AUTHOR=Ibrahim George M. , Morin Paul M. 

TITLE=Salmonella Serotyping Using Whole Genome Sequencing

JOURNAL=Frontiers in Microbiology

VOLUME=Volume 9 - 2018

YEAR=2018

URL=https://www.frontiersin.org/journals/microbiology/articles/10.3389/fmicb.2018.02993

DOI=10.3389/fmicb.2018.02993

ISSN=1664-302X

ABSTRACT=Until recently, traditional serology and the Kauffmann White Scheme (KWS) have been the gold standard for Salmonella serotyping. Whole Genome Sequencing (WGS) has now emerged as an alternative in this field.

In this study, we report a retrospective analysis of our laboratory inventory of 1041 Salmonella isolates collected between 1999 and 2017. These isolates are of great public health concern since they all came from food, feed and environmental swabs. They were all serotyped by both traditional serology and WGS. Both SeqSero (WGS) and traditional serotyping predicted identical Salmonella serotypes in most isolates, 899 (86.4%). Agreement between antigenic expression and genetic sequence is not always correlated but in fact it agrees quite well with the KW scheme. Among the 899 isolates, 36 (3.4%) were only predicted as monophasic, diphasic or non-motile by traditional serology without any further speciation, while SeqSero analysis of these same isolates was exactly as in KWS. However, for 105 (10 %) of the isolates, the SeqSero analysis provided two possible serotypes sharing the same predicted antigenic profile (or formula) which is not entirely wrong by SeqSero since antigenic formula can result in more than one serotype, while traditional serotyping predicted only one serotype from each isolate. For 26 (2.4%) isolates, traditional serotyping determined all of them to be Salmonella Newport and although SeqSero also assigned the Salmonella Newport designation, it also noted that Salmonella Bardo shares the identical antigenic profile however exceedingly rare. Moreover, traditional serology predicted 38 (4.2%) isolates as Salmonella Typhimurium, same predictions as SeqSero, however, genomic analysis was able to further detect the absence of the 05 epitope in 16 (42%) of them. Lastly, SeqSero assignments differed from traditional serological testing in 80 (7.7%) isolates and no serotype prediction was ascertained from 62 (5.9%) Salmonella isolates. This ratio will decrease overtime as more whole genomes are added to the Salmonella national database (NCBI). 

Overall this study clearly demonstrates that using WGS as the sole method for Salmonella serotyping can provide numerous advantages that include more molecular and genetic details regarding the characteristics of the Salmonella isolates compared to traditional KWS serotyping.