Role of Proteins Secreted by the Hrp-Pathways of Erwinia stewartii and E. herbicola pv. gypsophilae in Eliciting Water-Soaking Symptoms and Initiating Galls

Many bacterial pathogens of plants can inject pathogenicity proteins into host cells using a specialized type III secretion system encoded by hrpgenes. This system deliver effector proteins, into plant cells that function in both susceptible and resistant interactions. We have found that the virulence of Erwinia stewartii(Es; syn. Pantoea stewartii) and Erwinia herbicola pv. gypsophilae (Ehg, syn. Pantoea agglomerans), which cause Stewart's wilt of corn and galls on Gypsophila, respectively, depends on hrpgenes. The major objectives of this project were: To increase expression of hrpgenes in order to identify secreted proteins; to identify genes for proteins secreted by the type-III systems and determine if they are required for pathogenicity; and to determine if the secreted proteins can function within eukaryotic cells.   We found that transcription of the hrp and effector genes in Es and Ehg is controlled by at least four genes that constitute a regulatory cascade. Environmental and/or physiological signaling appears to be mediated by the HrpX/HrpY two component system, with HrpX functioning as a sensor-kinase and HrpY as a response regulator. HrpYupregulateshrpS, which encodes a transcriptional enhancer. HrpS then activates hrpL, which encodes an alternate sigma factor that recognizes "hrp boxes". All of the regulatory genes are essential for pathogenicity, except HrpX, which appears only to be required for induction of the HR in tobacco by Es. In elucidating this regulatory pathway in both species, we made a number of significant new discoveries. HrpX is unusual for a sensor-kinase because it is cytoplasmic and contains PAS domains, which may sense the redox state of the bacterium. In Es, a novel methyl-accepting protein may function upstream of hrpY and repress hrp gene expression in planta. The esaIR quorum sensing system in Es represses hrp gene expression in Es in response to cell-density.   We have discovered six new type III effector proteins in these species, one of which (DspE in Ehg and WtsE in Es) is common to both pathogens. In addition, Es wtsG, which is a homolog of an avrPpiB from P. syringae pv. pisi, and an Ehg ORF, which is a homolog of P. syringae pv. phaseolicola AvrPphD, were both demonstrated to encode virulence proteins. Two plasmidborne, Ehg Hop proteins, HsvG and PthG, are required for infection of gypsophilia, but interestingly, PthG also acts as an Avr elicitor in beets. Using a calmodulin-dependent adenylate cyclase (cyaA) reporter gene, we were successful in demonstrating that an HsvG-CyaA fusion protein can be transferred into human HeLa cells by the type-III system of enteropathogenic E. coli. This is a highly significant accomplishment because it is the first direct demonstration that an effector protein from a plant pathogenic bacterium is capable of being translocated into a eukaryotic cell by a type-III secretion system.   Ehg is considered a limiting factor in Gypsophila production in Israel and Stewart’s Wilt is a serious disease in the Eastern and North Central USA, especially on sweet corn in epidemic years. We believe that our basic research on the characterization of type III virulence effectors should enable future identification of their receptors in plant cells. This may lead to novel approaches for genetically engineering resistant plants by modifying their receptors or inactivating effectors and thus blocking the induction of the susceptible response. Alternatively, hrp gene regulation might also provide a target for plant produced compounds that interfere with recognition of the host by the pathogen. Such strategies would be broadly applicable to a wide range of serious bacterial diseases on many crops throughout the USA and Israel.