A unique xylose reductase from Thermomyces lanuginosus: Effect of lignocellulosic substrates and inhibitors and applicability in lignocellulosic bioconversion

Article abstract of DOI:10.1016/j.biortech.2019.02.102

In this study, the xylose reductase gene (XRTL) from Thermomyces lanuginosus SSBP was expressed in Pichia pastoris GS115 and Saccharomyces cerevisiae Y294. The purified 39.2 kDa monomeric enzyme was optimally active at pH 6.5 and 50 °C and showed activity over a wide range of temperatures (30–70 °C) and pH (4.0–9.0), with a half-life of 1386 min at 50 °C. The enzyme preferred NADPH as cofactor and showed broad substrate specificity. The enzyme was inhibited by Cu²⁺, Fe²⁺ and Zn²⁺, while ferulic acid was found to be the most potent lignocellulosic inhibitor. Recombinant S. cerevisiae with the XRTL gene showed 34% higher xylitol production than the control strain. XRTL can therefore be used in a cell-free xylitol production process or as part of a pathway for utilization of xylose from lignocellulosic waste.

Full text of DOI:10.1016/j.biortech.2019.02.102

BioresourceTechnology281(2019)374–381
Contents lists available at ScienceDirect
Bioresource Technology
journal homepage: www.elsevier.com/locate/biortech
A unique xylose reductase from Thermomyces lanuginosus: Effect of
lignocellulosic substrates and inhibitors and applicability in lignocellulosic
bioconversion
Meng Zhanga, Adarsh Kumar Puria,⁎, Zhengxiang Wangb, Suren Singha, Kugen Permaula
a Department of Biotechnology and Food Technology, Durban University of Technology, Durban, South Africa
b Key Laboratory of Industrial Fermentation Microbiology of Ministry of Education, Tianjin University of Science and Technology, Tianjin 300457, China
A R T I C L E I N F O
A B S T R A C T
Keywords:
In this study, the xylose reductase gene (XRTL) from Thermomyces lanuginosus SSBP was expressed in Pichia
pastoris GS115 and Saccharomyces cerevisiae Y294. The purified 39.2 kDa monomeric enzyme was optimally
active at pH 6.5 and 50 °C and showed activity over a wide range of temperatures (30–70 °C) and pH (4.0–9.0),
with a half-life of 1386 min at 50 °C. The enzyme preferred NADPH as cofactor and showed broad substrate
specificity. The enzyme was inhibited by Cu2+, Fe2+ and Zn2+, while ferulic acid was found to be the most
potent lignocellulosic inhibitor. Recombinant S. cerevisiae with the XRTL gene showed 34% higher xylitol pro-
duction than the control strain. XRTL can therefore be used in a cell-free xylitol production process or as part of a
pathway for utilization of xylose from lignocellulosic waste.
Xylose reductase
Lignocellulose
Thermomyces lanuginosus
Ferulic acid
Xylitol
1. Introduction
Microbial xylose reductases (EC 1.1.1.21) initiate the first step of
xylose metabolism, catalyzing the reduction of xylose to xylitol, and its
With annual global production exceeding 200 billion tons, lig-
nocelluloses are the most abundant biomass on planet Earth (Jiang
et al., 2018). The last decade has witnessed growing research interest
towards exploitation of lignocellulosic biomass for production of sev-
eral value-added compounds (Rao et al., 2016). Most of the research
has been focused on conversion of the recalcitrant lignocellulosic bio-
mass into viable products. Although chemical methods are economical,
manipulation of the enzymatic machinery of microbial systems for
enhanced bioconversion of lignocellulosic biomass is preferred due to
its environment-friendly nature and for long-term sustainability.
The pentose sugar, xylose, constitutes a major part of the lig-
nocellulosic hemicellulose and has the potential for production of sev-
eral value-added compounds. Xylitol, a low-calorie sugar substitute in
food, pharmaceuticals, nutraceuticals, and beverage industries, is one
such product with an annual market of USD 340 million per year, and
expected to reach USD 670 million by 2020 (Medina et al., 2018).
Current production of industrial xylitol is expensive and energy-in-
tensive which requires pure xylose and high temperature and pressure
for catalytic hydrogenation. Although biotechnological production of
xylitol is cost-effective and uses milder conditions, there are only a few
microorganisms that can metabolize xylose, and that too with poor
efficiency and yield.
properties affect efficiency of xylose fermentation (Kratzer et al., 2006).
The enzyme belongs to the aldo-keto reductase (AKR) super family,
commonly found in yeasts and filamentous fungi. Xylose reductase has
attracted significant scientific attention due to its application in the
fermentation of lignocellulosic biomass to ethanol and xylitol. Although
the use of lignocellulosic biomass for bioconversion of xylose to xylitol
using xylose reductase is promising, there is a lack of knowledge on the
effect of several lignocellulose-derived phenolic and non-phenolic by-
products produced during pretreatment processes. Therefore, it is cru-
cial to characterize xylose reductases from microorganisms to explore
maximum efficiency of bioconversion of xylose into value-added pro-
ducts such as polyols or ethanol. Genome and secretome analysis of the
thermophilic filamentous fungus Thermomyces lanuginosus has revealed
several enzymes of industrial importance (Mchunu et al., 2013). This
study was aimed at expressing a novel xylose reductase from T. lanu-
ginosus, characterizing and investigating the effect of different in-
hibitors and substrates. Additionally, the enzyme was expressed in
Saccharomyces cerevisiae to investigate its effect on xylitol production.
⁎ Corresponding author.
E-mail address: puriadarsh@gmail.com (A.K. Puri).
https://doi.org/10.1016/j.biortech.2019.02.102
Received 18 December 2018; Received in revised form 21 February 2019; Accepted 22 February 2019
Availableonline23February2019
0960-8524/©2019ElsevierLtd.Allrightsreserved.

M. Zhang, et al.
BioresourceTechnology281(2019)374–381
2. Materials and methods
2.3. Transformation of P. pastoris and S. cerevisiae with XRTL
2.1. Microorganisms, plasmids and cultural conditions
The recombinant plasmids XRTL-pPIC9K and 19up-TEF1-XRTL-
CYC-down-pAUR135 were linearized with SacI and SpeI, respectively.
These linearized plasmids were transformed into P. pastoris GS115 and
S. cerevisiae Y294 via electroporation (Gene pulser Xcell, Bio-Rad) by
following the manufacturer’s protocol. Positive transformants of P.
pastoris were selected on YPD-Geneticin plates containing Geneticin,
according to the manufacturer's instructions (Invitrogen). Similarly,
electroporated cells of S. cerevisiae, were incubated in YPD broth at
30 °C for 4 h and selected on YPD medium containing 0.5 μg/ml
Aureobasidin A (Takara). Selection of pAUR135 excised clones was
performed according to manufacturer's protocols (Takara). Colony PCR
was then performed to identify positive clones of P. pastoris and S.
cerevisiae, according to Looke et al. (2011).
The gene for xylose reductase (XRTL) was sourced from T. lanugi-
nosus SSBP cultivated in a medium containing (w/v) 2% xylose, 0.087%
K2HPO4, 0.068% KH2PO4, 0.02% KCl, 0.1% NH4NO3, 0.02% MgSO4
and 0.4% yeast extract; pH 6.0. Escherichia coli JM109 was used as a
host for plasmid construction and cultivated in a low salt Luria–Bertani
(LB) medium at 37 °C. Pichia pastoris GS115 (Invitrogen) and
Saccharomyces cerevisiae Y294 (MATα his3Δleu2Δlys2Δura3Δ) [ATCC
201160] were used for extracellular and intracellular expression of
recombinant proteins, respectively. The plasmid pPIC9K (Invitrogen)
was used to express XRTL in P. pastoris while pAUR135 (Takara) was
used to develop the S. cerevisiae integrative plasmid. P. pastoris was
cultivated in YPD (1% yeast extract, 2% peptone, 2% glucose) medium
at 30 °C while S. cerevisiae Y294 cultures were grown in SC minimal
medium (0.67% yeast nitrogen base without amino acids, 0.2% yeast
synthetic drop-out medium supplements and 2% glucose) at 30 °C.
2.4. Production and purification of recombinant xylose reductase from P.
pastoris
Recombinant P. pastoris GS115 harbouring the XRTL gene was cul-
tivated in YPD medium and induced by 0.5% (v/v) methanol in BMMY
2.2. Construction of recombinant XRTL expression plasmids for P. pastoris
GS115 and S. cerevisiae Y294
medium (1% yeast extract, 2% peptone, 1.34% YNB and 4 × 10−5
%
biotin in 100 mM phosphate buffer, pH 6.0) for the production of re-
combinant protein in shake flasks as per the manufacturer’s protocol. P.
pastoris GS115 transformed with pPIC9K was used as a control.
Total RNA isolation and cDNA synthesis was performed according
to Zhang et al. (2015b). The xylose reductase gene was amplified from
cDNA using XRTL-specific (XRTL-P) primers. The PCR products were
digested with EcoRI and XbaI, and ligated into the pPIC9K vector lin-
earized with EcoRI and AvrII, and the construct was named XRTL-
pPIC9K.
The XRTL gene was integrated into S. cerevisiae Y294 chromosome
XV at a solo long terminal repeat site according to Flagfeldt et al.
(2009). The homologous arms, promoter (TEF1) and terminator (CYC1)
of the expression cassette were amplified using S. cerevisiae Y294
genomic DNA. The final expression plasmid (Fig. 1) was named as
19up-TEF1-XRTL-CYC-down-pAUR135.
The recombinant XRTL was purified on an AKTA Purifier 100 (GE
Healthcare) using classical purification steps according to Zhang et al.
(2015a), using DEAE (HiTrap DEAE FF 5 ml, GE Healthcare) and gel
filtration (Superdex 200 Increase 10/300 GL, GE Healthcare) columns.
Homogeneity of the purified protein was estimated on SDS-PAGE gels.
Protein concentration was measured using the method of Bradford
(Bradford, 1976) with bovine serum albumin (BSA) as the standard.
Xylose reductase activity was estimated according to Kumar and
Gummadi (2011) by continuous spectrophotometric assay using com-
mercial xylose (Sigma) as substrate and NADPH (molar extinction
coefficient = 6.22 mol/l/cm) as a reductant. One enzyme unit was
Fig. 1. Steps involved in the construction of recombinant 19up-TEF1-XRTL-CYC-down-pAUR135 expression plasmid. AUR1-C: Aureobasidin A resistant gene in S.
cerevisiae; Amp: ampicillin resistance gene; GAL10p-GIN11M86: DNA sequence related to galactose-inducible growth inhibition in S. cerevisiae; ColE1 ori: origin of
replication in E. coli.
375

M. Zhang, et al.
BioresourceTechnology281(2019)374–381
defined as the amount of enzyme catalyzing the oxidation of 1 μmol of
NADPH per minute at the assay conditions.
2.7. Determination of inhibitory concentrations and inhibition mechanisms
of lignocellulose-derived by-products (LDBs)
Previously reported representatives of lignocellulose-derived in-
hibitory compounds, ferulic acid, vanillin, formic acid, acetic acid,
benzonic acid, coumaric acid, cinnamic acid and gallic acid, were used
to evaluate their inhibitory effect on XRTL. Their half maximal in-
hibitory concentrations (IC50) were determined by measuring XRTL
activities at varying gradients of increasing LDBs concentrations at
50 °C, pH 6.5 (100 mM of Tris-HCl buffer). The potent phenolic and
non-phenolic inhibitors were selected and their inhibition kinetics was
studied using the standard mixed inhibition model of Michaelis–Menten
equation:
2.5. Characterization of recombinant XRTL
2.5.1. Effect of pH and temperature and the thermal denaturation kinetics
To study the effect of pH on the enzyme, the activity was measured
at 50 °C, using 100 mM sodium citrate (pH 4.0–5.5); potassium phos-
phate (pH 6.0–8.0); and glycine-NaOH (pH 9.0–10). To determine the
optimum temperature, the XRTL activity was measured between 30 °C
and 80 °C at optimal pH. Thermostability and pH stability were in-
vestigated by incubating the enzyme at 40–70 °C in 20 mM phosphate
buffer (pH 7) and at pH 4–10 at 50 °C, respectively. Samples were
withdrawn every 15 min to calculate the residual enzyme activities.
The thermal denaturation kinetics of XRTL was studied in the range
of 50–70 °C using the standard first-order enzyme deactivation kinetics.
Change in enthalpy (ΔH, kJ/mol), change in free energy (ΔG, kJ/mol),
and change in entropy (ΔS, J/mol-K) for thermal denaturation of XRTL
were determined by using the following equations (Eyring and Stearn,
1939):
Vmax [S]
v0 =
[I]
[I]
Km 1 +
+ [S] 1 +
(
)
(
)
K
K
(6)
i
i
Ki values were estimated by the nonlinear regression analysis of
Graphpad Prism 6.0 (GraphPad Software, Inc.).
2.8. Xylitol production by recombinant S. cerevisiae Y294 with the XRTL
gene
KbT
h
kd
=
e
(−ΔH/RT)e(−ΔS/RT)
⎛
⎝
⎞
⎠
(1)
(2)
Shake-flask fermentations were performed to determine xylitol
production by S. cerevisiae Y294 harbouring the XRTL gene (S. cerevisiae
Y294-XRTL). Two pre-inoculation steps were used. During the first pre-
inoculation step, a single colony of S. cerevisiae Y294-XRTL was grown
in 25 ml YPD medium at 30 °C and 200 rpm. After 72 h, 1 ml of culture
was transferred into 50 ml fresh SC minimal medium and incubated at
30 °C and 200 rpm. This secondary pre-inoculation culture (0.5 ml)
were transferred into 50 ml of SC minimal medium with an additional
2% xylose (produced from sugarcane bagasse, data not shown). The
initial cell concentration was the same as that of the controlled ex-
periments with S. cerevisiae Y294 without XRTL gene. Glucose, xylose,
and xylitol concentrations were determined by the LCMS-2020 liquid
chromatography system (Shimadzu) using an Aminex HPX 87-H
column (Bio-Rad) and an ELSD-LT II detector (Shimadzu). The samples
on the column were eluted at 50 °C with water as the mobile phase, at a
flow rate of 0.5 ml/min.
ΔH = Ed − RT
Kdh
⎛
⎞
⎠
⎜
ΔG = −RTln
KbT
(3)
(4)
⎝
(ΔH − ΔG)
ΔS =
T
where kd = thermal inactivation rate constant; E0 and Et were the initial
(at t = 0) and residual (at t = t) XRTL activities; (Ed) = activation en-
ergy for denaturation of XRTL; Kb [Boltzmann’s constant (R/
N)] = 1.38 × 10−23 J/K; h (Planck’s constant) = 6.63 × 1034 J s; N
(Avogadro’s number) = 6.02 × 1023 per mol and
R (gas con-
stant) = 8.314 J/mol-K.
2.5.2. Effect of metal ions, additives and salts on purified recombinant
XRTL
The effects of metal ions (Ca2+, Cu2+, Fe2+, Mg2+, Mn2+, and
Zn2+) and reducing agents (EDTA and DTT) were studied by determi-
nation of XRTL specific activity in the presence of three concentrations
of these compounds (0.5, 1 and 2 mM). The experiments were done in
triplicate under optimal conditions and activity without any additive
was considered as 100%. Similarly, the effect of 0–1 M NaCl and KCl on
XRTL activity was determined under optimal conditions and activity
without any additive was considered as 100%.
3. Results and discussion
3.1. Expression, production and purification of XRTL
The expression plasmids, XRTL-pPIC9K and 19up-TEF1-XRTL-CYC-
down-pAUR135 were successfully constructed by subcloning the XRTL
gene from T. lanuginosus into pPIC9K and 19up-TEF1- CYC-down-
pAUR135 under control of the AOX1 and TEF1 promoters, respectively.
P. pastoris GS115 was used as a host for extracellular expression of the
XRTL protein for characterization and purification purposes as pur-
ification allowed testing of xylose reductase activity on the absence of
host aldo-keto reductase or its isozymes, while S. cerevisiae Y294 was
used to study the effect of intracellular expression of XRTL on xylitol
production. The XRTL gene sequence was deposited in the NCBI gene
bank with accession number MG437302.
2.6. Enzyme kinetics in the presence of different substrates and cofactors
Kinetic parameters of the purified recombinant XRTL were de-
termined against D-erythrose, D-ribose, D-arabinose, D-xylose, D-lyxose,
D-allose, D-altrose, D-glucose, D-mannose, L-gulose, D-idose, D-galactose,
D-talose, fructose, sucrose, maltose, lactose, cellobiose and xylobiose by
measuring the initial velocities at constant NADPH (0.15 mM) and
different concentrations of substrates ranging from 5 to 175 mM. The
specific activity data was used to estimate the kinetic constants by fit-
ting to the Michaelis–Menten equation:
The P. pastoris transformants grown on YPD plates containing 2 mg/
ml Geneticin were selected for XRTL production. Maximum XRTL ac-
tivity of 5.23
0.21 U/ml was achieved after 96 h of methanol in-
duction. The recombinant enzyme was purified to homogeneity with an
overall 13.77-fold increase in specific activity, through conventional
ammonium sulfate precipitation (29.76 U/mg), ion-exchange chroma-
tography (55.61 U/mg) and gel filtration chromatographic (78.06 U/
mg) steps. The purified protein had an apparent molecular mass of
∼39 kDa on SDS-PAGE gels (Fig. 2) and appears to migrate as a
monomer during Superdex gel filtration chromatography (data not
shown). A slightly higher molecular weight than the theoretically-
Vmax [S]
v0 =
[S] + Km
(5)
where Vmax is the maximal velocity; Km is the Michaelis-Menten con-
stant; S is the substrate concentration, and v0 is the mean initial velo-
city. The Michaelis-Menten model was also used to study the enzyme
specificity against 20–200 µM NADPH and NADH (disodium salt) as the
known cofactors.
376

M. Zhang, et al.
BioresourceTechnology281(2019)374–381
thermophilic fungus, Talaromyces emersonii, (Fernandes et al., 2009).
Xylose reductase from Neurospora crassa (Woodyer et al., 2005) and
Rhizopus oryzae (Zhang et al., 2015a) also showed optimal activities at
50 °C, similar to this study. The purified XRTL was extremely stable at
40 and 50 °C retaining 100% and 94.2% activity, respectively, after
120 min. The enzyme could only retain 24% activity at 60 °C, while the
activity was completely lost when incubated at 70 °C for 45 min. Half-
life (t1/2) of purified XRTL at 50 °C, 55 °C and 60 °C were 1386 min,
49.86 min and 14.53 min, respectively. It is evident that the thermo-
stability of XRTL is superior to the previously-reported xylose re-
ductases from R. oryzae, Debaryomyces nepalensis NCYC 3413, N. crassa
and Kluyveromyces marxianus NBRC 1777 which exhibited t1/2 values of
83 min at 50 °C, 2.47 min at 50 °C, 94 min at 40 °C and 10 min at 42 °C,
respectively (Malla and Gummadi, 2018; Woodyer et al., 2005; Zhang
et al., 2011; Zhang et al., 2015a). Therefore, XRTL is the most ther-
mostable xylose reductase reported thus far. Its high thermostability
may be attributed to the presence of a high number of alanine residues
(39 aromatic residues). Alanine has been considered as the best helix-
forming residue in thermophiles (Kumwenda et al., 2013).
Although there is a plethora of data available on thermal deacti-
vation kinetics of various enzymes, very limited information is avail-
able on the kinetics of microbial xylose reductases. Study of the thermal
deactivation kinetics of XRTL resulted in positive values of all ther-
modynamic parameters (Table 1). The high positive ΔH values de-
creased slightly when the temperature was increased from 50 to 70 °C.
This infers structural stability of the protein due to strong covalent
bonds. Gibbs free energy (ΔG) of activation for XRTL denaturation
decreased when temperature was increased from 50 to 60 °C, denoting
the decreasing stability of this enzyme at higher temperatures. A slight
increase in the value of ΔG from 60 to 70 °C was primarily entropy-
driven and associated with the concomitant decrease in ΔS value, in-
dicating that XRTL tends to defy disorders before the onset of unfolding,
as with many thermostable enzymes (Siddiqui, 2017). More recently,
Malla and Gummadi (2018) reported an enthalpy-driven thermal de-
activation of mesophilic xylose reductase from D. nepalensis NCYC 3413
showing negative ΔH and ΔG values and a positive ΔS value.
Fig. 2. SDS-PAGE analysis of XRTL-pPIC9K expressed in P. pastoris after protein
purification. Lane 1: supernatant of P. pastoris GS115 transformed with pPIC9K;
Lane 2: supernatant of P. pastoris GS115 transformed with XRTL-pPIC9K; Lane
3: Spectra multicolour broad range protein ladder; Lane 4: purified recombinant
xylose reductase after gel filtration chromatography; Lane 5: eluent fraction of
anion exchange chromatography; Lane 6: ammonium sulphate precipitation
fraction.
predicted molecular weight of 36.74 kDa indicates possible glycosyla-
tion of the protein during the secretion process in P. pastoris. In this
study, three potential N-glycosylation sites were identified at positions
26, 142 and 168 of XRTL using the NetNGlyc 1.0 Server (Blom et al.,
2004).
Marked inhibition of XRTL activity was observed in the presence of
2 mM Cu2+ and Fe2+ ions (Table 2). Another transition metal ion,
Zn2+, also emerged as a potent inhibitor at 0.5 mM, 1 mM and 2 mM
when it reduced the basal XRTL activity by 69%, 50% and 80%, re-
spectively. The strong inhibitory potential of Cu2+, Fe2+ and Zn2+ was
also observed in other known xylose reductases (Kumar and Gummadi,
2011; Lee et al., 2003; Paidimuddala et al., 2017a). Kumar and
Gummadi (2011) suggested a Cu2+-mediated site-specific oxidation
and a Fe2+-mediated reduction as possible mechanisms for inhibition of
xylose reductases. The other divalent metal-ions used in this study
could only show a slight inhibition at all concentrations. No significant
effect on XRTL activity was observed in the presence of EDTA, which
suggests that XRTL is a metal-free enzyme. Similar observations were
also reported in xylose reductases from Candida parapsilosis (Lee et al.,
2003) and D. nepalensis NCYC 3413 (Kumar and Gummadi, 2011). The
reducing agent, dithiothreitol (DTT) inhibited XRTL activity as ob-
served previously in the xylose reductases from D. nepalensis NCYC
3413 (Kumar and Gummadi, 2011). In contrast, addition of DTT sig-
nificantly enhanced the activity of xylose reductase from K. marxianus
NBRC1777 (Zhang et al., 2011).
3.2. Characterization of XRTL
Purified XRTL demonstrated activity in a broad range of pH
(4.0–9.0) and temperature (30–70 °C). The enzyme was optimally ac-
tive at pH 6.5 and exhibited remarkable stability at pH 6 and 7. It re-
tained 58.7%, 73.76% and 45.95% activity after 30 min at pH 5, pH 8
and pH 9, respectively. However, the activity at pH 5 and pH 8 declined
to 27% and 54%, respectively, after an incubation of 60 min. The half-
life (t1/2) at pH 5, 8 and 9 were 34.56 min, 69.14 min and 22.98 min,
respectively. These results indicated that XRTL was more stable in al-
kaline condition than acidic conditions. The optimum pH and pH sta-
bility of XRTL was similar to other known xylose reductases (Dasgupta
et al., 2016; Kumar and Gummadi, 2011). XRTL exhibited high stability
at neutral pH and similar pH stability profiles were observed for other
xylose reductases. The reason for these xylose reductases showing si-
milar pH stability profiles could be that xylose reductase is an in-
tracellular enzyme and the internal pH environment of cells is similar.
XRTL showed optimum activity at 50 °C, which is much higher than
the optimum (37 °C) of the only reported xylose reductase from another
Table 1
Estimated thermodynamic parameters of thermal deactivation of purified XRTL at different temperatures.
ΔH(kJ mol−1).
ΔG(kJ mol−1
)
ΔS(kJ mol−1
)
Temperature (K)
kd (min−1
)
t1/2 (min)
323.15
328.15
333.15
343.15
0.0005
0.0139
0.0477
0.0818
1386
216.00
215.96
215.92
215.84
99.78
92.298
90.33
91.58
359.66
376.87
376.99
362.10
49.856
14.528
8.472
377

M. Zhang, et al.
BioresourceTechnology281(2019)374–381
Table 2
Table 3
Effect of 0.5 mM, 1 mM and 5 mM divalent metal ions and reducing agents on
the activity of XRTL.
Apparent kinetic parameters of purified XRTL for NADPH-dependent reduction
of different sugars and cofactor preference.
Additives
Specific activity (U/mg)
Substrate
Vmax (μmol/mg/
min)
Km (mM)
Kcat (per min)
Kcat/Km
(mM/min)
0.5 mM
1 mM
2 mM
55
D-Erythrose
D-Ribose
D-Arabinose 88.48
76.79
79.54
0.54
1.16
1.50
1.27
10.07
39.13
64.2
15.36
ND
77.15
101.1
222.4
ND
0.33
1.74
2.69
10101
10463
11639
11656
ND
100
180
233
236
1003.08
267.39
181.29
758.85
ND
16.12
5.23
22.68
ND
11.25
11.12
92.21
ND
ND
ND
ND
ND
ND
ND
Kcat/Km
(μM/min)
403.45
Metal ions
Ca2+
70
68
71
74
70
52
2.5
1.9
2.5
2.2
1.8
2.3
62
65
67
70
68
38
1.1
2.2
1.3
2.8
2.3
1.3
2.8
Cu2+
4
0.3
D-Xylose
D-Lyxose
D-Allose
D-Altrose
D-Glucose
D-Mannose
D-Gulose
D-Idose
D-Galactose
D-Talose
Fructose
Sucrose
Maltose
Lactose
Cellobiose
Xylobiose
Cofactors#
88.61
ND*
9.459
4.02
38.35
ND
29.84
2.62
71.29
ND
ND
ND
ND
ND
ND
ND
0.89
Fe2+
24
63
59
15
1.7
1.5
2.0
1.3
Mg2+
Mn2+
Zn2+
0.33
0.17
1.67
5.30
4.86
14.93 5045
ND
54.53
3.86
7.23
1244
528.8
59
22
300
Reducing agents
EDTA
DTT
3.418
0.13
2.47
349
3925
344.2
9378
ND
ND
ND
ND
ND
ND
ND
633
16
459
75
68
75
1.21
1.0
1.7
72
58
2.1
1.3
69
51
1.6
0.8
30.95
101.7
ND
ND
ND
ND
ND
ND
ND
Control
Purified XRTL was highly tolerant to monovalent cations (Na+ and
K+). It retained 85% and 80% of its activity in the presence 100 mM
NaCl and KCl, respectively. About 45% of activity was retained at
600 mM of Na+ and K+, with residual activities of 33.72
1.28 and
Vmax (μmol/mg/
min)
94.43
5.14
Km (μM)
Kcat (per min)
35.31
0.86 U/mg, respectively. XRTL was more tolerant to Na+ than
the other moderately halotolerant xylose reductase reported by Kumar
and Gummadi (2011). The xylose reductase from D. nepalensis could
only retain 15% of activity in the presence of 500 mM NaCl. Higher
composition of acidic amino acids (12.42%) than basic amino acids
(10.56%) could be the reason for higher halotolerance of XRTL.
NADPH
NADH
2.24
0.13
30.44
112.1
2.69
6.02
12281
676.7
129
17.37 6.04
* ND: not detectable.
#
Kinetic parameters with NADPH and NADH were calculated using 1 mM
xylose as substrate
3.3. Substrate specificity and coenzyme preference
this study, no detectable activity was observed with fructose and tested
disaccharides. This indicates that XRTL may only catalyze the reduction
of aldehydic monosaccharides. Comparative kinetic studies of XRTL
towards monosaccharides having R-configured hydroxyl groups at C-2
with the corresponding C-2 epimers (D-ribose vs. D-arabinose; D-allose
vs D-altrose), showed that the OH group at the C2(R) position resulted
in higher catalytic efficiency and affinity than OH group at C2(S) po-
sition. These present results are consistent with xylose reductase from
C. intermedia (Nidetzky et al., 2003) and Cryptococcus flavus (Mayr
et al., 2003) where the OH group at the C2(R) was preferred for enzyme
activity. However, xylose reductase from D. hansenii binds sugars pro-
miscuously and the C2(R) hydroxyl group does not contribute to the
specificity of ligands binding to this enzyme (Biswas et al., 2012). XRTL
showed high catalytic efficiency and affinity toward sugars having
hydroxyl group at C2(R) and C3(S), but no detectable activities were
observed with their corresponding C-3 epimers (D-xylose vs D-lyxose; D-
glucose vs D-mannose; D-galactose vs D-talose). Interestingly, R-config-
ured hydroxyl groups, when both present at C-2 and C-3 positions, do
not favour the activity of XRTL. This has not been reported previously
from all aldose reductases. Recombinant XRTL exhibited unique sub-
strate specificity and kinetic characteristics as compared to other re-
ported xylose reductases. This may be attributed to an extended loop
(residue 314 to residue 318) at the C-terminal of XRTL, which has been
shown to be important for substrate recognition in other AKR proteins
(Yamamoto et al., 2016). The phylogenetic analysis clearly indicated
that XRTL evolved from diverse ancestors to well-known yeast and fi-
lamentous fungal xylose reductases (Fig. 3).
HPLC analysis of the reaction mixture containing 1 mM xylose
confirmed reductase activity of the purified protein by the conversion
of xylose to xylitol. To test XRTL affinity towards other sugars, its
preference to NADH and NADPH was studied. Purified XRTL exhibited
dual coenzyme specificity for both NADH and NADPH, however, the
protein preferred NADPH as indicated from a 3.68-times lower Km
value for NADPH than NADH (Table 3). Also, XRTL exhibited 66.8-fold
higher catalytic efficiency towards NADPH than NADH. Evidently, the
specific activity of XRTL with xylose as substrate and NADPH as
coenzyme was 78 U/mg, while it showed a specificity activity of
2.68 U/mg with NADH as the coenzyme. This dual coenzyme utilization
with preference for NADPH was also reported in many other xylose
reductases (Fernandes et al., 2009; Kumar and Gummadi, 2011). The
broad substrate specificity of purified XRTL is apparent and kinetic
parameters are shown in Table 3. XRTL showed higher affinity and
catalytic efficiency towards pentoses than hexoses. However, the
highest affinity (Km), turnover number (Kcat) and catalytic efficiency
(Kcat/Km) of the purified enzyme was observed towards erythrose at
10.07 mM, 10,101 per min and 1003.08 mM/min, respectively.
Neuhauser et al. (1997) also reported erythrose as the most preferred
substrate for xylose reductase from Candida tenuis, but xylose reductase
from C. parapsilosis (Lee et al., 2003) showed lower catalytic efficiency
and affinity towards erythrose than xylose. All attempts to use lyxose,
mannose, talose, fructose, sucrose, maltose, lactose, cellobiose and xy-
lobiose as alternative substrates in the presence of NADPH or NADH
failed. Different xylose reductases from different sources show entirely
different substrates preferences. While xylose reductase from D. nepa-
lensis NCYC 3413 (Kumar and Gummadi, 2011) could not show activity
against mannose, it was accepted as substrate for xylose reductase from
Candida intermedia (Nidetzky et al., 2003). Similarly, fructose – A ke-
tonic monosaccharide, acted as substrate for xylose reductase from C.
parapsilosis (Lee et al., 2003), but not for xylose reductase from N. crassa
(Woodyer et al., 2005). The disaccharide sucrose served as a substrate
for xylose reductase from Debaryomyces hansenii (Biswas et al., 2012),
but not for xylose reductase from N. crassa (Woodyer et al., 2005). In
3.4. Inhibition kinetics of XRTL due to LDBs
During pretreatment of lignocellulosic biomass, phenolic com-
pounds are the major by-products that inhibit microbial and enzymatic
biocatalysts (Jönsson and Martín, 2016). Among the tested plant-de-
rived phenolic compounds, ferulic acid exhibited highest inhibitory
effect on XRTL with an IC50 of 0.055 mM. This was followed by vanillin
(IC50 of 0.149 mM), coumaric acid (IC50 of 3.612 mM), cinnamic acid
378

M. Zhang, et al.
BioresourceTechnology281(2019)374–381
(IC50 of 8.15 mM), benzoic acid (IC50 of 8.785 mM) and gallic acid (IC50
of 12.27 mM). The non-phenolic compounds, acetic acid and formic
acid were the least potent inhibitors with IC50 values of 15.62 mM and
16.36 mM, respectively (Fig. 4).
Higher enzyme inhibition by phenolic compounds may be attributed
to the presence of hydroxyl and carbonyl groups as reported recently by
Zhai et al. (2018). However, in contrast to their findings where the
authors reported highest inhibition by smaller molecular weight com-
pound, this study showed strongest inhibition by a phenolic compound
with highest molecular weight among the tested inhibitors (ferulic acid,
194.19 g/mol). This can be due to steric hindrance caused by fer-
uloylation of the substrate and/or the enzyme and/or the enzyme-
substrate complex.
Gallic acid (IC50 of 12.27 mM) and acetic acid (IC50 of 15.62 mM)
were more inhibitory to XRTL than the xylose reductase from D. ne-
palensis, where IC50 values were reported as 40 mM and 60 mM, re-
spectively (Paidimuddala et al., 2017b). However, inhibition by vanillin
on xylose reductase from D. nepalensis was similar to the present study,
with an IC50 value of 0.14 mM. Acetic acid showed much less of an
effect on crude xylose reductase from Candida tropicalis with an IC50 of
190 mM (Rafiqul et al., 2015).
To understand the mechanism of inhibition, the enzyme kinetics of
XRTL was investigated in the presence of three different concentrations
of ferulic acid, vanillin and acetic acid in separate experiments. As the
concentrations of ferulic acid was increased, Km values increased from
Fig. 3. Phylogenetic analysis of XRTL amino acid sequence with existing xylose
reductases. The analysis was done using the neighbour-joining method avail-
able with MEGA7.0 software. The NCBI accession numbers of xylose reductases
have been provided after the name of each microbial source.
15.36
88.61
1.14 to 52.71
1.62 to 45.94
4.58 mM and Vmax values decreased from
1.56 U/mg (Table 4). This suggested that
ferulic acid inhibits XRTL by the mechanism of mixed inhibition. Si-
milar patterns of mixed inhibition of XRTL were also observed for va-
nillin and acetic acid. The Ki values of ferulic acid, vanillin and acetic
acid were estimated to be 0.015, 0.042 and 3.75 mM, respectively
(Table 4). Ferulic acid showed a strong inhibitory effect with lowest Ki
values. This is the first report on ferulic acid showing strong inhibition
towards a xylose reductase.
3.5. Xylitol production by recombinant S. cerevisiae Y294 with the XRTL
gene
XRTL was successfully expressed in S. cerevisiae Y294 with an ac-
tivity of 0.22 U/mg activity in the crude cell extract. Xylitol cannot be
efficiently metabolized by S. cerevisiae, so xylose conversion requires
the simultaneous metabolism of a co-substrate such as glucose and
ethanol for continuous regeneration of NAD(P)H and generation of
maintenance energy. S. cerevisiae Y294 and S. cerevisiae Y294-XRTL
were cultured in SC minimal medium with 2% glucose and xylose
(Fig. 5). Immediately after inoculation, both strains consumed glucose
and concomitantly produced biomass. The biomass formation was
halted and glucose was completely consumed after 24 h. Xylose de-
pletion and xylitol production was then detected. The xylose reduction
Fig. 4. Residual activity at different concentration of lignocellulose derived by-
products (LDBs). The concentrations of LDBs were expressed in terms of natural
logarithm.
Table 4
Kinetic parameters of XRTL in the presence of various LDBs.*
LDB
Concentration (mM)
Vmax (U/mg)
Km (mM)
Ki (mM)
–
Inhibition mechanism
Control†
Ferulic acid
–
88.61
67.05
56.58
45.94
1.62
15.36
30.72
37.98
52.71
1.14
2.36
3.71
4.58
0.025
0.05
0.06
1.65
1.93
1.56
0.015
0.002
0.003
Mixed
Mixed
Mixed
Vanillin
0.05
0.1
0.15
82.39
78.06
65.89
2.50
2.20
2.23
30.31
38.57
76.82
2.87
3.10
5.59
0.042
3.75
Acetic acid
10
15
20
76.14
64.53
52.43
2.13
2.20
1.72
44.36
3.33
5.28
57
4.72
0.59
73.24
* LDB: lignocellulose-derived by-products; Km: Michaelis-Menten constant; Vmax: maximal velocity; Ki: inhibition constant.
†
Controlled experiments were performed without any LDBs in the reaction mixture.
379

M. Zhang, et al.
BioresourceTechnology281(2019)374–381
be attributed to the much higher affinity and catalytic rate of XRTL
compared to other enzymes, since the intracellular environment is
much more complex than the extracellular environment. XRTL has a
low Km resulting in efficient binding with xylose. Those results com-
bined with the stability studies suggest that XRTL can be used as an
efficient component for construction of a xylose metabolic pathway in
S. cerevisiae, specifically in thermotolerant S. cerevisiae for the bio-
conversion of xylose to value-added compounds.
4. Conclusions
In this study, a thermostable recombinant xylose reductase from T.
lanuginosus SSBP was successfully expressed in P. pastoris and S. cere-
visiae. The enzyme showed broad substrate specificity and preferred
NADPH over NADH as the cofactor. The recombinant S. cerevisiae
harbouring XRTL successfully produced xylitol. Future studies will be
directed towards expressing the enzyme at high copy number to im-
prove xylitol production. Additionally, co-crystallization of XRTL with
potent inhibitors, substrates and cofactors will be helpful to identify key
amino acid residues and develop a robust enzyme using site-directed
mutagenesis.
Fig. 5. Growth and xylitol fermentation profiles of S. cerevisiae Y294 versus
recombinant S. cerevisiae Y294-XRTL in SC minimal medium without pH ad-
justment.
Acknowledgements
and xylitol formation slowed slightly after 100 h incubation. S. cerevi-
siae Y294-XRTL which contained a single copy of XRTL in its chromo-
The financial support from the National Research Foundation,
Republic of South Africa and Durban University of Technology is
gratefully acknowledged.
some, showed xylitol production of 4.4
cubation, about 34% higher than that of S. cerevisiae Y294 without
XRTL (3.27 0.15 g/L).
0.13 g/L after 120 h in-
S. cerevisiae is a promising candidate for industrial production of
biofuels and value-added bio-chemicals from lignocellulosic biomass.
High temperatures fermentation can effectively accelerate the rate of
reaction, avoid or minimize microbial contamination, shorten microbial
fermentation period and facilitate downstream product recovery.
Therefore, many researchers have attempted to improve the thermo-
tolerance of S. cerevisiae for ethanol and chemicals production
(Benjaphokee et al., 2012; Xu et al., 2018). However, such strains still
have low fermentation efficiency, possibly because the growth tem-
perature is high and not compatible with the optimum temperature of
the critical enzymes in the metabolic pathways. This decreases the
protein stability and leads to lower metabolic rates. As S. cerevisiae does
not have a specific xylose reductase, many researchers have attempted
to express xylose reductases from mesophilic yeasts such as Pichia sti-
pitis, Candida species and the S. cerevisiae GRE3 gene (coding for a non-
specific aldose reductase) in S. cerevisiae, mainly for lignocellulose
bioconversion. However, enzyme activity and stability of those xylose
reductase is not satisfactory at higher temperatures (Neuhauser et al.,
1997; Zhang et al., 2011). Interestingly, xylose reductase from the
thermotolerant yeast K. marxianus showed optimal activity only at
20 °C. Therefore, they are not suitable for construction of xylose me-
tabolic pathways in thermotolerant S. cerevisiae. In this study, one copy
of XRTL gene was integrated into the chromosome of S. cerevisiae, and it
had a significant effect on xylose reductase activity and xylitol pro-
duction. Kogje and Ghosalkar (2016) using a high copy number
plasmid, overexpressed the S. cerevisiae GRE3 gene and xylose reductase
from C. tropicalis and N. crassa in S. cerevisiae. Xylitol production of S.
cerevisiae harbouring the GRE3 gene and xylose reductase gene from C.
tropicalis was around 2 times more than the S. cerevisiae control. The N.
crassa gene showed no significant effect on xylitol production. Kim
et al. (1999) studied the effect of copy number of the P. stipitis xylose
reductase in S. cerevisiae and showed that the xylose reductase activity
was only 2.5 times higher when comparing a copy number of 30 to a
single copy. Pratter et al. (2015) pointed out that xylose reductase ac-
tivity did not correlate with the xylitol productivities and other factors
limited xylose conversion in S. cerevisiae. By comparison, a single copy
of XRTL caused a 1.70 times increase in enzyme activity and a 34%
increase in xylitol production. This increase in xylitol production may
Appendix A. Supplementary data
Supplementary data to this article can be found online at https://
doi.org/10.1016/j.biortech.2019.02.102.
References
Benjaphokee, S., Hasegawa, D., Yokota, D., Asvarak, T., Auesukaree, C., Sugiyama, M.,
Kaneko, Y., Boonchird, C., Harashima, S., 2012. Highly efficient bioethanol pro-
duction by a Saccharomyces cerevisiae strain with multiple stress tolerance to high
temperature, acid and ethanol. N. Biotechnol. 29, 379–386.
Biswas, D., Pandya, V., Singh, A.K., Mondal, A.K., Kumaran, S., 2012. Co-factor binding
confers substrate specificity to xylose reductase from Debaryomyces hansenii. PLoS
One 7, e45525.
Blom, N., Sicheritz-Ponten, T., Gupta, R., Gammeltoft, S., Brunak, S., 2004. Prediction of
post-translational glycosylation and phosphorylation of proteins from the amino acid
sequence. Proteomics 4, 1633–1649.
Bradford, M.M., 1976. A rapid and sensitive method for the quantitation of microgram
quantities of protein utilizing the principle of protein-dye binding. Anal. Biochem. 72,
248–254.
Dasgupta, D., Ghosh, D., Bandhu, S., Agrawal, D., Suman, S.K., Adhikari, D.K., 2016.
Purification, characterization and molecular docking study of NADPH dependent
xylose reductase from thermotolerant Kluyveromyces sp IIPE453. Process Biochem.
51, 124–133.
Eyring, H., Stearn, A.E., 1939. The application of the theory of absolute reacton rates to
proteins. Chem. Rev. 24, 253–270.
Fernandes, S., Tuohy, M.G., Murray, P.G., 2009. Xylose reductase from the thermophilic
fungus Talaromyces emersonii: cloning and heterologous expression of the native gene
(Texr) and a double mutant (TexrK271R + N273D) with altered coenzyme specificity. J.
Biosci. 34, 881–890.
Flagfeldt, D.B., Siewers, V., Huang, L., Nielsen, J., 2009. Characterization of chromosomal
integration sites for heterologous gene expression in Saccharomyces cerevisiae. Yeast
26, 545–551.
Jönsson, L.J., Martín, C., 2016. Pretreatment of lignocellulose: formation of inhibitory by-
products and strategies for minimizing their effects. Bioresour. Technol. 199,
103–112.
Jiang, Z., Zhao, P., Hu, C., 2018. Controlling the cleavage of the inter- and intra-mole-
cular linkages in lignocellulosic biomass for further biorefining: a review. Bioresour.
Technol. 256, 466–477.
Kim, Y.S., Kim, S.Y., Kim, J.H., Kim, S.C., 1999. Xylitol production using recombinant
Saccharomyces cerevisiae containing multiple xylose reductase genes at chromosomal
δ-sequences. J. Biotechnol. 67, 159–171.
Kogje, A., Ghosalkar, A., 2016. Xylitol production by Saccharomyces cerevisiae over-
expressing different xylose reductases using non-detoxified hemicellulosic hydro-
lysate of corncob. 3 Biotech 6, 127.
Kratzer, R., Wilson, D.K., Nidetzky, B., 2006. Catalytic mechanism and substrate se-
lectivity of aldo-keto reductases: insights from structure-function studies of Candida
380

M. Zhang, et al.
BioresourceTechnology281(2019)374–381
Paidimuddala, B., Rathod, A., Gummadi, S.N., 2017b. Inhibition of Debaryomyces nepa-
lensis xylose reductase by lignocellulose derived by-products. Biochem. Eng. J. 121,
73–82.
Pratter, S.M., Eixelsberger, T., Nidetzky, B., 2015. Systematic strain construction and
process development: xylitol production by Saccharomyces cerevisiae expressing
Candida tenuis xylose reductase in wild-type or mutant form. Bioresour. Technol. 198,
732–738.
tenuis xylose reductase. IUBMB Life 58, 499–507.
Kumar, S., Gummadi, S.N., 2011. Purification and biochemical characterization of a
moderately halotolerant NADPH dependent xylose reductase from Debaryomyces
nepalensis NCYC 3413. Bioresour. Technol. 102, 9710–9717.
Kumwenda, B., Litthauer, D., Bishop, O.T., Reva, O., 2013. Analysis of protein thermo-
stability enhancing factors in industrially important thermus bacteria species. Evol.
Bioinform. Online 9, 327–342.
Lee, J.K., Koo, B.S., Kim, S.Y., 2003. Cloning and characterization of the xyl1 gene, en-
coding an NADH-preferring xylose reductase from Candida parapsilosis, and its
functional expression in Candida tropicalis. Appl. Environ. Microbiol. 69, 6179–6188.
Looke, M., Kristjuhan, K., Kristjuhan, A., 2011. Extraction of genomic DNA from yeasts for
PCR-based applications. Biotechniques 50, 325–328.
Rafiqul, I.S.M., Sakinah, A.M.M., Zularisam, A.W., 2015. Inhibition by toxic compounds
in the hemicellulosic hydrolysates on the activity of xylose reductase from Candida
tropicalis. Biotechnol. Lett. 37, 191–196.
Rao, L.V., Goli, J.K., Gentela, J., Koti, S., 2016. Bioconversion of lignocellulosic biomass
to xylitol: an overview. Bioresour. Technol. 213, 299–310.
Malla, S., Gummadi, S.N., 2018. Thermal stability of xylose reductase from Debaryomyces
nepalensis NCYC 3413: deactivation kinetics and structural studies. Process Biochem.
67, 71–79.
Siddiqui, K.S., 2017. Defying the activity-stability trade-off in enzymes: taking advantage
of entropy to enhance activity and thermostability. Crit. Rev. Biotechnol. 37,
309–322.
Mayr, P., Petschacher, B., Nidetzky, B., 2003. Xylose reductase from the basidiomycete
fungus Cryptococcus flavus: purification, steady-state kinetic characterization, and
detailed analysis of the substrate binding pocket using structure-activity relation-
ships. J. Biochem. 133, 553–562.
Mchunu, N.P., Permaul, K., Rahman, A.Y.A., Saito, J.A., Singh, S., Alam, M., 2013.
Xylanase superproducer: genome sequence of a compost-loving thermophilic fungus
Thermomyces lanuginosus strain SSBP. . Genome Announc. 1 (3), e00388–e413.
Medina, J.D.C., Woiciechowski, A.L., Zandona, A., Brar, S.K., Magalhaes, A.I., Soccol,
C.R., 2018. Energetic and economic analysis of ethanol, xylitol and lignin production
using oil palm empty fruit bunches from a Brazilian factory. J. Clean. Prod. 195,
44–55.
Neuhauser, W., Haltrich, D., Kulbe, K.D., Nidetzky, B., 1997. NAD(P)H-dependent aldose
reductase from the xylose-assimilating yeast Candida tenuis. Isolation, characteriza-
tion and biochemical properties of the enzyme. Biochem. J. 326, 683–692.
Nidetzky, B., Bruggler, K., Kratzer, R., Mayr, P., 2003. Multiple forms of xylose reductase
in Candida intermedia: comparison of their functional properties using quantitative
structure-activity relationships, steady-state kinetic analysis, and pH studies. J. Agric.
Food Chem. 51, 7930–7935.
Woodyer, R., Simurdiak, M., Van der Donk, W.A., Zhao, H., 2005. Heterologous expres-
sion, purification, and characterization of a highly active xylose reductase from
Neurospora crassa. Appl. Environ. Microbiol. 71, 1642–1647.
Xu, K., Gao, L.M., UL Hassan, J., Zhao, Z.P., Li, C., Huo, Y.X., Liu, G.Y., 2018. Improving
the thermo-tolerance of yeast base on the antioxidant defense system. Chem. Eng. Sci.
175, 335–342.
Yamamoto, K., Higashiura, A., Suzuki, M., Shiotsuki, T., Sugahara, R., Fujii, T.,
Nakagawa, A., 2016. Structural characterization of an aldo-keto reductase (AKR2E5)
from the silkworm Bombyx mori. Biochem. Biophys. Res. Commun. 474, 104–110.
Zhang, B., Zhang, L., Wang, D., Gao, X., Hong, J., 2011. Identification of a xylose re-
ductase gene in the xylose metabolic pathway of Kluyveromyces marxianus
NBRC1777. J. Ind. Microbiol. Biotechnol. 38, 2001–2010.
Zhang, M., Jiang, S.T., Zheng, Z., Li, X.J., Luo, S.Z., Wu, X.F., 2015a. Cloning, expression,
and characterization of a novel xylose reductase from Rhizopus oryzae. J. Basic
Microbiol. 55, 907–921.
Zhang, M., Puri, A.K., Govender, A., Wang, Z.X., Singh, S., Permaul, K., 2015b. The multi-
chitinolytic enzyme system of the compost-dwelling thermophilic fungus
Thermomyces lanuginosus. Process Biochem. 50, 237–244.
Paidimuddala, B., Aradhyam, G.K., Gummadi, S.N., 2017a. A halotolerant aldose re-
ductase from Debaryomyces nepalensis: gene isolation, overexpression and biochem-
ical characterization. RSC Adv. 7, 20384–20393.
Zhai, R., Hu, J., Saddler, J.J.N., 2018. Extent of enzyme inhibition by phenolics derived
from pretreated biomass is significantly influenced by the size and carbonyl group
content of the phenolics. ACS Sustain. Chem. Eng. 6, 3823–3829.
381