A High-Throughput Screening Method for the Directed Evolution of Hydroxynitrile Lyase towards Cyanohydrin Synthesis

Article abstract of DOI:10.1002/cbic.202000658

Chiral cyanohydrins are useful intermediates in the pharmaceutical and agricultural industries. In nature, hydroxynitrile lyases (HNLs) are a kind of elegant tool for enantioselective hydrocyanation of carbonyl compounds. However, currently available methods for demonstrating hydrocyanation are still stalled at precise, but low-throughput, GC or HPLC analyses. Herein, we report a chromogenic high-throughput screening (HTS) method that is feasible for the cyanohydrin synthesis reaction. This method was highly anti-interference and sensitive, and could be used to directly profile the substrate scope of HNLs either in cell-free extract or fermentation clear broth. This HTS method was also validated by generating new variants of PcHNL5 that presented higher catalytic efficiency and stronger acidic tolerance in variant libraries.

Full text of DOI:10.1002/cbic.202000658

Combining Chemistry and Biology
Accepted Article
Title: A High-throughput Screening Method for the Directed Evolution
of Hydroxynitrile Lyase towards Cyanohydrin Synthesis
Authors: Yu-Cong Zheng, Liang-Yi Ding, Qiao Jia, Zuming Lin, Ran
Hong, Hui-Lei Yu, and Jian-He Xu
This manuscript has been accepted after peer review and appears as an
Accepted Article online prior to editing, proofing, and formal publication
of the final Version of Record (VoR). This work is currently citable by
using the Digital Object Identifier (DOI) given below. The VoR will be
published online in Early View as soon as possible and may be different
to this Accepted Article as a result of editing. Readers should obtain
the VoR from the journal website shown below when it is published
to ensure accuracy of information. The authors are responsible for the
content of this Accepted Article.
To be cited as: ChemBioChem 10.1002/cbic.202000658
Link to VoR: https://doi.org/10.1002/cbic.202000658
01/2020

10.1002/cbic.202000658
ChemBioChem
COMMUNICATION
A High-throughput Screening Method for the Directed Evolution
of Hydroxynitrile Lyase towards Cyanohydrin Synthesis
Yu-Cong Zheng,[a] Liang-Yi Ding,[a] Qiao Jia,[a] Zuming Lin,[b] Ran Hong,*[b] Hui-Lei Yu*[a] and Jian-He
Xu*[a]
Dedication ((optional))
[a]
[b]
Y.-C. Zheng, L.-Y. Ding, Prof. Dr. H.-L. Yu, Prof. Dr. J.-H. Xu
State Key Laboratory of Bioreactor Engineering, East China University of Science and Technology, Shanghai 200237 (China)
and State Key Laboratory of Bioreactor Engineering, Shanghai Collaborative Innovation Centre for Biomanufacturing and Frontiers Science Center for
Materiobiology and Dynamic Chemistry, East China University of Science and Technology, Shanghai 200237, China
E-mail: jianhexu@ecust.edu.cn; huileiyu@ecust.edu.cn
Z. Lin, Prof. Dr. G.-Q. Lin, Prof. Dr. R. Hong
CAS Key Laboratory of Synthetic Chemistry of Natural Substances,
Center for Excellence in Molecular Synthesis, Shanghai Institute of Organic Chemistry (CAS), Shanghai 200032 (China)
and University of the Chinese Academy of Sciences, Beijing 100049 (China)
E-mail: rhong@sioc.ac.cn
Supporting information for this article is given via a link at the end of the document.((Please delete this text if not appropriate))
are not strict to substrate. Methods for determination of the
released HCN were subsequently developed, which involving
using fluorescent probes (Scheme 1B),[7] Kꢀnig reaction assays
(Scheme 1C),[8] and colony-based Feigl–Anger reaction assays
(Scheme 1D)[9]. However, regarding fluorescent detection
methods, strong background interference and reaction
suppression are observed at acidic pH levels, which limited the
substrate scope only to the mandelonitrile at pH 6.5. Although
colony-based Feigl-Anger reaction allowed screening of massive
random mutagenesis library of E. coli harboured recombinant
GtHNL toward 2-Cl-mandelonitrile dehydrocyanation, all above
Abstract: Chiral cyanohydrins are useful intermediates in
pharmaceutical and agricultural industries. In nature, hydroxynitrile
lyases (HNLs) represent a kind of elegant tool for enantioselective
hydrocyanation of carbonyl compounds. Currently available methods
for demonstration of hydrocyanation reaction still stalled at precise but
low throughput GC or HPLC analyses. Herein we report
a
chromogenic high-throughput screening (HTS) method feasible for
cyanohydrin synthesis reaction. This method was highly anti-
interference and sensitive, and could be used to directly profile the
substrate scope of HNLs neither in cell-free extract or fermentation
clear broth. This HTS method was also validated by generating of new
variants of PcHNL5 which presented higher catalytic efficiency and
stronger acidic tolerance in variant libraries.
Hydroxynitrile lyases (HNLs; EC 4.1.2.10, 4.1.2.11, 4.1.2.46, and
4.1.2.47) catalyze the cleavage of cyanohydrins to form hydrogen
cyanide and carbonyl compounds (mostly aldehydes).[1] The
reverse hydrocyanation reaction has been successfully used for
the synthesis of chiral cyanohydrins in pharmaceutical and
agricultural industries.[2] As cyanohydrins easily degrade
spontaneously in aqueous systems at pH >5, asymmetric
biohydrocyanation is preferentially conducted under highly acidic
conditions (pH <4) to prevent spontaneous racemization of the
enzymatically formed product.[3] This presents a great challenge
to both the specific activity and pH stability of HNLs, with very few
enzymes tolerating such harsh conditions.[3a,4] Furthermore, to
meet the demands of synthetic applications, protein engineering
of HNLs is usually required to expand their substrate scope.
Directed evolution is a powerful tool for obtaining desired
biocatalysts for target transformations.[5] However, this strategy
relies on the mass screening of clones to obtain beneficial
mutants, which depends on the availability of a reliable high-
throughput screening (HTS) method. For HNLs, several HTS
methods have been established for HNL screening or substrate
profiling in the previous two decades. The most classic method
depended on UV absorbance measurement of the enzymatic
cleavage product (benzaldehyde) or its derivatives from
corresponding cyanohydrins (Scheme 1A),[6] the beauty of this
method relies on its very high sensitivity and easy to operate in
laboratorial conditions. Aiming to develop a general method which
Scheme 1. Examples of colorimetric assays reported based on cyanohydrin
cleavage. (A) UV absorbance of aromatic aldehydes. (B) Coupling the
fluorescent probe. (C) König reaction assays. (D) Feigl-Anger reaction assays.
1
This article is protected by copyright. All rights reserved.

10.1002/cbic.202000658
ChemBioChem
COMMUNICATION
HNLs is the reverse reaction, known as asymmetric
hydrocyanation. According to the first rule of directed evolution,
‘you get what you screen for’, positive hits with higher activity
toward the cleavage reaction do not always result in an improved
ability in the reverse synthetic reaction.[6c] Therefore, the direct
assay of hydrocyanation activity is still dependent on accurate but
low-throughput high-performance liquid chromatography (HPLC)
or gas chromatography (GC) analyses,[10] which are time
consuming, labor-intensive, and unsuitable for HTS during the
directed evolution of HNLs. Therefore, a general and rapid HNL
assay method targeting cyanohydrin synthesis is still needed for
the efficient evolution of HNLs. Herein, we report an HTS method
for monitoring cyanohydrin synthesis using a modified 2-
aminobenzamidoxime (ABAO) cycling assay (Scheme 2). This
method allows direct chromogenic quantification of the remaining
aromatic or aliphatic aldehyde after hydrocyanation.
Scheme 2. Quantification of residual aldehyde after cyanohydrin synthesis
using chromogenic reagent ABAO.
In a recent study, Rudoff and coworkers[11] reported a fast
cycling reaction of aldehydes using 2-amino-benzamidoxime
derivatives in an activity assay of carboxylate reductases (CARs).
As CARs only show activity under neutral conditions, a pH
adjustment (to pH <4.5) was necessary after the enzymatic
reaction to ensure iminium ion formation to facilitate cycling
reactions when using self-synthesized 2-amino-5-methoxyl-
benzamidoxime. We envisaged aldehyde quantification in the
hydrocyanation reaction system using commercially available 2-
amino-benzamidoxime (ABAO) as
a general chromogenic
reagent. Initially, the cycling reaction in the presence of ABAO (2
equiv.) was evaluated using aldehydes in an acidic monophase
aqueous citrate buffer system (0.8 M, pH 4.5), which both
suppressed background nonselective HCN addition to aldehydes
and facilitated the cycling reaction. Herein, benzaldehyde (α1)
was selected as a representative substrate because it is the
scaffold found in most aldehydes accepted by most HNLs.
However, the UV/Vis absorbance peak observed at 405 nm kept
increasing even 10 min after ABAO was added, indicating that
more ABAO might be needed. After further optimization 5 equiv.
of ABAO was found to be sufficient to complete the cycling
reaction within 2 min (Figure S2).
Figure 1. (A) Calibration curve of aldehyde α1 concentration in citrate buffer (1
M, pH 2.5, 100 μL) by monitoring UV/Vis absorption at 405 nm after adding a
stock solution of ABAO (50 mM) in citrate buffer (1 M, pH 2.5, 100 μL). Data
shown are mean values of three replicates with standard deviations. (B)
Biohydrocyanation progress curves of aldehyde α1 (10 mM) with PcHNL5
(supernatant of yeast culture broth), as monitored comparatively by ABAO and
GC assays, respectively. See Supporting Information for more experimental
details.
the HTS methods are based on the cleavage reaction of HNLs.
For the requirements of practical synthesis, desired capability of
Figure 2. Correlation study of ABAO cycling assays with GC analysis results in profiling the substrate spectrum of wild-type PcHNL5 (One unit of enzyme activity
was defined as the amount of enzyme that catalyzes the formation of 1 μmol cyanohydrin per minute under assay conditions). Specific activities toward various
aldehydes were determined in 0.5-mL aqueous reaction systems. See Supporting Information for experimental detail and data.
2
This article is protected by copyright. All rights reserved.

10.1002/cbic.202000658
ChemBioChem
COMMUNICATION
As spontaneous nonselective HCN addition is highly pH-
sensitive,[3a] cycling reactions at different pH values were next
investigated. Surprisingly, lowering the aqueous phase pH led to
a dramatic increase in the cycling reaction rate. At pH 2.5, the
cycling reaction was finished almost immediately after ABAO was
added to the system, which is ideal for hydrocyanation monitoring.
Furthermore, the absorption intensity at 405 nm increased
significantly with decreasing reaction pH (Figure S3). The assay
also showed compatibility with most ions and organic cosolvents
present in E. coli or yeast culture and the enzymatic reaction
systems (Figures S4–S6). The only exception was some
reductive sugars (such as glucose and lactose) that potentially
interfere with the target aldehyde to be detected. However, these
sugars could be replaced with other carbon sources, such as
methanol and glycerol, in the cell culture process. Therefore,
ABAO can be directly used to quench the reaction and start the
chromogenic assay. To investigate the accuracy of our HTS
assays, the measured level of coloration and resultant substrate
conversion were correlated to the GC analysis results for the
hydrocyanation of aldehyde 1 (10 mM) by PcHNL5 (in the form of
a supernatant of yeast culture broth)[12] in an aqueous system
(Figure 1). The assay demonstrated a high compatibility and
sensitivity, resulting in detectable UV/Vis absorption variance at a
conversion difference of only 1%.
With optimized assay conditions in hand, the versatility of the
ABAO assay methodology was investigated. The specific
activities of wild-type PcHNL5[12] toward a series of aldehydes (16
aromatic aldehydes (α1–α16), four heteroaromatic aldehydes
(α17–α20), and eight aliphatic aldehydes (α21–α24)) were
determined using the modified ABAO assay method. To confirm
the reliability of this assay methodology, the specific activities of
HNL for hydrocyanation were simultaneously determined by GC.
Good correlation was observed toward almost all aldehydes
tested (Figure 2), except substrate α12, owing to trace enzyme
activity (4.4 mU/mg). However, ABAO showed relatively low
reactivity toward ketones (α29–α32, Figures S7–S10) compared
with aldehydes, indicating that the current HTS method was not
suitable for ketone substrates.
and its variant L331A,[15] and industrially applied (R)-PaHNL5
variants N3I/I108M/A111G[6c] and V317A.[16] Four aldehydes, α2,
α4, α12, and α25, were used as representative substrates (Table
1). The UV/Vis absorbance of the control reaction without any
HNL preparation (E. coli cell lysate or yeast secretion
supernatant) remained unchanged after adding ABAO.
Hydrocyanation reactions catalyzed by the crude preparation of
HNLs with relatively high acidities showed good correlation with
the GC analytical results. Although both wild-type PcHNL5 and
PaHNL5V317A showed quite poor activities toward bulky aliphatic
aldehyde α26, a detectable difference was observed, with
PaHNL5V317A showing relatively high activity toward this substrate,
in accordance with the reported results. Similar results were also
obtained for the transformation of α12 by wild-type PcHNL5.
However, this is worth to note enzymatic consumption of yeast
endogenous proteins (e.g. aldehyde reductases or aldehyde
oxidases) in the secretion led to spontaneous transformation of
the aldehydes, but it didn’t disturb for indentication of the
beneficial variant. These results showed that the ABAO cycling
method allowed quantitative determination of the hydrocyanation
reaction at moderate conversion and qualitatitation of substrate
conversion when HNLs had only very low activity.
To demonstrate the practicability of this HTS methodology, a
round of PcHNL5 directed evolution toward aldehyde α7, whose
corresponding (R)-cyanohydrin is a key intermediate in the
synthesis of veterinary antibiotics, was attempted.[12] The
beneficial sites were predicted in a substrate binding model
according to a docking simulation. A total of 12 residues around
the substrate binding pocket (Figure S12) were subjected to
saturation mutagenesis, giving
a library of ≥1200 yeast
transformants. Based on the ABAO assay, positive hits with
improved conversion were selected from the saturation
mutagenesis library of site 343 (Figure S13). Rescreening of the
beneficial clones by chiral GC led to the identification of two
variants, L343I and L343F.
Variants L343I and L343F both showed significantly
enhanced acid tolerance and higher specific activity than the wild-
type (Figure 3A). In particular, L343F exhibited a half-life of 112
h at pH 2.5, showing stability much higher than that of the wild-
type (19.7 h), and far beyond that of industrialized PaHNL variants
(about 9 h at pH 2.6).[17] Single-site mutagenesis in the active
pocket leading to collaborative elevation of the acid tolerance and
hydrocyanation activity of the HNL is surprising. To gain deeper
insight into this phenomenon, the crystal structure of PcHNL5L343F
(PDB ID: 7CGS) was determined, indicating that five strong π–π
stacking interactions were formed between the benzene ring of
The freedom of applying structurally diverse aldehydes
prompted us to perform a substrate screening of several well-
known HNLs to validate the current HTS methodology. We tested
the enzymatic hydrocyanation reactions of reported (S)-
MeHNLH103L[13] and its variant H103L/W128A,[14] (R)-PcHNL5[12]
Table 1. Bioconversion of representative aldehydes by known HNLs or their
variants, analyzed by ABAO or GC assay.[a]
% Conv. by ABAO[b] vs.
Substrate
HNLs
GC (listed in brackets)[c]
α2
PcHNL5WT
27.7 ± 0.1 (32.2 ± 0.1)
73.8 ± 1.7 (75.4 ± 1.7)
33.0 ± 4.0 (25.6 ± 0.7)
94.7 ± 0.4 (76.9 ± 0.0)
8.5 ± 0.5 (0.39 ± 0.06)
42.0 ± 0.1 (43.0 ± 2.8)
4.2 ± 0.4 (0.2 ± 0.1)
7.9 ± 0.4 (1.1 ± 0.2)
PaHNL5N3I/I108M/A111G
MeHNLH103M
MeHNLH103M/W128A
PcHNL5WT
α4
α12
α26
PcHNL5L331A
PcHNL5WT
Figure 3. (A) Plot of residual activities of PcHNL5 mutants after agitation at 1000
rpm, pH 2.5, and 30 °C; (B) structural insight into strong π–π stacking
interactions around the Phe343 side chain (red stick) by superposition of the
crystal structures of wild-type PcHNL5 (PDB ID: 6JBY, yellow) and PcHNL5
L343F (PDB ID: 7CGS, cyan).
PaHNL5V317A
[a]Reactions performed on 0.5-mL scale by applying yeast fermentation
supernatant or E. coli cell free extract with HNL activity (see Supporting
Information for experimental details). [b]Conversions assayed using the ABAO
cycling method. [c]Conversions assayed by chiral GC.
3
This article is protected by copyright. All rights reserved.

10.1002/cbic.202000658
ChemBioChem
COMMUNICATION
residue Phe343 and the aromatic side chains of the surrounding
residues, including Phe336, Phe344, Phe356, His358, and
Trp458 (Figure 3B). As these residues are located at the flexible
loops in the active pocket and the outer shell of PcHNL5, these
newly introduced hydrophobic interactions might dramatically
enhance enzyme stability. Using aldehyde α7 at a concentration
of 1 M and a PcHNL5 L343F (purified enzyme) dose of 1.0 mg
per mmol substrate, the biohydrocyanation was complete within
12 h, affording 97% conversion and 99% ee. In contrast, the wild-
type enzyme gave only 80% conversion and 98% ee (Figure S14).
Approximately twice times of the wild-type enzyme compared with
the variant was needed to obtain the same result.
In summary, we have developed a rapid and reliable
colorimetric assay method that enables the high-throughput
quantification of hydrocyanation reactions. We hope that this
high-throughput assay methodology could provide opportunities
for rapid evaluation of the hydrocyanation ability of HNLs toward
tested aldehydes through directed evolution. Furthermore, as
ABAO showed low reactivity toward ketones, the search for
ketone-reactive chromogen compounds is now ongoing in our
laboratory.
[9]
B. Krammer, K. Rumbold, M. Tschemmernegg, P. Pöchlauer, H. Schwab,
J. Biotechnol. 2007, 129, 151–161.
[10] (a) Y. Asano, K. Tamura, N. Doi, T. Ueatrongchit, A. H. Kittikun, T.
Ohmiya, Biosci. Biotechnol. Biochem. 2005, 69, 2349–2357; (b) S.
Nanda, Y. Kato, Y. Asano, Tetrahedron: Asymmetry 2006, 17, 735–741;
J. Andexer, J. V. Langermann, A. Mell, M. Bocola, U. Kragl, T. Eggert, M.
Pohl, Angew. Chem. Int. Ed. 2007, 46, 8679–8681; (c) M. Dadashipoura,
Y. Ishidaa, K. Yamamotoa, Y. Asano, Proc. Natl. Acad. Sci. U.S.A. 2015,
112, 10605–10610. (d) E. Lanfranchi, T. Pavkov-Keller, E.-M. Koehler,
M. Diepold, K. Steiner, B. Darnhofer, J. Hartler, T. Van Den Bergh, H.-J.
Joosten, M. Gruber-Khadjawi, G. G. Thallinger, R. Birner-Gruenberger,
K. Gruber, M. Winkler, A. Glieder, Sci. Rep. 2017, 7, 46738. (e) K. Isobe,
A. Kitagawa, K. Kanamori, N. Kashiwagi, D. Matsui, T. Yamaguchi, K.
Fuhshuku, H. Semba, Y. Asano, Biosci. Biotechnol. Biochem. 2018, 82,
1760–1769; (f) E. Lanfranchi, B. Grill, Z. Raghoebar, S. Van Pelt, R. A.
Sheldon, K. Steiner, A. Glieder, M. Winkler, ChemBioChem 2018, 19,
312.
[11] A. K. Ressmann, D. Schwendenwein, S. Leonhartsberger, M. D.
Mihovilovic, U. T. Bornscheuer, M. Winkler, F. Rudroff, Adv. Synth. Catal.
2019, 361, 2538–2543.
[12] Y.-C. Zheng, J.-H. Xu, H. Wang, G.-Q. Lin, R. Hong, H.-L. Yu, Adv. Synth.
Catal. 2017, 359, 1185–1193.
[13] Y. Asano, M. Dadashipour, M. Yamazaki, N. Doi, H. Komeda, Protein
Eng. Des. Sel. 2011, 24, 607–616.
Acknowledgements ((optional))
[14] H. Bühler, F. Effenberger, S. Förster, J. Roos, H. Wajant, H.
ChemBioChem 2003, 4, 211–216.
This work was financially supported by the National Key Research
and Development Program of China (2019YFA09005000 and
2018YFC1706200), the National Natural Science Foundation of
China (21922804, 21871085, 21776085 and 21536004), the
Chinese Academy of Sciences (XDB20020000 and QYZDY-
SSWSLH026), and the Fundamental Research Funds for the
Central Universities (22221818014). We would like to thank Prof.
Florian Rudroff (TU Wien) for his helpful discussions and Mr Jun
Zhu (ECUST) for his experimental assistance. We are also
grateful for the access to beamline BL19U1 at Shanghai
Synchrotron Radiation Facility and thank the beamline staff for the
technical helps.
[15] Y. -C. Zheng, F. -L. Li, Z. Lin, G. -Q. Lin, R. Hong, H. -L. Yu, J. -H. Xu,
ACS Catal. 2020, 10, 5757−5763.
[16] B. Pscheidt, Z. B. Liu, R. Gaisberger, M. Avi, W. Skranc, K. Gruber, H.
Griengl, A. Glieder, Adv. Synth. Catal. 2008, 350, 1943–1948.
[17] A. Glieder, R. Weis, W. Skranc, P. Poechlauer, I. Dreveny, S. Majer, M.
Wubbolts, H. Schwab, K. Gruber, Angew. Chem. Int. Ed. 2003, 42, 4815–
4818.
Keywords: cyanohydrin • hydrocyanation • 2-amino-
benzamidoxime • high-throughput screening • directed evolution
[1]
[2]
M. Dadashipou, Y. Asano, ACS Catal. 2011, 1, 1121–1149.
P. Bracco, H. Busch, J. V. Langermann, U. Hanefeld, Org. Biomol. Chem.
2016, 14, 6375–6389.
[3]
(a) J. N. Andexer, J. V. Langermann, U. Kragl, M. Pohl, Trends
Biotechnol. 2009, 27, 599–607; (b) B. Pscheidt, M. Avi, R. Gaisberger,
F. S. Hartner, W. Skranc, A. Glieder, J. Mol. Catal. B: Enzym. 2008, 52-
53, 183–188.
[4]
[5]
J. V. Langermann, J. K. Guterl, M. Pohl, H. Wajant, U. Kragl, Bioprocess.
Biosyst. Eng. 2008, 31, 155–161.
(a) U. T. Bornscheuer, B. Hauer, K. E. Jaeger, U. Schwaneberg, Angew.
Chem. Int. Ed. 2019, 58, 36–40; (b) J. Wang, G. Li, M. T. Reetz, Chem.
Commun. 2017, 53, 3916–3928; (c) G. Qu, A. Li, C. G. Acevedo-Rocha,
Z. Sun, M. T. Reetz, Angew. Chem. Int. Ed. 2019, 58, 36–40; (d) C. Ren,
W. Xin, J. Mencius, Q. Shu, Bioresour. Bioprocess. 2019, 6, 53.
(a) Bove, E. E. Conn, J. Biol. Chem. 1961, 236, 207–210; (b) N.
Kawahara, Y. Asano, ChemBioChem, 2015, 16, 1891–1895; (c) Z. B. Liu,
B. Pscheidt, M. Avi, R. Gaisberger, F. S. Hartner, C. Schuster, W. Skranc,
K. Gruber, A. Glieder, ChemBioChem 2008, 9, 58–61.
[6]
[7]
[8]
H. Agarwalla, M. Gangopadhyay, D. K. Sharma, S. K. Basu, S. Jadhav,
A. Chowdhury, A. Das, J. Mater. Chem. B, 2015, 3, 9148–9156.
J. Andexer, J. K. Guterl, M. Pohl, T. Eggert, Chem. Commun. 2006, 4,
4201–4203.
4
This article is protected by copyright. All rights reserved.

10.1002/cbic.202000658
ChemBioChem
COMMUNICATION
Entry for the Table of Contents
Make biohydrocyanation measurable: Hydroxynitrile lyases (HNLs) catalyze the enantioselective cleavage/formation of the
cyanohydrins. To date, available methods for determination of hydrocyanation still stalls at HPLC or GC assays, which are not suitable
for screening of massive of mutants for the protein engineering of this class of biocatalysts. We therefore demonstrate a chromogenic
high-throughput screening method for cyanohydrin synthesis reaction, which was validated by either substrate profiling or directed
evolution of HNLs.
5
This article is protected by copyright. All rights reserved.