Background

Background Recently there has been considerable public debate about the role of foodborne pathogen carriage in live cattle and antibiotic use in livestock production on the safety of foods of bovine origin, specifically in ground beef, and their effects on multi drug-resistant (MDR) Salmonella infections in humans. Although it is recognized that foods of bovine origin can serve as a vehicle for Salmonella, and particularly MDR-Salmonella, the frequency of human diseases outbreaks caused by  the most commonly isolated Salmonella serotypes present in cattle and ground beef are highly disproportionate to the serotypic distribution of Salmonella in beef and live animals destine for slaughter (see table).

Our research team and others have demonstrated that live animals serve as a reservoir for a genetically diverse population of foodborne pathogens, but only a fraction of the genetic lineages of foodborne pathogens present in live animals actually spill over into the human population as causes of disease. It is also likely that a similar phenomenon occurs with Salmonella, with certain genotypes being more likely to cause human disease than others. For example, among S. Newport isolates there appear to be two distinct genetic lineages (S. Newport I and II). In a previous study, 89% of S. Newports isolated from humans were of lineage I, whereas only 40% of the S. Newports of animal origin were of Lineage I. It is possible that the true proportion of Lineage I S. Newports among cattle is far less than this because the isolates tested in Beltran et al’s study were from ill animals. Although most of the “bovine-origin Salmonella serotypes” have been reported at least once as a cause of human disease, their occurrence is rare and is usually associated with high risk populations such as infants, neonates or the elderly.

Human virulence of Salmonella is thought to be mediated primarily by the presence of virulence genes located within Salmonella pathogenicity islands (SPIs) and on virulence plasmids (van Asten et al., 2005). Several researchers have explored the distribution differences in virulence genes among pathogenic serovars of Salmonella, but to date there has been little information between pathogenic strains of Salmonella and those thought to be less pathogenic to humans. Given the presence of Salmonella in live animals, if cattle (or beef) are an important source of these serotypes, one would expect far more cases in humans than is reported. Having a genetic marker of human disease-causing potential among Salmonella, knowledge of the requirements/possibility of Salmonella serotypes of “low” virulence to acquire sufficient genes to become a strain of “higher” human virulence, and which host (Salmonella) factors are more closely associated with the acquisition of these specific genetic elements will provide valuable tools for investigating the epidemiology and emergence of Salmonella in the food production environment.

The stated objectives for this work were:

• To determine patterns of sequence variability (absence of specific genes) among Salmonella frequently isolated from cattle (but infrequently associated with human diseases) and the genetic sequences of highly pathogenic strains of Salmonella.

Methodology

State-of-the-art genomic resequencing using microarrays were used to determine the entire genomic sequence of Salmonella enterica serovars Mbandaka, Anatum, Montevideo. Salmonella enterica MDR-Newport and Typhimurium were used as positive control isolates. Briefly, a whole genome tiling microarray consisting of entire Salmonella enterica Typhimurium LT2 genome (GenBank NC_003197) was used.

Findings

Salmonella isolates of bovine origin were obtained colleages at the USDA MARC in Nebraska and Washington State University. Genomic DNA from the five Salmonella serovars (Mbandaka, Anatum, Montevideo, Newport, and Typhimurium) was extracted and purified. A gene chip represents 190,000, 24-mer probes covering 4,527 genes (with a minimum of 20 probes per gene) of Salmonella Typhimurium LT2 was synthesized and hybridized with the extracted DNA. From the hybridization data, the locations of SNPs, insertions, or deletions in the genomes in question were identified. Specific SNPs and insertions were then re-sequenced to identify novel genetic regions not present in the control DNA. Gene chip development and hypbrization was performed by NimbleGen Inc. SNPs, insertions, and deletions were identified using the SignalMap™ data browser software (NibleGen). Because of the enormous amount of data available, additional addendum to  this report will be submitted.

Implications

Many of the most frequently isolated types of Salmonella found in cattle and ground beef rarely show up as causes of human disease. The reason for this is unknown, but one of the most plausible explanations is that all Salmonella are not equally capable of causing disease. In other words, some organisms, although they are classified as “salmonella” pose little threat to health individual- “sheep in wolves’ clothing.” By comparing the DNA of Salmonella isolates from humans, cattle, and beef, it will be possible to develop specific genetic markers for Salmonella and predict which organisms should receive the most attention.