. Eddy, A. Bateman, R.D. Finn, Nucleic
Acids Res. 40 (2012) D290-301.
[27] J. Lees, C. Yeats, J. Perkins, I. Sillitoe, R. Rentzsch, B.H. Dessailly, C. Orengo, Nucleic Acids Res. 40
(2012) D465-471.
[28] C.J. Sigrist, E. de Castro, L. Cerutti, B.A. Cuche, N. Hulo, A. Bridge, L. Bougueleret, I. Xenarios,
Nucleic Acids Res. 41 (2013) D344-347.
[29] The UniProt Consortium, Nucleic Acids Res. 41 (2013) D43-47.
[30] S.F. Altschul, W. Gish, W. Miller, E.W. Myers, D.J. Lipman, J. Mol. Biol. 215 (1990) 403-410.
[31] A. Péter, E. Vékes, G. Török, Chromatographia 52 (2000) 821-826.
[32] A. Péter, E. Vékes, G. Tóth, D. Tourwé, F. Borremans, J. Chromatogr. A 948 (2002) 283-294.
[33] S. Moore, W.H. Stein, J. Biol. Chem. 211 (1954) 907-913.
[34] A.L. Duchateau, M.G. Hillemans-Crombach, A. van Duijnhoven, R. Reiss, T. Sonke, Anal. Biochem.
330 (2004) 362-364.
[35] R. Bauer, B. Hirrlinger, N. Layh, A. Stolz, H.J. Knackmuss, Appl. Microbiol. Biotechnol. 42 (1994) 1-
7.
[36] S. Rustler, A. Müller, V. Windeisen, A. Chmura, B.C.M. Fernandes, C. Kiziak, A. Stolz, Enzyme
Microb. Technol. 40 (2007) 598-606.
[37] D. Brady, A. Beeton, J. Zeevaart, C. Kgaje, F. Rantwijk, R.A. Sheldon, Appl. Microbiol. Biotechnol.
64 (2004) 76-85.
[38] F.O.M. Alonso, O.A.C. Antunes, E.G. Oestreicher, J.Braz. Chem. Soc. 18 (2007) 566-571.
[39] M. Madden, P. Weiner David, A.N.N. Chaplin Jennifer, R. Dr Dan, in: WO200148175, 2001.
[40] J.L. Seffernick, S.K. Samanta, T.M. Louie, L.P. Wackett, M. Subramanian, J. Biotechnol. 143 (2009)
17-26.
[41] O. Kaplan, A.B. Vesela, A. Petrickova, F. Pasquarelli, M. Picmanova, A. Rinagelova, T.C. Bhalla, M.
Patek, L. Martinkova, Mol. Biotechnol. 54 (2013) 996-1003.
[42] D.E. Robertson, J.A. Chaplin, G. DeSantis, M. Podar, M. Madden, E. Chi, T. Richardson, A. Milan,
M. Miller, D.P. Weiner, K. Wong, J. McQuaid, B. Farwell, L.A. Preston, X. Tan, M.A. Snead, M. Keller, E.
Mathur, P.L. Kretz, M.J. Burk, J.M. Short, Appl. Environ. Microbiol. 70 (2004) 2429-2436.
[43] C. Kiziak, D. Conradt, A. Stolz, R. Mattes, J. Klein, Microbiology 151 (2005) 3639-3648.
[44] A. Krebber, E. Mundoff, M. Mock, S. Kambourakis, in: US 0142063A1, 2012.
[45] Y. Shinoda, T. Tanaka, K. Tominaga-Yoshino, A. Ogura, PLoS One 5 (2010) e10390.
[46] K. Wisniewski, H. Car, CNS Drug Rev. 8 (2002) 101-116.
Page 18 of 20

*Graphical Abstract (for review)
6.0e6
5.0e6
4.0e6
3.0e6
2.0e6
1.0e6
0.0
chiral derivatization
NH
CO2H
NH
R
(S)
6.0e6
5.0e6
4.0e6
3.0e6
2.0e6
1.0e6
0.0
R
CO2H
(R)
11.0
12.0
13.0
14.0
Time, min
15.0
16.0
enantiospecific α-aminonitrilases
NH2
CO2H
NH2
CN
10.0
11.0
12.0
13.0
14.0
15.0
16.0
17.0
NH2
Time, min
R
R
LC-MS/MS analysis
R
CN
Page 19 of 20

*Highlights (for review)
Highlights
We performed a large nitrilase activity screening on -aminonitriles with diversified side
chains (alkyl, polar, aromatic)
The large collection of tested enzymes covers the whole biodiversity of the nitrilase
superfamily
An efficient liquid chromatography-electrospray ionization tandem mass spectrometry (ESI-
MS/MS) method was carried out after derivatization of reaction mixtures to access both
conversion rate and enantiospecificity of the nitrilase reaction
Five new -aminonitrilases were found, with no nitrile hydratase activity observed.
(S)-norvaline, (R)-phenylglycine and two phenylglycine derivatives were obtained from
corresponding -aminonitriles
Page 20 of 20