The results showed that E190 favors the binding of the initial portion of alkyl-type aglycones up to the sixth methylene group and also the first glucose unit of oligosaccharidic aglycone
Trang 1residues forming the substrate aglycone-binding site
of a b-glycosidase
Lu´cio M F Mendonc¸a and Sandro R Marana
Departamento de Bioquı´mica, Instituto de Quı´mica, Universidade de Sa˜o Paulo, Sa˜o Paulo, Brazil
The b-glycosidases from family 1 of the glycoside
hydrolases are widely distributed among living
organ-isms, being found in bacteria, archea and eukaria
These enzymes are involved in a high diversity of
physi-ological roles [1,2] b-glycosidases catalyze the
hydro-lytic removal of the monosaccharide from the
non-reducing end of b-glycosides [2,3] Their active site
may be divided into subsites, which are of sufficient size
to bind a monosaccharide unit The monosaccharide
forming the non-reducing end of the substrate, called
glycone, is bound at subsite)1, whereas the remaining
part of the substrate, called aglycone, interacts with the
aglycone-binding site, which may be composed of
several subsites identified by positive numerals The substrate is cleaved between subsites)1 and +1 [4] b-glycosidases are active upon a broad range of sub-strates, as evidenced by a total of 15 different EC numbers grouped in family 1 of the glycoside hydro-lases Fucose, glucose, galactose, mannose, xylose, 6-phospho-glucose and 6-phospho-galactose are recog-nized by the b-glycosidase subsite)1 Nevertheless, the diversity of aglycones is higher, including monosaccha-rides, oligosaccharides and aryl and alkyl moieties [2] Furthermore, the aglycone specificity is an important factor for determining the physiological functions of b-glycosidases
Keywords
aglycone; catalysis; glycoside hydrolase;
specificity; b-glycosidase
Correspondence
S R Marana, Departamento de Bioquı´mica,
Instituto de Quı´mica, Universidade de Sa˜o
Paulo, CP 26077, Sa˜o Paulo, 05513-970 SP,
Brazil
Fax: +55 11 3815 5579
Tel: +55 11 3091 3810
E-mail: srmarana@iq.usp.br
(Received 1 February 2008, revised 4 March
2008, accepted 13 March 2008)
doi:10.1111/j.1742-4658.2008.06402.x
The relative contributions to the specificity and catalysis of aglycone, of residues E190, E194, K201 and M453 that form the aglycone-binding site
of a b-glycosidase from Spodoptera frugiperda (EC 3.2.1.21), were investi-gated through site-directed mutagenesis and enzyme kinetic experiments The results showed that E190 favors the binding of the initial portion of alkyl-type aglycones (up to the sixth methylene group) and also the first glucose unit of oligosaccharidic aglycones, whereas a balance between interactions with E194 and K201 determines the preference for glucose units versus alkyl moieties E194 favors the binding of alkyl moieties, whereas K201 is more relevant for the binding of glucose units, in spite of its favorable interaction with alkyl moieties The three residues E190, E194 and K201 reduce the affinity for phenyl moieties In addition, M453 favors the binding of the second glucose unit of oligosaccharidic aglycones and also of the initial portion of alkyl-type aglycones None of the residues investigated interacted with the terminal portion of alkyl-type aglycones It was also demonstrated that E190, E194, K201 and M453 similarly contrib-ute to stabilize ES Their interactions with aglycone are individually weaker than those formed by residues interacting with glycone, but their joint catalytic effects are similar Finally, these interactions with aglycone
do not influence glycone binding
Abbreviations
BglB, b-glycosidase from Paenibacillus polymyxa; SbDhr1, b-glycosidase from Sorghum bicolor; Sfbgly, b-glycosidase from
Spodoptera frugiperda; ZmGlu1, b-glycosidase from Zea mays.
Trang 2Structural studies of complexes between
b-glycosid-ases and substrates or inhibitors revealed residues that
are present in subsite )1 and interact through
hydro-gen bonds with the glycone hydroxyls In addition, the
role of these interactions and residues in determining
glycone specificity has been characterized through
site-directed mutagenesis, enzyme kinetics and
bio-energetics [5–15]
Previous studies using oligocellodextrins showed
that diverse b-glycosidases have a different number
of subsites forming their aglycone-binding sites [15]
Nevertheless, just recently the structures of
b-glyco-sidases (ZmGlu1 from Zea mays, SbDhr1 from
Sorghum bicolor and BglB from Paenibacillus
poly-myxa) containing ligands in their aglycone-binding
site were obtained [13,16,17] These data revealed
that the subsite +1 is formed by two walls that
embrace the aglycone moiety Subsites +1 of
ZmGlu1 and BglB are narrow, whereas subsite +1
of SbDhr1 has a wider opening [13,17,18] One of
these walls (also called the basal platform) is formed
by the side chain of a conserved tryptophan (W378
in ZmGlu1, W376 in SbDhr1 and W328 in BglB),
whereas the residues forming the other wall (also
called the ceiling) are highly variable among these
enzymes In ZmGlu1 the subsite +1 is formed by
bulky and apolar amino acid residues T194, F198,
F205 and F466, whereas T192, V196, L203 and S462
are found in subsite +1 of SbDhr1 [13] These
residues correspond to C170, L174, H181 and A411
in BglB, which, together with Y169, N223, E225,
Q316 and W412, form subsite +1 Data from the
BglB–cellotetraose complex also revealed the residues
forming subsites +2 and +3 [13]
The type of residues forming subsite +1 of
ZmGlu1 indicate a prevalence of hydrophobic
interac-tions with the aglycone, which are interacinterac-tions that
present less restriction about the positioning of its
participant Hence, this could explain the broad
agly-cone specificity of ZmGlu1 On the other hand, the
presence of hydrogen bond-forming residues would
determine a more restricted aglycone positioning in
SbDhr1, which is only active on dhurrin [13,16] In
subsite +1 of BglB, which is active on
oligocellodext-rins, in spite of the presence of hydrophobic residues
L174 and A411, a network of hydrogen bonds hold
glucose units at subsites +1, +2 and +3 [17]
Therefore, in addition, to facilitate the
identifica-tion of amino acid residues composing the
aglycone-binding site, the structural data of b-glycosidase
complexes have been used to infer the molecular basis
of the specificity for aglycone Indeed, the balance
between hydrophobic interactions and hydrogen
bonds seems to be important for the aglycone speci-ficity However, this model, which is a general description of the interactions involved in the recog-nition of aglycone, is not complete, as demonstrated
by the site-directed mutagenesis experiments intending
to exchange the aglycone specificity between SbDhr1 and ZmGlu1 [15,19] Therefore, this model should be developed to establish the contribution of each resi-due in the aglycone-binding site to the binding of diverse types of aglycones Furthermore, these data may be of particular importance in understanding the physiological role of b-glycosidases and in designing inhibitors
In addition, another important issue in understand-ing the aglycone specificity is the contribution of the interactions with the aglycone to the catalysis in b-glycosidases Mutational studies with ZmGlu1 and SbDhr1 indicate that residues forming subsite +1 con-tribute to the catalysis, probably by stabilizing the ES complex of the glycosylation step [15,19] Besides that, structural data from the complexes of ZmGlu1 and SbDhr1 with their natural substrates (dimboa-glc and dhurrin, respectively) and an inhibitor (glucotetrazole) suggested that the interactions with the aglycone could affect the glycone positioning within subsite )1 [13] Nevertheless, details about the role of different resi-dues of the aglycone-binding site in the stabilization
of ESand the interdependence between the binding of aglycone and the positioning of glycone in subsite )1 are not known
In this study, the role in the substrate specificity and catalysis of four residues (E190, E194, K201 and M453) forming the aglycone-binding site of a digestive b-glycosidase from Spodoptera frugiperda (fall army-worm) (Sfbgly; EC 3.2.1.21; GenBank accession no.: AF052729) was evaluated through site-directed muta-genesis and enzyme kinetic experiments The roles of E190, E194, K201 and M453 in aglycone binding were characterized using competitive inhibitors with different types of aglycones (oligosaccharides, and alkyl and phenyl moieties), whereas p-nitrophenyl b-glycosides (fucosides, glucosides, galactosides and xylosides) were used to evaluate the catalytic contri-bution of E190, E194, K201 and M453 Sfbgly, which
is classified in family 1 of the glycoside hydrolases, has an active site composed of four subsites ()1, +1, +2 and +3) [20] The interactions formed between the substrate glycone and residues Q39 and E451, which are part of the Sfbgly active site, explain the preference of this enzyme for b-glucosides, b-galacto-sides and b-fucob-galacto-sides [10,14] Nevertheless, the molec-ular basis of the broad aglycone specificity of Sfbgly had not been studied previously
Trang 3Results and Discussion
Role of residues E190, E194, K201 and M453 in
aglycone binding
The spatial structures of ZmGlu1, SbDhr1 and BglB
revealed groups of amino acid residues that formed the
binding site of the substrate aglycone These
aglycone-binding sites share a common structural element, a
basal platform formed by a tryptophan residue (W378
in ZmGlu1, W376 in SbDhr1 and W328 in BglB), but
they also have a variable portion (called the ceiling)
This portion is formed by T194, F198, F205 and F466
in ZmGlu1, and by V196, L203 and S462 in SbDhr1
These residues correspond to C170, L174, H181 and
A411 in BglB, which, together with Y196, N223, E225,
Q316 and W412, form the aglycone-binding site
(Fig 1) Hence, these residues are potentially involved
in aglycone binding [13,17] In addition, other residues
form a ‘layer’ that helps in the positioning of the resi-dues that interact directly with the aglycone (basal platform and ceiling) [15] Nevertheless, the relative contributions of the aglycone-binding residues in the substrate binding and catalysis are not known for any b-glycosidase
Sequence alignment and structural comparison indi-cated that residues E190, E194, K201 and M453, which correspond to T194, F198, F205 and F466 in ZmGlu1, may be part of the aglycone-binding site of Sfbgly (Fig 1) Thus, in order to establish the relative contribution of these residues to the interaction with different types of aglycone, they were replaced through site-directed mutagenesis, generating the mutants E190A, E190Q, E194A, K201A, K201F and M453A Replacements with A were made to remove side chains that could interact with the aglycone Mutations E190Q and K201F were planned taking into account the residues found in the b-glycosidase from
E
D C
Fig 1 Comparison of the aglycone-binding site of some b-glycosidases (A) Aglycone-binding site of ZmGlu1 The aglycone and glycone of the substrate (dimboa-Glc) are also indicated (B) Aglycone-binding site of Sfbgly Residues E190, E194, K201 and M453 are structurally equivalent to T194, F198, F205 and F466 The substrate (dimboa-Glc) was included to facilitate the visualization of the active site (C) Agly-cone-binding site of SbDhr1, including the substrate dhurrin (D) AglyAgly-cone-binding site of BglB, including the inhibitor thiocellobiose (E) Sequence alignment of some b-glycosidases showing residues (boxes) forming the aglycone-binding site.
Trang 4Tenebrio molitor (AF312017) [21], which is closely
related to Sfbgly (45% identity; 65% similarity) These
insect b-glycosidases were previously characterized and
showed differences in their aglycone specificity [20,21]
The mutant enzymes were expressed in bacteria and
purified through hydrophobic chromatography
(sup-plementary Fig S1) Then, dissociation constants (Ki)
for the complex between these mutant enzymes and
different competitive inhibitors were determined and
compared with those from the wild-type Sfbgly
(Table 1)
A series of alkyl b-glucosides (hexyl b-glucoside to
nonyl b-glucoside) was initially used to characterize
the Sfbgly mutants (Table 1) Molecular models of
these alkyl b-glucosides indicate that each four
methy-lene groups of their aglycone may cover an area
simi-lar to that of one glucose unit Thus, this series of
alkyl b-glucosides is useful for probing mutational
effects along the aglycone-binding site
Based on the Kivalues (Table 1), the binding energy
(DG0) for the pentyl moiety formed by the methylene
groups 2 to 6 of an alkyl-type aglycone was calculated
by subtracting the DG0 for methyl b-glucoside from
that for hexyl b-glucoside These data indicated a
favorable binding (DG0=)2.9 kJÆmol)1) between the
pentyl moiety and the wild-type Sfbgly, whereas the
energy of that interaction was reduced by mutations
E194A, K201A and M453A Mutation K201F did not
affect the binding of that aglycone moiety (Fig 2)
However, the binding of the pentyl moiety in the
agly-cone-binding site was unfavorable for the mutants
E190A (DG0= +3.5 kJÆmol)1) and E190Q (DG0=
+3.3 kJÆmol)1) (Fig 2) Indeed, these mutations
caused the most drastic effects on the binding of the
initial pentyl moiety (converting a favorable interaction
to an unfavorable one) In addition, the binding energy
of the terminal portion of the alkyl-type aglycone (from the seventh to the ninth methylene group) was determined by subtracting the DG0 for hexyl b-gluco-side from that for nonyl b-glucob-gluco-side (Fig 2) The favorable interaction (DG0=)4.9 kJÆmol)1) observed for the wild-type Sfbgly was not significantly affected
by any of the mutations, except for K201F, which increased the energy of that interaction (DG0= )8.1 kJÆmol)1)
Table 1 Inhibition data for the wild-type and mutant Sfbgly All inhibitors were simple linear competitive K i values were calculated using
ENZFITTER software The data were obtained with at least five different concentrations of substrate (methylumbeliferyl b-glucoside) in the presence of at least five different concentrations of inhibitor heptylbglc, heptyl b-glucoside; hexylbglc, hexyl b-glucoside; methylbglc, methyl b-glucoside; nonylbglc, nonyl b-glucoside; octylbglc, octyl b-glucoside; phenylbglc, phenyl b-glucoside; wt, wild-type.
Enzyme
Inhibitor K i (m M )
Cellotriose 2.3 ± 0.2 0.070 ± 0.004 0.41 ± 0.01 0.09 ± 0.01 0.04 ± 0.01 8.6 ± 0.3 0.27 ± 0.01 Cellotetraose 2.4 ± 0.1 0.14 ± 0.01 0.54 ± 0.03 0.21 ± 0.01 0.25 ± 0.03 19.6 ± 0.7 0.20 ± 0.02 Cellopentaose 2.0 ± 0.1 0.20 ± 0.01 2.3 ± 0.1 0.26 ± 0.01 0.30 ± 0.02 16.2 ± 0.3 0.25 ± 0.02
Octylbglc 10.3 ± 0.6 1.49 ± 0.03 2.52 ± 0.09 1.30 ± 0.06 0.42 ± 0.03 11.0 ± 0.2 0.44 ± 0.02 Nonylbglc 5.9 ± 0.2 0.83 ± 0.02 1.16 ± 0.04 0.50 ± 0.03 0.13 ± 0.01 5.7 ± 0.3 0.22 ± 0.03
Fig 2 Binding energies for the interactions between different types of aglycone and the wild-type and mutant Sfbgly Negative values correspond to favorable interactions, whereas positive val-ues represent unfavorable binding Black bars indicate an aglycone formed by a phenyl moiety; white bars indicate the initial portion
of an alkyl-type aglycone (pentyl moiety formed by methylene groups 2 to 6); and gray bars indicate the terminal portion of an alkyl-type aglycone (moiety formed by methylene groups 7 to 9).
wt, wild-type.
Trang 5Therefore, these results indicate that residues E190,
E194, K201 and M453 are part of the
aglycone-bind-ing site of Sfbgly and interact with the initial pentyl
moiety formed by methylene groups 2 to 6 of the
alkyl-type aglycones On the other hand, the binding
of alkyl moieties in the more external portion of the
aglycone-binding site, which is occupied by methylene
groups 7 to 9, is not influenced by those residues
The binding energies for an aglycone formed by a
phenyl moiety were calculated using the Ki values for
phenyl b-glucoside and methyl b-glucoside (Table 1)
These data showed an unfavorable binding of the
phe-nyl-type aglycone with the wild-type and all-mutant
Sfbgly (Fig 2) Mutations E190 and E194A resulted in
only small reductions (about 40%) of that unfavorable
interaction, but they did not convert it to a favorable
binding Thus, by contrast with the binding of
alkyl-type aglycones, single mutations were unable to cause
large changes in the interaction of phenyl-type
agly-cones with Sfbgly
In addition to the analysis of the interaction with
alkyl-type and phenyl-type aglycones, the dissociation
constants (Ki) of the complexes between
oligocellodext-rins and the wild-type and mutant Sfbgly were also
determined (Table 1) Based on these results, the
bind-ing energies for each glucose unit of the aglycone
(DG0⁄ glucose unit) were calculated (Fig 3) For
orien-tation purposes, the glucose unit forming the
non-reducing end of the aglycone was called the ‘first
glucose unit’ and the other glucose units were
sequen-tially named in the direction of the reducing end of the
aglycone Thus, mutations E190A, E194A, K201A and K201F affected the binding of the first glucose unit of the aglycone (DDG0= +3.3, )3.9, +5.5, +4.1 kJÆmol)1, respectively), whereas M453A did not influence that interaction Mutations M453A, K201A and K201F increased the affinity for the second glucose unit of the aglycone (DDG0=)4.2, )7.3, )10 kJÆmol)1, respec-tively), whereas E190A and E194A did not affect that interaction It is noteworthy that E190 and E194 inter-act with the first glucose unit of the aglycone, whereas M453 interacts with the second glucose unit of the aglycone However, K201 interacts with both glucose units In addition, mutations E194A, K201A, K201F and M453A also significantly affected the binding of the third glucose unit of the aglycone (DDG0= +2.5, +1.4, +5.3, +2.8 kJÆmol)1, respectively) The fourth glucose unit of aglycone is not bound by any mutant sfbgly The same result has been previously observed for the wild-type sfbgly The exception to these trends
is the mutant E190Q, which showed a very low affinity for any glucose unit of the aglycone
Previously, it has been proposed that each glucose unit of the aglycone interacts with a specific portion of the aglycone-binding site, which was named subsite [4] Taking into account this definition and combining the results presented above it may be proposed that resi-dues E190 and E194 are positioned in an internal region of the aglycone-binding site, probably subsite +1, because they influence the binding of the first glu-cose unit of the aglycone On the other hand, M453 may be positioned in subsite +2 because these residues influence the binding of the second glucose unit of the aglycone Residue K201 may be part of both subsites +1 and +2 or may be located in the interface between them as it simultaneously affects the binding of the first and the second glucose unit of the aglycone This proposal is in agreement with the binding data for alkyl-type aglycones (Fig 2), because the pentyl moiety formed by the methylene groups 2 to 6 is large enough
to fill subsite +1 and to occupy part of subsite +2
As a result, residues E190, E194, K201 and M453 can still interact with that initial portion of an alkyl-type aglycone, even though they interact with different units
of an oligosaccharidic aglycone
Figure 3 also showed that mutations K201A and K201F affected the interactions with the first three glu-cose units of the aglycone, whereas M453A caused modifications in the interaction with the second and third glucose units This suggests that modifications in the interactions with a specific glucose unit of the agly-cone may alter the conformation and⁄ or freedom of the other glucose units, which are part of the aglycone, affecting its interactions with the aglycone site
Fig 3 Binding energies for the interactions between glucose units
of the aglycone and the wild-type and mutant Sfbgly Negative
values represent favorable interactions, whereas positive values
correspond to unfavorable binding Glucose units were numbered
from the non-reducing to the reducing end of the aglycone wt,
wild-type.
Trang 6Following the characterization of the binding of
different types of aglycone, a comparative analysis of
the mutational effect on their binding revealed that
residues E190, E194, K201 and M453 have different
roles in the determination of the Sfbgly specificity for
the substrate aglycone Thus, although the bulky and
apolar phenyl moieties may form favorable
interac-tions with the basal platform (W378 in Sfbgly), the
binding of these moieties is dominated by unfavorable
interactions with the polar residues E190, E194 and
K201, whereas M453 has little influence on this The
replacement of E190, E194 and K201 with A, which
has a small side chain, reduced those unfavorable
interactions (Fig 2), and, interestingly, the more
com-pact alkyl moieties overcame those unfavorable
inter-actions Indeed, residues E190, E194, K201 and M453
contributed to the favorable binding of the initial
pen-tyl moiety of the aglycone Of these residues, E190 is
the most important for that interaction because
muta-tions E190A and E190Q resulted in the largest
decrease of affinity for the pentyl moiety (DDG0= 6.4
and 6.2 kJÆmol)1, respectively); E194, K201 and M453
contribute similarly to the binding of pentyl moieties
given that the replacement of those residues by A
resulted in similar decrease (when taking into account
the experimental errors) of the affinity (DDG0
2 kJÆmol)1) for pentyl moieties However, these same
residues give different contributions to the binding of
the glucose units of the aglycone Thus, M453 does
not affect the binding of the first glucose unit
(DDG0 0; Fig 4); in fact, it interacts with the second
glucose unit of the aglycone (Fig 3) E194 has an
unfavorable interaction with the first glucose unit,
given that replacement of E194 with A increased, by
3.9 kJÆmol)1, the affinity for that glucose unit (Fig 4)
Conversely, E190 and K201 may form interactions
(probably hydrogen bonds) that are important for the binding of the first glucose units of the aglycone because their replacement with A caused a large decrease in the affinity for glucose (DDG0 = +3.9 and +5.6 kJÆmol)1, respectively; Fig 4) In agreement, K201 corresponds to H181 in BglB, a residue that inter-acts through a hydrogen bond with glucose units at sub-site +1 of that enzyme [17] Moreover, E190 corresponds to T194 in ZmGlu1, which may form a hydrogen bond with the aglycone of dimboa-Glc
It is noteworthy that, regarding aglycone binding, mutation K201F drastically reduced the glucose binding without affecting the affinity for alkyl moieties In addi-tion, mutation K201F also increased the affinity for the terminal portion of alkyl-type aglycones (Fig 2) Resi-due K201 is replaced by F in the b-glycosidases from Tenebrio molitor(Tmbgly; AF312017) and Cavia porce-lus (Cpbgly; U50545) Moreover, F179 is part of the aglycone-binding site of Cpbgly [22] In order to com-pare the aglycone-binding sites of these b-glycosidases, the effect of the aglycone size on the binding of several alkyl b-glucosides was measured Interestingly, these binding data showed that each aglycone methylene moiety of the aglycone increases in 1.6 kJÆmol)1 the affinity of the wild-type Sfbgly for alkyl b-glycoside (Fig 5), whereas mutation K201F increased that parameter to 2.6 kJÆmol)1, which was similar to that of Cpbgly (3.0 kJÆmol)1) and higher than that of Tmbgly (1.0 kJÆmol)1) [21,23] Therefore, residues in posi-tion 201 (and its equivalent in other b-glycosidases) may be an important factor in the determination of sub-site +1 preference for alkyl-type aglycones
In summary, the triad formed by E190, E194 and K201 controls the specificity for the substrate
Fig 4 Mutational effect (DDG 0 ) on the binding energy of alkyl
moi-eties and glucose units of the aglycone Black bars correspond to
alkyl moieties and white bars correspond to glucose units Positive
DDG 0 values correspond to a decrease in affinity, whereas negative
DDG 0 values represent an increase in affinity.
–10
–15
–20
–25
9
8
7
6
5
Aglycone carbon number
Fig 5 Effect of the aglycone size on the binding between alkyl b-glucosides and the wild-type and mutant Sfbgly ( ), wild-type; ( ), mutant K201F.
Trang 7aglycone This triad reduces the affinity for phenyl
moieties Residue E190 drives the binding of both alkyl
moieties (the initial portion up to the sixth methylene
group) and the first glucose unit of oligosaccharidic
aglycones, whereas a balance between interactions with
E194 and K201 determines the specificity for glucose
units versus alkyl moieties E194 favors the binding of
alkyl moieties, whereas K201 is more relevant for the
binding of glucose units, in spite of its favorable
interac-tion with alkyl moieties Therefore, the replacement of
E194 and K201 may be an important mechanism for
changing the specificity of Sfbgly for aglycone
The relative contribution to aglycone specificity of
residues corresponding to E190, E194, K201 and
M453 is probably different in other b-glycosidases, especially in view of the broad variety of residues occupying those positions Nevertheless, the differen-tial participation of several residues in the definition of the aglycone specificity may be a general trend
Catalytic role of residues E190, E194, K201 and M453 from the Sfbgly aglycone-binding site Steady-state kinetic parameters for the hydrolysis of p-nitrophenyl b-glycosides and methylumbelliferyl b-glucosides by the wild-type and mutant Sfbgly were determined (Table 2) In order to evaluate the role of residues E190, E194, K201 and M453 on the catalytic
Table 2 Steady-state kinetic parameters for the hydrolysis of p-nitrophenyl b-glycosides and methylumbeliferyl b-glucoside by wild-type and mutant Sfbgly Experiments were carried out using at least 10 different substrate concentrations Parameters were calculated using ENZFITTER
software MUbglc, methylumbeliferyl b-glucoside; NPbfuc, p-nitrophenyl b-fucoside; NPbgal, p-nitrophenyl b-galactoside; NPbglc, p-nitrophe-nyl b-glucoside; NPbxyl, p-nitrophep-nitrophe-nyl b-xyloside.
cat ⁄ K m (%)
Trang 8steps, these data were used to calculate the effect of
mutations of these residues on the stability of the ES
complex for p-nitrophenyl b-glucoside, p-nitrophenyl
b-fucoside, p-nitrophenyl b-galactoside and
p-nitrophe-nyl b-xyloside hydrolysis Figure 6 shows that all
mutations destabilized the ES complex for all
sub-strates tested, except mutation K201F, which resulted
in stabilization of the ES complex for p-nitrophenyl
b-galactoside and p-nitrophenyl b-xyloside hydrolysis
Destabilization of ES indicates a reduction in the rate
of substrate hydrolysis, whereas the opposite is valid
for the stabilization of ES Hence, the existence of
these mutational effects on ES stability indicate that
residues E190, E194, K201 and M453 participate in
the catalysis Considering that the catalytic mechanism
of b-glycosidases is divided into two steps
(glycosyla-tion and deglycosyla(glycosyla-tion), and that the aglycone is
released in the first step [2], the interactions of these
residues with the aglycone probably contribute to
sta-bilizing the EScomplex of the glycosylation step
Interestingly, a comparison of these mutational
effects with those resulting from the mutation of
resi-dues Q39 and E451, which are involved in the glycone
binding in Sfbgly [10], showed that the ‘ES
destabiliz-ing effects’ resultdestabiliz-ing from mutations of the
aglycone-binding residues are usually smaller than those from
mutations at subsite )1 (Q39A and E451A) (Fig 6)
Nevertheless, the total effect of the mutations of the
aglycone-binding residues (20, 20 and 16 kJÆmol)1 for
p-nitrophenyl b-fucoside, p-nitrophenyl b-glucoside
and p-nitrophenyl b-galactoside hydrolysis,
respec-tively) is similar to the combined effect of mutations
Q39A and E451A (33, 31 and 32 kJÆmol)1 for
p-nitro-phenyl b-fucoside, p-nitrophenyl b-glucoside and
p-nitrophenyl b-galactoside hydrolysis, respectively)
Hence, the energy of the non-covalent interactions
available to stabilize ES tend to be concentrated in a
few residues in subsite)1 (for instance Q39 and E451),
whereas these ‘ES-stabilizing’ interactions are more homogeneously distributed among the aglycone-bind-ing residues Besides, as the contributions of the agly-cone-binding residues to the stability of ES are relevant, they should be considered in the design of b-glycosidase inhibitors
Figure 6 also shows that the mutational effects (DDG) tend to be similar regardless of the substrate when considering the aglycone-binding residues This trend is not observed for mutations of residues (Q39 and E451) directly involved in the binding of glycone,
or for mutation K201F Taking into account that the substrates share a common aglycone, but differ in the glycone, the results obtained for the mutants E190A, E194A, K201A and M453A are those expected for res-idues that interact with the substrate aglycone but do not participate in the binding of the substrate glycone Therefore, an implication of these results is that the interactions with aglycone do not affect the binding of the glycone within the Sfbgly active site
This hypothesis was further investigated by deter-mining the influence on the glycone binding of the Sfbgly interactions with the aglycone Thus, the effect
of the alteration of the glycone structure in the stabil-ity of the ES complex (DDG) for the wild-type and mutant Sfbgly was determined using the kcat⁄ Km for the hydrolysis of p-nitrophenyl b-glucoside, p-nitrophe-nyl b-fucoside and p-nitrophep-nitrophe-nyl b-galactoside These DDG values represent the manner in which the inter-actions within subsite )1 are affected by the modifica-tion of the glycone structure, in particular the spatial positioning of its hydroxyl groups Hence, two b-glyco-sidases, presenting exactly the same interaction pattern within subsite )1, should be equally affected by the alteration of the substrate glycone generating the same DDG Thus, a plot of these DDG values for a pair of b-glycosidases presenting identical subsites )1 would
be a line showing a correlation coefficient and slope equal to 1 Indeed, comparison between wild-type and mutant Sfbgly using such plots revealed correlation coefficients and slopes close to 1 for all mutants except K201F, which presented a correlation coefficient of 0.58 (Table 3) These results indicate that despite the mutations in the aglycone-binding site, the interactions with the glycone remained very similar for the mutants
of Sfbgly analyzed, except for K201F Therefore, at least for the Sfbgly activity upon p-nitrophenyl b-gly-cosides, the interactions with aglycone do not affect glycone binding In addition, the catalytic contribu-tions of residues E190, E194, K201 and M453 result exclusively from interactions with the aglycone that stabilize ES Conversely, mutation K201F influenced the interaction pattern of the p-nitrophenyl
b-glyco-Fig 6 Mutational effect on the stability of ES (DDG ) involving
dif-ferent substrates Dark gray bars, p-nitrophenyl b-fucoside; white
bars, p-nitrophenyl b-glucoside; black bars, p-nitrophenyl
b-galacto-side; light gray bars, p-nitrophenyl b-xyloside.
Trang 9sides within subsite )1 through modifying the
interac-tions with their aglycones, suggesting that in mutant
K201F the binding of glycone and aglycone are
inter-dependent These results are not observed for mutant
K201A, revealing that such mutational effects are
related to the type of residue occupying position 201
Interestingly, the residue corresponding to K201 in
BglB is H181, which delineates a channel that could be
the pathway for the release of the aglycone in the
glyco-sylation step and the entrance of the water molecule
involved in the hydrolysis of the covalent intermediate
[17] Thus, the unusual effects of the K201F mutation
on the Sfbgly catalysis could be the result of a drastic
alteration in that putative channel, which would change
the rate of the glycosylation and deglycosylation steps
Materials and methods
Site-directed mutagenesis
Site-directed mutagenesis experiments were performed as
described in the instructions of the kit ‘QuikChange
site-directed mutagenesis’ (Stratagene, La Jolla, CA, USA) using
a plasmid pT7-7 [24] encoding the wild-type Sfbgly as the
template A pair of mutagenic primers was used to produce
each desired mutation Only the primers corresponding to
the sense strand are listed, as follows: mutation E190A,
5¢-caacgagcctagagcgatttgctttgagg-3¢; mutation E190Q, 5¢-caa
cgagcctagacagatttgctttgagg-3¢; mutation E194A, 5¢-gagagattt
gctttgcgggttatggatctgc-3¢; mutation K201A, 5¢-gttatggatctgct
accgcggctccgatcctaaacg-3¢; mutation K201F, 5¢-ggttatggatctg
ctacttcgctccgatcctaaacgc-3¢; and mutation M453A, 5¢-ggacaa
ctttgaatgggcggagggttatattgag-3¢ The incorporation of
muta-tions was verified by DNA sequencing
Expression of recombinant Sfbgly
BL21 DE3 cells (Novagen, Darmstadt, Germany) were
transformed with pT7-7 plasmids encoding the wild-type
and mutant Sfbgly Transformed bacteria were cultured
(37C, 150 r.p.m.) in 500 mL of Luria–Bertani (LB) broth
containing carbenecillin (50 lgÆmL)1) until an attenuance
(D) of 0.6–0.8 at 600 nm was reached Then, the production
of recombinant Sfbgly was induced (25C, 6 h, 150 r.p.m.)
by adding 1 mm isopropyl thio-b-d-galactoside The induced bacteria were harvested by centrifugation (7000 g,
30 min, 4C) and resuspended in 50 mm Hepes buffer containing 150 mm NaCl, 0.02% (w⁄ v) of hen egg-white lysozyme and 0.1% (v⁄ v) of Triton X-100 This suspension was incubated at room temperature for 45 min with slow shaking at 30 r.p.m Then, the suspension was exposed to four pulses (45 s each) of ultrasound using a Branson soni-fier (at output 4.0) adapted with a microtip Cell debris was harvested by centrifugation (7000 g, 4C, 30 min) and the supernatant was sequentially filtered through cheesecloth and 0.22-lm Millex filters (Millipore, Billerica, MA, USA)
Purification of recombinant Sfbgly The soluble material resulting from the lysis of the induced bacteria was mixed with 200 mm sodium phosphate (pH 7.0) containing 3.4 m (NH4)2SO4(2 : 1, v⁄ v) This mix-ture was incubated at 4C, without shaking, for 16h The precipitated material obtained after centrifugation (7000 g,
4C, 30 min) was discarded,and the supernatant was filtered through cheesecloth and 0.22-lm Millex filters (Millipore) Then, this supernatant was loaded onto a Resource ETH column (GE HealthCare, Chalfont, St Giles, UK) Non-retained proteins were eluted with 50 mm sodium phosphate (pH 7.0) containing 1.27 m (NH4)2SO4, whereas retained proteins were eluted using a linear gradi-ent of (NH4)2SO4 (from 1.27 to 0 m) prepared in 50 mm sodium phosphate (pH 7.0) Fractions of 1.0 mL were col-lected and analyzed for b-glycosidase activity using 4 mm p-nitrophenyl b-fucoside prepared in 50 mm citrate phos-phate buffer (pH 6.0)
Fractions containing b-glycosidase activity were pooled, mixed with (NH4)2SO4as described above and then loaded onto a Resource ISO column (GE HealthCare) Non-retained proteins were eluted with 50 mm sodium phos-phate (pH 7.0) containing 0.95 m (NH4)2SO4, whereas retained proteins were eluted using a linear gradient of (NH4)2SO4 (from 0.95 to 0 m) prepared in 50 mm sodium phosphate (pH 7.0) Fractions of 1.0 mL were collected and analyzed for b-glycosidase, as previously described To ascertain the purity of Sfbgly, fractions containing b-glyco-sidase activity were pooled and submitted to SDS-PAGE analysis followed by silver staining [25,26]
The protein concentrations were determined by using absorbance at 280 nm in the presence of 6 m guanidium hydrochloride prepared in sodium phosphate (pH 6.5) Extinction coefficients (e280 nm) were calculated based on the mutant Sfbgly sequences [27,28]
Enzyme kinetic analysis The initial rate of hydrolysis of at least 10 different concen-trations of p-nitrophenyl b-fucoside, p-nitrophenyl b-gluco-side, p-nitrophenyl b-galactob-gluco-side, p-nitrophenyl b-xyloside
Table 3 Parameters of the linear free energy relationships (LFER)
between the wild-type and mutant Sfbgly.
LFER
parameter
Mutant
E190A E194A K201A M453A K201F E190Q
Correlation
coefficient
Trang 10and methylumbeliferyl b-glucoside were separately measured
at 30C Substrate concentrations ranging from 0.2 to 4 Km
were used and their hydrolysis was detected by following the
production of p-nitrophenolate or methylumbeliferone All
substrates were prepared in 50 mm citrate phosphate
(pH 6.0) The enzyme kinetic parameters Kmand kcat were
determined by fitting these data on the Michaelis–Menten
equation using the enzfitter software (Elsevier-Biosoft,
Cambridge, UK)
The Kifor linear competitive inhibitors (cellobiose,
cello-triose, cellotetraose, cellopentaose, methyl b-glucoside, hexyl
b-glucoside, heptyl b-glucoside, octyl b-glucoside, nonyl
b-glucoside and phenyl b-glucoside) were determined by
measuring the initial hydrolysis rate of at least four different
concentrations of methylumbeliferyl b-glucoside in the
pres-ence of at least five different concentrations of inhibitors
(ranging from 0 to 4 Ki) The Kivalues were calculated from
replots of the inhibitor concentration versus the slope of the
lines observed in Lineweaver–Burk plots [29]
Calculation of thermodynamic parameters
Differences in the energy of the ES complexes (DDG)
between a pair of different enzymes (mutant and wild-type)
hydrolyzing the same substrate were calculated by the
equa-tion [30,31]:
DDGz¼ RT lnðkcat=KmÞmut=ðkcat=KmÞwt ð1Þ
where R is the gas constant (8.3144 JÆmol)1ÆK)1), T is the
absolute temperature (303 K) and mut and wt indicate
mutant and wild-type Sfbgly, respectively
The binding energy between an enzyme and a
competi-tive inhibitor was calculated using the equation:
DG0¼ RT lnð1=KiÞ ð2Þ where R is the gas constant (8.3144 JÆmol)1ÆK)1) and T is
the absolute temperature (303 K)
The binding energy corresponding to each glucose unit of
an oligocellodextrin was calculated using the equation:
DDG0n¼ DG0
n DG0
where DDG0 represents the binding energy of the ‘n’
glu-cose unit of the inhibitor, DG0 corresponds to the binding
energy of the oligocellodextrin presenting a degree of
poly-merization equal to ‘n’, and DG0(n)1) is the binding energy
of the oligocellodextrin presenting a degree of
polymeriza-tion equal to ‘(n)1)’ Methyl b-glucoside was used as an
inhibitor presenting a degree of polymerization equal to 1
Linear free energy relationships
The effect of the alteration of the substrate structure on the
stability of the complex ESwas evaluated using the
follow-ing equation [30,31]:
DDGz¼ RT lnðkcat=KmÞ1=ðkcat=KmÞ2 ð4Þ where R is the gas constant (8.3144 JÆmol)1ÆK)1), T is the absolute temperature (303 K) and kcat⁄ Km is the rate constant of hydrolysis of substrates 1 and 2 by the same enzyme (wild-type or mutant Sfbgly) The substrates were p-nitrophenyl b-fucoside, p-nitrophenyl b-glucoside and p-nitrophenyl b-galactoside
Then, DDG values calculated for the wild-type Sfbgly were plotted versus those for each mutant Sfbgly The simi-larity between the active sites of the wild-type enzyme and each mutant enzyme is related to the linear correlation coefficient and the line slope [32]
Structural comparison and sequence alignment The spatial structure of Sfbgly was modeled using chain A
of the myrosinase of the aphid Brevicoryne brassicae (1WCG) as the template [33] The homology modeling process was performed using the Swiss Model server The resulting structure was visualized using DeepView⁄ Swiss PDBViewer v3.7 [34] The structure of Sfbgly, Zea mays b-glucosidase 1 (ZmGlu1; 1E56), Sorghum bicolor (SbDhr1; 1V03) and Paenibacillus polymyxa (BlgB; 2O9R) were visu-alized and superimposed using DeepView⁄ Swiss
PDBView-er v3.7 Amino acid sequences of these b-glycosidases wPDBView-ere retrieved from the CAZy databank [2] and aligned using clustalx [35] Only the segment containing residues form-ing the aglycone-bindform-ing site of ZmGlu1 and SbDhr1 was presented
Acknowledgements
This project is supported by FAPESP (Fundac¸a˜o de Amparo a` Pesquisa do Estado de Sa˜o Paulo) and CNPq (Conselho Nacional de Desenvolvimento Cient-ı´fico e Tecnolo´gico) L M F Mendonc¸a is a graduate fellow from FAPESP and S R Marana is staff mem-ber of the ‘Departamento de Bioquı´mica – IQUSP’ and research fellow from CNPq
References
1 Esen A (1993) b-glucosidases Biochemistry and Molecular Biology ACS Symposium Series 553, Washington, DC
2 Coutinho PM & Henrissat B (1999) Carbohydrate Active Enzymes ServerAvailable at: http://afmb.cnrs-mrs.fr/CAZY
3 Kempton JB & Withers SG (1992) Mechanism of Agro-bacteriumb-glucosidase: kinetic studies Biochemistry
31, 9961–9969
4 Davies GJ, Wilson KS & Henrissat B (1997) Nomencla-ture for sugar-binding subsites in glycosyl hydrolases Biochem J 321, 557–559