Natural variation in a short region of the Acidovorax citrulli type III-secreted effector AopW1 is associated with differences in cytotoxicity

Bacterial fruit blotch (BFB) is a serious disease of melon and watermelon caused by the Gram-negative bacterium Acidovorax citrulli . The strains of the pathogen can be divided into two major genetic groups, I and II. While group I strains are strongly associated with melon, group II strains are more aggressive on watermelon. Like many pathogenic bacteria, A. citrulli secretes a variety of protein effectors to the host cell via the type III secretion system. In the present study, we characterized AopW1, an A. citrulli type III-secreted effector that shares similarity with the actin cytoskeletondisrupting effector HopW1 of Pseudomonas syringae and with effectors from other plant-pathogenic bacterial species. aopW1 is present in group I and II strains, encoding products of 485 amino acids. Although highly conserved in most of the sequence, AopW1 has a highly variable region (HVR) within amino acid positions 147 to 192, including 14 amino acid differences between groups. Here we show that group I AopW1 is more toxic to yeast and plant cells than group II AopW1, having a stronger actin filament disruption activity, and increased ability to reduce plant callose deposition. We demonstrate the role of some of these 14 amino acid positions in determining the phenotypic differences between the two versions of the effector. Moreover, cellular analyses revealed that in addition to the interaction with actin filaments, AopW1 is localized to the endoplasmic reticulum, chloroplasts, and early and recycling plant endosomes, with differences observed between the two AopW1 versions. Finally, we show that overexpression of the endosome-associated protein EHD1 that increases cellular recycling, attenuates the toxic effects exerted by AopW1 and increases defence responses. This study provides insights into the HopW1 family of bacterial effectors and their interactions with the plant cell and provides first evidence on the involvement of EHD1 in response to biotic stress. ### Competing Interest Statement The authors have declared no competing interest.