Characterization of NlSP7
Using the proteomics database established for BPH by our laboratory (Gong et al. 2019), we obtained a homologous fragment (312 base pairs) based on alignments with known insect transcriptome sequences. The open reading frame of the sequence was amplified via PCR with gene-specific primers. The complete sequence (678 base pairs) of the gene was obtained by RACE, and it comprised a 242-bp 5′ untranslated region, 98-amino acid open reading frame, and 82-base pair 3′ untranslated region (Fig. 1A). The predicted protein comprised 117 amino acid residues with a theoretical molecular mass of 12.88 kDa (Supplementary Fig. S1 and S2). We designated this sequence as NlSP7 (GenBank accession no. MF278695.1).
To investigate the spatio-temporal expression pattern of NlSP7 in 13 different developmental stages (1st, 2nd, 3rd, 4th and 5th instar nymphs, 1st, 3rd, 5th, 7th, 9th, 11th, 13th, 15th instar adults) and various parts of the adults (head, salivary glands, ovaries, midgut, leg, and fat body) of BPH. RT-qPCR showed that NlSP7 transcripts were expressed in all nymph and adult stages (Fig. 1B). The significantly higher expression in adults suggests that NlSP7 plays its most important role in the adult stage. Tissue-specific expression analysis in adults showed that the NlSP7 transcript levels were slightly higher in the ovaries, followed by the fat body, midgut, and salivary gland (Fig. 1C).
NlSP7 expression activated by tricin in different rice varieties
Tricin concentrations in the different rice varieties and organs were different (Fig. 2). Thetricin identification results for the TN1, Mudgo, IR26, ASD7, IR36, IR56, and RH varieties were shown in Fig. 2A. The tricin concentrations in the leaves were significantly higher than those in the stems and roots in both the resistant RH and susceptible TN1 varieties. There were no significant differences in the tricin concentrations between the stems and roots (Fig. 2B). The tricin concentration in the leaves was 28.1 ± 2.6 ng/g in the RH plant, which was significantly higher than that in the leaves of the TN1 plant (P = 0.002). The tricin concentration in the RH roots differed significantly from that in the TN1 roots (P = 0.004).
First, the optimal concentration of tricin had been chosed from two different tricin concentrations (50 mg/L and 100 mg/L) (Supplementary Fig. S6). The severity scores differed significantly for the tricin-resistant RH population of BPHs and tricin-sensitive TN1 population of BPHs. The mean severity scores were significantly lower for the tricin-sensitive population (P < 0.001) than the tricin-resistant population when feeding on the RH rice variety (control in this study, severity score = 9.0) (Fig. 2C). The NlSP7 gene mRNA expression level was higher in the tricin-resistant BPHs than in the tricin-sensitive population (Fig. 2C). These results suggest that NlSP7 may have a role in BPH feeding and virulence. BPH nymphs fed an artificial diet containing tricin had higher NlSP7 transcript levels at 12 h and 24 h, compared with those fed a diet without tricin (P < 0.05, P < 0.01, respectively) (Fig. 2D). Thus, the NlSP7 mRNA levels were upregulated after ingesting an artificial diet containing tricin.
Fitness of BPH
To elucidate the role of NlSP7 in BPHs, we synthesized dsRNA from NlSP7 and injected it into 1-day-old female adults to mediate RNAi (Supplementary Fig. S3). This treatment had strong silencing effects, reducing the NlSP7 transcript level significantly (~ 86%) in the first day after treatment, compared with the levels in the control group, in those injected with dsGFP RNA (P < 0.01), and in those injected with dsNlSP7 (P = 0.001 after injection with dsGFP or dsNlSP7 at 2 days, P < 0.001 after injection with dsGFP or dsNlSP7 at 3 days).
BPHs subjected to the NlSP7 RNAi treatment also exhibited less feeding activity, as indicated by significantly lower excretions of honeydew, compared with the two control groups (P < 0.001 for control and dsNlSP7; P = 0.003 for dsGFP and dsNlSP7), as shown in Fig. 3A. The NlSP7-RNAi BPHs also had lower weight gain values, regardless of whether they were fed RH rice plants (P < 0.001 for control and dsNlSP7, P = 0.001 for dsGFP and dsNlSP7) or an artificial diet (Fig. 3B), and smaller weight gain ratios (P = 0.006 for the control and dsNlSP7, P = 0.003 for dsGFP and dsNlSP7), as shown in Fig. 3C. Furthermore, silencing NlSP7 reduced the virulence of BPH (Fig. 3D). Compared with the two control groups, BPHs injected with dsNlSP7 had significantly lower survival rates from 3 to 11 days after microinjection (P = 0.003, P = 0.003, respectively), and most died on RH rice by 9 days after microinjection (Fig. 3E). Similar results were obtained when BPHs injected with dsNlSP7 were fed an artificial diet (100mg/L) (Fig. 3F). The BPHs injected with dsNlSP7 fed on TN1 rice and artificial diet without tricin, the mortality rate was significantly was significantly reduced (Supplementary Fig. S4 and S5).These results indicate that silencing the NlSP7 gene significantly reduces feeding and performance of BPHs on rice plants.
Impairment of BPH feeding
To determine the effect of NlSP7 on BPH feeding, we used the EPG technique to profile the feeding behavior of piercing–sucking insects (Yuan et al. 2020). Five main feeding phases can be distinguished by EPG: 1) the non-penetration phase, 2) the pathway phase (including penetration, salivation, stylet movement, and extracellular activity near the phloem), 3) the intracellular phase of activity in the phloem, 4) the phloem sap ingestion phase, and 5) the xylem phase. The top of Fig. 4A shows the representative EPG traces obtained from BPHs, showing the different phases. As shown in Fig. 4B, after NlSP7 knockdown in female adults, the EPG waveforms obtained over 6 h on TN1 rice plants showed that the duration of intracellular activity in the phloem exceeded those of the dsGFP and control groups (P = 0.001), and the durations of the non-penetration, pathway, xylem, and phloem sap ingestion phases decreased (P < 0.001, P = 0.011, P = 0.001, P = 0.007, respectively). On RH rice plants, the duration of the non-penetration phase increased significantly (P < 0.001), whereas the durations of the intracellular activity in the phloem and phloem sap ingestion phases decreased significantly over the 6-h period (P = 0.017, P < 0.001, respectively), as shown in Fig. 4C. These findings indicate that BPH individuals spent less time feeding on rice plants with high tricin levels after NlSP7 knockdown.
Furthermore, BPHs fed an artificial diet high in tricin produced significantly more salivary flanges than those in the control group and those fed an artificial diet containing low tricin levels (P < 0.001) (Fig. 5A–5C). In addition, the number of salivary flanges differed significantly between the two rice plant varieties. BPHs produced 71.43% more salivary flanges on RH plants than on TN1 plants after NlSP7 knockdown (Fig. 6A and 6B). Moreover, comparing salivary flanges among the control group, the numbers of salivary flanges were sparser per unit area and deeper on TN1 plants than on RH plants (Fig. 6C and 6D).
Tricin metabolism mediated by NlSP7
To determine whether the salivary protein NlSP7 influences the production of tricin in rice plants, we investigated tricin levels in rice plants infested by BPH adults, where the ability to produce NlSP7 was silenced or not silenced. The results showed that plants fed on by control and dsGFP groups BPH had decreased tricin levels than BPH group which injected by dsNLSP7 and plants fed on by BPH adults had decreased tricin levels compared to plants not fed on by BPHs (Fig. 7A–7C). After NlSP7 knockdown, the tricin level in RH rice plants increased by 86.36% (P = 0.005) but remained unchanged in TN1 rice plants. To further characterize the physiological properties associated with the blocked synthesis of flavonoids induced by NlSP7, we examined the expression levels of genes responsible for the synthesis of flavonoids in rice plants. Our results indicated that the level of tricin induced by NlSP7 shared similarities with the secondary metabolite synthetase induced by the flavonoid pathway marker genes CHS and CHI. Thus, NlSP7 may play important roles in defense-related signal transduction (Yang et al. 2001).