Characterization and Expression Profiling of Neuropeptides and G-Protein-Coupled Receptors (GPCRs) for Neuropeptides in the Asian Citrus Psyllid, Diaphorina citri (Hemiptera: Psyllidae)

Neuropeptides are endogenous active substances that widely exist in multicellular biological nerve tissue and participate in the function of the nervous system, and most of them act on neuropeptide receptors. In insects, neuropeptides and their receptors play important roles in controlling a multitude of physiological processes. In this project, we sequenced the transcriptome from twelve tissues of the Asian citrus psyllid, Diaphorina citri Kuwayama. A total of 40 candidate neuropeptide genes and 42 neuropeptide receptor genes were identified. Among the neuropeptide receptor genes, 35 of them belong to the A-family (or rhodopsin-like), four of them belong to the B-family (or secretin-like), and three of them are leucine-rich repeat-containing G-protein-coupled receptors. The expression profile of the 82 genes across developmental stages was determined by qRT-PCR. Our study provides the first investigation on the genes of neuropeptides and their receptors in D. citri, which may play key roles in regulating the physiology and behaviors of D. citri.


Introduction
The central nervous system (CNS) and its neuropeptide messengers rank the highest in the entity level regulating endogenous biochemical control function [1]. Neuropeptides are a diverse set of signaling molecules in multicellular organisms. In insects, neuropeptides play a significant role in the regulation of fundamental events such as development, reproduction, feeding, courtship, olfaction, circadian rhythm, and many other processes [2,3]. Neuropeptides are processed from their larger, inactive precursors by enzymes [4], up activation, and then work on target cells by binding to the specific signal-transducing membrane receptors [5]. Most of these receptors are subordinate to the G-protein-coupled receptors (GPCRs), and the GPCRs have a similar topographical structure with seven transmembrane domains which are highly conservative through evolution and constitute the largest superfamily of cell surface proteins [6]. In vivo studies showed that neuropeptides and their receptors appear to have key roles in the regulation of physiology and behavior in insects; injection of matched at least one GO term ( Figure 1). Among these unigenes, 13,536, 11,569, and 10,489 transcripts were assigned to biological processes, molecular function, and cellular component, respectively. Cellular process and metabolic processes represented the most abundant GO terms in the biological process category, binding and catalytic activity were most represented in molecular function, and the most unigenes that accorded with the cellular component category were involved in cell parts and organelles ( Figure 1). The raw data of the transcriptomic were submitted to the NCBI Short Read Archive (SRA) database as BioProject Accession Number SRP139008 (https://www.ncbi.nlm.nih.gov/sra/SRP139008).

G-Protein-Coupled Receptors (GPCRs) for Neuropeptides
A total of 42 putative neuropeptide GPCRs genes were identified in the transcriptomes of D. citri based on homology analysis (Table 2). Of these receptors, 35 GPCRs belong to the A family, four belong to B family, and three belong to leucine-rich repeat-containing GPCRs (LGRs). In order to assign putative functions of these GPCRs, we compared them with those of D. melanogaster, N. lugens, and other arthropods. The results are presented as a neighbor-joining tree in Figures 2-4.
Here the B-family neuropeptide GPCRs belong to the subfamily B1. Four B-family GPCRs associated with neuropeptide recognition were confirmed in D. citri. Each amino acid sequence of these candidate receptors contains the characteristic hormone receptor domain. In the phylogenetic tree of B-family GPCRs (Figure 3), four receptors were classified into four branches. DcB1 and DcB4 were identified as receptors of DH31 and DH44. DcB1 is the ortholog of DmCG32843, and DcB4 is the ortholog of CG12370 and DmCG8422. DcB2 was annotated as the receptor of PDF, which is involved in prolonging mating duration. DcB3 is an ortholog of DmCG4395, and it shows high similarity to the calcitonin-like diuretic hormone receptor 2 of R. prolixus (AHB86571, e-value 0.0).

Leucine-Rich Repeat-Containing GPCRs (LGRs)
Leucine-rich repeat-containing GPCRs are a class of GPCRs containing leucine-rich repeats (LRRs) at the N-terminus. Based on the number of LRRs, the type-specific hinge region and the presence or absence of a low-density lipoprotein receptor-like cysteine-rich motif (LDLa), LGRs can be identified as three main types (type A, B, and C). Furthermore, according to the number of LDLa motifs, Type C LGRs can be divided into two subtypes: Type C1, which contain only one LDLa, and Type C2, which contain multiple LDLa motifs. Three LGRs were identified in D. citri, two belong to Type A and one belong to Type C2. Both DcLGR1 and DcLGR2 contain eight LRRs, and this is a typical feature of Type A. In the phylogenetic tree of leucine-rich repeat-containing GPCRs (Figure 4), DcLGR1 and DcLGR2 are homologous to DmCG7665 which is the receptor of glycoprotein hormones in Drosophila. DcLGR3 contains eight LDLa motifs and two LRRs, and it belongs to Type C2. DcLGR3 is an ortholog of NlA47 which is also known as NlGRL101.

Tissue-Specific Expression Profiles of the Neuropeptides and Neuropeptide Receptors
In order to understand the potential function of the neuropeptides and neuropeptide receptors in D. citri, the expressions of the neuropeptides and neuropeptide receptors were profiled for different tissues of D. citri based on the transcriptome data ( Figure 5 In order to understand the potential function of the neuropeptides and neuropeptide receptors in D. citri, the expressions of the neuropeptides and neuropeptide receptors were profiled for different tissues of D. citri based on the transcriptome data ( Figure 5). AKH, DH31, SIF, AstB, CCH2, ILP, EH2, CNM, ELP, CCAP, and Kin are ubiquitous in the tissues of the female and male adult. AstA-C, TK, CAPA, ITG, CCH1-2, MS, sNPF, OKA, ILP1-2, SIF, NTL, EH2, AVLP, PBAN, NPLP1, and CNM show high expression levels in head, and AstA, CAPA, ITG, IPL2, NTL; PDF express in head and antenna; CNM also shows the highest expression level in legs of all tissues; and ELP and CCAP have a high expression level in abdomen terminal.

Developmental Stages Expression Analysis by RT-qPCR
Developmental stages expression profiles of the neuropeptides and neuropeptide receptors were confirmed by the RT-qPCR (Figures 7 and 8). AKH, AVLP, Burβ, DH31, DH45, PBAN, and OK had the highest expressional level in the 1st and 2nd instar nymphs. AstB, AstC, CAP2b, CCH1, CNM, GPb5, MS, and TK showed the highest expression of the adults at five days after eclosion (sexual maturity). The expression level of Kin, NP, and PTTH decreased gradually from egg to

Discussion
Control of D. citri is the key component of integrated control for citrus Huanglongbing. However, effective control measures are not currently available. As potential pesticides and new targets emerge on account of continuing development of existing insecticide resistance [1,25], the primary task of developing neuroendocrine-based insecticides is to identify the structure and function of neuropeptides and receptors, especially those involved in survival, development, and/or reproduction [11]. However, the molecular basis of the behavior of D. citri is poorly understood. The y-axis represents the relative expression level and the x-axis the life cycle. The standard error is represented by the error bar and significant differences are represented by the different letters (p < 0.05). E, egg; N1-2, 1st-and 2nd-instar nymphs; N3, 3rd instar nymphs; N4, 4th instar nymphs; N5, 5th instar nymphs; 1D-A, the adults of one day after eclosion; 5D-A, five days after eclosion.

Discussion
Control of D. citri is the key component of integrated control for citrus Huanglongbing. However, effective control measures are not currently available. As potential pesticides and new targets emerge on account of continuing development of existing insecticide resistance [1,25], the primary task of developing neuroendocrine-based insecticides is to identify the structure and function of neuropeptides and receptors, especially those involved in survival, development, and/or reproduction [11]. However, the molecular basis of the behavior of D. citri is poorly understood. Besides, the lack of genomic information prevents us from understanding the regulatory network of the neuropeptides. Fortunately, transcriptome analysis has provided the methods to identify and characterize multiple genes in insects [23], and using RNA-seq a number of genes encoding neuropeptides and their receptors were identified from D. citri.
The present transcriptional sequences of D. citri appear to contain most of the genes for insect neuropeptides and GPCRs except for several genes. Several neuropeptides (e.g., Trissin (TR) and IMFamide (IMF) orthologs) were also not identified in other Hemipteran insects such as A. pisum [26], N. lugens [23], and R. prolixus [11]. In fact, IMF is a unique neuropeptide of Lepidoptera and has not been found in other insects [20]. Neuropeptide F (NPF) is identified in almost all insect species and plays roles in feeding, metabolism, reproduction, and stress responses [20], whereas the BLAST search of transcriptomic data failed to find NPF in D. citri. Similarly, we have identified an ortholog of Proctolin; however, the receptor was not found. This may be due to the incomplete analyses of transcriptome information. Indeed, the gene coding for natalisin (NTL) and CNMamide (CNM) were identified from R. prolixus by reinvestigation [11] contrary to the previous suggestion that NTL and CNM were absent [24].

Neuropeptides Involved in Ecdysis and Development
Ecdysis-triggering hormone (ETH), eclosion hormone (EH), and crustacean cardioactive peptide (CCAP) are main players of the peptidergic circuit controlling ecdysis in insects. The functions of these peptides have been reported in the Drosophila and other insects [27,28]. In insects, EH is expressed in CNS, ETH production in endocrine cells in the epitracheal gland (Inka cells), and the CCAP-expressing neurons are located in abdominal ganglia [27]. Both ETH and EH regulate the release of CCAP from central CCAP neurons which inhibits pre-ecdysis [2]. In D. citri, one ETH encoding gene was identified, and two different genes encoding EH were identified. According to the expression profiles, ETH and EH1 show high expression levels in the nymph stage, and we can presume that ETH and EH1 play a vital role in molting processes.
Juvenile hormone (JH) is an important hormone and regulates development and growth in insects. Traditionally JH production in the corpora allata has been considered to be controlled by the peptides Allatotropins (AT) [29] and Allatostatins (Ast) [2]. The first insect AT was isolated from head extracts of Manduca sexta. This peptide was shown to stimulate JH biosynthesis by the corpora allata (CA) of adult females [29]. Allatostatins are diverse peptides derived from three different genes in insects, which were designated as AstA, AstB, and AstC. These peptides act as the inhibitors of JH production in corpora allata (CA) and have antagonism to AT [30]. Feeding the Lepidopteran AstC led to reduced growth and fecundity and caused significant mortality in A. pisum and in Myzus persicae [31,32]. The function of AstC receptors has been characterized in A. aegypti and T. castaneum, and they express in the central nervous system and gut [33,34]. Besides that, AstCC is a Type-C Ast and shows strong similarities to AstC. AstCC also has the function to regulate the biosynthesis of JH [35]. In D. citri, from 3rd instar nymphs to sexual maturity, the expression of AstCC shows a decreasing trend, and AT shows a growth trend. This result suggests that AT and AstCC might be involved in the regulation of JH production in D. citri.

Neuropeptides Control of Metabolism
Insulin-like peptides (ILPs) widely exist in insects and the insulin-like signaling pathway is conserved across higher multicellular animals [36]. In Drosophila, ablation of the insulin-producing cells, or deactivation of the ILPR leads to a series of phenotypes including grow logy, increased starvation resistance, increased levels of circulating carbohydrates cycling, elevated lipid storage, and lifespan extension [2]. Adipokinetic hormone (AKH) is suggested to be similar to mammalian glucagon and acts antagonistically to insulin by activating glycogen phosphorylase and mobilizing carbohydrates [37]. When the AKH-producing neuroendocrine cells were ablated in Drosophila, the trehalose levels of larvae and starved adults decreased, and these adults without AKH-cells become hypoactive, suggesting that AKH is involved in maintaining normal levels of circulating carbohydrates [38]. Ectopic expression of AKH in the fat body resulted in both increased circulating trehalose and a decrease in stored lipids [2]. In D. citri, two genes encoding ILP1 and ILP2 and one gene encoding AKH were identified, and more research is needed to reveal the functions of these genes.
The insect kinins are multifunctional neuropeptides shown to modulate hindgut contractions [39,40], diuretic activity [41,42], digestive enzyme release [40,43], and inhibit weight gain in larvae Lepidoptera [43,44]. Feeding the analogs of kinin to the pea aphid demonstrated that three of the biostable analogs showed antifeedant activity [45]. In D. citri, we identified one kinin precursor and one kinin receptor (DcA23), the kinin and the kinin receptor both express in the abdomens (Figures 5 and 6). On the present understanding of the expression profile of Kin and the receptor, we speculate that kinin and DcA23 may be involved in regulating the digestion process in D. citri and they represent a potential pesticide and target.

Reproductive-Related Neuropeptides
SIFamide (SIF) is strictly conserved and widespread in insects and modulates sexual behavior, and it was first found in 1995 in HPLC (high-performance liquid chromatography) fractions [46]. In Drosophila, the expression of SIF was restricted to only four neurons of the pars intercerebralis, when the SIF neurons was ablated, the lacking SIF-less males perform vigorous and indiscriminate courtship directed at either sex, while females appear sexually hyper-receptive. When the SIF gene was knocked down via RNAi, the decrease of SIF also led to a similar change of behavior [47]. SIFamides show a conserved sequence, X1-X2-RKPPFNGSIFamide, and the SIFs differ only in their N-terminal amino [46], in D. citri, SIF and the receptor of SIF (DcA4) were detected.
Neuropeptide pigment-dispersing factor (PDF) is involved in maintaining behavioral rhythms in D. melanogaster [48]. In addition, males usually prolong mating duration in the presence of other males to increase the chance of successful gene transfer, this effect in D. melanogaster requires both Neuropeptide F receptor 1 (NPFR1) and PDF expressing in four small ventrolateral neurons as well as the PDF receptor and expressing Neuropeptide F (NPF) in two dorsolateral neurons [49]. Apparently, PDF and NPF work together to regulate prolonged mating duration in D. melanogaster. In this study, PDF was identified in D. citri; however, NPF was not found. Natalisin (NTL) is an arthropod-specific neuropeptide which was recently identified in three holometabolous insect species: D. melanogaster, T. castaneum, and B. mori, and was proven to be involved in regulating mating behavior in D. melanogaster and T. castaneum [50]. The latest research on NTL shows that NTL is involved in modulating the mating of the oriental fruit fly, Bactrocera dorsalis [51]. The NTL precursors generally contain multiple repeats sequences of F-X1-X2-X3-Ra at the C-terminus. In hemipteran, X1, is usually W, and X2 is P [52]. In D. citri, NTL precursors and the receptor of NTL (DcA34) was identified, according to the amino acid sequences of the NTL precursor, four mature peptides were predicted (Supplementary Figure S1). The conservatism of NTLs sequences implies the conservatism of the functions; NTL will be a worthwhile option in developing novel control methods against D. citri.

Neuropeptides in Olfaction
Tachykinin (TK) is a multifunctional peptide, and it has been identified in many vertebrate and invertebrate species. In all vertebrate and a few invertebrate the TKs share a common C-terminal sequence motif, F-X-G-L-Ra [53]. Tachykinin is important in odor-based searching behavior of fruit flies, and several olfactory neurons contain high TK levels [54]. In Drosophila, (3R,11Z,19Z)-3-acetoxy-11,19-octacosadien-1-ol (CH503) is a gustatory sex pheromone. CH503 is detected by gustatory neurons on the male foreleg, a cluster of 8-10 neurons within the subesophageal region which mediates the pheromone response through the release of the TKs [55]. In D. citri, we identified four mature peptides of TK; all mature peptides have conserved sequences at the C-terminus (Supplementary Figure S1). Tachykinin shows a high expression level in antenna and DcA24 shows high expression in male antenna and head in D. citri, but beyond that, TK also showed the highest expression in adults at five days after sexual maturity. Based on the expression profile, it is probable that TK and DcA24 are involved in the recognition of the odor by adults. Therefore, TK and the receptor can be considered as a potential behavioral regulatory pesticide or target.
CCH1 and CCH2 were initially identified in the tsetse fly Glossina morsitans [56]. In Drosophila, CCH1 and CCH2 played a role in appetite regulation by activating their receptors respectively [57,58]. The receptor of CCH1 was also an important factor governing starvation-induced olfactory modifications [59]. Here, both CCH1 and CCH2 were identified in D. citri, and their corresponding receptors were excavated, A14 (CCH1) and A15 (CCH2), respectively. In D. citri, CCH1 and A14 presented high expression in adulthood, besides that, CCH1 is highly expressed in antennae and head. According to the results, it was assumed that CCH1 and A14 may play an important role in regulating starvation-induced host recognition behavior at the adult stage of D. citri. In addition to TK and CCHs, AstA, sNPF, and SIF also show high expression level in antenna and head, and it is consistent with neuropeptides expressed in brain and the antennal lobe of other insects as in previous reports [47,60,61].

Insect Rearing and RNA Extraction
The insects were collected from the Murraya exotica in the campus of South China Agricultural University, Guangzhou, Guangdong Province, China in 2013. The laboratory population of D. citri was reared in a greenhouse (26 • C, 80% RH) and 14:10 h (light:dark) photoperiod. D. citri adults were transferred to cages (40 × 40 × 50 cm) which contained the saplings of M. exotica for oviposition and feeding 4 days, the plants with eggs were transferred to new cages. When the eggs hatched, each instar of D. citri was collected during the process of growth and tissues of insect dissected from newly emerged adults (3 days old). Samples were frozen in liquid nitrogen and stored at −80 • C until extraction.

RNA-seq
For the small size of D. citri, we concentrated the tissues of the insects by dissecting newly emerged adults (3 days old). A total of 2000 antennas (includes a modicum of tissues of heads), 200 heads (remove antennas), 150 thoraxes, 300 legs, 150 abdomens, and 1000 terminal abdomens (cut from the 5th abdominal segments) were collected from males, and the tissues from females had equal numbers. Total RNA of each sample was extracted using TRIzol Reagent (Invitrogen, Waltham, MA, USA). Total RNA of each sample was quantified and qualified by Agilent 2100 Bioanalyzer (Agilent Technologies, Palo Alto, CA, USA), NanoDrop (Thermo Fisher Scientific, Waltham, MA, USA). NEBNext ® Ultra™ RNA Library Prep Kit for Illumina ® (Illumina, San Diego, CA, USA) was used for next-generation sequencing library preparations. Then, we employed Agilent 2100 Bioanalyzer (Agilent Technologies, Palo Alto, CA, USA) for library quality evaluation. The libraries were sequenced on an Illumina HiSeq2500 the clean data was assembled by Trinity [62].

Identification of the Neuropeptides and Their Putative G-Protein-Coupled Receptors in D. citri
The amino acid sequences of D. melanogaster, B. mori, N. lugens, C. suppressalis and other arthropods were used as BLAST queries to search for the candidate neuropeptides and neuropeptides receptor genes from D. citri transcriptomic data. The E-value threshold for neuropeptides was 10, and the E-value threshold for receptors was 10 −5 . The candidate neuropeptides and neuropeptides receptor genes were reconfirmed by means of BLASTX analysis with the non-redundant protein sequence (NR) at NCBI (http://www.ncbi.nlm.nih.gov/).

Phylogenetic Analysis
Phylogenetic analysis of D. citri neuropeptides receptors was performed by comparing the neuropeptide GPCRs with those of D. melanogaster, B. mori, N. lugens, and other arthropods. For the Drosophila, the CG numbers of the sequences were used, and for the other species the original names of GPCRs in the publications were used, and the names of ligands were added if identified, the amino acid sequences can be seen in Supplementary Text S1. Sequences were aligned by ClustalW [67], and Neighbor-Joining trees were constructed in MEGA6 with 1000 bootstrap replicates [68]. The dendrograms were viewed in FigTree and edited in Adobe PhotoShop CS6.

Gene Expression Profiling of the Neuropeptides and Neuropeptide Receptors
Gene expression levels for each tissues sample of male and female antenna (included a modicum of tissues of heads), head (remove antenna), thorax, leg, abdomen, and abdomen terminal were estimated by RSEM (RNA-Seq by Expectation-Maximization) (v1.2.6) [69]. The expression levels were given as FPKM values (fragments per kilobase of transcript, per million fragments sequenced). Each of the FPKM values were transformed into log2 (RPKM + 1) values, and the expression profiling of the putative genes was generated and visualized by Heatmap Illustrator version 1.0 (http://hemi. biocuckoo.org/) [70]. Differential expression analysis used the DESeq package [71], six genes were selected to verify the accuracy of expression level by qRT-PCR (Supplementary Figure S2). The data analysis was conducted using GraphPad Prism 7.01. Statistical significance was evaluated using a one-way ANOVA followed by Tukey's multiple range test at the 0.05 level.

Developmental Stages Expression Analysis
qRT-PCR was used with SYBR-green fluorescence. Total RNA was extracted from eggs, nymphs (1st to 5th-instars), and adults of one day and five days (sexual maturity) after eclosion. The PrimeScript™ RT reagent Kit with gDNA Eraser (Takara, Dalian, China) was used for cDNA synthesis following manufacturer's instructions, reference gene Dcitactin-1 was used as the internal controls [72]. The gene-specific primers were designed by Primer 3 program (http://elixir.ut.ee/Main/ Services) [73] (Supplementary Table S2). The qRT-PCR reactions run on a CFX96 Touch™ Real-Time PCR System (Bio-Rad, California, USA), and the following program was adopted: 95 • C for 3 min, 40 cycles of 95 • C for 5 s and 59 • C for 30 s, and a final melting cycle (from 60 to 95 • C). Each experiment consisted of three biological replicates and three technical replicates. The relative values of mRNA expression were calculated by The 2 −∆∆CT method [74]. Data analysis was conducted using GraphPad Prism 7.01. Statistical significance was evaluated using a one-way ANOVA followed by the Tukey's multiple range test at the 0.05 level.

Conclusions
The main purpose of this study was to identify the genes that encoding neuropeptides and their putative receptors. Transcriptomic analysis of these genes revealed the neuropeptide system in D. citri. In total, 40 neuropeptide and 42 neuropeptide receptor genes were identified. Most of these genes were annotated for the first time in D. citri. The expression of all these genes in different tissues was analyzed based on transcriptome profiling using RNA-seq data. The genes show high expression in the antenna and/or abdomen was speculated as representative potential pesticides or targets. Furthermore, these data contain more genes that are appropriate as new pesticides or targets which we did not mention; future research will prove that they are valuable in insect pest management.