A Highly Expressed Antennae Odorant-Binding Protein Involved in Recognition of Herbivore-Induced Plant Volatiles in Dastarcus helophoroides

Natural enemies such as parasitoids and parasites depend on sensitive olfactory to search for their specific hosts. Herbivore-induced plant volatiles (HIPVs) are vital components in providing host information for many natural enemies of herbivores. However, the olfactory-related proteins involved in the recognition of HIPVs are rarely reported. In this study, we established an exhaustive tissue and developmental expression profile of odorant-binding proteins (OBPs) from Dastarcus helophoroides, an essential natural enemy in the forestry ecosystem. Twenty DhelOBPs displayed various expression patterns in different organs and adult physiological states, suggesting a potential involvement in olfactory perception. In silico AlphaFold2-based modeling and molecular docking showed similar binding energies between six DhelOBPs (DhelOBP4, 5, 6, 14, 18, and 20) and HIPVs from Pinus massoniana. While in vitro fluorescence competitive binding assays showed only recombinant DhelOBP4, the most highly expressed in the antennae of emerging adults could bind to HIPVs with high binding affinities. RNAi-mediated behavioral assays indicated that DhelOBP4 was an essential functional protein for D. helophoroides adults recognizing two behaviorally attractive substances: p-cymene and γ-terpinene. Further binding conformation analyses revealed that Phe 54, Val 56, and Phe 71 might be the key binding sites for DhelOBP4 interacting with HIPVs. In conclusion, our results provide an essential molecular basis for the olfactory perception of D. helophoroides and reliable evidence for recognizing the HIPVs of natural enemies from insect OBPs’ perspective.


Introduction
The insect olfactory system is essential for maintaining an insect's life. Many important physiological behaviors, such as host-seeking, mating, and oviposition, primarily depend on sensitive olfactory organs, mainly the antennae [1][2][3]. In these processes, odorants in the environment, such as pheromones and plant volatiles, are considered vital semiochemicals [4,5]. Among them, herbivore-induced plant volatiles (HIPVs) are essential plant secondary metabolite that provide information on the host's location to natural enemies [6][7][8][9].

Odorant-Binding Proteins in D. helophoroides
For identification and classification of DhelOBPs, multiple amino acid sequence alignments were generated using the MAFFT program (https://www.ebi.ac.uk/Tools/msa/mafft/, accessed on 11 March 2021) using 23 DhelO-BPs (Table S2) and 50 OBPs from Tribolium castaneum [36] ( Figure S1). Based on the above alignment results, we provided more concise alignment results in Figure 2. Most DhelOBPs belong to the classic OBP subfamily and share a conserved cysteine pattern (C1X24-29C2X3C3X22-43C4X8-10C5X8C6) comparable to T. castaneum and other insects. However, one of the classic OBP, DhelOBP22, is similar to TcOBP8B, which has an extra cysteine of 16 AA before C1 and another additional cysteine of 8 AA after C6. Four DhelO-BPs (DhelOBP5, DhelOBP6, DhelOBP11, and DhelOBP22) belong to the Minus-C OBP subfamily and also show a conserved cysteine pattern, with only four cysteines lacking C2 and C5 (C1X18-30C3X38-39C4X16-17C6). The one Plus-C OBP, DhelOBP10, has an expanded N and C terminus containing six additional cysteines and a typical proline residue following C6. DhelOBP2 is a partial coding sequence without the C-terminal; we speculated that it might be a Minus-C OBP because it lacks C2 based on the current alignment. DhelOBP1 and DhelOBP23 were identified as complete coding sequences but was not consistent with the conserved sequence motif of any OBP family. As a result, we removed these three DhelOBPs (DhelOBP1, DhelOBP2, and DhelOBP23) in the subsequent studies.

Odorant-Binding Proteins in D. helophoroides
For identification and classification of DhelOBPs, multiple amino acid sequence alignments were generated using the MAFFT program (https://www.ebi.ac.uk/Tools/msa/mafft/, accessed on 11 March 2021) using 23 DhelO-BPs (Table S2) and 50 OBPs from Tribolium castaneum [36] ( Figure S1). Based on the above alignment results, we provided more concise alignment results in Figure 2. Most DhelOBPs belong to the classic OBP subfamily and share a conserved cysteine pattern (C1X24-29C2X3C3X22-43C4X8-10C5X8C6) comparable to T. castaneum and other insects. However, one of the classic OBP, DhelOBP22, is similar to TcOBP8B, which has an extra cysteine of 16 AA before C1 and another additional cysteine of 8 AA after C6. Four DhelO-BPs (DhelOBP5, DhelOBP6, DhelOBP11, and DhelOBP22) belong to the Minus-C OBP subfamily and also show a conserved cysteine pattern, with only four cysteines lacking C2 and C5 (C1X18-30C3X38-39C4X16-17C6). The one Plus-C OBP, DhelOBP10, has an expanded N and C terminus containing six additional cysteines and a typical proline residue following C6. DhelOBP2 is a partial coding sequence without the C-terminal; we speculated that it might be a Minus-C OBP because it lacks C2 based on the current alignment. DhelOBP1 and DhelOBP23 were identified as complete coding sequences but was not consistent with the conserved sequence motif of any OBP family. As a result, we removed these three DhelOBPs (DhelOBP1, DhelOBP2, and DhelOBP23) in the subsequent studies. 9 Terpinen

Odorant-Binding Proteins in D. helophoroides
For identification and classification of DhelOBPs, multiple amino acid sequence alignments were generated using the MAFFT program (https://www.ebi.ac.uk/Tools/msa/mafft/, accessed on 11 March 2021) using 23 DhelO-BPs (Table S2) and 50 OBPs from Tribolium castaneum [36] ( Figure S1). Based on the above alignment results, we provided more concise alignment results in Figure 2. Most DhelOBPs belong to the classic OBP subfamily and share a conserved cysteine pattern (C1X24-29C2X3C3X22-43C4X8-10C5X8C6) comparable to T. castaneum and other insects. However, one of the classic OBP, DhelOBP22, is similar to TcOBP8B, which has an extra cysteine of 16 AA before C1 and another additional cysteine of 8 AA after C6. Four DhelO-BPs (DhelOBP5, DhelOBP6, DhelOBP11, and DhelOBP22) belong to the Minus-C OBP subfamily and also show a conserved cysteine pattern, with only four cysteines lacking C2 and C5 (C1X18-30C3X38-39C4X16-17C6). The one Plus-C OBP, DhelOBP10, has an expanded N and C terminus containing six additional cysteines and a typical proline residue following C6. DhelOBP2 is a partial coding sequence without the C-terminal; we speculated that it might be a Minus-C OBP because it lacks C2 based on the current alignment. DhelOBP1 and DhelOBP23 were identified as complete coding sequences but was not consistent with the conserved sequence motif of any OBP family. As a result, we removed these three DhelOBPs (DhelOBP1, DhelOBP2, and DhelOBP23) in the subsequent studies.

Odorant-Binding Proteins in D. helophoroides
For identification and classification of DhelOBPs, multiple amino acid sequence alignments were generated using the MAFFT program (https://www.ebi.ac.uk/Tools/msa/mafft/, accessed on 11 March 2021) using 23 DhelO-BPs (Table S2) and 50 OBPs from Tribolium castaneum [36] ( Figure S1). Based on the above alignment results, we provided more concise alignment results in Figure 2. Most DhelOBPs belong to the classic OBP subfamily and share a conserved cysteine pattern (C1X24-29C2X3C3X22-43C4X8-10C5X8C6) comparable to T. castaneum and other insects. However, one of the classic OBP, DhelOBP22, is similar to TcOBP8B, which has an extra cysteine of 16 AA before C1 and another additional cysteine of 8 AA after C6. Four DhelO-BPs (DhelOBP5, DhelOBP6, DhelOBP11, and DhelOBP22) belong to the Minus-C OBP subfamily and also show a conserved cysteine pattern, with only four cysteines lacking C2 and C5 (C1X18-30C3X38-39C4X16-17C6). The one Plus-C OBP, DhelOBP10, has an expanded N and C terminus containing six additional cysteines and a typical proline residue following C6. DhelOBP2 is a partial coding sequence without the C-terminal; we speculated that it might be a Minus-C OBP because it lacks C2 based on the current alignment. DhelOBP1 and DhelOBP23 were identified as complete coding sequences but was not consistent with the conserved sequence motif of any OBP family. As a result, we removed these three DhelOBPs (DhelOBP1, DhelOBP2, and DhelOBP23) in the subsequent studies.

Odorant-Binding Proteins in D. helophoroides
For identification and classification of DhelOBPs, multiple amino acid sequence alignments were generated using the MAFFT program (https://www.ebi.ac.uk/Tools/msa/mafft/, accessed on 11 March 2021) using 23 DhelO-BPs (Table S2) and 50 OBPs from Tribolium castaneum [36] ( Figure S1). Based on the above alignment results, we provided more concise alignment results in Figure 2. Most DhelOBPs belong to the classic OBP subfamily and share a conserved cysteine pattern (C1X24-29C2X3C3X22-43C4X8-10C5X8C6) comparable to T. castaneum and other insects. However, one of the classic OBP, DhelOBP22, is similar to TcOBP8B, which has an extra cysteine of 16 AA before C1 and another additional cysteine of 8 AA after C6. Four DhelO-BPs (DhelOBP5, DhelOBP6, DhelOBP11, and DhelOBP22) belong to the Minus-C OBP subfamily and also show a conserved cysteine pattern, with only four cysteines lacking C2 and C5 (C1X18-30C3X38-39C4X16-17C6). The one Plus-C OBP, DhelOBP10, has an expanded N and C terminus containing six additional cysteines and a typical proline residue following C6. DhelOBP2 is a partial coding sequence without the C-terminal; we speculated that it might be a Minus-C OBP because it lacks C2 based on the current alignment. DhelOBP1 and DhelOBP23 were identified as complete coding sequences but was not consistent with the conserved sequence motif of any OBP family. As a result, we removed these three DhelOBPs (DhelOBP1, DhelOBP2, and DhelOBP23) in the subsequent studies.

Odorant-Binding Proteins in D. helophoroides
For identification and classification of DhelOBPs, multiple amino acid sequence alignments were generated using the MAFFT program (https://www.ebi.ac.uk/Tools/msa/ mafft/, accessed on 11 March 2021) using 23 DhelOBPs (Table S2) and 50 OBPs from Tribolium castaneum [36] ( Figure S1). Based on the above alignment results, we provided more concise alignment results in Figure 2. Most DhelOBPs belong to the classic OBP subfamily and share a conserved cysteine pattern (C1X 24-29 C2X 3 C3X 22-43 C4X 8-10 C5X 8 C6) comparable to T. castaneum and other insects. However, one of the classic OBP, DhelOBP22, is similar to TcOBP8B, which has an extra cysteine of 16 AA before C1 and another additional cysteine of 8 AA after C6. Four DhelOBPs (DhelOBP5, DhelOBP6, DhelOBP11, and DhelOBP22) belong to the Minus-C OBP subfamily and also show a conserved cysteine pattern, with only four cysteines lacking C2 and C5 (C1X 18-30 C3X 38-39 C4X 16-17 C6). The one Plus-C OBP, DhelOBP10, has an expanded N and C terminus containing six additional cysteines and a typical proline residue following C6. DhelOBP2 is a partial coding sequence without the C-terminal; we speculated that it might be a Minus-C OBP because it lacks C2 based on the current alignment. DhelOBP1 and DhelOBP23 were identified as complete coding sequences but was not consistent with the conserved sequence motif of any OBP family. As a result, we removed these three DhelOBPs (DhelOBP1, DhelOBP2, and DhelOBP23) in the subsequent studies.  Figure S1. Conserved cystine was annotated according to the classic OBP subfamily.

Spatio-Temporal Expression Patterns of DhelOBPs
To investigate the transcript abundance of 20 DhelOBPs in female and male adults, we measured the expression patterns in multiple organs (including antennae, head, thorax, abdomen, reproductive organ, leg, and wing) at different adult physiological states (emerging, virgin, and mated). Firstly, all amplification efficiencies for these twenty DhelOBPs and three reference genes (RPS, GAPDH, α-Tubulin) were tested in order to be directly used in the equation from Vandesompele et al. [37] (Table S3). The results showed that 20 DhelOBPs shared similar tissue and developmental expression patterns between female and male adults ( Figure 3); however, the expression levels of different DhelOBPs in the antennae varied greatly ( Figure 4). As shown in Figure 3, eight of these DhelOBPs (DhelOBP4, 9, 13, 14, 15, 16, 18, and 19) were specifically expressed in adults' antennae. The remaining 12 DhelOBPs were also expressed in other organs.  Figure S1. Conserved cystine was annotated according to the classic OBP subfamily.

Spatio-Temporal Expression Patterns of DhelOBPs
To investigate the transcript abundance of 20 DhelOBPs in female and male adults, we measured the expression patterns in multiple organs (including antennae, head, thorax, abdomen, reproductive organ, leg, and wing) at different adult physiological states (emerging, virgin, and mated). Firstly, all amplification efficiencies for these twenty DhelOBPs and three reference genes (RPS, GAPDH, α-Tubulin) were tested in order to be directly used in the equation from Vandesompele et al. [37] (Table S3). The results showed that 20 DhelOBPs shared similar tissue and developmental expression patterns between female and male adults ( Figure 3); however, the expression levels of different DhelOBPs in the antennae varied greatly ( Figure 4). As shown in Figure 3, eight of these DhelOBPs (DhelOBP4, 9, 13, 14, 15, 16, 18, and 19) were specifically expressed in adults' antennae. The remaining 12 DhelOBPs were also expressed in other organs. DhelOBP6, DhelOBP11, and DhelOBP21 were almost expressed in all organs. In particular, DhelOBP3 and DhelOBP20 were mainly expressed in the head. Intriguingly, the expression levels of DhelOBPs in antennae also differed considerably. DhelOBP4 was the most highly expressed OBP in the antennae of emerging adults ( Figure 4A,B), while DhelOBP9 was the most highly expressed OBP in the antennae of virgin and mated adults ( Figure 4C-F). In addition to DhelOBP4 and DhelOBP9, several DhelOBPs such as DhelOBP5, DhelOBP13, DhelOBP14, DhelOBP15, DhelOBP17, DhelOBP18, and DhelOBP21, showed consistently high expression levels in antennae at different adult physiological states. Correspondingly, DhelOBP3, DhelOBP7, DhelOBP12, and DhelOBP20 showed extremely low expression levels compared to others at all tested adult physiological states. Furthermore, DhelOBPs in antennae also showed a considerable variation in expression with adult physiological states. Thirteen DhelOBPs (DhelOBP3, 7, 8, 9, 11, 12, 13, 14, 15, 16, 18, 19, and 22) were upregulated with the adult physiological state, and conversely, four DhelOBPs (DhelOBP4, 5, 6, and 10) were downregulated ( Figure S2). DhelOBP6, DhelOBP11, and DhelOBP21 were almost expressed in all organs. In particular, DhelOBP3 and DhelOBP20 were mainly expressed in the head. Intriguingly, the expression levels of DhelOBPs in antennae also differed considerably. DhelOBP4 was the most highly expressed OBP in the antennae of emerging adults ( Figure 4A,B), while DhelOBP9 was the most highly expressed OBP in the antennae of virgin and mated adults ( Figure 4C-F). In addition to DhelOBP4 and DhelOBP9, several DhelOBPs such as DhelOBP5, DhelOBP13, DhelOBP14, DhelOBP15, DhelOBP17, DhelOBP18, and DhelOBP21, showed consistently high expression levels in antennae at different adult physiological states. Correspondingly, DhelOBP3, DhelOBP7, DhelOBP12, and DhelOBP20 showed extremely low expression levels compared to others at all tested adult physiological states. Furthermore, DhelOBPs in antennae also showed a considerable variation in expression with adult physiological states. Thirteen DhelOBPs (DhelOBP3, 7, 8, 9, 11, 12, 13, 14, 15, 16, 18, 19, and 22) were upregulated with the adult physiological state, and conversely, four DhelOBPs (DhelOBP4, 5, 6, and 10) were downregulated ( Figure S2).

AlphaFold2-Based Modeling and Molecular Docking of DhelOBPs
We selected three DhelOBPs (DhelOBP4, DhelOBP14, and DhelOBP18) that are specifically and highly expressed in antennae (DhelOBP4, DhelOBP14, and DhelOBP18), two DhelOBPs (DhelOBP5 and DhelOBP6) which are expressed in multiple organs, and one DhelOBP (DhelOBP20) which mainly expressed in the head as target proteins to conduct modeling and molecular docking. The AlphaFold2-Based Colab server (https://colab.research.google.com/github/deepmind/alphafold/blob/main/notebooks/Al phaFold.ipynb, accessed on 13 May 2022) was used to model the 3D structure of our

Binding Characteristic of Recombinant DhelOBPs
Recombinant proteins of DhelOBP4, 5, 6, 14, 18, and 20 were obtained using a prokaryotic expression system. The produced proteins were successfully expressed as soluble forms in Escherichia coli. Purified recombinant proteins on SDS-PAGE are presented in Figure 6A. A single protein band for each recombinant protein was seen on the gel with the predicted protein molecular weight (Table S2). Purified recombinant mature proteins were obtained by cleaving the His-tag off using the TEV protease.

Binding Characteristic of Recombinant DhelOBPs
Recombinant proteins of DhelOBP4, 5, 6, 14, 18, and 20 were obtained using a prokaryotic expression system. The produced proteins were successfully expressed as soluble forms in Escherichia coli. Purified recombinant proteins on SDS-PAGE are presented in Figure 6A. A single protein band for each recombinant protein was seen on the gel with the predicted protein molecular weight (Table S2). Purified recombinant mature proteins were obtained by cleaving the His-tag off using the TEV protease.
To determine the effect of decreasing the expression level of DhelOBP4 on D. helophoroides behavior choice, the adults were injected with dsDhelOBP4 and dsGFP, then examined using a qRT-PCR. As expected, a significant reduction in the transcript of DhelOBP4 occurred in the dsDhelOBP4 treatment, with 85.22% and 94.70% decreases compared with the dsGFP treatment on the 6th day post-injection for female and male adults, respectively ( Figure 7C,D). Then, we conducted Y-tube olfactometer experiments to determine whether DhelOBP4 deficit in D. helophoroides influences their behavioral responses to γ-terpinene and p-cymene. The results revealed that D. helophoroides with DhelOBP4 deficiency showed a significant loss of tendency to γ-terpinene and p-cymene compared to the dsGFP treatment ( Figure 7E-H). Notably, after silencing DhelOBP4, p-cymene and γ-terpinene elicited a significant repellent response from female adults ( Figure 7E,F), while they did not elicit a behavioral response from male adults ( Figure 7G,H).

Binding Conformation of DhelOBP4 with HIPVs
The post-RNAi behavioral assays indicated that DhelOBP4 might play a crucial role for D. helophoroides in detecting HIPVs components. To better understand the binding mechanisms of DhelOBP4, we further analyzed the key binding sites and essential amino acid residues of DhelOBP4 binding with HIPVs. The residues of DhelOBP4 interacting with HIPVs are presented in Figure 8 and Table 4. The hydrophobic residues most commonly interacting between DhelOBP4 and HIPVs included Leu 11, Met 50, Phe 54, Val 56, Phe 71, and Phe 126. Four hydrophilic residues, including His 51, Tyr 90, Cys 93, and Gly 113, were exposed to the binding cavity, possibly contributing to forming hydrogen bonds with ligands. Two active behavioral compounds, γ-terpinene, and p-cymene, shared the exact same ligand interaction mechanisms with DhelOBP4. For α-terpineol, which showed no behavioral effect on D. helophoroides, an additional hydrogen bond with residue His 51 was observed. The function of the key residues warrants further study using mutant proteins.

Discussion
Chemical communication is vital to the tri-trophic interactions among host plants, herbivores, and their natural enemies [4,38,39]. Previously, we have identified 23 putative OBP transcripts in D. helophoroides antennae using a transcriptomic approach [40]. However, in this study, we excluded two DhelOBPs (DhelOBP1 and DhelOBP23), which did not show a conserved sequence motif of any OBP subfamily ( Figure 2). Meanwhile, we also speculated DhelOBP2 was a putative Minus-C OBP as its C-terminal lost partial coding sequence and showed conserved C1, C3, and C4. Of course, a 3 RACE experiment is required to confirm this speculation in the future. Of the remaining 20 DhelOBPs, fifteen of them were Classic OBP, four of them were Minus-C OBP, and one of them was Plus-C OBP. Compared to other coleopteran insects, D. helophoroides showed a relatively average repertoire of OBPs (21 OBPs), which is similar to Holotrichia parallela (25 OBPs) [41], Harmonia axyridis (19 OBPs) [42], Colaphellus bowringi (26 OBPs) [43], and Tenebrio molitor (19 OBPs) [44]. Meanwhile, more OBPs were found in Anoplophora chinensis (46 OBPs) [45], Semanotus bifasciatus (32 OBPs) [46], M. alternatus (29 OBPs) [40], and T. castaneum (50 OBPs) [36]. The variation in the number of chemosensory genes could be linked to an adaptation to living habitats and other environmental factors. The average repertoire of OBPs is just one aspect of chemosensory genes in D. helophoroides; more aspects of the comparative research, including CSPs, ORs, IRs, and SNMPs, are required to explore the relationship between chemosensory genes and their living habitat.
Tissue-specific and developmental expression patterns of OBPs are typically associated with their potential physiological functions. We performed a thorough and comprehensive tissue and developmental expression profiling study. We found that DhelOBP4, 5,9,13,14,15,17,18, and 21 were predominantly expressed in adults' antennae with continuously high transcript levels (Figures 3 and 4), implying their potential roles in olfactory perception. Conversely, four DhelOBPs (DhelOBP3, 7, 12, and 20) shared deficient transcript levels in antennae at all tested adult physiological states. However, DhelOBP3 and DhelOBP12 were predominantly expressed in the head. The actual expression levels need to be further verified at the protein level using a Western blot. As reported by Yang et al. [47], they found DhelOBP10 was highly expressed at the protein level in D. helophoroides unmated adults; however, its transcript expression level was significantly lower than many DhelOBPs in this study ( Figure 4A-D). They also found that M. alternatus and D. helophoroides relied on the evolutionarily conserved OBPs-MaltOBP24 and DhelOBP10, respectively, to recognize (+)-fenchone. We also analyzed the expression of the OBP-encoding genes at different adult physiological states (emerging, virgin, mated). Other genes exhibited considerable variation at different adult physiological states ( Figure S2). For example, DhelOBP9 and DhelOBP14 were upregulated (by more than fivefold) in virgin and mated adults compared with emerging adults, suggesting that they might be linked to reproductive behavior or host-seeking for oviposition.
Six DhelOBPs (DhelOBP4, 5, 6, 14, 18, and 20) were evaluated using a reverse chemical ecology method to determine whether they could be involved in the recognition processes of HIPVs. Although they showed similar binding energies calculated via modeling and molecular docking (Table 2), the in vitro FCBAs revealed that only DhelOBP4 showed broad binding abilities (K i < 30 µM) with all tested HIPVs (Table 3). However, the FCBAs might also present some inaccurate results. A recent study indicated that the calculated affinities of FCBAs could be quite different when using different fluorescent reporters. In this study, the binding affinities of DhelOBP4, 5, 6, 14, 18, and 20 assessed via only one fluorescent reporter, 1-NPN, might also be imprecise. More fluorescent probes such as the 1-aminoanthracene and even more in vitro binding assay techniques are required to comprehensively determine these proteins' binding properties. Nevertheless, the strong binding abilities of DhelOBP4 with HIPVs imply that it might be a potential olfactoryfunctional protein.
Based on DhelOBP4 s binding affinities, three HIPVs with the best binding affinities (α-terpineol, γ-terpinene, and p-cymene) were evaluated using behavioral assays. Y-tube olfactometer experiments indicated that γ-terpinene and p-cymene showed significant attractional effects on female and male D. helophoroides adults compared with the control (Figure 7A,B). Further RNAi-mediated behavioral experiments demonstrated the olfactory function of DhelOBP4 during the detection of γ-terpinene and p-cymene by D. helophoroides, and both females and males lost their tendency responses to these two substances. The recognition of odorant substances by insects requires the corresponding OBPs to transport them across the lymph, where the ORs ultimately completes the signal transduction and elicits the behavioral responses of insects [10,48]. In this context, we hypothesized that in the absence of DhelOBP4, the key ORs that elicit the behavioral responses could not complete the signal transduction, resulting in adults losing their tendency to γ-terpinene and p-cymene. However, γ-terpinene and p-cymene could still elicit a significant repellent response in female adults without DhelOBP4 ( Figure 7E,F). It seemed to indicate that those female adults were more sensitive to HIPVs to the extent that they could still exhibit behavioral responses to both substances even with the loss of DhelOBP4. From an evolutionary point of view, more sensitive olfactory systems might help female adults locate their hosts more accurately and find a better place for mating or oviposition. Although the molecular mechanisms underlying the recognition of these two substances in the absence of DhelOBP4 by females were not clear, we could still identify DhelOBP4 as an important protein for the recognition of γ-terpinene and p-cymene by D. helophoroides.
Binding specificity has always been attributed to ligand structures and OBP 3D structures during their interactions. Similar ligands structures were observed in two attractive substances, γ-Terpinene and p-cymene, which mainly rely on hydrophobic force to bind with DhelOBP4 ( Figure 8). The additional hydrogen bond formed between DhelOBP4 and α-terpineol might be responsible for the stronger binding affinity compared to γ-Terpinene and p-cymene. Furthermore, γ-terpinene and p-cymene shared exactly the same residues when the bond to DhelOBP4, among which Phe 54, Val 56, and Phe 71 were the most commonly interacted residues of DhelOBP4, interacted with all the tested HIPVs. The functions of these critical sites can be verified via site-directed mutagenesis in the later stages.
A deep understanding of insect olfactory mechanisms will help us to regulate insect behavior more effectively for pest control. Although our study identified two attractive substances and the key OBP involved in their recognition, the above queries remain unexplained, revealing more about an complexity of an insect's olfaction. In the future, more ideas should be developed to explain the olfactory sensory mechanisms of insects and their potential applications for pest control.

Insect Culturing and Pine Woods Collection
D. helophoroides pupae were provided by the Research Institute of Forest Ecology, Environment, and Protection, Chinese Academy of Forestry. Adults were fed on the dried meat floss of M. alternatus larvae and pupae at 25 ± 1 • C. The healthy and infested Pinus massoniana Lamb wood sections were cut from the non-infected and infected areas of M. alternatus in Yichang, Hubei Province, China.

Collection and Analysis of P. massoniana HIPVs
After the fresh wood sections were cut down from the tree, they were immediately cut into 0.4 m sections to collect volatiles using a closed-loop dynamic headspace sampling system, as previously described by Sun et al. [49]. One milliliter of n-hexane was used to elute volatiles from traps. Plant volatiles were collected for 24 h at a 12:12 h light/dark cycle, and seven biological replicates were conducted for each treatment (healthy or infested by M. alternatus). After the volatiles collection was completed, all wood sections were cut into pieces to confirm whether they were healthy or infested by M. alternatus larvae. The collected volatiles were sent to gas chromatography-mass spectrometry (GC-MS) on a QP-2010 GC-MS instrument (Shimadzu, Japan) equipped with an HP-5 MS fused-silica column (30 m × 0.25 mm × 0.25 µm) (Agilent Technologies Inc., Santa Clara, CA, USA). Helium (1 mL/min) was used as the carrier gas, and the procedure was the same as previously reported by Sun et al. [50].

Tissue Sampling, RNA Extraction, and cDNA Synthesis
Emerging, virgin, and mated D. helophoroides adults were distinguished between males and females for the following tissue sampling. All dissecting tools were treated for RNase inactivation. Each whole insect was dissected into seven parts: antennae (An), head (H), thorax (T), abdomen (A), reproductive organ (R), leg (L), and wings (W). Then the dissected tissue was immediately placed in a RNase-free tube and stored in liquid nitrogen.
Total RNA was extracted from the above-mentioned preserved tissue using TRIzol reagent (Invitrogen, Carlsbad, CA, USA) and isolated following the manufacturer's instructions. The purity of the total RNA was determined via agarose gel electrophoresis, and the concentration was measured using a spectrophotometer (Eppendorf Bio Photometer Plus, Hamburg, Germany). cDNA was prepared from 0.5 µg of total RNA via a reverse transcription, using a PrimeScript II 1st Strand cDNA Synthesis Kit (TaKaRa Bio, Otsu, Japan) following the manufacturer's instructions.

Spatio-Temporal Expression Analyses
The expression level of verified DhelOBPs genes was analyzed using a reverse transcriptionmediated qRT-PCR. A qRT-PCR was performed on a LightCycler ® 96 System using the Hieff ® qPCR SYBR Green Master Mix (No Rox) (YEASEN, Shanghai, China). Conditions were the following: 95 • C 5 min, 40 × (95 • C 10 s, 60 • C 25 s), and completed with a final cycle for the post-amplification dissolution curve. Every reaction was systematically run in triplicate with three independent biological replicates. All DhelOBPs gene-specific primers were designed using Primer-BLAST (https://www.ncbi.nlm.nih.gov/tools/primer-blast/ index.cgi?LINK_LOC=BlastHome, accessed on 22 April 2020) (Table S3). Normalized expression and relative fold change were calculated based on a model by Vandesompele et al. [37] for normalization against several reference genes when the efficiencies of target and reference genes are not similar. In this study, ribosomal protein (RPS), 3-glyceraldehyde phosphate dehydrogenase (GAPDH), and the α-Tubulin gene were chosen as reference genes and were used in each cDNA sample to correct the relative expression levels of DhelOBPs. The amplification efficiency of each pair of primers was determined by generating standard curves with mixed cDNA samples (Table S3). The following equation from Vandesompele et al. [37] was applied for the calculations (E: primer efficiency, Cq: threshold cycle, Ref : reference gene, Tar: target gene): Cq(Re f 1)

Subcloning, Expression, and Purification of Target DhelOBPs
The primers without signal peptides and containing homology arms of target Dh-elOBPs were designed to homologous recombine with pET32b (Table S5). Recombinant plasmids were transformed into DH5α Escherichia coli. (chemically competent cells) (TransGen, Beijing, China) were grown on lysogeny broth (LB) solid medium containing ampicillin (50 µg/mL). Colonies were selected and then sequenced.
For the expression of recombinant proteins, each pET32b vector containing the target sequence was used for transformation into BL21 (DE3) Escherichia coli (chemically competent cells) (TransGen, Beijing, China). The positive clone verified using a PCR was inoculated in a 5 mL LB medium with ampicillin (50 µg/mL) with shaking at 200 rpm at 37 • C. After 8 to 10 h, the culture was diluted to 1 L LB medium and grown to an optical density of 0.8 to 1.0 at a wavelength of 600 nm. The protein expression was induced by adding isopropyl-beta-D-thiogalactopyranoside to a final concentration of 0.5 mM, followed by culturing for 14 h at 200 rpm at 16 • C. Then the culture was harvested by centrifugation and sonicated. The protein was purified using a combination of chromatographic steps on anion-exchange resins and His-tag affinity chromatography. All purification steps were monitored using SDS-PAGE.

Fluorescence Competitive Binding Assays
Purified proteins were dissolved in 30 mM Tris-HCl buffer (pH 7.4 and 5.0), and emission fluorescence spectra were recorded on an RF-5301PC fluorescence spectrophotometer (Shimadzu, Kyoto, Japan) at room temperature in a right-angle configuration with a 1-cm light path quartz cuvette. To measure the affinity of the fluorescent probe 1-NPN with the target proteins, a two mM solution of protein in 30 mM Tris-HCL (pH 7.4 and 5.0) was titrated with aliquots of 1 mM 1-NPN to final concentrations ranging from 0 to the approaching saturation. The probe 1-NPN was excited at 337 nm, and the emission spectra were recorded between 350 and 500 nm. The binding affinities of other ligands were measured using 1-NPN (2 µM) as the fluorescent probe with a stoichiometry of 1:1 (protein: probe), with the final concentration of each ligand ranging from 0 to 20 µM. All ligands, including 1-NPN were purchased from Sigma-Aldrich (St Louis, MO, USA) and TCI Shanghai (Shanghai, China) and dissolved in chromatogram class methanol (Aladdin, Shanghai, China).
The dissociation constants (K d ) of DhelOBPs with 1-NPN were calculated from Scatchard plots of the binding data using PRISM 8.0 software (GraphPad, La Jolla, CA, USA). Binding affinities (K i ) of DhelOBPs for each ligand were calculated based on the IC50 value (the initial fluorescence level of the competitor concentration reduced by half) using the  is the free concentration of 1-NPN, and K 1-NPN is the dissociation constant of the complex OBP/1-NPN. The analysis supposes that there is only one protein binding site, which is 100% active, and that the binding is reversible and at an equilibrium.

Y-Tube Olfactometer Experiments
Y-tube olfactometer experiments were conducted to evaluate the behavioral response to HIPVs with the best binding affinities in darkness at approximately 25 ± 1 • C. The central tube was 15 cm long, and the two branching arms, each linked to one vial, were 12 cm long with a 75 • angle between them. Twenty microliters of each chemical that had been diluted 1000 times by volume were added to a 1 × 1 cm filter paper in one of the vials, and 20 µL of liquid paraffin were dropped on the same size filter paper in the second vial as the blank control. An effective choice was recorded when adults crawled at least 5 cm into the arm within 1 min and stayed 30 s at least. After every five individuals, the two arms were exchanged to avoid position effects. After testing 10 individuals, the Y-tube olfactometer was cleaned with absolute ethanol and oven-dried at 150 • C for 2 h. Fifty adult D. helophoroides were analyzed in each treated group. The results were analyzed using the chi-square test.

Synthesis and Microinjection of dsRNA
dsRNA was generated via an in vitro transcription with completed coding sequence templates amplified via a PCR using the primers DhelOBP4 RNAi-F (5 -TAATACGACTCA CTATAGGGATGAAATCGTATTTGGTGTTTTTATGT-3 ) and DhelOBP4 RNAi-R (5 -TAAT ACGACTCACTATAGGGTTACATGAAAGGATCTTCAATGCC-3 ), which contained T7 RNA polymerase sequence (indicated by the underline) added to their 5 ends. dsRNA synthesis was performed using a T7 RiboMAXTM Express RNAi System (Promega, Madison, WI, USA). The dsRNA concentration was examined using a spectrophotometer at 260 nm and was then dissolved in diethyl pyrocarbonate-treated water to a final concentration of 5 µg/µL and was stored at −80 • C until use. 50 male and 50 female adults of D. helophoroides were chosen on day 1-4 after eclosion. The two treated groups were: dsGFP-injected and dsRNA-injected. dsGFP (250 ng) or dsRNA (250 ng) was respectively injected into the thorax internode of the adults with a microinjector (World Precision Instruments Inc., Sarasota, FL, USA) under a microscope. The RNAi efficiency was detected using a qRT-PCR for seven consecutive days. Following behavioral response was evaluated using the Y-tube olfactometer. Informed Consent Statement: The study does not concern any human subject; thus, informed consent was not applicable.

Data Availability Statement:
All data included in this study are available upon request by contact with the corresponding author.