Pathogenic yersiniae secrete about a dozen anti-host proteins, the Yops, by a pathway which does not involve cleavage of a classical signal peptide. The Yop secretory apparatus, called Ysc, for Yop secretion, is the archetype of type III secretion systems (which serve for the secretion of virulence proteins by several animal and plant pathogens) and is related to the flagellar assembly apparatus. The Yop secretion signal is N-terminal but has not been defined to date. Apart from the Ysc machinery, secretion of at least four Yops requires cytoplasmic proteins called Syc (for specific Yop chaperone). Each Syc protein binds to its cognate Yop. Unlike most cytoplasmic chaperones, these proteins do not have an ATP-binding domain, and are presumably devoid of ATPase activity. They share a few common properties: an acidic pl, a size in the range of 15-20 kDa, and a putative amphipathic alpha-helix in the C-terminal portion. They were recently shown to have counterparts in other pathogenic bacteria, where they appear to have a similar function.

Yersinia adhering at the surface of eukaryotic cells secrete a set of proteins called Yops. This secretion which occurs via a type III secretion pathway is immediately followed by the injection of some Yops into the cytosol of eukaryotic cells. Translocation of YopE and YopH across the eukaryotic cell membranes requires the presence of the translocators YopB and YopD. YopE and YopH are modular proteins composed of an N-terminal secretion signal, an internalization domain, and an effector domain. Secretion of YopE and YopH requires the presence of the specific cytosolic chaperones SycE and SycH, respectively. In this work, we have mapped the regions of YopE and YopH that are involved in binding of their cognate chaperone. There is only one Syc-binding domain in YopE (residues 15-50) and YopH (residues 20-70). This domain is localized immediately after the secretion signal and it corresponds to the internalization domain. Removal of this bifunctional domain did not affect secretion of YopE and YopH and even suppressed the need for the chaperone in the secretion process. Thus SycE and SycH are not secretion pilots. Instead, we propose that they prevent intrabacterial interaction of YopE and YopH with proteins involved in translocation of these Yops across eukaryotic cell membranes.

No abstract

Pathogenic yersiniae secrete antihost Yop proteins by a recently discovered secretion pathway which is also encountered in several animal and plant pathogens. The components of the export machinery are encoded by the virA (krA), virB (IcrB), and virC (IcrC) loci of the 70-kb pYV plasmid. In the present paper we describeyscU, the last gene of the virB locus. We determined the DNA sequence and mutated the gene on the pYV plasmid. After inactivation of yscU, the mutant strain was unable to secrete Yop proteins. The topology of YscU was investigated by the analysis of YscU-PhoA translational fusions generated by TnphoA transposition. This showed that the 40.3-kDa yscU product contains four transmembrane segments anchoring a large cytoplasmic carboxyl-terminal domain to the inner membrane. YscU is related to Spa4O from Shigelaflexneri, to SpaS from Salmonella typhimurium, to FlhB from Bacilus subtilis, and to HrpN from Pseudomonas solanacearum.

Bartonella henselae is the causative agent of cat scratch disease (CSD), which usually presents as a selflimiting lymphadenopathy. Occasionally, the bacteria will spread and be responsible for tissue and visceral involvement. Two B. henselae genotypes (genotypes I and II) have been described to be responsible for uncomplicated CSD on the basis of 16S rRNA sequence analysis. A type IV secretion system (T4SS) similar to the virulence-associated VirB system of Agrobacterium tumefaciens was recently identified in the B. henselae Houston-1 genotype I strain. We studied the correlations of the B. henselae genotypes with the clinical presentations and with the presence of T4SS. Isolates originated from CSD patients whose lymph nodes were prospectively analyzed. B. henselae genotype I was identified in 13 of 42 patients (30%). Among these, two teenage twins presented with hepatosplenic CSD and one immunocompetent adult presented with osteomyelitis. Genotype II was detected in 28 of 42 patients (67%), all of whom presented with uncomplicated CSD. The last patient was infected with both genotypes. T4SS was studied by PCR amplification of the virB4 gene. Amplification of virB4 codons 146 to 256, 273 to 357, and 480 to 537 enabled us to detect 66, 90, and 100% of the B. henselae isolates, respectively. Sequence analysis revealed sequence variations that correlated with genotype distribution. Our studies suggest that B. henselae genotype I strains harbor virB4 genes that are different from those harbored by genotype II strains and that genotype I strains might be more pathogenic.