Serotype Distribution and Antimicrobial Resistance of Streptococcus agalactiae Isolates in Nonpregnant Adults with Streptococcal Toxic Shock Syndrome in Japan in 2014 to 2021

ABSTRACT The incidence of streptococcal toxic shock syndrome (STSS) due to group B Streptococcus (GBS) has been increasing annually in Japan and is becoming a serious challenge. Furthermore, in recent years, penicillin- or clindamycin-resistant strains used in treating streptococcal toxic shock syndrome have been reported. However, no report analyzed >100 isolates of group B Streptococcus causing streptococcal toxic shock syndrome. Therefore, we aimed to perform serotyping and antimicrobial susceptibility testing of 268 isolated group B Streptococcus strains from streptococcal toxic shock syndrome cases involving nonpregnant adult patients in Japan between 2014 and 2021. The most prevalent serotype was Ib, followed by serotypes V, III, and Ia. Seven isolates were resistant to penicillin G, and 17.9% (48 isolates) were resistant to clindamycin. Of the penicillin-resistant group B Streptococcus isolates, 71.4% (5 isolates) were clindamycin resistant. In addition, group B Streptococcus strains resistant to penicillin and clindamycin were isolated from patients with streptococcal toxic shock syndrome. Therefore, before these strains become prevalent, introduction of the group B Streptococcus vaccine is essential for disease prevention. IMPORTANCE Group B Streptococcus (GBS) has been increasingly associated with invasive disease in nonpregnant adults. Such infections are responsible for substantial morbidity and mortality, particularly in individuals with underlying chronic conditions. Streptococcal toxic shock syndrome (STSS) is a severe invasive infection characterized by the sudden onset of shock, multiorgan failure, and high mortality. In this study, we assessed 268 GBS-related STSS cases in nonpregnant adults in Japan between 2014 and 2021. Serotype Ib was the most prevalent, followed by serotypes V, III, and Ia, which were identified in more than 80% of STSS isolates. We found that 48 clindamycin-resistant strains and 7 penicillin G-resistant strains were isolated between 2014 and 2021. We believe that our study makes a significant contribution to the literature because we show that the GBS vaccine, particularly the hexavalent conjugate vaccine, is important to reduce the number of patients with STSS.

ABSTRACT The incidence of streptococcal toxic shock syndrome (STSS) due to group B Streptococcus (GBS) has been increasing annually in Japan and is becoming a serious challenge. Furthermore, in recent years, penicillin-or clindamycin-resistant strains used in treating streptococcal toxic shock syndrome have been reported. However, no report analyzed .100 isolates of group B Streptococcus causing streptococcal toxic shock syndrome. Therefore, we aimed to perform serotyping and antimicrobial susceptibility testing of 268 isolated group B Streptococcus strains from streptococcal toxic shock syndrome cases involving nonpregnant adult patients in Japan between 2014 and 2021. The most prevalent serotype was Ib, followed by serotypes V, III, and Ia. Seven isolates were resistant to penicillin G, and 17.9% (48 isolates) were resistant to clindamycin. Of the penicillin-resistant group B Streptococcus isolates, 71.4% (5 isolates) were clindamycin resistant. In addition, group B Streptococcus strains resistant to penicillin and clindamycin were isolated from patients with streptococcal toxic shock syndrome. Therefore, before these strains become prevalent, introduction of the group B Streptococcus vaccine is essential for disease prevention.
IMPORTANCE Group B Streptococcus (GBS) has been increasingly associated with invasive disease in nonpregnant adults. Such infections are responsible for substantial morbidity and mortality, particularly in individuals with underlying chronic conditions. Streptococcal toxic shock syndrome (STSS) is a severe invasive infection characterized by the sudden onset of shock, multiorgan failure, and high mortality. In this study, we assessed 268 GBS-related STSS cases in nonpregnant adults in Japan between 2014 and 2021. Serotype Ib was the most prevalent, followed by serotypes V, III, and Ia, which were identified in more than 80% of STSS isolates. We found that 48 clindamycin-resistant strains and 7 penicillin G-resistant strains were isolated between 2014 and 2021. We believe that our study makes a significant contribution to the literature because we show that the GBS vaccine, particularly the hexavalent conjugate vaccine, is important to reduce the number of patients with STSS.
S treptococcus agalactiae (group B Streptococcus [GBS]) is a Gram-positive bacterium classified in Lancefield group B. GBS is found in the gastrointestinal tracts and vaginas of asymptomatic humans. It is well established as a colonizing agent in pregnant women. It is an important cause of neonatal sepsis and meningitis (1). Nevertheless, in the past few decades, GBS has been increasingly associated with invasive diseases in nonpregnant adults (2)(3)(4)(5). Such infections are responsible for substantial morbidity and mortality, particularly in individuals with underlying chronic conditions (5).
Streptococcal toxic shock syndrome (STSS) is a severe invasive infection characterized by the sudden onset of shock, multiorgan failure, and high mortality (6,7). It is caused by group A Streptococcus (Streptococcus pyogenes) (8,9). GBS causes invasive streptococcal infections, including STSS, similar to group A Streptococcus (6, 10). Furthermore, there are significant differences between GBS-related STSS and invasive GBS (iGBS), especially in prognosis in Japan (7,11). Serotype is a factor associated with invasiveness and prognosis often reported in adult iGBS but limited in GBS STSS (1,12). An approach to treating severe invasive infections, such as STSS, is combining penicillin and clindamycin (13). However, clinical GBS isolates with decreased susceptibility to penicillin have been reported in several countries (7,(14)(15)(16)(17). Resistance to clindamycin varies widely: up to 74.1% in China (18) and 65.9% in Taiwan (19).
Based on structural differences in the polysaccharide capsule, GBS isolates can be divided into 10 serotypes: Ia, Ib, and II to IX (20). GBS-related STSS is a devastating illness with a high mortality rate. However, as this infection is rare, no study has analyzed .100 isolates of GBS causing STSS, and the serotype prevalent in STSS isolates remains unknown. Herein, we aimed to describe the bacteriological analysis of 268 GBS-related STSS cases in nonpregnant adults.

RESULTS
Isolates. We determined the serotypes of the 268 STSS isolates by performing agglutination tests using antisera against each capsular type (Fig. 1) Erythromycin-and clindamycin-resistant isolates. Of 268 isolates, 83 (31.0%) were resistant to erythromycin or clindamycin. Forty-six isolates (17.2%) were resistant to erythromycin and clindamycin. Two isolates (0.7%) were resistant to clindamycin and susceptible to erythromycin. Thirty-five isolates (13.1%) were resistant to erythromycin and susceptible to clindamycin, of which 11 were erythromycin induced (Table 3). No association was observed between death prognosis and drug resistance.
All 43 ermB-carrying isolates were resistant to erythromycin and clindamycin. The two isolates carrying linB were resistant to clindamycin and susceptible to erythromycin. All 24 isolates carrying mefA/E were resistant to erythromycin and susceptible to clindamycin. The three isolates carrying ermA were resistant to clindamycin and erythromycin. The remaining 11 isolates carrying ermA were resistant to erythromycin, susceptible to clindamycin, and showed erythromycin-induced clindamycin resistance. The two isolates carrying ermA and msrD were resistant to erythromycin and clindamycin. The seven isolates carrying mefA/E and msrD were resistant to erythromycin and susceptible to clindamycin. Penicillin-and cephem-resistant isolates. Seven isolates (2.6% [7/268]) were resistant to penicillin G; one isolate, four isolates, one isolate, and one isolate were of serotypes Ia, Ib, II, and III, respectively. Four isolates, five isolates, and one isolate were resistant to ampicillin, cefazolin, and cefotaxime, respectively (Table 4). Three isolates were resistant to penicillin G and cefazolin, two to penicillin G and ampicillin, one to penicillin G, ampicillin, and cefazolin, and one to penicillin G, ampicillin, cefazolin, and cefotaxime. Penicillin G-resistant isolates were not necessarily resistant to other penicillinand cephem-based antibiotics. Additionally, the antibacterial resistance differed, depending on the strain. Not all strains were resistant to the same antimicrobial agents; therefore, specific resistant strains were probably not prevalent in the community.
Sequence analysis of pbp was performed for each of the seven penicillin G-resistant GBS isolates. The penicillin G-resistant GBS strains had several amino acid substitutions in the deduced amino acid sequence of the penicillin-binding protein PBP2X. In contrast, either or both PBP1A and PBP2B had amino acid substitutions. The V405A and Q557E amino acid substitutions commonly detected in PBP2X of penicillin-resistant strains were found in five of the seven strains (Table 4). According to Kimura et al. (21), PBPs are classified according to amino acid substitutions. Two strains belonged to subclasses IIc and IIIa, and one belonged to subclass IIIc. In addition, two strains belonged to a new class (Table 3).  Five of the seven penicillin-resistant GBS isolates (71.4%) were clindamycin resistant. Regarding outcomes for patients with penicillin-resistant GBS isolates, two of the seven patients (28.6%) died. Regarding outcomes in patients from whom penicillin Gand clindamycin-resistant GBS strains were isolated, two of the five patients (40.0%) died.

DISCUSSION
We investigated 268 STSS cases caused by GBS in nonpregnant adults. Serotype Ib was the most prevalent, followed by serotypes V, III, and Ia, which were identified in over 80% of STSS isolates. Of the 268 isolates, 48 clindamycin-resistant (17.9%) and 7 penicillin G-resistant (2.6%) strains were isolated between 2014 and 2021. Clindamycinresistant isolates had linB, ermA (including two isolates with ermA and msrD), or ermB genes. Penicillin G-resistant isolates contained amino acid substitutions in PBP2X and PBP1A or PBP2B. There have been studies on invasive GBS infections; however, this is the first to study strains isolated from over 100 STSS cases.
According to active, population-based surveillance for invasive GBS disease via the Active Bacterial Core surveillance network in the United States, the median age and case fatality rate of invasive GBS infection were 64 years and 5.6%, respectively. On the other hand, those for STSS were 73 years and 24.6%, respectively. The median age in patients with STSS was higher than that in invasive GBS infections. This may mean that older patients are more likely to progress to STSS. This may have resulted in increased severity and fatality rates.
In Japan, serotypes of isolates obtained from infants with invasive infections have been reported (22). The rates of serotypes Ib (P , 0.001) and V (P , 0.001) in strains isolated from STSS cases (this study) were significantly higher than those in strains isolated from infants with invasive infections (22). In contrast, serotypes III (P , 0.001) and Ia (P = 0.0192) were isolated more often in infants with invasive infections than in STSS isolates in this study (22). These data suggest that the serotype distributions of isolates caused by STSS and invasive infections in infants differ.
Various GBS vaccine formulations have been tested in clinical trials but are not currently approved (23). Based on this study's findings, the trivalent and hexavalent conjugate vaccines have efficacies of 61.2% (164/268) and 91.4% (245/268), respectively, for preventing STSS from GBS. The coverage of the hexavalent conjugate vaccine was considerably higher than that of the trivalent vaccine because the hexavalent conjugate vaccine contained the second most frequently isolated serotype V polysaccharide (Fig. 1). Therefore, vaccination of nonpregnant adults with the hexavalent conjugate vaccine could reduce STSS cases due to GBS by over 90%. The serotype distribution of GBS causing invasive infections in nonpregnant adults has been reported in various countries (24). Serotype V was the most prevalent serotype globally (43.5%) and in North America (46.7%). On the other hand, according to active, population-based surveillance for invasive GBS disease via the Active Bacterial Core surveillance network in the United States in 2008 to 2016, serotype Ia was the most prevalent serotype (5). Furthermore, serotype Ia was the second most prevalent serotype, followed by serotype III in Europe (25.0%) and Asia (29.5%). In Japan, the most prevalent serotype was Ib (28.5%) in isolates from patients with STSS (Fig. 1). Additionally, before 2013, serotype Ib was the most prevalent in STSS (11) and invasive infections (7). These findings suggest that GBS serotype distribution may differ geographically among nonpregnant adults, particularly in Japan.
In 15 of 19 studies, strains isolated from invasive GBS infection were susceptible to penicillin. Four studies reported some resistance to penicillin in Thailand (2%), Latin America (1.4%), the United States (0.5%), and Taiwan (2%) (5,24). In our study, seven isolates (2.6%) were resistant to penicillin, and the rates of penicillin-resistant isolates were similar to those of these studies. In Japan, the rate of penicillin-resistant strains isolated from invasive GBS infections in 2010 to 2013 was 0% (7). Therefore, paying attention to the emergence of penicillin-resistant strains in recent years in Japan is necessary. This study's proportion of erythromycin-resistant isolates was high (30.2%) ( Table 2). This high resistance rate is recognized in Japan and worldwide (24) and may pose a serious challenge in clinically managing patients allergic to penicillin. A recent study in Portugal reported an increase in macrolide resistance in GBS despite its decreased consumption, suggesting that the expansion of specific clones is the main factor for this variation (25). Thus, constantly monitoring changes in the GBS epidemic and implementing GBS vaccination covering a wide range of serotypes may help to reduce antimicrobial resistance.
We collected 268 GBS isolates from patients with STSS. Five isolates were resistant to penicillin and clindamycin ( Table 4). Two of the five patients (40.0%) died. This is the first report of mortality in patients with STSS caused by penicillin-and clindamycin-resistant GBS strains.
Notably, our study had some limitations. Patient data on antibiotic history and clinical characteristics were unavailable. Therefore, we focused on outcomes and analysis of in vitro data. Further investigation of the outcome and timing of antibiotic treatment in penicillin-and clindamycin-resistant strains is warranted. This study has interpretive limitations due to a lack of sequence type or clonal complex analysis.
Herein, we presented GBS isolated from over 100 patients with STSS in Japan. The distribution of serotypes differed from that worldwide and is characteristic of Japan. Penicillin-resistant bacteria were also isolated from GBS from patients with STSS; many isolates were also resistant to clindamycin. Therefore, introduction of the GBS vaccine, particularly the hexavalent conjugate, is vital to preventing infection and reducing the number of patients with STSS. Ongoing monitoring using antimicrobial susceptibility testing and serotyping of GBS that caused STSS will help monitor resistance to penicillin and clindamycin, lethality impact, and circulating serotypes.

MATERIALS AND METHODS
Bacterial isolates. STSS is classified as notifiable by the Infectious Disease Control law in Japan. Cases of STSS due to GBS are reported annually: 2014, 31; 2015, 36; 2016, 59; 2017, 80; 2018, 87; 2019, 123; 2020, 121; 2021, 127. STSS-causing GBS isolates were obtained from pathogen collections of the National Institute of Infectious Diseases and prefectural Public Health Institutes (PHIs) from all over Japan. Data on streptococcal infections and clinical isolates were sent to PHIs from cooperating hospitals. We collected data from seven reference center branch offices in the PHIs of Fukushima, Tokyo, Kanagawa, Toyama, Osaka, Yamaguchi, and Oita. For this study, adults were defined as individuals over 18 years old. The diagnostic criteria for STSS were established by the Centers for Disease Control and Prevention in 1993 (8). They included identification of GBS from a normally sterile site, septic shock, and multiorgan failure. All GBS isolates obtained were cultured from sterile body sites of patients with STSS. Isolates were part of standard patient care. This study complied with the guidelines of the Declaration of Helsinki and was approved by the institutional individual ethics committees for the use of human participants (National Institute of Infectious Diseases Ethic Review Board for Human Subjects approval no. 19). The study was considered public health surveillance as defined in Article 15 of the Act on the Prevention of Infectious Diseases and Medical Care for Patients with Infectious Diseases (1999). Thus, informed consent was not required. An API 20 Strep kit was used to identify the strains (bioMérieux, Marcy l' Etoile, France). The Lancefield grouping was determined using the Prolex streptococcal grouping latex kit (Iwaki & Co., Japan).
Capsular serotyping. Capsular serotyping of each isolate was performed using the latex agglutination method with specific antisera against Ia to IX capsular polysaccharide antigens (Statens Serum Institute, SSI Diagnostica, Denmark).
Detection of erythromycin resistance genes. The genes responsible for erythromycin resistance, ermA, ermB, mefA, and msrD, were detected via PCR using previously published primer sequences (see Table S1 in the supplemental material) (27)(28)(29). Briefly, the reaction mixture (25 mL) contained 1 mL DNA template, 1 mL 10 mM each primer, 0.2 mL 5 U/mL AmpliTaq Gold DNA polymerase, 2.5 mL 10 mM deoxynucleoside triphosphates, 2.5 mL 10ÂAmpliTaq Gold buffer (2.5 mM MgCl 2 Plus; Applied Biosystems, Waltham, MA, USA), and 17.3 mL water. The PCR conditions were 95°C for 20 s, 48°C for 25 s, and 72°C for 1 min for 30 cycles, followed by an initial denaturation of 95°C for 10 min and a final extension step of 72°C for 1 min.
Analysis of pbp. Five genes encoding high-molecular-weight penicillin-binding proteins (PBP1a, PBP1b, PBP2a, PBP2b, and PBP2x) were amplified and sequenced in STSS isolates as described previously (30). The sequencing products were compared with the S. agalactiae A909 NCBI reference strain.
Statistical analysis. Statistical analyses were performed using EkuseruToukei v.4.02 (Social Survey Research Information Co., Ltd., Tokyo, Japan). Data were compared using Fisher's exact test. Differences were considered significant at P values of ,0.05.
Data availability. The nucleotide sequences of the pbp genes from all penicillin G-resistant GBS strains tested in this study were deposited in EMBL/GenBank through DDBJ under accession no. LC731361 to LC731376.

SUPPLEMENTAL MATERIAL
Supplemental material is available online only. SUPPLEMENTAL FILE 1, PDF file, 0.02 MB.