RRSV infection regulates the spatial expression patterns of leaf phenotype-related miRNAs and their target genes
Leaf complexity in Arabidopsis is regulated by a conserved miRNA transcription factor regulatory network [28, 48]. To investigate the function of this network during rice virus infection, we identified key genes involved in Arabidopsis leaf morphogenesis (AtCUC2, AtCUC3, AtTCP4, and AtSPL9) and their corresponding rice homologs. Phylogenetic analysis demonstrated that OsCUC1 and OsCUC3 are closely related to AtCUC2 and AtCUC3, respectively (Fig. S1a; Table S1). Additionally, OsTCP1 aligns with AtTCP4 (Fig. S1b; Table S1). and OsSPL14 and OsSPL17 correspond to AtSPL9 (Fig. S1c; Table S1). We performed transcriptome and small RNA sequencing on RRSV-infected and control rice (Zhonghua 11, ZH11) at 7-, 14-, and 21-days post-inoculation (dpi). Our results revealed changes in miRNA expression: miR164 decreased at 7 and 21 dpi, miR319 exhibited dynamic expression patterns over time, and miR156 increased at 14 and 21 dpi (Fig. S2a, c; Table S2). Target gene analysis indicated that OsOMTN3 (OsCUC1 was not detected) expression increased progressively in infected rice, whereas OsTCP1 expression decreased at 14 and 21 dpi. OsSPL17 expression remained stable throughout the experiment (Fig. S2b, c; Table S3). To understand the long-term expression of miRNAs regulating leaf morphology, miR164, miR156, miR319, and their target genes were analyzed in mock and RRSV-infected rice from 7 to 56 dpi (Fig. S3a-c). Viral infection led to plant stunting and intensified leaf wrinkling (Fig. S3a, b). miR164 was downregulated during the symptomatic phase (Fig. S3d), while its target gene OsCUC1 was upregulated (Fig. S3e). miR319 and miR156 showed altered expression in infected rice, with miR319 upregulated and miR156 significantly higher during the symptomatic phase (Fig. S3f, h). The target genes OsTCP1 and OsSPL17 exhibited opposite and increasing expression patterns, respectively, in infected rice (Fig. S3g, i). These results indicated that RRSV infection causes dynamic temporal alterations in the expression patterns of miR164, miR319, miR156, and their target genes.
To explore the regulation of spatial expression patterns concerning miRNAs and their target genes during RRSV infection, rice leaves from mock and RRSV-infected samples were selected at 42 dpi. They were divided into four segments: C1, C2, C3, and C4 (for mock samples), as well as R1, R2, R3, and R4 (for RRSV-infected samples). Among these, C4 and R4 represent the respective proximal axis of the two types of leaves. The expression levels of the three miRNAs and their target genes in all segments were compared with the base C4 segment in mock leaves (Fig. 1a). In mock leaves, from the proximal to distal axis (C4 to C1), the expression patterns of miR164, miR319, and miR156 exhibited an initial decrease, followed by an increase (Fig. 1b-d). In contrast, in RRSV-infected leaves, the expression patterns of these three miRNAs demonstrated an initial increase, followed by a decrease (Fig. 1b-d). Intriguingly, compared with mock leaves, the expression levels of these three miRNAs were significantly increased in the middle segments of RRSV-infected leaves (Fig. 1b-d). The target genes regulated by these three miRNAs were also influenced by viral infection. In mock leaves, the three target genes OsCUC1, OsTCP1 and OsSPL17 showed a gradual decrease in expression from the proximal to distal axis (Fig. 1e-g), whereas. In RRSV-infected leaves, the expression patterns of OsCUC1 and OsSPL17 exhibited the opposite trend, gradually increasing from the proximal to distal axis (Fig. 1e, g), whereas OsTCP1 displayed a zig-zag expression pattern (Fig. 1f). Notably, compared with mock leaves, the expression levels of OsCUC1 and OsSPL17 were significantly increased in the visibly affected R2 and R1 segments of infected leaves (Fig. 1e, g), whereas OsTCP1 showed a significant increase in the R2 segment and a significant decrease in the R1 segment (Fig. 1f). Our results indicated that RRSV infection alters the spatial expression patterns of the three miRNAs and their target genes, causing opposite expression patterns in normal and RRSV-infected leaves. The overall changes among OsTCP1, OsCUC1, and OsSPL17 suggested an antagonistic relationship, similar to the antagonistic interaction observed among AtTCP4, AtCUC2, and AtSPL9 in Arabidopsis. These findings implied that OsTCP1, OsCUC1, and OsSPL17 regulate rice leaf development during infection in a manner similar to Arabidopsis, supporting the notion of a conserved regulatory mechanism in leaf morphogenesis before and after viral infection in rice.
OsCUC1 expression is downregulated in response to RRSV infection, contributing to viral pathogenesis
In the model plant Arabidopsis, CUC2 plays a key role in controlling leaf morphology through an antagonistic interaction with auxin, directly causing serration to form on leaf edges [19]. Previous studies have demonstrated that the absence of OsCUC1 in rice results in severe dwarfism, along with curled and shrunken leaves [36, 37]; this phenotype mimics the results of RRSV infection. To examine the regulatory effects of RRSV infection on OsCUC1, we infected pOsCUC1::rOsCUC1-GFP transgenic plants [37] with RRSV and compared them with mock samples. The results showed distinct fluorescence signals in mock samples across the peripheral, margin, and boundary regions of SAM; after RRSV infection, the fluorescence signals in these areas were weak or absent (Fig. 2a). To investigate the role of OsCUC1 in the RRSV infection process, we obtained OsCUC1 deletion mutant rice plants [37]. We initially assessed the expression levels of miR164, miR319, and miR156, along with their target genes OsCUC1, OsCUC3, OsTCP1, OsSPL14, and OsSPL17 in the OsCUC1 mutant lines. The results indicated a significant downregulation of miR164 expression, while the expression of OsCUC3 remains unchanged. Both miR319 and its target gene OsTCP1 exhibited notable downregulation in expression. Conversely, miR156 showed significant upregulation in expression, with its target genes OsSPL14 and OsSPL17 displaying marked downregulation (Fig. 2b, c). These findings suggest a regulatory role of OsCUC1 in modulating leaf phenotype-associated miRNAs and their target genes. We then conducted RRSV inoculation experiments on OsCUC1 deletion plants to assess the involvement of OsCUC1 in viral pathogenesis. The findings revealed that the OsCUC1 mutant rice exhibited greater disease severity compared with wild-type (WT) rice after RRSV inoculation (Fig. 2d). Statistical analyses of plant height showed that in both mock and RRSV-infected samples, the heights of OsCUC1 mutant rice plants were shorter than the height of WT rice plants (Fig. 2e). Virus accumulation in OsCUC1 mutant rice plants was greater than virus accumulation in WT rice (Fig. 2f), indicating that OsCUC1 mutation positively regulated RRSV pathogenesis. Given that OsCUC1 is a target of miR164, we further investigated the role of miR164 in viral pathogenesis (Fig. 2g; Fig. S4a). Our findings demonstrated that miR164 overexpression (OE) lines exhibited increased susceptibility to RRSV, whereas miR164 short tandem target mimic (STTM) transgenic lines (mimic) displayed enhanced resistance (Fig. 2h, i). These findings suggested that RRSV infection alters OsCUC1 expression in the SAM, potentially influencing leaf development. Accordingly, RRSV infection can attenuate and suppress the expression of OsCUC1 in the peripheral zone of the SAM, thereby affecting leaf development; genetic disruption of OsCUC1 expression contributes to viral pathogenesis.
Interaction network of OsTCP1, OsCUC1, and OsSPL14/ OsSPL17 proteins in rice
In Arabidopsis, the miR319 target gene TCP4 and miR156 target gene SPL9 jointly regulate the CUC protein complex, thereby dynamically regulates leaf margin serration [28]. To validate the existence of this target gene interaction network in rice, we investigated the interactions of the rice homolog OsTCP1 with CUC and SPL. Through yeast two-hybrid experiments, we confirmed that OsCUC1 can form homodimers,OsTCP1 interacts with OsCUC1, OsSPL14, and OsSPL17 (Fig. 3a). AlphaFold software was used to predict the protein structures of OsCUC1, OsTCP1, OsSPL14, and OsSPL17; rigid and flexible docking analyses were performed to identify interaction interfaces among these proteins. The results indicated that there were hydrogen bonds involving R28, H30, P31, Y40, E35, P25, and G26 was formed between OsCUC1 itself (Fig. 3b). The hydrogen bonds and salt bridges were present between D47, G48, D86, S160, S162 and T163 in OsCUC1 and R108, D119, D122 and R123 in OsTCP1 (Fig. 3c). There were hydrogen bonds involving K131, Q179, and L417 in OsSPL14, as well as R89, A99, and R108 in OsTCP1 (Fig. 3d). OsSPL17 residues involved in hydrogen bonds were T259, D260, S278, and N299, which interacted with A365, R372, Q321, and S262 in OsTCP1; a salt bridge was formed between S399 and L400 in OsSPL17 and D119 and S130 in OsTCP1 (Fig. 3e). Bimolecular fluorescence complementation (BiFC) experiments also confirmed that OsCUC1 can form homodimers, OsTCP1 interacts with OsCUC1, OsSPL14, and OsSPL17 (Fig. 3f). As transcription factors, OsCUC1, OsSPL14, and OsSPL17 are exclusively localized in the cell nucleus; conversely, OsTCP1 is expressed in both the nucleus and cytoplasm, exhibiting substantial regions of cytoplasmic expression (Fig. S5). The interactions of OsTCP1 with OsCUC1, OsSPL14, and OsSPL17 were not limited to the nucleus; fluorescence signals were also observed in the cytoplasm (Fig. 3f). These results indicated that OsTCP1 in rice may also function as an intermediate bridge, interacting with OsCUC1, OsSPL14, and OsSPL17; such interactions influence the nuclear localization and expression patterns of OsCUC1, OsSPL14, and OsSPL17.
In Arabidopsis, both homotypic and heterotypic dimers can form between CUC2 and CUC3, jointly regulating leaf margin serration [28]. To explore the potential conserved functions of CUC in rice, we conducted yeast two-hybrid experiments, which revealed that OsCUC1 and OsCUC3 can and interact with each other to form heterotypic dimers (Fig. S6a-d). As OsCUC1 expression contributes to viral pathogenesis. To explore whether OsCUC3 is involved in RRSV pathogenesis, we obtained OsCUC3 knockout mutant plants (Fig. S6e) and conducted virus inoculation experiments. In both mock and RRSV-infected samples, the height of OsCUC3 mutant plants were shorter than the height of WT rice plants (Fig. S6f, g). Virus accumulation was higher in OsCUC3 mutant rice than in WT rice plants (Fig. S6h), indicating that the absence of OsCUC3 has positive regulatory effects on RRSV pathogenesis. These results collectively demonstrated that OsCUC1 and OsCUC3 can form heterotypic dimers in rice, and both are involved in viral pathogenesis.
OsTCP1 regulates OsCUC1 self-complex formation through cellular localization and interaction with OsSPL14 and OsSPL17
Our previous research has shown that there is complex pairwise interactions among OsCUC1, OsTCP1, OsSPL14/ OsSPL17 proteins (Fig. 3). To further explore these interactions, we first co-expressed OsCUC1 and OsTCP1 proteins. The co-localization of OsCUC1 with OsTCP1 suggested that the signal of OsTCP1 was similar to its original nuclear localization (Fig. S5); the signal of OsCUC1, originally present solely in the nucleus (Fig. S5), was also found in the cytoplasm in a pattern that partially overlapped with OsTCP1 during co-localization (Fig. 4a). These observations suggested a potential role for OsTCP1 in regulating OsCUC1 self-interaction. We then added OsTCP1 to the OsCUC1 interaction system, with the unrelated protein GUS as a control. The results showed that compared to the control, the fluorescence of OsCUC1 self-interaction was weakened after adding OsTCP1 (Fig. 4b). Statistical results also showed that OsTCP1 significantly reduced the fluorescence signal intensity of OsCUC1 self-interaction (Fig. 4c), indicating that OsTCP1 could disrupt OsCUC1 self-interaction. To address the effect of OsSPL14 and OsSPL17 on the interaction between OsTCP1 and OsCUC1, we added OsSPL14 and OsSPL17 to the BiFC system of OsTCP1 and OsCUC1, respectively, with GUS protein as an unrelated control. Both fluorescence observation and statistical data showed that OsSPL14 and OsSPL17 could significantly disrupt the interaction between OsTCP1 and OsCUC1 (Fig. 4d, e). These results suggested that OsTCP1 may regulate OsCUC1 self-interaction in rice by altering subcellular localization or interacting with OsSPL14 and OsSPL17.
MiR319 overexpression and OsTCP1 knockout exacerbate viral infection
To investigate whether miR319 and OsTCP1 participates in viral pathogenesis, we established miR319 OE, mimic lines and OsTCP1 knockout mutant via CRISPR/Cas9 (Fig. S4b; Fig. S7a-c). Phenotypic observations and statistical analyses revealed that the Ostcp1 mutant displayed severe dwarfing and increased tillering compared with WT rice plants (Fig. S7a, c). We used quantitative reverse transcription polymerase chain reaction RT-qPCR) to assess the expression patterns of miR164, miR319, and miR156 in OsTCP1 mutants. The results indicated significant downregulation of the expression levels of these three miRNAs in Ostcp1 mutants compared with WT rice plants (Fig. 5d). The corresponding target genes—OsCUC1, OsCUC3, and OsSPL14—were significantly upregulated in Ostcp1 mutants (Fig. 5e), suggesting that OsTCP1 had negative regulatory effects on OsCUC1, OsCUC3, and OsSPL14. Virus inoculation experiments revealed that, in both mock and RRSV-infected samples, miR319 OE and Ostcp1 mutant plants exhibited more severe dwarfing compared with WT rice plants (Fig. 5a, b, f, g). Viral genome detection showed that virus accumulation levels were higher in miR319 OE and Ostcp1 mutant rice than in WT rice (Fig. 5c, h), indicating that the absence of OsTCP1 has positive regulatory effects on RRSV pathogenesis. These results demonstrated that OsTCP1 negatively regulates OsCUC1 and OsCUC3 in rice, and OsTCP1 loss leads to more severe viral pathogenesis.
MiR156 overexpression and OsSPL14/OsSPL17 knockout exacerbate viral infection
Mutants of OsSPL14 and OsSPL17 were also generated in rice (Fig. S7d-i), both of which displayed a dwarf phenotype (Fig. S7d, g). Initially, the expression levels of miR164, miR319, and miR156, along with their target genes OsCUC1, OsCUC3, OsTCP1, OsSPL14, and OsSPL17, were assessed in the rice mutants of OsSPL14 and OsSPL17. Significant downregulation of miR164 and its target genes OsCUC1 and OsCUC3 was observed in both rice mutants of OsSPL14 and OsSPL17. In the Osspl14 mutant, miR319 was significantly downregulated, with no change observed in the expression level of its target gene OsTCP1. Conversely, in the Osspl17 mutant, miR319 expression remained unchanged, while its target gene OsTCP1 was significantly upregulated. miR156 was significantly downregulated in both mutants of OsSPL14 and OsSPL17, accompanied by significant downregulation of their target genes OsSPL14 and OsSPL17 in their respective mutants (Fig. 6d, e). These findings suggest a potential regulatory role of OsSPL14 and OsSPL17 in leaf phenotypic-associated miRNAs and their target genes.
To access the role of miR156 and targets OsSPL14 and OsSPL17 in RRSV infection, we generated miR156 OE and mimic rice lines (Fig. S4c). Virus inoculation experiments revealed that, in both mock and RRSV-infected samples, miR164 OE and Osspl14/17 mutant plants exhibited more severe dwarfing compared with WT rice plants (Fig. 6a, b, f, g, I, j). Viral genome detection showed that virus levels were higher in miR319 OE and Osspl14/17 mutant rice than in WT rice (Fig. 6c, h, k). Conversely, miR156 mimic rice lines exhibited enhanced virus resistance relative to WT rice plants, characterized by milder symptoms and reduced virus accumulation (Fig. 6a-c), indicating that overexpression of miR156 or the absence of OsTCP1 has positive regulatory effects on RRSV pathogenesis.