Where are we heading with Salmonella molecular subtyping?

Abstract

Future MicrobiologyVol. 8, No. 10 EditorialFree AccessWhere are we heading with Salmonella molecular subtyping?Nikki Shariat & Edward G DudleyNikki ShariatDepartment of Food Science, The Pennsylvania State University, University Park, PA 16802, USASearch for more papers by this author & Edward G Dudley* Author for correspondenceDepartment of Food Science, The Pennsylvania State University, University Park, PA 16802, USA. Search for more papers by this authorEmail the corresponding author at egd100@psu.eduPublished Online:24 Sep 2013https://doi.org/10.2217/fmb.13.107AboutSectionsPDF/EPUB ToolsAdd to favoritesDownload CitationsTrack CitationsPermissionsReprints ShareShare onFacebookTwitterLinkedInReddit Keywords: bacterial pathogenCRISPRfoodborne outbreakgenome sequencingmolecular epidemiologymolecular subtypingPFGESalmonellaWe stop your regularly scheduled Editorials to bring you breaking news of a Salmonella enterica subspecies enterica serovar Saintpaul (Salmonella ser. Saintpaul) foodborne outbreak. Investigators have spoken to 49 of the suspected 84 patients (who live in 18 states), and report that 69% of these individuals ate cucumbers within 7 days of becoming ill. However, with a pathogen that causes approximately 3000 illnesses per day in the USA, there are more questions than answers: how can public health officials ascertain that all 84 cases, who live thousands of miles apart and became ill over a 3-month period, are epidemiologically related? If a small batch of cucumbers from a supermarket or an environmental sample from a supplier‘s farm is Salmonella positive, how will investigators know whether this is the same strain causing the illnesses?Enter molecular subtyping protocols: after serotyping, strains are identified by a DNA ‘fingerprinting‘ method known as Pulsed-field Gel Electrophoresis (PFGE). These DNA fingerprints are catalogued and coordinated in the international database PulseNet, which was first established by the US CDC in 1996. In the cucumber outbreak, the strain responsible happened to have a rare fingerprint so it was a relatively simple process to identify the culprit. It is not always so clear-cut and, in addition to PFGE, several other molecular subtyping approaches are being developed for use in Salmonella. Each technique must provide two critical things: strong discriminatory power and high epidemiological concordance [1]: ▪ Discriminatory power: the probability that two unrelated isolates can be distinguished;▪ Epidemiological concordance: the probability that two isolates are part of the same outbreak.Let us demonstrate what these terms mean: the PFGE fingerprint associated with a 2010 Salmonella ser. Enteritidis egg shell outbreak in the USA was an extremely common pattern, making it difficult to conclusively say whether all isolates analyzed during the outbreak timeframe were part of the outbreak itself. Thus, the discriminatory power in this instance was not strong. Conversely, during a 2011 Salmonella ser. Heidelberg outbreak that was linked to ground turkey, PFGE subtyping analyses showed two closely related fingerprints, suggesting the involvement of two different strains. Therefore in this case PFGE was too discriminatory and compromised the epidemiological concordance.These examples highlight the requirement for additional Salmonella subtyping approaches that can be used instead of, or as a complement to PFGE for routine disease surveillance and outbreak tracking. To date, tricky PFGE analyses have most often been resolved by the inclusion of other molecular subtyping techniques such as amplified fragment length polymorphism or multiple-locus variable-number tandem-repeat analysis. However, with the advancement of DNA sequencing technology, in terms of time and cost, there has been a shift toward protocols encompassing sequenced-based methodologies.For example, we and others have recently exploited the hypervariable nature of two genomic loci in Salmonella, known as clustered regularly interspaced short palindromic repeats (CRISPR)1 and CRISPR2, for subtyping purposes [2–5]. CRISPRs are present in nearly half of all bacterial species and these elements comprise several unique short sequences, called spacers, which are interspaced by conserved direct repeats [6]. CRISPR-based subtyping has also proven useful for rapid determination of the serotype of a Salmonella isolate [5], and for predicting the antibiotic resistance profile of Salmonella ser. Typhimurium isolates [7]. Additionally, CRISPR-based subtyping can also be used as an initial screen to immediately rule out certain isolates as being unassociated with an outbreak [5,8].To further highlight the success and applicability of sequence-based subtyping, in a recent blinded study, CRISPR-multivirulence-locus sequence typing (MVLST) was able to successfully identify isolates that were part of a 2012 Salmonella ser. Newport outbreak. Importantly, CRISPR-MVLST was able to separate outbreak isolates from control, sporadic-case isolates [8]. The advantages of CRISPR-typing and other sequenced-based subtyping methodologies lie largely in the unequivocal nature of sequence data as compared with analysis of DNA banding patterns on a gel: sequence data is much more tractable and interlaboratory data comparison is simpler.For all its potential shortcomings, it is indisputable that in Salmonella, PFGE has an excellent track record of success and provides epidemiologists with critical information needed to bring outbreaks to a conclusion. Additionally, given the extensive effort put forth to build and maintain PulseNet, replacing PFGE with a superior technology is not a simple ‘plug and play‘ proposition. However, stunning advances in throughput and cost of full genome sequencing over the past several years is undoubtedly the game changer that we (and others) predict will lead to the ultimate demise of PFGE. Only 12 years ago, genome sequencing was so expensive and specialized that the first Escherichia coli O157:H7 genome was worthy of a Nature publication [9]; however today the sequencing of 26 strains of this pathogen merits a two-page research note (although an excellent one!)[10]. The power of this methodology to quickly track and accurately identify the source of the outbreak strain has already been demonstrated in Salmonella[11] as well as several other high-profile cases involving other pathogens [12–14]. Full genome sequencing also provides information that PFGE cannot, such as the evolutionary relatedness of bacterial isolates. For Salmonella as well as other pathogens, it is well appreciated that not all strains are equally pathogenic [15], and information for making these predictions is contained within the DNA sequence. Other information that can be gleaned out of sequence data includes putative geographical origin [16] and antibiotic resistance profiles [17]. As huge sequencing efforts such as the 100 K Foodborne Pathogen Genome Project [101] are undertaken, collecting such metadata will be extremely important for ultimately using bacterial genome sequences to devise the best course of treatment.However, will these advanced technologies eliminate the need for targeted DNA sequence-based subtyping approaches? We would argue no. Even with the increased use of full genomes sequencing, it will likely be some time before these methods are universal. Hurdles such as data storage space and the complexity of analysis likely mean this technology will be constrained to high-profile cases for the time being. Lower throughput sequencing techniques may continue to be advantageous when the question asked may continue to be advantageous in certain circumstances.So in an era where PFGE, multilocus sequence typing (MLST), CRISPR-MVLST, and genome sequencing are just a few of the dozens of diverse molecular subtyping methods available – a situation poignantly referred to as YATM, or Yet Another Typing Method [18]– how does one determine the best way forward? Is there solid data to support replacing PFGE with full genome sequencing or might simpler sequence-based methods still be useful in public health laboratories? In addition, what about the field of metagenomics, which promises to further simplify outbreak investigations through simultaneous sequencing and subtyping of all organisms present within a blood, sputum or fecal sample without the need for culturing them first [19]? Fortunately, a multilaboratory study spearheaded by the US FDA‘s Center for Food and Applied Nutrition is in the process of initiating studies to answer such questions. In one study, subtyping methods running the gamut of the time-tested PFGE, lower-technological sequence-based methods such as MLST and CRISPR-MVLST, platforms for full genome analysis, and several others will be tested head-to-head in order to determine the ability of each to type a blinded reference collection of Salmonella isolates. Quantitative data such as rapidity, cost and hands-on time are being evaluated as well [101]. This study is the first step in developing an evidence-based argument for whether it is time to look beyond PFGE and, if so, what are the prime candidates for replacing it? While it seems extremely likely, given the number of data points offered by full genome analysis, that these methods will win the battle for discriminatory power, we suspect that a few advantages of the lower-technological approaches may become apparent as well.The outbreak scenario we began this piece with was not theoretical, but actually occurred in the USA this year [102]. As our food system becomes larger and more global, the importance of a rapid, international system of identifying the source of these outbreaks and bringing them to a close is only becoming more important. The power of full genome sequencing will transform this system by allowing us to track with high resolution the spatial and temporal spread of strains, and also provide us with the information needed to know what these strains are capable of and how to best minimize their impact on human health.AcknowledgementsThe authors wish to thank P Evans (FDA-Center for Food Safety and Applied Nutrition) and P Gerner-Smidt (US CDC) for their suggestions during the writing of this editorial.Financial & competing interests disclosureThe authors have no relevant affiliations or financial involvement with any organization or entity with a financial interest in or financial conflict with the subject matter or materials discussed in the manuscript. This includes employment, consultancies, honoraria, stock ownership or options, expert testimony, grants or patents received or pending, or royalties.No writing assistance was utilized in the production of this manuscript.References1 Van Belkum A, Tassios PT, Dijkshoorn L et al. Guidelines for the validation and application of typing methods for use in bacterial epidemiology. Clin. Microbiol. Infect.13(Suppl. 3),1–46 (2007).Crossref, Medline, CAS, Google Scholar2 Shariat N, Dimarzio MJ, Yin S et al. The combination of CRISPR-MVLST and PFGE provides increased discriminatory power for differentiating human clinical isolates of Salmonella enterica subsp. enterica serovar Enteritidis. Food Microbiol.34(1),164–173 (2013).Crossref, Medline, CAS, Google Scholar3 Liu F, Barrangou R, Gerner-Smidt P, Ribot EM, Knabel SJ, Dudley EG. Novel virulence gene and clustered regularly interspaced short palindromic repeat (CRISPR) multilocus sequence typing scheme for subtyping of the major serovars of Salmonella enterica subsp. enterica. Appl. Environ. Microbiol.77(6),1946–1956 (2011).Crossref, Medline, CAS, Google Scholar4 Liu F, Kariyawasam S, Jayarao BM et al. Subtyping Salmonella enterica serovar enteritidis isolates from different sources by using sequence typing based on virulence genes and clustered regularly interspaced short palindromic repeats (CRISPRs). Appl. Environ. Microbiol.77(13),4520–4526 (2011).Crossref, Medline, CAS, Google Scholar5 Fabre L, Zhang J, Guigon G et al. CRISPR typing and subtyping for improved laboratory surveillance of Salmonella infections. PLoS ONE7(5),e36995 (2012).Crossref, Medline, CAS, Google Scholar6 Barrangou R. CRISPR–Cas systems and RNA-guided interference. Wiley Interdiscip. Rev. RNA.4(3),267–278 (2013).Crossref, Medline, CAS, Google Scholar7 Dimarzio M, Shariat N, Kariyawasam S, Barrangou R, Dudley EG. Antibiotic resistance in Salmonellatyphimurium associates with CRISPR sequence type. Antimicrob. Agents Chemother. (2013) (In Press).Medline, Google Scholar8 Shariat N, Kirchner MK, Sandt CH, Trees E, Barrangou R, Dudley EG. CRISPR-MVLST Subtyping of Salmonella serovar Newport outbreak isolates and determination of the relationship between CRISPR-MVLST and PFGE. J. Clin. Microbiol.51(7),2328–2336 (2013).Crossref, Medline, CAS, Google Scholar9 Perna NT, Plunkett G, Burland V et al. Genome sequence of enterohaemorrhagic Escherichia coli O157:H7. Nature409(6819),529–533 (2001).Crossref, Medline, CAS, Google Scholar10 Eppinger M, Daugherty S, Agrawal S et al. Whole-genome draft sequences of 26 enterohemorrhagic Escherichia coli O157:H7 strains. Genome Annouc.1(2),11–12 (2013).Google Scholar11 Allard MW, Luo Y, Strain E et al. High resolution clustering of Salmonella enterica serovar Montevideo strains using a next-generation sequencing approach. BMC Genomics13(1),32 (2012).Crossref, Medline, CAS, Google Scholar12 Köser CU, Ellington MJ, Cartwright EJP et al. Routine use of microbial whole genome sequencing in diagnostic and public health microbiology. PLoS Pathog.8(8),e1002824 (2012).Crossref, Medline, CAS, Google Scholar13 Rohde H, Qin J, Cui Y et al. Open-source genomic analysis of Shiga-toxin-producing E. coli O104:H4. N. Engl. J. Med.365,718–724 (2011).Crossref, Medline, CAS, Google Scholar14 Chin CS, Sorenson J, Harris JB et al. The origin of the Haitian cholera outbreak strain. N. Engl. J. Med.364(1),33–42 (2011).Crossref, Medline, CAS, Google Scholar15 Heithoff DM, Shimp WR, House JK et al. Intraspecies variation in the emergence of hyperinfectious bacterial strains in nature. PLoS Pathog.8(4),e1002647 (2012).Crossref, Medline, CAS, Google Scholar16 Cao G, Meng J, Strain E et al. Phylogenetics and differentiation of Salmonella Newport lineages by whole genome sequencing. PLoS ONE8(2),e55687 (2013).Crossref, Medline, CAS, Google Scholar17 Sangal V, Harbottle H, Mazzoni CJ et al. Evolution and population structure of Salmonella enterica serovar Newport. J. Bacteriol.192(24),6465–6476 (2010).Crossref, Medline, CAS, Google Scholar18 Achtman MA. Surfeit of YATMs? J. Clin. Microbiol.34(7),1870 (1996).Crossref, Medline, CAS, Google Scholar19 Loman NJ, Constantinidou C, Christner M et al. A culture-independent sequence-based metagenomics approach to the investigation of an outbreak of Shiga-toxigenic Escherichia coli O104:H4. JAMA309(14),1502–1510 (2013).Crossref, Medline, CAS, Google Scholar101 UC-Davis School of Vet Med: 100K Foodborne Pathogen Genome Project (2013). http://100kgenome.vetmed.ucdavis.edu/index.cfmGoogle Scholar102 Salmonella Saint Paul Linked to Cucumbers, Final CDC Update (2013).http://outbreakdatabase.com/reports/Salmonella_Saint_Paul_Linked_to_Cucumbers,_Final_CDC_Update.pdfGoogle ScholarFiguresReferencesRelatedDetailsCited ByGenomic Evidence Reveals Numerous Salmonella enterica Serovar Newport Reintroduction Events in Suwannee Watershed Irrigation PondsApplied and Environmental Microbiology, Vol. 81, No. 24 Vol. 8, No. 10 Follow us on social media for the latest updates Metrics History Published online 24 September 2013 Published in print October 2013 Information© Future Medicine LtdKeywordsbacterial pathogenCRISPRfoodborne outbreakgenome sequencingmolecular epidemiologymolecular subtypingPFGESalmonellaAcknowledgementsThe authors wish to thank P Evans (FDA-Center for Food Safety and Applied Nutrition) and P Gerner-Smidt (US CDC) for their suggestions during the writing of this editorial.Financial & competing interests disclosureThe authors have no relevant affiliations or financial involvement with any organization or entity with a financial interest in or financial conflict with the subject matter or materials discussed in the manuscript. This includes employment, consultancies, honoraria, stock ownership or options, expert testimony, grants or patents received or pending, or royalties.No writing assistance was utilized in the production of this manuscript.PDF download