Nilaparvata lugens was collected from a rice field at the Plant Protection Station of Jiangpu County (Jiangsu, China). They were reared on Taichung Native 1 (TN1) rice seedlings in the laboratory. The rearing conditions were 27 ± 1°C, with 70 ± 10% relative humidity and a 16 h:8 h (Light:Dark) photoperiod.

The amino acid sequence of D. melanogaster TMC protein was used to screen against N. lugens genomic and transcriptomic databases for identification of its homologs in N. lugens. Open reading frames (ORFs) were predicted with EditSeq (version 5.02, DNAstar, Madison, WI, United States).

Total RNA was isolated from whole insects using the TRIzol Reagent (Invitrogen, Carlsbad, CA, United States) following the manufacturer’s protocol. Residual genomic DNA was removed by RQ1 RNase-Free DNase (Promega). Single-stranded cDNA was synthesized from the total RNA with M-MLV reverse transcriptase and oligo (dT)₁₈ (BioTeke, Beijing, China). The forward primer Nltmc-comp-F and the reverse primer Nltmc-comp-R were used to amplify the full-length or fragment gene by means of PCR on cDNA from adult N. lugens using TransTaq HiFi DNA Polymerase (TransGen Biotech, Beijing, China). The purified PCR products were sub-cloned into pGEM^®-T easy vector (Promega, Madison, WI) and then sequenced using the 3730 XL DNA analyzer (Applied Biosystems, Carlsbad, CA, United States. The primers corresponding to each gene are listed in Table 1.

The exon and intron architectures of Nltmc genes were predicted based on the alignments of putative cDNA against their corresponding genomic sequences in Spidey¹, and then structured on the website of GSDS v2.0² (Hu et al., 2015). The transmembrane segments and topology of NlTMC proteins were predicted by TMHMM v2.0³. Multiple alignments of the complete amino acid sequences were performed with Clustal Omega⁴. Phylogenetic tree was constructed using MEGA 5.2.2 software with the Maximum Likelihood method and bootstrapped with 1,000 replications. The branch support values are expressed as percentages. The accession numbers of the sequences used in the phylogenetic analysis are listed in Table 2.

Developmental stage samples were collected from eggs (n = 100∼120), first-instar (n = 80), second-instar (n = 60), third-instar (n = 230), fourth-instar (n = 15∼20), fifth-instar nymphs (n = 15), and adults of both sexes and wing forms: brachypterous female (BF), macropterous female (MF), brachypterous male (BM), and macropterous male adults (MM) (n = 10). Eggs were collected at 4 days (central development stage) after the oviposition since the egg stage is 6–7 days. Nymphs were collected every 24 h from the beginning of each instar until molting and all adults were collected 4 days after eclosion.

Different tissue samples including head, wing, gut, Malpighian tubule (MT), female reproductive organ (FRO), ovary (OA), oviduct (OU), copulatory pouch (CP), and spermatheca (SE) were dissected from brachypterous female adults collected 4 days after eclosion. The first-strand cDNA was synthesized with HiScript^® II Q RT SuperMix for qPCR (+ gDNA wiper) kit (Vazyme, Nanjing, China) using an oligo(dT)₁₈ primer and 500 ng total RNA template in a 10 μl reaction, following the instructions.

Real-time qPCRs were employed to investigate relative expression of Nltmc genes in the various samples using the UltraSYBR Mixture (with ROX) Kit (CWBIO, Beijing, China). The PCR was performed in 20 μl reaction including 4 μl of 10-fold diluted cDNA, 1 μl of each primer (10 μM), 10 μl 2× UltraSYBR Mixture, and 6 μl RNase-free water. The standard two-step PCR cycle conditions were as follows: 95°C for 10 min, and then 40 cycles of amplification consisting of 95°C for 15 s, 60°C for 40 s, followed by melting curve analysis. Pairs of gene-specific primers used for real-time qPCR were designed using the Primer Premier 5 Software (Table 1). The relative quantification of Nltmc was calculated according to the comparative 2^–ΔΔCT method (Livak and Schmittgen, 2001).

The fragment coding sequence of Nltmc genes and green fluorescent protein (gfp) were amplified by PCR using specific primers conjugated with the T7 RNA polymerase (Table 1). PCR-generated DNA templates were then used to synthesize dsRNA, which contains T7 promoter sequences at each end. We used a MEGAscript T7 transcription kit (Ambion, Austin, TX, United States) to produce the specific dsRNA of each gene as the manufacturer’s instruction. The quality and size of the dsRNA products were verified by 1% agarose gel electrophoresis. Thirty fully mated female adults (collected 4 days after eclosion) were injected with approximately 50 nl of purified dsRNA (5,000 ng/μl) via mesothorax and were reared with rice seedlings. A set of 6–10 insects at 3 days after injection was selected to verify dsRNA knockdown efficiency by qRT-PCR. The remaining individuals were used for observations of eggs laid and female survival. Four to six biological replications were performed.

For egg-laying assay, RNAi injected females (fully mated) were transferred vials with fresh rice seedlings. Number of laid eggs were counted under a stereomicroscope (Zeiss) after 3 days. The female survival was recorded 8 days after injection of dsRNA. At least four to six vials per treatment were observed. Ovaries of mated females were prepared under the stereomicroscope (Zeiss). Pictures of ovaries were taken using a light microscope with a digital video camera (Zeiss, ProgRes 3008 mF, Jenoptik, Jena, Germany). The number of eggs per ovary were measured and counted.

Experimental data was analyzed using GraphPad Prism 6 software (GraphPad Software Inc., San Diego, CA, United States). The two-tailed unpaired Student’s t-test or one-way analysis of variance (ANOVA) of Duncan’s multi-range test were used to test the differences between two or more than two normal distribution data.

We identified three tmc genes in the genome and transcriptome database of N. lugens. These three tmc genes were cloned by PCR and then confirmed by DNA sequencing (Table 1). One fragment and two full-lengths of different cDNA clones were obtained. These sequences were designated Nltmc3 (GenBank accession number: MT576068), Nltmc5 (MT576067), and Nltmc7 (MT576069) according to their similarity to other invertebrate and vertebrate tmc genes (Keresztes et al., 2003; Kurima et al., 2003).

We cloned the fragment of Nltmc3 gene that consists of 4,476-bp cDNA encoding 1,492 amino acids. We tried to clone the full-length of this gene using 5′-RACE and 3′-RACE technology. Unfortunately, we did not get the positive results. We then cloned the full-length of Nltmc5 and Nltmc7 gene. The complete ORF of Nltmc5 and Nltmc7 encodes 692 and 780 amino acids, respectively. Exon-intron organization was analyzed by comparing cloned cDNAs and the corresponding genomic sequence, revealing that Nltmc3 is located on scaffold 754 and scaffold 3202, Nltmc5 is located on scaffold 943, and Nltmc7 is located on scaffold 2298 (Figure 1). TMHMM2.0 strongly predicts the presence of eight or nine transmembrane-spanning domains in each of the TMC proteins. They all encode a conserved TMC domain that share the completely conserved amino acid triplet C (cysteine) – W (tryptophan) – E (glutamic acid), predicted to be located on the extracellular loop upstream of TM6 (Figures 2–4) (Keresztes et al., 2003). Amino-acid sequence comparisons between the NlTMC proteins and other species TMC proteins show high overall amino acid similarity at the transmembrane region and TMC domain (Figures 2–4). The encoded protein of NlTMC3, similar with CeTMC1, CeTMC2, and DmTMC, has large ORFs. BLASTP analyses of protein sequence alignment showed that NlTMC3 had 60, 66, and 66%, sequence similarity with the TMC proteins of Drosophila melanogaster and C. elegans. Interestingly, we found two internal repeats between TM5 and TM6 in the Nltmc3 gene (Figure 2). In mammals, eight TMC proteins can be grouped into three subfamilies A, B, and C, based on sequence homology (Keresztes et al., 2003). Phylogenetic tree comparison showed that NlTMC3 clustered with MmTMC1, MmTMC2, MmMTC3, CeTMC1, CeTMC2, and DmTMC, which belongs to A subfamily. NlTMC5 is assembled in a group that contains MmTMC5 and MmTMC6, which belongs to B subfamily. And NlTMC7 clustered with MmTMC7 and HsTMC7 that belongs to C subfamily (Figure 5).

The relative expression level of three Nltmc genes in different developmental stages and tissues were measured by qPCR (Figure 6). The results showed that the expression levels of the Nltmc genes varied between the developmental stages including egg, 1st–5th instar nymph, and 4-day old adults (MM, MF, BM, and BF). Among them, Nltmc3 and Nltmc7 were highly expressed in the nymphs compared with other developmental stages. Nltmc5 was more highly expressed in MM and MF than BM and BF adults indicated that Nltmc5 might involve in the wing polymorphism in BPH (Figures 6A,D,G).

We further investigated the relative expression level of three Nltmc genes in various female adult tissues, including the head, wing, gut, Malpighian tubules (MT) and FRO using qPCR method. In the examined tissues, all Nltmc genes were mostly expressed in reproductive organs compared with other tissues (Figures 6B,E,H). This indicated that these genes could be involved in reproduction in the BPH. We next examined the expression pattern of three Nltmc genes within the FRO (Figures 6C,F,I). Interestingly, Nltmc3 was highly expressed in the oviduct (OU). While, Nltmc5 was the most expressed in the ovary (OA). And Nltmc7 was almost expressed equally in the four examined tissues (Figures 6C,F,I).

Next, we tested whether Nltmc genes are involved in the egg laying of N. lugens. Using RNAi technology, we silenced all of the Nltmc genes in the N. lugens (Figures 7A–C). The dsRNA-injection did not negatively affect the survival of N. lugens (Figure 7D). However, the dsNltmc3-injected planthoppers showed the decreased eggs (Figure 7E). While silencing Nltmc5 and Nltmc7 had little impact on the egg-laying rate of BPH (Figure 7E).

Next, we investigated the underlying mechanism of Nltmc3 involved in the egg laying of N. lugens. We did not observe any developmental defects of BPH after Nltmc3 gene silencing (data not shown). We then examined ovary development in females at 7-day-adulthood (3-day after injection of dsRNA). For dsgfp-injected females, ovaries were fully developed (Figure 7F). By contrast, ovaries of dsNltmc3-injected females were small and poorly developed (Figure 7F). In the Nltmc3-RNAi planthoppers, we observed less detained eggs per ovary (Figure 7G). However, silencing Nltmc5 and Nltmc7 has no impact on the ovarian development in the BPH (Figures 7F,G).