Finding Your Perfect Match: Evolving Technologies for Bacterial Strain Typing

This is an open question and an area of active investigation and debate. The increased availability of whole genome sequencing (WGS) in recent years has allowed for the comparison of many strains at the genome level, both closely related and less closely related. The minimum difference that we can set between two strains and still call them identical is determined by the error rate of the WGS methods. Below this threshold, we cannot determine if the differences are strain differences or sequencing errors. This is typically less than five nucleotides across the whole genome.

Any characteristic that prevents the phage from attaching, replicating, or lysing the cell will change the susceptibility to a given phage in the panel. Phages use cell surface proteins or sugars as receptors to mediate phage attachment, and any mutation that alters these molecules so that they no longer bind to the phage effectively will prevent infection. For example, restriction enzyme systems originally evolved to cut up phage genomic material and prevent infection, so expression of specific restriction systems will target some phages. Alternatively, integrated prophages can express phage proteins that repress the replication of phage genomic material, which, in turn, can prevent coinfection by another phage of the same type.

PulseNet was originally developed for monitoring food-borne pathogens; this is reflected in the species represented in the database. PulseNet hold patterns and protocols for Shiga-toxin producing Escherichia coli, Campylobacter, Listeria monocytogenes, Salmonella, Shigella, Vibrio cholera, Vibrio parahaemolyticus, and Cronobacter. Researchers have used PFGE to type many other strains, but these results not maintained in same manner as PulseNet.

Genes in the core genome of a species determine serotype. These core genes are vertically transmitted from parental to daughter cells, so a lineage will typically inherit the same serotype. Some lineages will pick up virulence factors through horizontal exchange, and those lineages will transmit the serotype and the acquired virulence factors from parental to daughter cells, which can result in some specific serotypes becoming associated with sets of virulence factors.

Typically, methods used for strain identification are not used for species identification as well. Characteristics that can be resolved to distinguish strains (whether banding patterns or gene sequence) will be too variable (not enough shared characteristics) to be useful between species. Conversely, characters with enough shared information to distinguish between species are often too conserved (not enough variable characteristics) to distinguish strains of a species.

There is no technical or genetic limitation that would prohibit adding more species to the existing MVLA databases. Some specific species might be less amenable to the method due to fewer variable number tandem repeats (VNTRs). The primary limitation is the time and resources required by researchers to establish databases of patterns to compare strains, and to establish a structure to add new strains as they become available.

ATCC normally types strains to the species level, but in some cases strain level typing is performed to verify strain specific characteristics. The ATCC Quality Control testing group carries out O-group serotyping for many E. coli strains. Many of the strains in the ATCC MRSA collection have been typed by both PFGE and sequencing of the spa gene. WGS has been used to confirm the strain identity of a group of Escherichia strains in the collection.

Yes, in many cases a strong correlation exists between specific sequence types or clonal complexes and virulence. Multilocus sequence typing (MLST) is based on core genome genes that are vertically transferred (i.e. from parental to daughter cells). Some of these lineages acquire virulence genes through horizontal gene transfer and take on virulence characteristics. These virulence characteristics are then vertically inherited as along with the core genome alleles that define the MLST, so the ST or CC lineage takes on virulence characteristics (virulence factors, AMR, etc.). For example, CC1 and CC8 are associated with community acquired MRSA. E. coli ST 131 is a recently emerged strain associated with outbreaks of extraintestinal E. coli infections.

The Ridom and Kreisworth methods both examine the same segment of the spa gene, the polymorphic VNTR region in the 3’ coding sequence. Each of the unique repeat segments (usually 24 nucleotides) has been given an identifier, and each spa type is a unique combination of these repeats across the VNTR region. The Ridom and Kresiworth methods each provide a different naming scheme to describe the repeat sequences.