Functional Analysis Of SP Protein Encoded By Rice Stripe Virus And Its Interaction With Host Factor PsbP

Rice stripe virus (RSV) is one of the most economically significant pathogens of rice in China. Rice stripe disease caused by RSV has broken out many times in different provinces of China in recent years, causing disease epidemics and a considerable loss in rice production. The genome of RSV comprises four RNAs and the viral RNA of RNA4encodes the disease-specific protein (SP). Previous studies have shown that the accumulation of the SP protein correlates with symptom development. Thus, the SP protein is considered to be closely related to the pathogenicity of RSV. However, little progress has been made in characterizing the function of the SP protein. To better understand the biological roles of the SP protein during viral infection, the function of SP protein and its interaction mechanism with host factor were investigated in this study.SP was firstly expressed using a modified Potato virus X (PVX)-derived vector. Nicotiana benthamiana seedlings inoculated with PVX-SP exhibited more severe symptoms with leaves mosaic and chlorotic, in contrast to the symptomless phenotype caused by empty PVX at20day post inoculation. Five SP mutants were then constructed into PVX vector according to secondary structure analysis of the SP protein predicted using an online prediction software (http://www.predictprotein.org). The mutant PVX-muSP4lacked the C-terminal52amino acids induced much milder mosaic symptoms. We then generated trangenic N. benthamina and Oryza sativa plants expressing SP or muSP4under Cauliflower mosaic virus35S promoter (35S-SP/35S-muSP4). Expression of SP or muSP4was confirmed by Northern blot and Western blot. All transgenic plants grew normally without any apparent morphological differences, and these plants were then inoculated with RSV. As a result, symptoms observed on the35S-SP plants were earlier and severer than these on the vector-transgenic plants. Northern blot analysis showed that the35S-SP plants accumulated more viral RNA as compared with vector-transgenic plants. However, no difference on symptoms, infective curve and viral RNA accumulation was observed between the35S-muSP4plants and vector-transgenic plants after inoculation by RSV. Together, these results suggested that the SP protein enhances the infective efficiency of RSV. We also found that the C-terminal region is required for the pathogenicity of SP protein during RSV infection.A yeast two-hybrid screen was performed to identify RSV SP interacting host proteins from a rice cDNA library using the SP as bait. One cDNA clone designated as OsPsbP was selected, and its interaction with SP was further confirmed using yeast two-hybrid system. The OsPsbP gene shares high sequence identities with PsbP genes in N. benthamina, Triticum aestivum and Phyllostachys edulis. The full length NbPsbP was amplified from N. benthamina leaf cDNA and then inserted into the vector pGADT7and the mutant fragments of SP were cloned into the vector pGBKT7. The interaction between full-length NbPsbP and SP protein was confirmed by the yeast two-hybrid system. But, PsbP did not interact with muSP4in yeast two-hybrid system. The SP-PsbP interaction was also confirmed by bimolecular fluorescence complementation analysis. The fluorescence showed that these two proteins interacted both in nuclear and cytoplasm. The muSP4protein could not interact with PsbP in living N. benthamiana cells as well. GST pull-down assays in vitro futher verify the interaction between PsbP and SP using the purified SP and PsbP proteins.To understand the mechanism of SP-PsbP interaction, we generated transgenic N. benthamina and O. sativa plants overexpressing PsbP (35S-PsbP) and transgenic N. benthamiana plants carrying a hairpin-structure PsbP expression cassette (PsbP-RNAi), and also obtained a mutant of O. sativa (psbp) lacking PsbP gene. Efficient downregulation of PsbP expression was observed in PsbP-RNAi transgenic N. benthamiana plants and psbp rice plants while overexpression of PsbP was observed in35S-PsbP N. benthamina and O. sativa plants when evaluated by real-time RT-PCR, Northern blot and Western blot analyses.. Under our greenhouse conditions, the35S-PsbP plants grew normally and displayed no apparent phenotypic defects. However, the plants downregulating PsbP were slightly yellow and short than the wild-type (WT) plants. Besides, the net photosynthetic rates, maximal photochemical quenching efficiencies and the electron transfer rates measured on these downregulated N. benthamina plants exhibited significantly lower values as compared with the WT plant. Interestingly, the three parameters of photosynthesis detected in the35S-SP plants also exhibited significant decline as compared with these on WT plants, indicating that the photosynthesis was impaired in these plants.The electron microscopic studies showed that grana stacks were apparently disorganized and the thylakiod membranes were much looser in chloroplasts of PsbP-RNAi transgenic N. benthamiana plant leaves and psbp rice plant leaves. Similar versiform structures of chloroplasts were also observed in35S-SP and RSV-infected N. benthamiana and rice plants. To assess the subcellular localization of PsbP protein, the transient expression of the GFP-PsbP fusion protein in N. benthamiana cells was achieved through agro-infiltration. At36hour post inoculation, green fluorescence from the GFP-PsbP fusion protein co-colocalized with chloroplast autofluorencence was observed in WT. Surprisingly, the PsbP localization was altered when the GFP-PsbP was expressed in35S-SP plants with green fluorescence observed not only in chloroplasts but also in cytoplasm. To further confirm above results, we prepared thin sections from fixed and embedded N. benthamiana or O. sativa leaves and analyzed the localization of the PsbP protein using a gold-conjugated antibody against this protein. Under the electron microscope, the conjugated gold particles were only seen in chloroplasts from WT and PsbP-downregulated plants, and the average numbers of specific gold particles labeled in chloroplasts in PsbP-RNAi or psbp lines were almost a quarter of that in WT. As expected, we observed that the gold particles not only scattered in chloroplasts, but also in mitochondrias and cytoplasmin in the35S-SP and RSV-infected cells. And the numbers of gold particles in these chloroplasts were nearly35-60%of that counted in WT. Only few gold particles were obsvered in the cellular tissues except chloroplats in35S-muSP4transgenic plants. Thus, we conclud that the presence of SP protein reduces the amount of PsbP localization in chloroplasts.To illustrate the biological role of the SP-PsbP interaction during RSV infection, we used RNAi technology to inhibit PsbP expression by silencing vector DNA1and in O. sativa by silencing vector BMV. PsbP-silenced plants were confirmed by real-time RT-PCR analysis and those plants together with PsbP-RNAi plants,35S-PsbP plants, WT and psbp plants were then mechanically inoculated with RSV. The development of symptom appearance was correlated with the expression level of PsbP protein, and symptoms on the PsbP-downregulated lines appeared earlier and more severe as compared with WT plants, but symptoms on the35S-PsbP transgenic appeared later and mild as compared with WT plants. Northern blot analysis further confirmed that the accumulation levels of viral RNAs were higher in the PsbP-downregulated lines and significantly lower in leaves overexpressing PsbP as compared with WT. Therefore, these results suggest that the expression level of PsbP impacts the infection efficiency of RSV.In summary, our findings suggest that the SP interact with PsbP and reduce the amount of PsbP in chloroplasts, leading to the perturbation of chloroplast structure and function, and then increasing host susceptibility to RSV infection. These combined results indicate that SP is a pathogenicity determinant of RSV.