- Volume 149, Issue 5, 2003

The type III secretion system (TTSS) encoded by the Salmonella pathogenicity island 2 (SPI-2) is required for bacterial replication inside macrophages and for systemic infection in mice. Many TTSS secreted proteins, including effectors and components of the translocon, require chaperones which promote their stability, prevent their premature interactions or facilitate their secretion. In this study, the function of the first gene (sseA) of one of the SPI-2 operons (sseA–G) was investigated. This operon includes genes that encode translocon components (SseB, SseC and SseD), translocated proteins (SseF and SseG) and putative chaperones (SscA and SscB). sseA encodes a 12·5 kDa protein with a C-terminal region with the potential to form a coiled-coil structure, but no sequence similarity to other proteins. Mutation of sseA results in severe virulence attenuation and an intracellular replication defect. It is shown here that SseA is not a secreted protein, but is required for SPI-2-dependent translocation of two effector proteins (SifA and PipB). Furthermore, the translocon components SseB and SseD were not detected in an sseA mutant strain. By using a yeast two-hybrid assay and column binding experiments, it is demonstrated that SseA interacts directly with SseB and SseD. These results indicate that SseA is a chaperone for SseB and SseD. The inability of an sseA mutant to assemble the SPI-2 TTSS translocon accounts for its high level of virulence attenuation in vivo. To the authors' knowledge, this is the first chaperone described for the SPI-2 TTSS.

The plasmid-borne Rhizobium etli katG gene encodes a dual-function catalase-peroxidase (KatG) (EC 1.11.1.7) that is inducible and heat-labile. In contrast to other rhizobia, katG was shown to be solely responsible for catalase and peroxidase activity in R. etli. An R. etli mutant that did not express catalase activity exhibited increased sensitivity to hydrogen peroxide (H2O2). Pre-exposure to a sublethal concentration of H2O2 allowed R. etli to adapt and survive subsequent exposure to higher concentrations of H2O2. Based on a multiple sequence alignment with other catalase-peroxidases, it was found that the catalytic domains of the R. etli KatG protein had three large insertions, two of which were typical of KatG proteins. Like the katG gene of Escherichia coli, the R. etli katG gene was induced by H2O2 and was important in sustaining the exponential growth rate. In R. etli, KatG catalase-peroxidase activity is induced eightfold in minimal medium during stationary phase. It was shown that KatG catalase-peroxidase is not essential for nodulation and nitrogen fixation in symbiosis with Phaseolus vulgaris, although bacteroid proteome analysis indicated an alternative compensatory mechanism for the oxidative protection of R. etli in symbiosis. Next to, and divergently transcribed from the catalase promoter, an ORF encoding the regulator OxyR was found; this is the first plasmid-encoded oxyR gene described so far. Additionally, the katG promoter region contained sequence motifs characteristic of OxyR binding sites, suggesting a possible regulatory mechanism for katG expression.

Pseudomonas syringae produces the exopolysaccharide alginate, a copolymer of mannuronic and guluronic acid. Although alginate has been isolated from plants infected by P. syringae, the signals and timing of alginate gene expression in planta have not been described. In this study, an algD : : uidA transcriptional fusion, designated pDCalgDP, was constructed and used to monitor alginate gene expression in host and non-host plants inoculated with P. syringae pv. tomato DC3000. When leaves of susceptible collard plants were spray-inoculated with DC3000(pDCalgDP), algD was activated within 72 h post-inoculation (p.i.) and was associated with the development of water-soaked lesions. In leaves of the susceptible tomato cv. Rio Grande-PtoS, algD activity was lower than in collard and was not associated with water-soaking. The expression of algD was also monitored in leaves of tomato cv. Rio Grande-PtoR, which is resistant to P. syringae pv. tomato DC3000. Within 12 h p.i., a microscopic hypersensitive response (micro-HR) was observed in Rio Grande-PtoR leaves spray-inoculated with P. syringae pv. tomato DC3000(pDCalgDP). As the HR progressed, histochemical staining indicated that individual bacterial cells on the surface of resistant tomato leaves were expressing algD. These results indicate that algD is expressed in both susceptible (e.g. collard, tomato) and resistant (Rio Grande-PtoR) host plants. The expression of algD in an incompatible host–pathogen interaction was further explored by monitoring transcriptional activity in leaves of tobacco, which is not a host for P. syringae pv. tomato. In tobacco inoculated with DC3000(pDCalgDP), an HR was evident within 12 h p.i., and algD expression was evident within 8-12 h p.i. However, when tobacco was inoculated with an hrcC mutant of DC3000, the HR did not occur and algD expression was substantially lower. These results suggest that signals that precede the HR may stimulate alginate gene expression in P. syringae. Histochemical staining with nitro blue tetrazolium indicated that the superoxide anion () is a signal for algD activation in planta. This study indicates that algD is expressed when P. syringae attempts to colonize both susceptible and resistant plant hosts.

A three-dimensional computer model was used to evaluate the effect of chemically mediated detachment on biofilm development in a negligible-shear environment. The model, BacLAB, combines conventional diffusion-reaction equations for chemicals with a cellular automata algorithm to simulate bacterial growth, movement and detachment. BacLAB simulates the life cycle of a bacterial biofilm from its initial colonization of a surface to the development of a mature biofilm with cell areal densities comparable to those in the laboratory. A base model founded on well established transport equations that are easily adaptable to investigate conjectures at the biological level has been created. In this study, the conjecture of a detachment mechanism involving a bacterially produced chemical detachment factor in which high local concentrations of this detachment factor cause the bacteria to detach from the biofilm was examined. The results show that the often observed ‘mushroom’-shaped structure can occur if detachment events create voids so that the remaining attached cells look like mushrooms.