Genome-Wide Identification of the Odorant Receptor Gene Family and Revealing Key Genes Involved in Sexual Communication in Anoplophora glabripennis

Insects use a powerful and complex olfactory recognition system to sense odor molecules in the external environment to guide behavior. A large family of odorant receptors (ORs) mediates the detection of pheromone compounds. Anoplophora glabripennis is a destructive pest that harms broad-leaved tree species. Although olfactory sensation is an important factor affecting the information exchange of A. glabripennis, little is known about the key ORs involved. Here, we identified ninety-eight AglaORs in the Agla2.0 genome and found that the AglaOR gene family had expanded with structural and functional diversity. RT-qPCR was used to analyze the expression of AglaORs in sex tissues and in adults at different developmental stages. Twenty-three AglaORs with antennal-biased expression were identified. Among these, eleven were male-biased and two were female-biased and were more significantly expressed in the sexual maturation stage than in the post-mating stage, suggesting that these genes play a role in sexual communication. Relatively, two female-biased AglaORs were overexpressed in females seeking spawning grounds after mating, indicating that these genes might be involved in the recognition of host plant volatiles that may regulate the selection of spawning grounds. Our study provides a theoretical basis for further studies into the molecular mechanism of A. glabripennis olfaction.


Introduction
Insects have evolved a special olfactory system that can detect volatile substances in the air with specificity, providing accurate information about the environment to regulate behaviors such as feeding, mating, and egg-laying [1,2]. After the liposolubility odor molecules in the external environment enter the sensillum lymph of water solubility through the micropores of the insect sensillum epidermis, firstly, odorant-binding proteins (OBPs) and chemosensory proteins (CSPs) recognize and bind the odor molecules, and also assist odor molecules in transporting them to the periphery of the dendritic membrane of olfactory neurons (ORNs) and ultimately activate receptors. Secondly, after the receptors convert chemical signals into electrophysiological signals and transmit them to the central nervous system of insects for integration, the brain issues instructions to guide insects to conduct physiological reactions. [3,4]. Insect receptors include three gene families: odorant receptors (ORs) that are sensitive to alcohols, ketones, and esters; ionotropic receptors (IRs) that sense amines and acids; and gustatory receptors (GRs) that sense soluble chemicals [5][6][7]. As ligand-gated ion channels involved in odor recognition and signal transduction, ORs can specifically recognize different host volatiles and pheromone molecules [8]. In 1991, the first OR was identified in the mammal Rattus norvegicus, after which the first invertebrate OR was identified through the determination of the whole genome of nematodes. In 1999, the first OR of an insect was identified in Drosophila melanogaster [9][10][11]. This OR is different from the G protein-coupled receptors of vertebrates; it is a macromolecular hydrophobic protein with seven transmembrane helical structures [12]. Insect ORs can be divided into two categories; one category is common ORs that bind odor molecules, and the other is olfactory receptor co-receptors (Orco) [13]. Orco interacts with ORs to form heterodimers (Orco/OR), resulting in functional ion channels that do not react directly to odorant substances but are directly activated by intracellular cAMP or cGMP; therefore, unlike the olfactory recognition pathway of mammalian G-protein-coupled receptors that rely on second messengers to activate ion channels, insect ORs can recognize odor more quickly and efficiently [14,15].
The insect genome contains OR genes with different functions, each of these ORs express a protein that binds to a specific ligand molecule at a specific site [16]. Coleoptera, the largest order of Insecta, has recently been found to express OR families of nearly forty species at the genome and transcriptome levels [17][18][19][20][21][22][23][24][25][26][27][28][29][30]. Most polyphagous insects show diversity in the number of OR genes among different species. Indeed, the correlation between the number of ORs and insect feeding habits suggests a potential connection between the diversity of ORs and the species' ecological habitat and host range [19]. The rapid contraction and/or expansion of the OR family to adapt to complex and changing environments also increases the complexity of OR phylogenetic analysis and further limits our ability to predict the function of key ORs [19]. At present, only twelve ORs in Coleoptera, namely three in Megaxylene Caryae (McarOR3, McarOR5, and McarOR20), four in Ips typographus (ItypOR5, ItypOR6, ItypOR46, and ItypOR49), two in Hylobius abietis (HabiOR3 and HabiOR4), two in Dendroctonus ponderosae (DponOR8 and DponOR9), and one in Rhynchophorus ferrugineus (RferOR6) have been assigned to effective ligand components and classified as functional ORs. McarOR3, McarOR5, and McarOR20 are sensitive to components of aggregation pheromones released by males, including (S)-2-methyl-1-butanol, 2-phenylethanol and (25,3R)-2,3-hexanediol, while ItypOR46 and ItypOR49 recognize the male aggregation pheromone components of I. typographus, including ipsenol and ipsdienol [26,30]. HabiOR3, DponOR8, and ItypOR6 respond exclusively to 2-phenylethanol, and HabiOR4, DponOR9, and ItypOR5 respond to angiosperm green leaf volatiles [31]. RferOR6 is narrowly tuned to alpha-pinene [32]. Combining the identification of gene family with the specific expression patterns of the gene to explore possible functional ORs can provide insights that facilitate more comprehensive screening of target genes for pest control and prevention.
Anoplophora glabripennis (Coleoptera: Cerambycidae) is a worldwide quarantine pest native to Asia. As a polyphagous insect, it has possible hosts in 15 families, 37 genera (e.g., Acer, Populus, Salix, and Ulmus). Larvae damage the xylem and phloem, which results in the decline of tree vigor and death, causing a decline in forest productivity and loss of forest resources [33]. A. glabripennis is primarily distributed in Asia. However, in 1996, A glabripennis was identified in Brooklyn, New York, NY, USA, and then rapidly colonized other parts of North America and Europe making it a quarantine pest worldwide [34,35]. After eclosion, the beetles feed on tender leaves of the host twigs for one week, and then gradually arrive at sexual maturation. Virgin male and female beetles gather together under the influence of both host plant volatiles and male aggregation pheromones [36]. Virgin males select suitable mates through vision and volatiles released by females and complete the mating process under the stimulation of female contact pheromones [37]. After mating, females sense the host plant volatiles to locate a suitable spawning site. Therefore, host plant volatiles and sex pheromones play an important role in regulating the breeding activities of beetles [38]. Previous electrophysiological and behavioral studies have reported that the volatiles released by the host plants of A. glabripennis include terpenes, alcohols, aldehydes, and acetates. The A. glabripennis male aggregation pheromones are mainly comprised of two hydroxyl ethers, and the female contact pheromones include five unsaturated longchain hydrocarbons [39][40][41][42][43][44][45][46]. Attractant mixtures containing host plant volatiles and male aggregation pheromones have been used to trap A. glabripennis in forests, but with little success [47]. Previous studies have clarified the types of antennae receptors of beetles at different life stages. Olfactory sensory neurons located in the trichomes of the amphoteric antennal whipstock, for example, are more sensitive to the male-produced aggregation-sex pheromones 4-(n-heptyloxy)butan-1-ol and 4-(n-heptyloxy)butanal. Olfactory neurons of the conical sensilla of whipstock, on the other hand, have more obvious responses to alcohols, aldehydes, and terpenes, indicating that different types of A. glabripennis sensilla may have selective sensitivity to odor cues [48]. In a previous study, Hu et al. (2017) screened thirty-seven AglaORs on the A. glabripennis antennal transcriptome [49]. Mitchell et al. (2017) screened 121 AglaORs on the Agla1.0 genome, but they were not verified experimentally [25]. The four sex-biased AglaORs predicted in male and female wholebody transcriptomes are expressed at extremely low levels in antennae, suggesting that they might not be involved in the olfactory perception of volatile pheromone compounds (Zhang, unpublished data). Therefore, there has been no report of the key AglaORs involved in olfactory recognition in A. glabripennis. Later, the genome of A. glabripennis V2 version was published in U.S. DEPARTMENT OF AGRICULTURE (https://www.usda.gov/, accessed on 10 October 2022) by McKenna, which provided more accurate and complete genetic information. The Agla2.0 genome may help to identify members of the AglaOR gene family more comprehensively and provide reliable data for further screening of key genes related to smell.
In this study, ninety-eight candidate AglaORs were re-identified based on the Agla2.0 genome. We analyzed sequence characteristics, phylogenetic relationships, and gene and protein structures, to characterize in more detail the basic structural features of the gene family. In addition, the expression patterns of AglaORs in various sex tissues and at different developmental stages of the adult were analyzed by RT-qPCR. Moreover, key AglaORs participating in host localization or sexual communication were screened, providing a theoretical basis for elucidating the molecular mechanism of A. glabripennis olfaction and opening the perspective of identifying candidate target genes for pest control and prevention.

Genome-Wide Identification of OR Genes in A. glabripennis
In order to obtain more comprehensive information on the AglaORs gene family, we first identified 127 candidate AglaORs based on the Agla2.0 genome using blastP and the Hidden Markov Model. After filtering out nineteen atypical odor receptors belonging to the 7tm-4 subfamily and ten repetitive short sequences encoding less than one hundred amino acids, ninety-eight genes were finally identified for analysis in this study (AglaOR1-98). Two genes (AglaOR22/23) were newly identified in this study and had sequence similarities with AglaOR21 (XP_023311850.1) of 73.16% and 75.76%, respectively. Among the ninetyeight genes in the AglaOR gene family, fifty-eight genes had complete open reading frames encoding potentially functional proteins, and forty genes lacked 5' or 3' end structures. The full-length AglaORs had the potential to encode proteins of 302-540 amino acids, with molecular weights of 35.11-77.26 KDa, isoelectric points of 5.27-9.79, and grand average of hydropathicity values > 0. Subcellular localization showed that most of the AglaORs proteins were localized to the cell membrane (Table 1). These characteristics were consistent with the typical macromolecular hydrophobic transmembrane protein structures of insect ORs.

Phylogenetic Analysis of AglaORs
A phylogenetic tree was constructed using the OR gene families of A. glabripennis, T. castaneum, D. ponderosae, and A. planipennis, as well as ten functional ORs from M. Caryae, I. typographus, H. abietis, and R. ferrugineus. The results showed that the Orco of four species could be assembled into one single branch to form a clear homologous lineage. The OR of 573 other species formed different evolutionary branches and were divided into nine subfamilies, which is in line with the current phylogenetic map of the Coleoptera OR gene family (Figure 1). T. castaneum was mainly distributed in groups 5A, 2A, 1, and 3; D. ponderosae was mainly distributed in groups 7 and 5B; A. planipennis was mainly distributed in groups 4, 6, 2B; and ninety-seven ORs of A. glabripennis were mainly distributed in groups 3 (27), 2A (26), 7(17), 1(14), 2B (7), and 5A (5) subfamilies. Genes within the same subfamily were also phylogenetically dispersed, suggesting that they evolved rapidly to accommodate wide ranges of hosts and to respond to different environmental conditions. In our phylogenetic results, AglaOR29 was clustered with MacrOR3; AglaOR32/34/35 were clustered with MacrOR5; AglaOR61/62/64 were clustered with MacrOR20; AglaOR9 was clustered with RferOR6 and AglaOR30/33 was clustered with the functional receptor group (HabiOR3, DponOR8, ItypOR6, HabiOR4, DponOR9 and ItypOR5) in subfamily group 2B.

AglaOR Gene and Protein Structure
Comprehensive analysis of the conservative motif, conserved protein domain, and genetic structure of the AglaORs revealed differences in the motif structure between each subfamily ( Figure 2). Details of the ten motifs are provided in the supplementary information. In Group I, motifs 1 and 6 were considered to be conserved motifs. Motifs 1 and 7 were unique motifs in Group II. Group III has motifs 1, 2, 3, 5, 6, and 10 as the conserved motif, AglaOR86 and AglaOR55 lacked motif 6, and AglaOR82 lacked motifs 2, 5, and 10. Group IV has motifs 1, 4, 6, and 7 as the conserved motif. The gene structure map of introns-exons showed that Group I mostly contains 3-7 introns, of which, AglaOR1 and AglaOR9 contain 10 and 9 introns, respectively. Most of Group II has 2-6 introns. Group III mostly contains 4-6 introns, except that of AglaOR82, which had two introns. Group IV mostly contains 4-6 introns.

AglaOR Gene and Protein Structure
Comprehensive analysis of the conservative motif, conserved protein domain, and genetic structure of the AglaORs revealed differences in the motif structure between each subfamily ( Figure 2). Details of the ten motifs are provided in the supplementary information. In Group I, motifs 1 and 6 were considered to be conserved motifs. Motifs 1 and 7 were unique motifs in Group II. Group III has motifs 1, 2, 3, 5, 6, and 10 as the conserved motif, AglaOR86 and AglaOR55 lacked motif 6, and AglaOR82 lacked motifs 2, 5, and 10. Group IV has motifs 1, 4, 6, and 7 as the conserved motif. The gene structure map of introns-exons showed that Group I mostly contains 3-7 introns, of which, AglaOR1 and AglaOR9 contain 10 and 9 introns, respectively. Most of Group II has 2-6 introns. Group III mostly contains 4-6 introns, except that of AglaOR82, which had two introns. Group IV mostly contains 4-6 introns.

Analysis of Expression Pattern of AglaORs in Different Sex Tissues of Adults
By Sanger sequencing ninety-eight AglaORs, we filtered out twenty-six ORs that could not be amplified to the target sequence. The expression patterns of seventy-two AglaORs verified by sequencing in the antenna, leg, mandibular palps, head, external genitalia, and thorax of males and females were analyzed by RT-PCR. The results showed that although OR expression levels were low, they showed broad tissue expression profiles. Twenty

Analysis of Expression Pattern of AglaORs in Different Sex Tissues of Adults
By Sanger sequencing ninety-eight AglaORs, we filtered out twenty-six ORs that could not be amplified to the target sequence. The expression patterns of seventy-two AglaORs verified by sequencing in the antenna, leg, mandibular palps, head, external genitalia, and thorax of males and females were analyzed by RT-PCR. The results showed that although OR expression levels were low, they showed broad tissue expression profiles.    Actin was used as a housekeeping gene to normalize the expression level of each treatment. The relative expression levels represent mean ± standard error of the mean (SEM). Lowercase letters above the error bars indicate significant differences (p < 0.05, LSD). N/A means that the expression level is too low to be displayed. Subfigure (A) represents the genes with male antenna biased expression, (B) represents the genes with female antenna biased expression, and (C) represents the genes with no significant difference between two sexes.

Analysis of Expression Patterns of AglaORs in Adults at Different Stages of Development
RT-qPCR analysis of the expression patterns of twenty-three ORs with high antennal expression in different developmental stages of A. glabripennis adults showed that sixteen ORs were significantly upregulated in the sexual maturation stage and significantly downregulated after mating (AglaOR3/6/7/11/19/29/32/34/38/50/66/73/74/82/86/88) ( Figure  5A). AglaOR3 also showed significantly higher expression after 1 d nutrient supplementation. AglaOR33/91 was significantly overexpressed in the female post-mating grooving phase ( Figure 5C). There were no differences in AglaOR25/31/35/42/60 expression between different development stages ( Figure 5B). Actin was used as a housekeeping gene to normalize the expression level of each treatment. The relative expression levels represent mean ± standard error of the mean (SEM). Lowercase letters above the error bars indicate significant differences (p < 0.05, LSD). N/A means that the expression level is too low to be displayed. Subfigure (A) represents the genes with male antenna biased expression, (B) represents the genes with female antenna biased expression, and (C) represents the genes with no significant difference between two sexes.

Discussion
Prediction of gene family members based on sequence similarity and protein conservative domains depends largely on the maturity of genomic assembly and the integrity of genomic information. Incomplete OR gene sequences may be related to problems in genomic assembly or sequencing methods, or to the rapid evolution of the OR gene family within complex environments consisting of a wide range of odor molecules. In this study, we found many incomplete OR gene sequences in the OR gene families of T. castaneum, Rhaphuma horsfieldi, D. ponderosae, A. Planipennis, Aphis gossypii, Ambrostoma quadriimpres- The gene expression of A (eclosion in pupal chamber) was used as the control. The relative expression level is expressed as the standard error (SEM) of the mean value. The lowercase letters above the bar indicate significant differences in FA expression at different developmental stages, and the uppercase letters indicate significant differences in MA expression at different developmental states (p < 0.05, LSD). Subfigure (A) represents the genes with significant high expression in sexual maturity, (B) represents the genes with no significant difference expression in each development stage, and (C) represents the genes with significant high expression after female post-mating grooving phase.

Discussion
Prediction of gene family members based on sequence similarity and protein conservative domains depends largely on the maturity of genomic assembly and the integrity of genomic information. Incomplete OR gene sequences may be related to problems in genomic assembly or sequencing methods, or to the rapid evolution of the OR gene family within complex environments consisting of a wide range of odor molecules. In this study, we found many incomplete OR gene sequences in the OR gene families of T. castaneum, Rhaphuma horsfieldi, D. ponderosae, A. Planipennis, Aphis gossypii, Ambrostoma quadriimpressum, and Migratory locust [18,23,28,[50][51][52]. The ninety-eight AglaORs in this study, including thirty-seven AglaOR sequences identified by the antennal transcriptome [49], had ninety-six sequence matches compared to those identified by the Agla1.0 genome [25]. The members of OR gene family were obtained systematically and comprehensively, which will provide provides rich data for key gene mining.
Gene structure and conservative motif patterns are important for studying the evolutionary relationship of genes and the functional diversity of proteins in gene families. The conserved protein domain is important for OR structure, and the motif pattern can fine-tune the function of OR and lead to subtle differences in the binding of different odor molecules. We found that motifs 1, 6, and 7 are conserved motifs that are present in all genes and that they consist of 41, 16, and 11 amino acids, respectively. These motifs presumably play a conservative role in evolution. Motifs 2, 3, 5, and 10 are characteristic of subfamily Group III, which may regulate specific functions in this family. In the whole AglaOR gene family, the differences in the number and length of introns, which can affect gene expression, showed that they had been repeatedly acquired and lost [53,54]. AglaOrco (AglaOR1) had the most introns (ten), and the gene is highly expressed in olfactory tissues. The structure of the AgosOrco gene reported by Cao also has ten introns [50]. The single homologous Orco lineage can be fully explained by its conservation during evolution.
Using phylogenetic analysis, OR genes of the four species were assigned to nine subfamily lineages of the newly revised OR gene family of Coleoptera [19]. Notably, AglaOR30/31/32/34/35 are homologous to seven functional genes that recognize 2-phenylethanol or angiosperm green leaf volatiles, and they belong to the 2B subfamily, in which AglaOR31/32/34/35 exhibits male-antenna biased expression. It is speculated that these may be involved in the recognition of the above-mentioned compounds and we will conduct functional research on these genes in the future. There were no homologous genes in the AglaOR gene family that can be compared with the cluster of pheromone receptor Ityp46/49 of I. typographus. Ipsenol and ipsdienol are specific aggregation pheromone components of IPs. Males feeding on conifer species such as Pinus and Picea release a large amount of clastic pheromones, which signal male and female adults to hosts [55]. A. glabripennis selectively feeds on broad-leaved tree species. Therefore, differences in host selection may result in failure to evolve homologous pheromone receptors that recognize similar pheromone components.
Gene expression patterns are closely related to the function of encoded proteins. A typical OR is selectively expressed in olfactory neurons with low expression. In this study, RT-PCR analysis of the tissue expression profiles of AglaORs showed that the expression levels of twelve AglaORs were extremely low in all tissues. Low AglaOR expression may indicate that these genes exercise other functions, or that they are activated to recognize odorant molecules in other life states or under specific environmental conditions. The twenty-nine AglaORs were expressed in all tissues, indicating that they might also be involved in functions other than olfactory recognition, such as processing of olfactory bulb signals in the brain, detection of pheromone release, or regulating the reproductive process by affecting the development of sperm or egg cells [56,57]. Besides the antenna, the sensory organs such as the leg and mandibular palps also play a role in the perception of non-volatile compounds. For example, Hoover et al. (2014) reported that four sex-trace pheromones, 2-methyldocosane and (Z)-9-tricosene as major components and (Z)-9-pentacosene and (Z)-7-pentacosene as minor components, were action-labeled volatiles left by females [56]. It seems likely that the four highly expressed AglaORs in the leg or mandibular whisker (AglaOR27/45/47/49) are involved in the identification of contact and trace pheromones, and we propose that they could be used as important gene candidates to characterize further ligand binding for functional mining.
The antenna, which is the main organ used for detecting odor molecules in A. glabripennis, contains various types of receptors distributed on its surface. Odor molecules enter the sensillum lymph through the sensillum micropores and bind to ORs to activate the nerve center under the carriage of odorant-binding proteins (OBPs). Therefore, ORs specifically expressed in the antenna are considered to have an important function in recognizing odor molecules. Only 23 of the 98 AglaORs in the present study showed high antennal-biased expression, indicating the need to mine key genes using gene family identification in combination with gene expression pattern analysis. Among the 23 AglaORs, 20 showed significant male and female antennal-biased expression. Four AglaORs (AglaOR19/33/86/91) showed significant female antennal-biased expression. We speculate that they play a role in sensing male aggregation pheromones before mating or in finding spawning sites after mating, based on the analysis of expression patterns of adults at different stages of development. AglaOR19/86 expression was significantly upregulated at sexual maturation and significantly downregulated after mating, indicating that this gene plays a role in sexual communication. Moreover, because pheromone release is inhibited during mating, the ability of these receptors to recognize the pheromone is weakened, leading to a decrease in their expression. AglaOR33/91 expression was significantly upregulated in females searching for spawning sites after mating, suggesting that it might recognize host plant volatiles and determine the choice of spawning site. Another 16 AglaORs showed significant male antennal-biased expression, suggesting that these genes play an important role in the recognition of female pheromones prior to mating. We also found that the expression of eleven of the genes was significantly upregulated at sexual maturation and significantly downregulated after mating, suggesting that these genes are involved in premating sexual communication. Although the other five genes had male antennal-biased expression, they were not differentially expressed between the different development stages, suggesting that they might play an ongoing role in regulating olfaction. At the same time, we also found that AglaOR3 was highly expressed at 1 d post-eclosion after feeding on the host. We speculate that this gene may be involved in the selection of the feeding host after eclosion. Although AglaOR25 showed male antennae-biased expression, the expression in female antennae was significantly increased after mating. Thus, this gene may recognize a highly broad spectrum of odorants, allowing it to play different roles in various physiological states. After eclosion, A. glabripennis develops in the pupal chamber from 7-14 d, and then bites out of the circular eclosion hole and leaves the chamber [58]. However, here we found no significant difference in the expressed ORs between the two physiological states of eclosion in the pupal chamber and eclosion. We hypothesize that the beetle may rely on its tentacles and mouthparts to gnaw on the phloem and xylem of trees to explore the surrounding environment, and sensing odor molecules through the antennae may not dominate this process. Likewise, it is possible that candidate genes involved in this process have yet to be discovered, which is a direction for future research.

Identification of AglaOR Gene Family
Genomic information for A. glabripennis was obtained from the NCBI database (PRJNA167479). To identify AglaOR gene family members, the published OR proteins of Drosophila melanogaster and T. castaneum were used as queries to search against the genome of A. glabripennis using blastP with an E-value < 1 × 10 −5 . The HMM file for OR (PF02949:7tm odorant receptor) was then downloaded from the Pfam database. The 7tm-6 subfamily domain in the protein sequence of A. glabripennis was searched for using HMMER3.0 (E-value < 0.01) [59]. The presence of conservative domains in candidate proteins was confirmed using NCBI Preserved Domain Database (https://www.ncbi.nlm.nih.gov/Structure/cdd/wrpsb.cgi, accessed on 12 October 2022) and Smart Database (http://smart.embl-heidelberg.de/, accessed on 12 October 2022). The integration deleted the short sequence encoding less than 100 amino acids and confirmed the complete AglaOR gene family. All AglaORs sequences were analyzed by EXPASY (https://wolfpsort.hgc.jp/, accessed on 12 October 2022) to obtain amino acid number (AA), molecular weight (MW), isoelectric point (PI), and transmembrane domain (TM). The sequences were also subjected to protein avidity/hydrophobicity analysis (GRAVY). Subcellular localization was predicted by WOLF PSORT (https://wolfpsort, accessed on 12 October 2022).

Construction of Phylogenetic Tree
The OR gene family sequences identified from T. castaneum, Dendroctonus ponderosae, and Agrilus planipennis genomes were selected for participation in the construction of a phylogenetic tree to determine the branch positions of AglaORs, and ten functional Ors from M. Caryae, I. typographus, H. abietis, and R. ferrugineus were also included to determine whether they are homologous with AglaORs [18,22,23,31,32]. Maximum likelihood phylogenies were inferred using IQ-TREE [60] under a model automatically selected by IQ-TREE ('Auto' option in IQ-TREE) for 1000 ultrafast [61,62] bootstraps, as well as the Shimodaira-Hasegawa-like approximate likelihood-ratio test. Evolview (http: //www.evolgenius.info/evolview, accessed on 13 October 2022) was used to visualize the evolutionary tree.

Structural Characteristics Analysis of AglaORs
The gene structure information was extracted from the annotation file of A. glabripennis genome and TBtools (v1.09851) was used for visual analysis of the exon-intron structure of AglaORs [63]. The parameters were as follows: number of repetitions = arbitrary, maximum base number = 10, and optimal base width = 10-50 residues. The online tool MEME (http://meme-suite.org/, accessed on 17 October 2022) was used to perform a motif analysis of AglaORs. The structure of 10 motifs are shown in Figure S1. The conservative domain of AglaORs was analyzed by NCBI Preserved Domain Database (https: //www.ncbi.nlm.nih.gov/Structure/bwrpsb/bwrpsb.cgi, accessed on 17 October 2022) with search mode = automatic, E-value < 0.01, and maximum number of hits = 500.

Insect Collection and Processing
The natural poplar tree segments damaged by A. glabripennis were collected from Sanhe Forest Farm, Qingshui Town, Suzhou District, Jiuquan City, Gansu Province, China (39 • 34 N, 99 • 10 E). The sections were sealed with wax and shipped back to Beijing Laboratory for placement in a self-made wire mesh cage with dimensions 3 m × 3 m × 3 m. The indoor temperature was controlled at 25 ± 1 • C, and the relative humidity was within 60-70%. Tissues from both sexes of A. glabripennis were dissected using sterilized scissors and forceps, including the antenna, leg, maxillary palps, head (without maxillary palps), external genitalia, and thorax. Samples of male and female adults that were eclosed in the pupal chamber, eclosed and received supplemental nutrition for 1 d, 7 d, and 12 d, were collected. The reproductive organs gradually developed to sexual maturity 7-12 d after the beetles were supplemented with food. The males and females supplemented with food for 12 d were placed in insect feeding boxes in pairs and were collected after mating during female grooving behavior. Antennae of males and females are collected in different development stages mentioned above. The above process was repeated three times for each sample. Samples were frozen at −80 • C until use.

RNA Extraction and RT-qPCR Analysis
Total RNA from each tissue was extracted using TRIzol reagent (No. 15596026; Invitrogen, Carlsbad, CA, USA) and RNeasy Plus Mini Kit (No. 74134; Qiagen, Hilden, Germany) according to the manufacturer's instructions. RNA purity, concentration, and integrity were assessed by NanoDrop 8000 (Thermo, Waltham, MA, USA) and agarose gel electrophoresis. PrimeScript RT Reagent Kit with gDNA Eraser (No. RR047A; TaKaRa, Dalian, China) was used to extract 1 µg of total RNA for cDNA synthesis. Primer3Plus (http: //www.primer3plus.com/cgi-bin/dev/primer3plus.cgi, accessed on 18 October 2022) was used to develop specific primers for RT-PCR and RT-qPCR. Primers were designed and sent to Beijing Ruiboxingke Biotechnology Co., Ltd. for synthesis.RT-PCR specific primer sequences are shown in Table S1, RT-qPCR specific primer sequences are shown in Table S2. RT-PCR was performed using a 2×Taq PCR Master Mix (No. BN12045; Biorigin, Beijing, China). A 12.5 µL system was used for each PCR reaction, including 2×Taq PCR Master Mix, 1 µL per primer pair, 1 µL cDNA template, and 9.5 µLd2H2O. The amplification procedure was as follows: 94 • C for 1 min and 30 s; followed by 34 cycles of 94 • C for 20 s; 52 • C for 20 s; 72 • C for 30 s; 72 • C for 5 min; and 4 • C indefinitely. The PCR products were subjected to 1.2% agarose gel electrophoresis to examine gene expression (voltage 120 V, 30 min, 1× TAE as electrophoresis buffer). PCR products with bright and single bands were selected for Sanger sequencing to remove the false-positive gene. RT-qPCR was performed using the Bio-Rad CFX96 PCR system (Hercules, CA, USA) and SYBR Premix Ex Taq II (No. RR820A; TaKaRa, Dalian, China). Each PCR reaction was conducted in 12.5 µL of reaction mixture containing 6.25 µL of SYBR Premix Ex Taq II, 0.5 µL of each primer, 1 µL of cDNA template, and 4.25 µL of ddH2O. The amplification procedure was as follows: 95 • C for 30 s, followed by 40 cycles of 95 • C 0.05 s, 60 • C 30 s, and 95 • C for 10 s, followed by increments of 0.5 • C from 65-95 • C for 0.05 s each to generate a dissolution curve. Actin commonly used for tissue expression of A. glabripennis was selected as the internal reference gene to normalize the expression level [64][65][66]. Three biological replicates and three technical replicates were performed. The relative expression amount was calculated based on the 2 −∆∆ Ct method [67].

Data Analysis
One-way analysis of variance and least significant difference (LSD) tests were performed using SPSS 19.0 (IBM SPSS, Armonk, NY, USA). p < 0.05 indicated that the difference was statistically significant. The gene expression levels of female antennae were used as controls in the tissue expression profile, and the gene expression level at eclosed in the pupal chamber was used as the control for producing spatial-temporal expression profiles. Quantitative data were expressed as the standard error of the mean (SEM).

Conclusions
In this study, we identified the AglaOR gene family, constructed a phylogenetic tree, analyzed the structural characteristics of the gene proteins, and found that the AglaOR family exhibited structural and functional diversity. Using spatial-temporal differential expression analysis, twenty-three highly expressed AglaOR genes in the antenna were screened, revealing the key candidate genes involved in the premating sexual communication process (eleven male antennal-biased and two female antennal-biased) and the selection of female spawning grounds after mating (two female antennal-biased). To determine whether these genes have an olfactory recognition function and to assess their binding characteristics and sensitivity to ligand molecules, it is necessary to further explore genes using the xenopus oocyte/voltage clamp method. A comprehensive and systematic analysis of the OR gene family of A. glabripennis could provide a theoretical basis for further elucidating the molecular mechanism of olfactory recognition. The use of RNAi or CRISPR/Cas9 and other techniques to silence or knock out key AglaORs could be used to interfere with A. glabripennis olfaction, providing strategies for pest control and prevention over small environmental ranges.