Biochemical Process Engineering, Division of Chemical Engineering, Department of Civil, Environmental and Natural Resources Engineering, Luleå University of

Technology, Luleå, Sweden

Taros Chemicals GmbH & Co. KG, Dortmund, Germany

NzyTech LDA, Estrada Do Paco Do Lumiar, Campus Do Lumiar, Ed. E - R/C, Lisbon, Portugal

ARTICLE INFO

Keywords:

Directed evolution

High-throughput screening

Fusarium oxysporum

Library

Feruloyl esterase

ABSTRACT

The need to develop competitive and eco-friendly processes in the cosmetic industry leads to the search for new

enzymes with improved properties for industrial bioconversions in this sector. In the present study, a complete

methodology to generate, express and screen diversity for the type C feruloyl esterase from

Fusarium oxysporium

FoFaeC was set up in a high-throughput fashion. A library of around 30,000 random mutants of FoFaeC was

generated by error prone PCR of fofaec cDNA and expressed in

Yarrowia lipolytica

. Screening for enzymatic

activity towards the substrates 5-bromo-4-chloroindol-3-yl and 4-nitrocatechol-1-yl ferulates allowed the se-

lection of 96 enzyme variants endowed with improved enzymatic activity that were then characterized for

thermo- and solvent- tolerance. The

ve best mutants in terms of higher activity, thermo- and solvent- tolerance

were selected for analysis of substrate speci

city. Variant L432I was shown to be able to hydrolyze all the tested

substrates, except methyl sinapate, with higher activity than wild type FoFaeC towards methyl

-coumarate,

methyl ferulate and methyl ca

eate. Moreover, the E455D variant was found to maintain completely its hy-

drolytic activity after two hour incubation at 55 °C, whereas the L284Q/V405I variant showed both higher

thermo- and solvent- tolerance than wild type FoFaeC. Small molecule docking simulations were applied to the

ve novel selected variants in order to examine the binding pattern of substrates used for enzyme character-

ization of wild type FoFaeC and the evolved variants.

Introduction

The interest in feruloyl esterases (ferulic acid esterases, FAEs), also

known as cinnamoyl esterases (E.C. 3.1.1.73), is growing due to the

large number of potential biotechnological applications in several in-

dustrial sectors based on their ability to hydrolyze the ester bond be-

tween hydroxycinnamoyl motifs and sugars present in plant cell walls

[1]. The enzymes release phenolic components, such as ferulic,

-cou-

maric, ca

eic and sinapic acids [

2], with widespread industrial poten-

tial due to their antioxidant and antimicrobial properties [

3, 4]. The

modi

cation of these compounds via esteri

cation with aliphatic mo-

lecules (such as alcohols) can be used as a tool to alter solubility in oil-

based formulas and emulsions, making them ideal candidates for ap-

plication in oil-based industries, maintaining the antioxidant activity of

the starting acid. The reaction catalyzed by FAEs in the absence/low

content of water represents a further route of application of these en-

zymes that is receiving increasing interest for its industrial potential

[5]. As a consequence, the need to develop competitive and eco-

friendlier bioconversions based on esteri

cation reactions catalyzed by

FAEs leads to a search for new enzymes with improved properties in

https://doi.org/10.1016/j.nbt.2019.01.008

Received 6 December 2017; Received in revised form 9 January 2019; Accepted 23 January 2019

Abbreviations:

4NTC-Fe, 4-nitrocatechol-1-yl ferulate; ep-PCR, error prone polymerase chain reaction; FAEs, ferulic acid esterases; MCA, methyl ca

eate; MD,

molecular dynamics; MFA, methyl ferulate; M

CA, methyl

-coumarate; MSA, methyl sinapate;

NP-Fe, 4-nitrophenyl ferulate; RMSD, root-mean-square deviation;

SMD, small molecule docking; X-Fe, 5-bromo-4-chloroindol-3-yl ferulate

⁎

Corresponding author.

E-mail address:

vfaraco@unina.it

(V. Faraco).

These authors made equal contributions.

New BIOTECHNOLOGY 51 (2019) 14–20

Available online 24 January 2019

industrial applications. Directed evolution, which mimics natural evo-

lution, has proved to be a strategy suitable for improving or altering

enzyme properties such as speci

cities, activity, stability and solubility

by methods of genetic diversity integration [

6]. However, it is note-

worthy that this approach, of a time-consuming and cost-intensive

nature, is based on procedures that use model substrates to detect target

activities in order to provide more detailed qualitative data on enzyme

properties. So far, the analysis of FAE activity has not been straight-

forward, mainly due to a lack of suitable compounds for practical high-

throughput assays [

7, 8].

In a previous study, it was reported that the fungus

Fusarium oxy-

sporum

showed multiple FAE enzymes enabling its ability to grow on

varied materials such as wheat straw and corn cobs [

9, 10]. In parti-

cular, heterologous recombinant expression of type C FAE, belonging to

the SF2 subfamily, was carried out in

Pichia pastoris

and the re-

combinant enzyme was puri

ed and characterized using di

erent

substrates including methyl esters of hydroxycinnamates [

11].

This study was aimed at developing evolved variants of FoFaeC with

higher activity than the wild type enzyme and improved resistance to

temperature and solvent exposure. The objectives were therefore to

generate a library of mutants by error-prone polymerase chain reaction

(ep-PCR) in

Yarrowia lipolytica

and apply this platform in conjunction

with a high-throughput method to select the best variants. In addition,

docking studies were employed to examine the a

ffi

nity of the di

erent

substrates with the wild type and the selected evolved variants of

FoFaeC.

Materials and methods

Chemicals

Yeast extract, bacto tryptone, bacto peptone and yeast nitrogen base

(without amino acids and without ammonium sulphate) were pur-

chased from Difco (Difco, Paris, France). QIAprep kit from Qiagen

(Hilden, Germany) was used for plasmid extraction and PCR fragment

puri

cations. Enzymes were purchased from Promega, Wisconsin, USA

and methyl cinnamate substrates were provided by Apin Chemicals ltd,

Oxford, UK. 5-Bromo-4-chloroindol-3-yl ferulate (X-Fe) was provided

by LISBP (Université de Toulouse, CNRS, INRA, INSA, Toulouse,

France). 4-nitrophenyl ferulate (

NP-Fe) and 4-nitrocatechol-1-yl fer-

ulate (4NTC-Fe) [

7] were provided by Taros Chemicals (Dortmund,

Germany). Other chemicals were purchased from Sigma

–

Aldrich (Sig-

–

Aldrich, St. Louis, MO).

Vectors, strains and culture media

The

Escherichia coli

strain Top 10 was used for transformations and

manipulations of recombinant plasmids and its growth was performed

at 37 °C in Luria

–

Bertani (LB) medium (10 g/L bacto tryptone, 10 g/L

NaCl, and 5 g/L yeast extract) supplemented with 100

g/ mL of am-

picillin or 40

g/mL of kanamycin to select transformed clones.

The JMP62-TEF-ppLIP2-LIP2 expression vector [

12] was used for

the cloning of the

faeC

cDNA from

F. oxysporum

and its mutants in

lipolytica

strain JMP1212. The Ura3 transformants obtained by yeast

transformation,

were selected on YNBG medium (1.7 g/L yeast nitrogen

base, sterilized by

ltration; 10 g/L glucose; 5 g/L NH

Cl; 50 mM

phosphates bu

er pH 6.8 phosphates bu

er Na/K (10%, v/v); 2 g/L

casamino acids, sterilized by

ltration) and grown in rich medium YPD

(10 g/L bactopeptone; 10 g/L yeast extract; 10 g/L glucose) and

YT2DH5 (10 g/L yeast extract; 20 g/L tryptone; 50 mM phosphates

er pH 6.8 Na/K (20%, v/v) supplemented with 50 g/L glucose). For

solid media, 20 g/L agar was added.

Recombinant vectors construction

Two types of cloning were performed using the cDNA sequence of

fofaec

cDNA, synthesized and optimized following

Y. lipolytica

codon

usage (NZYTech, Lisbon, Portugal), in which three di

erent restriction

sites, BamHI, BsrGI and AvrII, were inserted. The cDNA was cloned

either with its own optimized native signal sequence (indicated as

FoFaeC + SP) or by fusing the sequence of the mature protein with

preproLIP2 (indicated as FoFaeC

–

SP), a pro-peptide that has been

shown in some cases to increase the level of extracellular recombinant

protein [

13].

The plasmid JMP62-TEF-ppLIP2-LIP2 and

fofaec

cDNA were di-

gested by restriction enzymes BamHI/AvrII and BsrgI/AvrII (Promega,

Wisconsin, USA). In order to prevent circularization and re-ligation of

linearized DNA, dephosphorylation of the linearized vector ends was

performed using Calf intestinal alkaline phosphatase (CIAP) (Promega,

Wisconsin, USA).

The ligation of DNA fragments with cohesive ends was carried out

overnight at 4 °C in presence of T4 DNA ligase (Promega, Wisconsin,

USA) and, after plasmid ampli

cation in

E. coli

, linearization by NotI

(Promega, Wisconsin, USA) was performed according to manufacturer

’

instruction.

Error-prone PCR strategy

The expression cassette dedicated to the library construction of

mutants was obtained by overlapping PCR ampli

cation as described

by Bordes et al. [

14]. PCR reactions were carried out in order to amplify

DNA sequence of interest using MyCyclerTIM thermal cycler (Bio-Rad,

North America). Primers used in PCR reactions are listed in

Table 1

Y. lipolytica preparation and transformation

Competent cells preparation and transformation of

Y. lipolytica

wild

type was performed as previously described in [

13].

Wild type FoFaeC puri

fi

cation

The crude cell-free extract was concentrated by ultra

ltration

(Amicon chamber 8200, cut o

10 kDa membrane Millipore, Billerica,

MA) and puri

ed by immobilized metal ion a

ffi

nity chromatography

(IMAC). The concentrated sample was loaded onto a HisTrap 1 mL

column (GE Healthcare). The column was

rst washed with 20 mM

sodium phosphate bu

er (pH 7) containing 100 mM sodium chloride,

10 mM imidazole and then a linear salt gradient was applied, at a

rate of 1 mL/min, from 0 to 100% of elution bu

er (20 mM sodium

phosphate bu

er pH 7 containing 100 mM NaCl and 500 mM

Table 1

Primers used into the PCR strategy.

Primer name

PCR reaction

Sequence

PCR1_d

upstream region/faithfull

GATCCCCACCGGAATTGC

PCR1_RT

GCACCTGGGGAATGAAGCCGAGCACGAGCACG

PCR2_d

cDNA of interest + downstream Zeta/epPCR

CGTGCTCGTGCTCGGCTTCATTCCCCAGGTGC

PCR2_RT

GGAGTTCTTCGCCACCCC

PCR3_d

Fusion of PCR fragments PCR 1and 2

CCGCCTGTCGGGAACCGCGTTCAGGTGGAACAGGACCACC

PCR3_RT

CCGCACTGAAGGGCTTTGTGAGAGAGGTAACGCCG

G. Cerullo et al.

New BIOTECHNOLOGY 51 (2019) 14–20

Previous PDF in this issueNext PDF in this issue

Outline