Tài liệu Báo cáo khoa học: Secondary substrate binding strongly affects activity and binding affinity of Bacillus subtilis and Aspergillus niger GH11 xylanases docx

Secondary substrate binding strongly affects activity and
binding affinity of Bacillus subtilis and Aspergillus niger
GH11 xylanases
Sven Cuyvers, Emmie Dornez, Mohammad N. Rezaei, Annick Pollet, Jan A. Delcour and
Christophe M. Courtin
Laboratory of Food Chemistry and Biochemistry & Leuven Food Science and Nutrition Research Centre (LFoRCe), Katholieke Universiteit
Leuven, Belgium

Keywords
arabinoxylan; GH11; noncatalytic binding;
single domain xylanase; surface binding
Correspondence
C. Courtin, Laboratory of Food Chemistry
and Biochemistry & Leuven Food Science
and Nutrition Research Centre (LFoRCe),
Katholieke Universiteit Leuven, Kasteelpark
Arenberg 20 - PO Box 2463, B-3001
Leuven, Belgium
Fax: + 32 16 321997
Tel: +32 16 321917
E-mail:

The secondary substrate binding site (SBS) of Bacillus subtilis and Aspergillus niger glycoside hydrolase family 11 xylanases was studied by site-directed mutagenesis and evaluation of activity and binding properties of
mutant enzymes on different substrates. Modification of the SBS resulted
in an up to three-fold decrease in the relative activity of the enzymes on
polymeric versus oligomeric substrates and highlighted the importance of
several amino acids in the SBS forming hydrogen bonds or hydrophobic
stacking interactions with substrates. Weakening of the SBS increased Kd
values by up to 70-fold in binding affinity tests using natural substrates.
The impact that modifications in the SBS have both on activity and on
binding affinity towards polymeric substrates clearly shows that such structural elements can increase the efficiency of these single domain enzymes

on their natural substrates.

(Received 12 October 2010, revised 11
January 2011, accepted 20 January 2011)
doi:10.1111/j.1742-4658.2011.08023.x

Introduction
Glycoside hydrolases can possess noncatalytic polysaccharide binding sites that facilitate attack on the natural substrate. Most of these sites belong to separate
domains, referred to as carbohydrate-binding modules
(CBMs), linked to the catalytic domain through flexible linker regions. Elaborate research has clarified the
functional relevance of these CBMs [1–4]. CBMs are
considered to target the enzyme towards specific cell
wall regions and to keep it in proximity of the substrate. In some cases, distortion of the substrate structure by the CBMs is considered to facilitate hydrolysis
[5]. CBMs can also be involved in binding the bacterial

cell wall, thereby anchoring the attached enzyme onto
the bacterial surface [6,7].
Despite the clear advantage of having CBMs, some
glycoside hydrolases consist of a catalytic domain only
[8]. Studies investigating the structure of carbohydrateactive enzymes have revealed the presence of other
substrate binding regions situated on the surface of the
structural unit that contains the catalytic site, rather
than on an auxiliary domain [9]. These substrate binding sites are located at a certain distance from the
active site and are called secondary binding sites
(SBS). The presence of one or more SBS has been

Abbreviations
AX, arabinoxylan; AzU, unit of enzyme activity on Azo-wheat AX; CBM, carbohydrate-binding module; GH, glycoside hydrolase family;
OSX, insoluble xylan from oat spelts; SBS, secondary binding site; X6, xylohexaose; X6U, unit of enzyme activity on xylohexaose;
XAN, Aspergillus niger xylanase; XBS, Bacillus subtilis xylanase; XyU, unit of enzyme activity on Xylazyme AX.

1098

FEBS Journal 278 (2011) 1098–1111 ª 2011 The Authors Journal compilation ª 2011 FEBS


S. Cuyvers et al.

Secondary substrate binding in GH11 xylanases

reported in enzymes belonging to glycoside hydrolase
family (GH) 8, 10, 11, 13, 14, 15, 16 and 77 [9–20].
The widespread occurrence of these binding sites indicates that incorporation of a SBS provides an evolutionary benefit for these enzymes. However, the
function of these SBS in many enzymes remains to be
unraveled. To date, most work aiming to understand
the role of SBS has been performed on starch degrading enzymes. Human salivary a-amylase contains several SBS. Mutational analysis demonstrated that these
SBS residues are important for the activity on starch
and that they play a role in the binding of the enzyme
to bacteria of the oral cavity [15]. Nielsen et al. [9]
concluded that the two SBS in barley a-amylase each
have a distinct binding specificity, although they both
play a role in substrate targeting. In two single domain
glucoamylases from Saccharomycopsis fibuligera, a SBS
was also found to enhance binding to starch granules
[18].
In xylanases (EC 3.2.1.8), the existence of a SBS was
also discovered in several single domain enzymes: one

belonging to GH8 [10], one to GH10 [11] and three to
GH11 [12,13]. In GH11, the existence of these SBS

was recently identified by NMR-monitored titrations
of Bacillus circulans xylanase [13] and X-ray analysis of
crystals of catalytically incompetent mutants of the
xylanases of Bacillus subtilis (XBS) and Aspergillus
niger (XAN) soaked with xylo-oligosaccharides [12].
GH11 xylanases have a b-jelly roll fold structure,
which is often compared to a partially closed right
hand [21]. The SBS are present in different regions of
the GH11 xylanases. In the B. circulans xylanase and
in XBS, the SBS is located on the ‘knuckles’ of the
enzyme, whereas, in XAN, it is located at the ‘tip of
the fingers’ (Fig. 1). Because SBS in these enzymes are
located distant from the active site, their impact on
substrate hydrolysis is expected to be limited to longer
substrates.
In the present study, the impact of the presence of a
SBS on the biochemical properties of single domain
GH11 xylanases was investigated by extensive mutational analysis of XBS and XAN. This allowed the

A

B

Fig. 1. Superposition of the overall structures of the xylanases from A. niger (green, PDB 2QZ2) and B. subtilis (blue, PDB 2QZ3) in complex
with oligosaccharides shows the presence of a secondary substrate binding site in different surface regions of these enzymes (on the left).
The figure was drawn using PYMOL ( On the right, schematic representations are shown of the oligosaccharides bound at the secondary binding sites of the A. niger xylanase (A) and the B. subtilis xylanase (B). The diagrams of protein–ligand interactions were generated using LIGPLOT [37] based on Vandermarliere et al. [12]. Amino acid residues that form direct protein–ligand
interactions with their main chain only are indicated by an asterisk.

FEBS Journal 278 (2011) 1098–1111 ª 2011 The Authors Journal compilation ª 2011 FEBS

1099


Secondary substrate binding in GH11 xylanases

S. Cuyvers et al.

investigation of whether the SBS has a similar functionality in different GH11 xylanases and hence
whether the occurrence of SBS is a more general strategy of GH11 xylanases to compensate for their lack of
CBMs.

could not entirely be ruled out at this stage. The
screening procedure was performed on E. coli cell
lysates containing (mutant) XBS and on P. pastoris
expression media containing (mutant) XAN. The
results of the screening for XBS and XAN with a
modified SBS are shown in Table 1.

Results and Discussion
XBS
Genetic engineering of the SBS
Residues of the SBS of both XBS and XAN involved
in substrate interaction were selected based on the
crystal structures of XBS and XAN soaked with xylotetraose and xylopentaose, respectively [12] (Fig. 1),
and were subjected to genetic engineering using sitedirected mutagenesis. Amino acid residues reported to
potentially play a role in secondary substrate binding
were mutated to Ala aiming to investigate their importance. Several mutations were also combined to assess
the importance of the SBS as a whole for the biochemical properties of XBS and XAN. In an attempt to
increase substrate binding affinity of the SBS, aromatic
residues were introduced at certain places to create

extra or stronger hydrophobic stacking interactions or
residues were replaced to create new hydrogen bonds.
Screening procedure
To examine the impact of different mutations on the
functionality of the SBS in XBS and XAN, a screening
method on nonpurified enzyme samples was developed.
Because the SBS is located far from the active site, it
was hypothesized that the hydrolysis of soluble, oligomeric substrates, such as xylohexaose (X6), is not influenced by the presence of a SBS because this substrate
cannot interact with both the SBS and the active site
at the same time. By contrast, larger polymeric substrates, such as Xylazyme arabinoxylan (AX), can
reach both sites simultaneously. Previously, it was
demonstrated that X6 binds independently to the active
site and the SBS of the B. circulans xylanase, whereas
larger substrates, such as xylododecaose, bind the two
sites cooperatively [13]. Accordingly, a screening ratio
was defined as the activity on Xylazyme AX divided
by the activity on X6. This ratio is considered to reflect
the impact of a modification in the SBS on its functionality, independent of the expression efficiency of
the protein. Because Escherichia coli and Pichia pastoris do not produce xylanolytic enzymes, this ratio of
two activities enables a comparison of nonpurified
enzymes. However, the possibility that other proteins
present in the nonpurified enzyme samples might have
an unforeseen effect on the activity of XBS or XAN
1100

The results of the screening clearly show that modification of the SBS of XBS leads to a lower relative efficiency towards the water-unextractable Xylazyme AX
compared to that towards X6. These results are in
agreement with the observations of Ludwiczek et al.
[13] on GH11 B. circulans xylanase, which has a SBS
equivalent to that of XBS. Replacement of residues

considered to play a role in secondary binding with
Ala leads to a lower screening ratio for all enzymes.
For the G56A-T183A-W185A mutant, a large drop in
the screening ratio is seen, resulting in a ratio that is
only half the ratio obtained for the wild-type XBS.
The results shown in Table 1 also demonstrate the
importance of the hydrophobic stacking interaction
that Trp185 makes with bound substrate. Hydrogen
bonds with other residues also appear to be of major
importance. Mutation of residues Thr183, Asn181 and
Gly56 to Ala leads to a large drop in the screening
ratio. A smaller effect is also observed upon mutation
of Asn54 and Asn141. These results correspond well
with the results obtained in the previous study by Vandermarliere et al. [12] that reported residues important
for secondary substrate binding. The results obtained
for mutants where an attempt was made to increase
substrate binding affinity show that the efficiency on
polymeric versus oligomeric substrate was not
increased. Lower screening ratios emerged for all these
mutants compared to the wild-type. This shows that
the intended fortification of the SBS failed or that a
stronger SBS does not lead to more efficient hydrolysis
of polymeric substrate.
XAN
Screening of XAN mutants where amino acid residues
involved in secondary substrate binding are replaced
by Ala shows a trend similar to that observed for the
set of XBS mutants, although the differences are smaller. This indicates that the SBS in both enzymes probably has the same functionality. The screening ratio
goes down by a maximum of approximately 30% in
mutants E31A and Y29A-E31A, whereas, for XBS,

decreases of up to 50% were observed. Glu31 appears
to be indispensable for the SBS of XAN because the

FEBS Journal 278 (2011) 1098–1111 ª 2011 The Authors Journal compilation ª 2011 FEBS


S. Cuyvers et al.

Secondary substrate binding in GH11 xylanases

Table 1. The effect of genetic engineering of the secondary binding site of the B. subtilis and A. niger xylanases on the screening
ratio. The screening ratio is defined as the ratio of activity on Xylazyme AX and activity on X6. Screening ratios are expressed relative
to the ratio of the wild-type enzyme (100%) and were calculated
based on two independent activity measurements on X6 and two
independent activity measurements on Xylazyme AX on the same
unpurified enzyme sample, with each independent assay comprising three replicates. Data are shown as the mean ± SD.
Screening
ratio (%)
XBS
Wild-type
N54A
G56A
N141A
N181A
T183A
W185A
N54A-G56A
N54A-T183A
N181A-T183A
G56A-T183A-W185A

N54F
N54W
N141Q
N54W-N141Q
XAN
Wild-type
D16A
Y29A
E31A
D32A
D16A-E31A
Y29A-E31A
D16A-Y29A-E31A
G15W
D16Y
Y29W
E31Q
E31T
D32E
D32F
D32N
D32Q
D32W

100
86
70
81
70
63

62
68
65
66
54
83
76
80
79

±
±
±
±
±
±
±
±
±
±
±
±
±
±
±

13
19
15*
13

21
13*
12*
13*
16*
23
14*
10
17
26
21

100
92
82
70
101
76
73
82
98
83
89
101
104
105
86
99
98
95

±
±
±
±
±
±
±
±
±
±
±
±
±
±
±
±
±
±

13
6
3
13*
2
16
5*
5*
12
8

8
10
8
13
8
11
11
7

*Significantly different from the wild-type enzyme by Student’s
t-test (P < 0.05). To account for multiple comparisons, the signifi´
cance levels were adjusted according to Scheffe’s method.

screening ratio obtained for mutant E31A is not lowered further when extra Ala mutations are introduced.
Glu31 can make several hydrogen bonds with substrate
bound in the SBS [12] (Fig. 1). Tyr29, which can make
a hydrophobic stacking interaction with bound substrate, also appears to be an important residue for the
SBS. The screening ratio of D32A is similar to that of

the wild-type, indicating the minor importance of the
acidic side chain of Asp32. This result is logical
because Asp32 makes only one hydrogen bond with
surface bound substrate through a main chain amine
group that is not abolished by the D32A mutation
˚
(Fig. 1). The acidic side chain is 3.7 A away from a
hydroxyl group of the bound substrate and this distance is too far to form a relevant hydrogen bond [12].
The introduction of amino acid residues to create new
or stronger hydrophobic stacking of hydrogen bonds
with substrate in the SBS has led to screening ratios

similar to that of the wild-type XAN for most
enzymes. Subtle changes in the hydrogen bonding
appear to have no (or only a very minor) effect on the
functionality of the SBS. In some cases, the introduction of aromatic residues even lowered the screening
ratio, as was seen for some of the XBS mutants.
Activity measurements
After the screening procedure, a smaller set of enzymes
was selected for purification and further biochemical
characterization. The activity of these enzymes was
determined on X6 and two chromophoric polymeric
substrates: the water-unextractable Xylazyme AX and
the water-extractable Azo-wheat AX. Table 2 lists
these results along with temperature and pH optima of
the enzymes. Most mutations lead to a lower temperature optimum, whereas little or no change is observed
in the pH optimum. The lowered temperature optimum is possibly explained by decreased enzyme stability at higher temperatures. The solubilization of
water-unextractable AX isolated from wheat flour and
insoluble oat spelt xylan (OSX) by the different
mutants was also examined. For several mutants, the
solubilization in function of the enzyme concentration
used in the assay is shown in Fig. 2.
XBS
Mutations in the SBS of XBS appear to have no (or
very little) effect on the activity on X6 because all XBS
mutants give results similar to those of the wild-type
xylanase. The substrate is probably too small, so the
enzyme cannot benefit from the presence of an additional SBS at the enzyme surface. However, the activities on Xylazyme AX and Azo-wheat AX drop upon
modification of the SBS. These results suggest that the
functionality of the SBS is limited to larger substrates
that can reach both the active site and SBS at the same
time. Because the activity on X6 remains the same

upon engineering of the SBS of XBS, the results
obtained in the previous screening experiment directly

FEBS Journal 278 (2011) 1098–1111 ª 2011 The Authors Journal compilation ª 2011 FEBS

1101


Secondary substrate binding in GH11 xylanases

S. Cuyvers et al.

Table 2. Biochemical characterization of B. subtilis and A. niger xylanases with a modified secondary binding site. Values shown are
expressed relative to the activity of the wild-type enzyme (100%). The activity on X6 was calculated from values obtained from three independent trials, each comprising five independent samples; the activity on Xylazyme AX and Azo-wheat AX was from five independent trials,
each comprising three replicates. For each enzyme, values with the same letter in one column are not significantly different from each other
according to Tukey’s tests (P < 0.05) performed with SAS, version 9.2 (SAS Institute). For the wild-type XBS, 1 X6U = 1.34 · 10)10 M,
1 XyU = 9.76 · 10)10 M and 1 AzU = 50.8 · 10)10 M and for the wild-type XAN, 1 X6U = 1.02 · 10)10 M, 1 XyU = 7.72 · 10)10 M and
1 AzU = 9.65 · 10)10 M for the activity on X6, Xylazyme AX and Azo-wheat AX, respectively (data are shown as the mean ± SD). Kd values
are expressed in mgỈmL)1 and are apparent Kd values in many cases as a result of substrate concentration limitations in the test (data are
shown as the mean ± SE from the fit on a single curve). The reported temperature and pH ranges indicate the intervals in which the
observed activity was at least 95% of the maximal activity of the enzyme.
Activity (%) on
Xylazyme
AX

Xylohexaose
XBS
Wild-type
N181A
T183A

W185A
G56A-T183A-W185A
N54W
N141Q
N54W-N141Q
XAN
Wild-type
D16A
Y29A
E31A
D32A
Y29A-E31A
D16A-Y29A-E31A
Y29W
E31Q
E31T
D32E

Affinity (Kd) towards
Azo-wheat
AX

Waterunextractable AX

OSX

Temperature

pH

100
113
100
100
94
102
103
102

±
±
±
±
±
±
±
±

4
6
4
8
5
4
3
6

AC
C
AB

AB
A
B
AB
B

100
81
73
70
52
89
96
92

±
±
±
±
±
±
±
±

4
6
3
6
2
1

4
3

E
C
B
B
A
D
DE
DE

100
77
66
53
33
85
92
92

±
±
±
±
±
±
±
±

9
8
7
4
5
8
14
11

E
CD
C
B
A
DE
DE
DE

8.8
11
11
17
25
8.1
6.1
5.3

±
±
±

±
±
±
±
±

0.4
2
1
2
4.5
0.5
0.3
0.4

0.4
8.1
6.2
13
29
5.6
1.9
2.5

±
±
±
±
±
±

±
±

0.01
0.6
0.3
1
3
0.4
0.1
0.2

48–54
43–49
38–42
39–46
37–42
43–50
47–52
46–51

°C
°C
°C
°C
°C
°C
°C
°C

5.2–6.8
5.2–6.8
5.2–6.6
5.3–6.6
5.2–6.5
5.3–6.6
5.2–6.8
5.2–6.8

100
86
97
51
102
61
39
111
94
102
77

±
±
±
±
±
±
±
±
±

±
±

10
4
19
8
11
8
7
9
12
8
10

DE
DE
DE
B
E
C
A
E
DE
E
D

100
85
80

44
99
50
35
102
99
103
72

±
±
±
±
±
±
±
±
±
±
±

3
4
9
5
3
5
2
2
3

4
4

F
E
E
B
F
C
A
F
F
F
D

100
82
72
45
91
48
29
98
93
99
77

±
±
±

±
±
±
±
±
±
±
±

9
1
3
3
4
5
6
7
5
5
1

E
CD
C
B
DE
B
A
E
DE

E
C

24
70
93
57
57
117
122
66
28
45
36

±
±
±
±
±
±
±
±
±
±
±

3
17
27

11
10
51
64
14
3
7
4

3.8
15
41
12
10
51
61
17
6.3
10
5.2

±
±
±
±
±
±
±
±
±

±
±

0.5
2
7
1
1
23
32
2
0.4
1
0.5

47–51
44–50
43–47
43–47
43–49
40–45
39–41
43–47
43–47
44–50
46–51

°C
°C
°C

°C
°C
°C
°C
°C
°C
°C
°C

3.4–4.1
3.4–4.1
3.4–3.8
3.4–4.1
3.6–4.1
3.4–4.1
3.8–3.8
3.5–4.1
3.6–4.1
3.5–4.1
3.4–4.1

reflect the activity of the enzymes on Xylazyme AX.
The activity drop on polymeric substrates is the largest
for the G56A-T183A-W185A, mutant with an activity
that dropped to half on Xylazyme AX and even to
one-third on Azo-wheat AX compared to that of wildtype XBS. The trends observed on chromophoric
substrates are confirmed on natural substrates because
modification of the SBS decreases the rate at which
XBS solubilizes water-unextractable AX and OSX.
Especially in the case of OSX, the solubilization by the

enzymes is greatly hampered upon modification of the
SBS. At the same enzyme concentration, the wild-type
XBS solubilized the most. The maximal attainable solubilization of these substrates by different mutants was
also measured, although no clear differences were
observed (results not shown). This indicates that the
SBS influences the rate of hydrolysis, most likely by
enhancing substrate recognition, rather than affecting
the real catalytic potential and substrate specificity of
the enzyme.
1102

Optimal conditions

XAN
By contrast to results on XBS, the activity on X6 is
affected by a number of the mutations made in the
SBS of XAN. A much lower activity is observed especially for those enzymes containing the E31A mutation
display. This single mutation leads to an activity on X6
that is only half that of the wild-type XAN. However,
a loss of activity on X6 in the set of purified XAN
mutants does not appear to be correlated with a weakening of the SBS. For example, Y29A does not display
a lower activity on X6, whereas it is regarded as one of
the most important residues for secondary binding.
The decrease also cannot be explained by differences
in enzyme stability under the assay conditions
(Fig. S1). The small decrease in activity under assay
conditions observed for a few XAN mutants is not
proportional to the decrease in activity on X6. The loss
of activity on X6 for certain mutants might be attributed to these mutations exerting an effect on the

FEBS Journal 278 (2011) 1098–1111 ª 2011 The Authors Journal compilation ª 2011 FEBS


S. Cuyvers et al.

C

40

WU-AX solubilisation
(% of WU-AX)

WU-AX solubilisation
(% of WU-AX)

A

Secondary substrate binding in GH11 xylanases

30

20

10

40

30

20

10

0

0
0

30

20

10

0

10
20
30
40
Enzyme concentration (×10–8 M)

40

30

20

10

0

10
20
30
40
Enzyme concentration (×10–8 M)
LEGEND
XBS wild-type
XBS W185A
XBS G56A-T183A-W185A

10
20
30
40
Enzyme concentration (×10–9 M)

0

D

40

0

0

OSX solubilisation
(% of OSX)

OSX solubilisation
(% of OSX)

B

10
20
30
40
Enzyme concentration (×10–10 M)

XBS N54W
XBS N54W-N141Q

LEGEND
XAN wild-type
XAN Y29A
XAN E31A

XAN Y29A-E31A
XAN D16A-Y29A-E31A

Fig. 2. Solubilization of water-unextractable arabinoxylan (WU-AX) (A) and oat spelt xylan (OSX) (B) by B. subtilis xylanase mutants with a
modified secondary binding site and of water-unextractable arabinoxylan (C) and oat spelt xylan (D) by A. niger xylanase mutants.

overall catalytic efficiency of the enzyme. Changes of
SBS residues located on the outer b-sheet of the XAN
structure might induce subtle changes in the position
of important binding or catalytic residues in the active

site located on the inner b-sheet of the b-jelly roll. The
results for the activity on Xylazyme AX and Azowheat AX in Table 2 therefore do not show clear
trends at first sight. The ratios of activities on Xylazyme AX over X6, however, do confirm the results of
the screening. Solubilization experiments with the natural substrates OSX and water-unextractable AX also
clearly demonstrate that the solubilizing capacity is
strongly decreased upon modification of SBS residues.
The observed drops in solubilizing capacity are especially spectacular on OSX.
Comparison of XBS and XAN
For XBS, it is clear that the residues involved in secondary substrate binding play no (or a very minor)

role in the hydrolysis of oligomeric substrates. The
SBS is located too far from the active site to influence
the binding and catalysis of these substrates in the
active site. The same statement is probably true for
most residues in the SBS of XAN, although, in this
case, some mutants (especially those containing the
E31A mutation) display a strong decrease in activity
on X6. The reason for this is not clear. Possibly, these
mutations provoke subtle positional changes of residues located in the active site. The results of activity
measurements on purified enzymes, as presented in the
present study, support the results already obtained by
screening and thereby indicate that the screening procedure can indeed provide a valuable tool for the initial selection of mutant enzymes. The SBS in XBS and
XAN mainly function to increase the efficiency of the
enzyme on polymeric substrates, as demonstrated by
the results obtained for both chromophoric and natural substrates. The drop in activity on polymeric substrates upon mutations in the SBS is substantial,

FEBS Journal 278 (2011) 1098–1111 ª 2011 The Authors Journal compilation ª 2011 FEBS

1103

Secondary substrate binding in GH11 xylanases

S. Cuyvers et al.

especially when considering that these mutations are
located far from the active site. In general, the drop in
activity on Azo-wheat AX is slightly larger than that
on Xylazyme AX, as shown in Table 2. One parameter
that is often used to refer to the ratio of activity
towards water-unextractable and water-extractable AX
is substrate selectivity [22]. It is often expressed as a
substrate selectivity factor, which can be calculated as
the activity on Xylazyme AX over the activity on Azowheat AX [22], and is a determinant for functionality
of xylanases in several applications [23]. Table 3 lists
substrate selectivity factors for XBS and XAN with a
modified SBS. Although the differences are small, a
general trend can be seen in which weakening of the
SBS increases substrate selectivity. In XBS, especially
for those mutants containing W185A, significant differences in substrate selectivity are observed. In XAN,
the differences are smaller, although the same general
trend is seen. Water-unextractable AX are probably
less flexible and therefore it might be more difficult
for the SBS to exploit its full functionality with respect
to the hydrolysis of these substrates. Whether the
observed differences in substrate selectivity are relevant
in applications remains to be explored. Strikingly, the
W185A mutation in XBS has already been characterized with regard to its effect on substrate selectivity.
However, Moers et al. [24] reported a drop in the substrate selectivity factor for the W185A mutant. One
possible explanation for this discrepancy could be the

use of a His-tagged protein in their study. The C-terminal location of this His-tag suggests a likely interference with the functionality of the SBS because three
important residues for secondary binding are located
near the C-terminus.
Binding affinity towards insoluble polymers
As outlined above, the effect of the SBS on activity
towards different substrates was studied. Obviously,
substrate binding is closely linked to activity. Therefore, the binding of the different XBS and XAN
mutants towards water-unextractable AX and OSX
was assessed by constructing binding curves, as shown
for several mutants in Fig. 3. Table 2 gives the overall
dissociation constants (Kd) derived from these curves.
XBS
Weakening of the SBS of XBS clearly increases Kd values and therefore lower affinities towards both waterunextractable AX and OSX. The differences on OSX
are more pronounced than those on water-unextractable AX. The G56A-T183A-W185A mutant has a Kd
1104

Table 3. Substrate selectivity factors of B. subtilis and A. niger
xylanases with a modified secondary binding site. The substrate
selectivity factor is calculated as the ratio of activity on Xylazyme
AX over the activity on Azo-wheat AX, with both activities calculated based on the values obtained from five independent trials,
each comprising three replicates. All values are expressed relative
to the wild-type enzyme (1.00). Data are shown as the mean ± SD.
Substrate
selectivity factor
XBS
Wild-type
N181A
T183A
W185A
G56A-T183A-W185A

N54W
N141Q
N54W-N141Q
XAN
Wild-type
D16A
Y29A
E31A
D32A
Y29A-E31A
D16A-Y29A-E31A
Y29W
E31Q
E31T
D32E

1.00
1.05
1.07
1.31
1.57
1.04
1.01
1.00

±
±
±
±
±

±
±
±

0.10
0.10
0.08
0.10*
0.09*
0.08
0.14
0.11

1.00
1.03
1.11
0.98
1.08
1.04
1.18
1.04
1.06
1.03
0.94

±
±
±
±
±

±
±
±
±
±
±

0.09
0.04
0.11
0.06
0.05
0.07
0.08*
0.07
0.06
0.06
0.04

*Significantly different from the wild-type enzyme by Student’s
t-test (P < 0.05). To account for multiple comparisons, the signifi´
cance levels were adjusted according to Scheffe’s method.

of 25 mgỈmL)1 on water-unextractable AX and
29 mgỈmL)1 on OSX, respectively, compared to
8.8 mgỈmL)1 and 0.4 mgỈmL)1 for the wild-type XBS.
It can probably be regarded as an enzyme without a
SBS. The elimination of the possibility of substrate
forming a hydrophobic stacking interaction with
Trp185 (W185A) leads to the largest increase in Kd

for a single mutation, giving rise to Kd values of
on
water-unextractable
AX
and
17 mgỈmL)1
13 mgỈmL)1 on OSX. On water-unextractable AX,
N54W and N141Q appear to have slightly lower Kd
values, similar to the combined N54W-N141Q mutant,
which has a Kd of 5.3 mgỈmL)1. By contrast, on OSX,
all mutants give higher Kd values than the wild-type
enzyme. It is difficult to pinpoint the reason for the
higher affinity of these mutants towards water-unextractable AX. One possibility is that the attempt to
enhance substrate binding was successful for waterunextractable AX, although a higher affinity is
not necessarily correlated with a higher activity.

FEBS Journal 278 (2011) 1098–1111 ª 2011 The Authors Journal compilation ª 2011 FEBS


S. Cuyvers et al.

Secondary substrate binding in GH11 xylanases

C
100

Enzyme bound to substrate (%)

Enzyme bound to substrate (%)

A

80
60
40
20
0
0

10

20
30
WU-AX (mg·mL)

80
60
40
20
0
0

10

20
30
WU-AX (mg·mL)

40

0

40

B

10

20
30
OSX (mg·mL)

40

D
100

Enzyme bound to substrate (%)

Enzyme bound to substrate (%)

100

80
60
40
20
0
0

10

20
30
OSX (mg·mL)

LEGEND
Wild-type XBS
XBS W185A
XBS G56A-T183A-W185A

100
80
60
40
20
0

40

LEGEND
Wild-type XAN
XAN Y29A
XAN E31A

XBS N54W
XBS N54W-N141Q

XAN Y29A-E31A
XAN D16A-Y29A-E31A

Fig. 3. Binding of B. subtilis xylanase mutants with a modified secondary binding site to water-unextractable arabinoxylan (WU-AX) (A) and
oat spelt xylan (OSX) (B) and of A. niger xylanase mutants to water-unextractable arabinoxylan (C) and oat spelt xylan (D).

A substrate may be bound too tightly to the enzyme
to allow efficient hydrolysis. For CBMs, it has also
been suggested that too strong a binding affinity
between substrate and CBM may limit the activity of
the attached enzyme [1,25]. Another possibility is that
these mutations result in stronger substrate binding
but that this binding, for example, orients the substrate wrongly to assist in its catalysis in the active
site. Binding experiments have shown that N54W,
N141Q and N54W-N141Q also display higher affinity
towards some other polysaccharides such as cellulose
and barley b-glucan than the wild-type XBS (results
not shown). This might indicate that these mutations
create a ‘sticky patch’ causing aspecific binding to all
kinds of substrates, rather than enhancing the specific
binding of xylan substrates in a correct orientation to
help provide the catalytic site with substrate for
hydrolysis.

XAN
Affinity towards both water-unextractable AX and
OSX is decreased upon modification of the SBS of
XAN. The wild-type XAN has lower Kd values than
mutants that weaken the SBS, as well as mutants
aimed at creating a SBS with increased substrate binding affinity. The Kd of wild-type XAN is 24 mgỈmL)1
for water-unextractable AX and 3.8 mgỈmL)1 for
OSX. The Y29A mutation increases the Kd values to

93 mgỈmL)1 and 41 mgỈmL)1 for water-unextractable
AX and OSX, respectively. The largest increase is seen
for the D16A-Y29A-E31A mutant, with Kd values of
122 mgỈmL)1 and 61 mgỈmL)1 on water-unextractable
AX and OSX, respectively. The binding of mutants
aimed at obtaining an enzyme with an increased substrate binding affinity in the SBS also appeared to be
negatively affected. However, for most of these

FEBS Journal 278 (2011) 1098–1111 ª 2011 The Authors Journal compilation ª 2011 FEBS

1105


Secondary substrate binding in GH11 xylanases

S. Cuyvers et al.

enzymes, this is not reflected by the activity measurements. Unexpectedly, Kd values of D32A are higher
than those of the wild-type, whereas the activity of the
mutant was unaffected.
Comparison of XBS and XAN
In general, large differences in affinity are observed for
both XBS and XAN upon modification of the SBS.
Even single mutations in the SBS can lead to a drastic
increase in the Kd value. Aromatic residues (Trp185 in
XBS and Tyr29 in XAN) appear to play an essential role
in the substrate binding. In most cases, a lower activity
and a lower affinity appear to be satisfactorily correlated. However, in XBS, stronger binding does not necessarily increase activity. This result suggests that the
function of the SBS is not merely to bring the enzyme
into contact with insoluble substrates, as is often proposed as one of the functions of CBMs [1,2]. The SBS

possibly has a more pronounced role, such as assisting
catalysis by leading the substrate into the active site.
Relevance of the present findings
Many glycoside hydrolases contain one or more CBMs
that are considered to function as an aid to target substrates and to keep enzymes in proximity with their
substrate [1,2]. The discovery of a SBS in two single
domain xylanases from B. subtilis and A. niger led to
the suggestion that these structures compensate for the
lack of CBMs in these enzymes [12]. In the B. circulans
xylanase, it was found that the SBS assists the active
site by binding larger substrates cooperatively, thereby
facilitating their hydrolysis [13]. In the present study,
the demonstrated effects of modification in the SBS of
XBS and XAN on binding affinities, as well as on
activity, indicate that these sites are of significant
importance for the enzymes. Previously, the deletion of
CBMs from (or fusion of CBMs to) xylanases was
shown to lead to lower activities in the absence of the
CBM comparable to those seen for the elimination of
the SBS in the present study [26–28]. In most studies,
however, the presence of CBMs is correlated with a
higher activity on insoluble substrate, whereas the
activity on soluble substrate is often unaffected [26,27].
However, in the present study, the presence of a SBS
gives rise to higher activities on all tested polymeric
substrates (i.e. both water-extractable and waterunextractable). The presence of a SBS was even more
beneficial for enzyme activity on water-extractable substrate. This difference might be explained by the fact
that a water-extractable AX chain is probably more
flexible and therefore can be guided more easily into
1106

the active site cleft once it is bound to the SBS.
Although the exact functional role of the SBS in
GH11 xylanases is not clear from the data obtained in
the present study, it might be speculated that SBS is
involved in targeting the enzymes towards their substrate and, subsequently, in anchoring them onto the
substrate or in feeding the substrate chain to the active
site cleft, corresponding to their suggested role in barley a-amylase [29]. Whatever the case, the demonstrated effect of the SBS on affinity and activity
towards polymeric substrate confirms that the sites are
of great assistance to XBS and XAN with respect to
overcoming the lack of CBMs.
The large beneficial effect of a SBS and its presence
in two different regions of two different single domain
GH11 xylanases leads to the assumption that XBS and
XAN are representatives of a larger group of single
domain xylanases that have evolved this feature. The
sequence alignment of known GH11 enzymes reported
by Sapag et al. [30] reveals that residues important for
SBS in XBS and XAN also occur in other single
domain xylanases. The residues important for secondary substrate binding in XAN occur in a subgroup of
fungal GH11 that is defined as ‘group II’ by Sapag
et al. [30]. The SBS of XBS appears to be present in
several xylanases of other Bacillus species. Furthermore, it is possible that other GH11 xylanases also
contain undiscovered SBS in different regions of the
enzyme. The concept of a SBS in GH11 xylanases
demonstrated in the present study is possibly also valid
for other single domain xylanases. As noted in the
Introduction, the presence of a SBS has also been suggested in GH8 and GH10 xylanases [10,11]. Future
work on these enzymes will aim to clarify whether the
SBS has the same functional relevance for these

enzymes.

Conclusions
Screening of a large set of XBS and XAN with a modified SBS clearly established that the SBS raises the relative activity of single domain xylanases on polymeric
versus oligomeric substrate. Activity measurements on
purified enzymes confirmed these findings. For XBS,
the activity on X6 is independent of the strength of
the SBS; for XAN, this is probably also the case,
although, for a few mutations, the overall activity was
seriously decreased. Although the differences are small,
the efficiency of the SBS in XBS and XAN appears to
be higher for water-extractable substrate, probably
because it is less rigid or more accessible than waterunextractable substrate, which is tightly associated
with other components in the cell wall matrix.

FEBS Journal 278 (2011) 1098–1111 ª 2011 The Authors Journal compilation ª 2011 FEBS


S. Cuyvers et al.

The effects on activity on different substrates and
the binding affinity towards the natural substrates,
water-unextractable AX and OSX, as described in the
present study, imply that the SBS is of significant
importance for XBS and XAN. The SBS in these
single domain xylanases may be an example of a very
efficient strategy that targets the enzyme towards a
substrate to anchor and feed it into the active site,
which are functions often attributed to CBMs in multidomain enzymes. Future work will need to unravel the
mechanism by which this SBS can assist hydrolysis. In

addition, the extent to which SBS occur amongst xylanases and other glycoside hydrolases will need to be
determined.

Materials and methods
Materials
All chemicals, solvents and reagents were purchased from
Sigma-Aldrich (Bornem, Belgium) and are of analytical
grade, unless specified otherwise. Xylazyme AX tablets,
liquid Azo-wheat AX, water-unextractable AX isolated
from wheat flour, barley b-glucan and xylooligosaccharides
up to X6 were obtained from Megazyme (Bray, Ireland).
Pustulan was obtained from Calbiochem (Darmstadt, Germany). Oligonucleotide primers were obtained from SigmaAldrich. The insoluble fraction of oat spelt xylan was
obtained by removing the soluble component from the
starting material. First, an extract was made of 1.0 mg
xylan per 20 mL of water by shaking the solution for
15 min at 4 °C. After centrifugation (1500 g for 10 min),
the residue was boiled for 30 min in 20 mL water per milligram of starting material. After a new centrifugation step
(11000 g for 30 min), the remaining soluble components
were removed from the residue by shaking in 20 mL of
water per milligram of starting material for 15 min at room
temperature. After centrifugation (11000 g for 30 min), the
insoluble fraction was lyophilized.

Secondary substrate binding in GH11 xylanases

E. coli TOP10 cells (Invitrogen, Groningen, the Netherlands) were transformed with the modified plasmids (3 lL)
by heat shock (30 s at 42 °C). Success of mutagenesis was
verified by sequence analysis (Genetic Service Facility, VIB,
Wilrijk, Belgium).

Recombinant expression
XBS
E. coli * BL21 (DE3) pLysS cells transformed with pEXP5CT-xynA (or mutant constructs) were used to express XBS
(and its mutant variants) in accordance with a method
previously described by Pollet et al. [33]. The enzyme yield
of recombinant XBS after purification was typically
20–50 mgỈL)1 culture.

XAN
PmeI (New England Biolabs, Beverly, MA, USA) linearized
pPicZaC-exlA was used to transform P. pastoris strain X33
by electroporation. Extracellular expression of XAN was
performed using the EasySelect Pichia expression kit (Invitrogen) in accordance with the manufacturer’s instructions.
More specifically, 500 lL of an overnight culture of transformed cells in ‘yeast extract peptone dextrose medium’
containing 100 lgỈmL)1 Zeocine (InvivoGen Cayla, Toulouse, France) was used for the inoculation of 90 mL of
‘buffered complex medium containing glycerol’. The resulting culture was then grown for 16–20 h at 30 °C under continuous shaking before the cells were harvested by
centrifugation (2500 g for 5 min) and transferred to 20 mL
of ‘buffered complex medium containing methanol’ for
induction of protein expression. This culture was then incubated for an additional 4 days at 30 °C and 400 lL of pure
methanol was added each day. After centrifugation (2500 g
for 10 min), the supernatant was collected for further purification (see below). One culture typically yielded 5–15 mg
of XAN after purification.

Protein purification
Site-directed mutagenesis
Expression plasmid pEXP5-CT-xyna was used for heterologous expression of XBS (UniProtKB P18429) in E. coli
[31]. For the heterologous expression of XAN (differing in
three amino acids from UniProtKB P55329, namely K50N,
E57D and M167V) in P. pastoris, the pPicZaC-exlA plasmid was used [32]. In both plasmids, a stop codon was
incorporated after the last nucleotide encoding for the Cterminal amino acid of the native protein (Trp185 and

Ser184 for XBS and XAN, respectively). All mutants were
constructed using a QuikChange site-directed mutagenesis
kit (Stratagene, La Jolla, CA, USA). The template DNA
and oligonucleotide primers used are shown in Table S1.

XBS
XBS and its mutant variants were purified from the E. coli
cell lysates with cation exchange chromatography, as previously described by Pollet et al. [33]. To remove the last contaminating proteins, an additional gel filtration step was
performed on a Sephacryl S-100 column (GE Healthcare,
Uppsala, Sweden) with sodium acetate buffer (250 mm,
pH 5.0) as elution buffer.

XAN
XAN and its mutant variants were purified with anion
exchange chromatography, as previously described by Van-

FEBS Journal 278 (2011) 1098–1111 ª 2011 The Authors Journal compilation ª 2011 FEBS

1107


Secondary substrate binding in GH11 xylanases

S. Cuyvers et al.

dermarliere et al. [12]. Xylanase containing fractions were
pooled and dialyzed against a sodium acetate buffer
(200 mm, pH 4.0).

XBS and XAN

All purified XBS, XAN and mutant variants were free
from protein impurities as verified by SDS ⁄ PAGE and
silver staining was performed on a PhastSystem Unit
(GE Healthcare) (Fig. S2).

Protein quantification
XBS
The protein concentration of purified XBS samples was
determined by measurement of E280 with a Nanodrop-1000
spectrophotometer (Thermo Fisher Scientific, Waltham,
MA, USA) using molar extinction coefficients calculated
with protparam software ( based on the known amino acid sequence of
the different mutants. The measurement was performed in
triplicate.

XAN
Purified XAN samples had a faint yellow colour as a result
of components originating from the expression in P. pastoris X33 that interfered with the spectrophotometric protein
quantification methods. Therefore, protein quantification of
the XAN samples was performed by hydrolysis of the protein and subsequent separation and quantification of the
resulting single amino acids using high-performance anion
exchange chromatography with integrated pulsed amperometric detection. A 1.0-mL sample containing 6.0 m HCl,
0.1% phenol, 30 lm norleucine (used as internal standard)
and an estimated final protein level of 0.01–0.10 mg was
incubated for 24 h at 110 °C for hydrolysis after the head
space was flushed with nitrogen gas. HCl was removed by
evaporation and the obtained residue was dissolved in
1.0 mL of water and analyzed on a Dionex BioLC system
(Dionex Corporation, Sunnyvale, CA, USA) as previously
described by Rombouts et al. [34]. The protein concentration was calculated based on the values obtained for

alanine, threonine, glycine, valine, phenylalanine and aspartate. The analysis, starting from protein hydrolysis, was
performed in quintuplicate.

Activity assays
For all activity measurements of XBS and its mutants, a
McIlvaine buffer (100 mm citric acid + 200 mm sodium
phosphate) (pH 6.0) was used, whereas activity measurements of XAN and its mutants were performed in a
McIlvaine buffer at pH 4.0. Both buffers contained

1108

0.50 mgỈmL)1 BSA. The incubation time in all assays was
60 min and the pre-incubation and incubation temperature
was 40 °C. The stability of the different purified enzymes in
these measurement conditions was tested to avoid erroneous conclusions with respect to differences in activity
(Fig. S1). The substrates used in these measurements were
X6, Xylazyme AX, Azo-wheat AX, water-unextractable AX
and OSX. X6 is a soluble, linear, oligomeric substrate.
Xylazyme AX and Azo-wheat AX are polymeric chromophoric AX that are water-unextractable and waterextractable, respectively. Water-unextractable AX is a
substrate isolated from wheat flour. Its water-unextractability is mainly a result of ferulic acid cross-links between AX
molecules and interactions with other cell wall components.
OSX has a low degree of substitution and is considered to
be insoluble as a result of partial alignment of unsubstituted regions [35]. The water-unextractable AX used in the
present study had an average arabinose ⁄ xylose ratio of 0.60
and the average arabinose ⁄ xylose ratio of the OSX was
0.08, as assessed by hydrolysis of the materials and analysis
of the noncellulosic monosaccharide content by GC following derivatization, as described previously [36]. For measurements on X6, Xylazyme AX and Azo-wheat AX,
enzyme dilutions were chosen in the range where a linear
correlation between enzyme concentration and hydrolyzed
substrate was valid and therefore no substrate exhaustion

occurred.

Activity on X6
After 10 min of pre-incubation, 20 lm X6, 37.5 lm rhamnose (used as internal standard) and an appropriate
enzyme dilution were mixed in a final volume of 650 lL.
To ensure measurement at substrate saturation, enzyme
dilutions were chosen in such a way that, in the final sample, less than 30% of the X6 was hydrolyzed. After
60 min of incubation, the reaction was terminated. For
XBS samples, this was carried out by the addition of
20 lL of NaOH (2.00 m) and subsequent boiling of the
sample for 30 min. XAN samples were simply boiled for
30 min. Samples were filtered (Millex-GP, 0.22 lm, polyethersulfone; Millipore, Carrigtwohill, Ireland) before separation and quantification of the hydrolysis products by
high-performance anion exchange chromatography with
integrated pulsed amperometric detection on a CarboPac
PA-100 column (250 · 4 mm) performed on a Dionex
ICS-3000 system (Dionex Corporation). A linear gradient
of 0–125 mm sodium acetate in 100 mm sodium hydroxide
over 30 min was used for elution (1.0 mLỈmin)1). A standard solution containing xylooligosaccharides (xylose to
X6) and rhamnose was used to identify and quantify the
hydrolysis products. One unit of enzyme activity on X6
(X6U) corresponds to the enzyme concentration needed
for the formation of 1.0 lm xylotriose from excess X6
under the conditions of the assay.

FEBS Journal 278 (2011) 1098–1111 ª 2011 The Authors Journal compilation ª 2011 FEBS


S. Cuyvers et al.

Activity on Xylazyme AX

After 10 min of pre-incubation, a Xylazyme AX tablet was
added to 1.0 mL of an appropriate enzyme dilution. The
reaction was terminated by the addition of 10.0 mL of Tris
solution (1.0% w ⁄ v), vigorous vortex-mixing and immediate
filtration. E590 of the filtrate was measured. One unit of
enzyme activity on Xylazyme AX (XyU) corresponds to the
enzyme concentration required to obtain E590 = 1.0 after
subtraction of the control value (no enzyme) under the conditions of the assay.

Activity on Azo-wheat AX
An appropriate enzyme dilution and liquid Azo-wheat AX
substrate were pre-incubated separately for 10 min before
500 lL of the substrate was added to 500 lL of enzyme
solution. The reaction was terminated after 60 min by the
addition of 2.5 mL of ethanol and vigorous vortex-mixing.
The mixture was put on ice for 10 min before being centrifuged (3000 g for 10 min at 4 °C). Then, the E590 of the
supernatant was measured. One unit of enzyme activity on
Azo-wheat AX (AzU) corresponds to the enzyme concentration needed to obtain E590 = 1.0 after subtraction of the
control value (no enzyme) under the conditions of the assay
and the assumption that no substrate exhaustion occurs.

Solubilization of water-unextractable AX and OSX
Water-unextractable AX isolated from wheat flour and the
insoluble fractions isolated from oat spelt xylans were used
as substrates. To determine the solubilization of the substrates by XBS, XAN and its variants, 20 mg of substrate
was incubated with enzyme in a final volume of 7.80 mL
after 10 min of pre-incubation. During incubation, the samples were continuously stirred. For OSX, the reaction was
terminated by the addition of 230 lL of 4.00 m NaOH and
by separating the nonsolubilized from the solubilized material by ltration (MN615; Macherey-Nagel, Duren, Geră
many). To avoid alkaline solubilization of waterunextractable AX, the reaction with water-unextractable

AX was terminated by first separating the nonsolubilized
from the solubilized material by filtration and by then
immediately mixing the filtrate with 230 lL of 4.00 m
NaOH. The monosaccharide content in the filtrate was analyzed by GC following hydrolysis and derivatization, as
previously described by Gebruers et al. [36].

Temperature and pH optima
The optimal temperature for enzyme activity under the conditions of the activity measurement assays was determined
by measuring activity on Xylazyme AX, under the conditions described above, using different incubation tempera-

Secondary substrate binding in GH11 xylanases

tures, in the range 30–60 °C at temperature intervals of
5 °C. The optimal pH for enzyme activity was determined
by measuring activity on Xylazyme AX, under the conditions described above, in McIlvaine buffers with different
pH, in the pH range 4.5–7.5 for XBS and 2.5–6.5 for XAN
at intervals of 0.5 pH.

Binding affinity towards insoluble polymers
To study the binding affinity of enzymes towards insoluble
substrates, 1.0 mL of enzyme solution (91.4 · 10)10 m),
diluted in the same buffer as that used for activity measurements, was incubated for 10 min on ice with different
substrate concentrations. The 14 concentrations used varied in the range 0–40.0 mg for water-unextractable AX
and 0–30.0 mg for OSX. The mixtures were then centrifuged (13000 g for 2 min at 4 °C) and the residual activity
in the supernatants was measured with Xylazyme AX tablets, as described above. For affinity towards water-unextractable AX and OSX, the percentage of bound protein
was plotted as a function of the concentration of polymer
used for the incubation. The obtained graphs were fitted
in graphpad prism, version 3.03 (GraphPad Software
Inc., San Diego, CA, USA), with a ‘one site binding
model’ that results in an overall dissociation constant

(Kd): fraction bound protein = (maximum fraction bound
protein · [polysaccharide]) ⁄ (Kd + [polysacchaide]). For most
measurements, an apparent Kd value is presented because
the substrate concentration was a limiting factor in the
assay. Indeed, at too high substrate concentrations, it is
impossible to hydrate all material. To evaluate whether
the enzymes bind to cellulose, zymosan, paramylon, pustulan and barley b-glucan, a single substrate concentration
(25 mgỈmL)1) was tested.

Acknowledgements
The authors thank Nele Schoonens and Koen Vanderlinden for technical assistance. The ‘Fonds voor
Wetenschappelijk
Onderzoek-Vlaanderen’
(FWO,
Brussels, Belgium) is gratefully acknowledged for the
postdoctoral fellowship of E. Dornez. This study is
part of the Methusalem programme ‘Food for the
Future’ at the Katholieke Universiteit Leuven.

References
1 Boraston AB, Bolam DN, Gilbert HJ & Davies GJ
(2004) Carbohydrate-binding modules: fine-tuning polysaccharide recognition. Biochem J 382, 769–781.
2 Gilbert HJ (2010) The biochemistry and structural biology of plant cell wall deconstruction. Plant Physiol 153,
444–455.

FEBS Journal 278 (2011) 1098–1111 ª 2011 The Authors Journal compilation ª 2011 FEBS

1109

Secondary substrate binding in GH11 xylanases

S. Cuyvers et al.

3 Shoseyov O, Shani Z & Levy I (2006) Carbohydrate
binding modules: biochemical properties and novel
applications. Microbiol Mol Biol Rev 70, 283–295.
4 Hashimoto H (2006) Recent structural studies of carbohydrate-binding modules. Cell Mol Life Sci 63, 2954–
2967.
5 Gao PJ, Chen GJ, Wang TH, Zhang YS & Liu J (2001)
Non-hydrolytic disruption of crystalline structure of cellulose by cellulose binding domain and linker sequence
of cellobiohydrolase I from Penicillium janthinellum.
Acta Biochim Biophys Sin 33, 13–18.
6 Montanier C, van Bueren AL, Dumon C, Flint JE,
Correia MA, Prates JA, Firbank SJ, Lewis RJ, Grondin
GG, Ghinet MG et al. (2009) Evidence that family 35
carbohydrate binding modules display conserved specificity but divergent function. Proc Natl Acad Sci USA
106, 3065–3070.
7 Ezer A, Matalon E, Jindou S, Borovok I, Atamna N,
Yu ZT, Morrison M, Bayer EA & Lamed R (2008) Cell
surface enzyme attachment is mediated by family 37
carbohydrate-binding modules, unique to Ruminococcus
albus. J Bacteriol 190, 8220–8222.
8 Warren RAJ (1996) Microbial hydrolysis of polysaccharides. Annu Rev Microbiol 50, 183–212.
9 Nielsen MM, Bozonnet S, Seo ES, Motyan JA, Andersen JM, Dilokpimol A, Abou Hachem M, Gyemant G,
Naested H, Kandra L et al. (2009) Two secondary carbohydrate binding sites on the surface of barley alphaamylase 1 have distinct functions and display synergy in
hydrolysis of starch granules. Biochemistry 48, 7686–
7697.
10 De Vos D, Collins T, Nerinckx W, Savvides SN, Claeyssens M, Gerday C, Feller G & Van Beeumen J
(2006) Oligosaccharide binding in family 8 glycosidases:

crystal structures of active-site mutants of the beta-1,4xylanase pXyl from Pseudoalteromonas haloplanktis
TAH3a in complex with substrate and product. Biochemistry 45, 4797–4807.
11 Schmidt A, Gubitz GM & Kratky C (1999) Xylan binding subsite mapping in the xylanase from Penicillium
simplicissimum using xylooligosaccharides as cryoprotectant. Biochemistry 38, 2403–2412.
12 Vandermarliere E, Bourgois TM, Rombouts S, Van
Campenhout S, Volckaert G, Strelkov SV, Delcour
JA, Rabijns A & Courtin CM (2008) Crystallographic
analysis shows substrate binding at the -3 to +1
active-site subsites and at the surface of glycoside
hydrolase family 11 endo-1,4-beta-xylanases. Biochem J
410, 71–79.
13 Ludwiczek ML, Heller M, Kantner T & McIntosh LP
(2007) A secondary xylan-binding site enhances the catalytic activity of a single-domain family 11 glycoside
hydrolase. J Mol Biol 373, 337–354.
14 Ramasubbu N, Ragunath C & Mishra PJ (2003) Probing the role of a mobile loop in substrate binding and

1110

15

16

17

18

19

20

21

22

23

24

25
26

enzyme activity of human salivary amylase. J Mol Biol
325, 1061–1076.
Ragunath C, Manuel SGA, Venkataraman V, Sait
HBR, Kasinathan C & Ramasubbu N (2008) Probing
the role of aromatic residues at the secondary saccharide-binding sites of human salivary alpha-amylase in
substrate hydrolysis and bacterial binding. J Mol Biol
384, 1232–1248.
Nielsen MM, Seo ES, Dilokpimol A, Andersen J, Abou
Hachem M, Naested H, Willemoes M, Bozonnet S,
Kandra L, Gyemant G et al. (2008) Roles of multiple
surface sites, long substrate binding clefts, and carbohydrate binding modules in the action of amylolytic
enzymes on polysaccharide substrates. Biocatalysis Biotransformation 26, 59–67.
Mikami B, Adachi M, Kage T, Sarikaya E, Nanmori T,
Shinke R & Utsumi S (1999) Structure of raw starchdigesting Bacillus cereus beta-amylase complexed with
maltose. Biochemistry 38, 7050–7061.
Sevcik J, Hostinova E, Solovicova A, Gasperik J,
Dauter Z & Wilson KS (2006) Structure of the complex
of a yeast glucoamylase with acarbose reveals the presence of a raw starch binding site on the catalytic
domain. FEBS J 273, 2161–2171.

Allouch J, Helbert W, Henrissat B & Czjzek M (2004)
Parallel substrate binding sites in a beta-agarase suggest
a novel mode of action on double-helical agarose.
Structure 12, 623–632.
Przylas I, Terada Y, Fujii K, Takaha T, Saenger W &
Strater N (2000) X-ray structure of acarbose bound to
amylomaltase from Thermus aquaticus - implications for
the synthesis of large cyclic glucans. Eur J Biochem 267,
6903–6913.
Collins T, Gerday C & Feller G (2005) Xylanases, xylanase families and extremophilic xylanases. FEMS
Microbiol Rev 29, 3–23.
Moers K, Courtin CM, Brijs K & Delcour JA (2003)
A screening method for endo-beta-1,4-xylanase
substrate selectivity. Anal Biochem 319, 73–77.
Dornez E, Gebruers K, Delcour JA & Courtin CM
(2009) Grain-associated xylanases: occurrence, variability, and implications for cereal processing. Trends Food
Sci Technol 20, 495–510.
Moers K, Bourgois T, Rombouts S, Belien T, Van Campenhout S, Volckaert G, Robben J, Brijs K, Delcour JA
& Courtin CM (2007) Alteration of Bacillus subtilis
XynA endoxylanase substrate selectivity by site-directed
mutagenesis. Enzyme Microb Technol 41, 85–91.
Linder M & Teeri TT (1997) The roles and function of
cellulose-binding domains. J Biotechnol 57, 15–28.
Sun JL, Sakka K, Karita S, Kimura T & Ohmiya K
(1998) Adsorption of Clostridium stercorarium xylanase
A to insoluble xylan and the importance of the CBDs
to xylan hydrolysis. J Ferment and Bioeng 85, 63–
68.

FEBS Journal 278 (2011) 1098–1111 ª 2011 The Authors Journal compilation ª 2011 FEBS

S. Cuyvers et al.

27 Ali MK, Hayashi H, Karita S, Goto M, Kimura T,
Sakka K & Ohmiya K (2001) Importance of the carbohydrate-binding module of Clostridium stercorarium
Xyn10B to xylan hydrolysis. Biosci Biotechnol Biochem
65, 41–47.
28 Mangala SL, Kittur FS, Nishimoto M, Sakka K, Ohmiya K, Kitaoka M & Hayashi K (2003) Fusion of family VI cellulose binding domains to Bacillus halodurans
xylanase increases its catalytic activity and substratebinding capacity to insoluble xylan. J Mol Catal B-Enzym 21, 221–230.
29 Kramhoft B, Bak-Jensen KS, Mori H, Juge N, Nohr J
& Svensson B (2005) Involvement of individual subsites
and secondary substrate binding sites in multiple attack
on amylose by barley alpha-amylase. Biochemistry 44,
1824–1832.
30 Sapag A, Wouters J, Lambert C, de Ioannes P, Eyzaguirre J & Depiereux E (2002) The endoxylanases from
family 11: computer analysis of protein sequences
reveals important structural and phylogenetic relationships. J Biotechnol 95, 109–131.
31 Belien T, Verjans P, Courtin CM & Delcour JA (2008)
ă
Phage display based identification of novel stabilizing
mutations in glycosyl hydrolase family 11 Bacillus subtilis endoxylanase XynA. Biochem Biophys Res Commun
368, 74–80.
32 Rombouts S (2007) Molecular and mutational analysis
of a wheat thaumatin-like xylanase inhibitor (TLXI).
PhD thesis, pp. 186. Katholieke Universiteit Leuven,
Leuven (Belgium).
33 Pollet A, Vandermarliere E, Lammertyn J, Strelkov SV,
Delcour JA & Courtin CM (2009) Crystallographic and
activity-based evidence for thumb flexibility and its relevance in glycoside hydrolase family 11 xylanases. Proteins Struct Funct Bioinformat 77, 395–403.

34 Rombouts I, Lamberts L, Celus I, Lagrain B, Brijs K &
Delcour JA (2009) Wheat gluten amino acid composi-

Secondary substrate binding in GH11 xylanases

tion analysis by high-performance anion-exchange chromatography with integrated pulsed amperometric
detection. J Chromatogr A 1216, 5557–5562.
35 Saulnier L, Sado PE, Branlard G, Charmet G & Guillon F (2007) Wheat arabinoxylans: exploiting variation
in amount and composition to develop enhanced varieties. J Cereal Sci 46, 261–281.
36 Gebruers K, Courtin CM & Delcour JA (2009) Quantification of arabinoxylans and their degree of branching
using gas chromatography. In Healthgrain Methods –
Analysis of Bioactive Components in Small Grain Cereals
(Shewry PR & Ward JL eds), pp. 177–189. AACC
International, Inc., Minnesota.
37 Wallace AC, Laskowski RA & Thornton JM (1995)
Ligplot – a program to generate schematic diagrams of
protein ligand interactions. Protein Eng 8, 127–134.

Supporting information
The following supplementary material is available:
Fig. S1. Enzyme stability under the conditions of the
activity assays.
Fig. S2. Purity of XBS, XAN and their mutant variants as verified by SDS ⁄ PAGE and silver staining.
Table S1. Summary of template DNA and oligonucleotide primers used in site-directed mutagenesis for
the genetic engineering of XBS and XAN.
This supplementary material can be found in the
online version of this article.
Please note: As a service to our authors and readers,
this journal provides supporting information supplied
by the authors. Such materials are peer-reviewed and

may be re-organized for online delivery, but are not
copy-edited or typeset. Technical support issues arising
from supporting information (other than missing files)
should be addressed to the authors.

FEBS Journal 278 (2011) 1098–1111 ª 2011 The Authors Journal compilation ª 2011 FEBS

1111