Comparative Transcriptome Analysis of Chemoreception Organs of Laodelphax striatellus in Response to Rice Stripe Virus Infection

Many vector-borne viruses possess the ability to manipulate vector behaviors to facilitate their transmission. There is evidence that the mechanism of this phenomenon has been described in part as direct manipulation through regulating vector chemosensation. Rice stripe virus (RSV) is transmitted by the small brown planthopper, Laodelphax striatellus (Fallen), in a persistent, circulative–propagative manner. The effect of RSV infection on the olfactory system of L. striatellus has not been fully elucidated. Here, we employed transcriptomic sequencing to analyze gene expression profiles in antennae, legs and heads (without antennae) from L. striatellus females and males with/without RSV infection. Comparisons of the differentially expressed genes (DEGs) among antennae, legs and heads indicated that tissue-specific changes in the gene expression profile were greater than sex-specific changes. A total of 17 olfactory related genes were differentially expressed in viruliferous antennae as compared to nonviruliferous antennae, including LstrOBP4/9, LstrCSP1/2/5, LstrGR28a/43a/43a-1, LstrIR1/2/NMDA1, LstrOR67/85e/56a/94 and LstrSNMP2/2-2. There are 23 olfactory related DEGs between viruliferous and nonviruliferous legs, including LstrOBP2/3/4/12/13, LstrCSP13/5/10, LstrIR1/2/Delta2/Delta2-1/kainate2/NMDA2, LstrOR12/21/31/68 and LstrORco. A low number of olfactory related DEGs were found between viruliferous and nonviruliferous heads, including LstrCSP1, LstrOBP2, LstrOR67 and LstrSNMP2-2. Among these DEGs, the expression patterns of LstrOBP2, LstrOBP3 and LstrOBP9 in three tissues was validated by quantitative real-time PCR. The demonstration of overall changes in the genes in L. striatellus’ chemoreception organs in response to RSV infection would not only improve our understanding of the effect of RSV on the olfactory related genes of insect vectors but also provide insights into developing approaches to control the plant virus transmission and spread as well as pest management in the future.


Introduction
Vector-borne viruses contribute to a substantial portion of the global plant disease burden, accounting for more than 70% of all known plant viruses [1,2]. The horizontal transmission and spread of vector-borne viruses relies on the movement of insect vectors, which could be manipulated by vector-borne viruses [3,4]. The mechanism by which viruses manipulate vector behaviors has largely been described as indirect manipulation through modifications of the quality, color, odor and associated plant traits. For instance, cucumber mosaic virus (CMV) accelerates the emigration of aphid vectors by reducing host plant quality, and increased the attractiveness of infected plants to aphid vectors by inducing elevated emissions of a volatile blend [3]. Potato leafroll virus (PLV) infection changes the volatile blend of potato to attract vector aphids, thereby increasing the probability of virus acquisition [5]. Furthermore, some vector-borne viruses could manipulate vector behavior via directly affecting the nervous and olfactory system. Tomato yellow leaf curl virus (TYLCV) reduces whitefly preference to viruliferous plant by disabling the nervous system of vectors [6]. Southern rice black streaked dwarf virus (SRBDV) and tomato chlorosis virus (TCV) are able to regulate the gene expression of odorant-binding protein (OBP) 2 and OBP3 in their vectors, Sogatella furcifera and Bemisia tabaci, respectively, and to reverse the preferences of nonviruliferous/viruliferous vectors, finally affecting the virus transmission [7,8]. Therefore, various vector-borne viruses have developed the ability to utilize a direct or indirect manipulation mechanism to induce behavioral changes of insect vectors, to benefit their spread. However, the effect of vector-borne virus infection on the olfactory system of insect vectors has not been fully elucidated.
The critical roles of the insect's olfactory system in their locomotion, host seeking, probing and feeding are mainly executed via detecting semiochemicals by sensilla on the main periphery, such as the main chemoreception organ, the antennae [9,10]. A number of protein families located in these peripheral sensory organs are responsible for the recognition of semiochemicals, which include two classes of carrier proteins: OBPs and chemosensory proteins (CSPs), and four classes of membrane proteins: ionotropic (IR), gustatory (GRs), membrane-bound odorant receptors (ORs)/olfactory co-receptor receptor (ORco) and sensory neuron membrane proteins (SNMPs) [11][12][13][14][15][16]. When environmental chemicals penetrate the pores of sensilla, OBPs and CSPs as carriers bind these chemicals in the sensilla lymph and transport the chemicals to ORs or IRs located on the olfactory sensory neurons. Then, chemical signals are converted into electrical signals, and conveyed to the insect's brain to eventually modulate the insect's behavior [14,17]. Thus far, the role of olfactory-related proteins in virus transmission remains unclear.
Rice stripe virus (RSV), a typical member of the genus Tenuivirus, has inflicted serious yield losses of rice by causing the notorious rice stripe disease in Eastern Asia [18,19]. It is transmitted by the small brown planthopper, Laodelphax striatellus (Fallen), in a persistent, circulative-propagative manner [20][21][22]. RSV particles enter L. striatellus via the alimentary canal during feeding and initially establish infection in the midgut epithelium, propagating in midgut visceral muscles. After transferring into the hemolymph, these virus invade several key organs via lymph circulation and, ultimately, arrive at the salivary gland, from which RSV particles are inoculated back to the plant hosts along with saliva. RSV transmission in L. striatellus requires specialized interactions between components of the virus and vector, evidenced by the transcriptome analyses of the expression changes of overall mRNA genes and the regulation of RSV-derived siRNA in L. striatellus after RSV infection [23][24][25]. In addition, the organ-specific transcriptomes of L. striatellus' alimentary canal and salivary gland characterize the responses of two organs in confronting RSV infection [26]. However, the molecular mechanism underlying the interactions between RSV and chemoreception organs remains unknown at the omic level.
Our preliminary antennae-specific transcriptome analysis identified 14 OBPs, 12 CSPs, 7 SNMPs and 95 ORs in L. striatellus [27]. Meanwhile, viruliferous L. striatellus might have a stronger olfactory and seeking ability for rice than nonviruliferous insect [28]. Thus, the current study aims to further investigate the host preference mechanism of viruliferous L. striatellus and the role of the olfactory system in RSV transmission. We compared the response of the antennae, legs and heads (without antennae) from both L. striatellus females and males to RSV infection using next-generation deep-sequencing techniques. Furthermore, olfactory related genes from differentially expressed genes (DEGs) of chemoreception organ transcriptomes were enriched between viruliferous and nonviruliferous organs. Lastly, a qPCR was performed to confirm the expression changes of three OBPs between viruliferous and nonviruliferous tissues. Our results would not only improve our understanding on the role of the olfactory chemoreception organs system in RSV infection and transmission but also provide new insights into the symbiosis interaction between virus, insect and plant. A deeper understanding of this interaction provides new avenues for controlling plant viruses and their vectors.  (Table S1). GC percentages (%) of sequence reads from the twelve libraries were all approximately 40%, and Q20 (%) were more than 95.88%, which showed that the accuracy and quality of the sequencing data met the standard for further analysis (Table S1). Of the reads from each transcriptome, 58.06~68.94% paired-end clean reads were mapped to the genome of L. striatellus (Table S1) and assembled into 20,409 genes with function annotation. Additionally, the normalized read count with Fragments per Kilobase of transcript per Million mapped reads (FPKM) of these genes were calculated.

Results
Principal component analysis (PCA) was used to compare the twelve transcriptomes from viruliferous and nonviruliferous organs of females and males. Overall, the principal component axis 1 and 2 (PC1 and PC2, accounting for 47.24% and 27.05% of the observed gene expression variation, respectively), separated these libraries into three groups: Antennae libraries (VMA, VFA, NMA and NFA), Legs libraries (VML, VFL, NML and NFL) and Heads libraries (VMH, VFH, NMH and NFH) ( Figure 1A). The comparison between the three tissues demonstrated the relative most number of DEGs between heads and legs (especially in VMH versus (vs.) VML), modest in antennae vs. heads, and least in antennae vs. legs. However, the comparison between female tissues and male tissues showed fewer DEGs ( Figure 1B). These results demonstrated that organ-specific and sex-specific DEGs profile in L. striatellus might be involved in the host preference of viruliferous L. striatellus.

Gene Ontology Classification of Differentially Expressed Genes among the Chemoreception Organs of Females and Males
The DEGs among the antennae, legs and heads were functionally annotated and classified into different molecular function categories by gene ontology (GO) analysis ( Figure 2). Eight pairwise comparisons between antennae and other tissues revealed a total of 1628, 1234, 1510, 1267, 1451, 1107, 1514 and 1409 DEGs with GO analysis. There were approximately 37 significantly enriched GO categories concerning molecular function seen in the DEGs of eight comparisons, of which the top three significant GO terms were protein binding (16.42~21.77%), DNA/RNA/chromatin binding (6.63~9.77%) and odorantbinding/olfactory receptor activity (6.94~10.12%) (Figures 2A,B,D,E and S1A,B,D,E). In the DEGs' molecular function of heads vs. legs, approximately 35 significantly enriched GO categories were identified with the top three GO terms involved in protein binding, DNA/RNA/chromatin binding and G-protein coupled receptor activity or peptidase activity (Figures 2C,F and S1C,F).

Gene Ontology Classification of Differentially Expressed Genes among the Chemoreception Organs of Females and Males
The DEGs among the antennae, legs and heads were functionally annotated and classified into different molecular function categories by gene ontology (GO) analysis ( Figure  2). Eight pairwise comparisons between antennae and other tissues revealed a total of 1628, 1234, 1510, 1267, 1451, 1107, 1514 and 1409 DEGs with GO analysis. There were approximately 37 significantly enriched GO categories concerning molecular function seen in the DEGs of eight comparisons, of which the top three significant GO terms were protein binding (16.42~21.77%), DNA/RNA/chromatin binding (6.63~9.77%) and odorantbinding/olfactory receptor activity (6.94~10.12%) (  When comparing the transcriptomes between females and males, only 152, 93, 286, 41, 115 and 121 genes were differentially expressed between VFA vs. VMA, VFL vs. VML, VFH vs. VMH, NFA vs. NMA, NFL vs. NML, NFH vs. NMH, respectively (Figures 2G-I and S1G-I). The enriched GO terms for these sex-specific DEGs among the three tissues include protein binding, structural constituent of the ribosome, phosphatase activity, peptidase activity, and so on.
Taken together, these results demonstrated that a large number of genes involved in the regulation of cellular environment and peripheral chemosensory were more or less expressed in chemoreception organs, antennae and legs, but a low number of DEGs were found between females and males. It suggested that these organ-and sex-specific DEGs might participate in the host preference of L. striatellus on rice, not considering whether the samples are from viruliferous or nonviruliferous L. striatellus. When comparing the transcriptomes between females and males, only 152, 93, 286, 41, 115 and 121 genes were differentially expressed between VFA vs VMA, VFL vs VML, VFH vs VMH, NFA vs NMA, NFL vs NML, NFH vs NMH, respectively (Figures 2G-I and S1G-I). The enriched GO terms for these sex-specific DEGs among the three tissues include protein binding, structural constituent of the ribosome, phosphatase activity, peptidase activity, and so on.

Differentially Expressed Genes Induced by RSV Infection
Thus, we next inspected the gene expression profile of viruliferous and nonviruliferous organs through a comparative heatmap. The gene expression patterns for the same organ clustered together. Among these clustered samples, the viruliferous samples were outgrouped from the nonviruliferous samples (

Gene Ontology Classification of Differentially Expressed Genes Induced by RSV Infection
Hundreds of DEGs between viruliferous and nonviruliferous tissues were functionally annotated by GO analysis and were classified into 28~33 significantly enriched categories between viruliferous and nonviruliferous organs ( Figure 5A-F). The top ten GO categories included protein binding, peptidase activity, structural constituent of cuticle, DNA/RNA/nucleotide/chromatin binding, transferase activity, transmembrane transporter activity, hydrolase activity, catalytic activity, ion binding and odorant-binding/olfactory receptor activity ( Figure 5A,B,D,E). There were only 18 and 24 significantly enriched GO terms from nonviruliferous vs viruliferous male and female heads, respectively ( Figure 5C,F). In particular, a large number of olfactory-related genes (7.14%, 4.51%, 2.14% and 4.17%) were found to be more or less enriched in viruliferous chemoreception organs (antennae and legs) as compared to nonviruliferous chemoreception organs ( Figure  5A,B,D,E). The data indicated a link of the olfactory related genes to the difference between the viruliferous and nonviruliferous chemoreception organs, however, deep analysis of these genes may shed light on the interaction between RSV and the vector's chemoreception organs.

Gene Ontology Classification of Differentially Expressed Genes Induced by RSV Infection
Hundreds of DEGs between viruliferous and nonviruliferous tissues were functionally annotated by GO analysis and were classified into 28~33 significantly enriched categories between viruliferous and nonviruliferous organs ( Figure 5A-F). The top ten GO categories included protein binding, peptidase activity, structural constituent of cuticle, DNA/RNA/nucleotide/chromatin binding, transferase activity, transmembrane transporter activity, hydrolase activity, catalytic activity, ion binding and odorant-binding/ olfactory receptor activity ( Figure 5A,B,D,E). There were only 18 and 24 significantly enriched GO terms from nonviruliferous vs. viruliferous male and female heads, respectively ( Figure 5C,F). In particular, a large number of olfactory-related genes (7.14%, 4.51%, 2.14% and 4.17%) were found to be more or less enriched in viruliferous chemoreception organs (antennae and legs) as compared to nonviruliferous chemoreception organs ( Figure 5A,B,D,E). The data indicated a link of the olfactory related genes to the difference between the viruliferous and nonviruliferous chemoreception organs, however, deep analysis of these genes may shed light on the interaction between RSV and the vector's chemoreception organs.

Changes in the Expression of Olfactory-Related Genes in the Chemoreception Organs Infected by RSV
In order to validate the effect of RSV infection on the olfactory system, a more indepth expression profiling the olfactory-related chemosensation and olfaction DEGs was performed. A total of 17 olfactory-related genes were differentially expressed in viruliferous antennae compared to nonviruliferous antennae (Figure 6 left). A cluster of olfactory genes, including LstrOBP4, LstrCSP2/5, LstrGR28a, LstrIR1/NMDA1, LstrOR67/85e and LstrSNMP2/2-2, showed higher abundance after infection in both male and female antennae (Figure 6 left). The expression level of six other olfactory genes (LstrOBP9,

Validation of the OBPs Expressions in RSV-Infected Chemoreception Organs by Quantitative Real-Time PCR
To confirm the above transcriptome results, we selected a significant differentially expression OBP from each organ transcriptome for qPCR examination. The expression of selected LstrOBP9 gene in viruliferous female antennae was significantly down-regulated by 33.1% compared to that in nonviruliferous female antennae, while it exhibited no significant change in viruliferous vs. nonviruliferous male antennae ( Figure 7A). On the contrary, the expression level LstrOBP3 in viruliferous female legs was about 4-fold of that in nonviruliferous female legs ( Figure 7B). Similarly, the LstrOBP2 was also highly expressed in the heads of viruliferous female, but no changes were found between viruliferous and nonviruliferous male heads ( Figure 7C). In general, the expression trends of the three LstrOBPs were consistent with our transcriptome data, supporting the involvement of these organ-and sex-specific olfactory-related LstrOBPs in mediating the host preference of L. striatellus on rice plant behavior as well as RSV transmission.

Validation of the OBPs Expressions in RSV-Infected Chemoreception Organs by Quantitative Real-Time PCR
To confirm the above transcriptome results, we selected a significant differentially expression OBP from each organ transcriptome for qPCR examination. The expression of selected LstrOBP9 gene in viruliferous female antennae was significantly down-regulated by 33.1% compared to that in nonviruliferous female antennae, while it exhibited no significant change in viruliferous vs nonviruliferous male antennae ( Figure 7A). On the contrary, the expression level LstrOBP3 in viruliferous female legs was about 4-fold of that in nonviruliferous female legs ( Figure 7B). Similarly, the LstrOBP2 was also highly expressed in the heads of viruliferous female, but no changes were found between viruliferous and nonviruliferous male heads ( Figure 7C). In general, the expression trends of the three LstrOBPs were consistent with our transcriptome data, supporting the involvement of these organ-and sex-specific olfactory-related LstrOBPs in mediating the host preference of L. striatellus on rice plant behavior as well as RSV transmission.

Discussion
The small brown planthopper, as one of the most destructive rice pests, is notorious for causing serious yield loss by feeding damage as well as transmitting RSV [29]. Before the release of the L. striatellus genome sequence in 2017 [30], several previous pyrosequencing-based transcriptome studies have characterized the whole body or organ transcriptomic response [23,26]. In the current study, based on a high-quality genome assembly and annotation of L. striatellus, we performed transcriptomic RNA sequencing to analyze gene expression profiles in three tissues (antennae, legs and heads) from L. striatellus females and males with/without RSV infection. The identification of these DEGs would provide insights into the host preference of L. striatellus on rice plant behavior as well as RSV transmission.
The GO analysis for these DEGs found that the top listed GO terms included protein binding, DNA/RNA/chromatin binding, odorant-binding/olfactory receptor activity, Gprotein coupled receptor activity and peptidase activity, suggesting a tissue-or sex-specific signaling reaction. Moreover, the comparison of DEGs between antennae vs legs/heads was more enriched in the regulation of the cell cycle, cellular environment and

Discussion
The small brown planthopper, as one of the most destructive rice pests, is notorious for causing serious yield loss by feeding damage as well as transmitting RSV [29]. Before the release of the L. striatellus genome sequence in 2017 [30], several previous pyrosequencingbased transcriptome studies have characterized the whole body or organ transcriptomic response [23,26]. In the current study, based on a high-quality genome assembly and annotation of L. striatellus, we performed transcriptomic RNA sequencing to analyze gene expression profiles in three tissues (antennae, legs and heads) from L. striatellus females and males with/without RSV infection. The identification of these DEGs would provide insights into the host preference of L. striatellus on rice plant behavior as well as RSV transmission.
The GO analysis for these DEGs found that the top listed GO terms included protein binding, DNA/RNA/chromatin binding, odorant-binding/olfactory receptor activity, G-protein coupled receptor activity and peptidase activity, suggesting a tissue-or sexspecific signaling reaction. Moreover, the comparison of DEGs between antennae vs. legs/heads was more enriched in the regulation of the cell cycle, cellular environment and peripheral chemosensory than those of heads vs. legs. Since most olfactory sensilla of insects are distributed on the surface of primary chemoreception organs, the antennae, while insect legs and heads are secondary chemoreception organs and olfactory signal processing organs, respectively [31], it is reasonable that olfactory-related genes were higher expressed in the antennae than that in the legs and heads. Thus, the increase or decrease in olfactory-related genes in antennae after virus infection has dire implications for vectorial capacity.
To the best of our knowledge, there has been no systematic study on the effect of virus infection on the olfactory genes expression of insect vector's antennae. Our study demonstrated overall changes in insect chemoreception organs in response to infection with plant virus RSV. Transcriptomic comparisons of viruliferous vs. nonviruliferous chemoreception organs indicated that a large number of olfactory related genes were differentially expressed in viruliferous antennae and legs but not in the heads. A previous report also found that dengue virus could infect the mosquito's antennae and result in changes in the transcript abundance of two AgOBPs [32]. The SRBDV infection led to a decrease in SfOBP2 and SfOBP11 in S. furcifera [7]. However, the TCV infection increased the expression of BtOBP3 in B. tabaci [8]. Li and his colleagues found that the LstrORco was stimulated by RSV in the head (with antennae) [28]. However, our data showed that RSV infection down-regulated the LstrORco expression in female L. striatellus legs and did not affect the LstrORco expression in antennae and heads (without antennae). These different results may be due to different sampling and detection methods. The qPCR method was used to detect the LstrORco expression in mixed samples of heads and antennae in Li and his colleagues' study, while transcriptome sequencing was used to detect the LstrORco expression in antennae and heads (without antennae) in our study. The protein level expression and the location of LstrORco in chemoreception organs need to be further verified by Western blot and in situ hybridization.
Previous reports have demonstrated that plant viruses could regulate olfactory genes to affect the behavior of insect vectors and their associations with host plants. SRBDV and TCV reversed the preferences of insect vectors by regulating OBPs [7,8]. Li and his colleagues reported that silencing of the LstrORco expression inhibited the host seeking behavior and increased the 'no response' percent and the response time of L. striatellus [28]. Our transcriptome analysis revealed 17 olfactory-related genes differentially expressed in viruliferous antennae (including two LstrOBPs, three LstrCSPs, three LstrGRs, three LstrIRs, four LstrORs and two LstrSNMPs) as well as five LstrOBPs, three LstrCSPs, six LstrIRs, four LstrORs and LstrORco differentially expressed in viruliferous legs. Additionally, only four olfactory genes were differentially abundant in viruliferous heads. During insect chemical perception, OBPs and CSPs can selectively transport specific odorants to four classes of membrane proteins for odorant recognition [14]. It is possible that the expression changes in olfactory-related genes alter the plant's volatile organic compound (VOC) recognition and affect insect behavioral responses. Taken together, it was implied that the RSV induction of olfactory related genes, including LstrORco, in antennae and legs could alter the feeding behavior of L. striatellus on rice with the host preference and, thus, at least theoretically, affect virus transmission. In the next step, the plant VOCs recognized by differentially expressed olfactory genes can be screened using gas chromatographyelectroantennographic detection (GC-EAD). The two electrodes voltage clamp technique and field trapping experiments can be conducted to validate the interaction between differentially expressed olfactory genes and plant VOCs. Through these techniques, we will further reveal the role of olfactory genes in RSV-induced behavior.

Nonviruliferous and Viruliferous L. striatellus Rearing
Nonviruliferous and viruliferous strains of L. striatellus were a gift from Prof. Yijun Zhou's laboratory of Jiangsu Academy of Agricultural Sciences, Jiangsu province, China, and were reared independently on seedlings of rice cv. Wuyujing 3 in a growth incubator at 25 ± 1 • C, with 80% ± 5% RH and a 12-h light-dark photoperiod.
To ensure that insects were viruliferous, individual female insects were allowed to feed independently. The presence of RSV in their offspring were collected and detected via Dot-ELISA with the monoclonal anti-CP antibody [33]. Highly viruliferous colonies were screened and prepared for subsequent studies.

Samples Preparation and Transcriptomic Sequencing
The antennae, legs and heads (without antennae) of more than 3000 L. striatellus adults were dissected independently. Viruliferous female/male and nonviruliferous female/male samples were collected and stored at −80 • C for each RNA library. TRIzol method was used to extract total RNA from organ samples as recommended by the manufacturer (Ambion, Austin, TX, USA). RNA purity and concentration were checked using the NanoPhotometer ® spectrophotometer (IMPLEN, Westlake Village, CA, USA) and RNA assay kit in Qubit ® 2.0 Fluorometer (Life Technologies, Carlsbad, CA, USA). Twelve paired-end RNA-seq libraries were established from the extracted RNA using the NEBNext ® Ultra™ RNA Library Prep Kit (New England BioLabs, Ipswich, MA, USA). First and second strand cDNA were synthesized according to the methods described previously [27]. After the adapters had been attached, PCR was performed with the adapter-ligated cDNA as templates, phusion high-fidelity DNA polymerase, universal PCR primers and index (X) Primer to generate cDNA libraries. These libraries were sequenced with Illumina HiSeq 2000 sequencer (Illumina, San Diego, CA, USA) and 125-150 bp paired-end reads were generated.

Transcriptomic Assembly and GO Annotation
After sequencing, FastQC was employed to check the quality distribution of the raw data (http://www.bioinformatics.babraham.ac.uk/projects/fastqc, accessed on 20 December 2018). The clean reads were obtained by removing reads containing adapter, reads containing ploy-N and low quality reads from raw reads. The Q20, Q30 and GC content of the clean data were calculated. All paired-end clean reads were aligned to a reference genome of L. striatellus using HISAT2 v2.0.5 (http://www.ccb.jhu.edu/software/hisat, accessed on 25 December 2018) and assembled by StringTie software [34] to generate mapped unigenes. The reference genome of L. striatellus (GigaDB RRID: SCR004002) and gene model annotation files were released by Zhu's lab in the GigaScience repository [30]. All mapped unigenes were annotated by Gene Ontology (GO) enrichment analysis. GO terms with corrected p-value less than 0.05 were considered significantly enriched.

Analysis of Differentially Expressed Genes
Read counts mapped to each unigene were calculated by featureCounts v1.5.0-p3 [35]. FPKM of each unigene was obtained based on the length of the gene and reads count mapped to this gene. Unigene expression profiles were performed by heatmapping and Heml software [36]. DEGs between two treatments were identified using the edgeR R package [37]. Corrected p-value of 0.05 and absolute fold change of 2 were set as the threshold for significantly differential expression according to Benjamini and Hochberg method [38].

Identification and Comparative Expression Profiles of Olfactory Related Genes
Based on GO term described as odorant-binding/olfactory receptor activity, all unigenes were manually retrieved by keywords (OBP, CSP, OR, IR. GR SNMP, chemosensory protein, olfactory protein). Then, retrieved olfactory-related genes were matched with 128 olfactory genes in previous description [27] and identified using a BLASTx algorithmbased search in NCBI website. Expression profiles of these olfactory-related genes were displayed using Heml software according to FPKM of each gene [36].

Quantitative Real-Time PCR Analysis
Total RNA was isolated from 600 antennae, 100 legs and 50 heads of viruliferous and nonviruliferous adults using the TRIzol Total RNA Isolation Kit (Takara, Dalian, China).
First-strand cDNA was synthesized using One Step SYBR PrimeScript RT-PCR kit (Takara, Dalian, China). Primers for LstrOBPs and LstrActin (control) (Table S2) were designed as previous description [27]. The qPCR was conducted with SYBR Premix Ex Taq (Takara) and a CFX96™ Real-Time PCR Detection System (Bio-Rad, Hercules, CA, USA) as follows: denaturation for 3 min at 95 • C, followed by 40 cycles at 95 • C for 10 s, and 60 • C for 30 s. Relative expression levels for triplicate samples were calculated using the 2 −∆∆Ct method [39], and expression levels of target genes were normalized to the LstrActin gene. Three technical repeats were performed for each of the three biological replicates. The T-test method was used to analyze the significance of relative expression level from various samples [40].

Conclusions
In summary, our transcriptomic analysis revealed the expression profiles of genes in antennae, legs and heads of L. striatellus infected by RSV. RSV infection regulated the expression of multiple olfactory-related genes in the a chemoreception organs. In particular, qPCR confirmed the down-regulation of LstrOBP2, and increased LstrOBP3 and LstrOBP9 in the chemoreception organs of females L. striatellus post RSV infection. It is suggested that RSV may alter host preference of the insect vector by changing these olfactory-related genes, which are ultimately beneficial to its transmission. Therefore, it is promising to develop approaches (such as RNAi based gene-silencing technology) to interrupt the olfactory-related genes in chemoreception organs of L. striatellus to control the transmission and spread of RSV in the future. In this regard, this work is a starting point for research, which will eventually help the control of RSV and provide new strategies for the control of other vector-borne viruses.
Supplementary Materials: The following are available online at https://www.mdpi.com/article/10 .3390/ijms221910299/s1, Figure S1: GO analysis of DEGs among male organs (or between female and male viruliferous organs), Table S1: Summary statistics of Illumina sequence data, Table S2: The primers used in this study.