Journal of Agricultural and Food Chemistry
Article
the abundant clusters [root-mean-square deviation (RMSD) = 2 Å]
that had lower binding energies. Weak intermolecular interactions,
such as hydrogen bonding and hydrophobic interactions, were
Table 2. Inhibitory Activity of Compounds 5a−5c at 10 μM
against SmChiB
23
analyzed by LigPlus+ and visualized by PyMOL (DeLano Scientific
LLC, San Carlos, CA, U.S.A.) and Discovery Studio, client version 20
(
Dassault System
̀
es, San Diego, CA, U.S.A.).
Crystallization and Data Collection. Protein−inhibitor com-
plex crystals were prepared by co-crystallization methods. For
SmChiB−6k and SmChiB−6p crystal complex crystallization, 10
mg/mL protein solution was mixed with an equal volume of well
solution containing 5 mM 6k and 6p, 5% (v/v) DMSO, 50 mM
24
citrate (pH 5.6), 0.5 M Li SO , and 0.25 M (NH ) SO . These
2
4
4
2
4
crystals were soaked for 10 s in a reservoir solution containing 20%
v/v) glycerol as a cryoprotection reagent and subsequently flash-
(
cooled in liquid nitrogen. X-ray diffraction data of the complexes were
collected on BL-19U1 at the Shanghai Synchrotron Radiation Facility
in China. The diffraction data of the complexes were processed using
the HKL-3000 package.
Structure Determination and Refinement. The structures of
SmChiB−6k and SmChiB−6p were solved by molecular replacement
25
with Phaser using the structure of free SmChiB as a model (PDB
26
entry 4Z2G). The PHENIX suite of programs was used for
structure refinement. Coot was used for manually building and
27
Discovery and Modification of Compound 6a. Our
strategy was to construct compound 6a by introducing a rigid
pyridine moiety based on compound 5a to extend the π system
and increase hydrophobicity. An additional carbonyl group was
expected to form a hydrogen bond with residues in the binding
pocket (Figure 2). To realize the simple synthetic procedure,
extending the molecular models. The stereochemical quality of the
2
8
models was checked by PROCHECK. The coordinates of
SmChiB−6k and SmChiB−6p were deposited in the PDB as entries
7C92 and 7CB1, respectively. The structural figures were generated
using the PyMOL program. The statistics for the diffraction data and
the structure refinement are summarized in Table 1.
we successfully constructed the key Friedlander synthon 4,
̈
enabling us to obtain compound 6a by following the same
synthetic procedure as the procedure used for the synthesis of
compound 5a (Figure 3).
RESULTS AND DISCUSSION
■
Design and Synthesis of 1,8-Naphthyridine Deriva-
tives 5a−5c. 1,8-naphthyridine derivatives with a planar
structure have attracted significant attention for their broad-
spectrum biological activities, such as antitumor, antibacterial,
To our delight, compound 6a showed good improved
activity toward SmChiB in comparison to compound 5a.
Compound 6a showed 61.4% inhibitory activity at 10 μM
(IC50 = 3.7 μM) (Table 3). With the inspiring results, we
sought to modify compound 6a by replacement of the phenyl
group with different aromatic heterocycles, substituted phenyl
ring, alkyl cycle, alkyl chain, and alkenyl group. First, the
methyl group was added in the benzyl position (6b), resulting
in decreased activity (IC50 = 6.3 μM). The additional methyl
group may bias compound 6b toward less bioactive
conformation and hamper the hydrophobic interaction with
29,30
anti-inflammatory, and protein kinase inhibition.
Initially,
with key interaction analysis of previously reported chitinase
inhibitors FQ1 and TP11, we designed compound 5a with a
simple synthetic procedure as a probe to explore the binding
pocket and tested our hypothesis. The 1,8-naphthyridine π
system of compound 5a was expected to form a stacking
interaction with aromatic residues in the binding subsites of
chitinase. Besides, the amino group could form a pseudo-ring
through the intramolecular hydrogen bonding (IMHB) to
stabilize the co-plane conformation to fit the binding cleft.
Three 1,8-naphthyridine derivatives 5a−5c were constructed
and evaluated their activity against SmChiB. However, the
inhibition activity was not satisfactory. Only compound 5a
showed marginal inhibition against SmChiB, with 12.6%
inhibition activity at 10 μM (Table 2). To further improve
inhibition activity, the docking-predicted binding mode of
compound 5a in the SmChiB active sites was analyzed. A
nomenclature for sugar-binding subsites was proposed by
SmChiB. The substituted phenyl group with F, CF , F and Cl,
3
and OMe was also introduced to replace the phenyl group
(6c−6e). However, none of them improved the activity,
indicating that the phenyl ring was not tolerant to
substitutions. Then, we replaced the phenyl group with several
heterocycles; both 6g (IC50 = 0.77 μM) and 6k (IC50 = 1.18
μM) showed enhanced activity (Table 4). To establish the
molecular basis of binding for further design of derivatives, we
sought to obtain the co-crystal complex of SmChiB with
compounds 6g and 6k. However, only the crystal complex of
SmChiB−6k was determined at a resolution of 2.3 Å.
3
1
Davies et al., where subsite + n represents a subsite at the
reducing end, subsite − n represents a subsite at the non-
reducing end, and the cleavage point is between the −1 and +1
subsites. The target region of our molecular docking
calculations was based on the position of the substrate in the
chitinase crystal structures, which occupies the −1, −2, and −3
subsites of SmChiB according to the crystal structure, with
PDB access code 1E6R. Compound 5a was predicted to bind
in the active site of SmChiB. However, it seemed that the
hydrophobic surface of compound 5a bicyclic moiety was not
big enough to form a strong π−π stacking to occupy the active
site of chitinase (Figure 4). This result prompted us to increase
the hydrophobicity of the compounds.
Binding Mechanism of Compound 6k with SmChiB.
Two binding modes were found in the asymmetric unit of
SmChiB−6k: two molecules of compound 6k were bound in
the bound +1 and +2 subsites of the binding cleft one protein
(Figure S1A of the Supporting Information), and the other
protein accommodates only a single compound 6k (Figure
5A). Although the two-molecule binding mode has been
32
reported , our previous isothermal titration calorimetry (ITC)
33
study indicated that only a single inhibitor was bound to
chitinase SmChiB under similar experimental conditions. The
existence of the two-molecule binding mode might be due to
higher concentrations of compound 6k used to grow crystals.
D
J. Agric. Food Chem. XXXX, XXX, XXX−XXX