We conclude that VPEgamma is a caspase-like enzyme that has been recruited in plants to regulate vacuole-mediated cell dismantling during cell death, a process that has significant influence in the outcome of a diverse set of plant-pathogen interactions.

Plants are unique in their ability to store proteins in specialized protein storage vacuoles (PSVs) within seeds and vegetative tissues. Although plants use PSV proteins during germination, before photosynthesis is fully functional, the roles of PSVs in adult vegetative tissues are not understood. Trafficking pathways to PSVs and lytic vacuoles appear to be distinct. Lytic vacuoles are analogous evolutionarily to yeast and mammalian lysosomes. However, it is unclear whether trafficking to PSVs has any analogy to pathways in yeast or mammals, nor is PSV ultrastructure known in Arabidopsis vegetative tissue. Therefore, alternative approaches are required to identify components of this pathway. Here, we show that an Arabidopsis thaliana mutant that disrupts PSV trafficking identified TERMINAL FLOWER 1 (TFL1), a shoot meristem identity gene. The tfl1-19/mtv5 (for ''modified traffic to the vacuole'') mutant is specifically defective in trafficking of proteins to the PSV. TFL1 localizes to endomembrane compartments and colocalizes with the putative ␦-subunit of the AP-3 adapter complex. Our results suggest a developmental role for the PSV in vegetative tissues.protein trafficking ͉ vegetative tissue ͉ adaptin

Vacuoles are very prominent compartments within plant cells, and understanding of their function relies on knowledge of their content. Here, we present a simple vacuole purification protocol that was successfully used for large-scale isolation of vacuoles, free of significant contamination from other endomembrane compartments. This method is based on osmotic and thermal disruption of mesophyl-derived Arabidopsis protoplasts, followed by a density gradient fractionation of the cellular content. The whole procedure, including protoplast isolation, takes approximately 6 h. INTRODUCTIONLarge-scale isolation and characterization of cellular organelles offers excellent opportunities in plant systems biology 1 . Detailed qualitative and quantitative knowledge of the composition of a particular organelle allows for investigation of subcellular pathways through the identification of novel protein components and the analysis of cellular sinks. This also facilitates detailed comparative studies between species or between mutant and wild-type plants. In contrast to some organelles, such as mitochondria and chloroplasts, that are relatively easy to obtain in a pure form 2,3 , plant vacuoles are extremely fragile, and their isolation using conventional tissue homogenization and fractionation schemes can be remarkably challenging and often results in a subcellular fraction without the same properties as intact mature vacuoles 4 . The first successful technique to overcome these difficulties was based on protoplast isolation and subsequent gentle release of vacuoles by osmotic shock 5,6 . Later large-scale isolation protocols employed gradient fractionation of osmotically released organelles 7,8 , and the method was further optimized by application of both osmotic and thermal shocks to the isolated protoplasts 9,10 . As an alternative to the protoplast methods, which are applicable only to relatively soft tissues, vacuoles can be isolated by slicing the plant material and releasing contents into a medium with a perfectly adjusted osmoticum 11 . In subsequent studies, numerous vacuolar constituents have been determined, including sugars and amino acids 12 , hydrolases 13 and ions 14 .

No abstract