The complete genomes of 30 Soybean mosaic virus (SMV) isolates and strains were sequenced in this study. Together with fourteen previously reported sequences, we analyzed the genetic structure of the SMV population. Analyses of genetic diversity showed that different genomic regions of SMV are under different evolutionary constraints and that there was no significant genetic differentiation between East Asian and North American populations of SMV. Phylogenetic analyses revealed a significant correlation between phylogeny of the cylindrical inclusion (CI) gene of SMV and SMV resistance gene 3 (Rsv3)-relating pathogenicity of SMV, suggesting CI might be a pathogenic determinant in Rsv3-mediated disease response. Interestingly, recombination analyses identified 19 'clear' recombination events in the SMV population. Furthermore, as several resistance-breaking strains were identified as recombinants, it appears that recombination might contribute to overcome host resistance in SMV-soybean pathosystem. Our finding suggests that recombination as well as mutation is an important evolutionary process in the genetic diversification of SMV population.

Effector-triggered immunity (ETI) is an active immune response triggered by interactions between host resistance proteins and their cognate effectors. Although ETI is often associated with the hypersensitive response (HR), various R genes mediate an HR-independent process known as extreme resistance (ER). In the soybean-Soybean mosaic virus (SMV) pathosystem, the strain-specific CI protein of SMV functions as an effector of Rsv3-mediated ER. In this study, we used the soybean (Rsv3)-SMV (CI) pathosystem to gain insight into the molecular signaling pathway involved in ER. We used genome-wide transcriptome analysis to identify a subset of the type 2C protein phophatase (PP2C) genes that are specifically up-regulated in Rsv3-mediated ER. Gain-of-function analysis of the most significantly expressed soybean PP2C gene, GmPP2C3a, showed that ABA-induced GmPP2C3a functions as a key regulator of Rsv3-mediated ER. Our results further suggest that the primary mechanism of ER against viruses is the inhibition of viral cell-to-cell movement by callose deposition in an ABA signaling-dependent manner.

Cucumber mosaic virus (CMV) encodes two viral replication proteins, 1a and 2a. Accumulating evidence implies that different aspects of 1a-2a interaction in replication complex assembly are involved in the regulation of virus replication. To further investigate CMV replicase assembly and to dissect the involvement of replicase activities in negative- and positive-strand synthesis, we transiently expressed CMV RNAs and/or proteins in Nicotiana benthamiana leaves using a DNA or RNA-mediated expression system. Surprisingly, we found that, even in the absence of 1a, 2a is capable of synthesizing positive-strand RNAs, while 1a and 2a are both required for negative-strand synthesis. We also report evidence that 1a capping activities function independently of 2a. Moreover, using 1a mutants, we show that capping activities of 1a are crucial for viral translation but not for RNA transcription. These results support the concept that two or more alternate states of replicase assembly are involved in CMV replication.