Early pathogenesis of Yersinia ruckeri infections in rainbow trout (Oncorhynchus mykiss, Walbaum)

Yersinia ruckeri is the causative agent of yersiniosis or enteric redmouth disease (ERM) and causes significant losses in salmonid aquaculture worldwide. Although infection with this agent has been reported in other fish species, salmonids and especially rainbow trout are most susceptible to ERM. Vaccination of fish with a commercial Y. ruckeri bacterin provides good levels of protection, however, disease outbreaks do occur from time to time. The understanding of how this pathogen causes disease is crucial in order to develop new preventive and therapeutic approaches. The aim of this thesis was to study the early pathogenesis of Y. ruckeri infections, with emphasis on the identification of the portal of entry and the investigation of different traits in vitro that have been associated with bacterial virulence. The General Introduction gives an overview of the micro-organism, Y. ruckeri, and the disease it causes. From this review it is clear that only little information is available about the pathogenesis of Y. ruckeri. Only recently, the mechanisms by which the pathogen causes disease have started to be unravelled. In the first experimental study, the portal of entry of Y. ruckeri and its tissue distribution at different time intervals were investigated using an immersion infection model in rainbow trout. This model strongly mimics the natural way of infection and therefore, is suitable to study infectious diseases in fish. Additional experimental infections were carried out to determine whether a different tissue distribution is related to differences in virulence. Bacteriological and histological examination revealed the presence of high numbers of bacteria in the gills immediately after infection. Many Y. ruckeri were found to be attached to the gill mucus and some were detected within the gill capillaries between 0 and 2.5 h post infection (p.i.). This indicates that bacteria first adhere to gill mucus and thereafter invade the branchial vasculature leading to colonization of the internal organs. These findings show that the gills are an important portal of entry. Nevertheless, only a virulent strain was able to survive and multiply in the host, causing septicaemia and death several days after infection whereas exposure to avirulent strains resulted in a complete removal of bacteria from the internal organs. This could indicate that immune evasion is a major virulence property of Y. ruckeri. In the second experimental study, the early interactions of virulent and avirulent Y. ruckeri strains with the gill and gut tissue of rainbow trout were determined using a standardized gill and gut perfusion model. For this purpose, perfused gill and gut explants were inoculated for 1 h with Y. ruckeri and thereafter, processed for bacteriological, histological and immunohistochemical examination for demonstration of bacterial adherence. For the first time, both perfusion models were also used to study bacterial invasion. Therefore, the perfusion fluid at the efferent gill arch during gill perfusion and the organ bath fluid following gut perfusion were examined bacteriologically. A non-invasive Escherichia coli strain was included as negative control for invasion. In both the gill and gut perfusion models, virulent and avirulent Y. ruckeri strains highly adhered to gill and gut tissue and significant differences were not observed. Bacteria were seen in the mucus, associated with the epithelium and invaded in the lamina propria of the gills. Y. ruckeri was also observed in the crypts, attached to the villi and within the epithelial layer of the gut. Few bacteria were also detected in the intestinal lamina propria mucosae. Our results showed that virulent and avirulent Y. ruckeri invaded the gill and gut whereas a non-invasive E. coli strain was not isolated from the perfusion fluid and organ bath fluid in the gill and gut perfusion model, respectively. These findings indicate that Y. ruckeri is able to colonize and invade the gills and intestine of rainbow trout. Significant differences in adherence and invasive capacity were not observed between virulent and avirulent strains. In the third experimental study, different traits that have been associated with bacterial virulence were compared between virulent and avirulent Y. ruckeri strains, including adhesion to mucus and cell lines, invasion and intracellular survival in cell lines and serum resistance. The adhesion and invasion mechanisms were further characterized for selected Y. ruckeri strains. The virulent isolates highly adhered to isolated gill and gut mucus of rainbow trout, whereas the majority of the avirulent strains demonstrated significantly lower adhesion. A decrease in adherence capability following bacterial treatment with sodium metaperiodate and proteolytic enzymes suggested that carbohydrates as well as proteins play a role and consequently, lectins may be involved in the adhesion. All Y. ruckeri strains were able to adhere to and invade CHSE-214, FHM and R1 cells. One avirulent strain showed high adherence and invasive capacity in the three cell lines, whereas the virulent strains were moderately adhesive and invasive. The internalization of several isolates was inhibited by colchicine and cytochalasin-D, suggesting the involvement of microtubules and microfilaments. Monodansylcadaverine never inhibited Y. ruckeri internalization and therefore, the formation of clathrin-coated pits in receptor-mediated endocytosis may not play a role. For all strains, intracellular survival assays showed a decrease of viable bacteria in the cells 6 h after inoculation. This suggests that Y. ruckeri is not able to multiply or survive inside cultured cells or remains intracellularly for only a short period. Analysis of the susceptibility to the bactericidal effect of rainbow trout serum demonstrated that virulent strains were serum resistant, whereas avirulent strains were generally serum sensitive. Taken together, adherence to mucus and resistance to the bactericidal effect of serum seems to be correlated with in vivo virulence. In the General Discussion, the results of the three experimental studies are discussed and hypothetical virulence mechanisms are proposed in relation to the known virulence factors of Y. ruckeri. Finally, perspectives are given on how these findings may contribute to the prevention of ERM. In conclusion, the gills and gut were identified as possible important portals of entry for Y. ruckeri in rainbow trout. This pathogen is shown to be able to adhere to and invade in non-phagocytic cells. The adherence of the pathogen seems to involve bacterial lectins that bind specific carbohydrate structures on host surfaces. Invasion appears to be a microtubule and microfilament dependent process. Virulence of Y. ruckeri isolates is probably a complex combination of adhesion and colonization capacity, resistance to the bactericidal effect of serum and other factors which still needs to be determined.