Background

The visual system is essential for insects to collect abundant information from their surroundings and respond appropriately, including making choices among alternative behaviors, foraging, and mating [1,2,3,4]. Generally, the spectral sensitivity of photoreceptor cells is determined by the photopigment, a complex of chromophore and opsin [5]. The chromophore, usually the 11-cis-retinal derived from vitamin A, forms a covalent bond with a lysine residue (K) in the opsin, a light-activated G-protein coupled receptor, via a Schiff base [6,7,8]. Opsins can be usually categorized into three groups based on their spectral peaks, namely, long-wavelength-sensitive (LW), blue-sensitive (B), and ultraviolet (UV)-sensitive proteins [9]. The majority of insects possess this ancestral trichromatic visual system, such as Drosophila melanogaster (Diptera) [10, 11], Apis mellifera (Hymenoptera) [12], Spodoptera exigua (Lepidoptera) [13], and Nephotettix cincticeps (Hemiptera) [14]. As more and more genome databases have been constructed, insects are found to possess various numbers or types of opsins due to gene duplications or losses [15]. For instance, one UV-sensitive, five B-sensitive, and up to ten LW-sensitive opsins are present in Sympetrum frequens (Odonata) [16]; meanwhile, the ancestral B-sensitive opsin is lost in Coleoptera insects, but the trichromacy at least twelve times is regained by duplicating the UV-sensitive and LW-sensitive opsin genes [17, 18].

In Drosophila, Rh1-6 opsins are expressed in the compound eyes, among which, Rh1 long-sensitive opsin is expressed in photoreceptor cells 1–6, Rh3/4 UV-sensitive opsins in photoreceptor 7 cells, whereas photoreceptor 8 cells express the Rh5 opsin for blue light detection and the Rh6 opsin for green light detection [11, 19, 20]. Another opsin, Rh7, found in the central brain, helps regulate circadian light entrainment by circadian pacemaker neurons [21]. In addition to the above ancestral visual functions, insect opsins have also evolved additional functions that are crucial for insect survival and fitness [22, 23]. For instance, in Drosophila, Rh1 plays a crucial role in the initiation of thermosensory signaling cascades for thermotaxis [24]; additionally, two other opsins, Rh5 and Rh6, function in the brain and body wall to support thermal preference during the late-3rd instar larvae [25]; furthermore, Rh6 is also responsible for the cooling-induced activation of bitter gustatory receptor neurons [26]; and opsins are also suggested to be linked to morphological differentiation, like the evolution of distinct yellow wing pigment along with a second UV opsin in Heliconius butterflies [27]. Despite the extensive research regarding the impact of opsins on the visual function across various insect species [23], the unconventional roles of these opsins in insects remain largely mysterious.

The Nilaparvata lugens (Hemiptera), belonging to the Delphacidae family, is the most destructive rice pest in Asia. It is particularly notable for its wing dimorphism and the capability of long-distance migration, allowing it to exploit its exclusive rice host in temperate regions and inflict significant damage on rice crops [28]. A previous study has identified three opsin genes in N. lugens genome, including one LW- and two UV - absorbing opsins (Nllw, and NlUV1/2, respectively) [14]. However, it is still unknown whether there are other functions except for visual sensing. In this study, opsin genes in the N. lugens were systematically investigated based on transcriptomic and genomic data, and an additional new opsin with hypothesized UV peak sensitivity (NlUV3-like) was identified. Thereafter, the functions of these four opsin genes were investigated, respectively. According to our results, Nllw took part in the melanization of body color and might be involved in the tyrosine-mediated cuticle melanism pathway. As far as we know, this study is the first to reveal the unconventional function of opsins in Hemiptera and suggests a cross-talk between gene pathways underlying the visual system and the morphological differentiation in insects.

Results

Identification of opsin genes from the genome and transcriptome databases

To identify the potential opsin genes in N. lugens, the Drosophila ninaE protein sequence (NP_524407.1) was used as a query to search the corresponding genome and transcriptome databases of N. lugens [29, 30]. As a result, altogether four opsin genes were identified, including the previously reported Nllw (GenBank accession number: AB761147.1), NlUV1 (GenBank accession number: AB761148.1), NlUV2 (GenBank accession number: AB761149.1), and an additional paralog. Blast analysis of the paralog revealed a 99% identity with a UV-sensitive opsin predicted from an N. lugens genome sequence (GenBank accession number: NC_052505.1), indicating that a UV-sensitive-like opsin (NlUV3-like) (GenBank accession number: XP_022205432) was detected in N. lugens.

Sequence and phylogenetic analysis

The four opsins, Nllw, NlUV1/2, and NlUV3-like were predicted to be G protein-coupled receptors with seven transmembrane α-helices (TMH1-7) [6] and contained a conserved K in TMH7, which covalently binds to the chromophore [7, 8]. A total of 386, 380, 380, and 499 amino acids (aa) peptides were predicted in Nllw, NlUV1/2, and NlUV3-like, respectively. The conserved motif histidine-glutamine-lysine (HEK) involved in G-protein binding [31] was present at the C-terminal side of TMH5 of Nllw and NlUV1/2, but not in NlUV3-like (Fig. 1A). Moreover, these opsins were located on different chromosomes in the N. lugens genome, with Nllw and NlUV3-like on chromosomes (Chr) 13 and 01, separately; whereas NlUV1/2 with tandem distribution on Chr 04 (Fig. 1B). Structural analysis revealed differences in the number of exons spanning TMH1-7, namely, 2nd − 8th (E2-8) in Nllw, 2nd − 7th (E2-7) in NlUV1/2, and all the four exons (E1-4) in NlUV3-like (Fig. 1C). Furthermore, a phylogenetic tree analysis suggested that the Nllw, NlUV1, and NlUV2 clustered well with the long-wave opsins, UVopsin1, and UVopsin2 in Sogatella furcifera (Hemiptera) and Laodelphax striatellus (Hemiptera), respectively, which was consistent with a previous study [14]. However, the NlUV3-like was distinct from NlUV1/2, which demonstrated strong clustering with the Rh7 (GenBank accession number: NP_524035.2) in D. melanogaster, with Rh4 (GenBank accession number: XP_021702394.1) in Aedes aegypti (Diptera), as well as with Rhodopsin7 (GenBank accession number: QDQ16904.1) and a UV-like opsin (GenBank accession number: XP_032513103.1) in Danaus plexippus (Lepidoptera) (Fig. 2). Overall, the sequence characteristics, Chr location, gene structure, and phylogenetic analysis highlights that Nllw, NlUV1/2, and NlUV3-like are divided into three distinct subfamilies. And the emergence of NlUV1/2 is likely attributed to the duplication of a common ancestor opsin. Meanwhile, NlUV3-like shares homology with other opsin variants, such as Rh7 in D.melanogaster, which helps regulate circadian light entrainment in circadian pacemaker neurons [21].

Fig. 1
figure 1

Sequence characteristics analysis. A. Alignment of the amino acid sequences of Nllw, NlUV1/2, and NlUV3-like. Seven transmembrane α-helices (TMH1-7) are indicated with red frames; The conserved motif histidine-glutamine-lysine (HEK) is denoted by a blue frame; The conserved lysine (K) in TMH7 is marked below; Black background: 100% conserved amino acid sites; Gray background: 80% conserved amino acid sites. Red and green words: the 1st and 2nd regions for RNAi against Nllw, respectively. B. An ideogram of Nllw, NlUV1/2, and NlUV3-like on three chromosomes. Chr 13/04/01: chromosome 13/04/01; The arrow representing Nllw pointing to the left: means the Nllw transcript is in the opposite direction to Chr 13; The arrow representing NlUV1/2 and NlUV3-like pointing to the right means the NlUV1/2 and NlUV3-like transcripts are in the same direction to Chr 04 and 01; Solid line means tandem replicates of NlUV1/2. C. The distribution of Nllw, NlUV1/2, and NlUV3-like coding sequences within the genomic sequences. Gray boxes indicate exons (E), and the width of each box is proportional to the size of the corresponding exon. Figures below each exon are the sizes of corresponding exons. Black lines represent introns, which are not proportional to the size of the corresponding introns. The location of seven transmembrane α-helices on exons is marked with black dotted frames

Full size image
Fig. 2
figure 2

Phylogenetic analysis. A phylogenetic tree with the amino acid sequences of Nllw, NlUV1/2, NlUV3-like, and 28 homologs from other insects. The names represent the abbreviated Latin name – GenBank accession number as well as functional annotation. LW, UV, B: long-wave-, ultraviolet-, and blue-sensitive opsin; Blue arrow: Rh7 in Drosophila. Detailed information regarding the 28 homologs is listed in Supplementary Table 3. Only > 50 bootstrap values are shown

Full size image

Spatiotemporal expression patterns

Temporal expression patterns were analyzed with the fragments per kilobase per million mapped fragment (FPKM) values based on the transcriptome sequencing data. According to our results, Nllw was expressed continuously after the hatching of the 1st instar nymph, with relatively stable expression levels in the nymph and female adult stages, but it reached the highest expression level in the male adult stage. For NlUV1/2 and NlUV3-like, the expression patterns varied obviously. The expression levels of NlUV1/2 peaked in the male adult stage, similar to Nllw, but NlUV1 was barely expressed from the egg stage to the 4th instar nymph stage and showed low expression in the female adult stage. NlUV2, on the other hand, exhibited continuous expression from late embryogenesis, a stage when the cuticle of the 1st instar nymph began to form [32]. Besides, NlUV2 was periodically expressed with molting, with high levels being detected at the beginning and the end of each instar. Whereas NlUV3-like showed a similar periodic expression pattern to NlUV2, the expression level of NlUV3-like peaked in the hatching stage rather than in the male adult period (Fig. 3A). Overall, Nllw, NlUV1/2, and NlUV3-like exhibited exceptionally high expression levels in eyes (Fig. 3B), indicating that all the four opsins are involved in visual processing, albeit with the possibility of divergent functions depending on age-related expression patterns.

Fig. 3
figure 3

Spatiotemporal expression patterns of Nllw, NlUV1/2, and NlUV3-like. A. Age expression patterns of Nllw, NlUV1/2, and NlUV3-like with transcript abundances. B. Tissue expression patterns of Nllw, NlUV1/2, and NlUV3-like. I: integument; DT: digestive tract; O: ovary; T: testis; Fb: fat body; E: eye. For A and B, data represent the means of three biological replicates, (means + SD), and raw data are provided in Supplementary Table 5

Full size image

Nllw is associated with melanin-type pigment

The expression levels of Nllw, NlUV1/2, and NlUV3-like were detected in all the RNAi-treated groups (Fig. 4A). The target genes were all effectively silenced by RNAi with the early 4th instar nymph N. lugens, compared with the dsGFP control group. Moreover, the expression levels of the other three opsins were also detected with each RNAi treatment. When Nllw was knocked-down, the expression levels of NlUV1/2 were both significantly up-regulated. Conversely, silencing of either NlUV1/2 resulted in a significant up-regulation of Nllw, suggesting a possible functional complementation between Nllw and NlUV1/2. However, the down-regulation of either NlUV1/2 led to the silencing of the other one, probably attributed to their 77.97% sequence similarity. It is noteworthy that the silencing of NlUV3-like did not impact the expression of Nllw and NlUV1/2. Similarly, the silencing of each of Nllw and NlUV1/2 did not affect the expression of NlUV3-like, indicating that the expression of NlUV3-like is independent of Nllw and NlUV1/2. Among the four opsins in N. lugens, RNAi against each target gene did not lead to an apparent lethal phenotype, and there was no significant difference in survival rate compared with the dsGFP control group (Fig. 4B). However, an unconventional function was observed in the dsNllw-treated group (Fig. 4C). To be specific, gradual accumulation of melanin occurred when Nllw was knocked-down with the 4th instar nymph. When N. lugens developed into the 5th instar, the rigid cuticles of the mesothorax, abdominal dorsal cuticle, leg, wing bud, and head became darker; while the other soft and flexible cuticles, including the abdominal ventral plate as well as the segmacoria, were only slightly affected. When N. lugens further developed into adults, melanin-type pigments occurred on almost the whole body surface, except for soft parts of leg bases of the mesothorax. In addition, the short-hind wings also became darker. While in dsGFP-treated control group, N. lugens usually remained brown, with only slight darkening in male adult, but the darkening degree was far less than the dsNllw-treated group N. lugens, and the hind-wing of adult was light color in general. To avoid the off-target effect of RNAi, the other non-overlapping RNAi region of dsNllw was designed (Fig 1A). RNAi in two different regions featured similar phenotypes, demonstrating that off-targeting did not occur in this experiment.

Fig. 4
figure 4

RNAi against each of Nllw, NlUV1/2, and NlUV3-like. A. Relative expression levels of Nllw, NlUV1/2, and NlUV3-like after RNAi against each target gene, dsGFP was used as the control group. B. The survival rates of N. lugens. For A and B, data represent the means of three biological replicates (means + SD). A two-tailed unpaired t-test was applied. *P < 0.05 and ***P < 0.001 indicate significant differences compared with the dsGFP control group. C. Melanism of N. lugens after knock-down of Nllw. 5th: the 5th instar nymph; FM: female adule; M: male adult. RNAi was performed with the 4th instar nymph. In the dsNllw-treated group, when N. lugens developed into the 5th instar, the rigid cuticle of the mesothorax and abdominal dorsal cuticle, leg, wing bud, and head became darker; when N. lugens eclosed into adult, melanin-type pigments occurred in almost the whole body surface, except for soft parts of the leg bases of the mesothorax. The short-hind wing which was marked with red line and arrow became darker. In the dsGFP-treated control group, N. lugens usually remained brown, with a little darker in male adult, and the hind-wing of adult was light color

Full size image

RNA-seq of the transcriptomes revealed potential downstream genes of Nllw in the melanism pathway

Among the various types of pigmentation in insects, melanin-type pigment is prevalent [33, 34]. RNA-seq analysis after RNAi targeting Nllw revealed a total of 184 differentially expressed genes (DEGs) in 0 h-female adults, comprising 35 down-regulated and 149 up-regulated genes, compared with the dsGFP control group (Supplementary Table 1 A, B). Moreover, Gene Ontology (GO) analysis indicated that 3 genes related to the regulation of the circadian sleep/wake cycle (GO:0042749), which was consistent with the classical function of opsin [21], 3 genes associated with positive regulation of morphogenesis of an epithelium (GO:1,905,332), and 2 genes involved in negative regulation of glucose transmembrane transport (GO:0010829) were down-regulated. In comparison, the up-regulated GO terms were primarily linked to various catabolic processes of organic compounds (Fig. 5A, Supplementary Table 2 A, B).

Fig. 5
figure 5

Analysis of potential downstream genes of Nllw. A. Gene Ontology (GO) analysis after RNAi against Nllw. All of the 3 down-regulated GO terms are marked with a red dotted frame; The others are the top 20 up-regulated GO terms. The result is analyzed with three biological replicates. dsGFP is used as the control group. B. qRT-PCR verification of NlTH and NlaaNAT after RNAi against Nllw. Data represent the means of three biological replicates (means + SD). A two-tailed unpaired t-test was applied. *P < 0.05 and **P < 0.01 indicate significant differences compared with the dsGFP control group

Full size image

Particularly, a tyrosine hydroxylase gene (NlTH) and an arylalkylamine-N-acetyltransferases gene (NlaaNAT) in N. lugens were up- and down-regulated after knockdown of Nllw, respectively. The TH and aaNAT were key genes involved in the tyrosine-mediated melanism pathway [35]. Consistent expression levels of NlTH and NlaaNAT were verified by qRT-PCR with the newly emerging insects after Nllw knockdown (Fig. 5B).

Discussion

As a kind of signal-transducing molecule, opsins play key roles in receiving environmental light signals and subsequently translating them into chemical-free energy [6, 9, 36]. Apart from the typical visual function, Nllw is found to be involved in the melanization of body color in N. lugens via the tyrosine-mediated melanism pathway, which is important for insect pigmentation [37]. In general, the tyrosine-mediated melanism pathway begins with the hydroxylation of tyrosine into 3,4-dihydroxyphenylalanine (DOPA) under the action of TH [38], followed by decarboxylation into dopamine. Afterward, a series of subsequent oxidation and polymerization reactions result in the formation of melanin from DOPA or dopamine [37]. Different from the TH, the aaNAT acts as a melanin-suppressing factor by N-acetylating dopamine to form N-acetyldopamine (NADA), a major precursor for colorless sclerotin [35]. To date, both TH and aaNAT involved in melanization have been extensively studied in several insects, of them, the loss of TH function is suggested to diminish black pigmentation [39, 40]; while the loss of aaNAT function contributes to dark coloration [41, 42]. In this study, RNAi against Nllw up-regulated the expression of NlTH and down-regulated that of NlaaNAT, which suggested that Nllw participated in regulating the changes in melanin precursor substances, resulting in the excessive accumulation of DOPA or dopamine and ultimately the black individuals, as observed in B. mori [43]. Interestingly, such result is consistent with the neuropeptide, Elevenin, as well as its receptor (NlA42) in N. lugens: when either Elevenin or NlA42 is knocked down, the tyrosine-mediated cuticle melanism pathway is activated with up-regulation of NlTH and down-regulation of NlaaNAT, thus inducing melanization, furthermore, the Elevenin was scarcely expressed in integument but extremely highly expressed in brain, while its receptor NlA42 was mainly expressed in both of brain and integument [44]. Similarly, the tissue-expression patterns showed that Nllw was extremely highly expressed in eyes but had a low expression level in integuments, consistent with a previous report indicating that Nllw is expressed in compound and ocelli eyes of N. lugens by in-situ hybridization [14].

A variety of body colors are observed in wild-type N. lugens, with predominant colors being black and light which displays yellow or brown. Furthermore, as suggested in a survey, the dominant body color of N. lugens in Guangzhou, China, varies across different periods [45, 46]. Additionally, under our phytotron conditions, the majority of N. lugens exhibit a brown coloration, with only a minority displaying a black coloration [44]. This observation confirms that body color polyphenism is an important phenomenon, alongside the wing polyphenism in N. lugens [28]. The diversity of body colors enables insects to better adapt to different environments. For instance, the black body color achieves the advantageous purpose of ecological adaptation to climatic conditions under low temperatures [47]. In the meantime, melanization takes part in the innate immune system to resist pathogens and non-self-substances, that are involved in a phenoloxidase [48]. However, previous studies have indicated that distinct photoperiod does not significantly affect the changes in body color composition of N. lugens, and less than half of the black 5th -instar nymphs retain their black coloration after molting [44]. Typically, the phenomenon of melanization in Biston betularia (Lepidoptera) is a well-known example that highlights the crucial role of visual perception in body color change [49]. While in the case of N. lugens, the regulatory mechanism underlying body color may be much more intricate than our current understanding, just like in Locusta migratoria (Orthoptera), the pigment of body color is complex and can be modified by various factors, such as endogenous hormone and population density [50]. Moreover, previous research has demonstrated that both body color and wing morphology of N. lugens contribute to the regulation of development [51]. N. lugens is widely distributed among rice-growing regions in the tropical and temperate Asian Pacific region [52], and there are variations in migration and reproductive preferences among different regional populations of N. lugens [53, 54]. Consequently, it is important to conduct extensive investigations into the relationship between body color and the environment in various regions, to provide further insights into the polyphenism of body color in N. lugens. In a word, it is speculated that Nllw is not only responsible for long-wave spectrum absorption but also plays a crucial role in the complex regulatory network between external signals and body color via the tyrosine signal pathway, with the regulation of key enzymes NlTH and NlaaNAT.

In addition, a predicted NlUV3-like was identified in N. lugens, but the B-sensitive opsin was lacking, which was different from Nephotettix cincticeps, another rice pest in Hemiptera, that possesses classic trichromacy [14]. Notably, the down-regulation of Nllw led to the up-regulation of NlUV1/2, and vice versa, suggesting partial functional compensation between Nllw and NlUV1/2. However, the expression of NlUV3-like was independent of the other three opsins, aligning with the phylogenetic analysis result. Additionally, as revealed by the phylogenetic tree analysis, NlUV1/2 exhibited higher conservation within the Delphacidae family, whereas NlUV3-like diverged from NlUV1/2. Until now, the mechanisms of vision in N. lugens remain unknown, including whether the replicated UV-sensitive opsins can compensate for the missing B-sensitive opsin, like that in beetles [18]. Besides, although the silencing of any one of the Nllw, NlUV1/2, and NlUV3-like did not lead to significant mortality, it did not necessarily imply that it has no impact on the fitness of N. lugens. As crucial components of visual pigments, the changes in preference for spectral sensitivity in N. lugens after the silencing of the four opsins are still unknown. Given that the N. lugens is a kind of long-distance migratory insect with a specialized feeding on rice, the influence of the spectral preference deserves further investigation. For example, a study in monarch has demonstrated that the specific opsin expansions are involved in sensing skylight cues [55]. Additionally, there is also another opsin reported in Ciona intestinalis (Phlebobranchia), which plays an important role in photoresponse during swimming behavior [56].

Conclusions

In summary, this study has highlighted an unconventional role of one opsin (Nllw) in the regulation of cuticle melanization through tyrosine signal pathway, implying a cross-talk between the gene pathways underlying the visual system and the morphological differentiation in insects and a complex mechanism existed in them. Besides, another UV-like opsin (NlUV3-like), which was homolog to the Rh7 in Drosophila was described, and the detailed study of Nllw, NlUV1/2, and NlUV3-like revealed the conservation and evolution of opsins in N. lugens.

Methods

Insect rearing

N. lugens used in this study is reared on the Xiushui rice plant in a phytotron under 26.5℃, 65–75% relative humidity, and a 14/10 h light/dark photoperiod at Ningbo University, Ningbo, China.

Sequence analysis

DNAMAN software was used to predict gene CDS. NCBI blast (https://blast.ncbi.nlm.nih.gov/Blast.cgi) and NCBI CD search (https://www.ncbi.nlm.nih.gov/Structure/cdd/wrpsb.cgi) websites were used to reveal potential gene function and conserved domains. Protein sequences were aligned using Clustal X software. Furthermore, through the NCBI Splign website (https://www.ncbi.nlm.nih.gov/sutils/splign/splign.cgi?textpage=online&level=form), the distribution of gene CDS on N. lugens genomic sequence was displayed.

Phylogenetic analysis

28 amino acid sequences from ten insect species of five orders were downloaded from the NCBI database (Supplementary Table 3). A phylogenetic tree was thereby constructed based on Nllw, NlUV1/2, NlUV3-like, and the above 28 proteins with the software MEGA7, following the maximum likelihood method with 1000 bootstrap replications [57]. Furthermore, the NlaaNAT amino acid sequence was selected as an outgroup member.

Real-time quantitative PCR (qRT-PCR) analysis

The total RNA of the target sample was extracted using the Trizol RNA isolater Total RNA Extraction Reagent (Vazyme Biotech, Nanjing, China) following the manufacturer’s protocol. The quality and concentration of the isolated RNAs were assessed using a NanoDrop One Spectrometer (Thermo Fisher Scientific, USA). Next, 400 ng of each RNA was reverse transcripted into cDNA using the HiScript II Q RT SuperMix (+ gDNA wiper) kit (Vazyme Biotech, Nanjing, China) in line with the protocol. Finally, qRT-PCR was carried out with an ABI QuantStudio 5 equipment (Thermo Fisher Scientific, USA) in a 10-µL reaction system, containing 5µL of ChamQ SYBR Color qPCR Master Mix (Vazyme Biotech, Nanjing, China), 2.4µL of ddH2O, 2µL of 20-fold diluted cDNA, and 0.3µL of each 10 µmol primer. The procedure of qPCR comprised an initial denaturation step at 95 °C for 5 min, followed by 40 cycles of denaturation at 95 °C for 10 s, and primer annealing together with amplification at 60 °C for 30 s. With 2−ΔΔCt (Ct: cycle threshold) method [58], the relative expression level of the target gene was calculated, with N. lugens 18 S ribosomal RNA (Nl18S, GenBank accession number: JN662398.1) serving as the housekeeping gene. Primers used for qRT-PCR are listed in (Supplementary Table 4 A).

Spatiotemporal expression analysis

Temporal expression patterns of Nllw, NlUV1/2, and NlUV3-like were analyzed with the FPKM values. Data from three biological replicates of each sample were used (Supplementary Table 5 A). By dissecting different tissues containing integument, digestive tract, ovary, testis, fat body, and eye from the 5th instar nymphs and adults, the Nllw, NlUV1/2, and NlUV3-like expression patterns in tissues were analyzed by qRT-PCR. Data from three biological replicates are listed in (Supplementary Table 5B).

RNA interference (RNAi)

One specific sequence of the target gene was cloned to synthesize the double-stranded RNA (dsRNA) with a MEGAscript T7 Transcription Kit (Ambion, Austin, TX), following the manufacturer’s protocol. The green fluorescent protein (GFP) gene from Aequorea victoria (Leptothecata) was used as a negative control. Primers containing the T7 promoter sequences used for cloning are listed in (Supplementary Table 4B). With a FemtoJet Microinjection system (Eppendorf, North America), around 250 ng of dsRNA was injected into each of the 4th instar nymphs. The gene silencing efficiency was detected at 72 h post-injection by randomly selecting five insects from each sample. The insect survival rate after RNAi was counted for 10 days and the insect phenotypes were recorded using a stereomicroscope (Leica 151 S8AP0, Germany).

Transcriptome sequencing (RNA-seq)

Firstly, dsRNA of Nllw was injected into the 3rd instar N. lugens nymphs, and female adult N. lugens was collected at 0 h after emergence since the body color difference was more pronounced than male adults after RNAi. Afterward, ten individuals were homogenized for total RNA extraction, and dsGFP-treated sample was taken as the negative control. Each treatment was replicated thrice for cDNA library preparation and sequencing on the Illumina Novaseq 6000 platform (Novogene, Tianjin, China). Clean reads were aligned to the N. lugens reference genome using HISAT2. Furthermore, transcripts per million (TPM) expression values were calculated based on feature counts for genes by Cufflink. DEGs were identified using the DESeq2 package upon the threshold of P-value < 0.05 and log2 ratio > 1. To explore the biological progress of DEGs, GO analysis was conducted with eggNOG, and GO terms were classified as biological process (BP), cellular component (CC), and molecular function (MF) categories.

Statistical analysis

The data are presented as means + SD in graphic instructions, and three biological repetitions were performed for each experiment. Statistical analysis was performed using GraphPad Prism 7.00 software and Mircosoft Excel. The two-tailed Student’s t-test was conducted to compare the difference. The significance level was set at * p < 0.05, ** p < 0.01, and *** p < 0.001.