Edwardsiella tarda is an important Gram-negative enteric pathogen affecting both animals and humans. It possesses a type III secretion system (T3SS) essential for pathogenesis. EseB, EseC and EseD have been shown to form a translocon complex after secretion, while EscC functions as a T3SS chaperone for EseB and EseD. In this paper we identify EscA, a protein required for accumulation and proper secretion of another translocon component, EseC. The escA gene is located upstream of eseC and the EscA protein has the characteristics of T3SS chaperones. Cell fractionation experiments indicated that EscA is located in the cytoplasm and on the cytoplasmic membrane. Mutation with in-frame deletion of escA greatly decreased the secretion of EseC, while complementation of escA restored the wild-type secretion phenotype. The stabilization and accumulation of EseC in the cytoplasm were also affected in the absence of EscA. Mutation of escA did not affect the transcription of eseC but reduced the accumulation level of EseC as measured by using an EseC-LacZ fusion protein in Ed. tarda. Co-purification and coimmunoprecipitation studies demonstrated a specific interaction between EscA and EseC. Further analysis showed that residues 31-137 of EseC are required for EseC-EscA interaction. Mutation of EseC residues 31-137 reduced the secretion and accumulation of EseC in Ed. tarda. Finally, infection experiments showed that mutations of EscA and residues 31-137 of EseC increased the LD 50 by approximately 10-fold in blue gourami fish. These results indicated that EscA functions as a specific chaperone for EseC and contributes to the virulence of Ed. tarda.

No abstract

With a uniform linear antenna array, a new direction-of-arrival (DOA) estimation method is proposed for wideband coherent signals in the presence of unknown correlated noise but with structured covariance matrix. Based on this proposed structure, i.e. Hermitian Toeplitz, a spatial differencing operation that exploits this symmetry is applied to remove the effect of the unknown noise and a new matrix is constructed accordingly at each frequency bin. Following this step, a focusing operation is performed to give the corresponding aligned covariance matrix. Finally, an eigenstructure-based DOA estimation method is applied. The validity of the method is supported by numerical simulation under various conditions.