Sequence-based and other methods for molecular serotyping of E. coli

Authors: Fratamico, Pina

Submitted to: United States-Japan Cooperative Program in Natural Resources (UJNR)
Publication Type: Proceedings
Publication Acceptance Date: 8/27/2017
Publication Date: 11/12/2017
Citation: Fratamico, P.M. 2017. Sequence-based and other methods for molecular serotyping of E. coli. In: Proceedings of the United States-Japan Cooperative Program in Natural Resources (UJNR), November 12-18, 2017, Japan. p.1.

Interpretive Summary:

Technical Abstract: Escherichia coli strains are classified based on their O- and H-types. Currently there are approximately 188 different O-serogroups; however, a number of O-group designations have been withdrawn. There are 56 H-types, but H13, H22, and H50 have been withdrawn. Traditionally, determination of O- and H-antigens has been performed by serotyping using antibodies. The O-groups are defined by the genes in the O-antigen gene clusters, with wzx (O-antigen flippase) and wzy (O-antigen polymerase) genes as unique markers for many O-groups or the wzm (transporter protein) and wzt (ATP-binding component) genes in strains that produce O-antigens via an ABC transporter-dependent pathway. To determine H-types, unique sequences in one of the genes involved in formation of flagella can be targeted in genetic-based assays. Information on genes involved in formation of the E. coli O-antigens has been published [1]. The polymerase chain reaction (PCR) has been widely used to detect pathogens in human specimens, food, animals, and the environment. Real-time PCR monitors the amplification of the targeted DNA sequence during the PCR (in real-time) through the use of fluorescent probes. In recent years, miniaturization of reactions by designing nanoliter-scale PCR platforms has been a major step forward in real-time PCR. One such platform is the Biomark system from Fluidigm, which is a nanoscale, real-time high-throughput PCR platform that can perform thousands of reactions in a single run, reducing reagent and sample consumption. For example, 48 assays can be performed on 48 different samples (2,304 reaction simultaneously) or 96 assays on 96 samples for 9,216 reactions. The FDA E. coli identification (FDA-ECID) microarray was designed for characterizing pathogenic E. coli, and it contains a molecular serotyping component [2]. This high-density microarray was designed for use with the GeneAtlas Personal Microarray System (Fig. 1) and can detect, discriminate, identify, and characterize E. coli strains. All known E. coli genes (>40,000) and >10,000 single nucleotide polymorphism biomarkers are represented on the array, and it has probes that can identify all 53 H-types and 108 E. coli O-groups. AmpliSeq is a highly multiplexed PCR-based targeted sequencing technology that can be accomplished at low cost using small amounts of DNA. An Ion AmpliSeq panel consists of a pool of up to thousands of PCR primers for single reactions. For E. coli, primer pairs targeting unique regions in all O-groups and H-types, as well as primers for virulence or other genes can be included in the pool. After performing PCR, remaining primers are digested, and a library containing the remaining amplicons are ligated with up to 768 different barcodes and prepared for sequencing (Fig. 2). The data are then analyzed with Ion Reporter or other software to determine which sequences were amplified. Whole genome sequencing (WGS) is being utilized as a tool for E. coli typing, subtyping, characterization, diagnostics, and for outbreak investigation. A publicly available tool at the Center for Genomic Epidemiology (http://www.genomicepidemiology.org/) website called SerotypeFinder identifies the E. coli serotype from WGS data [3]. Results and Discussion The BioMark real-time PCR system (Fluidigm) was used for amplification of 69 E. coli-associated virulence genes, as well as a number of genes targeting specific O-groups (N=14) and H-types (N=11) in Shiga toxin-producing E. coli (STEC) strains recovered from swine [4]. There were 11 distinct virulence gene profiles identified among the strains, with 16 different combinations based on serotype and virulence gene profiles. A large portion of the strains belonged to serotype O59:H21, and they carried the same virulence gene profile. Some STEC strains could be considered hybrids carrying genes found in more than one E.