Salmonella in the tropical household environment--Everyday, everywhere.

OBJECTIVES: To determine the prevalence of Salmonella in the environment of case and control houses, and compare serovars isolated from cases and their houses. METHODS: From 2005 to 2008, we tested samples from houses of 0-4 year old cases and community controls in Darwin and Palmerston for Salmonella. Case isolates were compared with environmental isolates. S. Ball and S. Urbana isolates were compared using Multiple Amplification of Phage Locus Typing (MAPLT) and Multiple-Locus Variable number of tandem repeat Analysis (MLVA). RESULTS: Salmonella were found in 47/65 (72%) case houses and 18/29 (62%) control houses; these proportions were not significantly different. In 21/47 (45%) houses, case and environmental isolates (from animal faeces, soil and vacuums) were indistinguishable. Multiple serovars were isolated from 20 (31%) case and 6 (21%) control houses. All but one environmental isolate are known human pathogens in the Northern Territory (NT). Each of the four pairs of S. Ball and S. Urbana were indistinguishable. CONCLUSIONS: Animal faeces were the most likely source of salmonellosis in cases. The similar prevalence of house isolates suggests that Salmonella is ubiquitous in this environment. The distinction of S. Ball and S. Urbana subtypes enabled linkage of human illness to environmental exposure. Environmental contamination with Salmonella is an important source of sporadic infection in children in the tropics.

Pathogenicity of Salmonella strains isolated from egg shells and the layer farm environment in australia.

In Australia, the egg industry is periodically implicated during outbreaks of Salmonella food poisoning. Salmonella enterica serovar Typhimurium and other nontyphoidal Salmonella spp., in particular, are a major concern for Australian public health. Several definitive types of Salmonella Typhimurium strains, but primarily Salmonella Typhimurium definitive type 9 (DT9), have been frequently reported during egg-related food poisoning outbreaks in Australia. The aim of the present study was to generate a pathogenicity profile of nontyphoidal Salmonella isolates obtained from Australian egg farms. To achieve this, we assessed the capacity of Salmonella isolates to cause gastrointestinal disease using both in vitro and in vivo model systems. Data from in vitro experiments demonstrated that the invasion capacity of Salmonella serovars cultured to stationary phase (liquid phase) in LB medium was between 90- and 300-fold higher than bacterial suspensions in normal saline (cultured in solid phase). During the in vivo infection trial, clinical signs of infection and mortality were observed only for mice infected with either 10(3) or 10(5) CFU of S. Typhimurium DT9. No mortality was observed for mice infected with Salmonella serovars with medium or low invasive capacity in Caco-2 cells. Pathogenicity gene profiles were also generated for all serovars included in this study. The majority of serovars tested were positive for selected virulence genes. No relationship between the presence or absence of virulence genes by PCR and either in vitro invasive capacity or in vivo pathogenicity was detected. Our data expand the knowledge of strain-to-strain variation in the pathogenicity of Australian egg industry-related Salmonella spp.

MLVA and phage typing as complementary tools in the epidemiological investigation of Salmonella enterica serovar Typhimurium clusters.

In South Australia serotyping and phage typing are employed for routine Salmonella surveillance. Molecular techniques such as Multiple-locus variable number tandem repeat analysis (MLVA) are increasingly utilized to aid outbreak investigations. During 2007 three Salmonella enterica serovar Typhimurium outbreaks involving phage types DT9, DT29, and DT44 were investigated. Human, food and environmental isolates were also typed by MLVA. In the DT9 outbreak cluster MLVA demonstrated distinct groupings that corresponded to epidemiological differences in time, place, and descriptive information on potential transmission mechanisms. In contrast, the human and food isolates of both the DT29 and DT44 clusters had identical MLVA profiles for all but one case. These data correlated with the epidemiology suggesting that these isolates were closely related and probably a single agent. These findings illustrate that phage typing and MLVA can provide different but complementary information for epidemiological investigations of Salmonella outbreaks.