Virulence and pathogenicity of group B Streptococcus: Virulence factors and their roles in perinatal infection

ABSTRACT

This review summarizes key virulence factors associated with group B Streptococcus (GBS), a significant pathogen particularly affecting pregnant women, fetuses, and infants. Beginning with an introduction to the historical transition of GBS from a zoonotic pathogen to a prominent cause of human infections, particularly in the perinatal period, the review describes major disease manifestations caused by GBS, including sepsis, meningitis, chorioamnionitis, pneumonia, and others, linking each to specific virulence mechanisms. A detailed exploration of the genetic basis for GBS pathogenicity follows, emphasizing the roles of capsules in pathogenesis and immune evasion. The paper also examines the molecular structures and functions of key GBS surface proteins, such as pili, serine-rich repeat proteins, and fibrinogen-binding proteins, which facilitate colonization and disease. Additionally, the review discusses the significance of environmental sensing and response systems, like the two-component systems, in adapting GBS to different host environments. We conclude by addressing current efforts in vaccine development, underscoring the need for effective prevention strategies against this pervasive pathogen.

Introduction

Streptococcus agalactiae (group B Streptococcus; GBS) is a gram-positive, low-GC content, encapsulated bacterium that causes several forms of serious infection in pregnant women, fetuses, and infants. While GBS is occasionally identified as a cause of local or systemic infection in older children or adults, the preponderance of GBS disease worldwide occurs during the perinatal period.

The developmental specificity of GBS raises questions about what molecular interactions between GBS and the pregnant, fetal, or infant host predispose these populations to GBS infections. This review will present an overview of virulence mechanisms that contribute to colonization and disease. To emphasize connections between specific bacterial molecular features and clinical aspects of GBS infection, the discussion of virulence factors is organized around the different manifestations of perinatal GBS disease.

Historical context

At the beginning of the 20th century, GBS was known as a cause of zoonotic disease, mainly mastitis in infected cattle udders, but not a health threat to people. Rebecca Lancefield first characterized GBS in 1932 based on the presence of a carbohydrate-containing surface moiety, the group B antigen, whose reactivity to specific antisera in a precipitation assay could be used to distinguish the species from other hemolytic streptococci with reactivity to group A, C, D, or E antisera [1].

Starting in the early 1960s, scattered reports emerged from maternity and newborn care facilities of severe neonatal infections caused by GBS [2–4]. This emergence was initially viewed as a curiosity. “The group B Streptococcus is a stranger to most pediatricians,” wrote one author in 1973 [5]—a statement that was soon to become outdated as an alarming acceleration of GBS neonatal infection incidence occurred over the subsequent decade. By 1979, a prospective study of GBS colonization and infection rates revealed a frightening attack rate of approximately 0.5% of newborns experiencing GBS sepsis or meningitis [6].

This history prompts consideration of whether the observed emergence of GBS perinatal infection reflected a true epidemiologic shift or, alternatively, if the mid-century explosion of GBS diagnoses was due to new recognition of a disease process that had been previously missed or mischaracterized. Strong evidence indicates in the former interpretation – that GBS in fact leapt from bovine hosts to humans following World War II, driven by tetracycline overuse in the years following the discovery of antibiotics. Cunha et al. performed phylogenetic analysis of genome sequences from a set of 229 human pathogenic isolates. The group used imputed molecular clock mutation rates to determine that the major GBS clades responsible for human infection arose from a tetracycline selective bottleneck in the mid-twentieth century, corresponding closely with the observed clinical emergence. The convincing interpretation of the data is that human virulence genes (some discussed later in this review) were co-selected with tetracycline resistance islands, facilitating zoonotic transmission from antibiotic-treated cows to humans in an early example of selection for pathological bacteria due to indiscriminate antimicrobial overuse [7].

Disease manifestations (Figure 1)

Figure 1.

Major disease manifestations of GBS in the fetus, neonate, and adult (left) [8–26] and modes of transmission with timing of neonatal disease manifestations (right) [27–29].

Sepsis

Worldwide, the most common manifestation of GBS infection in the perinatal period is bacteremia [8,9]. When systemic spread of GBS through the bloodstream co-occurs with organ dysfunction – examples include hypotension, respiratory distress, or digestive symptoms such as ileus – the condition is termed GBS sepsis.

GBS sepsis can occur in isolation or in combination with any of the other disease manifestations described in this section. Clinical signs of GBS sepsis in the newborn can develop precipitously, with respiratory and vital sign instability spiraling over a few hours, or more insidiously – such as with mild hypothermia or blood glucose abnormalities punctuating otherwise normal newborn appearance and behavior.

If left untreated, however, these subtle initial signs will generally escalate and evolve into more severe clinical features. The variable, sometimes initially mild, clinical attributes of GBS sepsis make it a challenging disease for pediatric healthcare professionals, requiring a high index of suspicion and careful vigilance to effectively intervene during the earliest stages before severe multiorgan injury takes place.

Meningitis

Meningitis. GBS has neonatal central nervous system (CNS) tropism, with a high propensity to invade and survive within cerebrospinal fluid and brain parenchyma, where it often causes severe permanent injury [10–12]. Some of the molecular features that potentiate neonatal GBS meningitis have been elucidated and are described below. Major facilitators of GBS homing to the blood–brain barrier and transiting across it include adhesins, such as HvgA, as well as molecular enablers for intracellular survival and escape, such as two-component systems that allow environmental adaptation and immune evasion.

GBS meningitis typically begins with bacteremia. Therefore, many early signs of GBS meningitis are the same as GBS sepsis: respiratory distress, temperature and blood glucose instability, and lethargy. Once CNS invasion has occurred, seizure likelihood increases. Seizures from GBS can be focal or generalized and may or may not be clinically apparent. Therefore, any newborn with unexplained mental status changes must have GBS meningitis considered as part of the differential diagnosis and should be tested microbiologically and by electroencephalogram as soon as possible.

Chorioamnionitis

GBS colonizes the adult intestine and reproductive tract. Up to one-third of healthy adult women has GBS vaginal colonization, which can fluctuate in presence or density over time [13–15], but which very rarely causes disease in the nonpregnant state. During pregnancy, however, GBS can ascend through the protective cervicovaginal mucous plug, entering the uterine cavity.

Following uterine ascension, GBS can invade and persist within multiple pregnancy tissues. Infiltration of the placental decidua, villi, and fetal membranes permits entry into the amniotic fluid and the fetus. Clinically, these processes present as chorioamnionitis: a constellation of maternal fever, pro-inflammatory laboratory findings such as leukocytosis, and signs of labor such as contractions, rupture of the amniotic sac, and cervical dilation. GBS chorioamnionitis during early pregnancy is increasingly recognized as a major global contributor to stillbirth [8,16]. When it occurs after the point of fetal extrauterine viability, GBS chorioamnionitis often leads to premature delivery of an already septic neonate. These are challenging neonatology cases, where the effects of physiologic immaturity are exacerbated by bacterial infection and an already powerful pro-inflammatory cascade.

Pneumonia

Neonatal GBS pneumonia usually presents in cases of early-onset infection (within the first week of life) and often co-occurs with GBS chorioamnionitis. The mechanism of pathogenesis in these cases is likely fetal inhalation of GBS-infiltrated amniotic fluid, permitting bacterial seeding of the airways and lung parenchyma. GBS pneumonia inactivates surfactant, leading to atelectasis, air leak syndrome, and poor overall neonatal lung function. Because of the limited barrier capabilities of neonatal alveoli, GBS early-onset pneumonia is often complicated by bacteremia and systemic GBS spread.

Soft-tissue infections

While bacteremia, sepsis, and meningitis are the most common forms of neonatal GBS infection, infants and toddlers outside of the neonatal window can develop localized GBS infections such as cellulitis, fasciitis, osteomyelitis, and septic arthritis [17–19]. These localized skin and joint infections can also occur in adults, among whom diabetic patients are particularly susceptible to GBS growing in foot ulcers [20].

Urinary tract infections

Heavy vaginal GBS colonization is sometimes complicated by urinary tract infection [21–23], history of which is considered an independent risk factor for neonatal GBS complications, warranting automatic intrapartum antibiotic prophylaxis (discussed below) and heightened vigilance among those caring for the newborn.

Disease in older adults

While GBS is an uncommon cause of infection among healthy children, teens, and adults after the neonatal period, susceptibility rises among elderly populations and those with immunodeficiency from other diseases [17,24,25]. Sepsis, meningitis, and severe soft tissue infections are possible among these medical subgroups, with some evidence accruing that incidence of adult GBS disease may be becoming more common over time [26].

Modes of transmission (Figure 1)

Early-onset disease. GBS infection in the newborn is referred to as early-onset disease (EOD) if it occurs during the first 7 days of life after birth. Confusingly, some GBS research publications use a 72-hour-of-life cutoff to distinguish between EOD and late-onset disease (LOD, discussed below). However, in recent years the major professional and regulatory agencies involved in tracking GBS epidemiology and issuing guidelines (such as the United States Centers for Disease Control and Prevention, the American Academy of Pediatrics, and the American College of Obstetricians and Gynecologists) have adopted the seven-day definition [27–29].

The key distinction between EOD and LOD is based on mode-of-transmission rather than clinical signs, treatment approaches, or outcomes, which are similar for EOD and LOD regardless of the exact definition employed. The crucial difference is that EOD results from fetal exposure to GBS, either through exposure to chorioamnionitis or (more commonly) exposure to maternally asymptomatic colonization of the vaginal canal. EOD is therefore partially preventable through administration of preventative antibiotics to the mother during labor, known as intrapartum antibiotic prophylaxis (IAP).

Late-onset disease

GBS infection after the first week of life is classified as LOD. Unlike EOD, LOD stems from postnatal exposure to GBS. This may result from vertical transmission of GBS from the mother’s skin or breast milk or may be from acquisition of GBS from another close contact or environmental exposure to a contaminated surface. Because GBS colonization can be transient and may be established stochastically in previously uncolonized hosts, prior maternal negative screening for GBS colonization does not confer significantly lower LOD risk to the infant, nor does maternal receipt of IAP. In most cases, the source of GBS in the case of LOD is never determined. As an infant matures, however, the risk of LOD decreases, such that most presentations occur within the first 3 months of life, with only rare and tapering rates of “very late onset” GBS infection occurring later in the postnatal period.

Treatment and prevention

Treatment

GBS remains universally sensitive to β-lactam antibiotics, including first-generation agents such as penicillin and ampicillin. While there have been a few reports of borderline GBS sensitivity to β-lactam antibiotics [30,31], there has never been documented emergence of high-level resistance. This fact is both fortunate and curious. Given the frequency and promiscuity of GBS genetic recombination, sustained universal susceptibility to penicillin and derivative antimicrobials cannot be considered assured. The potential for GBS to become resistant to clinically relevant antibiotics is highlighted by the significant (and increasing) rates of erythromycin and clindamycin resistance among patient-derived isolates [32]. Both are antibiotics that have played roles in GBS prevention in the past, but which must now be deployed cautiously and only after careful resistance testing of the isolate in question.

For now, patients with suspected or confirmed GBS infection are generally treated with intravenous penicillin or ampicillin. Combination therapy with a β-lactam and another class of antibiotic, such as an aminoglycoside, is sometimes used to speed GBS clearance [33]. Duration of therapy is determined by the site of the infection (meningitis and bony infections require more prolonged treatment than uncomplicated sepsis or a urinary tract infection) and – in some cases – follow-up microbiological testing to ensure complete GBS eradication.

Prevention

The only widespread approach to preventing GBS perinatal infection is administration of IAP to women with documented rectovaginal colonization who present in labor. Pregnancy screening for GBS dates to the 1990s, based on recommendations by the Centers for Disease Control and Prevention and the American Academy of Pediatrics following observations of the increased neonatal risk of infection after exposure to birth canal colonization. Starting in 2002, the United States adopted practice policies of universal microbiological screening for maternal GBS colonization and administration of IAP during labor in cases of positive colonization screens [28].

Penicillin and ampicillin are the mainstay antimicrobial agents for IAP. The mechanism of action is believed to be twofold: direct reduction of the intravaginal colonization burden and transplacental passage of the antibiotic to the fetus, conferring enhanced neonatal colonization protection during the first week of life when vulnerability to overgrowth and invasion is highest.

Several rounds of refinement to universal GBS screening and IAP guidelines have occurred in the United States over the past two decades. Specific recommendations for different case scenarios, including protocols for maternal β-lactam allergy and presentation in preterm labor, exist but are outside the scope of this review. Interested readers are directed to relevant publications [27,29].

Three key points regarding IAP for GBS prevention are worth emphasizing. First, these policies – adopted by a handful of countries worldwide, but by no means all – are effective at lowering the risk of EOD but have had no effect on LOD. EOD reduction in the United States has been roughly fivefold since universal screening and IAP were introduced, yet LOD incidence has not measurably changed [28]. This fact reflects the distinct modes of pathogenesis for the two categories of GBS disease. IAP can only influence intrapartum vertical transmission, therefore only affects EOD. Second, universal screening and IAP practices are best suited to term deliveries and offer suboptimal protection in the setting of preterm labor and delivery. Most mothers who present in preterm labor have not yet undergone routine screening (now recommended between pregnancy weeks 36 and 37 in the United States) so their colonization status is not known. IAP can be started in this setting and is often recommended depending on the precise circumstances. However, if labor began due to GBS chorioamnionitis or proceeds too quickly for IAP to effectively clear the birth canal of GBS and reach therapeutic levels in the fetus, its administration may be subtherapeutic. Finally, universal screening and IAP programs constitute a major healthcare expense for any country undertaking them and require logistical coordination that is not available in many places around the world [34]. Therefore, IAP – while undeniably a major public health success in the battle against perinatal GBS infection – remains an imperfect approach.

GBS Cell Biology

Genetic makeup and recombination. The single, circular chromosome encoding the GBS genome ranges in size from 1.53 Mb to 2.38 Mb [35]. It contains approximately 2,000 protein-encoding genes and a relatively high number of tRNA-coding sequences (approximately 80), ABC transporter genes (approximately 60), and two-component system genes (see below) [36–38]. Approximately 20% of putative open reading frames encoded on the GBS chromosome remain uncharacterized by automated annotation engines, leaving open opportunities for discovery of new molecular mechanisms and systems [39]. Large-scale bioinformatic analyses of GBS genomes from diverse sources have revealed evidence of internal recombination, prophage and plasmid integration, and migration of mobile genetic elements [40–45].

Evidence of GBS genetic recombination lays the foundation for the concept of the GBS open pan-genome introduced by Tettelin et al. in 2005 [46]. The pan-genome of a species encompasses a core genome containing a set of genes shared by all strains and a dispensable genome made up of genes only present in a subset of strains. An open pan-genome – a characterization that applies to GBS – implies that the dispensable genome is theoretically infinite. In GBS, the core genome consists of approximately 1,800 genes, continually recombined in nature with a limitless set of newly introduced genes and genetic material.

Different genetic classification systems exist for GBS. The capsular polysaccharide synthesis operon can be specified based on which of the 10 capsular serotypes its encoded enzymes synthesize (Figure 2) [48–51]. Non-typable and acapsular strains are also found in nature, thought to result from mutations in capsule synthesis or regulatory genes or hybridization between existing capsule types – the latter phenomenon highlighting the possibility of emergence of new serotypes [52]. Multilocus sequence typing (MLST) is a serotype-independent genotyping approach, which in GBS has revealed the existence of 11 major clonal clusters (CC) [53]. These two classification systems are correlative but independent and can also be combined with newer classification systems such as CRISPR-based strategies [54].

Figure 2.

GBS capsule mediates functions essential to the colonization of the host and immune system evasion. The varying arrangements of five monosaccharides (key in top right) give rise to the molecular structures of the 10 known GBS polysaccharide capsule types. The percentage contribution of each serotype toward global incidence of human disease (in red) varies, suggesting a pathogenic advantage of certain serotypes over others [47]. Sialic acid is found in every capsular serotype and its location at the outward facing ends of the polysaccharide chains cloaks GBS from the immune system through host cell mimicry.

Genetic basis for human pathogenicity

The majority of human pathogenic GBS strains are capsular serotype III, CC17. The next most common are serotype Ia, CC23. However, all 10 serotypes and all 11 CCs are, theoretically, cross-compatible with each other; common serotype-CC linkages are not immutable and many counterexamples are described in the literature [55–58]. Several studies have used diverse approaches to investigate whether specific genes or chromosomal regions – perhaps those associated with the most common pathogenic GBS isolates – directly confer virulence potential. A 2019 study used ELISA analysis of multiple cytokine responses by macrophage-like THP-1 cells to a panel of GBS with different genotypes. The results indicated that serotypes and CCs more associated with human disease (serotype III, CC17, for example) induced stronger pro-inflammatory responses by cultured leukocytes [59]. A pan-genome-wide association study examined the genome sequences of approximately 2,000 strains and classified them by host species and whether the isolate was from asymptomatic colonization or invasion. The authors identified 279 CC-specific genes, including genes whose presence correlated with virulence potential. Some of the CC-specific genes also presented SNP variants with potential functional significance, pointing to GBS selective adaptation to niches and sub-niches [44].

Another study used bioengineering to decouple capsular serotypes from MLST backgrounds. A serotype switched CC17 strain, in which the genetic locus encoding serotype III was exchanged for a serotype Ia genotype, was compared to wild type and acapsular strains of both backgrounds in a mouse model of vaginal colonization. Encapsulated strains outcompeted unencapsulated strains, while the serotype switched capsule-type Ia/CC17 strain persisted more than the WT capsule type III/CC17 and the capsule-type Ia/CC7 progenitors [60]. These results suggest that while the presence of capsule is important for vaginal colonization, the combined influence of capsule and non-capsule genetic determinants is likely complex and difficult to predict, again pointing to GBS adaptability to highly specific environmental niches.

Capsule structure and contribution to virulence

(Figure 2). The structural composition and orientation of GBS capsule from all 10 serotypes have been biochemically determined [61–67] and this work has been condensed and summarized in reviews [68,69]. GBS capsule is synthesized from the cross-linked glucose, galactose, N-acetylglucosamine, rhamnose, and N-acetylneuraminic acid (Figure 2).

Early mutagenesis experiments demonstrated significant reductions in GBS virulence in animal models when the capsule was lost [70]. A robust subsequent body of work showed that among the structural components of the GBS capsule, terminal N-acetylneuraminic acid decorations on the outermost capsule surface play a critical role in capsule-mediated immune evasion. N-acetylneuraminic acid is the most abundant form of sialic acid in human biology (Figure 2). An initial observation supporting the hypothesis that this sialic acid moiety was critical in protecting GBS from the host was the observation that GBS treatment with the terminal sialase enzyme neuraminidase rendered the organism susceptible to antibody-independent bactericidal activity through alternative pathway deposition of the complement molecule C3b [71]. Further experimental refinement demonstrated a structural mechanism by which terminal sialic acid decorations on serotype III GBS lead to impaired complement deposition by the alternative pathway [61].

Several studies have described GBS enzymatic modification of N-acetylneuraminic acid and incorporation into the crosslinked polysaccharide capsule [72–75]. While some bacteria are capable of de novo sialic acid synthesis, this process has not been conclusively demonstrated in GBS. Clear evidence does exist, however, of GBS utilization of environmental sialic acid. A conserved sialic acid catabolic operon, when mutated, significantly affected GBS ability to grow in semidefined media in which sialic acid was the major carbohydrate source. Expression of genes in this operon was upregulated in the presence of sialic acid and downregulated in human blood or in the presence of high glucose concentrations [76]. Furthermore, exogenous sialic acid supplementation was shown to enhance GBS colonization of nasal, pulmonary, and intravaginal epithelial surfaces in mice [76]. GBS growth in human serum was shown to increase expression of sialic acid on the exterior of serotype III GBS, with corresponding decreased sensitivity to complement deposition [77]. This result, in combination with the whole blood-mediated effect on decreased sialic acid catabolism, suggests multifactorial GBS regulation of sialic acid utilization and externalization in different colonization and invasion scenarios.

Importantly, for efforts to develop GBS vaccines targeting the outer surface of the polysaccharide capsule, the inhibitory effect of sialic acid on complement deposition can be overcome through the classical complement pathway in antiserum with capsule-specific antibody above a threshold concentration [71]. The exact antibody concentration necessary for GBS serologic protection remains incompletely defined, and varies by capsular serotype, with significant implications for vaccine development and regulatory approval. While earlier studies on naturally occurring antibody concentrations suggested that anti-GBS capsule titers in the 1–10 μg/mL range were required for protection across all serotypes [78], a more recently developed capsule targeting candidate vaccine demonstrated likely 75% and 95% protection from early-onset or late-onset disease at neonatal IgG concentrations of 0.184 and 0.827 μg/mL, respectively [79].

An additional mechanism by which the GBS capsule contributes to bacterial evasion of the immune system is through molecular mimicry surface interactions with sialic acid-recognizing immunoglobulin superfamily lectins (Siglecs) (Figure 2). Human Siglecs occur in multiple subclasses. Several Siglec subclasses are involved in inhibitory signaling that suppress autoimmune responses to widespread sialic acid surface features on self-cells. Naturally sialylated GBS interacts with multiple Siglec subclasses in a sialic acid-dependent manner [80]. GBS interactions with inhibitory Siglecs dampen innate immune responses to GBS encounters, such as neutrophil oxidative burst, release of extracellular traps, and impaired GBS killing [81]. Macrophage depletion of the inhibitory Siglec-E subclass led to increased pro-inflammatory cytokine release and increased GBS clearance in low-dose in vivo mouse models of GBS bacteremia, meningitis, and pneumonia. However, in high-dose challenging conditions, the increased reactivity of Siglec-E deplete macrophages led to a cytokine storm with accelerated lethality compared to wild-type hosts [82]. Further study of Siglec-sialic acid influences on GBS pathogenesis has suggested that developmental mismatch between immune inhibitory Siglec-E expression (present at adult levels in neonatal lung alveolar macrophages) and expression of immune activating Siglec-1 (relatively underexpressed in neonatal lung alveolar macrophages) may contribute to neonates’ developmental susceptibility to GBS pneumonia [83].

Environmental sensing and response systems

The major mechanism by which bacteria sense and respond to external molecular stimuli is through the coordinated action of two-component signal transduction systems (two-component systems; TCS) [84]. TCS consist of two proteins, most often encoded by paired genes in a co-regulated two-gene operon. The first component is a surface-anchored, sensing histidine kinase that detects a specific molecular stimulus and transmits that reception event through a conformational change that leads to phosphorylation of an intracellular signal transmission domain. The second component is an intracellular response regulator with its own phosphorylation site(s). Dynamic activation and deactivation of the response regulator by the surface histidine kinase modulates transcription of a specific set of genes (the “regulon” of that TCS) due to DNA binding and release by the response regulator.

Compared to closely related bacteria, GBS encodes a relatively high number of TCS—38 were identified in a 2014 study, well-above the typical range of 15 to 32 TCS in other members of the order Lactobacillales (adjusted for genome size) [38]. Not all TCS are part of the GBS core genome (17 are universally shared), implying that different GBS TCS may be specifically adapted to particular host species or environmental niches. The majority of GBS TCS has not been experimentally investigated; hence, their roles in colonization and virulence are not determined. On the other hand, a few key TCS have been unequivocally shown to contribute to GBS disease processes by enabling survival and propagation in protected host environments. The GBS TCS that contribute to virulence are discussed below. A recent review specifically focused on GBS TCS provides additional details on the complete set [85].

CovR/S

The most-studied GBS TCS is CovR/S, which in early papers following its discovery was also labelled CsrR/S [86,87] due to its homology to the group A Streptococcus CsrR/S system with virulence regulatory activity [88]. While nonessential for GBS survival in laboratory conditions [88,89], CovR/S is an important cellular switch that controls a regulon influencing the balance between GBS commensal and invasive lifestyles. Initial observations of GBS ΔcovRS knockout mutants revealed increased production of the pigmented β-hemolysin/cytolysin toxin (Figure 4) described below, decreased production of the sphingomyelinase-sensitive pore-forming CAMP factor (Christie, Atkins, Munch-Petersen factor, also described below), increased adherence to epithelial cell culture surfaces, increased sensitivity to micronutrient limitation, and decreased virulence in a neonatal rat model [88].

Figure 4.

GBS β-hemolysin/cytolysin production is regulated through control of the biosynthetic cyl operon, which is typically repressed by the CovRS two-component system. However, repression can be alleviated through Stk1 alternative phosphorylation of CovR, Abx1 regulation of CovS, or direct increase in cyl transcription by RovS transcriptional regulator [90–94]. (top right) the 12 genes of the cyl operon and their designated biosynthetic roles, colored to match corresponding portions of the structure (center) of GBS β-hemolysin/cytolysin [95–99]. (bottom) the contribution to pathogenicity and typical cell types targeted by GBS β-hemolysin/cytolysin [100–106].

GBS CovR/S senses environmental pH, although its environmental ligand has not been identified. Under acidic conditions, such as would be expected in the vaginal lumen, GBS adhesion to cultured epithelial cells and extracellular matrix proteins is generally reduced compared to neutral conditions, where adhesion is promoted. Deleting covRS results in significantly increased host epithelial cell adhesion, suggesting regulated suppression of pro-adhesion genes in the regulon [86,107], an interpretation supported by the fact that pH modulation of adhesion is reduced (although not entirely abolished) in ΔcovRS mutant GBS [86]. While ΔcovRS mutants have enhanced constitutive adhesion phenotypes, they show impaired intracellular survival in epithelial and phagocytic cells [107,108]. Overall, these results indicate that the CovR/S system balances phenotypic responses to host environments in a way that optimizes fitness.

Several molecular mechanisms related to CovR/S have been established. In addition to phosphorylation-driven control by CovS, CovR activity has also been shown to intersect with the phosphorylation status of the eukaryotic-like serine-threonine kinase Stk1. CovS and Stk1 have opposing influences on CovR activity across its regulon. Phosphorylation of the CovR D53 residue by CovS leads to mostly repressive transcriptional influence across the CovR regulon, a major exception being the CAMP factor gene, which is upregulated by ^PD53 CovR. Residue T65 phosphorylation by Stk1, however, leads to a mostly derepressed CovR regulon, with the exception of CAMP factor, whose expression is decreased by ^PT65 CovR [90,91] (Figure 4). These dual influences on CovR activity are thought to allow GBS to fine-tune toxin, immune evasion, and metabolic responses to diverse surrounding environments.

Another surface enzyme, Abx1, has been shown – using a two-hybrid reporter system – to interact directly with CovS, modulating its kinase and phosphatase activity in response to as-yet undetermined environmental signals. Overexpression of Abx1 led to CovS becoming fixed in a phosphatase conformation, deactivating CovR and generally upregulating the expression of virulence genes including adhesins and β-hemolysin/cytolysin (Figure 4). Congruently, deletion of the abx1 gene froze CovS in its activating kinase conformation, tipping the CovR regulon into a generally repressed state with minimal expression of virulence genes under its influence [92].

A recent multimodality examination of the CovR/S regulon across multiple strains included genomic analysis, transcriptomics, and genome-wide chromatin-binding studies. This work demonstrated multiple ways in which GBS can adapt the CovR/S response network – including through gene mutation, promoter changes, and evolution of CovR itself – to optimize its fitness across multiple environments and maximize its versatility in response to potential host defenses [109].

SaeR/S

The SaeR/S TCS was discovered based on its upregulation during exposure to vaginal lavage fluid, suggesting responsiveness to factors present during female genitourinary colonization [110]. The GBS SaeR/S protein sequences resemble proteins from a homologous signal transduction system in Staphylococcus aureus [111,112]. A GBS ΔsaeR mutant was found to have differential expression of 300–470 genes during laboratory outgrowth, depending on the media employed (chemically defined medium or Todd-Hewitt broth supplemented with yeast). However, the same study examined the ΔsaeR differential transcriptome during vaginal colonization and revealed only three differentially expressed genes in this more stringent environment. One of the three differentially expressed genes during vaginal colonization encodes a plasminogen binding adhesin, pbsP; deletion of saeR led to a significant decrease in pbsP expression in vivo. Targeted deletion of the pbsP gene, in turn, led to significantly reduced GBS density in the mouse vaginal colonization model – an effect that could be complemented by re-adding the pbsP gene in trans on an expression plasmid [110]. This suggests that the SaeR/S system responds to environmental stimuli in the vaginal tract by upregulating an adhesin that promotes epithelial attachment. Additional investigation of the SaeR/S system [113] demonstrated CovR/S co-regulation of pbsP and another SaeR/S controlled gene, bvaP, the latter encoding a secreted factor that contributes to vaginal colonization and biofilm formation [114]. In both cases, the SaeR/S and CovR/S influences on gene expression were inverted, with SaeR/S acting as a dominant enhancer and CovR/S acting as a counterbalancing repressor of gene expression [113].

CiaR/H

A GBS transposon mutant library screen examined mutants with enhanced susceptibility to antimicrobial peptides [115]. The goal was to identify genes that allow GBS persistence in intracellular environments where host innate immune responses – including elaboration of antimicrobial peptides – would be overwhelming without resistance mechanisms.

The screen identified a TCS locus called ciaR/H, which also has homologs in related Streptococci [116–118]. GBS ΔciaR and ΔciaH transposon insertion deletion mutants both had increased sensitivity to the antimicrobial peptide mCRAMP [115]. Confirmatory targeted deletion of the ciaR gene resulted in increased sensitivity to mCRAMP, polymyxin B (a different cationic antimicrobial peptide), reactive oxygen species that GBS would be exposed to during neutrophil phagocytosis or degranulation, and the cell wall degrading enzyme lysozyme. The apparent contribution of ciaR/H to innate immune resistance was consistent with observed defects in intracellular survival following coincubation with cultured hBMEC cells [115].

The molecular mechanism of GBS CiaR/H control of virulence is, in large part, due to the regulation of small RNAs that affect coding RNA elongation, intracellular persistence, and translation [119]. Four small RNA transcripts, which had previously been identified through GBS RNA-seq [120], were shown to be significantly regulated by CiaR/H activity. Altered expression of the CiaR/H small RNA regulon affects expression of pullulanase – an enzyme that influences human extracellular matrix binding and carbon utilization – and biofilm propensity in GBS [121].

Intriguingly, the CiaR/H regulatory system may also intersect with the native type-II clustered regularly interspaced short palindromic repeats (CRISPR)/Cas9 system. A Δcas9 GBS strain had genome-wide differential gene expression revealed through RNA-seq. In a mouse model of vaginal colonization, the Δcas9 mutant showed decreased persistence relative to wild-type control. It also showed decreased cerebral invasion in a mouse model of GBS meningitis. The differentially expressed gene loci in Δcas9 were not obviously related to targeting protospacers in the CRISPR array but overlapped significantly with the CiaR/H regulon. Of 58 differentially expressed genes in the ΔciaR mutant, 53 were also dysregulated in Δcas9 [122]. The mechanism for regulatory crosstalk between CiaR/H and GBS CRISPR/Cas9 is not known.

LiaR/H

Part of a three-gene operon that also includes liaF, which encodes a negative regulator of LiaS [123], the LiaR/H system – like CiaR/H – influences GBS sensitivity to cell wall damaging innate immune molecules and antibiotics. Targeted deletion of liaR increased sensitivity to cell wall-active antibiotics vancomycin and bacitracin and antimicrobial peptides polymyxin B, colistin, and nisin [124]. In an in vivo sepsis model, the inoculum required for 50 percent lethality (LD₅₀) was approximately 100 times higher for ΔliaR compared to the wild type, and mouse mortality in a pneumonia model was significantly decreased following infection with the mutant [124]. The major gene components of the LiaR/H regulon encode a pilus subunit (pilus island 2B) and cell wall structural molecules such as penicillin-binding protein 2B, a peptidoglycan cross-linking peptide synthetase (MurN), and a membrane phospholipid modification enzyme thought to affect surface charge. This collection of regulated genes points to likely GBS surface feature changes in the ΔliaR mutant that account for its decreased virulence.

RgfA/C

The RgfA/C TCS was first described as regulating GBS surface features that control fibrinogen binding – a phenotype thought to be important for colonization of tissue surfaces and invasion into protected spaces such as the blood–brain barrier [125]. The RgfA/C system is encoded as part of a four-gene operon (rgfBDAC) that shows sequence homology to quorum-sensing gene networks in S. aureus and other gram-positive bacteria. In S. aureus, the homologous TCS (labeled AgrA/C) is responsive to a peptide pheromone AgrD, which is translated as a pre-peptide that is matured and secreted by the transmembrane endopeptidase AgrB [126,127], then sensed by neighboring S. aureus through AgrC, which relays signal reception to the effector molecule AgrA. The analogous predicted domains in the GBS rgfBDAC operon, which shares the same layout as S. aureus, suggest a similar interbacterial signaling mechanism operating in GBS.

The RgfA/C TCS is not universally encoded in GBS, but its presence is characteristic of the highly virulent CC17 clade, where it is generally co-expressed with two genes for fibrinogen-binding adhesins: fbsA and fbsB. When the three genes are present, as in CC17, the RgfA/C system regulates fbsA and fbsB, enhancing fibrinogen binding [128]. When the RgfA/C system was experimentally deleted and deficient mutants used in a mouse sepsis model, the knockout ΔrgfA/C strain showed enhanced lethality compared to wild-type control, with increased CFU densities from blood, brain, and lung samples [38]. Together with the finding that rgfA/C expression is growth-phase dependent – with maximal transcription during the late-log growth phase [125]—the results imply that this GBS TCS is involved in inter-GBS communication that coordinates expression of surface features, such as adhesins. Increased expression of fbsA and fbsB, two genes regulated by RgfA/C, increases GBS tissue association and invasion in animal models of disease.

HssR/S

Iron acquisition and homeostasis are essential metabolic functions necessary for bacterial survival. In contrast, iron sequestration is a defense mechanism evolved in eukaryotic hosts to deprive bacteria of iron access, limiting their fitness in protected anatomical compartments such as the bloodstream [129]. Heme, the iron-containing compound in hemoglobin within erythrocytes, is present in low concentrations in whole blood, and can be further released from erythrocytes through hemolysis. As an important potential source of nutritive iron, pathogenic bacteria such as GBS have developed systems for sensing and importing heme to harvest its iron for biochemical redox reactions.

The major heme sensing system in GBS is the HssR/S TCS, which coordinates transcription of the heme export system HrtA/B, genes for which are co-transcribed with the TCS on an operon. Deletion of hssRS leads to near-complete silencing of hrtA/B expression, indicating that – as in other gram-positives [130,131]—heme detection by the HssS sensor initiates signal transduction that turns on transcription of heme export system genes [132]. This pathway relieves toxicity that would otherwise arise from unchecked intracellular heme accumulation. ΔhrtA/B GBS without a functioning heme export system showed decreased infection density in blood-rich organs such as the heart, kidney, liver, and brain, demonstrating that regulation of intracellular heme concentrations is necessary for GBS survival during systemic invasion [132].

The mechanism by which heme is transported into GBS is not currently clear. A comparative genomic analysis investigating genes differentially expressed during vaginal colonization by group A and group B Streptococci identified a heme-binding protein, HupY, which may chaperone heme through an importer channel in group A Streptococcus [133]. While a HupY homolog is encoded by GBS, it does not appear to have the same role in chaperoning heme to an import channel; in fact, no such importer system has been identified in GBS, leaving the full picture of GBS heme utilization incomplete.

Surface proteome

GBS uses amino acid encoding of trafficking motifs to coordinate transport of externalized proteins to the cell surface, where they can be secreted or anchored in place [134,135]. Proteins that are anchored to the GBS cell wall can be expected to intercalate within the polysaccharide capsule, which may partially shield them [136]. However, the fact that surface-anchored GBS proteins show evidence of interaction with the external environment (such as surface adhesion and antibody binding) indicates that they are incompletely cloaked by the capsule.

The main and best-studied GBS cell surface-anchoring motif is a five-residue LPXTG pattern recognized by the sortase A (SrtA) enzyme. Deletion of srtA is expected to interfere with surface anchoring of the entire GBS LPXTG-bearing surface protein arsenal, allowing examination of the role of surface proteins en masse in virulence phenotypes. An early study of the GBS surface proteome used ΔsrtA and plasmid-complemented variants to show that the knockout had significant defects in cultured cell adherence, fibronectin binding, lethality in a rat pup sepsis model, and colonization of neonatal rat intestines – all of which were reversible through reintroduction of the functional srtA gene [137]. While this study points to the importance of GBS surface proteins in virulence mechanisms, it was not designed to pinpoint which of the LPXTG surface proteins were most important in disease processes. Additional study, however, has identified several key GBS surface proteins that predispose it to perinatal colonization, invasion, and spread.

Pilus

GBS encodes non-flagellar, polymeric externalized pili involved in adhesion, biofilm formation, and defense from host immunity. Two main pilus-encoding genetic islands were first described in 2006, each also encoding sortase variants (classified as sortase C enzymes) that assist with pilus biosynthesis. One of the pilus islands (Pilus Island 2; PI-2) has two variants (PI-2A and PI-2B) distinguished by the order of genes in the operon and the presence or absence of accessory signal peptidase and transcriptional regulatory genes [138,139]. GBS clinical isolates reliably express at least one of the two pilus islands [140]. The pilus island-encoded SrtC enzymes coordinate with SrtA to chaperone pilus anchoring, as a ΔsrtA mutant showed absence of externalized pili by microscopy and Western blot techniques [138].

GBS pili increase adhesion to epithelial surfaces in vitro and in vivo [138,141,142] and play an especially important role in adhesion to and invasion of brain microvascular endothelial cells-one facet that contributes to GBS central nervous system tropism [141,142]. One pilus protein in particular – the pilus tip adhesin PilA – was shown to have dual adhesion and immune activation roles through collagen binding and integrin-mediated signaling. This dual functionality establishes a pathogenic cascade of cytokine release and neutrophil recruitment to the brain microvasculature, resulting in localized barrier permeability at the exact site of GBS adhesion, creating a gap in the blood–brain barrier that sets the stage for GBS meningitis [142].

One GBS pilus subtype, Pilus 2A (encoded by the PI-2A genomic island) contributes to biofilm formation, which in turn is thought to promote persistent colonization and potentially protect GBS from immune clearance during infection [140]. Targeted mutations that completely remove pilus subtypes 1 or 2B had no effect on biofilm formation, whereas removal of pilus subtype 2A significantly limited biofilm formation, a result recapitulated with antibody targeting of pilus subtype 2A. Further targeted mutation of PI-2A operon genes revealed that one of the ancillary proteins in pilus subtype 2A, ancillary protein 2 (AP2), is the necessary component for pilus subtype 2A biofilm establishment. Deleting the ap2 gene, the srtC-1 gene that chaperones AP2, or the srtA gene universally required for pilus assembly and anchoring, all led to biofilm deficiency, whereas deletion of a different pilus subtype 2A ancillary protein (AP1) or its Sortase C gene (sortC-2) had no effect.

Finally, GBS pili have been shown to actively thwart immune clearance during infection by inhibiting phagocytic killing by macrophages and neutrophils [143]. As described earlier in the context of TCS protective effects, the mechanism of pilus-mediated immune evasion seems to be resistant to cationic antimicrobial peptides. The same study showed reduced persistence of a GBS ΔpilB mutant lacking a key pilus backbone protein when incubated with the antimicrobial peptide mCRAMP [143]. Resistance to phagocytic clearance, mCRAMP, and increased persistence in the mouse bloodstream were all inducible in nonpathogenic Lactococcus lacti when transformed with an expression plasmid that induced externalization of the GBS PilB protein. Remarkably, the recombinant L. lacti strain expressing GBS pilB, when used in a mouse model of sepsis, led to 100% lethality over 72 hours whereas no mice died in the L. lactis wild-type control infection [143].

Serine rich repeat (Srr) proteins

Another important class of GBS surface protein virulence factors are members of the Srr family. Srr proteins have been described in other Streptococci and Staphylococci species, in which they are recognized to play roles in adhesion to host surfaces [144,145]. In GBS, there are two Srr variants: Srr-1, which is more common, and Srr-2, whose expression is limited to strains from the hypervirulent serotype III, CC 17 lineage [146]. Srr proteins are transported by the SecA2 secretory system; during transport, Srr-1 and Srr-2 are extensively glycosylated by a pair of enzymes GtfA and GtfB. Once on the bacterial surface, these high molecular-weight, post-translationally modified molecules function as mucin-like structures that contribute to aggregation, attachment, and successful invasion of epithelial and endothelial barriers [147]. Srr protein size is variable, even among closely related GBS, due to recombination between the repeating nucleotide sequences in the coding sequence [148].

GBS Srr-1 was identified on the basis of sequence homology to other Srr proteins and was first shown to contribute to keratin binding and adhesion to cultured laryngeal epithelial cells with high keratin matrix expression [148]. Srr-2 was initially discovered as an immunoactive antigen separated from a serotype-specific capsule, followed by immunoprecipitation and characterization. The first report describing Srr-2 also found that a deletion mutant strain had an LD₅₀ approximately four log-fold higher than wild-type controls in a mouse model of sepsis, indicating Srr-2 plays a significant role in virulence [146]. Subsequent studies of GBS Srr-2 showed that this serine repeat-rich subtype binds to plasminogen, which Srr-1 does not, and shows greater binding affinity to fibrinogen than Srr-1. Plasma protein binding by Srr-2 contributes to in vivo aggregation of multi-cell GBS clumps, which increases the rate of phagocytosis. Nevertheless, in a mouse model of GBS meningitis, Srr-2 deficient GBS showed less persistence in brain tissue than Srr-2 expressing serotype III, CC 17 control GBS. Recombinant Srr-2 N-terminus injections also showed promise as a candidate vaccine or vaccine component, reducing meningitis lethality by approximately 40% [149].

Fibrinogen binding proteins

Fibrinogen is a soluble plasma glycoprotein that acts as a precursor to the important blood clotting polymer fibrin. Bacterial binding to fibrin can facilitate transit in the bloodstream, cloaking from phagocytes, and bacterial fixation at sites of injury, which can contribute to further spread (see Figure 3) [150]. The first GBS fibrinogen binding protein identified was FbsA. After carefully characterizing GBS-fibrinogen binding and showing that proteinase treatment of GBS ameliorated the interaction (indicating an externalized protein factor as key to the phenomenon), the team used an Escherichia coli cosmid library screen to identify the GBS gene responsible for conferring fibrin affinity [151].

Figure 3.

Molecular crosstalk and resulting pathogenesis phenotypes of GBS during systemic invasion of the bloodstream. GBS can invade the bloodstream with strategies including: evading the immune system (top left), adhering to damaged tissue (middle left), directly damaging tissue (bottom left), and effectively utilizing heme as a source of iron (right side of coccus). These responses are regulated by a molecular network within GBS (coccus to right). The environment is sensed by histidine kinases anchored in the plasma membrane (CovS, RgfC, HssS, and SaeS). The histidine kinases communicate with response regulators, indicated by the ovals (CovR, RgfA, HssR, and SaeR) which then act as transcription factors for various virulence genes including surface expressed proteins, peptidases, export channels, and toxins. These virulence factors contribute to the phenotypic responses observed when GBS is in the blood. Bloodstream enriched ligands of GBS surface proteins include fibrinogen (which binds FbsA, FbsB, and C5a peptidase), plasminogen (binds PbsP), and C5a complement factor (binds C5a peptidase). Toxins include β-hemolysin/cytolysin (βH/C), CAMP factor, and other enzymes that can be transported in membrane vesicles. Dashed lines indicate contributions to phenotype. Dotted lines indicate an indirect mechanism.

The FbsA protein structure is composed of variable numbers of repeated amino acid motifs with a canonical “GNVLERRQRDAENRSQ” sequence, subject to minor variation in some of the repeats. The number of repeats among strains expressing fbsA tested in the initial sample set ranged between 3 and 30. The repeated motifs comprise the fibrinogen-binding site on the protein, and the number of repeats correlates directly with fibrinogen binding affinity when tested with recombinant FbsA [151]. A later structural inquiry suggested that GBS FbsA folds into a coiled helix, the length of which is determined by the number of repeated units. The primary sequence of GBS FbsA does not resemble that of closely related fibrinogen-binding proteins from other gram-positive bacteria, suggesting a unique evolutionary history of this protein [152]. An increased susceptibility to clearance in human blood by ΔfbsA suggests a role for this surface factor in resisting opsonophagocytosis during pathogenesis [151].

Two years after the discovery of GBS FbsA, a second GBS fibrinogen binding protein – FbsB – was reported [153]. Its identification stemmed from the same E. coli cosmid screen whose performance revealed FbsA. Sequencing showed a modest 22% identity to a fibronectin-binding protein in Streptococcus pyogenes. Recombinant GBS FbsB demonstrated saturable binding to human fibrinogen in vitro. Interestingly, however, a ΔfbsB GBS mutant was not fibrinogen binding deficient, but did show a defect in adherence and invasion to cultured lung epithelial cells [153]. A later study, however, demonstrated an approximately 80% decrease in fibrinogen binding by a ΔfsbB mutant in a CC17 strain background. These discrepant findings suggest that the roles of fibrinogen-binding protein subtypes in GBS may be dependent on capsule subtypes or other surface features of specific lineages.

C5a peptidase

Human pathogenic GBS strains bind and inactivate complement factors that threaten their survival in the bloodstream [154]. The major effector of this virulence capability is a surface-anchored C5a peptidase encoded by the scpB gene with sequence homology to the group A Streptococcus scpA C5a peptidase gene [155]. C5a cleavage prevents GBS opsonophagocytosis as evidenced by foundational studies showing decreased neutrophil attachment among GBS strains with constitutively high scpB expression relative to those with lower expression in coincubations that included recombinant C5a [156].

Subsequent studies indicated that C5a peptidase plays an additional, moonlighting role as an adhesin. A phage display library constructed in a GBS serotype Ia strain and selected in a fibronectin adhesion screen pointed to scpB as a mediator of extracellular matrix attachment in this experimental design. The screen finding was confirmed by comparing fibronectin binding between a ΔscpB knockout, wild type, and a complemented control strain in which scpB had been reintroduced to a plasmid. Fibronectin binding was found to be partially scpB-dependent, pointing to its dual role in pathogenesis [157]. Almost simultaneous to this study, a second, independent group identified the adhesin moonlighting role of GBS C5a peptidase, demonstrating that ΔscpB GBS had decreased invasion of two cultured human cell lines and decreased adhesion to fibronectin and that GBS invasion of some cell lines could be enhanced by co-treatment with soluble fibronectin [158]. Examination of naturally occurring GBS strains with scp coding variants that eliminate C5a peptidase activity but do not affect fibronectin binding indicate that the two functions of the protein are mediated by physically distinct domains [159]. A later study showed that only when fibronectin surface density nears saturation – juxtaposing multiple N-termini of the molecule next to each other – does the C5a peptidase-binding capability become revealed, suggesting a cryptic interaction with a binding pocket formed by a dense fibronectin web as might be encountered at a local wound site [160].

Plasminogen binding surface protein (PbsP)

Another GBS surface molecule that interacts with ubiquitous circulating host protein, PbsP was identified from sequence homology with a plasminogen binding adhesin in pneumococcus [161]. Plasminogen is a proenzyme whose activation contributes to clot breakdown and extracellular matrix turnover. Like earlier examples of important GBS surface-trafficked virulence factors, PbsP promotes GBS disease by latching the bacteria to proteins that inherently localize to sites of local injury. Expression of PbsP is regulated by the SaeR/S and CovR/S systems, the latter of which represses pbsP at baseline but can derepress the gene upon GBS exposure to the bloodstream or other immune surveilled spaces [162].

Since its initial discovery in 2016, PbsP has been shown to facilitate plasminogen-dependent invasion of brain microvasculature endothelial cells, promoting meningitis in the setting of bacteremia [161,162]. Deletion of the pbsP gene in a serotype III, CC17 GBS background significantly decreased brain tissue CFU density, release of pro-inflammatory cytokines, and mortality in a mouse model of meningitis. Vaccination with PbsP protein conjugated to a glutathione-S-transferase adjuvant significantly decreased lethality and bacterial burden [162].

Hypervirulent GBS adhesin (HvgA)

First discovered in 2010, HvgA was identified from genomic analysis of hypervirulent, serotype III, CC17 isolates from neonatal and adult infections [163]. Like multiple other GBS virulence factors, its expression is upregulated in vivo compared to during in vitro laboratory growth.

HvgA is an adhesin and its expression in L. lactis confers ability to adhere to diverse epithelial and endothelial cell cultures [163]. Importantly for neonatal pathogenesis, HvgA is especially crucial for translocation across intestinal and blood–brain barriers-two steps believed to be indispensable for late-onset meningitis pathogenesis [163]. In the intracranial bloodstream, HvgA demonstrates robust interactions with clathrin, which facilitates transcytosis across the epithelium of the choroid plexus located in the lateral, third, and fourth ventricles of the brain [164].

SHP/RovS system

Bacterial cell-to-cell communication can play an important role in population survival and coordination of responses to environmental threats. A recently described Streptococcal communication system is centered around a secreted short hydrophobic peptide (SHP) that can be imported by a receiver cell after diffusion through intercellular space. Once internalized, SHP activates a transcriptional regulator of the Rgg subclass, which in GBS is encoded by rovS [93,94] (Figure 4).

The SHP/RovS system is part of the core GBS genome, reliably present in clinical isolates [94]. First demonstrated to be functional in GBS through reporter gene constructs, it was then shown that the SHP/RovS system was important in regulating the FbsA fibrinogen-binding gene, and that disruption of this signaling chain ameliorated liver and spleen invasion in a mouse model of sepsis [94].

β-hemolysin/cytolysin (βH/C)

GBS grown on plates or in monoculture produce a distinctive, brick-, or rust-colored pigment. This coloration typically appears in the center of a GBS colony, starting as a central “bullseye” with a surrounding paler ring and gradually expanding outward as the colony ages. If grown in liquid medium, a centrifuged GBS pellet may display varying degrees of red/brown pigmentation, which is also sometimes visible in the supernatant.

This pigmentation comes from the expression of the cytotoxin known as βH/C, or – in earlier publications – sometimes referred to as Granadaene (from the Granada-based laboratory of Manuel Rosa-Fraile, et al., who performed foundational studies on the molecule and its related phenomena) [95,165]. βH/C – expression of which can be induced in specialized media containing methotrexate or other folate metabolism inhibitors [166] and which can be isolated in aqueous solvents saturated with starch or other dissolved polysaccharide matrices [167–169]—plays an important role in GBS pathogenesis, although its exact mechanism is not fully elucidated.

Early descriptions of βH/C recognized its relation to cytotoxicity. GBS strains with high constitutive expression of βH/C form wide bands of β-hemolysis on blood-containing solid media, and starch-purified βH/C is hemolytic in solution [169]. Rosa-Fraile et al. performed NMR studies that pointed to an unusual molecular structure: a non-proteinaceous ornithine-rhamno-polyene, the polyene linker between the ornithine and rhamnose containing 24 carbons [165] (Figure 4). This structure is a carotenoid-like pigment (explaining its coloration and characteristic spectrophotometric absorbance spectrum); carotenoids are produced by various other bacteria, although generally they are not directly cytotoxic [170]. Carotenoid molecules have antioxidant properties, and βH/C has been shown to neutralize ROS in chemical- and cell-based assays [100,101]. Controversy over whether this GBS carotenoid pigment is sufficient to cause cytotoxicity or whether it is associated with an as-yet undefined protein product has persisted since its description. A pair of papers describing a) induction of cytotoxicity from L. lactis transformed with the GBS cyl operon, whose enzymes are responsible for pigment production [96] and b) cytotoxicity from laboratory-synthesized βH/C have seemingly resolved the question [97] in favor of the ornithine-rhamno-polyene being sufficient to cause cytotoxicity without a protein cofactor.

βH/C seems to play multiple roles in GBS biology and pathogenesis. Like other carotenoids, it is an antioxidant with protective effects for the organism [100]. However, it is also directly linked to the ability of GBS to invade a variety of epithelial, immune, and parenchymal cell types in vitro and in vivo [102–106]. It has been demonstrated to be particularly important in intrauterine infection, promoting GBS invasion of placental and fetal tissues, and driving pyroptotic inflammation associated with preterm birth and stillbirth [98,171]. The immunomodulatory role of βH/C in other tissues and cell types appears significant in disease processes beyond chorioamnionitis and prematurity, including meningitis [103] and urinary tract infection [172]. Generally, βH/C seems to induce a pro-inflammatory valence in cells it affects, although exceptions have been reported, such as a potential anti-inflammatory effect on macrophages after GBS phagocytosis [105]. The immune influences of βH/C therefore remain an active area of investigation.

Genetic regulation of βH/C

The genes that encode the β/HC biosynthetic cascade are reliably grouped and co-transcribed in an operon labeled cyl [96]. The 12 genes in the cyl operon have been examined from multiple perspectives, both experimentally and in functional domain-based models, conclusively establishing this locus as the necessary and sufficient chromosomal region for βH/C production [95,96,98,99] (Figure 4).

The cyl operon is negatively regulated by the CovR/S system, such that transition to acidic or other environmentally inhospitable sites leads to de-repression and increased expression of βH/C (Figure 4). Alterations to covR/S operon expression have been identified in some hypervirulent strains of GBS in which, generally, constitutively decreased CovR/S activity causes upregulation of cyl transcription and overproduction of βH/C [99,173]. Other environmental stress responses, such as the bacterial stringent response, have been implicated in cyl regulation [174], as have significant changes in the environmental nutritional milieu, such as shifts in arginine or glucose abundance [175,176].

CAMP factor

Historically used as a diagnostic tool for identifying GBS from clinical swabs, CAMP factor is a secreted peptide with a molecular weight of approximately 25 kDa. CAMP factor accelerates GBS-mediated β-hemolysis when potentiated with supplemental sphingomyelinase, which releases ceramide from erythrocyte membranes. This phenomenon creates a characteristic arrowhead-shaped zone of hemolysis when CAMP factor-producing GBS is streaked on a blood agar plate perpendicular to a sphingomyelinase (also known as β-factor)-producing strain of S. aureus [177]. Unlike βH/C, CAMP factor is an oligomerizing, protein-based, pore-forming toxin that creates 12–15 nm, circular pores in target cells, although the exact role that ceramide plays in its biochemical reactivity is not determined [178,179].

GBS membrane vesicles

Surface budding of diffusible membrane vesicles is a mechanism by which GBS transports protein and non-protein virulence factors into the surrounding environment. First identified microscopically and reported in a 2016 paper, GBS membrane vesicles were shown to contain CAMP factor, PcsB, hyaluronate lyase, and other degradative enzymes that promote tissue breakdown and GBS invasion [180]. Cell-free GBS membrane vesicles were shown to travel anterograde through the uterus in a mouse pregnancy model, causing progressive inflammation and tissue breakdown to fetoplacental units along both arms of the bicornate murine uterus. Brownian motion of these small (the average diameter is around 25 nm), virulence factor-laden structures may partly explain how non-motile GBS is able to passage from the vaginal canal into the intrauterine space [180]. GBS membrane vesicles also contain βH/C – in proportion to the production of the toxin by their bacterial source cells – which likely contributes to cytolytic breakdown of tissues against which these structures make contact [181] and pyroptotic injury to pregnancy tissues during GBS chorioamnionitis [181,182] (Figure 4).

Current efforts at eradication and conclusion

The past decade has seen increasingly coordinated and successful efforts to develop a GBS vaccine, premised on the concept of transplacental IgG trafficking to the fetus to neutralize systemically invading GBS during early life [183]. While no set protocol has been established and no vaccine is currently authorized, clinical trials up to phase II have shown promise for single- or dual-dosed vaccine administration to pregnant people during their first trimester. Such a regimen confers antibody titers to newborns that have, over decades of observation and experimentation, been seen to be efficacious at reducing sepsis risk.

Two vaccine designs are currently in lead contention for future adoption: a conjugated, polyvalent capsular polysaccharide formulation [79,184,185] and a surface-protein-based formulation [186–188]. While questions of efficacy against EOD, premature chorioamnionitis, and LOD have yet to be resolved, and patient willingness to accept a new vaccine for administration during pregnancy may be a hurdle, there is no doubt that efforts to achieve a scalable, targeted, and safe way to reduce the human toll of GBS – a major global perinatal pathogen – are needed and welcome. Readers seeking a detailed discussion of the current state of GBS vaccine development may be interested in a selection of several focused reviews on this topic [69,189,190].

While GBS infection remains a significant threat to pregnant and newborn hosts worldwide, significant strides in understanding its molecular mechanisms of colonization, invasion, and interconversion between these distinct lifestyles give hope to the view that – within the near future – new preventative and treatment approaches will render GBS perinatal infection a problem of the past.

Funding Statement

The author(s) reported there is no funding associated with the work featured in this article.

Disclosure statement

No potential conflict of interest was reported by the author(s).

Author contributions statement

CJM assisted with writing and editing; SMC, EJG, and GHH assisted with writing, editing, and figure preparation; TAH conceptualized the article, contributed writing, and assisted with editing. All contributing authors have read and approved the final manuscript.

Data availability statement

No new data are included in this article. Data in referenced prior studies may be accessible in the original works.