Available via license: CC BY-NC-ND 4.0
Content may be subject to copyright.
Phylogenetic, functional and structural characterization of a GH10
xylanase active at extreme conditions of temperature and alkalinity
David Talens-Perales
a
, Elena Jiménez-Ortega
b
, Paloma Sánchez-Torres
a
, Julia Sanz-Aparicio
b,
⇑
,
Julio Polaina
a,
⇑
a
Institute of Agrochemistry and Food Technology, Spanish National Research Council (CSIC), Paterna, Valencia, Spain
b
Institute of Physical-Chemistry Rocasolano, Spanish National Research Council (CSIC), Madrid, Spain
article info
Article history:
Received 25 February 2021
Received in revised form 29 April 2021
Accepted 1 May 2021
Available online 03 May 2021
Keywords:
Crystal structure
Glycoside hydrolase
Pulp and paper
Xylose
Xylooligosaccharides
abstract
Endoxylanases active under extreme conditions of temperature and alkalinity can replace the use of
highly pollutant chemicals in the pulp and paper industry. Searching for enzymes with these properties,
we carried out a comprehensive bioinformatics study of the GH10 family. The phylogenetic analysis
allowed the construction of a radial cladogram in which protein sequences putatively ascribed as ther-
mophilic and alkaliphilic appeared grouped in a well-defined region of the cladogram, designated TAK
Cluster. One among five TAK sequences selected for experimental analysis (Xyn11) showed extraordinary
xylanolytic activity under simultaneous conditions of high temperature (90 °C) and alkalinity (pH 10.5).
Addition of a carbohydrate binding domain (CBM2) at the C-terminus of the protein sequence further
improved the activity of the enzyme at high pH. Xyn11 structure, which has been solved at 1.8 Å resolu-
tion by X-ray crystallography, reveals an unusually high number of hydrophobic, ionic and hydrogen
bond atomic interactions that could account for the enzyme’s extremophilic nature.
Ó2021 The Authors. Published by Elsevier B.V. on behalf of Research Network of Computational and
Structural Biotechnology. This is an open access article under the CC BY-NC-ND license (http://creative-
commons.org/licenses/by-nc-nd/4.0/).
1. Introduction
Enzymes play a fundamental role in the transformation of the
traditional chemical industry into a future, more sustainable, green
alternative. The great diversity of enzymes existing in Nature can
supply suitable catalysts to any type of biochemical reaction.
Although most enzymes are labile material, unable to stand the
harsh physicochemical conditions used in industry, life shows a
surprising plasticity, being able to generate organisms that thrive
under harsh conditions of temperature, pH, pressure, salinity,
etc., which a priori would be unimaginable. Extremophilic organ-
isms, mainly archaea and bacteria, adapted to such extreme condi-
tions become a source of enzymes that widen the range of
processes and operational conditions in which enzymes can be
applied [1,2].
Use of xylanases by the paper industry, for pulp bleaching, offer
a challenging case for stretching the enzyme capability to the limit.
Xylanases facilitate the removal of residual lignin that causes the
dark color of the pulp and can be a total or partial replacement
to the use of hazardous chemical agents [3,4]. The predominant
method used by the paper industry, the kraft process, requires
harsh conditions. Therefore, the availability of xylanases active
under extreme environments of temperature and alkalinity repre-
sents an economic bonus.
Enzymes with xylanolytic activity appear in at least nine fami-
lies of the CAZY database [5]. However, most true xylanases (EC
3.2.1.8), considered as such because of their high substrate speci-
ficity, are classified in families GH10 and GH11 [6,7]. Family
GH10 enzymes are characterized by the presence of a catalytic
domain of ca. 40 kD, with (
a
/b)
8
-barrel structure. Family GH11
enzymes contain a smaller (ca. 30 k Da) catalytic domain with b-
jelly roll structure. There is abundant published information about
enzymes from both families. Up to the present the number of
annotated protein sequences and resolved atomic structures is
about 5000 and 50 for GH10, respectively and 2000 and 30 for
GH11 (CAZY database, accessed on January 2021). Although only
a relatively small fraction of the sequences corresponds to
enzymes whose activity has been characterized experimentally,
the available data provide very useful information to predict the
enzymatic properties of an uncharacterized sequence. Several
bioinformatic approaches to this end have been reported.
https://doi.org/10.1016/j.csbj.2021.05.004
2001-0370/Ó2021 The Authors. Published by Elsevier B.V. on behalf of Research Network of Computational and Structural Biotechnology.
This is an open access article under the CC BY-NC-ND license (http://creativecommons.org/licenses/by-nc-nd/4.0/).
⇑
Corresponding authors.
E-mail addresses: xjulia@iqfr.csic.es (J. Sanz-Aparicio), jpolaina@iata.csic.es
(J. Polaina).
Computational and Structural Biotechnology Journal 19 (2021) 2676–2686
journal homepage: www.elsevier.com/locate/csbj
Conventional strategies used to identify enzymes with extre-
mophilic properties relied on their isolation from microorganisms
with such properties, for instance, Thermotoga and Thermoascus [8].
However, this approach is too limited to yield enzymes with opti-
mal physicochemical and catalytic properties that fulfill industry.
Protein engineering techniques provide a complementary tool for
obtaining enzymes with enhanced thermostability or alkalinity
[9–11].In silico screening of protein sequences from databases,
often derived from genomic and metagenomic analysis, has been
shown to be a powerful methodology that has been successfully
applied to identify extremophilic xylanases in GH11 family [12]
and for discovery of hidden properties of carbohydrate active
enzymes [13–15].
In this communication we describe an in silico analysis of GH10
family, searching for xylanases with extreme thermophilic and
alkalophilic properties, and the functional and structural charac-
terization of the enzyme with best characteristics. Additionally,
we show that the modification of the protein structure by the addi-
tion of carbohydrate binding domains improves the enzyme per-
formance, which is relevant for biotechnological applications.
2. Materials and methods
2.1. Phylogenetic analysis of GH10 sequences
Accession numbers of GH10 protein sequences were obtained
from the CAZy database [5]. The amino acid sequences were
obtained from the NCBI database, using the Batch Entrez Tool
(https://www.ncbi.nlm.nih.gov/sites/batchentrez). Pfam [16] pro-
vided protein domain composition and coordinates for each
sequence.
The linear composition of domains of a given sequence in N-
terminal to C-terminal order, defined what has been called domain
architecture (DA). The assignment of specific DA from the
sequences of the GH10 family was performed following the
methodology previously described for the GH2 family [13]. Only
sequences containing a catalytic domain Glyco_hydro_10 (GH10)
matching at least 80% with the Pfam consensus were included in
this study.
Sequence alignment of GH10 catalytic domains was performed
with CLC sequence viewer (Quiagen), using ClustalO MSA algo-
rithm [17]. Trees were built using W-IQ-TREE Maximum Likehood
algorithm with JTT matrix [15] and a bootstrap of 1000 replicates.
Results were analyzed on Dendroscope Software [18] and repre-
sented using FigTree (http://tree.bio.ed.ac.uk/software/figtree/)
and MESQUITE software (https://www.mesquiteproject.org).
2.2. Molecular biology
DNA sequences and amino acid selected for experimental anal-
ysis were edited before cloning. Signal peptide sequences were
detected using the Phobius Tool [19] and removed. The coding
sequences were optimized for E. coli expression by using the Inte-
grated DNA Technologies (IDT) Codon Optimization Tool (www.
idtdna.com). Native restriction sites were eliminated and SacI
and SalI restriction sites were added in 5
0
and 3
0
respectively, to
facilitate the cloning in vector pQE80L (Quiagen).
Synthetic genes (codon-optimized) of sequences encoding five
selected putative xylanases (Xyn10-14) were purchased from IDT
(Supplementary Table S3). The DNA fragments (except Xyn10
and Xyn14) were digested with endonucleases SacI and SalI and
cloned into pQE80L plasmid cut with the same enzymes. Xyn10
and Xyn14 could not be synthesized as a single piece and were
therefore assembled by joining two fragments (F1 and F2).
Xyn10-F1 and Xyn14-F1 were cut with SacI and EcoRI, whereas
Xyn10-F2 and Xyn14-F2 were cut with EcoRI and SalI. The
corresponding fragments were then cloned as one piece in pQE80L.
Fast Digest enzymes and T4 ligase were purchased from
ThermoScientific.
The xylanase encoding plasmids were used for transforming
E. coli XL1 Blue. Selection of E. coli clones expressing xylanase genes
was performed as described in (Talens-Perales et al., 2020). The
cloned sequences were confirmed by sequencing. Once checked,
the plasmids were transferred to E. coli Rosetta (Stratagene) for
protein production. Hybrid enzymes Xyn11-CBM2 and Xyn11-
CBM9 were constructed from plasmids Xyn5-CBM2 PQE80L and
Xyn5-CBM9, respectively [12], replacing the Xyn5 gene by the
Xyn11 gene.
2.3. Protein purification
Cell crude extracts were obtained from E. coli cultures grown at
37 °C until reaching OD
600
of 0.6 and then were induced with 1 mM
IPTG, either at 37 °C for 5 h or 16 °C overnight. Buffer A (20 mM
phosphate buffer, pH 7.4, 10 mM imidazole, 500 mM NaCl) was
used to disrupt the cells by sonication. Protein extracts were
obtained by centrifugation at 12,000 g during 25 min. Protein
purification was performed using nickel affinity chromatography
with 1 mL HisTrap FF crude column (GE Healthcare) mounted in
AKTA-Purifier (GE). Buffer B (20 mM phosphate buffer, pH 7.4,
500 mM imidazole, 500 mM NaCl) was used for elution. Eluted
fractions with xylanase activity were dialyzed against buffer C
(20 mM Tris-HCl, pH 7.0, 50 mM NaCl). Protein concentration in
the eluted fractions was determined in a NanoDrop spectropho-
tometer (Thermo Fisher). The purity of the protein recovered was
analyzed by SDS-PAGE, using Blue Safe staining (Nzytech). An
image of the gel was taken with a Proxima AQ-4 gel documentation
system (Isogen) and the amount of protein in the gel bands was
quantified using FIJI software [20].
2.4. Evaluation of xylanase activity at different conditions of
temperature and pH
Xylanase activity of purified enzymes was assayed at range of
temperature and pH. Enzyme reactions for temperature assays
were prepared by mixing 180
l
L of substrate (1% oat spelt xylan,
in Tris-HCl 50 mM buffer pH 9.0) and 20
l
L of purified protein
(protein concentrations were adjusted to assure linear enzyme
response) and then incubated at 60 °C, 70 °C, 80 °Cor90°C, for
10 min. The reaction was stopped by putting the tubes on ice.
Activity at different pH was determined using 50 mM buffered
solutions, at the following pH values: 5.0 (acetate), 6.0 and 7.0
(phosphate), 8.0, 9.0 and 10.0 (Tris-HCl). The enzyme reactions
were prepared as above described, using the appropriate buffer.
The reactions were incubated at 65 °C during 10 min and then
stopped on ice. Control reactions, without enzyme, were carried
out for each assay condition.
Measurement of reducing sugars was carried out by adding
100
l
L of DNS reactive to the reaction tubes that were then boiled
for 10 min. Next, 900
l
L of miliQ H
2
O were added and the tubes
were centrifuged. Measurements were done at OD
540
in 96-well
plates by transferring 300
l
L of the supernatant using PowerWave
HT equipment, from BioTek Instruments (Winooski, VT, USA).
Cleavage pattern of xylooligosaccharides was analyzed by using
500
l
L of 2 mM of the corresponding substrate, from 2 to 6 units
(Megazyme), dissolved in 50 mM Tris-HCl buffer pH 9.0. To each
tube, 20
l
L of enzyme solution were added. The amount of enzyme
used was estimated to obtain 1
l
mol of reducing sugars min
1
mg
1
of enzyme. The reactions were carried for 5 h at 90 °C and
stopped on ice. Analysis of reaction products was carried out by
ion exchange chromatography using a Dionex (Thermo Fisher Sci-
D. Talens-Perales, E. Jiménez-Ortega, P. Sánchez-Torres et al. Computational and Structural Biotechnology Journal 19 (2021) 2676–2686
2677
entific) with CarbonPac PA100 column and a pulsed amperometric
detector.
2.5. Protein crystallization, data collection and structure
determination
Firstly, different crystallization screens were explored, using the
sitting-drop vapor-diffusion method. A wide variety of crystals
grew in several commercial kits such as Index (Hamptom
Research), JBScreen JCSG++ (Jena Bioscience), JBScreen Classic (Jena
Bioscience). Different manuals grids were used in order to optimize
the quality of the crystals. Finally, the best crystals were obtained
with 10% PEG 3350, 0.2 M Proline, 0.1 M Hepes pH 7.5. Each drop
presented the same proportion of protein (6.47 mg/mL) and reser-
voir solution (250 nl). Mother liquor solution was supplemented
with 25% of glycerol to cryoprotect the crystal during the data col-
lection. Diffraction data were collected at the ALBA synchrotron
station of Barcelona (Spain). The X-ray images were processed with
XDS [21], and merged using Aimless from CCP4 suite package [22].
P2
1
2
1
2
1
space group was obtained with cell parameters 91.45
95.27 100.53 Å
3
and two molecules in the asymmetric unit.
The structure was solved by molecular replacement using MOL-
REP from CCP4 [23]. The xylanase CbXyn10C from Caldicellulosirup-
tor bescii [24], with the Protein Data Bank code 5OFJ (56.5%
sequence homology) was selected as a template. Restraint refine-
ment was carried out using REFMAC5 from CCP4 with local non-
crystallographic symmetry (NCS) [25]. This was combined with
several rounds of model building in COOT [26]. Figures were pre-
pared with PyMOL [27]. Final crystallography results are given in
Supplementary Table 4.
2.6. Analysis of protein stability
Different web servers were employed to analyze the structure
and thermostability of the protein. Protein Blast [28] and END-
script server [29] were used to compare protein sequence and
the tridimensional structures were inspected in Dali server [30].
ProtParam tool (ExPASy) [31] was used to compute amino acid
composition. Intraprotein hydrophobic and ionic interactions,
hydrogen bonds, aromatic and cation-
p
interactions were deter-
mined by PIC server (Protein Interactions Calculator) [32]. The
accessible surfaced area (ASA) and the exposed charged accessible
area were both calculated with VADAR server [33]. The formula for
compactness was: ASA/number of total residues, and for % exposed
charged accessible surface area was: exposed charged accessible
area /ASA *100.
3. Results and discussion
3.1. In silico screening of GH10 sequences for thermophilic alkaliphilic
xylanases
The nearly 5000 entries currently available in the CAZy data-
base were expurgated discarding repetitions, incomplete
sequences and those having less than 20% matching with the Pfam
Glyco_Hydro_10 domain (GH10). This yielded a total of 2309
sequences that were further processed (Supplementary material,
Table S1). About 80% of these sequences (1848) corresponded to
bacteria whereas the remaining 20% (461) were from eukaryotes.
No archaeal sequences matched the established requirements.
GH10 protein sequences showed a great variety of domain archi-
tectures. Up to 197 different domain architectures (DA) could be
distinguished (Supplementary material,Table S2), in contrast with
the lesser diversity (69 DA) described for functionally related GH11
family [12]. About one half of the bacterial sequences and one
tenth of the eukaryotic, showed the simplest possible DA, consisted
of just the catalytic GH10 domain. Regarding more complex DA,
the presence of CBM4-9, which binds amorphous cellulose and sol-
uble oligosaccharides [34,35], was ubiquitous and detected in 14%
of the sequences, in some cases in tandem repeats at N-terminal
position. CBM2 which binds cellulose and chitin [36], or Ricin-B-
lectin domains that mediate sugar recognition [37], were quite
abundant, present in more than 5% of the sequences. A small frac-
tion of the sequences (1–2%) contained C-terminal extensions not
corresponding with identified motifs. These tails were labeled by
their length, as Ct1 (50–150), Ct2 (150–200), Ct3 (200–300), Ct4
(300–400) and Ct5 (400–500) (Supplementary material,Table S2).
Fig. 1A shows a schematic representation of the cladogram
resulting from the phylogenetic analysis of the Glyco_Hydro_10
domain of the 2309 selected sequences. A detailed version of the
cladogram is presented as supplementary material (Fig. S1). In
Fig. 1 and S1 and in Table S2, different DA with common domains
are represented by the same color code. In the cladogram, eukary-
otic sequences appear in three groups, generated by different evo-
lutionary events, at a late stage in the evolutionary process. The
simplest DA, a single GH10 domain, appears in some instances
mingled with composed DA, but most sequences with additional,
non-catalytic domains, emerge from separate nodes. Sequences
putatively identified as thermoresistant, are marked by a gray label
in Fig. 1A. Among these, a cluster (hereinafter designated as TAK)
marked by a triangle, corresponds to sequences that putatively
are thermostable and alkaliphilic. The sequences included in the
TAK cluster, belong to genera the Caldicellulosiruptor, Thermotoga,
Pseudothermotoga, Petrotoga and Defluviitoga, and are listed in
Fig. 1B. Most of these sequences contain a multidomain DA, with
CBM4-9 present in ca. 70% of them. Another domains present are
CBM9, CBM3 and the catalytic domain GH48 with cellulase activity
[38]. CBM9 and CBM3 have been described to have xylan-binding
activity [38,39]. Five of these sequences, representing different
DA, were selected for experimental analysis: Xyn10 from Dyc-
tioglomus turgidum, Xyn11 from Pseudothermotoga thermarum,
Xyn12 and Xyn13 from Caldicellulosiruptor bescii and Caldicellu-
losiruptor owensensis, respectively and Xyn14 from Thermotoga sp.
3.2. Production, purification and functional characterization of
selected enzymes
Synthetic, codon-optimized gene sequences encoding the
selected putative xylanases were cloned and expressed in E. coli.
Synthetic DNA coding sequences could be obtained as a single frag-
ment, except for Xyn10 and Xyn14, which had to be reconstructed
from two fragments (Xyn10-F1, Xyn14-F1 and Xyn10-F2, Xyn14-
F2). Proteins were purified from bacterial crude cell extracts by
thermal treatment and nickel affinity chromatography. Xyn12
was discarded at this point because no substantial amount of pro-
tein could be recovered from the recombinant E. coli cultures.
Semiquantitative evaluation of Xyn14 showed very low level of
xylanase activity (results not shown) and therefore was also dis-
carded. The other three enzymes were purified and analyzed by
SDS-PAGE (Fig. 2). In all three cases, electrophoretical mobility
was in accordance with the expected molecular mass of the
polypeptides, predicted by ProtParam [40].
Oat spelt xylan hydrolysis by Xyn10, Xyn11, and Xyn13 was
measured at different pH and temperature. Assays at different val-
ues of pH were carried out at 90 °C(Fig. 3A). Xyn10 and Xyn11
showed the expected profile for alkaliphilic enzymes, with optimal
activity at pH between 8.0 and 9.0. Xyn13 showed maximal activ-
ity at pH between 6.0 and 7.0, but retained high activity at pH 9.0.
In buffered solution at pH 9.0, the three xylanases showed a ther-
mophilic profile, with optimal temperature 70 °C and 90 °C. The
highest activity measured corresponded to Xyn11 at 90 °C
D. Talens-Perales, E. Jiménez-Ortega, P. Sánchez-Torres et al. Computational and Structural Biotechnology Journal 19 (2021) 2676–2686
2678
(Fig. 3B). These results are remarkably different from a report by
Shi et al., [41], describing the properties of a xylanase encoded
by the Xyn10B gene of T. thermarum (now Pseudothermotoga ther-
marum), same sequence as Xyn11, but optimal activity at 80 °C,
and weak activity at pH > 8.0. A different xylanase from the
GH10 family from P. thermarum has also been described. This
one displays a complex DA, which three tandem CBM4-9 at N-
terminal position and two CBM9 at C-terminal position. Despite
of the differences in DA, the catalytic domain of this enzyme,
named Xyn10A, is very similar to Xyn11 being located very close
in the phylogenetic tree (Figs. 1 and S1). Xyn10A, was reported
to be optimally active at 95 °C and pH 7.0 [8]. Other thermostable
xylanases (from GH10 and GH11 families) have been characterized
but their optimal temperatures are around 70 °C and their opti-
mum pH ranged from 6.5 to 9.0. Only Bacillus halodurans TSEV
xylanase combines thermostability and alkaliphility, with optimal
values of pH of 9.0 and temperature 80 °C[42], but still below the
score of Xyn11. Xylanases from Geobacillus, Caldicellusiruptor [43]
and Thermotoga, showed high optimum temperatures. Remarkably,
Thermotoga naphthophila RKU-10 and Thermotoga petrophila RKU-1
(95 °C) or Thermotoga maritima MSB8, (90 °C), but optimum pH val-
ues for these enzymes were between 5.0 and 6.0 [44–46]. Xyn11
represented an extraordinary performance under simultaneous
conditions of high temperature and alkalinity, showing at pH
Fig. 1. (A) Cladogram of the GH10 family. The gray triangle marks the TAK cluster. (B) Phylogenetic tree of the TAK cluster sequences. Asterisks mark the sequences that were
selected for experimental analysis.
Fig. 2. SDS-PAGE analysis and domain architecture of selected enzymes. Protein ladder mass in kDa.
D. Talens-Perales, E. Jiménez-Ortega, P. Sánchez-Torres et al. Computational and Structural Biotechnology Journal 19 (2021) 2676–2686
2679
10.5 and 90 °C, an activity value of 600 mmol min
1
mg
1
of
enzyme.
A previous study carried out with GH11 xylanases showed that
in some instances, the addition of a Carbohydrate Binding Domain
(CBM2, CBM9) to the protein structure has a positive effect on
enzyme performance [12]. Therefore, we undertook the construc-
tion of hybrid enzymes derived from Xyn11, by fusing this enzyme
with CBM2 and CBM9. The SDS-PAGE profile and DA of the hybrids
are shown in Fig. 4A. Addition of CBM9 caused a decrease of Xyn11
activity at different pH (Fig. 4B). This result was unexpected as it is
the opposite to what was observed by the addition of CBM9 to a
GH11 xylanase [12]. However, the addition of CBM2 improved
the activity of the enzyme at high pH. Hybrid Xyn11-CBM2 showed
an activity two times higher than Xyn11 at 90 °C and pH 10.5
(Fig. 4B).
The cleavage pattern of xylanases was analyzed using as sub-
strate oligoxylosides from 2 to 6 units. Enzyme reactions were car-
ried out for 5 h at pH 9.0 and 90 °C. The products released from the
reactions were analyzed chromatographically. As expected, the
enzymes showed characteristic behavior of endoxylanases, but
with some interesting differences. Xyn10 and Xyn13 were unable
to hydrolyze xylobiose and xylotriose. Xyn10 hydrolyzed xylote-
traose, xylopentaose and xylohexaose, yielding mainly xylobiose
and xylotriose (Fig. 5A). Xyn13 was inactive towards xylotetraose,
but excised xylopentaose and xylohexose, yielding mainly xylo-
biose and xylotriose (Fig. 5C). Xyn11, as well as its hybrids
Xyn11-CBM9 and Xyn11-CBM2, were unable to cut xylobiose,
but were active against xylotriose, xylotetraose, xylopentaose
and xylohexaose yielding mainly xylose and xylobiose (Fig. 5B).
These results agree with previous studies on GH10 xylanases
[42]. The observed differences in the cleavage pattern of Xyn10,
Xyn11 and Xyn13 is relevant from a biotechnological point of view,
as it can be used to produce short-chain oligoxylosides from xylan.
These oligosaccharides are compounds with recognized value as
prebiotics, whose administration confers a proven heath benefit
[47,48].
3.3. Crystallographic structure of Xyn11
The structure of Xyn11 was solved at 1.8 Å resolution by X-ray
crystallography. The protein folds into an (b/
a
)
8
barrel (TIM-barrel)
architecture (Fig. 6A), typical of GH10 xylanases. From the loops
linking the C-terminal of each b-strand to the succeeding
a
-helix,
the long L4, L7 and L8 loops protrude markedly and define an
extended groove to accommodate the substrate. An additional b-
hairpin motif is present in the loop L2, linking b2to
a
2.
According to its primary structure, the closest homologues of
Xyn11 are thermoresistant multidomain proteins from Caldicellu-
losiruptor species (N-terminal domain of CbXyn10C-Cel48B [24],
PDB code 5OFJ, and GH10 domain of WP_045175321 [49], PDB
code 6D5C, showing 55% sequence identity, both presenting the
Fig. 3. Activity of selected enzymes at different values of pH (A) and temperatures
(B). Activity values correspond to
l
mol of reducing sugars min
1
mg
-1
of enzyme.
Fig. 4. (A) Physical characterization (DA and SDS-PAGE analysis) of hybrid xylanases. (B) Activity of hybrid xylanases assayed at 90 °C and different pH. Displayed values
correspond to activity expressed as
l
mol of reducing sugars min
1
mmol
1
of enzyme. (C) Activity of hybrid xylanases at pH 9.0 and different temperatures.
D. Talens-Perales, E. Jiménez-Ortega, P. Sánchez-Torres et al. Computational and Structural Biotechnology Journal 19 (2021) 2676–2686
2680
same b-hairpin observed in Xyn11. Next homologues are two
intracellular xylanases, IXT6 from Geobacillus stearothermophilus
[50], PDB code 1 N82, with 50% identity, and the mesophilic
Xyn10B from Paenibacillus barcinonensis [51], PDB code 3EMC,
showing 44% identity. These enzymes show very similar topology,
with equivalent L4, L7 and L8 loops both in length and position.
Despite fairly conserved topology, loops L7 and L8 are highly
variable in sequence (Fig. 7). Therefore, these loops shape very dif-
ferent active site crevices. To fully depict the Xyn11 active site, the
binding mode of an heptaxylose has been inferred from structural
superimposition of its complex from Caldicellulosiruptor bescii
Xyn10C [43], PDB code 5OFK, onto the Xyn11 coordinates
(Fig. 6B–D). As it is observed in Fig. 6, Xyn11 seems able to accom-
modate at least seven units of xylose spanning its catalytic tunnel.
The most peculiar feature of Xyn11 is the presence at the non-
reducing end of three consecutive Phe residues (Phe312, Phe313
and Phe314) at L8 (Fig. 7) that define a narrow entrance to the cat-
alytic channel. Only the central Phe313 is conserved in homo-
logues, which have Ser/Thr and Pro at the other two positions.
This feature places Phe313 in a more prominent position within
a narrower cavity and suggests a potential subsite 3, by
hydrophobic interaction to Phe312, not seen in the Xyn11 ana-
logues. Apart from this hydrophobic wall, subsite 2 is well-
characterized by the presence of a great amount of hydrogen bonds
at the opposite wall with a number of very conserved residues.
Thus, and according to the complex model, O2 for the xylose would
interact with Trp301 and Glu50, O3 with Glu50 and Asn51, O4
would be stabilized by Trp301 and Lys54, which also form a hydro-
gen bond with O5. Furthermore, the xylose bound at subsite 1
would also present polar interactions to conserved residues at
the same wall of the active site, i.e. to Lys54 (O3), His91 (O2 and
O3), and Asn143 and Gln220, (O2).
The xylose bound at subsite + 1 is mainly fixed by stacking to
Tyr189 and hydrophobic interaction to Trp309 and Phe313, all con-
served residues, and lacking direct polar interactions as in other
xylanases. However, in Xyn11, the presence of the pair Phe313-
Phe314 narrow the cavity at this subsite + 1 and, more important,
fixes the conformation of the Arg260 side-chain in a very
restrained position pointing to the inner part of the cavity, in
which it could be hydrogen linked to O2 and O3 of xylose bound
at subsite + 2. As a consequence, Xyn11 does not present at this
subsite + 2 the pocket observed in its homologues that seems able
to allocate a putative xylan decoration [51]. Furthermore, Ser259, a
position in which most analogues present an aromatic residue with
a protruding side-chain that makes a more restricted subsite + 3,
precedes Arg260, unique to Xyn11 at loop L7 (Fig. 7). Thus, the
reducing-end of the Xyn11 tunnel exhibit rather open subsites
from + 3 that are markedly different to that observed in its homo-
logues. Only the bifurcated polar link from Asn190 to the xyloses
bound at subsite + 2 (O5) and + 3 (O3) is conserved.
A last interesting feature of the channel is the presence of a
small pocket created by a cluster of three aromatic residues located
at the end of the tunnel, Phe192, Phe193 and Trp231, only Phe193
being conserved (Fig. 6D).Interestingly, a glycerol molecule from
Fig. 5. Cromatographic profile of reaction products obtained from the action of xylanases on xylooligosaccharides of different degree of polymerization.
D. Talens-Perales, E. Jiménez-Ortega, P. Sánchez-Torres et al. Computational and Structural Biotechnology Journal 19 (2021) 2676–2686
2681
the cryoprotectant has been trapped within this pocket in our crys-
tals, suggesting a putative role in accommodating a potential O3
substituent at the succeeding xylose unit, which would represent
a putative subsite + 5. As it is shown in Fig. 6D, an additional glyc-
erol molecule has been trapped in a groove close to the free O2 and
O3 hydroxyls of xylose at subsite + 3. The particular shape of both
cavities might reflect a specific pattern of xylan decorations that
would be recognized and bound at subsite + 3 and + 5.
In summary, Xyn11, as its closest thermoresistant homologues, is
able to degrade xylan, and presents 6–7 subsites (from 2 to + 4/+5)
for binding this polysaccharide. However, inspection of its active site
crevice reveals that substitutions are not allowed at 2 and 1, nor
at + 2 and + 4. Therefore, two consecutive unsubstituted xylose units
are required to bind at positions 2, 1, for hydrolysis to occur, and a
subsequent sequence of alternate substituted/unsubstituted
xyloses is envisaged from subsite + 1. This result is in accordance
with the cleavage pattern observed in Fig. 5. The especial shape of
the cavities described above may match the chemical structure of
its natural substrate.
3.4. Structural basis of Xyn11 extremophilic properties
A great effort has been directed to disclose the molecular basis
of protein adaptability to extreme conditions [52–55]. The general
agreement is that this ability results from the cumulative effect of
multiple stabilizing factors that may be present to a different
degree in each extremophilic protein. Nevertheless, despite the
ample variation observed, some trends can be observed between
the sequences and structures of extremophilic vs their mesophilic
orthologues.
In order to explain the extremophilic properties of Xyn11, its
primary and tertiary structure was compared to other family
GH10 xylanases, including thermoresistant CbXyn10C (PDB code
5OFJ) [49] and three mesophiles: PbXynB from Paenibacillus barci-
nonensis; (3EMC) [51], CmXyn10B from Cellvibrio mixtus; (2CNC)
[53] and SoXyn10A from Streptomyces olivaceoviridis (1V6Y) [56].
Several features analyzed in Xyn11 and its homologues are sum-
marized in Tables 1 and 2.
As a general rule, extremophilic proteins contain low number of
residues that decompose easily, such as hydroxy amino acids: Ser
and Thr [57], amino acids with an amide terminal group: Asn,
Gln [58] or sulfur containing amino acids: Cys, Met [57]. Accord-
ingly, Xyn11 complies this rule (Table 1). On the other hand, the
relative abundance of Pro observed in Xyn11, compared with its
homologues, can been related to increased stabilization provided
by a higher chain rigidity that reduces fluctuation of the secondary
structure and helps to keep the proper folding at high temperature
[59]. Finally, an increased number of charged residues, meaning an
increment in the ion pair interactions, derive in thermostability
[58,60]. Thus, Xyn11 presents the highest value of Lys, Arg and
Glu compared with the homologues.
Considering the tertiary structure (Table 2), hydrophobic interac-
tions have been reported to play a key role in protein folding [53].
Thus, the buried hydrophobic interactions in the core of the protein
increase the van der Waals contacts, reducing the exposure of
hydrophobicresiduestosolventand promoting rigidity [53]. In addi-
tion, aromatic clusters have also been commonly considered to con-
tribute to the extreme stability of the proteins, resulting in internal
stacking interactions that enhance the stability of the protein [61].
In this respect, Xyn11 presents the highest number of both,
hydrophobic and aromatic interactions. Whereas Xyn11 compact-
ness is similar to its mesophilic analogues, its accessible surface area
(ASA) is higher. Hydrogen bonding is correlated with an increase in
Fig. 6. Crystal structure of Xyn11. (A) Overall folding as an (b/
a
)
8
barrel
(TIM-barrel) architecture, represented in green (helices) and violet (strands). The
most relevant loops that participate in the substrate binding and the catalytic
residues are colored in orange. (B) The active site tunnel with a heptaxylose
molecule, modelled from structural superimposition with its complex from
Caldicellulosiruptor bescii Xyn10C (PDB code 5OFK). (C) Proposed atomic interac-
tions of heptaxylose at Xyn11 active site. The catalytic residues are colored in
orange, while the non-conserved residues are colored in yellow. Polar interac-
tions are marked in dashed lines. (D) Detail of the reducing-end moiety of the
active site, showing two glycerol molecules (blue) trapped in the crystals, and
the three aromatic residues shaping a pocket in the tunnel. (For interpretation of
the references to color in this figure legend, the reader is referred to the web
version of this article.)
D. Talens-Perales, E. Jiménez-Ortega, P. Sánchez-Torres et al. Computational and Structural Biotechnology Journal 19 (2021) 2676–2686
2682
rigidity in the core of the protein [62] and is considered one of the
major contributors to extreme-stability, especially when consider-
ing the main chain-main chain interactions [53,63]. Theoretical
studies show the stabilizing effect of electrostatic interactions and
salt-bridges at high temperature [64]. Again, Xyn11 presents a high
number of hydrogen bonds and ionic interactions.
It is worth noting that, even when Xyn11 is first-ranked in only
5 out of 11 stabilizing parameters that we have analyzed (Tables 1
and 2), it is one of the two best proteins in all cases, revealing that
its molecular structure is well suited to stand denaturing condi-
tions. All the atomic interactions considered in this work have
been previously identified as molecular mechanism underlying
resistance in both, thermophilic and also alkaliphilic proteins
[53], which explains not only Xyn11 thermal resistance but also
its activity at high pH. Furthermore, the increased number of
atomic interactions observed in Xyn11 must derive in protein
rigidity, which is the key to high-temperature adaptation by pre-
venting unfolding. In this context, Xyn11 presents a markedly
low chain flexibility, as deduced from the analysis of the atomic
B factors values along the polypeptide chain that is shown in
Fig. 8A.
Lastly, an interesting feature in Xyn11 structure is the long loop
L2 (Fig. 6A), connecting b2to
a
2, also present in close homologues
from Caldicellulosiruptor species (xylanases from C. bescii, PDB code
5OFK and C. danielli, PDB code 6D5C). This loop is making a
b-hairpin which is protruding from the (
a
/b)
8
barrel that,
Fig. 7. Sequence variability of loops L7 and L8. Sequences are of xylanases CbXynb10C from C. bescii (5OFJ), GH10 module from C. danielli (6D5C), IXT6 from G.
stearothermophilus (1N82), and Xyn10B from P. barcinonensis (3EMC). (ESPript - http://espript.ibcp.fr).
Table 2
Analysis of the atomic interactions and packing. Number of hydrophobic and aromatic interactions, main chain-main chain hydrogen bonds and ionic interactions in Xyn11,
CbXyn10C from Caldicellulosiruptor bescii (5OFJ), XynB from Paenibacillus barcinonensi (3EMC), CmXyn10B from Cellvibrio mixtus (2CNC) and SoXyn10A from Streptomyces
olivaceoviridis (1V6Y). Compactness and percentage of exposed residues in the accessible surface area are also computed and shown. In each column, the two first-ranked proteins
are shown in bold type, underlined number indicate the most favorable value.
Atomic Interactions and Packing
HYDROPHOBIC AROMATIC HYDROGEN BONDS (mch-mch) IONIC COMPACTNESS EXPOSED CHARGED ASA (%)
Xyn11 381 32 449 47 42.4 21.8
CbXyn10C 361 21 497 39 40.6 19.1
PbXynB 323 24 428 38 42.7 25.1
CmXyn10B 347 27 419 56 43.1 21.4
SoXyn10A 274 18 439 34 37.9 16.2
Table 1
Analysis of amino acid composition. Percentage of Asn, Gln, Ser, Thr, Pro, Met, Cys, Glu, Arg and Lys in Xyn11, CbXyn10C from Caldicellulosiruptor bescii (5OFJ), XynB from
Paenibacillus barcinonensi (3EMC), CmXyn10B from Cellvibrio mixtus (2CNC) and SoXyn10A from Streptomyces olivaceoviridis (1V6Y). In each column, the two first-ranked proteins
are shown in bold type, underlined number indicate the most favorable value.
AA Composition (%)
Asn + Gln Ser + Thr Pro Met + Cys Glu + Arg + Lys
Xyn11 7.4 10.0 4.9 3.0 19.0
CbXyn10C 10.0 13.0 4.1 3.3 17.1
PbXynB 9.0 10.8 3.3 3.0 18.4
CmXyn10B 7.2 9.6 3.4 3.1 18.6
SoXyn10A 10.8 11.8 2.8 4 13.9
D. Talens-Perales, E. Jiménez-Ortega, P. Sánchez-Torres et al. Computational and Structural Biotechnology Journal 19 (2021) 2676–2686
2683
nevertheless is stabilized by two cation-
p
interactions keeping a
close packing of L2 (Fig. 8B). Thus, Arg61 is stacking to Trp102,
located at L3, while Arg131, at the end of helix
a
3, is stacking to
Tyr69. Only the last interaction is conserved in the other thermore-
sistant homologues. Cation-
p
interactions have been described as
another form of electrostatic interaction that make an important
contribution to protein stability, which appears to increase at
higher temperatures [65]. Thus, Xyn11 might have developed addi-
tional local molecular mechanisms that combined to the above
parameters make a very resistant protein.
4. Conclusions
The comprehensive bioinformatics screening of the GH10 fam-
ily reported in this study represents a powerful methodological
approach that allowed the identification and characterization of
xylanases active at extreme conditions of pH and temperature.
The enzyme with best performance, Xyn11, corresponds to a xyla-
nase from the bacterium Pseudothermotoga thermarum. The
enzyme shows an exceptional level of activity at 90
°
C and pH
10.5. Fusion of a carbohydrate binding module (CBM2) to Xyn11
further increased its activity at extreme conditions. In addition to
Xyn11, two other xylanases, Xyn10 and Xyn13 showed activity
high at extreme conditions. Interestingly, these enzymes differ in
their cleavage pattern on xylan derived substrates, yielding differ-
ent proportions of xylose, xylobiose and xylotriose as final prod-
ucts, which may have practical consequences from a
biotechnological point of view. The crystallographic resolution at
1.8 Å of Xyn11 structure provides an explanation of its function
at extreme conditions. Not surprisingly, qualitatively, the atomic
interactions responsible for Xyn11 resistance are the same known
to sustains protein stability, namely hydrogen bonds, ion pairs,
hydrophobic and aromatic interactions. However, the number
and distribution in which these interactions appear and its extre-
mophilic enzyme properties makes Xyn11 an outstanding case
for study.
5. Accesion code
PDB 7NL2
CRediT authorship contribution statement
David Talens-Perales: Formal analysis, Investigation. Elena
Jiménez-Ortega: Investigation. Paloma Sánchez-Torres: Investi-
gation. Julia Sanz-Aparicio: Supervision, Funding acquisition. Julio
Polaina: Supervision, Funding acquisition.
Declaration of Competing Interest
The authors declare that they have no known competing finan-
cial interests or personal relationships that could have appeared
to influence the work reported in this paper.
Acknowledgements
This work was funded as part of the European Project WOOD-
ZYMES, by the Bio Based Industries Joint Undertaking, under the
European Union’s Horizon 2020 research and innovation program
(Grant Agreement H2020-BBI-JU-792070) and by grants
BIO2016-76601-C3-3-R and PID2019-105838RB-C33 from the
Spanish Government.
Appendix A. Supplementary data
Supplementary data to this article can be found online at
https://doi.org/10.1016/j.csbj.2021.05.004.
References
[1] Kohli I, Joshi NC, Mohapatra S, Varma A. Extremophile - an adaptive strategy
for extreme conditions and applications. Curr Genomics 2020;21(2):96–110.
[2] Krüger A, Schäfers C, Schröder C, Antranikian G. Towards a sustainable
biobased industry - highlighting the impact of extremophiles. N Biotechnol
2018;40(Pt A):144–53.
[3] Walia A, Guleria S, Mehta P, Chauhan A, Parkash J. Microbial xylanases and
their industrial application in pulp and paper biobleaching: a review. 3 Biotech
2017;7(1):11.
[4] Kumar V, Marín-Navarro J, Shukla P. Thermostable microbial xylanases for
pulp and paper industries: trends, applications and further perspectives.
World J Microbiol Biotechnol 2016;32(2):34. https://doi.org/10.1007/s11274-
015-2005-0.
[5] Lombard V, Golaconda Ramulu H, Drula E, Coutinho PM, Henrissat B. The
Carbohydrate-active enzymes database (CAZy) in 2013. Nucleic Acids Res
2014;42(D1):D490–5.
[6] Verma D, Satyanarayana T. Cloning, expression and applicability of thermo-
alkali-stable xylanase of Geobacillus thermoleovorans in generating
xylooligosaccharides from agro-residues. Bioresour Technol 2012;107:333–8.
[7] Paës G, Berrin J-G, Beaugrand J. GH11 xylanases: structure/function/properties
relationships and applications. Biotechnol Adv 2012;30(3):564–92.
[8] Shi H, Zhang Yu, Li X, Huang Y, Wang L, Wang Ye, et al. A novel highly
thermostable xylanase stimulated by Ca2+ from Thermotoga thermarum:
cloning, expression and characterization. Biotechnol Biofuels 2013;6(1):26.
https://doi.org/10.1186/1754-6834-6-26.
[9] Paës G, O’Donohue MJ. Engineering increased thermostability in the
thermostable GH-11 xylanase from Thermobacillus xylanilyticus. J Biotechnol
2006;125(3):338–50.
[10] Li He, Kankaanpää A, Xiong H, Hummel M, Sixta H, Ojamo H, et al.
Thermostabilization of extremophilic Dictyoglomus thermophilum GH11
xylanase by an N-terminal disulfide bridge and the effect of ionic liquid
[emim]OAc on the enzymatic performance. Enzyme Microb Technol 2013;53
(6-7):414–9.
Fig. 8. Putative structural elements of Xyn11 thermoresistance. (A) Crystallo-
graphic atomic B factors in the polypeptide chain shown in rainbow code, from low
(blue) to high (red) values. (B) Cation-
p
interactions stabilizing the b-hairpin
present in loop L2. A relevant hydrogen bond is represented as dashed line. (For
interpretation of the references to color in this figure legend, the reader is referred
to the web version of this article.)
D. Talens-Perales, E. Jiménez-Ortega, P. Sánchez-Torres et al. Computational and Structural Biotechnology Journal 19 (2021) 2676–2686
2684
[11] Bai W, Cao Y, Liu J, Wang Q, Jia Z. Improvement of alkalophilicity of an alkaline
xylanase Xyn11A-LC from Bacillus sp. SN5 by random mutation and Glu135
saturation mutagenesis. BMC Biotech 2016;16(1).
[12] Talens-Perales D, Sánchez-Torres P, Marín-Navarro J, Polaina J. In silico
screening and experimental analysis of family GH11 xylanases for
applications under conditions of alkaline pH and high temperature.
Biotechnol Biofuels 2020;13(1):1–15.
[13] Talens-Perales D, Górska A, Huson DH, Polaina J, Marín-Navarro J, Mallela K.
Analysis of domain architecture and phylogenetics of family 2 glycoside
hydrolases (GH2). PLoS ONE 2016;11(12):e0168035. https://doi.org/10.1371/
journal.pone.016803510.1371/journal.pone.0168035.g00110.1371/journal.
pone.0168035.g00210.1371/journal.pone.0168035.g00310.1371/journal.
pone.0168035.g00410.1371/journal.pone.0168035.g00510.1371/journal.
pone.0168035.g00610.1371/journal.pone.0168035.g00710.1371/journal.
pone.0168035.s00110.1371/journal.pone.0168035.s00210.1371/journal.
pone.0168035.s00310.1371/journal.pone.0168035.s00410.1371/journal.
pone.0168035.s00510.1371/journal.pone.0168035.s00610.1371/journal.
pone.0168035.s00710.1371/journal.pone.0168035.s008.
[14] Jones DR, Thomas D, Alger N, Ghavidel A, Inglis GD, Abbott DW. SACCHARIS: an
automated pipeline to streamline discovery of carbohydrate active enzyme
activities within polyspecific families and de novo sequence datasets.
Biotechnol Biofuels 2018;11(1):1–15.
[15] Nguyen LT, Schmidt HA, von Haeseler A, Minh BQ. IQ-TREE: a fast and effective
stochastic algorithm for estimating maximum-likelihood phylogenies. Mol
Biol Evol 2015;32:268–274.
[16] El-Gebali S, Mistry J, Bateman A, Eddy SR, Luciani A, Potter SC, Qureshi M,
Richardson LJ, Salazar GA, Smart A, Sonnhammer E, Hirsh L, Paladin L, Piovesan
D, Tosatto S, Finn RD. The Pfam protein families database in 2019. Nucleic
Acids Res 2019;47(D1):D427–D432.
[17] Sievers F, Higgins DG. Clustal Omega for making accurate alignments of many
protein sequences. Protein Sci 2018;27(1):135–45.
[18] Huson DH, Scornavacca C. Dendroscope 3: an interactive tool for rooted
phylogenetic trees and networks. Syst Biol 2012;61:1061–1067.
[19] Käll L, Krogh A, Sonnhammer EL. Advantages of combined transmembrane
topology and signal peptide prediction–the Phobius web server, Nucleic Acids
Res 2007;35(Web Server issue):W429–W432.
[20] Schindelin J, Arganda-Carreras I, Frise E, Kaynig V, Longair M, Pietzsch T, et al.
Fiji: an open-source platform for biological-image analysis. Nat Methods
2012;9(7):676–82.
[21] Kabsch W. Xds. Acta Crystallogr Sect D Biol Crystallogr 2010;66(Pt 2):125–
132.
[22] Evans PR. An introduction to data reduction: Space-group determination,
scaling and intensity statistics. Acta Crystallogr D Biol Crystallogr 2011;67
(4):282–92.
[23] Lebedev AA, Vagin AA, Murshudov GN. Model preparation in MOLREP and
examples of model improvement using X-ray data. Acta Crystallogr D Biol
Crystallogr 2007;64(1):33–9.
[24] Chu Y, Tu T, Penttinen L, Xue X, Wang X, Yi Z, et al. Insights into the roles of
non-catalytic residues in the active site of a GH10 xylanase with activity on
cellulose. J Biol Chem 2017;292(47):19315–27.
[25] Murshudov GN, Vagin AA, Dodson EJ. Refinement of macromolecular
structures by the maximum-likelihood method. Acta Crystallogr D Biol
Crystallogr 1997;53(3):240–55.
[26] Emsley P, Cowtan K. Coot: Model-building tools for molecular graphics. Acta
Crystallogr D Biol Crystallogr 2004;60(12 I):2126–32.
[27] DeLano WL. Pymol: an open-source molecular graphics tool. CCP4 Newslett
Protein Crystallogr 2002;40(1):82–92.
[28] Madden T. The BLAST sequence analysis tool. The NCBI Handbook
[Internet]. National Center for Biotechnology Information (US); 2013.
[29] Robert X, Gouet P. Deciphering key features in protein structures with the new
ENDscript server. Nucleic Acids Res 2014;42(W1):W320–4.
[30] Holm L, Laakso LM. Dali server update. Nucleic Acids Res 2016;44(W1):
W351–5.
[31] Wilkins MR, Gasteiger E, Bairoch A, Sanchez JC, Williams KL, Appel RD, et al.
Protein identification and analysis tools in the ExPASy server. Methods Mol
Biol (Clifton N.J.) 1999;112:531–52.
[32] Tina KG, Bhadra R, Srinivasan N. PIC: protein interactions calculator. Nucleic
Acids Res 2007;35(SUPPL.2):473–6.
[33] Willard L, Ranjan A, Zhang H, Monzavi H, Boyko RF, Sykes BD, et al. VADAR: a
web server for quantitative evaluation of protein structure quality. Nucleic
Acids Res 2003;31(13):3316–9.
[34] Johnson PE, Joshi MD, Tomme P, Kilburn DG, McIntosh LP. Structure of the N-
terminal cellulose-binding domain of Cellulomonas fimi CenC determined by
nuclear magnetic resonance spectroscopy. Biochemistry 1996;35
(45):14381–94.
[35] Nakamura T, Mine S, Hagihara Y, Ishikawa K, Ikegami T, Uegaki K. Tertiary
structure and carbohydrate recognition by the chitin-binding domain of a
hyperthermophilic chitinase from Pyrococcus furiosus. J Mol Biol 2008;381
(3):670–80.
[36] Hirabayashi J, Arai R. Lectin engineering: the possible and the actual. Interface
Focus 2019;9(2):20180068. https://doi.org/10.1098/rsfs.2018.0068.
[37] Chu Y, Hao Z, Wang K, Tu T, Huang H, Wang Y, et al. The GH10 and GH48 dual-
functional catalytic domains from a multimodular glycoside hydrolase
synergize in hydrolyzing both cellulose and xylan. Biotechnol Biofuels
2019;12(1). https://doi.org/10.1186/s13068-019-1617-2.
[38] Notenboom V, Boraston AB, Kilburn DG, Rose DR. Crystal structures of the
family 9 carbohydrate-binding module from Thermotoga maritima xylanase
10A in native and ligand-bound forms. Biochemistry 2001;40(21):6248–56.
[39] Sakka M, Higashi Y, Kimura T, Ratanakhanokchai K, Sakka K. Characterization
of Paenibacillus curdlanolyticus B-6 Xyn10D, a xylanase that contains a family 3
carbohydrate-binding module. Appl Environ Microbiol 2011;77(12):4260–3.
[40] Gasteiger E, Hoogland C, Gattiker A, Duvaud S, Wilkins MR, Appel RD, Bairoch
A. Protein identification and analysis tools on the ExPASy server. In: John M.
Walker (Ed.), The Proteomics Protocols Handbook, Humana Press; 2005, p.
571–607.
[41] Shi H, Zhang Yu, Zhong H, Huang Y, Li X, Wang F. Cloning, over-expression and
characterization of a thermo-tolerant xylanase from Thermotoga thermarum.
Biotechnol Lett 2014;36(3):587–93.
[42] Wang K, Cao R, Wang M, Lin Q, Zhan R, Xu H, et al. A novel thermostable GH10
xylanase with activities on a wide variety of cellulosic substrates from a
xylanolytic Bacillus strain exhibiting significant synergy with commercial
Celluclast 1.5 L in pretreated corn stover hydrolysis. Biotechnol Biofuels
2019;12(1):1–13.
[43] An J, Xie Y, Zhang Y, Tian D, Wang S, Yang G, et al. Characterization of a
thermostable, specific GH10 xylanase from Caldicellulosiruptor bescii with high
catalytic activity. J Mol Catal B: Enzymatic 2015;117:13–20.
[44] ul Haq I, Akram F. Insight into kinetics and thermodynamics of a novel
hyperstable GH family 10 endo-1, 4-b-xylanase (TnXynB) with broad
substrates specificity cloned from Thermotoga naphthophila RKU-10T.
Enzyme Microb Technol 2019;127:32–42.
[45] ul Haq I, Hussain Z, Khan MA, Muneer B, Afzal S, Majeed S, Akram F. Kinetic
and thermodynamic study of cloned thermostable endo-1,4-b-xylanase from
Thermotoga petrophila in mesophilic host. Mol Biol Rep 2012;39(7):7251–61.
[46] Yu Tianyi, Anbarasan Sasikala, Wang Yawei, Telli Kübra, Aslan Asßkın Sevinç, Su
Zhengding, et al. Hyperthermostable Thermotoga maritima xylanase XYN10B
shows high activity at high temperatures in the presence of biomass-
dissolving hydrophilic ionic liquids. Extremophiles: life under extreme
conditions. Extremophiles 2016;20(4):515–24.
[47] Amorim Cláudia, Silvério Sara C, Prather Kristala LJ, Rodrigues Lígia R. From
lignocellulosic residues to market: production and commercial potential of
xylooligosaccharides. Biotechnol Adv 2019;37(7):107397. https://doi.org/
10.1016/j.biotechadv.2019.05.003.
[48] Santibáñez Luciana, Henríquez Constanza, Corro-Tejeda Romina, Bernal
Sebastián, Armijo Benjamín, Salazar Oriana. Xylooligosaccharides from
lignocellulosic biomass: a comprehensive review. Carbohydr Polym
2021;251:117118. https://doi.org/10.1016/j.carbpol.2020.117118.
[49] Conway Jonathan M, Crosby James R, Hren Andrew P, Southerland Robert T,
Lee Laura L, Lunin Vladimir V, et al. Novel multidomain, multifunctional
glycoside hydrolases from highly lignocellulolytic Caldicellulosiruptor species.
AIChE J 2018;64(12):4218–28.
[50] Solomon V, Teplitsky A, Shulami S, Zolotnitsky G, Shoham Y, Shoham G.
Structure-specificity relationships of an intracellular xylanase from Geobacillus
stearothermophilus. Acta Crystallogr D Biol Crystallogr 2007;63(8):845–59.
[51] Gallardo Óscar, Pastor FI Javier, Polaina Julio, Diaz Pilar, Łysek Robert, Vogel
Pierre, et al. Structural insights into the specificity of Xyn10B from
Paenibacillus barcinonensis and its improved stability by forced protein
evolution. J Biol Chem 2010;285(4):2721–33.
[52] González-Blasco Gracia, Sanz-Aparicio Juliana, González Beatriz, Hermoso Juan
A, Polaina Julio. Directed evolution of b-glucosidase A from Paenibacillus
polymyxa to thermal resistance. J Biol Chem 2000;275(18):13708–12.
[53] Chakravorty Debamitra, Khan Mohd Faheem, Patra Sanjukta. Multifactorial
level of extremostability of proteins: can they be exploited for protein
engineering?. Extremophiles 2017;21(3):419–44.
[54] Panja Anindya S, Maiti S, Bandyopadhyay B. Protein stability governed by its
structural plasticity is inferred by physicochemical factors and salt bridges. Sci
Rep 2020;10(1):1822. https://doi.org/10.1038/s41598-020-58825-7.
[55] Xie Hefang, Flint James, Vardakou Maria, Lakey Jeremy H, Lewis Richard J, Gilbert
Harry J, et al. Probing the structural basis for the difference in thermostability
displayed by family 10 xylanases. J Mol Biol 2006;360(1):157–67.
[56] Kaneko Satoshi, Ichinose Hitomi, Fujimoto Zui, Kuno Atsushi, Yura Kei, Go
Mitiko, et al. Structure and function of a family 10 b-xylanase chimera of
Streptomyces olivaceoviridis E-86 FXYN and Cellulomonas fimi cex. J Biol Chem
2004;279(25):26619–26.
[57] Ihsanawati, Kumasaka Takashi, Kaneko Tomonori, Morokuma Chihiro,
Yatsunami Rie, Sato Takao, et al. Structural basis of the substrate subsite and
the highly thermal stability of Xylanase 10B from Thermotoga maritima MSB8.
Proteins Struct Funct Genet 2005;61(4):999–1009.
[58] Cambillau Christian, Claverie Jean-Michel. Structural and genomic correlates
of hyperthermostability. J Biol Chem 2000;275(42):32383–6.
[59] Watanabe K, Hata Y, Kizaki H, Katsube Y, Suzuki Y. The refined crystal
structure of Bacillus cereus oligo-1, 6-glucosidase at 2.0 Å resolution: structural
characterization of proline-substitution sites for protein thermostabilization. J
Mol Biol 1997;269(1):142–53.
[60] Arrizubieta Marı
´a Jesús, Polaina Julio. Increased thermal resistance and
modification of the catalytic properties of a b-glucosidase by random
mutagenesis and in vitro recombination. J Biol Chem 2000;275(37):28843–8.
[61] Feller G, Gerday C. Psychrophilic enzymes: molecular basis of cold adaptation.
Cel Mol Life Sci 1997;53(10):830–41.
[62] Auerbach G, Ostendorp R, Prade L, Korndörfer I, Dams T, Huber R, et al. Lactate
dehydrogenase from the hyperthermophilic bacterium Thermotoga maritima:
D. Talens-Perales, E. Jiménez-Ortega, P. Sánchez-Torres et al. Computational and Structural Biotechnology Journal 19 (2021) 2676–2686
2685
The crystal structure at 2.1 Å resolution reveals strategies for intrinsic protein
stabilization. Structure 1998;6(6):769–81.
[63] Kollman Peter A, Allen Leland C. Theory of the hydrogen bond. Chem Rev
1972;72(3):283–303.
[64] Panja Anindya Sundar, Bandopadhyay Bidyut, Maiti Smarajit, Bhattacharjya
Surajit. Protein thermostability is owing to their preferences to non-polar
smaller volume amino acids, variations in residual physico-chemical
properties and more salt-bridges. PLoS ONE 2015;10(7):e0131495. https://
doi.org/10.1371/journal.pone.013149510.1371/journal.pone.0131495.
g00110.1371/journal.pone.0131495.g00210.1371/journal.pone.0131495.
g00310.1371/journal.pone.0131495.g00410.1371/journal.pone.0131495.
g00510.1371/journal.pone.0131495.g00610.1371/journal.pone.0131495.
g00710.1371/journal.pone.0131495.g00810.1371/journal.pone.0131495.
t00110.1371/journal.pone.0131495.t00210.1371/journal.pone.0131495.
t00310.1371/journal.pone.0131495.s001.
[65] Prajapati RS, Sirajuddin M, Durani V, Sreeramulu S, Varadarajan R.
Contribution of cation-pi interactions to protein stability. Biochemistry 2006
Dec 19;45(50):15000–10.
D. Talens-Perales, E. Jiménez-Ortega, P. Sánchez-Torres et al. Computational and Structural Biotechnology Journal 19 (2021) 2676–2686
2686
