Fig 1.
LsST6 interacted with viral proteins in vivo and in vitro.
(A) Interaction between LsST6 and the outer capsid or nucleocapsid protein of different viruses detected using an in vivo yeast two-hybrid assay. Yeast strain NMY51 was cotransformed with LsST6 and the respective viral proteins. After 10−1 to 10−4 times dilution, the yeast cells were plated onto DDO (SD-trp-leu) and QDO (SD-trp-leu-his-ade-20 mM 3-AT) medium plates, respectively. Clones grown on DDO were selected for β-galactosidase activity assay. Large T + P53 was used as the positive control; Large T + LsST6 served as the negative control. (B) Confirmation of the interaction by in vitro coimmunoprecipitation (Co-IP). Sf9 cells were cotransfected with the respective recombinant bacmids (LsST6 fused with His-tag and viral proteins linked with Myc-tag) for protein expression. After 72 h, cells were harvested, lysed in lysis buffer, then incubated with anti-Myc antibody and protein A/G plus agarose beads for IP. The anti-Myc antibody or anti-His antibody was used for western blots.
Fig 2.
LsST6 affected the localization of interacting viral proteins in Sf9 cells.
(A) Localization of LsST6 and outer capsid or nucleocapsid protein of different viruses in Sf9 cells. The recombinant bacmids (LsST6 and viral proteins) were individually used to transfect Sf9 cells, then observed with LSCM. (B) Colocalization of LsST6 and outer capsid or nucleocapsid protein of different viruses in Sf9 cells. LsST6 and the respective viral proteins were coexpressed in Sf9 cells and observed with LSCM 72 h after transfection. LsST6 was labeled with Dylight 488 (green); viral proteins were labeled with Dylight 549 (red). Cell nucleus was stained with DAPI (blue). Noninfected Sf9 cells served as the negative control. Scale bars, 10 μm.
Fig 3.
LsST6 is involved in rice stripe virus invasion of the midgut epithelial cells of L. striatellus.
(A) LsST6 expression in different tissues. Total RNA was extracted from the salivary gland, gut, hemolymph and ovary for assay by RT-qPCR. The whole insect body served as the control. Means ± SEM of three independent experiments are shown (P < 0.01; one-way ANOVA, least significant difference [LSD] test). WB: whole body; SG: salivary glands; Gu: gut; He: hemolymph; Ov: ovary. (B) Localization of LsST6 in midgut epithelial cells. Excised midgut was incubated with anti-LsST6 antibody labeled with Dylight 488 (green) and observed with LSCM. Dylight 633 phalloidin was used to label actin (blue) of midgut epithelial cells. Scale bars, 50 μm. (C-E) Colocalization of LsST6 and RSV on membrane of midgut epithelial cell during RSV invasion. Excised midguts were incubated with anti-LsST6 labeled with Dylight 488 (green) and anti-RSV labeled with Dylight 549 (red) at 2 days (C), 4 days (D) and 8 days (E) after a 2-day access acquisition period (AAP), then observed with CLSM. Scale bars, 50 μm.
Fig 4.
Immunoelectron microscopy to show colocalization of RSV particles and LsST6 in epithelial cells of L. striatellus allowed to feed for 48 h on RSV-infected rice plants.
(A) Immunoelectron micrograph of midgut epithelial cell. (B, C) Colocalization of RSV particle and LsST6 in microvilli and cytoplasm. Red box: Magnified area. Red arrows: 10-nm gold-conjugated goat anti-rabbit IgG against RSV used to detect virus; yellow arrows: 5-nm gold-conjugated goat anti-mouse IgG against LsST6.
Fig 5.
RSV particles successfully invaded Sf9 cells that expressed the heterologous gene LsST6.
(A) RSV particle invasion of Sf9 cells was mediated by binding with LsST6. RSV particles were added to Sf9 cells, which had been transfected by recombinant bacmid plasmid of LsST6 at 48 h. After 7 h, 15 h and 20 h incubation with RSV particles, Sf9 cells were treated with anti-LsST6 labeled with Dylight 488 (green) and anti-RSV labeled with Dylight 549 (red) and observed with LSCM. Scale bars, 10 μm. (B-D) The accumulation level of RSV particles increased when LsST6 was expressed in Sf9 cells. Total RNA was extracted from Sf9 cells at 7, 15 and 20 h after adding virus particles and then probed by DIG-northern blot using specific probes (B) and by RT-qPCR (C and D) to analyse the levels of LsST6 and RSV RNA. The mRNA levels of the housekeeping gene ECD were used as an internal control, and mRNA for LsST6, and RNA for RSV level of LsST6 and RSV at 7 h was set to 1. (B–D) Mean ± SEM of three independent experiments (P < 0.01, one-way ANOVA, LSD test).
Fig 6.
Immunoelectron microscopy showing RSV particle entered Sf9 cells.
(A) RSV particles and LsST6 colocalized in the cell membrane by 7 h. (B) RSV particles had entered the cytoplasm. (C) Numerous RSV particles were distributed in the cytoplasm. Red arrows: 10-nm gold-conjugated goat anti-rabbit IgG against RSV used to detect virus; yellow arrows: 5-nm gold-conjugated goat anti-mouse IgG against LsST6.
Fig 7.
Knockdown of LsST6 expression interfered with virus entry into midgut epithelial cells of healthy planthoppers.
(A) RNA interference mediated by microinjection with dsLsST6. Third-instar and nonviruliferous nymphs were injected with dsLsST6 or dsGFP using an Auto-Nanoliter Injector, then fed on healthy rice seedling. LsST6 mRNA expression level was quantified by RT-qPCR at different times. (B) LsST6 RNA levels were reduced after injection with dsLsST6 as quantified by RT-qPCR. (C) RNA level of RSV quantified by RT-qPCR. (D) RNA levels of RSV and LsST6 were quantified by DIG-northern blot. Nonviruliferous SBPHs that had been injected with either dsLsST6 or dsGFP were fed on RSV-infected rice for a 2-day acquisition access period (AAP), then collected at 2, 4 and 8 days to quantify RNA. Level of the housekeeping gene actin was used as an internal control. (E) Percentage of RSV-positive insects and transmission efficency decreased significantly compared with the control that was injected with dsGFP. Total RNA was extracted from insects at 8 days after a 2-day AAP or from rice seedling at 21 days after a 10 h inoculation access period and used to detect RSV by RT-PCR using specific primers. (A–D) Mean ± SEM of three independent experiments (P < 0.01, one-way ANOVA, LSD test). (F) Localization of LsST6 and RSV in midgut epithelial cells after knockdown of LsST6 expression. Excised midguts were incubated with anti-LsST6 (green) and anti-RSV (red) antibodies and observed with LSCM. Scale bars, 50 μm.
Fig 8.
RNAi knockdown of LsST6 had no effect on virus titre and transmission efficiency in viruliferous planthoppers.
(A) LsST6 RNA expression was inhibited by injection with dsLsST6. Viruliferous insects were fed on health rice seedlings and collected at 2, 4 and 8 days after the dsRNA injection, then RNA levels quantified by RT-qPCR. (B-C) RSV titre in viruliferous insects after injection. RSV RNA levels were quantified by (B) RT-qPCR and (C) DIG-northern blots. (D) Transmission efficiency of virus to rice seedlings by viruliferous insects after injection with dsLsST6. (A-D) Mean ± SEM of three independent experiments (P < 0.01, one-way ANOVA, LSD test). (E) Localization of LsST6 and RSV in midgut epithelial cells after injection with dsLsST6. Excised midguts at 2, 4 and 8 days were incubated with anti-LsST6 (green) and anti-RSV (red) antibodies and observed with LSCM. Scale bars, 50 μm.
Fig 9.
Model of virus entry into the midgut epithelial cell.
The viruses (RSV, RBSDV, SRBSDV and RGSV) enter the lumen of the midgut with sap of virus-infected rice plants ingested by L. striatellus, then the virions bind to the cell membrane of midgut epithelial cells by specific interaction with LsST6. Only a compatible interaction results in the virions crossing the cell membrane to enter the epithelial cells, where they replicate and are disseminated to other body parts of the insect.