The parasitoid wasp, M. cingulum Brischke were reared at 25°C, in 16 h light:8 h dark on the host insect, O. furnacalis in the laboratory. Adult parasitoid wasps were fed with 20% honey solution. The antennae of 3 days-old male and female (200 of each sex) adults were excised at the base and immediately transferred into tubes immersed in liquid nitrogen. The preparation of all other samples including the head (without antennae), thorax, leg, and abdomen were carried out in the same way. Prepared samples were stored at −80°C until used. The total RNA of each sample was isolated according to the manual of TRIzol reagent (Invitrogen, Carlsbad, CA, USA). The RNA integrity was verified by 1% agarose gel electrophoresis and quantity was assessed with a Nanodrop ND-2000 spectrophotometer (NanoDrop products, Wilmington, DE, USA). First-strand cDNAs were synthetized by reference to the handbook of EasyScript First-Strand cDNA Synthesis SuperMix (TransGen Biotech, Beijing, China), and employed as templates for latter PCR amplification.

The McinOBP4 has been identified from the antennal transcriptome analysis (GenBank no. KY270854). The SignalP 3.0 program (http://www.cbs.dtu.dk/services/SignalP/) was used to predicted signal peptides from N-terminal. The McinOBP4 and other insects OBPs amino acids sequences were align with ClustalX 1.83. MEGA 5.2 software was used to construct phylogenetic tree followed by the neighbor-joining method (Tamura et al., 2007).

One μL cDNA with gene specific primers were amplified in a TC-5000 thermal cycler (Bibby Scientific Ltd, UK) as follows:

McinOBP4-sense: 5′- CGGATCCAAGCGACCAGATTTC -3′ and

The BamHI and HindIII are employed in the sense and antisense primers, respectively. The PCR amplification conditions are 3 min at 94°C for denaturation, 30 s at 94°C, 30 s at 57°C and 1 min at 72°C for 37 cycles and last 10 min at 72°C.

The pGEM-T easy vector was used into PCR gel extraction products ligation and kept at 4°C for overnight. The ligation products were transformed into Trans1-T1 competent cell (TransGen Biotech, Beijing, China) which belonged to E. coli. Positive colonies were selected by PCR using with M13 (F and R) primers, and cultured in ampicillin enriched LB liquid medium. The custom sequencing was at Sangon Biotech, Beijing, China.

The McinOBP4 sequence with endonuclease restriction sites BamHI and HindIII were incorporated into PET-32a(+), and the generated constructs were transformed into BL21 (DE3) E. coli strains and cultured on ampicillin enriched solid LB medium. Positive clone was grown in medium with ampicillin at 220 rpm in 37°C for 8 h.

Recombinant McinOBP4 protein was induced with 0.5 mM Isopropyl β-D-1-thiogalactopyranoside (IPTG) under the condition of 37°C, 160 rpm, and purification of the recombinant McinOBP4 was initiated by harvesting of McinOBP4-expressing DE3 cells from 1 liter cultures by centrifugation. All expressed proteins were found in the supernatant, and the recombinant proteins were purified by Ni ion affinity chromatography (GE-Healthcare, Sangon Biotech Co., Ltd., Shanghai, China). SDS-PAGE was used to monitor protein expression and purification. The purified proteins were stored at −20°C in 50 mM Tris-HCl buffer with a pH of 7.4.

The C-terminal tail truncated form of the McinOBP4 was obtained by removing the last 8 amino acids and inserting a stop codon after Leu112. The McinOBP4-Wt/pGEM-T Easy construct was used as a template. A pair of primer was designed manually and as listed sense primer 5′- CGGATCCAAGCGACCAGATTTC -3′ (underlined with BamHI) and antisense primer, 5′- CGAAGCTTTTACTTCGATTGCAC -3′ (underlined with HindIII) where 113th position residues used as a stop codon. The mutant M2 (Met119) was prepared along with the pair of primer as sense primer 5′- CGGATCCAAGCGACCAGATTTC -3′ (under lined with BamHI), and antisense primere 5′- CGAAGCTTTTACACCAAGAACCACATC -3′ (under lined with HindIII). The following reaction conditions were employed: 4 min at 94°C for initial denaturation, followed by 36 cycles of 30 s at 94°C, 30 s at 57°C and 1 min at 72°C for 36 cycles and 10 min at 72°C. The selected mutant was cloning, expression; protein induction and purification were done according to our previous works (Ahmed et al., 2014).

The qRT-PCR was employed to measure the tissue-specific transcripts level of McinOBP4 by using cDNA from antennae, head, leg and abdomen from female and male wasps. The gene specific primers, F: 5′-GCGACCAGATTTCGTGACT-3′; R: 5′-TGTCGTTTGGAATATCGCC-3′ were used. The ABI 7500 fast real-time PCR system (Applied Biosystems, USA) were employed according to our previous study (Ahmed et al., 2014).

One cm light path fluorimeter quartz cuvette was used on a FluoroMax-4 (Horiba Scientific, USA) spectrophotometer at room temperature for fluorescence spectra recording and excitation wavelength of 337 nm was chosen and 300–500 nm for the emission with a 5 nm slit width. Spectra were recorded with 2 μM McinOBP4 in 50 mM Tris-HCl of different pH value, while ligands were added as 1 mM methanol solutions.

To measure the affinity of the fluorescence ligand N-phenyl-1-naphthylamine (1-NPN) to McinOBP4-wt and mutants, a 2 μM solution of the protein in 50 mM Tris-HCl, pH 7.4, was titrated with aliquots of 1 mM ligand in high-performance liquid chromatography (HPLC) purity grade methanol to final concentrations of 2–20 μM. The probe was excited at 337 nm and emission spectra were recorded between 300 and 500 nm. While the other competitor were measured in competitive binding assays, where 1-NPN using as the fluorescent reporter at 2 μM concentration and each competitor (odorants) over concentration ranges 2–14 μM, depending on the different odorants.

It was assumed that McinOBP4 was 100% active and the protein-ligand binding was 1:1 ratio at saturation. The obtained data were processed with Graphpad Prism software (Graphpad Software, Inc.) and using a software package SPSS Version 16 (http://www.spss.com). The maximum intensity values of emission were plotted against the 1-NPN concentration for the K_D determination, and the values of intensity were used for the evaluation the bound ligands. The curve was linearized using Scatchard Plots to verify the confidence of K_D. IC₅₀ for other corn odorants were calculated using with Microsoft excel 2010, and the equation, K_D = [IC₅₀]/[1+(1–NPN)/K_1−NPN] was used for the K_D values calculation. In this equation [IC₅₀] stands for the ligand concentration where the ligand quenching the fluorescence intensity of 1-NPN to 50%, [1-NPN] and K_1−NPN mean free 1-NPN concentration and McinOBP4/1-NPN complex K_D values, respectively (Campanacci et al., 2001; Ban et al., 2002).

Three dimensional model of McinOBP4 was generated through the I-TASSER web server (http://zhanglab.ccmb.med.umich.edu/I-TASSER/.) (Wu et al., 2003; Roy et al., 2010). The protein sequences on RCSB Protein Data Bank (PDB) (http://www.rcsb.org) were employed to compare the McinOBP4 sequence. The AmelPBP1 (Leite et al., 2009) model was used as template for the structure of McinOBP4. Molecular dynamics program CDOCKER (Wu et al., 2003) was used to investigate the binding characteristics of two corn odorants, limonene, and trans-3-hexen-1-ol-acetate at structural levels and total interaction energy was calculated from these two odorants and respective residues of McinOBP4.

McinOBP4 was identified in the antennal transcriptomic analysis, is a protein highly expressed in olfactory tissues specifically expressed in antennae. Through the biochemical approach full length McinOBP4 was cloned. McinOBP4 encoded 143 amino acids residues including a 22-aa-long signal peptide, with a predicted MW of 15.71 kD. The theoretical isoelectric points and molecular weight of the mature proteins were 4.00 and 13.43 kD, respectively. The amino acid sequences alignment of McinOBP4 with the Hymenopteran and other insects (Figure 1A). To analyze the phylogenetic relationship of McinOBP4 with other insects, we constructed phylogenetic tree comprising 24 OBPs from different insect species (Figure 1B). McinOBP4 was phylogenetically closest to MmedPBP1 from Microplitis mediator, which belongs to the same order. This result suggested that McinOBP4 is a classical OBP/PBP, need to further study to confirm.

Real-time quantitative PCR (RT-qPCR) was employed to check the transcripts level of McinOBP4 in male and female antennae, de-antennated heads, legs, thoraxes, and abdomens. Tissue specific tRNA were isolated from male and female wasps, and the reverse transcribed. The RT-qPCR results showed that the target protein was largely expressed in antennae of both sex, with low in other tissues, might be the McinOBP4 was specifically expressed in adult antennae (Figure 2). Generally, very low transcripts levels were in other test tissues except for the adult antennae.

We measured affinities of 1-NPN with purified McinOBP4. According to the changes in fluorescence intensity, McinOBP4 bound strongly to 1-NPN with binding affinity of 5.69 ± 0.83a μM (Figure 3A). The binding affinities of McinOBP4-Wt to corn odorants were tested belongs the three groups. List of odorants and resulted affinities curves are shown in Table 1 and Figures 3B–D, respectively. The results suggested that most of the selected corn odorants displaced 1-NPN from the McinOBP4-Wt/1-NPN complex. Among the tested compounds, limonene and trans-3-hexen-1-ol-acetate had the highest affinities to McinOBP4-Wt with K_D values of 4.12 ± 0.70 and 6.99 ± 1.23 μM respectively (Figures 3B,D). The linalool, trans-2-octanal, and β-ionone had medium affinities to McinOBP4-wt, with K_D of 9.61 ± 0.23, 11.78 ± 0.57, and 11.92 ± 0.49 μM, respectively.

Additionally, we investigated the affinity of 1-NPN to McinOBP4-Wt and McinOBP-M1 over the pH range from 4.4 to 9.4. Binding curves are showed in Figures 3A–D at pH 7.4 and the wild and mutant proteins affinities at several pH levels (Figure 4B). We did not observe significant loss of binding activity in acidic condition (pH 4.4) while a basic condition increase in affinity was measured. At different pH levels, the C-terminal truncated protein has no effect of their activities. Moreover, we have measured binding affinity of the corn odorant to the mutant McinOBP4-M1 (Figure 4C). The corn odorants, showed poor binding abilities to the M1 mutant compared to the McinOBP4-Wt, the results unlike to 1-NPN, which affinity was not depends on the segments of C-terminus.

We prepared a second mutant to identify one of key amino acid residue by mutated Met119 to leucine, predicted candidates forming a Schiff base with the aldehyde group or hydrogen bonds with other odorants compounds. We reasoned that Phe118 could establish an electrostatic interaction with Met116, and Met113 could be linked to Glu115. This would leave Met119 free to interact with the odorants. Furthermore, limonene and trans-3-hexen-1-ol-acetate were bound in a large flask-shaped hydrophobic pocket and the Met13, Gly38, Lue53, Val59, Glu62, Val67, Leu71, Phe74, Ile79, Gly83, Leu87, Glu93, and Met119 residues were take part in binding to these corn odorants. Residues and odorants interaction energy are shown in Table 2. By the total interaction energy comparison, limonene was showed a more interaction energy than trans-3-hexen-1-ol-acetate, especially at Met119 (Table 2). This phenomenon leads higher binding affinity (lower K_D) of McinOBP4-limonene complex. Additionally, validated key binding sites through the sequence alignment of McinOBP4 to other insects OBPs such as Val14 of LstiGOBP1 (Yin et al., 2015), IIe80 to HoblOBP1 (Zhuang et al., 2014) and V87A to LmigOBP1 (Jiang et al., 2009; alignment results not shown in here).

The same fluorescence probe 1-NPN was used under the previously described conditions. The M2 bound 1-NPN with a K_D of 8.85a μM, little more than the McinOBP4-Wt; it indicated that McinOBP4-M2 and 1-NPN interaction was not influenced by the mutation (Figure 4A). Binding affinities of the mutants M2 was then evaluated with five corn odorants namely limonene and trans-3-hexen-1-ol-acetate, linalool, trans-2-octanal, and β-ionone and are shown in Figure 4C. Compared to wild type and first mutant, the second mutant showed lower binding affinities to the tested odorants.

McinOBP4 sequence was compared to all known proteins sequences in the protein data bank through a Blastp search; the AmeliBP1 as template 3D structure to the McinOBP4 in I-TASSER. The best model was chosen from 10 candidates through homology modeling. The verification score and Ramachardran plot were used for assessed the quality of the best model (data not shown). McinOBP4 model C-score were 0.83 (ranges −5 to 2) suggests a high level of confidence in the predicted model (Roy et al., 2010). Estimated TM-score and RMSD of predicted model McinOBP4 are 0.83 ± 0.08 and 2.7 ± 2.0 A°, respectively. The 3D structure of McinOBP4 (Figure 5) was composed with six α-helices which are located into the 8–24 (α1), 29–36 (α2), 46–57 (α3), 68–74 (α4), 77–89 (α5), and 98–111 (α6) residues. Four (α1, α2, α3, α4, and α5) antiparallel helices were converged to form a large flask-shaped hydrophobic binding cavity with the slender opening and the closing being capped with the helix, α6. The hydrophobic binding pocket of McinOBP4 was formed by the residues of leucine phenylalanine, tryptophan, valine, and methionine).

The authors declare that the research was conducted in the absence of any commercial or financial relationships that could be construed as a potential conflict of interest.