Three Amino Acid Residues Bind Corn Odorants to McinOBP1 in the Polyembryonic Endoparasitoid of Macrocentrus cingulum Brischke

Odorant binding proteins (OBPs) play a central role in transporting odorant molecules from the sensillum lymph to olfactory receptors to initiate behavioral responses. In this study, the OBP of Macrocentrus cingulum McinOBP1 was expressed in Escherichia coli and purified by Ni ion affinity chromatography. Real-time PCR experiments indicate that the McinOBP1 is expressed mainly in adult antennae, with expression levels differing by sex. Ligand-binding experiments using N-phenyl-naphthylamine (1-NPN) as a fluorescent probe demonstrated that the McinOBP1 can bind green-leaf volatiles, including aldehydes and terpenoids, but also can bind aliphatic alcohols with good affinity, in the order trans-2-nonenal>cis-3-hexen-1-ol>trans-caryophelle, suggesting a role of McinOBP1 in general odorant chemoreception. We chose those three odorants for further homology modeling and ligand docking based on their binding affinity. The Val58, Leu62 and Glu130 are the key amino acids in the binding pockets that bind with these three odorants. The three mutants, Val58, Leu62 and Glu130, where the valine, leucine and glutamic residues were replaced by alanine, proline and alanine, respectively; showed reduced affinity to these odorants. This information suggests, Val58, Leu62 and Glu130 are involved in the binding of these compounds, possibly through the specific recognition of ligands that forms hydrogen bonds with the ligands functional groups.


Introduction
Olfaction is a finely tuned sense, essential for sensory assessment of the environment, which plays a crucial role for insects in foraging, host-seeking, mating, ovipositing and avoiding toxin substances. Olfaction is mediated by specific olfactory sensory neurons, which project their dendrites into a lymphatic cavity where odorant binding proteins (OBPs) are present at high concentrations [1]. The major proteins involved in the selectivity and sensitivity of the insect olfactory system are odorant-binding proteins (OBPs) [2] and odorant receptors (ORs) [3]. The small, water soluble polypeptides OBPs are present at the interface between the external environment and chemoreceptors [4]. The first step in the recognition of odorants are air-born small hydrophobic molecules mediated by OBPs [5,6] that are involved in odorant reception, where they bind, solubilize and deliver odorant molecules to ORs [2,7]. ORs are heterodimers comprised of highly variable odorant-binding subunits associated with one conserved co-receptor (OR83b) [8] and these are most widely expressed in the dendritic membranes of olfactory sensory neurons (OSNs) that are housed in sensory hairs called olfactory sensilla [9].
The insect OBP was discovered at the beginning of 1980s in the giant moth Antheraea polyphemus [10]. Many studies related to function and structure have occurred, since their discovery [10,11,12,13], yet mode of action and specific role in chemoreception of OBPs remains largely unknown [4]. In the last few years, several studies have demonstrated requirement of OBPs for chemoreception in insects, which has lead to useful models. Studies demonstrate electrophysiological responses are abolished when A. polyphemus OBPs are removed [14]. Acyrthosiphon pisum OBP3 was shown to interact with its alarm pheromone, (E)-bfarnesol [15]. Recently studies show presence of OBPs increases sensitivity and specificity to the olfactory system of Bombyx mori [16,17]. In Drosophila melanogaster OBP76 LUSH gene knock out, abolishes response to 11-cis-vaccenyl acetate (cVA), but is reactivated when the gene is expressed [18]. So, OBPs are not only involved in odorant detection, but also the discrimination of olfaction stimuli in certain insect behaviors [4]. The specific behavioral response is linked to two genes that encode OBPs, namely 57d and 57e that are found in D. simulans and D. sechellia [19]. The specific OBP was deficient in each of 17 strains of D. melanogaster, which influenced fly responses to several odorants [20].
Interactions between insect OBPs and ligand-binding have been explored by considering key amino acid residues, 3D (dimensional) structures and their physiochemical properties [4,21,22]. Studies show that the pheromone-binding protein 1 (BmorPBP1) of B. mori binds to bombykol when Ser56 interacts with hydroxyl group in the binding pocket [23]. The C-terminal segment of BmorPBP1 is unstructured at pH 7.0, but a major conformational change occurs at pH,7.0, when the seventh a-helix enters into the binding pocket [24]. The acidic pH of the membrane surface triggers the conformational change [4]. Several studies report that this type of conformational change is clearly observed with several OBPs [4,25]. Other OBPs, due to the short length of their Cterminus, cannot make an additional helix and push into the binding cavity, but covers the binding pocket with a lid, possible to contribute to the binding of specific ligands [26,27].
Drosophila LUSH seems to have another mechanism. The single amino acid residue is involved with the conformational change when this protein binds with cVA. LUSH undergoes a conformational change that triggers binding of the complex to the specific olfactory receptor [4], which generates the electrophysiological response of olfactory receptor OR67d in the T1 neuron [28]. This has been very elegantly demonstrated by showing a mutant of LUSH, which mimicks the conformation of the protein when bound to cVA, activates the electrophysiological signal from the corresponding neuron, even in the absence of the pheromone [28].
The Macrocentrus cingulum Brischke (Hymenoptera: Braconidae) is distributed widely in Europe, Japan, Korea and China [29] and is considered as a specialist larval polyembryonic endoparasitoid of the genus Ostrinia (Lepidoptera: Crambidae) [30,31]. Parasitic effectiveness of M. cingulum depends on location and infection by pathogens and other factors like temperature and humidity. Female M. cingulum oviposit in all instars but prefer second to fourth [32]. Parasitization rates are variable, but typically high, for example, reaching up to 80% in studies conducted in Northern France [33,34]. In our present research, we have identified a novel OBP protein gene, McinOBP1 (GenBank accession no. KF900276) from the antennal cDNA library of M. cingulum; which was expressed and purified in vitro. We also have measured the fluorescence binding affinities of McinOBP1 to 23 corn odorants belonging to three groups: aldehydes, terpenoids, and aliphatic alcohols with others aromatic compounds. On the basis of binding affinity results, we predicted the putative binding sites of McinOBP1 by three dimensional structure modeling and molecular docking methods. We confirm these predictions with sitedirected mutagenesis experiments and fluorescence binding assays. The expression patterns of McinOBP1 in different tissues of male and female adults also were detected with real-time quantitative PCR (qPCR).

Insects
M. cingulum were collected from O. furnacalis larvae living on corn plants at the Langfang Experiment Station of the Institute of Plant Protection, Chinese Academy of Agricultural Sciences, China. A laboratory colony was established on host larvae of O. furnacalis that were reared on an artificial diet as described by Zhou et al. [35] and maintained at 25uC with a photoperiod of 16 h:8 h, L:D. Adult parasitoid wasps were fed with 20% honey solution.

RNA Extraction and cDNA Synthesis
Antennae, heads with maxillary palps (excluding antennae), legs, thoraxes and abdomen of female and male individuals were dissected 1-2 days after eclosion and immediately frozen in liquid nitrogen, then stored at -80uC until RNA extraction. Total RNA was extracted from the antennae or other tissues using TRIzol reagent (Invitrogen, Carlsbad, CA, USA), following the manufacturer's instructions. Before transcription, the RNAs were treated with DNase I (Invitrogen) to remove residual genomic DNA. cDNA was prepared from total RNA by reverse transcription, using the RT-PCR system (Promega) in accordance with the manufacturers protocol.
The restriction enzymes are NcoI and HindIII in the forward and reverse primer, are underlined, respectively. The PCR condition consisted of an initial 3 min step at 94uC followed by 37 cycles of 94uC for 30 sec, 56uC for 30 sec and 72uC for 1 min and a final 10 minute step at 72uC. Real-time qPCR experiments were performed using 96 well plates (Applied Biosystems, Carlsbad, CA), ABI Prism 7500 Fast Detection System (Applied Biosystems, Carlsbad, CA) and Brilliant II SYBR Green qPCR master mix (Takara). qRT-PCR was conducted in 20 mL reactions containing 506 SYBR Premix Ex Taq 10 mL, primer (10 mM) 0.462 mL, ROX reference dye II 0.4 mL (506), sample cDNA 1 mL, sterilized ultra-pure grade H 2 O 7.8 mL. Cycling conditions were: 95uC for 30 sec, 40 cycles of 95uC for 05 sec and 60uC for 30 sec. Afterwards, the PCR products were heated to 95uC for 15 sec, cooled to 60uC for 1 min and heated to 95uC for 15 sec to measure the dissociation curves. No-template and no-reverse transcriptase controls were included in each experiment. To check reproducibility, each test sample was done in triplicate technical replicates and three biological replicates. Relative quantification was performed by using the comparative 2 2DDCT method [36]. All data were normalized to endogenous b-actin levels from the same tissue samples and the relative fold change in different tissues was calculated with the transcript level of the abdomen as calibrator. Thus, the relative fold change in different tissues was assessed by comparing the expression level of each OBP in other tissues to that in the abdomen.

Expression Pattern of McinOBP1
For the verified the DCT values, each sample was diluted 5 serial fold (1, 10, 100, 1000, 10000). The amplifications were performed from each dilution in triplicate by using the gene specific primer and b-actin. The absolute values of the slopes of all lines from the template dilution plots (log cDNA dilution vs. DCT) were close to zero (data not shown). Quantification of gene transcript level, in the DDCT calculation for the comparative 2 2DDCT method assumes that the amplification efficiencies of the target and reference are approximately equal.

Cloning and Sequencing
The PCR products were ligated into the pGEM-T Easy vector (Promega, Madison, WI, USA), using a 1:5 (plasmid:insert) molar ratio and incubated overnight at 4uC. The ligation products were transformed into TransT1 E. coli competent cell and grown on LB solid medium with 10 mg/mL ampicillin, 1 M isopropyl b-Dthiogalactoside (IPTG) 40 mL, and X-gal 40 mL. Positive colonies were selected by PCR using the plasmid's primers SP6 and T7, grown in LB liquid medium with ampicillin and sequencing.

Cloning of McinOBP1-Wild Type in Expression Vectors
The pGEM plasmid containing the positive clones and the pET32a (+) were digested with NcoI and HindIII restriction enzymes for 2 hrs at 37uC and then the products were separated on agarose gel. The target fragments were purified from the gel and ligated into the digested pET32a (+) plasmid and the recombinant plasmids were transformed into TransT1 competence cells and grown on LB solid medium with 10 ml ampicillin (10 mg/mL). Selected colonies were grown in LB liquid medium with ampicillin and then incubated in a shaker at 200 rpm, E. coli and transformed into BL21 (DE3)-competent cells. A single clone was identified and cultivated overnight in LB liquid medium including ampicillin on a shaker at 200 rpm in 37uC. The resulting plasmids were sequenced and shown to encode the mature proteins.
Purification of Proteins pET32a (+) vectors containing the verified sequences encoding the McinOBP1-wt and mutants proteins were grown overnight in 5 mL LB with 10 mg/mL ampicillin. The cultures were diluted to 1:100 in fresh medium with ampicillin and then shaken at 200 rpm for 2.5 hrs at 37uC. Then, 0.5 mM IPTG was added to the culture and incubated for additional 5 hrs for the induction of target proteins. The bacterial cells were collected by centrifugation and re-suspended in 16 phosphate-buffered saline (PBS), then sonicated; the pellet and supernatant were collected by centrifuge and analyzed by 12% SDS-PAGE. Proteins were present both in supernatant and inclusion body. We chose the supernatant expressed proteins for further uses. Proteins were purified by means of Ni ion affinity chromatography then desalted and concentrated. The purified proteins were digested with rProtease to remove the His-tag. The size and purity of proteins were checked by 12% SDS-PAGE analysis. The concentrations of the proteins were measured along with Bradford method [37] by the microplate reader Multiskan Ascent (Thermo Labsystem, Germany).

Preparation of Site-Directed Mutants
The three mutants of McinOBP1, m1 (mutation of amino acid, valine to alanine at 58 position), m2 (mutation of amino acid, leucine to proline at 62 position) and m3 (mutation of amino acid, glutamic to alanine at 130 position), were developed using the kit of Quick-change lightning site-directed mutagenesis (Stratagene, La Jolla, CA, USA). The McinOBP1-wt/pGEM-T Easy construct was used as a template. The primers were designed manually and mutation sites are underlined as listed below.
Val58 The PCR conditions were 95uC for 3 min for initial denaturation, followed by 37 cycles of 95uC for 30 sec, 57uC for 1 min and final extension at 72uC for 10 min. The selected mutants were subcloned into pGEM-T Easy vector (Promega) then sequenced. The same expression vector and competent cells were used as for McinOBP1-wt. The mutant's expression and purification were performed as described for the wild-type protein.

Fluorescence Binding Assays
Fluorescence binding assays were recorded on Shimadzu fluorescence spectrophotometer (Shimadzu, Japan) with a 1 cm light path quartz cuvette; both excitation and emission silts widths were 5 nm. The probe was excited at 337 nm and emission spectra were recorded between 350 and 500 nm.

Molecular Modeling and Ligands Docking
A three dimensional model of McinOBP1 was generated using the I-TASSER Protein Structure and Function Predictions web server (http://zhanglab.ccmb.med.umich. edu/I-TASSER/) [40,41]. The sequence of McinOBP1 was compared to all known protein sequences that had a high sequence similarity with the target sequence on RCSB Protein Data Bank (PDB) (www.rcsb. org). The homology modeling of McinOBP1 was performed using build homology model of AaegOBP1 [27]. The best model was confirmed using the evaluation of PDF total energy, verify score and Ramachardran plot.
Molecular docking can fit molecules together in a favorable configuration to form a complex system. Three dimensional chemical structures were used to search for trans-2-nonenal, cis-3hexen-1-ol and trans-caryophelle on PubChem Compound (www. pubchem.ncbi.nlm.hih.gov), where the structural information from the theoretically modeled complex may helped clarify the binding mechanism between McinOBP1 and odorants. The advanced docking program CDOCKER [42] was used to perform the automated molecular docking to gain insights into binding mode of McinOBP1 with these three odorants. CHARMm-based molecular dynamics scheme was used to dock ligands into a receptor binding site in the CDOCKER program. The energy minimization was used to refine the ligand poses. The ligandprotein docked complexes were selected according to their interaction energy with geometric matching quality. Interaction energy was calculated from odorants and key residues of McinOBP1.

Coding and Sequence Analysis of McinOBP1
McinOBP1of the M. cingulum was obtained from the antennal cDNA library is a protein specifically expressed in antennae. By using gene specific primers, a full length cDNA encoding McinOBP1 was cloned. Sequence analysis showed that the full length (ORF) consists of 423 nucleotides that encode 140 amino acids residues, with a predicted MW of 15.71 kD. For McinOBP1 SingnalP predicted a peptide with 20 amino acid residues, with calculated molecular weight and isoelectric point of 13.46 kDa and 4.73, respectively. An alignment of the amino acid sequences of McinOBP1 with the corresponding OBP from other species of Hymenoptera and additional insects is shown in Figure 1A.
McinOBP1 had the typical six-cysteine signature of OBPs with a common pattern: C 1 -X 15-39 -C 2 -X 3 -C 3 -X 21-44 -C 4 -X 7-12 -C 5 -X 8 -C 6 , (X is any amino acid). The phylogenetic tree was constructed based on the amino acid sequences of OBP from other insects (Fig. 1B). The dendrogram shows that the McinOBP1 has closer ancestry with insects from the same order. This suggests McinOBP1 is a classical OBP.

Tissues-Specificity Expression Pattern of McinOBP1
We examined the expression pattern of McinOBP1 in different tissues (male and female antennae, de-antennated heads, legs, thoraxes, and abdomens) by qPCR. Total RNA of each sample was isolated and separated from male and female that was reverstranscribed; the expected product was largely amplified in male and female antennae, with low transcripts in other tissues, suggesting that McinOBP1 is mainly expressed in adult antennae, with expression levels differing by sex (Fig. 2). In general, the levels of transcripts were very low in all tissues except the antennae, where McinOBP1 was highly expressed.

Recombinant Protein McinOBP1-wt Expression and Purification
The recombinant McinOBP1-wt (wild type) protein was expressed in E. coli as a soluble protein in supernatant with high yield (,10 to 20 mg/L). The recombinant protein was directly purified from the total protein (supernatant) by the Ni ion affinity chromatography. The His-tag was digested from the purified recombinant protein by rProtease, and then Ni ion affinity chromatography was used to separate His-tagged and uncleaved His-tagged protein (Fig. 3). The purified recombinant protein, McinOBP1-wt was then used for the further experiments in this study.

Ligand-Protein Binding Specificity
The binding affinity of the fluorescence probe 1-NPN (Nphenyl-1-naphthylamine) to protein was measured with the excitation wavelength at 337 nm and emission spectra were recorded between 350 and 500 nm. The dissociation constant of 1-NPN-bound recombinant McinOBP1-wt, approximately 6.60 mM, was calculated according to the changes in fluorescence intensity (Fig. 4A), which was used to calculate the dissociation constants (K D ) of ligands.
The competitive binding affinities of McinOBP1-wt to 23 ligands belonging to aldehyde, terpenoids and aliphatic alcohol with others aromatic compounds are listed in Table 1; and curves for a few representative ligands tested are shown in Figure 4B-4D.  Table 1 lists the IC 50 values (concentration of ligand that reduces by half the initial fluorescence value), and when possible, the calculated dissociation constants (K D ) for OBP/ligand combination.

Molecular Modeling of McinOBP1-wt Protein and Ligands Docking
Sequence of McinOBP1-wt was compared to all known proteins with AaegOBP1 as template in order to model the 3D structure of McinOBP1-wt with I-TASSER. Following the homology modeling, the best model (Fig. 6) was chosen from 10 candidates, and its quality was further checked by Ramachardran plot and verification score (Data not shown). C-score for McinOBP1 of 0.24 (ranges 25 to 2) suggests high confidence in the model [41]. Considering the three dimensional model of McinOBP1, most of the residues were confirmed to have hydrophobicity at their binding site. Ligand poses and consensus score programs were used to evaluate binding pose affinities for the residues. Subsequently, the optimal 3D binding conformations of protein ligand complexes were selected Figures 7A-7C. The interaction energies between key residues and the ligands are listed in Table 2. All of the residues are located in the cavity formed by six helices. From the docking simulation results, several residues including Phe54, Val58, Phe61, Leu62, IleE65, Ile67, Phe74, Ala79, Val83, Leu87, Ser91, Ile95, Leu121, Leu128 and Glu130 seemingly play important roles in the binding of McinOBP1-wt to trans-2nonenal, cis-3-hexen-1-ol and trans-caryophelle.

Mutagenesis and Binding Assays
On the basis of molecular modeling and ligand docking simulation results, Val58, and Glu130 were mutated to alanine, and Leu62 was mutated to proline. These recombinant mutant proteins were expressed and purified as described for the wild type protein and analyzed by SDS-PAGE. The same fluorescence probe 1-NPN was used under the previously described conditions. The mutants Val58, Leu62 and Glu130 bind 1-NPN dissociation constants of 14.6261.07, 13.6960.71, and 6.5360.50 mM,    Aliphatic alcohols and others  respectively, (Fig. 8A). This was calculated from the ligand dissociation constants (K D ) shown in Table 3.

Discussion
Insects recognize complex environmental stimuli such as volatile and non-volatile compounds with their well-developed olfactory system. This system plays an important role in several behaviors, such as searching for hosts, food, mates, evading predators and locating oviposition sites. Insect OBPs are very diverse proteins with an average of only 14% amino acid identity [43]. OBPs are soluble polypeptides present at the interface between the external environment and chemoreceptors for insects and other animal species [4]. McinOBP1 gene was obtained from our previous study about the construction of antennal cDNA library of M. cingulum (unpublished). Alignment analysis of McinOBP1 shows it has 6 typical conservative cysteines in the sequence (Fig. 1A), which is consistent with previous reports [44]. Quantitative examination of transcript levels shows McinOBP1 is expressed in male and female antennae but there is very low if any expression in others tissues (Fig. 2). These results are similar to those found in other insects [45,46,47,48]. Some of these studies suggested that McinOBP1 plays an important role in the olfactory systems response to chemical stimuli.
In this study to better understand McinOBP1 structure, function and binding characteristics, site directed mutagenesis and a fluorescence binding specificity were conducted. Odorants analogues of chemicals released from damaged corn plants of Asian corn borer (ACB), Ostrinia furnacalis (Guenée) [49] were selected as ligands. A total 23 odorants were tested, including ten aldehydes, nine terpenoids and four aliphatic alcohols/aromatic compounds. Three odorant compounds (decanal, undecanal and 2,6 dimethyl octane) have no binding affinity to McinOBP1. Among the tested compounds, trans-2-nonenal, trans-2-octenal, trans-2-hexenal acetate, trans-caryophelle, b-pinene, b-ionone, and cis-3-hexen-1-ol have binding affinities to McinOBP1. In general, ''b'' isomer odorants have higher binding affinities than ''a'' isomer, which suggests McinOBP1 could discriminate the chiral structure of chemical molecules. Future research, however, needs to investigate whether odorant solubility affects binding capacity. Furthermore, a limitation to note is fluorescence assays only provide an indirect measure of binding affinity [50,51].
Our modeling suggests trans-2-nonenal, cis-3-hexen-1-ol and trans-caryophelle bind to McinOBP1 in a cisoid conformation similar to bombykol in BmorGOBP2 [21]. Formation of hydrogen bonds of hydrophilic amino acid residues are involved in the recognition of ligands [50]. Hydrogen bonds make networks to hold the ligand in the central cavity through the locked Cterminus. Similar results were reported with PBP-ligand interactions in Lymantria dispar [56]. In the moth Helicoverpa armigera, HarmOBP7 binds with good affinity to both pheromone components Z-11-hexadecenal and Z-9-hexadecenal as well as linear aldehydes, alcohols and esters.
Binding affinity of insect OBP is influenced by pH. A study reported that binding affinities of HarmOBP7 and a mutant lacking the last 6 residues do not substantially decrease in acidic conditions, but increases at basic pH values, with no significant differences between wild-type and mutant [4]. Furthermore, a second mutant, where C-terminus one (Lys123) of the three lysine residues was replaced by methionine showed reduced affinity to pheromone components, as well as to their analogues due to less hydrogen bonds formation affinity [4]. pH-dependent ligandrelease mechanisms were reported in Culex quinquefasciatus, CquiOBP1 [57], Aedes aegypti, AaegOBP1 [27] and Anopheles gambiae, AgamOBP1 [58], in which a decrease in pH, disrupts the hydrogen bond between the C-terminal loop and the rest of the protein, and then opens the loop to release the ligands [50].
We measured binding affinities of trans-2-nonenal, cis-3-hexen-1-ol and trans-caryophelle to the mutant McinOBP1-m1 and m2 (Fig. 8B). Unlike 1-NPN, whose binding to selected odorants was not affected by the change of the respective amino acids, mutant McinOBP1 showed poorer affinities compared to the wild type McinOBP1. The trans-caryophelle showed the weakest binding affinity with McinOBP1-m1 whereas trans-2-nonenal had the lowest binding affinity with McinOBP1-m2 (Fig. 8B). The third mutant McinOBP1-m3 binds 1-NPN with a dissociation constant These three mutants showed lower binding affinity to selected three odorants with respect to the wild type, indicating that the respective amino acids are involved in binding these ligands. doi:10.1371/journal.pone.0093501.g008 Table 3. Affinities of McinOBP1-wt and three mutants to trans-2-nonenal, cis-3-hexen-1-ol and trans-caryophelle.

Proteins
Trans-2-Nonenal Cis-3-Hexen-1-ol Trans-Caryophelle IC 50  slightly better (K D = 6.5360.50 mM) than the wild type, indicating that the mutation did not significantly affect binding and that Glu130 is not involved in binding 1-NPN (Fig. 8A). However, binding of these three odorants to McinOBP1-m3 was much weaker than to the wild type (Fig. 8A). These results suggest Glu130 is involved in binding to aldehyde odorants, as well as alcohol and terpenoid odorants. The molecular modeling of three mutants, McinOBP1m1-m3 did not change the overall confirmation of the protein structure, McinOBP1-wt (supplementary Fig.  S1). These results also confirm the observation, reported with other OBPs [59,60], that different ligands bind equally well to the OBP. Presumably, different orientations in the binding pocket make it possible to for, different residues to bind. These results suggest McinOBP1 is a general OBP that plays an important role in recognizing several plant volatiles. The three amino acids Val58, Leu62 and Glu130 appear to be involved in binding the aldehyde trans-2-nonenal as aldehyde and the terpenoid trans-caryophelle. Seemingly these compounds interact with McinOBP1 with their hydrophobic chain inside the binding pocket and the functional group on the mount of the cavity. Future investigations with NMR tests or X-ray diffraction are needed to confirm these ligand-OBP interaction findings.