Enhanced nasopharyngeal infection and shedding associated with an epidemic lineage of emm3 group A Streptococcus

Abstract

Background: A group A Streptococcus (GAS) lineage of genotype emm3, sequence type 15 (ST15) was associated with a 6 month upsurge in invasive GAS disease in the UK. The epidemic lineage (Lineage C) had lost 2 typical emm3 prophages, Φ315.1 and Φ315.2 associated with the superantigen ssa, but gained a different prophage (ΦUK-M3.1) associated with a different superantigen, speC and a DNAse spd1.

Methods and results: The presence of speC and spd1 in Lineage C ST15 strains enhanced both in vitro mitogenic and DNase activities over non-Lineage C ST15 strains. Invasive disease models in Galleria mellonella and SPEC-sensitive transgenic mice, revealed no difference in overall invasiveness of Lineage C ST15 strains compared with non-Lineage C ST15 strains, consistent with clinical and epidemiological analysis. Lineage C strains did however markedly prolong murine nasal infection with enhanced nasal and airborne shedding compared with non-Lineage C strains. Deletion of speC or spd1 in 2 Lineage C strains identified a possible role for spd1 in airborne shedding from the murine nasopharynx.

Conclusions: Nasopharyngeal infection and shedding of Lineage C strains was enhanced compared with non-Lineage C strains and this was, in part, mediated by the gain of the DNase spd1 through prophage acquisition.

Keywords: DNases; Streptococcus pyogenes; epidemic upsurge; genotype emm3; group A Streptococcus; invasive group A streptococcal disease; nasopharyngeal infection; prophage; serotype M3; superantigens.

Figure 1.

Phylogenetic representation of the ST15 emm3 population. Within the ST15 population (indicated in black) was the identified epidemic Lineage C ST15 (indicated in red), characterized by the unique prophage profile; Φ315.1 and Φ315.2 are absent in all 78 Lineage C strains and 65/78 carried ΦUK-M3.1 with speC and spd1. The short reads for 172 ST15 isolates were mapped to the ST15 reference strain MGAS315 and concatenated SNPs from the core genome (excluding all prophage regions) were used to generate the maximum likelihood phylogenetic tree with RAxML.

Figure 2.

Lineage CST15 emm3 strains have greater DNase activity and enhanced mitogenicity compared with non-Lineage CST15 strains. (A) Human MNC proliferation was greater in response to broth culture supernatant of Lineage C ST15 strains (with ΦUK-M3.1-speC) (red, n = 51) compared with non-Lineage C ST15 strains (with Φ315.2-ssa) (blue, n = 41). Data represent counts per minute (cpm) of thymidine incorporation of a single human MNC donor minus the background level (delta cpm). All emm3 strains carried additional superantigens speA, speG, speK and the non-functional M3-smeZ. The experiment was repeated with a second donor and gave similar results. (B) DNase activity was greater in culture supernatant of Lineage C ST15 strains (with the ΦUK-M3.1-spd1) (red, n = 51) than non-Lineage C ST15 strains (blue, n = 41). Data represent DNase activity in units per ml of culture supernatant calculated relative to a standard. ***; p ≤ 0.0001 (Mann-Whitney). Horizontal line represents the median (C) Human MNC proliferation was reduced in response to culture supernatant of Lineage C ST15 strains M3-C1 and M3-C2 when speC was deleted by mutagenesis (M3-C1ΔspeC and M3-C2ΔspeC, white bars) compared with wild-type parental strains M3-C1 and M3-C2 (black bars). (D) DNase activity was reduced in culture supernatant of Lineage C ST15 strains M3-C1 and M3-C2 when spd1 was deleted by mutagenesis (M3-C1Δspd1 and M3-C2Δspd1, white bars) compared with wild-type parental strains M3-C1 and M3-C2 (black bars). Data represent the mean of 3 experimental repeats (+standard deviation).

Figure 3.

SPEC is more potent than SSA in both human and murine HLA-DQ8 transgenic cell culture and in vivo. (A) Human MNCs were stimulated with recombinant SPEC (red dotted line, left axis) or SSA (blue dotted line, left axis) at increasing concentrations and compared with murine splenocytes from HLA-DQ8 transgenic mice also stimulated with SPEC (red solid line right axis) or SSA (blue solid line, right axis). Data represent the mean ( ± standard deviation) of a single human donor measured in triplicate or splenocytes from 3 mice measured in triplicate. Human MNC data were reproduced in one other donor. (B) To see if this would be replicated in vivo, human MNCs were stimulated with infected thigh tissue (circles) or sera (squares) obtained from HLA-DQ8 mice after 24 h of infection with a non-Lineage C associated ST15 strain M3–1 (blue) or a Lineage C associated ST15 strain, M3-C1 (red); 8 mice per strain. A greater mitogenic response was observed when human MNCs were exposed to thigh or serum from mice infected with a Lineage C strain compared with a non-Lineage C strain. Control; uninfected mouse thigh tissue (black circles) or uninfected mouse serum (black squares). ***; p ≤ 0.0001 (Mann-Whitney). Each data point represents the mean of an individual mouse sample measured in triplicate. Horizontal lines represent the median.

Figure 4.

Nasopharyngeal infection with Lineage C strains resulted in prolonged nasal shedding and enhanced airborne shedding. Mice infected intranasally demonstrated direct nasal shedding of GAS over a longer period of time when infected with Lineage C strains (A) M3-C1 (red line) or (C) M3-C2 (red line) compared with mice infected with ST15 strains (A) M3–1 (blue line) or (C) M3–2 (blue line). Direct nasal shedding was monitored by daily nose press. *; p = 0.0167 (A) or **; p = 0.0062 (C) (Mantel-Cox Log-rank test). (B) Nasal shedding of GAS by each mouse was significantly greater in mice infected with M3-C1 (red line) compared with M3–1 (blue line) on day 2 (p = 0.0122) and day 3 (p = 0.021) of infection. Airborne shedding of GAS was also significantly greater in the cage of mice infected with M3-C1 compared with the cage of mice infected M3–1 on days 2 (p = 0.0294) and 3 (p = 0.0256). (D) Nasal shedding of GAS by each mouse was no different between those infected with M3-C2 compared with M3–2 but airborne shedding of GAS was significantly greater in the cage of mice infected with M3-C2 compared with the cage of mice infected M3–3 on day 2 (p = 0.0294). *; p ≤ 0.05 (Mann-Whitney). N = 7 per group. Nasal and airborne shedding data represent mean ( ± SEM).

Figure 5.

Nasopharyngeal infection with Lineage C associated ST15 strains resulted in enhanced nasal mucosal damage. Four mice were infected intranasally for 2 d with either one of 2 ST15 strains, M3–1 or M3–2, or one of 2 Lineage C-associated ST15 strains, M3-C1 or M3-C2. (A) and (B) photomicrographs of haemotoxylin and eosin stained nasal cavity sections following infection with either a non-Lineage C ST15 strain (A) or a Lineage C ST15 strain (B). Damage to the nasal mucosa with surface neutrophilic exudate was observed following Lineage C strain infection. (C) Semi quantitative histopathological analysis of all sections was performed to assess the damage to the nasal cavity based on level of rhinitis and associated neutrophilic exudate within the nasal cavity. +++; marked, ++; moderate, +; mild, −; no abnormality. Each row in each column represents an individual infected mouse. Free and phagocytosed bacteria were observed in all sections that scored above ‘no abnormality’.

Figure 6.

The DNase Spd1 contributes to nasal and airborne shedding of Lineage C strains. HLA-DQ8 transgenic mice were infected intranasally with either the parental wild-type strain (M3-C1, red line) or the speC deleted strain of M3-C1 (M3-C1ΔspeC black line) and nasal shedding was monitored daily over a period of 7 d (A) along with the number of nasal GAS shed by each mouse and airborne GAS shed (B). N = 8 per group. HLA-DQ8 transgenic mice were also infected intranasally with either the parental wild-type strain (M3-C1, red line) or the spd1 deleted strain of M3-C1 (M3-C1Δspd1, black line) and nasal shedding was monitored daily over a period of 7 d (C) along with the number of nasal GAS shed by each mouse and airborne GAS shed (D). N = 7 per group. Deletion of speC had limited effect on the nasal infection with, in fact an increase in airborne GAS shedding of M3-C1ΔspeC on day 1 (p = 0.0286). Deletion of the DNase spd1 significantly reduced the number of nasal GAS shed by each mouse on day 2 and reduced airborne GAS shedding on days 2 (p = 0.0284) and 3 (p = 0.0275). *; p ≤ 0.05 (Mann-Whitney). Nasal and airborne shedding data represent mean ( ± SEM).