Streptococcal toxic shock syndrome (STSS) is typically caused by Streptococcus pyogenes (group A Streptococcus [GAS]) (1). Major investigations on host-pathogen interactions have been performed to determine why some persons experience uncomplicated pharyngitis, but STSS develops in others. On a molecular level, mutations in covS (a sensor gene of the major virulence regulator CovR/S) have been frequently associated with invasive GAS disease (2). In 2009, we reported a case of STSS caused by S. agalactiae (group B Streptococcus [GBS]) and covS mutation (3). Here, we reassess those findings in a larger collection of GBS isolates causing STSS.

We tested 26 GBS isolates from 25 patients (22 adults, 3 children) (Table) that were pooled from 3 countries; the United States (22 strains collected 2004–2005), Germany (1 strain, 2006), and Switzerland (2 strains, 2005). For 1 of the 2 case-patients from Switzerland, 2 isolates (same clone) were available for mutation analyses (patient 23 [4]). The isolate from our previously published case report (Sweden, 2005 [3]) served as a control strain for molecular analyses; the corresponding case-patient was included in the demographic analyses (i.e., 26 patients: 23 adults, 3 children). The median age of the included adult patients was 59 years (interquartile range 45.5–68 years); mortality rate was 35% (8/23). The ages of the 3 children were 0, 30, and 60 days (1 death).

We used standard molecular biology techniques for nucleic acid preparation and analysis. We performed molecular typing by multilocus sequence typing (MLST) as described (5) and capsular typing by using latex agglutination and PCR serotype determination (6) (Table). To analyze the cov gene locus, we amplified the genes covS and covR by PCR (Technical Appendix Table). Resulting PCR products underwent DNA sequencing with internal cov primers on an ABI PRISM 310 Genetic Analyzer (Applied Biosystems, Weiterstadt, Germany).

Nucleotide sequence analysis showed that, in 1 of the strains (from patient 18), both genes, the sensor histidine kinase covS and the response regulator covR, had mutated. In covR, at nucleotide position 242, cytosine was replaced by thymine, leading to an amino acid exchange from alanine to valine. In addition, the covS gene of this strain showed a 1-bp deletion of adenine at position 895 of the gene, causing a frame shift and leading to a truncated CovS with a stop codon at nucleotide position 926 of the gene. In the control strain that harbored a 3-bp deletion, our previously published finding was confirmed (3). In the remaining strains, cov alleles matched the gene sequences of completely sequenced GBS strains in the GenBank database (NEM316, 2603V/R, 909A).

During the past few decades, the overall incidence of invasive GBS infections has increased substantially. This trend is particularly noticeable in the elderly and in persons with co-morbid conditions (7). Our results regarding age distribution of patients, mortality rate, and frequencies of different GBS serotypes are in line with results of previous studies. Twelve (52%) of 23 patients were >59 years old. The mortality rate for group B STSS (>30%) was similar to that reported for group A STSS (1). In a previous case series, which included 13 patients with group B STSS, the mortality rate was 23% (3/13) (8). Three-quarter of our strains (19/26 strains) were attributed to serotype V or Ia/Ib. Large epidemiologic studies have frequently implicated serotype s Ia/Ib, III, and V GBS isolates in the etiology of invasive disease in adults (9). Apart from sequence type (ST) 1 and ST23 (15/26 strains), the distribution of MLSTs among the GBS isolates was heterogeneous. The highly virulent GBS lineage ST17 was found in only 2 patients (1 adult and 1 child).

The role of covR/S mutations in the switch from colonization to invasion has been demonstrated for GAS in a mouse model (2). Consistent with these findings, GAS strains isolated from STSS patients frequently carry mutations in this operon (10). Similarly, a 3-bp deletion in the covR gene was detected in a GBS strain that caused STSS and necrotizing fasciitis (3). However, our investigations on a larger collection of GBS isolates did not confirm a high cov mutation rate. Only 1 of 25 GBS clones demonstrated a mutation. From 1 patient, a colonizing and invasive isolate (same clone) was available (4); that isolate showed no mutation in covR/S.

Our results should be interpreted with caution because the absolute number of included patients is small, and the cases were pooled from various centers. Nonetheless, Ikebe et al. found covR/S mutations in 76 (46.3%) of 164 GAS strains causing STSS and in only 1.7% of 59 strains without invasive disease (10). In light of these results, the low frequency of mutations found in our collection is surprising. Yet, the association with covR/S mutations and GBS TSS in a case report has been shown previously and confirmed here. However, GBS harbors multiple 2-component systems and stand-alone regulators. Our findings indicate that different virulence regulators may be involved in the pathogenesis of fulminant GBS disease.