Cross-kingdom RNA interference mediated by insect salivary microRNAs may suppress plant immunity

Significance Growing evidence has highlighted the role of small RNAs (sRNA) as trafficking effectors for cross-kingdom RNA interference (RNAi) between interacting organisms. Yet, it remains unknown whether insect-derived miRNAs can serve as the cross-kingdom effectors to regulate plant physiology. Here, we report that Nilaparvata lugens secretes sRNAs into host plants to enhance feeding. Specifically, miR-7-5P, which is highly conserved in sequence, serves as a salivary effector targeting multiple host genes. This effector systemically migrates after secretion and may be crucial for insect feeding on rice plants, but not on artificial diets. Our findings illustrate a type of salivary effector that insects may use to manipulate plant immunity.


This PDF file includes:
Figs. S1 to S13 Tables S1 to S5 Fig. S1.Secondary structures of miRNA precursors (pre-miRNAs).The mature miRNA located in the 5' end was highlighted by blue, whereas the mature miRNA located in the 3' end was highlighted by yellow.Genome locations of these pre-miRNAs were displayed in Table S1.The sRNAminer software was used to visualize the stem-loop structure.Rubisco staining (RbcL) was conducted to visualize the amount of sample loading.Vectors encoding GFP protein alone or recombinant β-glucuronidase (GUS)-GFP protein were served as negative controls.Bar = 50 μm.The target genes and their corresponding codes are as follows: Os02g49560 (OsbZIP43), Os03g38330 (NBS-LRR-like resistance protein), Os04g02860 (disease resistance protein RPM1), Os10g30040 (BTB/POZ domain containing protein), Os11g36719 (lipoxygenase), and Os10g18990 (receptor kinase 2).The target band was indicated by red arrows.Three independent biological replicates were performed, and the representative fluorescence and Western-blotting images were displayed.The intensity/density value from three biological replicates were calculated using ImageJ, and the mean value in the controls were set as 1.0.Data are presented as mean ± SEM (n= 3 independent biological replicates).The small triangle indicates the different concentration (OD 600= 0.3 and 1.0) of Agrobacterium harboring p35S: miR-7-5P.Motor domain

Fig. S5 .
Fig. S5.Validation of the interaction between miR-7-5P and target sequences.DNA fragments approximately 100 bp upstream and downstream of the target sites were ligated to the 3' end of the GFP gene.The recombinant plasmid was co-infiltrated into Nicotiana benthamiana leaves along with varying concentration of p35S: miR-7-5P.The miR-7-5P level in leaves was determined by stem-loop qRT-PCR.Confocal microscopy (left) and Western-blotting (right) were performed.Rubisco staining (RbcL) was conducted to visualize the amount of sample loading.Vectors encoding GFP protein alone or recombinant β-glucuronidase (GUS)-GFP protein were served as negative controls.Bar = 50 μm.The target genes and their corresponding codes are as follows: Os02g49560 (OsbZIP43), Os03g38330 (NBS-LRR-like resistance protein), Os04g02860 (disease resistance protein RPM1), Os10g30040 (BTB/POZ domain containing protein), Os11g36719 (lipoxygenase), and Os10g18990 (receptor kinase 2).The target band was indicated by red arrows.Three independent biological replicates were performed, and the representative fluorescence and Western-blotting images were displayed.The intensity/density value from three biological replicates were calculated using ImageJ, and the mean value in the controls were set as 1.0.Data are presented as mean ± SEM (n= 3 independent biological replicates).The small triangle indicates the different concentration (OD 600= 0.3 and 1.0) of Agrobacterium harboring p35S: miR-7-5P.

Fig. S6 .
Fig. S6.Effects of OsbZIP43 overexpression on rice plants.(a, b) Detection of OsbZIP43 in transgenic plant oebZIP#1 and oebZIP#2 using Western-blotting (a) and qRT-PCR assays (b).The empty vector (EV) transgenic plants served as a control.The full open reading frames of OsbZIP43 with a flag-tag at the C-terminus was overexpressed.The flag-tag antibody was used to detect the presence of OsbZIP43-flag protein.Histone H3 antibody was used to visualize the amount of sample loading.(c, d) Effects of OsbZIP43 overexpression on insect honeydew excretion (c) and reproduction (d).Nilaparvata lugens were allowed to feed on indicated rice plants, respectively.For honeydew analysis, n= 10 independent biological replicates.For reproduction analysis, n= 16, 15, and 15 independent biological replicates on EV, oebZIP#1, and oebZIP#2, respectively.P-values were determined by two-tailed unpaired Student's t test.***P < 0.001; **P < 0.01; *P<0.05.Data are presented as mean ± SEM.

Fig. S7 .
Fig. S7.Suppression of OsbZIP43 expression by miR-7-5P, but not miR-184-3P.At 48-h after co-infiltration, green fluorescence of the OsbZIP43-GFP was observed in Nicotiana benthamiana leaves (a), and the protein level of GFP-fused proteins was determined by a Western-blotting assay (b).Bar = 100 μm.miR-184-3P was used as a negative control.Rubisco staining (RbcL) was conducted to visualize the amount of sample loading.Three independent biological replicates were performed, and the representative fluorescence and Western-blotting images were displayed.The intensity/density value from three biological replicates were calculated using ImageJ, and the mean value in the controls were set as 1.0.Data are presented as mean ± SEM (n= 3 independent biological replicates).The small triangle indicates the different concentration (OD600=0.05,0.3, and 1.0) of Agrobacterium harboring p35S: miR-7-5P or p35S: miR-184-3P.

Fig. S8 .
Fig. S8.Effects of miR-7-5P on the transcript level of OsbZIP43 and OsbZIP43-M.Coinfiltration of miR-7-5P was performed with OsbZIP43 and OsbZIP43-M, respectively.The relative transcript levels of OsbZIP43 (a) and OsbZIP43-M (b) were determined using qRT-PCR.Different lowercase letters indicate statistically significant differences at P < 0.05 level according to one-way ANOVA test followed by Tukey's multiple comparisons test.Data are presented as mean ± SEM (n= 3 independent biological replicates).The small triangles below the figure indicate the different concentration (OD600=0.05,0.3, and 1.0) of Agrobacterium harboring p35S: miR-7-5P.

Fig. S9 .
Fig. S9.Effects of miR-7-5P overexpression on insects.Nilaparvata lugens were allowed to feed on transgenic plant oemir7#1 and oemir7#2.Honeydew excretion (a) and reproduction (b) were determined.The empty vector (EV) transgenic plants served as a control.n= 11 and 16 independent biological replicates in honeydew and reproduction analyses, respectively.P-values were determined by two-tailed unpaired Student's t test.**P < 0.01; *P<0.05;ns, not significant.Data are presented as mean ± SEM.

Fig. S10 .
Fig. S10.Effects of OsbZIP43-GFP and OsbZIP43-M-GFP overexpression on Nicotiana tabacum.The OsbZIP43-GFP, OsbZIP43-M-GFP, and GFP were transiently expressed in N. tabacum, respectively.At 48 h post-infiltration, cell morphology and fluorescence signal derived from GFP expression were observed.No necrosis effect was detected in GFP-, OsbZIP43-GFP-or OsbZIP43-M-GFP-expressing leaves at 48 h post-infiltration.In contrast, leaves infiltrated with InF1, a necrosis-inducing protein from Phytophthora infestans, exhibited significant cell morphological changes.

Fig. S12 .
Fig. S12.The proposed model for the cross-kingdom RNA interference between Nilaparvata lugens and rice plants.During the feeding process, N. lugens secreted a few salivary miRNAs into rice plants, including the miR-7-5p, which is specifically expressed in the salivary glands.The miR-7-5P enhances insect feeding on rice plants by suppressing multiple immune-associated genes in host plants, such as the bZIP transcription factor 43 (bZIP43), pattern-recognition receptors (PRRs), and nucleotide-binding domain, leucine-rich repeat domain-containing receptors (NLRs).

Fig. S13 .
Fig. S13.The diagram of inserted sequences in constructing vectors for transgenic rice plants.The pBAW(V)HS of pCAMBIA1300 backbone was selected for vector construction.For oemir7 plants (a), the looped sequence was inserted after CaMV 35S promoter (35S Pro), while for oebZIP plants (b), the full open reading frames of OsbZIP43 with a flag-tag at the C-terminus was inserted after 35S Pro.

Table S1 . Identification of insect-derived miRNA in saliva
2The abundance of miRNAs was displayed as transcripts per million (TPM)