Introduction

During plant virus transmission by piercing-sucking insects, arboviruses are inoculated into the plant phloem via the secreted insect’s saliva (Arca & Ribeiro, 2018; Conway et al., 2016; Wu, Yi, Lv, Mao, & Wei, 2022). Thus, insect saliva acts as an interface for the virus-insect-host tripartite interaction and it can directly promote pathogen transmission to, and infection of, the hosts (Sun et al., 2020; Wu et al., 2022). However, despite the importance of insect salivary proteins in the tripartite interaction, there is still much to learn about how these proteins enable successful viral infection.

Previous studies have revealed that there are two ways in which insect saliva facilitates viral infection (Arca & Ribeiro, 2018). One is an indirect approach whereby the saliva modulates the host microenvironment at the feeding site, and saliva effectors work together to allow the arthropod to go unnoticed while it feeds on the host plant (Acevedo et al., 2019; Arca & Ribeiro, 2018; Sun et al., 2020). The other is direct regulation, whereby saliva proteins promote virus transmission through specific molecular interactions (Wen et al., 2019). Our recent work indicated that a Laodelphax striatellus mucin protein, LssaMP, enables the formation of the salivary sheath and facilitates the transmission of rice stripe virus (RSV) into the rice phloem (Huo et al., 2022). A recent study on leafhoppers revealed that the expression of a saliva calcium-binding protein is inhibited by rice gall dwarf virus (RGDV), thus causing an increase of cytosolic Ca²⁺ levels in rice and triggering callose deposition and H₂O₂ production. This increases the frequency of insect probing, thereby enhancing viral horizontal transmission into the rice phloem (Wu et al., 2022). Direct saliva protein-pathogen interactions have been reported in animal pathogens. For example, during transmission of Borrelia burgdorferi by Ixodes scapularis, the saliva protein Salp15 of I. scapularis binds to the bacterial outer surface protein C, which prevents the bacterium from being recognized by the animal immune system. In this way, the saliva protein enables the pathogen to infect the animal host (Ramamoorthi et al., 2005; Schuijt et al., 2008). The Anopheles gambiae saliva sporozoite-associated protein (AgSAP) binds to the surface of the Plasmodium berghei sporozoite. The binding of AgSAP to heparan sulfate inhibits local inflammatory responses, thereby facilitating early Plasmodium infection in the vertebrate host. Although most plant viruses are completely dependent on insect vectors for plant-to-plant transmission (Gray, 2008), the direct function of insect saliva proteins in mediating virus transmission remains largely uninvestigated.

During sap-feeding, arthropods produce two distinct types of saliva at different stages of the feeding process; gel saliva and watery saliva (Bonaventure, 2012; Lou et al., 2019). The former forms a salivary sheath to provide a smooth path for the stylet penetration (Lou et al., 2019). The latter is secreted into the phloem sieve elements to prevent them from plugging up, suppresses plant defense responses, induces pattern-triggered immunity (PTI) through herbivore-associated molecular patterns (HAMPs), and induces effector-triggered immunity (ETI) through avirulent gene, etc. (Huang, Liu, Xu, Bao, & Zhang, 2017; Ji et al., 2017; Yi, Wu, & Wei, 2021).

For phloem-feeding insects, callose deposited on phloem sieve tubes can function as a defense mechanism by reducing insect feeding and preventing long-distance viral movement (Hao et al., 2008; Hipper, Brault, Ziegler-Graff, & Revers, 2013; Will & Vilcinskas, 2015; Zavaliev, Ueki, Epel, & Citovsky, 2011). Callose is a β-(1,3)-D-glucan polysaccharide that is synthesized by callose synthases and degraded by β-(1,3)-glucanases. Plants can defend themselves by depositing callose at the sieve tubes in response to virus infection, whereas viruses can counter this by activating β-(1,3)-glucanases to degrade callose and unplug the sieve tube occlusions (Bucher et al., 2001; Hao et al., 2008; Wu et al., 2022; Zavaliev et al., 2011).

RSV is the causative agent of rice stripe disease, a serious disease of rice crops that has occurred repeatedly in China, Japan, and Korea (Xu, Fu, Tao, & Zhou, 2021). RSV is completely dependent on insect vectors for transmission to its host plants, and L. striatellus is the main vector (Xu et al., 2021; Zhao et al., 2018). L. striatellus transmits RSV in a persistent-propagative manner. The virus initially infects the midgut, then disperses from the hemolymph into the salivary glands, and is inoculated into the plant host during L. striatellus feeding (Huo et al., 2022). L. striatellus belongs to the order Hemiptera, whose members mainly feed from sieve tubes through their mouthparts (stylets) that penetrate plant tissues and reach sieve tubes to ingest the phloem sap (Tjallingii, 2006; van Bel & Will, 2016). RSV is secreted into the rice phloem via the watery saliva (Huo et al., 2022; Wang & Blanc, 2021).

In this study, we identified a molecular interaction among RSV, an L. striatellus saliva protein, and a plant β-1,3-glucanase. The insect saliva protein directly binds to the RSV nucleocapsid protein (NP) and then binds to a rice thaumatin-like protein to activate its β-1,3-glucanase activity. The activation of β-1,3-glucanase helps facilitate RSV infection of the host plant by inhibiting callose deposition in response to viral infection.

Results

L. striatellus salivary carbonic anhydrase is an essential factor in enhancing RSV infection

To investigate the influence of L. striatellus salivary proteins on RSV infection levels in rice plants, two experiments were conducted and compared: viruliferous insect feeding (VIF) and microinjection of purified virus (MPV) (Fig. 1A). The results showed that VIF caused high RSV infection levels and obvious rice stripe disease symptoms, while MPV did not lead to disease symptoms, and only a low level of RSV NP protein was detected inside the plant (Fig. 1B-E). These results suggested that the insect saliva factors are necessary for enabling RSV infection.

Figure 1.

To identify the salivary proteins that play roles in mediating RSV infection, we have sequenced the L. striatellus saliva proteins and focused on the watery proteins that were inoculated into the plant phloem sieve tubes together with RSV (Huo et al., 2022). A salivary protein, Sap6 (GenBank accession No. RZF48846.1), was identified to have a high expression level in L. striatellus salivary glands (Fig. 2A). Western blotting assay showed that Sap6 was secreted into the plant as a watery saliva protein (Fig. 2B-C). Sequence homology analysis revealed Sap6 as a carbonic anhydrase-like protein, and enzyme active site analysis and enzyme activity experiment indicated Sap6 as a carbonic anhydrase (Fig. 2D-E). Sap6 was then named L. striatellus salivary carbonic anhydrase (LssaCA).

Figure 2.

To determine the role of LssaCA in RSV infection of rice plants, a LssaCA-specific double-stranded RNA (dsLssaCA) was synthesized and delivered into the hemocoel of RSV-infected 3^rd instar nymphs to interfere with the gene expression. At 2 days post-microinjection (dpi), the LssaCA mRNA levels were measured by RT-qPCR and compared with the dsGFP control. The results showed that dsLssaCA treatment reduced the LssaCA mRNA level by 80% (Fig. 3A). The two groups of insects were then allowed to feed on healthy rice seedlings (5 insects per rice seedling) for 2 days. The RSV NP RNA levels (indicative of RSV titers) in the plants were measured at 14 days post-feeding (dpf). Compared with the rice seedlings fed by dsGFP-treated insects, those fed by dsLssaCA-treated insects had significantly reduced RSV titers (Fig. 3B), indicating that LssaCA played an essential role in mediating RSV infection of the rice plants.

Figure 3.

LssaCA enhances RSV infection not before the viruses have been inoculated into the plant phloem

To determine at which step LssaCA-deficiency affected RSV infection, we analyzed the influence of LssaCA-deficiency on RSV titers during several steps of the RSV transmission process, including the virus load in the salivary glands and saliva, and the initial inoculation levels in planta. First, at 2 days after dsLssaCA treatment of the viruliferous insects when LssaCA was significantly downregulated, RT-qPCR was used to detect RSV titer in the insect salivary glands and the results showed that LssaCA deficiency did not reduce RSV titers in this tissue (Fig. 3C). Then, the LssaCA-deficient insects were allowed to feed on an artificial diet and their saliva was collected for RSV level analysis, which revealed that LssaCA-deficiency did not reduce the RSV titer in the secreted saliva (Fig. 3D). Finally, plants were fed with 10 viruliferous insects for 24 h respectively and the RSV titers were measured immediately after feeding, which showed that LssaCA-deficiency did not affect the RSV initial inoculation levels in planta (Fig. 3E). These results suggested that LssaCA promoted RSV infection by a mechanism occurred not in insects or early stage of viral entry in plants, but in planta after viral inoculation.

As shown in Fig. 1, VIF but not MPV is able to enable RSV infection of rice plants, suggesting the necessity of insect saliva factors in enabling RSV infection. To further confirm whether LssaCA played a role in directly enhancing RSV levels in planta, we produced a recombinant LssaCA protein to pre-incubate with purified RSV particle solution and used bovine serum albumin (BSA) as a control for comparison. The two groups were then microinjected into leaves on different plants, and at 14 dpi their viral titers were measured using NP gene-specific RT-qPCR. The results showed that compared to those inoculated with BSA plus RSVs particles, those inoculated with LssaCA plus these same particles exhibited significantly higher RSV titer values (Fig. 3F-G). These findings clearly indicate that LssaCA plays an important role in promoting RSV infection of plants after RSV has been inoculated into the plant.

LssaCA interacts with rice thaumatin-like protein to increase its endo-β-1,3-glucanase activity

Because LssaCA played its role in planta, we performed two experiments to investigate the mechanisms by which LssaCA enhances RSV infection. The first was to create LssaCA mutant proteins lacking enzymatic activity to determine if the enhanced infection was related to the LssaCA enzymatic activity. The second was to conduct a yeast two-hybrid screening to identify rice proteins that interact with LssaCA.

LssaCA has 7 residues constituting the enzymatic active center, including H₁₁₁, N₁₃₉, H₁₄₁, H₁₄₃, E₁₅₃, H₁₆₆, and T₂₅₃. Seven recombinant mutant proteins were constructed and were labeled as LssaCA^H111D, LssaCA^N139H, LssaCA^H141D, LssaCA^H143D, LssaCA^E153H, LssaCA^H166D, LssaCA^T253E (Fig. S2A). Enzymatic activity assays indicated that five of the seven mutants had significantly reduced enzymatic activity (Fig. S2B). To measure the influence of these mutants on RSV infection, they were pre-cultured with purified RSV particle solution and were then microinjected into plant leaves. The RSV titers were measured using RT-qPCR at 14 dpi. The results showed that, compared to LssaCA, none of the mutants decreased the protein function in enhancing RSV infection (Fig. S2C). These results suggest that LssaCA enables RSV infection in an enzymatic activity-independent manner.

The yeast two-hybrid screening identified a rice thaumatin-like protein (XP_025879846.1), designated as Oryza sativa thaumatin-like protein (OsTLP) (Fig. 4A-B and S3). The open reading frame of OsTLP encoded for a polypeptide of 327 amino acids and a thaumatin family (THM) domain from residues 27 to 267. A BLAST algorithm-based analysis against sequence databases revealed the sequence identities to TLP 1b. The N-terminal 24 amino acids constituted a secretion signal peptide, suggesting OsTLP as a secreted protein. Pull-down and MST assays were performed to confirm the molecular interaction between OsTLP and LssaCA. The pull-down assays showed that OsTLP and LssaCA had a specific interaction (Fig. 4C). The MST assay revealed a K_D of 1.3 ± 2.2 μM for the OsTLP-LssaCA interaction, further confirming the molecular interaction (Fig. 4D).

Figure 4.

Previous studies have reported that TLP orthologs in apples, cherries, tomatoes, barley, etc. have endo-β-1,3-glucanase activity, which works on degrading callose (de Jesus-Pires et al., 2020). To confirm the endo-β-1,3-glucanase activity of OsTLP, an enzyme activity experiment was conducted with the recombinantly expressed OsTLP protein. The results revealed an average endo-β-1,3-glucanase activity of 60 units/mg OsTLP (Fig. 4E). To determine the in-planta activity of OsTLP, OsTLP overexpression transgenic plants were constructed. Enzymatic activity assays confirmed that the transgenic plants had significantly increased β-1,3-glucanase activity (Fig. 4F).

To investigate the influence of the molecular interaction between LssaCA and OsTLP on OsTLP activity, OsTLP was pre-incubated with either LssaCA or a BSA control for 2 h before the enzymatic activity was measured. The results showed that pre-incubation with LssaCA significantly increased OsTLP activity, compared to the BSA control (Fig. 4G). These results imply that LssaCA may facilitate RSV infection by enhancing the activity of OsTLP to degrade callose.

LssaCA inhibits callose deposition induced by RSV infection

Callose deposition on sieve plates is an important mechanism for slowing virus long-distance transmission and reducing the virus infection level in plants (Li et al., 2012; Zavaliev et al., 2011). To investigate whether RSV infection induces callose deposition, rice seedlings were fed by RSV-free or viruliferous L. striatellus for 24 h, and then callose deposition of the plants was measured. The results showed that VIF led to more callose deposition compared with that induced by feeding with RSV-free L. striatellus (Fig. 5A-B). ELISA assays also revealed an increased callose concentration in the plant fed by viruliferous L. striatellus (Fig. 5C). To further confirm the role of RSV in inducing callose deposition, the transcript levels of callose synthase genes were measured using RT-qPCR. We found that RSV infection upregulated gsl3, gsl4, gsl5, and gsl10 (Fig. 5D–G). These results demonstrate that RSV infection induces callose deposition.

Figure 5.

To determine if LssaCA influences callose deposition induced by RSV infection, we downregulated LssaCA in viruliferous L. striatellus by dsRNA microinjection. On 2 dpi when LssaCA expression was significantly reduced, the insects were allowed to feed on healthy plants for 24 h, and then the callose deposition was measured. The results showed that the plants fed by LssaCA-deficient insects deposited more callose compared to the plants fed by the control-group insects (dsGFP treatment) (Fig. 5H–J), indicating that LssaCA can impair the callose caused by RSV infection.

OsTLP degrades callose to facilitate RSV infection

To confirm if OsTLP degrades the RSV-induced callose in planta, viruliferous L. striatellus were allowed to feed on either OsTLP overexpression plants or control plants, and then callose deposition was measured. The results showed that transgenic plants exhibited significantly lower callose deposition compared to wild-type ones (Fig. 6A-B), indicating that OsTLP can degrade the deposit callose induced by RSV infection. Moreover, we measured the RSV infection levels that were influenced by OsTLP overexpression. Viruliferous L. striatellus fed on either OsTLP overexpression plants or control plants for 2 days, and then the RSV titers were measured on 14 dpf. The RT-qPCR analyses indicated that, compared with wild-type plants, plants overexpressing OsTLP had significantly higher RSV titers (Fig. 6C), indicating that OsTLP facilitates RSV infection in planta. It is hypothesized that the overexpressed OsTLP degrades the deposited callose that is caused by RSV infection, therefore facilitating RSV infection.

Figure 6.

RSV NP interacts with LssaCA to further enhance OsTLP enzymatic activity

We have confirmed that LssaCA binds to OsTLP to promote callose degradation and facilitates RSV infection in planta, we then investigated whether there is a direct interaction between LssaCA and RSV by which RSV directly participates in the function played by LssaCA. By using RSV and LssaCA-specific antibodies to conduct immunofluorescence assays (IFA), we found that LssaCA and RSV co-localized in salivary glands (Fig. 7A), suggesting a potential molecular interaction between them. A pull-down assay then revealed a specific interaction between LssaCA and RSV NP (Fig. 7B). An MST assay determined that the K_D of the interaction between LssaCA and NP was 2.7 ± 2.2 μM, further confirming the molecular interaction (Fig. 7C).

Figure 7.

As LssaCA interacts with both RSV NP and OsTLP, we measured the tripartite interaction. Through pull-down assays, we discovered that when LssaCA was pre-incubated with RSV to form an NP-LssaCA complex, the complex still had the ability to bind to OsTLP (Fig 7D). This result demonstrates a tripartite interaction among these three proteins.

We then explored how the tripartite interaction affects OsTLP enzymatic activity. OsTLP was pre-incubated with LssaCA-NP complex or LssaCA control. After incubation for 2 h, the enzyme activity was measured. The results showed that pre-incubation with LssaCA-NP complex significantly increased OsTLP activity compared with LssaCA control (Fig. 7E), indicating that stronger activation of β-1,3-glucanase activity was achieved through the NP-LssaCA-OsTLP tripartite interaction.

Taken together, these results suggested that when RSV is delivered into plant phloem, it can induce callose deposition. This process is overturned through a tripartite interaction involving the LssaCA protein which binds to RSV in insect salivary glands and then the LssaCA-RSV complex binds to OsTLP in plants. Once this tripartite interaction occurs, OsTLP activity of β-1,3-glucanase is significantly increased which degrades callose thus allowing for efficient RSV infection.

Discussion

Insects are known to be vectors of many plant viruses, and some of these plant viruses rely entirely on insect vectors for their plant-to-plant transmission. Previous studies have suggested that insect salivary proteins may facilitate arbovirus infection of plants through various mechanisms, such as breaking down plant cell walls, inhibiting plant immune defenses and callose deposition, directly binding to arbovirus to prevent their detection by plant immune systems, and more. Further research is needed to fully understand the role that insect salivary proteins play in arboviruses infection of plants. In this study, we have found that RSV infection of rice plants triggers the deposition of callose. However, this process is reversed by a tripartite interaction involving the central component of an insect salivary protein LssaCA, which subsequently binds to RSV in insect salivary glands and then OsTLP in the plants. The activity of OsTLP’s β-1,3-glucanase is significantly increased when this tripartite interaction occurs; leading to the degradation of callose and thus allowing for efficient RSV infection (Fig. 8).

Figure 8.

Callose is a complex polysaccharide, and its deposition in plant phloem plays an important role in defending against insect attacks and viral spread (Zavaliev et al., 2011). When attacked by piercing insects or viruses, plants activate callose deposition on the sieve plates to occlude the flow of phloem sap, thus discouraging feeding and reducing viral long-distance spread. This accumulation of callose in plant phloem can be switched on or off through the action of callose synthases and hydrolases (Li et al., 2012; Shangguan et al., 2018). Studies have shown that expression levels of these glucanase enzymes differ between resistant and susceptible plant species. This suggests that some plants may be more successful at resisting infection due to high levels of glucanase enzymes in their tissues, which allow them to more quickly and effectively form callose barriers against insects and viruses (Hao et al., 2008). Some insects have evolved β-1,3-glucanases enzymes that can degrade the deposited callose, thus enabling them to bypass the plant’s defense mechanisms and continue feeding (Bucher et al., 2001; Hao et al., 2008). In this study, we demonstrate a novel mechanism by which an insect saliva protein binds to a plant β-1,3-glucanases protein, thus activating the enzymatic activity and promoting callose degradation.

In this study, we found that plants fed on by viruliferous insects showed higher callose deposition than did plants fed on by uninfected insects (Fig. 5A–C), suggesting that arbovirus infection via insect saliva may represent a double stress to the plant, inducing stronger callose deposition that must be overcome for successful insect feeding and viral spread. To analyze how this happens, we looked at the tripartite interaction between LssaCA-NP-OsTLP. We found that this interaction induced significantly higher β-1,3-glucanase activity than the LssaCA-OsTLP interaction alone, suggesting that SBPH and RSV have evolved mechanisms to increase β-1,3-glucanase activity in order to bypass the plant’s defense of callose deposition and facilitate successful viral spread.

LssaCA is carbonic anhydrase, but its role in aiding RSV infection does not rely on its enzymatic activity (Fig. S2). Our results found that even though LssaCA mutants had significantly reduced enzymatic activity, their ability to facilitate RSV infection was not significantly decreased. Because LssaCA interacts with RSV NP, we measured the interactions between LssaCA mutants and NP. Pull-down assays indicated that all the mutants still interacted with RSV NP (Fig. S4), while the mutants with a stronger interaction (LssaCA^H111D and LssaCA^E153H) showed better facilitation for RSV infection levels (Fig. S2). These results further supported the importance of LssaCA-NP interaction in mediating RSV infection, while the exact role of LssaCA enzymatic activities in aiding RSV infection is still unclear.

In conclusion, this study elucidated a tripartite interaction among a plant virus, an insect saliva protein, and a host phloem β-1,3-glucanase. This interaction inhibited the plant defense of callose deposition and enabled arboviral phloem transmission. This study provides new insights into the tripartite virus-insect vector-plant interaction, which is relevant to many agriculturally important plant arboviruses whose transmission is facilitated via insect saliva proteins.

Figure S1. Nucleotide sequence and deduced amino acid sequence of LssaCA.

Underlining indicates the predicted signal peptide sequence. Seven predicted catalytically active residues are shown in red. Bold font indicates eukaryotic-type carbonic anhydrase sequences.

Figure S2. LssaCA enzymatic activity does not affect RSV infection.

A. Summary of the predicted enzymatic active sites and the muteins.

B. Relative esterase activities of recombinant LssaCA protein and its muteins. *, p < 0.05; **, p < 0.01; ****, p < 0.0001.

C. Schematic diagram to show the process of microinjecting RSV into a plant. The RSV was pre-incubated with recombinantly expressed LssaCA, the LssaCA muteins, or a BSA control, before being microinjected into the rice phloem.

D. The effects of LssaCA, its muteins, or the BSA control, on the levels of RSV infection as determined by RT-PCR. The titers of RSV (copy number of NP) were measured at 14 dpi.

Figure S3. Nucleotide sequence and deduced amino acid sequence of OsTLP.

Underlining indicates the predicted signal peptide sequence. Bold font indicates the conserved thaumatin family protein sequence.

Figure S4. The molecular interaction between RSV and LssaCA muteins is important for RSV to infect rice plants.

A. Pull-down assays showing the interactions between the LssaCA muteins (expressed with a His-tag) and RSV NP (GST-NP). GST was used as a negative control. Anti-His and Anti-GST represent antibodies against 6*His and GST, respectively. Both LssaCA^H111D and LssaCA^E153H (marked in red) have a stronger interaction with NP compared to the wild-type LssaCA protein.

B. Preincubation of RSV with either LssaCA^H111D or LssaCA^E153H facilitates RSV infection of plants after being delivered via microinjection into the phloem. The titers of RSV (copy number of NP) were measured at 14 dpi. BSA was used as a negative control. *, p < 0.05; **, p < 0.01.

Table S1. Primers used in this study.

Materials and methods

Viruses, insects, host plants, and antibodies

The RSV-free and RSV-infected L. striatellus individuals used in this study were originally captured in Jiangsu Province, China, and were maintained in our laboratory. All plants used to rear L. striatellus were grown in a growth incubator (14.5 cm height, 3.2 cm radius) at 25℃ under a 16-h light/8-h dark photoperiod. Insects were transferred to fresh seedlings every 15 days to ensure adequate nutrition and insect activity.

The RSV-specific antibody was produced using RSV ribonucleoproteins (RNPs) as antigens. The LssaCA polyclonal antibody was produced using recombinantly expressed LssaCA protein (with a His-tag at the C-terminus) as the antigen.

Tissue collection

Insect dissection and tissue collection were performed using previously described procedures (Huo et al., 2018). In brief, insects were anesthetized at 4°C for 10 min, and the forelegs were severed at the coxa-trochanter joint using forceps. The hemolymph was expelled and drawn to the tip of clean forceps. The insects were then dissected from the abdomen in pre-chilled PBS buffer. The fat body was collected with forceps, aided by the tension of the liquid. Tissues, including the guts, salivary glands, and ovaries, were washed three times in PBS to remove contaminants from the hemolymph.

Bioinformatic analysis

The signal peptide and transmembrane helix were predicted using tools at the SignalP 5.0 server (https://services.healthtech.dtu.dk/service.php?SignalP-5.0) and the TMHMM-2.0 server (https://services.healthtech.dtu.dk/service.php?TMHMM-2.0). Homology searches of the protein sequence of LssaCA and OsTLP were conducted using the BLAST algorithm at the National Center for Biotechnology Information (http://blast.ncbi.nlm.nih.gov/Blast.cgi). Conserved domains were predicted with SMART (http://smart.embl-heidelberg.de/).

Collection of watery and gel saliva

Saliva was collected as described previously (Huo et al., 2022). Watery saliva was obtained by centrifuging the collected artificial diet at 10,000 ×g for 10 min to remove the solid gel. To collect gel saliva proteins, the membrane was washed twice with PBS. Sheathes adhering to the parafilm were denatured and solubilized in a solution consisting of 8 M urea, 4% v/v CHAPS, 0.1% w/v sodium dodecyl sulfate (SDS), and 2% v/v DTT. The parafilm was incubated on an orbital shaker at room temperature for 45 min. Then, the denatured gel saliva was concentrated to a final volume of 100 µL. Saliva samples were collected from 2,000 insects for silver staining and western blotting analyses.

Measurement of carbonic anhydrase activity

Carbonic anhydrase activity was characterized by measuring its esterase activity. P-nitroethyl acetate (p-NPA) was used as the substrate. The LssaCA enzyme solution was prepared as follows: LssaCA was expressed in E. coli with a His-tag at the C-terminus and the purified LssaCA protein was diluted in PBS buffer (pH 7.0) to a final concentration of 1 mg/mL. Details of the reaction conditions have been described elsewhere (Verpoorte et al., 1967). Briefly, 1 mL of freshly prepared 3 mM p-NPA dissolved in acetone was added to 1.9 mL 100 mM PBS (pH 7.0), followed by the addition of 0.1 ml LssaCA enzyme solution (0.5 mg/mL in 100 mM PBS). As the blank control, 2 mL 100 mM PBS (pH 7.0) was added to 1 mL 3 mM p-NPA. The absorbance at 384 nm was measured at the start of the reaction and 3 min later. One unit (U) of enzyme activity is defined as the amount of enzyme that hydrolyzes 1 μmol p-nitroethyl acetate per minute at pH 7.0 at 25 °C.

Immunofluorescence assays

Immunofluorescence assays (IFA) were performed to determine the localization of RSV and LssaCA in L. striatellus salivary glands. The dissected salivary gland samples were fixed in 4% v/v paraformaldehyde at 4°C overnight and then permeabilized in 2% v/v Triton X-100 for 2 h. The samples were incubated with rabbit anti-RSV serum and mouse anti-LssaCA serum (1:250 dilution in PBST/FBS: PBS containing 2% v/v Triton X-100 and 2% fetal bovine serum) for 1 h at room temperature. After washing twice with PBST, the samples were incubated with 1:200 diluted Alexa Fluor 488-conjugated goat anti-mouse antibody and Alexa Fluor 594-conjugated goat anti-rabbit antibody (Invitrogen) for 1 h. After washing three times with PBST, the samples were examined under a Leica TCS SP8 microscope (Leica, Wetzlar, Germany).

Evaluation of callose deposition in rice plants

Five insects were allowed to feed on a 10 d-old rice plant in a leaf area of 2 × 2 cm for 24 h. The leaf samples were immersed in 5 mL alcoholic lactophenol (1 volume phenol: glycerol: lactic acid: water (1:1:1:1) and 2 volumes ethanol), then evacuated with a syringe for 15 min, and then incubated at 65℃ for 30 min until the chlorophyll was completely cleared. Cleared leaf samples were rinsed with 50% v/v ethanol, then with water, and then stained in 150 mM K₂HPO₄ (pH 9.5) containing 0.01% w/v aniline blue for 30 min in the dark. Stained leaf samples were sectioned and mounted in 50% v/v glycerol on slides. The slides were examined under a Zeiss observer Z1 immunofluorescence microscope (Carl Zeiss MicroImaging GmbH, Gottingen, Germany), and fluorescence intensity was determined using ImageJ software.

Callose concentrations were measured using a plant callose ELISA kit (Catalog number: JL46407; Jonln, Shanghai, China), according to the manufacturer’s protocol. The callose solution was prepared as follows: leaf material from the feeding site was homogenized with PBS buffer (pH 7.4) and the mixture was centrifuged at 500–1000 g for 20 min. The supernatant was collected for the detection of callose concentration. The absorbance at 450 nm was measured and callose concentrations were calculated based on the standard curve.

RNA interference

We conducted RNAi analyses to determine the function of LssaCA in mediating RSV infection and regulating callose deposition. The LssaCA-specific gene fragment was PCR-amplified with the primer pair LssaCA-T7F/LssaCA-T7R (Table S1). Then, dsRNA was synthesized using a commercial kit (T7 RiboMAX Express RNAi System, Promega, Madison, WI, USA) and purified by phenol: chloroform extraction and isopropanol precipitation. Finally, 36.8 nL dsRNA at 1 ng/nL was delivered into the insect hemocoel for gene silencing. Microinjection was performed using a glass needle through a Nanoliter 2000 (World Precision Instruments, Sarasota, FL, USA). The negative control, GFP dsRNA, was synthesized and microinjected following the same protocol.

RT-qPCR

Gene transcript levels were determined by RT-qPCR. The RNA was extracted from L. striatellus tissues using an RNeasy Mini Kit (Qiagen, Hilden, Germany), and from plant leaves using TRIzol. Reverse transcriptional PCR (iScript cDNA Synthesis Kit, Bio-Rad, Hercules, CA, USA) and SYBR-Green-based qPCR (SYBR Green Supermix, Bio-Rad) was performed according to the manufacturer’s protocols. The primer pair used to amplify LssaCA was LssaCA-q-F/LssaCA-q-R (Table S1). Viral RNA copies were measured using the primers NP-q-F/NP-q-R (Table S1). L. striatellus translation elongation factor 2 (EF2) and rice actin were amplified as internal controls to ensure equal loading of cDNA isolated from different samples. The primer pairs EF2-q-F/EF2-q-R and actin-q-F/actin-q-R (Table S1) were used to amplify EF2 and actin, respectively. Water was used as a negative control.

Yeast two-hybrid assays

Yeast two-hybrid screening was performed to identify rice proteins interacting with LssaCA. A high-complexity rice cDNA library was fused to Gal4 AD and transformed into the yeast strain Y187. The LssaCA gene lacking the signal-peptide sequence (LssaCA^sp-) was fused to Gal4 DNA-BD and transformed into the yeast strain Y2HGold. The two yeast strains were co-cultured overnight to allow mating to create diploids. After library screening, positive clones were selected on quadruple dropout medium: SD/–Ade/–His/–Leu/–Trp supplemented with X-a-Gal and aureobasidin A (QDO/X/A), and prey plasmids were isolated for sequencing. To further confirm the interaction between LssaCA and OsTLP, LssaCA^sp- was cloned into the prey vector pDB-Leu, and the OsTLP gene lacking the signal-peptide sequence (OsTLP^sp-) was cloned into the bait vector pPC86. The two plasmids were co-transformed into the yeast stain Mav203 and the transformants were selected on SD/–His–Leu–Trp + 20 mM 3-AT medium.

Pull-down assays

Pull-down assays were performed to determine the molecular interactions between LssaCA and RSV NP, LssaCA muteins and RSV NP, LssaCA and OsTLP, and OsTLP and the LssaCA-NP complex.

The interaction between LssaCA and RSV NP was determined using GST pull-down analyses as described previously (Huo et al., 2014). Briefly, LssaCA^sp- was cloned into the pGEX-3x vector and LssaCA was expressed as a GST-fusion protein (GST-LssaCA). The NP gene was cloned into the pET-30a (+) expression vector and the protein was expressed with an N-terminal His-tag (NP-His). The GST-LssaCA construct was expressed in, and purified from, E. coli stain BL21. The GST-LssaCA protein or the GST control protein was bound to Glutathione Sepharose beads (Cytiva, Marlborough, MA, USA), then NP-His protein was added to the beads. The bead-bound proteins were separated by SDS-PAGE and detected by western blotting with antibodies against GST or His-tag.

We conducted His pull-down assays to determine the interactions between the LssaCA muteins and RSV NP. The gene sequences encoding LssaCA muteins (without the signal-peptide sequence) were each cloned into the pET-30a (+) expression vector, and each mutein was expressed with an N-terminal His-tag. The NP gene was cloned into the pGEX-3x vector and NP was expressed as a GST-fusion protein (GST-NP). Each of the His-tag-fused LssaCA muteins was bound to Ni Sepharose (GE Healthcare, Uppsala, Sweden), then the GST-NP protein or the GST control protein was added to the beads. The bead-bound proteins were separated by SDS-PAGE and detected by western blotting with antibodies against GST or His-tag.

Next, MBP pull-down analyses were performed to determine the interaction between OsTLP and LssaCA. The OsTLP^sp- sequence was cloned into pMAL-p2X to produce the MBP-fusion protein (MBP-OsTLP). The MBP-OsTLP protein or the MBP control protein was bound to amylose resin (New England Biolabs, Beverley, MA, USA), and then the LssaCA-His protein was added to the beads. To determine the interaction between OsTLP and the LssaCA-NP complex, the LssaCA-His and GST-NP proteins were preincubated before being added to the beads. The bead-bound proteins were separated by SDS-PAGE and detected by western blotting with the MBP antibody (Cell Signaling Technology, Danvers, MA, USA), His-tag antibody (Tiangen, Beijing, China) and GST antibody (Tiangen).

Microscale thermophoresis (MST) assays

We conducted MST assays to detect the interactions between LssaCA and NP, and between LssaCA and OsTLP. First, 10 μM purified LssaCA^sp- protein (His-fusion protein) was labelled with a Monolith NT Protein Labeling Kit RED-NHS (Nano Temper Technologies GMBH, München, Germany) using red fluorescent dye NT-647 N-hydroxysuccinimide (amine-reactive), according to the manufacturer’s instructions. The binding assays were performed on a Monolith NT.115 Microscale Thermophoresis instrument (Nano Temper Technologies GMBH) using capillaries treated using the standard method. The labeled protein LssaCA^sp- was added to serially diluted NP (GST-NP, GST was used as a negative control) or OsTLP^sp- (MBP-fusion protein, MBP was used as a negative control). The initial concentration of each protein was 40 µM with 0.1% (v/v) Tween 20. The KD Fit function of NanoTemper Analysis software version 1.2.214.1 was used for curve fitting and to calculate the dissociation constant (Kd).

RSV inoculation through midrib microinjection

Midrib microinjection was performed as described previously (Zhao et al., 2017). Briefly, 200 RSV-infected L. striatellus were homogenized with 200 μL PBS (pH 7.0), and the homogenate was centrifuged twice at 8000 g for 10 min at 4℃. The crude virus extract was transferred to a new centrifuge tube. The concentration of RSV in the extract was measured using a Nanodrop One instrument. The RSV extract was adjusted to 1 mg/mL for microinjection. Rice seedlings were grown in an incubator for 2 weeks until the leaf veins became clearly visible. Then, 23 nL RSV extract was microinjected into the midrib. The RSV titers in plants were detected at 7–14 days post microinjection.

β-1,3-glucanase activity

β-1,3-glucanase hydrolyzes laminarin and cuts the β-1,3-glucoside bond to produce a reducing terminus. Therefore, the rate of reducing sugars production was used to calculate enzyme activity. One unit of β-1,3-glucanase activity was defined as the amount of enzyme that hydrolyses laminarin to generate 1 μg reducing sugars per minute.

The β-1,3-glucanase activity of OsTLP was measured using a beta-1,3-glucanase microplate assay kit (Cat. ASK1028, Bioworld, Bloomington, MN, USA) according to the manufacturer’s instructions.

To determine the enzymatic activity of the recombinant OsTLP proteins, the MBP-fusion OsTLP^sp- protein or the MBP control protein was adjusted to 0.5 mg/mL. To determine the enzymatic activity of the in vivo-expressed OsTLP^sp-, 0.1 g rice leaf material was homogenized with 1 mL assay buffer on ice, and then the mixture was centrifuged at 12,000g at 4°C for 10 min. The supernatant was transferred into a new centrifuge tube and kept on ice until detection of enzymatic activity.

For the assay, a 50-μL aliquot of the protein solution was mixed with laminarin substrate and incubated at 37°C for 30 min. The reaction was terminated by incubating the reaction tube in a boiling water bath for 10 min. The reaction supernatant was transferred into a microplate and incubated at 90°C for 10 min before measuring the absorbance at 540 nm.

Generation of transgenic plants

The full-length OsTLP gene was PCR-amplified with the primer pair TLP-F/TLP-R (Table S1), and cloned into the plant expression vector pCAMBIA-1300-eGFP to produce the plasmid pCAMBIA-1300-OsTLP. The plasmid pCAMBIA-1300-OsTLP was transformed into seedlings of the Japonica rice cultivar Nipponbare through Agrobacterium-mediated transformation as described previously (Hiei et al., 1994). Thirty independent T₀ transformants were obtained and were screened by germinating seeds from T₂ lines on medium containing 50 μg/ml hygromycin. The hygromycin-resistant homozygous seeds were used for further experiments. Seeds from homozygous lines transformed with the empty vector pCAMBIA-1300-eGFP were used as the negative control.

Statistical analyses

Graphs were constructed and statistical analyses were performed using Prism 9.0 software (GraphPad Software, San Diego, CA, USA). Data are expressed as the mean ± standard deviation (SD) calculated from three independent experiments. The significance of differences between two groups was determined by unpaired Student’s t-test. Differences among three or more groups were compared using one-way analysis of variance (ANOVA).

Acknowledgements

This work is supported by the National Key R&D Program of China (2022YFD1400800), Major Program of the National Natural Science Foundation of China (32090013) and Youth Innovation Promotion Association CAS (2021084).

Author contributions

Conceptualization, Z.L.L., H.Y., Z.J.; Methodology, H.Y., Z.J. and Y.J.; Investigation, Z.J., Y.J., M.X.Y., H.Y.; Writing-Original Draft, H.Y. and Z.J.; Writing-Review & Editing, Z.L.L. and F.R.X.; Funding Acquisition, Z.L.L., H.Y. and F.R.X.; Supervision, Z.L.L., and F.R.X.

Declarations of competing interests

The authors declare no competing interests.