Bioinformatic and biochemical analysis of the key binding sites of the pheromone binding protein of Cyrtotrachelus buqueti Guerin-Meneville (Coleoptera: Curculionidea)

Materials

Three compounds were chosen to investigate the ligand-binding specificity of CbuqPBP1. Ligands of the highest purity were purchased from Aladdin (Shanghai, China) and stored in accordance with the manufacture’s specifications. The sequence of CbuqPBP1 was taken from the GenBank with accession number KU845733.1.

Alignment and homology modeling

The amino acid sequence of CbuqPBP1 was downloaded from the GenBank and Blast was used to search against the CbuqPBP1 protein sequence in the Protein Data Bank to identify a structural template. Software Modeller 9.19 (http://salilab.org/modeller/) was used for homology modeling based on the sequence comparison results with the structural template sequence identified. The 3D structure obtained from modeling was evaluated with SAVES v5.0 (https://servicesn.mbi.ucla.edu/SAVES/). After confirming the models, the Chiron on-line server was used for optimization. Modeller 9.19 was used to optimize loop regions and PyMOL was used to analyze structural characteristics and to search for ligand binding sites.

Molecular docking

Based on the established homology model, the docking program AUTODOCK vina 1.1.2 was used to find the potential binding mode between CbuqPBP1 and the ligand dibutyl phthalate. Dibutyl phthalate with strong affinity is a female pheromone of the giant bamboo weevil, which plays a role in the process of male individual searching for female individual. ChemBioDraw Ultra 14.0 was used to simulate the structure of dibutyl phthalate and to generate a 3D structure of the ligand. Energy optimization was conducted using the MMFF94 force field and Autodock Tools 1.5.6 was used to create the PDBQT format (Huey et al., 2007; Morris et al., 2009). Binding coordinates of CbuqPBP1 and dibutyl phthalate were set to: center_x = 22.389, center_y = − 25.143, center_z = 1.08, and size_x = 15, size_y = 15, size_z = 15. Parameter exhaustiveness was set to 20 and default values were used for other parameters to increase the calculation accuracy. Finally, the conformation with the highest score was selected and PyMoL 1.7.6 was used for visual inspection and analysis of the structural data.

Site-directed mutagenesis

The CbuqPBP1 coding sequence was mutated to yield the two mutants CbuqPBP1-Phe69A (phenylalanine to alanine at position 69) and CbuqPBP1-His53A (histidine to alanine at position 53). PCR reactions were used to form overlapping chains. The extension of overlapping chains was used to splice segments in a superimposed manner. Primer5 was used to design primers (Table 1). Three rounds of PCR amplification were conducted after designing primers. Expression vectors (pET-28a(+)/PBP1-Phe69A, pET-28a(+)/PBP1-His53A and pET-28a(+)/PBP1) were generated and transformed into Escherichia coli BL21(DE3) competent cells for protein overexpression. Recombinant proteins produced were detected by SDS-PAGE analysis.

Expression and purification of the native protein and mutants

Expression plasmids were transformed into E. coli TOP10 competent cell and plated on agar plates. Several colonies were selected randomly for overnight cultivation in LB media and plasmids were extracted for sequencing. Mutant plasmids pET-28a(+)/PBP1- Phe69A and pET-28a(+)/PBP1- His53A with the correct sequence were transformed into E. coli BL21(DE3) competent cells, and cells were grown to an optical density at 600 nm (OD₆₀₀) of 0.6. IPTG was added to the culture to a final concentration of 1 mM and cells were further grown at 37 °C with shaking for 3 h to induce protein expression (Deng et al., 2011). After harvesting cells by centrifugation, ultrasound sonication was used to disrupt cells (200 W, 3/4 s, 25–30 min). The supernatants and sediments were collected under low temperature centrifugation (16, 000 g-force, 50 min) and SDS-PAGE detection was conducted. Nickel affinity (Ni-NTA) was used to purify recombination proteins, and the purified proteins were stored in Tris-HCl buffer (pH 7.4, 50 mM). To avoid the function of the protein being affected by the His-tag, recombinant bovine enterokinase was used to remove the His-tag and the protein was re-purified and collected. Purity was confirmed by SDS-PAGE analysis.

Fluorescence assay

To measure the affinity of the fluorescent ligand N-phenyl-1-naphthylamine (1-NPN) toward CbuqPBP1, a 2 µM solution of protein in 50 mM Tris-HCl, pH 7.4, was titrated with aliquots of 1 mM 1-NPN dissolved in methanol to a final concentration of 16 µM. The probe was excited at 337 nm and emission spectra were recorded between 350 and 550 nm. To evaluate the effect of pH on the binding affinity of CbuqPBP1, we also measured its binding with 1-NPN over a pH range of 4.5–9.0. The displacement of 1-NPN by selected ligands was measured in a competitive binding assay using both the protein and 1-NPN at 2 µM. The mixtures were titrated with 1 mM methanol solutions of each competitor at concentrations of 2–16 µM. The fluorescence of the mixture was recorded after 5 min. Dissociation constants for 1-NPN and the stoichiometry of binding were obtained from Scatchard plots of the binding data using the Prism software. For other competitor ligands, the dissociation constants were calculated from the corresponding half maximal inhibitory concentration (IC₅₀) values using the equation: inhibitory constant Ki = [IC₅₀]/(1+[1-NPN]/K_1-NPN), where [1-NPN] is the free concentration of 1-NPN and K_1-NPN is the dissociation constant of the protein/1-NPN complex.

Three-dimensional model of CbuqPBP1

On the basis of the Blast search against the Protein Data Bank, two types of insect odor proteins with known structures and quite similar sequences to the CbuqPBP1 sequence were found. These two odorant binding proteins were Nasonovia ribisnigri OBP3 (NribOBP3 PDB ID: 4Z45_A) and Megoura viciae OBP3 (MvicOBP3 PDB ID: 4Z39_A). The total sequence identity between the target (CbuqPBP1) and the template protein (NribOBP3) is 33% (Cavasotto & Phatak, 2009) (Fig. 1A). The resolution of the template is 2.02 Å.

After homology modeling, the 3D structure of CbuqPBP1 (Fig. 1B) is clearly very similar to the 3D structure of the template NribOBP3 (Fig. 1C). The structural characteristics of CbuqPBP1 are similar to other sex pheromone binding proteins and include six α-helices: residues 26–36 (α1), 44–51 (α2), 59–72 (α3), 83–94 (α4), 101–114 (α5) and 123–137 (α6). Six conserved cysteine residues stabilize the protein structure by forming three disulfide bonds. Disulfide bond Cys36–Cys67 connects α1 and α3, Cys63–Cys121 connects α3 and α6, and Cys110–Cys130 connects α5 and α6. Five of the α-helices adopt an antiparallel arrangement (α1, α3, α4, α5 and α6) and form an internal binding pocket. α2 forms a cover-type structure or lid above the pocket, which stabilizes this structure.

The result of further rationality estimates by Pro-CHECK (Fig. 1D) was that 88.4% residues were in the favored regions (red area A, B and L), 10.1% of the residues fall into additionally allowed regions (bright yellow area a, b, l, p) and 0.8% residues have backbone torsion angles that fall into generously allowed regions (light yellow area ∼a, ∼b, ∼l, ∼p). The percentage sum of residues in the allowed regions was 99.3%, which was higher than 95%. This result showed that the constructed 3D structure of CbuqPBP1 was a high-quality model.

Energy assessment was performed on ProSa Fig. 1E. The shadow part is Z-score value of all proteins similar to Cbuq PBP1 protein in PDB database, and the black spot is Z-score value of Cbuq PBP1 protein, which is −4.35. The Z-score value of template protein NOBP3 is −5.87 in the range of Z-score of known reasonable structural proteins, which indicates that the modeling structure is more stable than template structure. This indicates that the homologous modeling Institute is more stable than template structure. The constructed CBuq PBP1 protein is reasonable in energy.

Molecular docking

To research characteristics of CbuqPBP1 binding with odor molecules, dibutyl phthalate (Fig. 2A), which interacts with CbuqPBP1 favorably, was selected to construct a complex between CbuqPBP1 3D model and dibutyl phthalate. Such a model should clarify the mode of interaction of dibutyl phthalate with CbuqPBP1 at the molecular level. We have docked dibutyl phthalate with the active pocket of CbuqPBP1, with a binding energy of −6.4 kcal/mol. Generally, compound dibutyl phthalate bound to the active pocket of CbuqPBP1 with a compact conformation (Fig. 2B).

The benzene ring and one aliphatic chain of dibutyl phthalate were located in the hydrophobic region at the bottom of the pocket. Strong hydrophobic interactions formed between the ligand and residues Leu3, Leu4, Leu5, Leu29, Leu50, Pro56, Ile65 and Phe69. Another side chain of dibutyl phthalate was located at the opening of the pocket. Based on detailed analysis, a CH-π interaction may occur between the benzene ring of dibutyl phthalate and residue Phe69. Moreover, an important long-range hydrogen bond (3.3 Å) can form between one ester carbonyl oxygen of dibutyl phthalate and residue His53 (Fig. 2C). All aforementioned interactions enable the formation of a stable complex between dibutyl phthalate and CbuqPBP1.

Site-directed mutagenesis of CbuqPBP1 and binding specificities of mutants

After double enzyme digestion with restriction enzymes Nde I and Xha I, mutant plasmids pET-28a(+)/CbuqPBP1-His53A and pET-28a(+)/CbuqPBP1-Phe69A, and the original plasmid pET-28a(+)/PBP1 formed bands in an agarose gel that were ∼400 bp in length (Fig. 3). After SDS-PAGE analysis of protein overexpression, three specific bands with molecular weights of 16 kDa were observed in the SDS-PAGE gel, which is consistent with expected molecular weight of the target proteins (Fig. 4).

After ultrasonication to disrupt the bacteria and release the recombinant proteins (including His tag), SDS-PAGE analysis could be conducted (Fig. 5). All recombinant proteins were found in the supernatant part of the disrupted cells. After purification, recombinant bovine enterokinase was used to cleave the His-tag and following a further round of purification pure recombinant protein samples were obtained.

1-NPN was selected as the fluorescent probe. Fluorescence competitive binding assays were conducted for the purified wild-type CbuqPBP1, mutant CbuqPBP1-His53A and CbuqPBP1-Phe69A proteins. The fluorescence peak maximum in the presence of the recombinant proteins was recorded at different concentrations. The Scatchard equation was used to calculate the equilibrium binding constant (K_d) between CbuqPBP1, CbuqPBP1-His53A, CbuqPBP1-Phe69A and 1-NPN, which were determined to be 2.725, 3.352 and 2.260 µM, respectively. When the final concentration of odor substance was higher than 50 µM, the fluorescence peak did not decrease to half its value. This showed that almost no affinity was established between protein and the odor substance, and the binding constant could not be calculated (Fig. 6).

Dibutyl phthalate, benzothiazole and cedar camphor were selected based on previous fluorescence binding assay test (Yang et al., 2017b). Fluorescence competitive binding assays were conducted with CbuqPBP1, CbuqPBP1-His53A and CbuqPBP1-Phe69A (Fig. 7). Based on the results, CbuqPBP1 bound favorably with dibutyl phthalate, benzothiazole and cedar camphor. The binding ability of CbuqPBP1-His53A with the three types of odor substances was essentially lost. The binding ability of CbuqPBP1-Phe69A mutant with cedar camphor was significantly reduced, whereas affinity toward the other two odor substances was not significantly different from that of the wild-type protein (Table 2).