The NIa
protease of potyviruses is a
chymotrypsin-like
cysteine protease related to the picornavirus 3C
protease. It is also a multifunctional
protein known to play multiple roles during
virus infection. Picornavirus
3C proteases cleave hundreds of host
proteins to facilitate
virus infection. However, whether or not potyvirus NIa
proteases cleave
plant proteins has so far not been tested. Regular expression search using the cleavage site consensus sequence [EQN]xVxH[QE]/[SGTA] for the plum pox virus (PPV)
protease identified 90 to 94 putative cleavage events in the
proteomes of Prunus persica (a crop severely affected by PPV), Arabidopsis thaliana, and Nicotiana benthamiana (two experimental hosts). In vitro processing assays confirmed cleavage of six A. thaliana and five P. persica
proteins by the PPV
protease. These
proteins were also cleaved in vitro by the
protease of turnip mosaic virus (TuMV), which has a similar specificity. We confirmed in vivo cleavage of a transiently expressed tagged version of AtEML2, an EMSY-like
protein belonging to a family of nuclear
histone readers known to be involved in pathogen resistance. Cleavage of AtEML2 was efficient and was observed in plants that coexpressed the PPV or TuMV NIa
proteases or in plants that were infected with TuMV. We also showed partial in vivo cleavage of AtDUF707, a
membrane protein annotated as
lysine ketoglutarate reductase trans-splicing
protein. Although cleavage of the corresponding endogenous
plant proteins remains to be confirmed, the results show that a plant virus
protease can cleave host
proteins during
virus infection and highlight a new layer of plant-virus interactions. IMPORTANCE Viruses are highly adaptive and use multiple molecular mechanisms to highjack or modify the cellular resources to their advantage. They must also counteract or evade host defense responses. One well-characterized mechanism used by vertebrate viruses is the proteolytic cleavage of host
proteins to inhibit the activities of these
proteins and/or to produce cleaved
protein fragments that are beneficial to the
virus infection cycle. Even though almost half of the known plant viruses encode at least one
protease, it was not known whether plant viruses employ this strategy. Using an in silico prediction approach and the well-characterized specificity of potyvirus NIa
proteases, we were able to identify hundreds of putative cleavage sites in
plant proteins, several of which were validated by downstream experiments. It can be anticipated that many other plant virus
proteases also cleave host
proteins and that the identification of these cleavage events will lead to novel
antiviral strategies.