Page Banner

United States Department of Agriculture

Agricultural Research Service

Research Project: Genetic Analysis of Poultry-Associated Salmonella enterica to Identify and Characterize Properties and Markers Associated with Egg-Borne Transmission of Illness

Location: Egg Safety and Quality

Title: Improving Food Safety by Understanding the Evolution of Egg-contaminating Salmonella Enteritidis

Author
item Guard, Jean

Submitted to: Meeting Proceedings
Publication Type: Proceedings
Publication Acceptance Date: March 11, 2011
Publication Date: April 6, 2011
Citation: Guard, J.Y. 2011. Improving Food Safety by Understanding the Evolution of Egg-contaminating Salmonella Enteritidis. In: Proceedings of XXX International Seminar of AMEVEA. April 6-8, 2011, Bogota, Columbia. 2011CDROM.

Technical Abstract: Improving Food Safety by Understanding the Evolution of Egg-contaminating Salmonella Enteritidis Jean Guard, Veterinary Medical Officer U. S. Department of Agriculture, Athens, GA USA (jean.guard@ars.usda.gov) The curious case of egg contamination by Salmonella enterica serovar Enteritidis S. Enteritidis is currently the world’s leading cause of human food-borne salmonellosis. Its emergence worldwide was well-documented in the early 1980s, but its first appearance was probably during the 1970s (6,7). It is the only Salmonella enterica serotype that routinely contaminates and survives in the internal contents of eggs produced by otherwise healthy hens (8). Other Salmonella enterica serotypes such as S. Pullorum and S. Gallinarum are well-documented for their ability to contaminate the internal contents of eggs, but they sicken birds and do not cause food borne disease in humans (2,9). Comparison of the two prevalent serotypes, S. Enteritidis and S. Typhimurium Egg-contaminating S. Enteritidis has retained a host range essentially as broad as that of S. Typhimurium (11,12). Together, these two serotypes cause approximately 40% of all human illnesses. Comparison of outer membrane properties revealed that strains of S. Enteritidis can produce a capsular-like lipopolysaccharide O-antigen region that mitigates signs of disease in hens (14,15). S. Enteritidis is especially capable of surviving within the internal contents of eggs, because it makes this capsule (13,16). S. Typhimurium does not appear to make this capsule with any efficiency. Playing poker with Salmonella A useful analogy for understanding that both large and small scale genetic events contribute to the ability of S. Enteritidis to contaminate eggs is to consider a deck of cards where constant reshuffling and cutting of the deck eventually results in one individual in a game drawing a combination of higher value. Several hands have the potential to win the game. Certain patterns of genetic inheritance add value over all the other hands in play. This situation accounts for the emergence of one serotype more than another. The winner appears to have linked together traits such as the ability to survive stressful conditions, to grow to high cell density and to access new niches (8). Both S. Enteritidis and S. Typhimurium have a group of genes that give these serotypes an advantage in the environment over the other 1400 serotypes belonging to subspecies I Salmonella enterica. However, S. Enteritidis developed the ability to contaminate eggs and thus it has access to an entirely new niche. Methods used for analysis of evolutionary trends can introduce bias S. Enteritidis appears to be an unusual case of evolution within the Salmonellae, because small scale genetic events are driving the biology which impacts human health. These events are deletions, and nucleotide substitutions that may or may not alter an open-reading frame. All of these events are referred to as single nucleotide polymorphisms (SNPs). For example, the largest SNP discovered in S. Enteritidis is a 215bp deletion that removed the N-terminus and some upstream region of the gene SEN4316. SNPs are not usually detected by DNA-DNA microarray hybridization (3). However, microarrays can be designed to detect SNPs (4,5). Conversely, a SNP that disrupts gene function can be missed by a DNA-DNA microarray even if it registers that the gene is present. Analysis of mRNA is thus a necessary supporting assay to assess which genes are functional. The problem with analysis of mRNA is that gene expression is dependent upon environmental conditions. It is quite possible that a gene could have a very important role within a specific environment that is absent in the conditions for analysis chosen by an investigator. Again, this situation results in a false interpretation. All methods have a limit of detection and it is important to combine methods to avoid bias. At least two approaches are required to document broader evolutionary trends, but one assay can be chosen over another to achieve specific purposes. S. Enteritidis provides a window into the world of small scale evolution Wherever it is isolated from around the world, S. Enteritidis generates little genomic variation in comparison to other serotypes (18,19). However, animal infection trials, cell invasion assays, survivability assays and other biological approaches consistently report that there are significant differences in the ability of any one S. Enteritidis strain to produce distinctive outer membrane structures and to grow, invade, colonize and survive within the internal contents of eggs (8). A turning point in understanding evolution of S. Enteritidis was when it became evident variation occurring within a single bacteriophage type (PT) appeared to be as great as that occurring between serotypes. In other words, phage type was not strongly linked to genomic determinants of egg contamination. It means that two clearly different strains isolated from within the same phage type lineage could have all of the evolution consequential for egg contamination. Thus, a comparison of whole genomes from strains within a single PT of S. Enteritidis became a useful approach for finding the minimal number of high consequence genomic differences that could be associated with evolution of egg contamination. Concentrating on finding all SNPs that differentiated two phenotypic variants within a single PT lineage would increase the likelihood of finding those of consequence, because background genomic noise associated with random evolution would be silenced as much as possible. It was important to reduce genetic noise, because analysis of known SNPs across serotypes showed that genetic noise was prevalent (1). Conversely, phylogenetic analysis and complementary phenotype microarrays also confirmed that there were true clones in circulation, even if they did not come from the same host or geographic region (21). That which had escaped detection is finally detected Two PT13a strains that varied in their ability to contaminate eggs and to form biofilm were mutationally mapped and compared to the Sanger Institute reference PT4 S. Enteritidis strain P125109 (GenBank AM933172). Strain PT13a 21027 formed biofilm, but did not contaminate eggs (EC-BF+). In contrast, PT13a 21046 contaminated eggs, but did not form biofilm (EC+BF-). The PT4 template strain, PT4 22079, contaminated eggs and formed biofilm (EC+BF+). A total of 247 evolutionary events were detected that distinguished the two PT13a strains from each other. The completed database of evolutionary events is available within a public in NCBI dbSNP Build 131 (http://www.ncbi.nlm.nih.gov/genomes/static/Salmonella_SNPS.html). None of these events that differentiated the genomes of the two PT13a strains were detectable by DNA-DNA hybridization microarray. Two ribosomal gene SNPs had been previously detected by pulsed field gel electrophoresis (PFGE) and SNPs present within cyaA served as internal controls to assess the sensitivity of methodology. Of all the events, 82 (33.2%) were synonymous and thus did not alter predicted amino acid sequences of proteins, 38 (15.4%) occurred in intergenic regions, 99 (40.1%) introduced amino acid substitutions, 12 (4.9%) were in ribosomal genes and 15 (6.1%) SNPs were small deletions or early terminations in genes. The largest event was a 215 bp deletion, but most events were non-contiguous single nucleotide substitutions. PT13a 21046 (BF-EC+) had 3 gene mutations disrupting ORFs that differentiated it from the reference PT4 strain (BF+EC+); conversely, PT13a 21027 (BF+EC-) had 12 unique mutations that disrupted coding regions and differentiated it from the reference PT4 strain. Approximately 250 SNPs are thus linked to phenotypic differentiation occurring within phage type lineage and some of these are clearly of more consequence than others. These data provide a basis for

Last Modified: 11/28/2014
Footer Content Back to Top of Page