Submitted to: National Center for Biotechnology Information (NCBI)
Publication Type: Other
Publication Acceptance Date: 2/16/2011
Publication Date: N/A
Technical Abstract: Previous research identified that the 5S ribosomal (rrn) gene and associated flanking sequences that are closely linked to the dkgB gene of Salmonella enterica were highly variable between serotypes, but not between subpopulations within the same serotype (PMID: 17005008). The degree of variability observed suggested that the region might be useful for linking genovar to serotype. Intergenic Sequence Ribotyping (ISR) was used to obtain sequence from a total of 140 strains of Salmonella enterica collected either as field isolates or from reference collections. The ISR region of interest was defined as beginning at the 1st base pair after the 23S rrn gene closest to dkgB and ending at the base pair preceding the sequence tag from S. enterica serovar Enteritidis tagttcagttggttagaatacctgcctgtc. ISR analysis includes base pairs 299304-299802 (499bp) of the S. enterica serovar Enteritidis reference genome NC_011294. The size of the ISR region varied from 257 to 530bp among S. enterica isolates. Historical data about the serotype as determined by the Kauffman-White O- and H-antigen schema was obtained from the provider and compared to ISR results. In addition, a commercially available AOAC approved DNA-array microtube assay approved for serotyping in Europe was included to further test isolates to confirm the relationship between ISR and serotype. Of 140 isolates examined, 6 (4.2%) showed no agreement between the three methods. However, both ISR and the AOAC DNA microarray identified that these strains were not reported in the public database, which suggests that each method detected unique strains. Evidence of mixed serotypes was recovered for 12 (8.6%) of isolates by ISR and DNA microarray genotyping when multiple colonies per strain were analyzed. Thus, mixtures of serotypes should be expected when results from different laboratories or tests do not agree with another. It is possible that mixtures of three or more serotypes occur and retrospective analysis of living culture does not always lead to recovery of all strains that had at one time been present during earlier steps of isolation. Both ISR and DNA microarray methods were superior to the Kauffman-White schema for assigning a genovar to isolates that lacked an O-antigen (a.k.a. rough strains). A total of 62 ISR sequences were detected, of which 28 are unique and not yet in the Salmonella enterica genome database. The 28 sequences will be deposited at NCBI in association with the release of this abstract. Comparison of costs for conducting 140 determinant assays by the three methods suggest that ISR was the least expensive ($840 at $6 each assay), Kauffman-White serotyping was intermediate ($4,900 at $35 each assay), and DNA microarray analysis was the most expensive ($7,000 at $50 each assay). The need to process multiple colonies from a single isolate in order to detect the presence of multiple isolates is facilitated by having an inexpensive assay such as ISR to confirm a serotype-associated genovar. An additional benefit is that analysis of a discrete region using universal primers should facilitate robotic applications and further reduce costs associated with serotyping of Salmonella.