Mechanistic insights into enzymatic catalysis by trehalase from the insect gut endosymbiont Enterobacter cloacae

Article abstract of DOI:10.1111/febs.14760

Energy metabolism in the diamondback moth Plutella xylostella is facilitated by trehalase, an enzyme which assists in trehalose hydrolysis, from the predominant gut bacterium Enterobacter cloacae. We report the biochemical and structural characterization of recombinant trehalase from E. cloacae (Px_EclTre). Px_EclTre showed K_M of 1.47 (±0.05) mm, k_cat of 6254.72 min^?1 and V_max 0.2 (±0.002) mm·min^?1 at 55 °C and acidic pH. Crystal structures of Px_EclTre were determined in the ligand-free form and bound to the inhibitor Validoxylamine A. The crystal structure of the ligand-free form, unavailable until now for any other bacterial trehalases, enabled us to delineate the conformational changes accompanying ligand binding in trehalases. Multiple salt bridges were identified that potentially facilitated closure of a hood over the substrate-binding site. A cluster of five tryptophans lined the ?1 substrate-binding subsite, interacted with crucial active site residues and contributed to both trehalase activity and stability. The importance of these residues in enzyme activity was further validated by mutagenesis studies. Many of these identified residues form part of signature motifs and other conserved sequences in trehalases. The structure analysis thus led to the assignment of the functional role to these conserved residues. This information can be further explored for the design of effective inhibitors against trehalases.

Full text of DOI:10.1111/febs.14760

Received Date : 05-Jul-2018
Revised Date : 21-Oct-2018
Accepted Date : 15-Jan-2019
Mechanistic Insights into Enzymatic Catalysis by Trehalase from the insect gut
endosymbiont Enterobacter cloacae
Anmol Adhav1#, Shrikant Harne2#, Amey Bhide3, Ashok Giri3*, Pananghat
Gayathri2*, Rakesh Joshi1*
1Institute of Bioinformatics and Biotechnology, Savitribai Phule Pune University,
Ganeshkhind, Pune 411007, Maharashtra, India
2Indian Institute of Science Education and Research, Dr. Homi Bhabha Road, Pune
411008, Maharashtra, India
3Division of Biochemical Sciences, CSIR National Chemical Laboratory, Dr. Homi Bhabha
Road, Pune 411008, Maharashtra, India
*Corresponding Authors
Dr. Rakesh Joshi, Institute of Bioinformatics and Biotechnology, Savitribai Phule Pune
University,
Ganeshkhind,
Pune
411007,
Maharashtra,
India
Email:
rakesh.joshi@unipune.ac.in
Dr. Pananghat Gayathri, Indian Institute of Science Education and Research, Dr. Homi
Bhabha Road, Pune 411008, Maharashtra, India Email: gayathri@iiserpune.ac.in
Dr. Ashok Giri, Division of Biochemical Sciences, CSIR National Chemical Laboratory, Dr.
Homi Bhabha Road, Pune 411008, Maharashtra, India Email: ap.giri@ncl.res.in
This article has been accepted for publication and undergone full peer review but has not
been through the copyediting, typesetting, pagination and proofreading process, which may
lead to differences between this version and the Version of Record. Please cite this article as
doi: 10.1111/febs.14760
This article is protected by copyright. All rights reserved.

Running Title: Enzymatic Catalysis by Trehalase
Article type : Original Article
#Authors contributed equally
Abstract
Energy metabolism in the diamondback moth Plutella xylostella is facilitated by
trehalase, an enzyme which assists in trehalose hydrolysis, from the predominant gut
bacterium Enterobacter cloacae. We report the biochemical and structural
characterization of recombinant trehalase from Enterobacter cloacae (Px_EclTre).
Px_EclTre showed KM of 1.47 (±0.05) mM, kcat of 6254.72 min-1 and Vmax 0.2 (±0.002)
mM/min at 55 C and acidic pH. Crystal structures of Px_EclTre were determined in the
ligand-free form and bound to the inhibitor Validoxylamine A. The crystal structure of
the ligand-free form, unavailable until now for any other bacterial trehalases, enabled
us to delineate the conformational changes accompanying ligand binding in trehalases.
Multiple salt bridges were identified that potentially facilitated closure of a hood over
the substrate-binding site. A cluster of five tryptophans lined the -1 substrate binding
subsite, interacted with crucial active site residues, and contributed to both trehalase
activity and stability. The importance of these residues in enzyme activity was further
validated by mutagenesis studies. Many of these identified residues form part of
signature motifs and other conserved sequences in trehalases. The structure analysis
thus led to the assignment of the functional role to these conserved residues. This
information can be further explored for the design of effective inhibitors against
trehalases.
Keywords: Enterobacter cloacae, Structural analysis, Trehalase, Plutella xylostella,
Validoxylamine A
This article is protected by copyright. All rights reserved.

Abbreviations: Ec_Tre, E. coli trehalase; Px_EclTre, Enterobacter cloacae trehalase; Suc,
Sucrose; Sc_Tre, Yeast trehalase; Unlig, Unliganded; ValA, Validoxylamine A
Introduction
Plutella xylostella (diamondback moth) infestation is one of the prominent biotic
stresses on cruciferous crops. It exhibits remarkable adaptability to its environment and
has developed resistance against most of the existing pest control strategies [1].
Recently, it was found that a few bacterial species from gut microbiota of P. xylostella
are vital for their digestive and detoxification processes. P. xylostella gut hosts 148
bacterial species belonging to 10 different phyla. Amongst these, Enterobacter cloaca
was found to be the most abundant species of larval gut microbes. These bacteria
produce various carbohydrate modifying and glycolytic enzymes like cellulases,
trehalases and other glucosidases, which participate in food digestion and nutrition
acquisition process of the host [2]. Thus, structural characterization of these enzymes of
the symbiont provides scope to explore them as novel targets for future biological or
chemical insect control methods.
Amongst different carbohydrate metabolic pathways, trehalose metabolism has a
variety of physiological roles in most organisms. Trehalose serves as a major energy
source, and also functions as stress protectant, in phytophagous insects [3]. Trehalase
catalyses the breakdown of trehalose into two glucose molecules (Fig. 1A) and the high
amount of energy released in the process is utilized for various physiological activities
[4]. Trehalase inhibition in insects leads to abnormalities like hampered growth,
unsuccessful stress recovery, weight loss, hypoglycemia, reduced chitin synthesis, lethal
metamorphosis and decreased flight capacity [5,6]. Hence, trehalase inhibitors,
Validoxylamine A being one of them (Fig. 1A), will be a good pest control strategy
against insects. With this goal, we have initiated biochemical and structural
characterization of trehalase from Enterobacter cloacae (Px_EclTre).
Trehalases are classified as either acidic or neutral depending on their pH optima
and also, as soluble or membrane-bound, based on their cellular localization [5]. Crystal
structures of bacterial and yeast trehalases co-crystallized with various ligands and
activation partners are summarized in Table 1. Crystal structure of trehalase from
Escherichia coli (PDB ID: 2WYN) showed an (α/α)6 barrel fold similar to other α-toroidal
This article is protected by copyright. All rights reserved.

glycosidases. The complexes of E. coli trehalase with Validoxylamine A (Ec_Tre-ValA;
2JF4) and 1-thiatrehazolin (Ec_Tre-TTZ; 2JG0) revealed the interactions in the -1 and +1
subsites, where -1 subsite is “catalytic site” and +1 subsite is “recognition site” for the
two glucose subunits in trehalose [7,8]. The corresponding unliganded structure of E.
coli trehalase has not been determined. The requirement of an activation partner for
catalysis differentiates a yeast trehalase from the bacterial counterpart. The yeast
trehalase crystal structures include two unliganded forms and complexes with
activation partners, however, an active conformation with the specific substrate and
activation partner is unavailable. Hence, despite the presence of multiple crystal
structures, features contributing to the molecular mechanism of trehalose catalysis are
still unexplored.
In the present report, we have performed biochemical and structural
characterization of trehalase (Px_EclTre) from the P. xylostella bacterial endosymbiont
E. cloacae. We determined the crystal structures of Px_EclTre in the unliganded and
Validoxylamine A bound forms. A comprehensive structural analysis of the unliganded
and liganded structures of Px_EclTre, were performed by comparing them with the E.
coli and S. cerevisiae trehalase structures, to understand the salient features
accompanying substrate recognition and catalysis. Characterization of mutants
validated the role of conserved residues involved in catalysis and structural stability of
Px_EclTre.
Results
Px_EclTre exhibited acidic α, α-trehalase activity with high optimum temperature
Px_Ecltre gene coding for periplasmic trehalase (Refseq Accession ID:
WP_045355849.1) was recombinantly expressed in E. coli and the protein purified by
affinity chromatography. The purified protein showed a single protein band at ~65 kDa
on a 12% SDS-PAGE gel (Fig. 1B). The activity of purified Px_EclTre was measured by
colorimetric estimation of product formed upon reaction of glucose with 3,5-
Dinitrosalicylic acid (DNSA) [9]. Px_EclTre exhibited a KM of 1.47 (±0.05) mM, kcat of
6254.72 min-1 and Vmax 0.2 (±0.002) mM/min (Fig. 1C). Px_EclTre displayed optimum
This article is protected by copyright. All rights reserved.

activity at pH 5 (Fig. 1D), similar to the characteristic feature of acidic trehalases (pH 5
to 7) [6,10–13]. The optimum temperature for Px_EclTre activity was found to be 55 °C
(Fig.1E), which is higher as compared to previously reported trehalases from other
organisms such as Zunongwangia sp. (50 ⁰C), Apis mellifera L. (47 ⁰C), Saccharomyces
cerevisiae (40 ⁰C) and Rhodotorula rubra (30 ⁰C)[13–16].
Validoxylamine A showed strong competitive inhibition of Px_EclTre activity
Validoxylamine A, a trehalose analogue, is an inhibitor of trehalases, with demonstrated
inhibitory activity against bacterial and insect trehalases (Fig. 1A) [17]. In the present
study, inhibition kinetics of Validoxylamine A against Px_EclTre at optimum conditions
(pH 5, 55 °C) was measured to estimate its trehalase inhibitory activity. The
concentration-dependent decrease in the activity of Px_EclTre was observed with
increasing concentrations of Validoxylamine A (Fig. 1F). Enzyme kinetic analysis
suggested competitive inhibition of Px_EclTre by Validoxylamine A (Fig. 1G).
Furthermore, IC50 and Ki of Validoxylamine A for Px_EclTre was found to be 7.5 (± 0.2)
and 1.9 (± 0.075) nM, respectively (Fig. 1F). A similar Ki value (1.9 nM) was previously
reported for inhibition of trehalase from Rhizoctonia solani by Validoxylamine A [17,18].
Microscale thermophoresis analysis showed that Validoxylamine A bound strongly to
Px_EclTre with a dissociation constant (KD) of 206 (±10) nM (Fig. 1H). Further,
structural studies were carried out on Px_EclTre in unliganded form (Px_EclTre-Unlig)
and co-crystalized with Validoxylamine A (Px_EclTre-ValA) to investigate its catalytic
mechanism.
Crystal structures of Px_EclTre in unliganded form and Validoxylamine A bound
state
Crystal structures of Px_EclTre in unliganded (Px_EclTre-Unlig) and Validoxylamine A
bound (Px_EclTre-ValA) forms were determined to resolutions 2.3 Å and 1.8 Å (PDB
IDs: 5Z6H and 5Z66), respectively. Data collection and refinement statistics are given in
Table 2. The protein consists of (/)6 toroidal fold, with a hood-like domain that closes
This article is protected by copyright. All rights reserved.

over the ligand binding pocket (Fig. 2A). Strong electron density for the ligand was
observed, confirming that the inhibitor was indeed bound to trehalase (Fig. 2B-D).
Interestingly, in the substrate-binding pocket of the unliganded form (Px_EclTre-Unlig;
5Z6H), a glycerol moiety from the crystallization condition was observed, whose atoms
superposed very well with corresponding atoms of the bound ligand (Fig. 2E). In the
absence of a ligand, a glycerol moiety and water molecules substituted for satisfying the
hydrogen bonds formed by the active site and ligand-binding residues. Since glycerol is
a non-specific ligand with very few contacts with the protein compared to the substrate
(trehalose) and trehalase activity is not affected (specific activity of 1392.7 (±22.27)
U/mg with 10 % glycerol compared to 1421.5 (±18.32) without glycerol), we consider
glycerol as part of the solvent and the glycerol-bound form as the unliganded form of
Px_EclTre (Px_EclTre-Unlig; 5Z6H). Structure determination of a ligand-free form of the
enzyme (Px_EclTre-Unlig; 5Z6H) allowed us to compare the specific conformational
changes accompanying ligand binding.
Structural comparison of trehalases in unliganded and ligand-bound states
The Px_EclTre shares about 79% sequence identity with E. coli trehalase over the entire
sequence length and 29% with S. cerevisiae trehalase comparing the homologous region
of the hood domain. While all four structures of trehalase from E. coli are in the ligand-
bound forms, S. cerevisiae trehalase structures in the ligand-free and liganded forms are
available. Since yeast trehalases are activated in the presence of an activator Bmh1, the
complexed structure with Bmh1 was also included in the structural analysis [19]. The
availability of these structures permitted us to carry out a comprehensive comparison
of different conformations of trehalases, including ligand derivatives with different
sugar linkages and activator bound structures. Table 1 contains the root-mean-square
deviation (RMSD) of (α/α)6 domain of all the unliganded and liganded conformations of
trehalases with respect to (α/α)6 domain of Px_EcTre-Unlig (5Z6H).
The extensive structural comparison of liganded/unliganded and active/inactive
structures of trehalases led us to identify the following changes upon ligand binding - 1)
subtle loop movements 2) salt bridges specific to ligand binding, 3) interaction between
This article is protected by copyright. All rights reserved.

conserved residues lining active site (+1 and -1 subsite), 4) predict roles in substrate
recognition and hydrolysis for some of the residues conserved across trehalases. Most
of these findings were not discussed in earlier reports on trehalases due to
unavailability of a structure in catalytically active form with the specific substrate (for
yeast) and unliganded form (for E. coli) in the respective organisms.
Conformational changes and ordering of the ‘lid loop’ upon ligand binding
Comparison of the ligand-free and inhibitor-bound structures illustrated that the hood
domain Q36-L100, and loop comprising residues Y147-Y159 and G506-G519 (hereafter
referred to as side loop and lid loop respectively) might undergo conformational
changes upon ligand binding (Fig. 3A). We compared the lid loop, side loop and hood
domain positions with other available structures of trehalases (Table 1). Liganded
structure of Px_EclTre (Px_EclTre-ValA; 5Z66) superimposed on E. coli trehalase
structures complexed with Validoxylamine A (Ec_Tre-ValA; 2JF4) and 1-thiatrehazolin
(Ec_Tre-TTZ; 2JG0) showed RMSD of 0.334 (389 C atoms) and 0.311 Å (359 C atoms),
respectively. The lid loop, side loop and the hood domain were oriented identically in
these structures. Compared to the other inhibitor-bound trehalase structures of E. coli,
one of the interesting observations was the difference in conformation of the lid loop in
the structures liganded with derivatives of casuarine-6-O--D-glucoside (Ec_Tre-
LG9+Glu; 2WYN and Ec_Tre-Casu+Glu; 2JJB)[20,21]. The presence of a sulphate ion at
the active site disrupted the interaction between the corresponding glutamate (E511
)
and the substrate, thereby leading to incomplete lid loop closure in Ec_Tre-Casu+Glu
(2JJB) and Ec_Tre-LG9+Glu (2WYN) (Fig. 3B). The lid loop was disordered in all the
yeast trehalase structures except for Sc_Tre-Suc+Bmh1 (complexed with sucrose and
Bmh1; 5N6N), where it was not closed completely (Fig. 3B). In Sc_Tre-Suc+Bmh1
(5N6N), the presence of non-specific substrate (sucrose) probably kept the
corresponding glutamate (E690) distant from the ligand. The superposed structures
showed that lid loop residues E511, Y512 and F518, which are conserved across trehalases,
contacted the ligand, and binding of the specific ligand ensured complete loop closure.
Thus, we hypothesized that residues from lid loop play a major role in recognizing the
correct substrate (Fig. 3B). Mutation of the glutamate (E511) to alanine in Px_EclTre
This article is protected by copyright. All rights reserved.

resulted in 23% activity of the mutant enzyme compared to the wildtype (WT)
Px_EclTre (specific activity of 397.4 U/mg compared to 1734 U/mg for wildtype
enzyme; Fig. 1C), confirming the importance of E511 in substrate recognition and
catalysis.
In addition to the lid loop, the side loop (Y147-Y159) also underwent a conformational
change upon substrate binding (Fig. 3C). Out of the available yeast neutral trehalase
structures (Table 1), the loop position was open in the unliganded structures (Sc_Tre-
Unlig1; 5JTA and Sc_Tre-Unlig2; 5NIS). This loop superposed well with the
corresponding loop position in the unliganded structure of Px_EclTre (5Z6H) (Fig. 3C).
In the liganded form (5Z66), residues R153, F154 and Y158 interacted with the substrate,
all of which are conserved residues. Loop position in the liganded state appeared to be
driven not only by the presence of a substrate in the pocket but also by the specificity of
the correct ligand at the +1 subsite (Table 1). In the sucrose bound yeast trehalase
structure Sc_Tre-Suc+Bmh1 (5N6N), despite the substrate-binding site being occupied,
the loop position matched with that of the unliganded conformations (Px_EclTre-Unlig;
5Z6H) (Fig. 3C). Since the glucoside linkage between the monosaccharide units is
different in sucrose compared to trehalose, the +1 subsite occupancy of sucrose
sterically prevented F303 (corresponding to F154 in Px_EclTre) from closing over the
ligand. Hence, the loop closure is one of the features that contribute towards recognition
of α, α (1→1) linkage sugar in the active site pocket of trehalases. In order to investigate
the role of F154 in catalysis, we mutated it to alanine (F154A) in Px_EclTre. Though the
mutant was well folded (Fig. 4A-D), it showed complete loss of trehalase activity
(undetectable levels of activity for the mutant, Fig. 1C), emphasizing the significance of
F154 in the enzyme function.
The hood domain closed over the ligand-binding pocket in the Validoxylamine A
bound Px_EclTre structure (5Z66). An RMSD of 1.125 Å (91 C atoms of the hood
domain, upon superposition of the toroid domain) highlighted the domain movement
compared to the unliganded structure (Fig. 3A, D). Though there were no direct
interactions of hood domain residues with the ligand, interactions of the hood domain
with residues of the side loop appeared to drive the conformational changes.
Specifically, R153 from side loop formed a salt bridge with D59 from the hood domain.
The position of the hood domain matched with those of all liganded structures of E. coli
This article is protected by copyright. All rights reserved.

trehalase (Ec_Tre-TTZ; 2JG0, Ec_Tre-LG9+Glu; 2WYN, Ec_Tre-ValA; 2JF4 and Ec_Tre-
Casu+Glu; 2JJB) (Table 1). Interactions that led to domain closure were dependent on
occupancy of the specific ligand in the +1 subsite, and conformation of the side loop.
These observations prompted us to look at the specific interactions of each residue that
drive active site closure in trehalase structures.
A cluster of conserved salt bridges lines the substrate-binding pocket and
mediates the active site closure
Salt bridges lining the substrate-binding pocket (Fig. 5A) constituted major
determinants for closed conformation of the active site. A salt bridge between R277 and
E511 is formed when the lid loop closes over the active site. Both R277 and E511 formed
bridging hydrogen bonds with O7’ and O4’ of Validoxylamine A (+1 subsite)
respectively, that led to substrate binding and loop closure (Fig. 5B and C).
Furthermore, as discussed earlier, the decrease in activity of E511A mutant verified the
crucial role of the residue in Px_EclTre glycosidic activity (Fig. 5C). The salt bridge also
serves as a determinant of substrate specificity. In the E. coli trehalase structure bound
to casuarine-6-O--D-glucoside (2WYN), the presence of a sulphate ion bound at the
active site disrupted the salt bridge between E511 and R277 and resulted in an
intermediate position of the lid loop.
The side loop closure over the substrate-binding site is accompanied by the
formation of another salt bridge between R153 from the side loop and D59 from the hood
domain. R153 formed a hydrogen bond with O7’ of Validoxylamine A, which coordinated
the substrate binding with the closure of both side loop and hood domain. R206
interacted with Validoxylamine A (O3’ and O4’ atoms respectively) and E279, thus
forming a bridging set of interactions with the ligand (Fig. 5B and C). Mutation of E279 in
Px_EclTre affected the activity of the enzyme (undetectable activity, Fig. 1C). Two other
conserved residue pairs that link the toroid and hood domains are K61 and E156; D65 and
R200, respectively. These residues formed a cluster of salt bridges in both unliganded
and ligand-bound states and did not interact with ligand (Fig. 5D). Thus, they may not
contribute to the switch between conformational states accompanying ligand binding.
This article is protected by copyright. All rights reserved.

Structural superimposition of bacterial and yeast trehalases demonstrated that
despite the difference in hood domain fold (Fig. 3D), the network of salt bridges was
conserved at the corresponding positions in yeast trehalases also (Fig. 5C). This
evolutionary conservation of salt bridge network between prokaryotic and eukaryotic
trehalases, despite differences in the fold of the hood domain, indicates their
significance in the catalytic activity of the enzyme. In yeast trehalase, R277 that is part of
a salt bridge interaction with the conserved E511 is replaced with a proline. Interestingly,
guanidium group of R473 in yeast occupied the same position as the side chain of R277 in
Px_EclTre and was positioned optimally to conserve the salt bridge interaction as well
as interaction with the ligand [eg. yeast trehalase structure complexed with trehalose
(Sc_Tre-Trehalose; 5M4A; Fig. 5C)]. The salt bridge was not formed in Sc_Tre-Suc+Bmh1
(5N6N: yeast trehalase structure complexed with sucrose and Bmh1) due to the
presence of the non-specific substrate, though the lid loop was ordered. R302 and E426
were poised to form the salt bridge pair in yeast trehalase for the side loop closure. The
potential salt bridge interaction was not observed among yeast trehalase structures,
probably due to them being either catalytically inactive or unliganded conformations.
The salt bridge by the R206 and E279 pair and interaction with the ligand is conserved in
the yeast trehalase structures (R355 and E424 in yeast). The interaction with the ligand
(sucrose) was disrupted in Sc_Tre-Suc+Bmh1 (5N6N), and the salt bridge interaction
also was not observed. D212-R349and K214-E305 in yeast trehalases formed pairs of
conserved salt bridges corresponding to D65-R200, and K61-E156, respectively in bacterial
trehalases. D212 and K214 from the hood domain of yeast were spatially conserved with
K61 and D65 of bacterial trehalases, though the hood domains are of different folds.
All the salt bridges described above coordinated ligand binding and closure through
recognition of the monosaccharide moiety at the +1 subsite. Further analysis was
carried out to understand the effect of ligand binding at the -1 subsite.
This article is protected by copyright. All rights reserved.

Substrate binding pocket at the -1 subsite is lined by tryptophan residues
Analysis of contacts to -1 subsite of the ligand highlighted the presence of conserved
tryptophans in the ligand-binding pocket. The active site of trehalases is surrounded by
five tryptophans, namely W160, W311, W447, W453 and W520, out of which all except W311
are conserved (Fig. 6A, B). N2 atom of W447 formed a hydrogen bond with O2 of
Validoxylamine A, while W447 and W453 formed approximately parallel stacking
interactions (Fig. 6B, C; the angle between the planes of side chains <12˚; Table 1) in
close proximity to the substrate. W453 formed hydrophobic interactions with K497 (Fig.
6C). Only when the correct ligand or its analogue was bound, K497 oriented such that its
terminal amino group formed crucial hydrogen bonds with Q446 and E496. The conserved
Y512 also interacted with Q446 and E496 side chains in catalytically active trehalase
conformations (Fig. 6C). Since the density of K675 in yeast trehalase structures
(corresponding to K497 in bacterial trehalase) was unavailable beyond C, presence of
hydrogen bonds has been inferred only on basis of bacterial trehalase structures (PDB
IDs: 2WYN, 2JG0, 2JJB, 2JF4, 5Z66, 5Z6H) and yeast trehalase structure 5N6N (Sc_Tre-
Suc+Bmh1) where the side chain is structured for K675. W453 was sandwiched between
the lysine (K497) and W447 that interacts with the ligand. These interactions might serve
towards recognition of the cognate ligand for optimally orientating the active site
cluster. Px_EclTre mutant W447A showed structural destabilization (Fig. 4C-D) and
complete loss of activity (undetectable activity, Fig. 6C).
W311, adjacent to the W447 side chain, formed a cation- interaction with R316, while
R316, in turn, formed salt bridging interactions with D336 and D389. W311 is replaced by a
histidine (H477) in yeast trehalase, whose imidazolium ring coincides with the 5-
membered ring of tryptophan side chain (Fig. 6D). These interactions are conserved
across all trehalases and might be involved in structure stabilization. Two other
tryptophans in the vicinity of the active site and which interact with the ligand are W160
and W520, whose side chains are perpendicularly oriented (Fig. 6B).
Though many of the residues discussed above did not interact with the ligand
directly, they appear to form crucial interactions that modulate allosteric regulation of
enzyme activity. A conservation analysis of the implicated residues among 185
This article is protected by copyright. All rights reserved.

sequences showed the level of conservation of each of these residues, highlighting the
maintenance of the interactions across trehalases (Fig. 6E).
Discussion
Molecular mechanism of trehalose recognition and hydrolysis
Glycoside hydrolases catalyze the hydrolysis of sugar by two mechanisms – one involves
retention of the anomeric configuration while the other occurs through an inversion
step. [22]. Trehalases are known to show the inversion of anomeric configuration of
glucose i.e. it forms both α-glucose and β-glucose upon trehalose hydrolysis [23–25]. In
inverting enzymes, an average distance of ~10 Å between the catalytic residues is
observed [23]. In the case of liganded form and unliganded structures of Px_EclTre, the
distances between the Cγ and Cδ atoms of the respective catalytic residues D312 and E496
were 9.2 Å and 11.6 Å, respectively. This suggests that Px_EclTre possesses an inverting
type of mechanism of catalysis. Interestingly, it was also observed that the catalytic D312
is in a cis-peptide conformation in both the liganded and unliganded structures.
The atoms from the substrate at the -1 subsite are surrounded by five tryptophans
(W160, W311, W447, W453 and W520), four of which (except W311) are conserved (Fig. 6A,
B). Tryptophans have been observed in other carbohydrate metabolizing enzymes such
as endoglucanase CelA and β-mannanase. In these examples, the aromatic residue is
near the active site and stacks with the sugar ring of the substrate ring by orienting the
aromatic ring of the side chain in parallel [26,27]. However, in trehalases, substrate
recognition does not involve stacking interactions of aromatic rings. The aromatic
residue (W447) interacts with the substrate by formation of a hydrogen bond with the -
OH group at the C2 position of the ligand. Interestingly, earlier studies with derivatives
of trehalose have shown that substitutions in the C2 position affect the catalytic activity
of trehalase, implying that the interaction of W447 plays a significant role in the catalytic
activity [20]. Interaction of a tryptophan residue with the substrate that assists
substrate binding in active site pocket has been demonstrated in other glycosyl
hydrolases, including lysozyme and chitinases. In these hydrolases, tryptophan residues
were involved in hydrogen bonding with the substrate through the N2 atom of the side
This article is protected by copyright. All rights reserved.

chain (as observed in Px_EclTre) or formed a stacking interaction with sugar rings of the
substrate [28–30].
In the present study, we have highlighted a set of 5 salt bridges. While three of these
salt bridges were formed upon binding of the correct ligand at the active site, the
remaining four residues formed a network of constitutive salt bridges (present
independent of ligand binding). Apart from salt bridges, the residues from side loop and
lid loop are involved in substrate recognition and interaction at the +1 subsite. The
binding of a correct ligand at the active site led to reorienting the Q446, E496, K497 and Y512
such that residues K497 and Y512 formed H-bonds with Q446 and E496. These hydrogen
bonds were not formed when there is a non-specific ligand at the active site or when
there is no ligand. Hence, the interactions identified as specific to the active
conformation of Px_EclTre could be signatures of the active form of trehalases,
regulated by either the correct substrate or by allostery.
Role of conserved motifs and signature sequences in trehalose catalysis
Two signature motifs (PGGRFIEFYYWDSY and QWDFPNVWPP) and a glycine-rich
region (GGGGEY) are conserved across all the insect trehalases [31]. Additionally, the
dipterans also have been reported to contain five conserved sequences (DSKTFVDMK,
IPNGGRV/IYY, RSQPPF/LL, GPRPESYREDI and ELKAG/ AAESGMDFSSRWFV) [5,31].
Though the trehalase sequences from multiple sources (bacteria, yeast, fungi, insects,
dipterans etc.) belonging to different life forms are known, a comprehensive analysis
correlating the conserved sites and sequence motifs to their function was not reported
earlier.
Upon aligning trehalase sequences from 100 different sources, as picked up from
UniProt, we find that the conserved motifs from N-terminus to C-terminus are:
[D/R]59X[K/Q/Y]XXXDXX67, P150GGRFXEXYXWDXY163
S309G[W/H]DX[S/T]XR316 P334[I/V/A][E/D]LN338
,
N197XXRXXYXX[R/H]SQPP210
[D/Y]389[Y/F/W][N/D]391
,
,
,
,
Q446WDXPXXWXP455 and E511YXXXXGFGWTN522 (numbering according to Px_EclTre).
These also correspond to the conserved sequence stretches in dipteran trehalase
sequences. The conserved motif E511YXXXXGFGWTN522 is preceded by GGGG, a ‘Glycine-
This article is protected by copyright. All rights reserved.

rich’ region, in most of the trehalases. Apart from the aforesaid motifs, a few other
residues are also conserved: H118, W123, Y158, G168, L169, M215, E237, Y269, [R/P]277, E279
L337, Y488, E496, K497. As mentioned earlier, in yeast structures where there is a P422
corresponding to R277 in bacterial trehalases, the salt bridge is formed by R473
,
,
.
Interestingly, we observe that these sequence motifs contain most of the residues
implicated in the salt bridges and tryptophan residues identified as residues of potential
functional significance. While E496 is a conserved catalytic base, the binding of the
correct substrate led to the formation of hydrogen bonds by E496 and Y512 with Q446 and
K497. The residues R277, E279 were involved in the formation of salt bridges with residues
from conserved motifs. The rest of the conserved residues formed interacting pairs
(H118 & Y488, W123 & L169, Y269 & L337, Y158 & E237) and are probably involved in structural
stability. Mutational studies on these residues can provide further insights into their
function. Conserved residues and their functions are summarized in Table 3.
Based on our structural, biochemical and mutational studies, we have assigned a
functional role to the conserved residues in trehalases. The study highlights the role of
salt bridges and a cluster of tryptophans in substrate specificity and catalysis. These
strong interactions surrounding the active site might contribute to the optimal activity
of the enzyme at a high temperature and acidic pH.
Design of Validoxylamine A or Validamycin A based trehalase inhibitors
Many life forms including bacteria, yeast, plants, mice and humans have been reported
to produce trehalase enzyme [19,32–35]. Interestingly, the sensitivity of trehalase from
different sources to inhibitors such as validamycin A or Validoxylamine A seems to vary.
For instance, feeding rats with 500 and 10,000 ppm validamycin A or Validoxylamine A
is not toxic and no abnormalities were observed in F0 and F1 generations [36]. In normal
humans, the expression of trehalase is at very low levels. Elevated expression of
trehalase is observed only in cases of kidney disorders or untreated celiac disease [37].
The use of trehalase inhibitors Validoxylamine A and Validamycin A as anti-fungal
agents against plant pathogenic fungi has already been reported [18,38].
Validoxylamine A, a breakdown product of Validamycin A, can be naturally degraded to
non-hazardous products Valienamine and Validamine [35].
This article is protected by copyright. All rights reserved.

Thus, in addition to its use as an antifungal agent, the use of validamycin A or
validaxylamine A based trehalase inhibitors appears to be a reasonable choice against
insect pests, based on our study of inhibition of trehalase from an essential insect gut
bacterium by Validoxylamine A [39]. The information about the interaction of
Validoxylamine A with Px_EclTre binding pocket can be further explored for the
structure-activity relationship to optimize the inhibition potential and specificity of
inhibitor molecule. Understanding of binding mode of trehalases with novel inhibitors
that disrupt the salt bridges and the tryptophan interactions can be utilized for fine
tuning enzyme-inhibitor interaction for the design of more effective and specific
trehalase inhibitors.
Materials and Methods
Production and purification of recombinant Px_EclTre
The sequence for the Px_Ecl_tre (Px000318) was obtained from the Diamondback moth
Genome Database (http://iae.fafu.edu.cn/DBM)[40]. This sequence codes for an 890 aa
protein. The trehalase coding sequence from Px000318 (1599 bp) was identified. Signal
peptide from the amino acid sequence of Px_EclTre was predicted using SignalP 4.1
server and was excluded while designing primers. RNA was isolated from 100 mg of
larval tissue using TriZol (Thermo Fisher Scientific, Massachusetts, USA). RNA was
treated with DNase enzyme to remove DNA contamination (Promega, Wisconsin, USA).
Furthermore, cDNA was prepared using High-Capacity cDNA Reverse Transcription kit
(Applied Biosystems, California, USA) as per manufacturer’s instructions. Random
primers were used for cDNA preparation from total RNA. Gene was amplified using
specific primers and cloned in Zero Blunt® vector (Thermo Fisher Scientific,
Massachusetts, USA), followed by cloning in pET28a (Novagen, New Jersey, USA) vector
with EcoRI and HindIII restriction sites. Sequencing confirmed that cloned sequence
corresponds to Enterobacter cloacae periplasmic trehalase (Refseq Accession ID:
WP_045355849.1). cDNA prepared using oligo dT primers do not show amplification
for this gene, which in turn indicates its bacterial origin.
This article is protected by copyright. All rights reserved.

BL21 Star (DE3) cells transformed with Px_EclTre construct were grown in
Luria-Bertani (LB) media containing 50 µg/mL Kanamycin to OD600 0.6-0.8. The cells
were then induced with isopropyl-β-D-1-thiogalactopyranoside (IPTG) at the final
concentration of 0.5 mM and incubated at 16°C for 10 h. Cells were harvested by
centrifugation and lysed by sonication in lysis buffer (50 mM Tris pH 8.0, 500 mM NaCl
and 10% Glycerol). The lysate was centrifuged at 12000 rpm for 15 min at 4°C. The
clear supernatant thus obtained was allowed to bind with 700 μl Pro Bond Ni–NTA
resin (Thermo Fisher Scientific, Massachusetts, USA) pre-equilibrated with binding
buffer (50 mM Tris pH 8.0 and 500 mM NaCl) and then loaded on Bio-Rad gravity
column (Bio-Rad, Hercules, CA, USA). Non-specifically bound proteins were washed
away with 25 mL wash buffer (50 mM NaH2PO4, 300 mM NaCl and 20 mM Imidazole pH
8.0). The protein bound to resin was eluted with 200 mM Imidazole in the wash buffer.
The protein was concentrated by passing through 10 kDa cutoff membrane (Millipore,
Billerica, MA, USA). The concentrated protein was further purified by size exclusion
chromatography using Superdex-75 column (GE Healthcare, UK) on ÄKTA system (GE
Healthcare, UK) and eluted into 50 mM Tris pH 8.0 and 50 mM NaCl. Purified fractions
were pooled and concentrated for further activity studies and crystallization
experiments. The mutants were purified using a Ni-NTA affinity chromatography as
described earlier (Fig.4A-B).
Biochemical characterization of Px_EclTre and mutants
Trehalase activity was estimated by monitoring release of glucose from α, α-Trehalose
using DNSA (Dinitrosalicylic acid) reagent [9]. A pre-mix containing 50mM Tris-HCl pH
7.0 (130 μL) and equal units (32 nM) of purified Px_EclTre protein (20 μL) was
prepared to which 150 μL of Trehalose (0.25%) was added. The reaction was incubated
for 15 min at 37⁰C and then stopped by adding 500 μL of DNSA reagent. The reaction
tubes were incubated in a boiling water bath for 5 min and absorbance was measured at
540 nm. One unit of trehalase was defined as the amount of enzyme required to release
1 μM glucose/min at 37°C from trehalose under the given assay conditions.
This article is protected by copyright. All rights reserved.

Enzyme activity of Px_EclTre was measured at different pH values ranging from
2 to 12. Equal units of enzyme in citrate-phosphate, sodium phosphate, Tris and glycine-
NaOH buffers (50 mM each) were used for the activity assay at specified pH values 2 to
6, 7, 8 to 9, and 10 to 12, respectively. For determining the optimum temperature for
trehalose hydrolysis, the activity of purified Px_EclTre was measured at pH 5.0 using 50
mM citrate sodium phosphate buffer at various temperatures ranging from 15°C to
95°C.
Substrate saturation assay (Michaelis-Menten kinetics) was performed to
calculate KM and Vmax of the recombinant enzyme. The assay was performed at 37°C and
pH 5 using various concentrations of trehalose (0.8 to 6.5 mM) as substrate. KM and Vmax
for Px_EclTre were calculated by fitting the data to a Michaelis-Menten plot using
GraphPad Prism v6.0 (GraphPad Software, California, USA). For determination of
inhibition kinetics and Ki of Validoxylamine A (obtained by acid hydrolysis of
Validamycin A), enzyme activity assays were performed at 6.6 nM and 33 nM inhibitor
concentration. Trehalase inhibition values obtained were used to generate Michaelis-
Menten and double reciprocal plot to determine Ki and mode of inhibition, respectively.
Results were analysed by plotting double reciprocal plots and data fitting into
competitive inhibition model of GraphPad Prism v6.0. This model uses the equation of
Obs
KMObs=KM*(1+[I]/Ki), where KM
is apparent KM and [I] is inhibitor concentration.
All assays were performed as two independent experiments in triplicate and
obtained data were subjected to analysis of variance (ANOVA) using GraphPad Prism
v6.0 (GraphPad Software, San Diego, CA). The level of significance between different
groups was judged by F-test, while data were compared using the least significant
differences (LSD) at P ≤ 0.05. In all the figures, error bars represent ±standard error for
the observations of each treatment.
Crystallization of Px_EclTre
Purified Px_EclTre (11 mg/ml in the buffer) was used to set up a crystallization screen
of 480 commercially available conditions by sitting drop method with the help of
Mosquito Nanodrop Crystallization Robot (TTP Labtech, UK). 100 nL of protein was
This article is protected by copyright. All rights reserved.

mixed with 100 nl of the crystallization condition and the plates were incubated at 18
°C. Among the various initial hits obtained, one of the conditions (0.17 M Ammonium
Sulfate, 25% PEG 4000 and 15% glycerol) was optimized further. The optimized
crystallization condition consisted of 0.5M ammonium sulphate, 25% PEG 4000 and
15% glycerol. Equal volumes of optimized crystallization condition and protein (31
mg/ml) were mixed and incubated at 18°C to obtain crystals. Plate-like crystals were
observed around 48-72 hrs. A mixture of 1 mM (final concentration) Validamycin A
(Duchefa Biochemie B.V., Haarlem, Netherlands) and purified Px_EclTre (10 mg/ml)
was used for crystallization of the liganded complex. Diffraction quality crystals were
obtained in the condition with 0.2M ammonium sulphate and 28% PEG 4000. The cryo
protectant for unliganded Px_EclTre consists of 0.17M ammonium sulphate, 25.5% PEG
4000, 20% glycerol while 0.2M ammonium sulphate, 28% PEG 4000 and 20% glycerol
was used for Px_EclTre crystals in complex with Validamycin A.
Data collection, structure determination and refinement
X-ray diffraction data was collected at beamline I04 of Diamond Light Source (DLS)
Synchrotron facility, United Kingdom (UK). Data was processed using XDS [41] and
SCALA [42]. The structure was determined by molecular replacement using Phaser [43],
with the trehalase structure 2JG0 (Ec_Tre-TTZ) as a model. Model building and
structure refinement were carried out using COOT and Phenix, respectively [44,45]. The
Data collection and refinement statistics are summarized in Table 2.
Interestingly, upon solving the co-crystal structure, we found that the protein
had in fact Validoxylamine A bound, instead of Validamycin A used for setting up
crystallization. Composite omit map calculated using Phenix unambiguously showed
that the bound ligand is indeed Validoxylamine A [46]. We believe this is possibly
because of hydrolysis of Validamycin A to Validoxylamine A under acidic conditions
present in the crystallization mixture for the prolonged time period.
This article is protected by copyright. All rights reserved.

Analysis of functional sites, mutagenesis and activity
Site-directed mutagenesis of Px_EclTre for F154A, E279A, W447A and E511A was
carried out using Quick Change Lightning Site Directed mutagenesis kit (Agilent
Technology, Santa Clara, USA) as per manufacturer’s protocol. The primers were
designed as described in the manual, and the mutation was confirmed by sequencing.
Mutant plasmids were transformed in E. coli BL21 (DE3) cells and protein was
expressed. The protocol as described for wildtype protein purification was followed for
purification of mutants. Enzyme activity measurements were also carried out as
described above and compared with wildtype Px_EclTre.
Thermal shift assay
Thermal shift assay was performed to gain information about the folded state of
Px_EclTre mutants by comparing their thermal denaturation curves as a function of
temperature, with respect to the wildtype protein. The assay was set up by adding
following components to wells in a BIO-RAD (Bio-Rad, California, USA) Multiplate® PCR
plate™ 96-well: protein (final concentration 2µM), Sypro Orange (Sigma-Aldrich)
(diluted to 50X in DMSO: buffer 1:10) dye at a final concentration of 5X, buffer (20 mM
sodium phosphate buffer pH 8.0 and 100 mM KCl) was added to make the final reaction
volume to 25 µL. The wells were sealed with adhesive seals (Microseal® ‘B’ seals, Bio-
Rad). Plates were briefly spun and put into a BIO-RAD CFX96 Real-Time System RT-PCR
machine. The temperature was increased from 4⁰C to 90⁰C with an increment of
0.4⁰C/20 sec. Fluorescence was monitored using FRET channel 6 (excitation λ = 450-
490 nm and detection λ = 560-580 nm). Melting temperature (Tm) of wildtype and the
mutants was estimated from the plots of relative fluorescence vs temperature.
This article is protected by copyright. All rights reserved.

Circular Dichorism Spectroscopy
Circular Dichroism (CD) assays were carried out using Jasco spectropolarimeter (Jasco,
Tokyo, Japan) under constant nitrogen flow[47]. Far-UV spectra were recorded using
0.1 cm path length quartz cuvette. Five consecutive measurements were accumulated to
obtain mean spectra. The observed ellipticities were converted into molar ellipticities
(θ) based on molecular mass per residue of wildtype and mutants.
MicroScale Thermophoresis (MST) of the Px_EclTre interaction with
Validoxylamine A
Microscale thermophoresis experiments were carried out using the Monolith NT 115
instrument from Nanotemper technologies [48]. His-tag labelled Px_EclTre was kept
constant at a concentration of 5 nM, whereas the concentration of unlabelled partner
Validoxylamine A varies from 0.61E-3 to 1.25 µM in 1X PBS. LED power of 100% and
MST power of 60% were used for this experiment. Data analysis was carried out using
NTanalysis (Nanotemper technologies, Munchen, Germany) [49].
Conservation analysis
In order to identify the residues conserved across trehalases Consurf server was used
[50]. Conservation pattern was visualized and depicted using The PyMOL Molecular
Graphics System, Version 2.0 Schrödinger, LLC.
Acknowledgements
The project work is supported by the research grant from the Department of Science
and Technology - Science and Engineering Research Board (DST-SERB), Government of
India under ECR/2015/000502 grant and Board of College and University
Development, Savitribai Phule Pune University, Pune 411007, Maharashtra, India. Work
in the lab of PG is supported by INSPIRE Faculty Fellowship (IFA12-LSBM-52); Indian
National Science Academy Research grant; Women Excellence Award, Department of
This article is protected by copyright. All rights reserved.

Science and Technology; and Innovative Young Biotechnologist Award, Department of
Biotechnology. AA acknowledges UPE II for fellowship. SH acknowledges IISER Pune for
fellowship. Authors also acknowledge Dr. Sivaramaiah Nallapeta and Dr. Saji Menon,
NanoTemper Technologies for their assistance in Microscale thermophoresis analysis.
Synchrotron facilities at ESRF, Grenoble, and Diamond Light Source, UK are
acknowledged. We thank EMBL-ESRF and DBT India for facilitating data collection at
ID29, ESRF; and beamline scientists at I03 and I04 (MX14198, MX15369, MX16452 and
MX17251) at the Diamond Light Source, UK.
Author contributions
AA, SH, AB and RJ designed and carried out the experiments, SH and PG solved crystal
structures, PG provided structural insights, APG, PG and RJ supervised overall work and
wrote paper with comments from all authors.
Note
The authors declare no competing financial interest.
References
1 Furlong MJ, Wright DJ & Dosdall LM (2013) Diamondback Moth ecology and
management: Problems, Progress, and Prospects. Annu. Rev. Entomol. 58, 517–541.
2 Xia X, Gurr GM, Vasseur L, Zheng D, Zhong H, Qin B, Lin J, Wang Y, Song F, Li Y, Lin H &
You M (2017) Metagenomic sequencing of diamondback moth gut microbiome unveils
key holobiont adaptations for herbivory. Front. Microbiol. 8. 663.
3 Becker A, Schlöder P, Steele JE & Wegener G (1996) The regulation of trehalose
metabolism in insects. Experientia 52, 433–439.
4 Wyatt GR (1957) The Chemistry of Insect Hemolymph: II. Trehalose And Other
Carbohydrates. J. Gen. Physiol. 40, 833–847.
5 Barraza A, Sánchez F, Barraza A & Sánchez F (2013) Trehalases: A neglected carbon
metabolism regulator? Plant Signal. Behav. 8, e24778.
6 Forcella M, Cardona F, Goti A, Parmeggiani C, Cipolla L, Gregori M, Schirone R, Fusi P &
This article is protected by copyright. All rights reserved.

Parenti P (2010) A membrane-bound trehalase from Chironomus riparius larvae:
Purification and sensitivity to inhibition. Glycobiology 20, 1186–1195.
7 Silva MCP, Terra WR & Ferreira C (2004) The role of carboxyl, guanidine and imidazole
groups in catalysis by a midgut trehalase purified from an insect larvae. Insect Biochem.
Mol. Biol. 34, 1089–1099.
8 Gibson RP, Gloster TM, Roberts S, Warren RAJ, Storch De Gracia I, García Á, Chiara JL &
Davies GJ (2007) Molecular basis for trehalase inhibition revealed by the structure of
trehalase in complex with potent inhibitors. Angew. Chemie - Int. Ed. 46, 4115–4119.
9 Miller GL (1959) Use of dinitrosalicylic acid reagent for determination of reducing sugar.
Anal Biochem 31, 426–428.
10 Forcella M, Mozzi A, Bigi A, Parenti P & Fusi P (2012) Molecular cloning of soluble
trehalase from Chironomus riparius larvae, its heterologous expression in Escherichia
coli and bioinformatic analysis. Arch. Insect Biochem. Physiol. 81, 77–89.
11 Shukla E, Thorat L, Bhavnani V, Bendre AD, Pal JK, Nath BB & Gaikwad SM (2016)
Molecular cloning and in silico studies of physiologically significant trehalase from
Drosophila melanogaster. Int. J. Biol. Macromol. 92, 282–292.
12 Silva MCP, Terra WR & Ferreira C (2010) The catalytic and other residues essential for
the activity of the midgut trehalase from Spodoptera frugiperda. Insect Biochem. Mol.
Biol. 40, 733–741.
13 Cheng Q, Gao H & Hu N (2016) A trehalase from Zunongwangia sp .: characterization
and improving catalytic efficiency by directed evolution. BMC Biotechnol., 16, 9.
14 Lee J-. H, Saito S, Mori H, Nishimoto M, Okuyama M & Kim D (2007) Molecular cloning of
cDNA for trehalase from the European honeybee, Apis mellifera L., and its heterologous
expression in Pichia pastoris. Biosci Biotechnol Biochem 71, 2256-65.
15 Biswas N & Ghosh AK (1996) Characterisation of an acid trehalase of Saccharomyces
cerevisiae present in trehalase-sucrase aggregate. BBA-Gen Subj. 1290, 95-100.
16 Mansure JJ, Silva JT & Panek AD (1992) Characterization of trehalase in Rhodotorula
rubra. Biochem 28, 693–700.
17 Asano N (2003) Glycosidase inhibitors: Update and perspectives on practical use.
Glycobiology 13, 93-104.
This article is protected by copyright. All rights reserved.

18 Asano N, Takuji Y, Yukihiko K & Katsuhiko M (1987) Effect of validamycins on
glycohydrolases of Rhizoctonia solani. J. Antibiot. (Tokyo). 40, 526–532.
19 Alblova M, Smidova A, Docekal V, Vesely J, Herman P, Obsilova V & Obsil T (2017)
Molecular basis of the 14-3-3 protein-dependent activation of yeast neutral trehalase
Nth1. Proc. Natl. Acad. Sci. U. S. A. 114, 9811–9820.
20 Cardona F, Goti A, Parmeggiani C, Parenti P, Forcella M & Fusi P (2010) Casuarine-6-O-
α-d-glucoside and its analogues are tight binding inhibitors of insect and bacterial
trehalases. Chem Commun 46.
21 Cardona F, Parmeggiani C, Faggi E, Bonaccini C, Gratteri P, Sim L, Gloster TM, Roberts S,
Davies GJ, Rose DR & Andrea G (2009) Total Syntheses of Casuarine and Its
6‐O‐α‐Glucoside: Complementary Inhibition towards Glycoside Hydrolases of the GH31
and GH37 Families. Chem. – A Eur. J. 15, 1627–1636.
22 Koshland DE (1953) Stereochemistry and the mechanism of enzymatic reactions. Biol.
Rev. 28, 416–436.
23 Davies G & Henrissat B (1995) Structures and mechanisms of glycosyl hydrolases.
Structure 3, 853–859.
24 Henrissat B, Callebaut I, Fabrega S, Lehn P, Mornon JP & Davies G (1995) Conserved
catalytic machinery and the prediction of a common fold for several families of glycosyl
hydrolases. Proc. Natl. Acad. Sci. U. S. A. 92, 7090–7094.
25 Zechel DL & Withers SG (2000) Glycosidase mechanisms: anatomy of a finely tuned
catalyst. Acc. Chem. Res. 33, 11–18.
26 Alzari PM, Souchon H & Dominguez R (1996) The crystal structure of endoglucanase
CelA, a family 8 glycosyl hydrolase from Clostridium thermocellum. Structure 4, 265–
275.
27 Hilge M, Gloor SM, Rypniewski W, Sauer O, Heightman TD, Zimmermann W,
Winterhalter K & Piontek K (1998) High-resolution native and complex structures of
thermostable beta-mannanase from Thermomonospora fusca - substrate specificity in
glycosyl hydrolase family 5. Structure 6, 1433–1444.
28 Phillips DC, Cheetham JC & Ox O (1992) Refinement of an enzyme complex with
inhibitor bound at partial occupancy. Hen egg-white lysozyme and tri-N-
acetylchitotriose at 1.75 A resolution. J Mol Biol. 224, 613–628.
This article is protected by copyright. All rights reserved.

29 Maenaka K, Matsushima M, Kawai G, Kidera A, Watanabe K, Kuroki R & Kumagai I
(1998) Structural and functional effect of Trp-62 Gly and Asp-101 Gly substitutions on
substrate-binding modes of mutant hen egg-white lysozymes. Biochem J. 333, 71–76.
30 Terwisscha van Scheltinga AC, Armand S, Kalk KH, Isogai A, Henrissat B & Dijkstra BW
(1995) Stereochemistry of Chitin Hydrolysis by a Plant ChitinaseLysozyme and X-ray
Structure of a Complex with Allosamidin: Evidence for Substrate Assisted Catalysis.
Biochemistry 34, 15619–15623.
31 Shukla E, Thorat LJ, Nath BB & Gaikwad SM (2015) Insect trehalaseꢀ: Physiological signi
fi cance and potential applications. Glycobiology 25, 357–367.
32 Kendall EJ, Adams RP & Kartha KK (1990) Trehalase activity in plant tissue cultures.
Phytochemistry 29, 2525–2528.
33 Sacktor B (1968) Trehalase and the transport of glucose in the mammalian kidney and
intestine. Proc. Natl. Acad. Sci. U. S. A. 60, 1007.
34 Sasai-Takedatsu M, Taketani S, Nagata N, Furukawa T, Tokunaga R, Kojima T KY (1996)
Human Trehalase: Characterization, Localization, and Its Increase in Urine by Renal
Proximal Tubular Damage. Nephron 73, 179–185.
35 Chen X, Lu Y, Fan Y & Shen Y (2017) An Introduction to Validamycins and Their
Derivatives. In Validamycin and Its Derivatives, 1st Edition (Chen X, Lu Y, Fan Y & Shen Y,
eds), 1–8. Elsevier.
36 Moriya M, Ohta T, Watanabe K, Miyazawa T, Kato K & Shirasu Y (1983) Further
mutagenicity studies on pesticides in bacterial reversion assay systems. Mutat. Res.
Toxicol. 116, 185–216.
37 Alpers DH (2003) Carbohydrates: Digestion, Absorption, And Metabolism. In
Encyclopedia of Food Sciences and Nutrition, 2nd Edition (Caballero B, ed), 881–887.
Academic Press, Oxford.
38 Wu Q, Zhang L, Xia H, Yu C, Dou K, Li Y & Chen J (2017) Omics for understanding
synergistic action of validamycin A and Trichoderma asperellum GDFS1009 against
maize sheath blight pathogen. Sci. Rep. 7, 40140.
39 Asano N, Takeuchi M, Kameda Y, Matsui K & Kono Y (1990) Trehalase inhibitors,
Validoxylamine A and related compounds as insecticides. J. Antibiot. (Tokyo). 43, 722–
726.
This article is protected by copyright. All rights reserved.

40 You M, Yue Z, He W, Yang X, Yang G, Xie M, Zhan D, Baxter SW, Vasseur L, Gurr GM,
Douglas CJ, Bai J, Wang P, Cui K, Huang S, Li X, Zhou Q, Wu Z, Chen Q, Liu C, Wang B, Li X,
Xu X, Lu C, Hu M, Davey JW, Smith SM, Chen M, Xia X, Tang W, Ke F, Zheng D, Hu Y, Song
F, You Y, Ma X, Peng L, Zheng Y, Liang Y, Chen Y, Yu L, Zhang Y, Liu Y, Li G, Fang L, Li J,
Zhou X, Luo Y, Gou C, Wang J, Wang J, Yang H & Wang J (2013) A heterozygous moth
genome provides insights into herbivory and detoxification. Nat. Genet. 45, 220–225.
41 Kabsch W (2010) XDS. Acta Crystallogr. D. Biol. Crystallogr. 66, 125–132.
42 Collaborative Computational Project, Number 4 (1994) The CCP4 suite: programs for
protein crystallography. Acta Crystallogr. D. Biol. Crystallogr. 50, 760–763.
43 McCoy AJ (2007) Solving structures of protein complexes by molecular replacement
with Phaser. Acta Crystallogr. D. Biol. Crystallogr. 63, 32–41.
44 Emsley P & Cowtan K (2004) Coot: model-building tools for molecular graphics. Acta
Crystallogr. D. Biol. Crystallogr. 60, 2126–2132.
45 Adams PD, Grosse-Kunstleve RW, Hung LW, Ioerger TR, McCoy AJ, Moriarty NW, Read
RJ, Sacchettini JC, Sauter NK & Terwilliger TC (2002) PHENIX: building new software for
automated crystallographic structure determination. Acta Crystallogr. D. Biol.
Crystallogr. 58, 1948–1954.
46 Bhat TN (1988) Calculation of an OMIT map. J. Appl. Crystallogr. 21, 279–281.
47 Mishra M, Joshi RS, Gaikwad S, Gupta VS & Giri AP (2013) Structural–functional insights
of single and multi-domain Capsicum annuum protease inhibitors. Biochem. Biophys. Res.
Commun. 430, 1060–1065.
48 Jerabek-Willemsen M, André T, Wanner R, Roth HM, Duhr S, Baaske P & Breitsprecher D
(2014) MicroScale Thermophoresis: Interaction analysis and beyond. J. Mol. Struct.
1077, 101–113.
49 Adhav AS, Kokane SR & Joshi RS (2018) Functional characterization of Helicoverpa
armigera trehalase and investigation of physiological effects caused due its to inhibition
by Validamycin A formulation. Int. J. Biol. Macromol. 112, 638–647.
50 Glaser F, Pupko T, Paz I, Bell RE, Bechor D & Martz E (2003) ConSurf: identification of
functional regions in proteins by surface mapping of phylogenetic information.
Bioinformatics 19, 163–164.
This article is protected by copyright. All rights reserved.

Table 1: Summary of bacterial and yeast trehalase structures
Source of
Angle between
planes of W447
and W453 side
chains
Position of
RMSD of
(α/α)6
domain*
PDB ID
Bound ligand(s)/
protein partner
Occupancy of +1 /
-1 subsite
the
Side
loop
(abbreviation)
Lid loop
enzyme
2JG0
(Ec_Tre-TTZ)
1-Thiatrehazolin
(TTZ)
E. coli
E. coli
α-D-Glucopyranosilamine/
Tetrahydrocyclopentathiazole
0.489
(378)
Closed
Closed
Closed
5.12⁰
LG9 (Casuarine
analogue) and
α-D-Glucose
(LG9+Glu)
Validoxylamine A
(ValA)
Casuarine and
α-D-Glucose
(Casu+Glu)
2WYN
(Ec_Tre-LG9+Glu)
α-D-Glucose/
LG9 (Casuarine analogue)
0.389
(342)
Closed
2.56⁰
6.27⁰
7.69⁰
2JF4
(Ec_Tre-ValA)
2JJB
(Ec_Tre-
Casu+Glu)
5M4A
(Sc_Tre-
Trehalose)
5N6N
E. coli
E. coli
C7-Cyclitol/
C7-Cyclitol
0.498
(422)
Closed
Closed
Closed
Closed
α-D-Glucose/
Casuarine
0.530
(397)
S.
α-D-Glucopyranose/
α-D-Glucopyranose
1.317
(253)
Trehalose
Closed
Open
11.47⁰
cerevisiae
Sucrose/
Protein Bmh1
(Suc+Bmh1)
S.
β-D-Fructofuranose/
α-D-Glucopyranose
1.259
(289)
(Sc_Tre-
Suc+Bmh1)
Open
Open
Intermediate
Not modeled
10.88⁰
8.88⁰
cerevisiae
5NIS
(Sc_Tre-Unlig2)
S.
None/
None
1.406
(265)
None
None
cerevisiae
5JTA
(Sc_Tre-Unlig1)
5Z66
(Px_EclTre-ValA)
5Z6H
(Px_EclTre-Unlig)
S.
None/
None
C7-Cyclitol/
C7-Cyclitol
None/
1.372
(263)
0.217
(374)
Open
Closed
Open
Not modeled
Closed
8.5⁰
7.25⁰
3.62⁰
cerevisiae
Validoxylamine A
(ValA)
E. cloacae
E. cloacae
None
Open
-
None
* Root-mean-square deviation of (α/α)6 domain calculated by superposing the structures onto Px_EclTre-Unlig (5Z6H) structure. The number of
superposed atoms is indicated in bracket.
This article is protected by copyright. All rights reserved.

Table 2: Data collection and refinement statistics
Trehalase-unliganded
Trehalase-
Validoxylamine A
5Z66
PDB ID
5Z6H
Data Collection statistics
Collected at
DLSI04
DLSI04
Wavelength (Å)
Space Group
0.979
0.979
P212121
C121
a, b, c (Å)
55.25, 127.91, 156.14
90, 90, 90
122.03, 51.45, 96.4
90, 106.95, 90
30.5 - 1.8 (1.84 – 1.8)
53185 (3172)
9.9 (62.8)
α, β, γ (˚)
Resolution (Å)*
Number of unique reflections*
Rmerge (%)*
33.32 - 2.30 (2.38-2.3)
48818 (4363)
12.1 (59.2)
8.6 (42.1)
Rpim (%)*
8.2 (52.1)
CChalf
*
0.98 (0.685)
9.5 (2.5)
0.992 (0.634)
9.4 (1.9)
Mean I/I
Completeness (%)*
Redundancy*
97.43 (96.0)
4.2 (4.2)
97.9 (100)
3.0 (3.1)
Refinement statistics
Resolution (Å)
33.32 – 2.3
30.49 - 1.8
Number of unique reflections
(test set)
48768 (2422)
53176 (2698)
Rwork / Rfree (%)
16.84/22.96
16.28/19.36
Average B-factor (Å2)
Wilson B-factor (Å2)
RMS deviations
31
21
27.1
15.6
Bond lengths (Å)
Bond angles (˚)
0.007
0.926
0.004
0.729
Ramachandran map statistics
Favoured (%)
96
03
00
97
03
02
Allowed (%)
Outliers (%)
* Values in parentheses denote last resolution shell.
This article is protected by copyright. All rights reserved.

Table 3: Residues conserved across trehalases and summary of their
functional roles
Residue*
Function
D59& R153, K61& E156, D65& R200, R206& E279, and
R277& E511
Salt bridging pairs
R153, R277, E279, E511
W160, W311**, W447, W453 and W520
D312 and E496
Conserved residues interacting at +1 site
Conserved tryptophans near -1 site
catalytic residues
R153, F154, Y158, E511, Y512, F518
Q446, E496 and K497
Substrate recognition
H118& Y488, W123& L169, Y269& L337, Y158& E237
R316, D336 and D389
Interacting pairs of residues, function not
assigned
*numbering according to bacterial trehalase residues in Px_EclTre
**replaced by a histidine in some and methionine in one structure
This article is protected by copyright. All rights reserved.

Fig. 1: Biochemical characterization of Px_EclTre. (A) Px_EclTre catalyzes the
hydrolysis of α, α-trehalose to break it down to α-D-glucose and β-D-glucose. This
reaction is inhibited by the trehalose analogue, Validoxylamine A. (B) 12% SDS-PAGE
gel showing purified Px_EclTre protein. (C) The rate of reaction as a function of
substrate (trehalose) concentration for wildtype and mutants of Px_EclTre. (D) Effect of
pH on trehalase activity by Px_EclTre. Px_EclTre shows optimum activity at pH 5. (E)
Effect of temperature on trehalase activity by Px_EclTre. The optimum temperature for
Px_EclTre activity was found to be 55⁰C. (F) Enzyme inhibition of Px_EclTre activity by
Validoxylamine A. Velocity vs trehalose concentration for different concentrations of
Validoxylamine A has been plotted. (G) Lineweaver-Burk plot shows that
Validoxylamine A functions as a competitive inhibitor. (H) MicroScale Thermophoresis
analysis of Validoxylamine A binding to Px_EclTre. The chemical formula for
Validoxylamine A is given in inset. In all the figures, error bars represent ±standard
error for the observations of each treatment.
This article is protected by copyright. All rights reserved.

Fig. 2. Crystal structures of Px_Ecl trehalase in its unliganded form and bound to
the inhibitor Validoxylamine A. (A) Superposition of unliganded (green) and ligand
bound (orange) forms of Px_EclTre. (B) The electron density (composite omit map of F0-
Fc at 2.5) of Validoxylamine A from the liganded form of Px_EclTre is shown with
atoms labelled and -1 and +1 subsites (SS) marked. (C, D) Ligand binding site of
Px_EclTre showing residues (orange) interacting with Validoxylamine A (wheat). For
clarity, interactions of the residues shown above and below the ligand are shown in two
separate figure panels (C, D), with the ligand maintained in the same orientation. (E)
Ligand binding site of superposed Px_EclTre unliganded and ligand-bound structures.
The glycerol moiety and water molecules present in unliganded structure contact the
protein in a similar manner as Validoxylamine A in the liganded form. Color scheme:
Validoxylamine A (wheat), glycerol (light green), the unliganded form of Px_EclTre
(dark green) and interactions between ligands and protein (dotted green/yellow lines).
Subsite has been abbreviated as ‘SS’ in all the figure panels.
This article is protected by copyright. All rights reserved.