BackgroundDuring a substantial elevation in scarlet fever (SF) notifications in 2014 a national genomic study was undertaken of Streptococcus pyogenes (Group A Streptococci, GAS) isolates from patients with SF with comparison to isolates from patients with invasive disease (iGAS) to test the hypotheses that the increase in SF was due to either the introduction of one or more new/emerging strains in the population in England or the transmission of a known genetic element through the population of GAS by horizontal gene transfer (HGT) resulting in infections with an increased likelihood of causing SF. Isolates were collected to provide geographical representation, for approximately 5% SF isolates from each region from 1st April 2014 to 18th June 2014. Contemporaneous iGAS isolates for which genomic data were available were included for comparison. Data were analysed in order to determine emm gene sequence type, phylogenetic lineage and genomic clade representation, the presence of known prophage elements and the presence of genes known to confer pathogenicity and resistance to antibiotics.Results555 isolates were analysed, 303 from patients with SF and 252 from patients with iGAS. Isolates from patients with SF were of multiple distinct emm sequence types and phylogenetic lineages. Prior to data normalisation, emm3 was the predominant type (accounting for 42.9% of SF isolates, 130/303 95%CI 37.5–48.5; 14.7% higher than the percentage of emm3 isolates found in the iGAS isolates). Post-normalisation emm types, 4 and 12, were found to be over-represented in patients with SF versus iGAS (p < 0.001). A single gene, ssa, was over-represented in isolates from patients with SF. No single phage was found to be over represented in SF vs iGAS. However, a “meta-ssa” phage defined by the presence of :315.2, SPsP6, MGAS10750.3 or HK360ssa, was found to be over represented. The HKU360.vir phage was not detected yet the HKU360.ssa phage was present in 43/63 emm12 isolates but not found to be over-represented in isolates from patients with SF.ConclusionsThere is no evidence that the increased number of SF cases was a strain-specific or known mobile element specific phenomenon, as the increase in SF cases was associated with multiple lineages of GAS.Electronic supplementary materialThe online version of this article (doi:10.1186/s12864-017-3603-z) contains supplementary material, which is available to authorized users.

Currently there are 93 validated M serotypes of Streptococcus pyogenes, Lancefield group A streptococcus (GAS), and >130 emm genotypes. A marked increase in the number of non-typable GAS isolates (2 % in 2000, 4 % in 2001 and 9 % in 2002) from invasive disease cases referred to the authors' reference laboratory was noted during 2000-2002. A total of 217 (92 %) were from blood cultures, 14 (6 %) from deep abscesses and five (2 %) from aspirates. The clinical manifestations included bacteraemia, septicaemia, cellulitis, meningitis, necrotizing fasciitis and toxic-shock syndrome. In order to establish whether this increase was due to the emergence of novel types or the unavailability of M-typing sera, these isolates were subjected to emm sequencing. A total of 144 isolates (61 %) belonged to M types for which sera were no longer available; 112 (48 %) belonged to higher M types, including emm83.1 (9 %), emm94 (8 %) emm87 (6 %) and emm89 (6 %); and 32 (13 %) belonged to lower M types that were not commonly isolated in the UK, and included M25, M43, M49, M64, M73 and M74. Sixty-six (28 %) of the isolates belonged to newly designated emm types. Other isolates belonged to the novel emm types st2147, STNS1033 and st854, recently registered in the Centers for Disease Control (CDC) database by other laboratories. One novel emm type, st2161, was isolated from an injecting drug user. There were differences in the type distribution of these isolates according to geographic location. However, 90 % of emm93, one of seven predominant emm types identified amongst the collection of M nontypable (MNT) isolates, were isolated from the London region.

Genomic discrimination can be applied to assist outbreak investigation. It enabled confirmation or exclusion of GAS cases from epidemiologically defined outbreaks. Colonization of healthcare workers and environmental contamination with the outbreak strain was demonstrated for several outbreaks.

No abstract