Conventional typing of Australian Salmonella isolates
During the course of the project the chicken industry sent an average of 2,670 samples per annum (2000-2002 inclusive) strains to The Australian Salmonella Reference Centre in Adelaide for conventional typing. S. Sofia is still the predominant isolate from chickens. It first appeared in significant numbers in 1980 and steadily increased to the point where it currently represents around 50% of all Salmonella isolates from chicken and now appears to have stabilised at this level of isolation after some variation in previous years. There have been no significant changes in Salmonella serovar distribution or incidence in isolation from humans and food animals, particularly chickens, over the course of this project.

A major industry outbreak due to Salmonella Typhimurium phage type 126 occurred between May and November 2001. While commercially cooked products were not implicated, the outbreak once again focussed attention on raw chicken as a potential source of Salmonella into the community.

Over the course of the project, the phage type distribution within S. Typhimurium human and chicken isolates changed dramatically. Most notably, the previously important DT 64 is now a relatively minor isolate from both chickens and humans. In 2001 phage type 126 was implicated in a large outbreak and has become a frequently isolated phage type in both humans and chickens. This trend has continued into 2002.

In contrast to S. Typhimurium, S. Virchow phage type distribution did not markedly changed in the period 2000-2002 and, for both chickens and humans, phage type 8 has predominated, together with phage type 34. Although other phage types are seen, they are not consistently isolated from year to year.

Amplified fragment length polymorphism typing (AFLP)
The underlying principle of phage typing is the host specificity of bacteriophages in the typing panel.

Phage typing schemes for Salmonella enterica serovars are based on patterns of lysis produced by distinct phages isolated from a variety of sources. Phage typing has been used to subdivide isolates within serovars Typhi, Typhimurium, Enteritidis, Virchow, Hadar and Heidelberg.

Although phage typing is essential for the subdivision of Salmonella serovars, the method can prove inadequate for serovars in which a small number of phage types predominate. Molecular typing is used when conventional methods fail to give sufficient discrimination between isolates. Pulsed field gel electrophoresis (PFGE) has been in use for some considerable time and is the accepted gold standard adopted by organisations such as Centers for Disease Control in Atlanta. Newer methods of molecular typing such as amplified fragment length polymorphism (AFLP) may offer significant advantages in speed and discriminatory power over established methods such as PFGE. The potential use of this method as a regular typing method was therefore investigated in this project.

AFLP was used in conjunction both with PFGE and classical typing methods (i.e. serotyping and phage typing) to establish levels of discrimination between isolates, both individual strains and those suspected to belong to outbreaks. The most common serotype isolated from chickens, S. Sofia, was studied in this respect. It was found that S. Sofia was quite variable by AFLP, but no specific clone could be correlated to the time or place of isolation. It was concluded that there does not appear to be an Australian specific clone of S. Sofia that could account for its persistence in Australian chickens for such a long period of time.

S. Typhimurium DT126, the most common isolate from humans in 2001, was also examined by AFLP. The strains investigated ranged from epidemiologically unrelated strains to isolates associated with three outbreaks. Comparison with PFGE suggested that AFLP would provide greater discrimination between non-epidemiologically related isolates when compared to PFGE.

Significance of bacteriophages in S. Typhimurium
It has been observed for some time that certain phage types tend to predominate in S. Typhimurium in Australia. In 1998 work was begun to examine any temperate phages that could be carried by S.

Typhimurium, in particular phage type or definitive type (DT) 64 which was at the time this project commenced one of the most common phage types from S. Typhimurium in chickens. The carriage of lysogenic (or temperate) bacteriophages can strongly influence phage type designation. In 1998, in work undertaken by this group, two lysogenic phages were induced from S. Typhimurium DT64.

The induced phages were capable of mediating phage type conversion. The two phages have been designated ST64T and ST64B. An objective of this project was to determine whether genes carried by these phages might influence the virulence or distribution of the bacteria and this report describes the genetic characterisation of these two previously unknown temperate phages carried by S. Typhimurium DT 64.

It was demonstrated that ST64B is a defective phage, and could not mediate phage type conversion.

In contrast, phage type conversion could be mediated by phage ST64T and all converted phage types were shown to be lysogens of ST64T. Phage type conversion can have serious epidemiological consequences. In addition ST64T, like it’s close relative, P22 (another Salmonella phage) also mediates generalised transduction, which is a method of bacterial genetic exchange and has the potential to transfer genes between bacteria for traits like virulence and antibiotic resistance.

Although the ST64B and ST64B phages are carried by the same S. Typhimurium DT 64 isolate and are morphologically indistinguishable, they share very little DNA sequence similarity and this proves that both phages are distinct.

Southern hybridisation analysis using both phage genomes (ST64T and ST64B) as probes indicated that ST64B genome was found in most S. Typhimurium definitive phage types tested whereas ST64T genome was found only in DTs 64 and 29. Interestingly, when ST64B phage was used as a probe to screen other Salmonella enterica serovars, the results indicated that sequences hybridising to ST64B are found in a number of non-Typhimurium Salmonella serovars. Since temperate phage in Salmonella account for much of the genetic diversity between closely related Salmonella, the possibility exists that this diversity an be exploited to develop new, rapid typing systems.

Colonisation studies with S. Sofia
Salmonella Sofia currently represents 53.4% (2002 figures) of the approximately 2,500-3,000 Salmonella chickens isolates annually submitted to the IMVS for typing from the poultry industry.

However, it is almost never associated with human disease. Therefore S. Sofia is not a pathogen for humans or chickens but it is an effective coloniser of chickens. It has been demonstrated in a previous project that certain S. Sofia strains are better colonisers of chickens and appear to be able to persist in the chicken even when challenged with S. Typhimurium. This raises the possibility that S.

Sofia may be able to be used to decrease the level of colonisation of chickens with S. Typhimurium.

Consequently, this possibility was investigated further in the current study.

The molecular mechanism of S. Sofia colonisation was investigated. Filamentous fimbrial structures on the cell surface of bacteria can be essential for efficient colonisation of a host. The presence of fimbrial genes on the S. Sofia chromosome was confirmed by PCR and sequence analysis. The agfA gene appears to be universally found in S. Sofia isolates, where other fimbrial genes were less widespread. The agfA gene encodes the SEF17 fimbriae.

Studies were undertaken to determine what effect a mutation in the agfA gene has on the ability of S.

Sofia to colonise chickens. A mutant strain, K4, developed in a previous RIRDC project was used and its colonising ability tested in an experimental protocol also developed in that project. In these experiments, birds initially colonised with S. Sofia (day 0) were exposed by direct contact with S.

Typhimurium DT64 infected birds at day 5, rather than the usual protocol where in-contact exposure occurs at day 8. The rationale behind earlier exposure was that exposure to S. Typhimurium when S.

Sofia numbers were higher (as they often are early in colonisation) might prevent or lessen subsequent S. Typhimurium colonisation of the birds that were initially colonised with S. Sofia on day 0. The results obtained suggest that strain K4 is no worse at colonising chickens than the nonmutant parental strain. Earlier exposure (day 5) to S. Typhimurium DT64 infected birds had no effect on the colonisation of the previously S. Sofia colonised birds by S. Typhimurium DT64 than those exposed at day 8.

The more than two decade persistence of S. Sofia in Australian chickens as the predominant Salmonella isolate is a unique situation and not seen elsewhere in the world. Examination of S. Sofia strains isolated over 20 years from both here and overseas by AFLP indicated that there does not appear to be a specific dominant clone of this organism in Australia. It is also clear that S. Sofia can coexist in chickens with S. Typhimurium, but is unlikely to influence colonisation by that serovar.

Implications

• AFLP was found to be a useful typing tool but will not replace other established methods for typing, although in some cases it is clearly superior to established methods and should be retained.
• A molecular phage-typing scheme can be developed using genomic data gathered in this project, which will have the advantage of greater and less subjective discrimination and be more rapid.
• Additional work on S. Sofia is considered unlikely to demonstrate a significant potential for its role in displacing more pathogenic strains from the chicken production environment, and it is therefore felt that no further work in this area is warranted.

It is recommended that a molecular based typing system should be investigated to complement rather than replace classical phage typing. This would ensure continuity between the old system and the new.

Site map Contact us Feedback Links Disclaimer Privacy Statement

Last updated: 19 December 2004  Copyright RIRDC
http://www.rirdc.gov.au/reports/CME/04-138sum.html