Bacterial Pathogens in Plants: Life up against the Wall.

Higher plants contain potentially vast sources of nutrients for the myriad bacterial species in their environment, and most bacteria are small enough to pass through stomates and other natural openingsinto the apoplast-the anteroom for these riches. However, surprisingly few bacteria raid the nutrient stores of living plant cells, apparently because the metabolic intimacy involved in parasitism requires the work of specialists. Of these specialists, some in the Rhizobiaceae produce hypertrophies that are genetically engineered or developmentally tricked into providing an undefended, nutritive niche in root cortical tissues and rhizospheres (see Long, 1996; Sheng and Citovsky, 1996, in this issue), whereas others, mostly Gramnegative bacteria in the Pseudomonadaceae and Enterobacteriaceae, specialize in colonizing the apoplast. It is the apoplastic colonizers that are the common pathogens that produce the rots, spots, wilts, cankers, and blights afflicting virtually all crop plants, and their relationship with the host is defined by two features. They spend their parasitic life up against the wall of plant cells, in the intercellular spaces of various plant organs or in the xylem, and they are necrogenic-able to cause the death of plant cells. Their ability to multiply and then sooner or later to kill plant cells depends on secreted enzymes that degrade the wall or on molecules that pass through it. This review addresses our progress in understanding this molecular traffic and how it may enable necrogenic bacterial pathogens to colonize the apoplast. The present picture of pathogenesis has been strongly determined by three developments. The first was the discovery that bacteria elicited the defense-associated hypersensitive response (HR) in plants during incompatible interactions. The HR was first observed as a rapid localized collapse of tobacco leaf tissue after infiltration of high numbers of bacterial pathogens that are host specific for other plant species (Klement, 1963; Klement et ai., 1964). Because the ability to elicit the HR is a unique attribute of the necrogenic pathogens and these bacteria can avoid or suppress its elicitation in their hosts, the HR phenomenon appears central to bacterial pathogenicity and host specificity and has attracted much attention (Klement, 1982; Goodman and Novacky, 1994; see also Dangl et al., 1996; Hammond-Kosack and Jones, 1996, in this issue). The second development was the application of the molecular tools