ABSTRACT
Streptococcus agalactiae (group B Streptococcus [GBS]) is an important cause of invasive infection in newborns, maternal women, and older individuals with underlying chronic illnesses. GBS has many mechanisms to adapt and survive in its host, and these mechanisms are often controlled via two-component signal transduction systems. In GBS, more than 20 distinct two-component systems (TCSs) have been classified to date, consisting of canonical TCSs as well as orphan and atypical sensors and regulators. These signal transducing systems are necessary for metabolic regulation, resistance to antibiotics and antimicrobials, pathogenesis, and adhesion to the mucosal surfaces to colonize the host. This minireview discusses the structures of these TCSs in GBS as well as how selected systems regulate essential cellular processes such as survival and colonization. GBS contains almost double the number of TCSs compared to the closely related Streptococcus pyogenes and Streptococcus pneumoniae, and while research on GBS TCSs has been increasing in recent years, no comprehensive reviews of these TCSs exist, making this review especially relevant.
INTRODUCTION
Streptococcus agalactiae, also called group B Streptococcus (GBS), is a beta-hemolytic Gram-positive bacterium found to asymptomatically colonize the vaginal and gastrointestinal tracts of approximately 10% to 30% of the healthy population. In neonates, maternal women, the elderly, and immunocompromised adults, GBS is an important cause of mortality and morbidity, causing diseases such as bacteremia, sepsis, meningitis, pneumonia, urinary tract infections, osteomyelitis, and skin and soft tissue infections (1–6).
GBS is most commonly recognized as an important pathogen of neonates. Neonatal GBS infections are classified as early onset disease (EOD), occurring in the first 7 days after birth, or late onset disease (LOD), which develops between 1 week and 3 months postpartum. GBS is transmitted vertically from mother to child before or during labor due to the aspiration of contaminated amniotic and bodily fluids (7–9), potentially resulting in invasive neonatal diseases such as pneumonia and meningitis. GBS has also been shown to be transmitted in the breast milk, which may play a role in the development of diseases such as sepsis and meningitis, especially in preterm infants (10, 11). In some countries, including the United States, colonization-positive maternal women are treated with intrapartum antibiotic prophylaxis (IAP). Though IAP has aided in the reduction of EOD, it has little to no impact on LOD rates. Additionally, preliminary evidence indicates IAP delays the development of a healthy gut microbiota of the newborns (12, 13), potentially increasing the risk for other types of infections such as Escherichia coli (14).
GBS encounters many different environmental niches and, as such, has evolved numerous systems to perceive changes in its external environment and make adaptations as needed to ensure survival (15–17). The adaptation dynamics exhibited by GBS to successfully colonize its host and cause disease are often under the control of one or more regulatory systems controlling bacterial processes such as adherence and invasion of host tissues and resistance to host immune responses (18). One such mechanism employed by GBS to adapt to a changing environment is the use of two-component regulatory systems (TCSs). These systems are widely distributed and structurally varied between and within species.
Canonical TCSs contain a histidine kinase (HK) protein associated with the cell membrane. Environmental sensing results in the autophosphorylation of the cytoplasmic domain of sensor HK, which then phosphorylates the cognate cytoplasmic response regulator (RR) protein (19, 20). RRs generally control cellular behaviors through direct transcriptional repression or activation of one or more genes, resulting in adaptation to the particular environmental signal (21). To date, at least 22 distinct TCSs have been identified in GBS (22). This number is almost double that of any other Streptococcus species, with 13 identified in S. pneumoniae (23) and S. pyogenes (24). This suggests GBS may have a higher capability of monitoring environmental conditions and reacting to changing stimuli.
Surprisingly, despite the obvious importance of TCSs in GBS, no comprehensive reviews of the structure and function of these TCSs exist. This review will discuss the domain architecture of many of the GBS TCSs described to date. Importantly, the downstream effects of signaling through TCSs, specifically regarding GBS virulence, basic metabolic processes, and adherence to mucosal surfaces, will be discussed. While many of the TCSs have been named and described, many remain uncharacterized. In this review, we will primarily use the names and GBS strain A909 (serotype Ia) gene numbers or the numbered nomenclature previously published (22) unless specifically stated.
STRUCTURE OF GBS TWO-COMPONENT REGULATORY SYSTEMS
Sensor kinases.Sensor HKs are generally made up of sensory and signal transduction domains. The sensory domains can vary greatly in both sequence and structure and can sense signals extracellularly, in the membrane, or intracellularly in the cytoplasm. In GBS, approximately half of the known TCSs are predicted to be transmembrane sensors and the other half extracellular sensors (Table 1). It is uncertain as to whether any of the GBS TCSs contain cytoplasmic sensor kinases. It was previously hypothesized that HK-8 (Sak_0846) may be a cytoplasmic sensor, but there is no direct evidence for this as yet (22).
Two-component systems of GBSe
Extracellular sensors generally consist of two transmembrane domains separated by a large extracellular protein domain. As seen in Fig. 1 and Table 1, the GBS histidine kinases containing putative extracellular sensing domains include BgrS (Sak_0188), VncS (Sak_0702), CiaH (Sak_1079), HK-12 (Sak_1358), CovS (Sak_1638), DltS (Sak_1813), FspS (Sak_1907), HK-18 (Sak_1921), HK-19 (Sak_1993), and HK-20 (Sak_2062). Extracellular sensors often directly interact with their cognate signals or environmental cues (25) such as solutes (26) and changes in nutrient conditions (21, 27).
Structure and domain architecture of GBS TCS histidine kinases. (Left) Schematic of intramembrane histidine kinases (IM-HK) and individual sensor kinase domains of predicted IM-HKs in GBS. (Right) Schematic of extracellular histidine kinases and sensor kinase domains of predicted GBS extracellular sensors. Domains predicted using Pfam (https://pfam.xfam.org/) and SMART (http://smart.embl-heidelberg.de/). Predicted transmembrane domains are shown in blue, HAMP domains are represented by green squares, histidine kinase domains are red diamonds, and H-ATPase domains are gray hexagons. Pink lightning bolts signify the likely site of signal sensing. Domain architecture is not fully to scale. (Adapted from mBio [22].)
Some GBS HKs primarily contain sensing components located in the membrane rather than large extracellular domains. These intramembrane histidine kinases (IM-HKs) may have just two transmembrane regions with a short linker peptide in the extracellular space, while others have five or more transmembrane domains connected by linkers (Fig. 1). IM-HKs often sense membrane changes, integral membrane components, or compounds that affect the integrity of the membrane (26). GBS has at least nine potential IM-HKs associated with TCSs (Fig. 1 and Table 1), including HssS (Sak_0175), HK-2 (Sak_0248), HK-3 (Sak_0380), LiaS (Sak_0391), SaeS (Sak_0468), BceS/NsrK (Sak_1072), LtdS (Sak_1112), RgfC (Sak_1917), and HK-21 (Sak_2066).
Generally, following the transmembrane domains, sensor kinases have a short peptide region that links the sensor domain to the HK portion of the protein. Some HKs contain a HAMP domain (contained in histidine kinases, adenylyl cyclases, methyl-accepting proteins, and phosphatases) made up of alpha-helices of approximately 50 amino acids that play a role in regulating phosphorylation levels through conformational changes in ligand-binding domains (28). As shown in Fig. 1, in GBS, several of the HKs are predicted to contain HAMP domains (SMART; http://smart.embl-heidelberg.de).
The C-terminal HK portion of the sensor generally contains a dimerization and phosphorylation domain with a conserved His residue as well as an ATPase domain known as H-ATPase (histidine kinase-like ATPase). Following signal sensing, it is generally a conformational change in the sensor portion of the protein that signals autophosphorylation of the conserved His residue via phosphate transfer from ATP. Details of the mechanisms and structures of histidine kinases have been reviewed and published elsewhere and are beyond the scope of this review (21, 29).
Response regulators.TCS response regulators (RRs) usually contain a conserved Asp residue in the N-terminal receiver domain which is phosphorylated following autophosphorylation of the His residue in the HK. Phosphorylation of this conserved residue triggers protein changes, allowing the C-terminal effector domain of the RR to alter gene expression. In GBS, all of the identified RRs are predicted to bind DNA to directly affect transcription (22). The C-terminal effector domains of GBS RRs mostly fall into three categories: the LytTR family, the LuxR family, and the OmpR/PhoB or family (Table 1). More than half of GBS RRs fall into the OmpR/PhoB family (also known as Trans_reg_C family; SMART [http://smart.embl-heidelberg.de/]). Structurally, these RRs contain a winged helix or helix-turn-helix (HTH) domain that binds DNA (30). LytTR family RRs are slightly overrepresented in GBS compared to that in prokaryotes in general (22) and include RR-2 (Sak_0249), LtdR (Sak_1111), and RgfA (Sak_1918). These RRs contain a novel winged HTH domain that differs in structure and sequence from the OmpR winged helix domains (31) and, in many bacteria, is associated with regulation of virulence factors (32).
PATHOGENESIS AND SURVIVAL IN A HOST
Bacterial virulence factors enable the organism to replicate and spread in the host through host tissue damage and successful subversion of host immune responses. TCSs often play a central role in the production of virulence factors, and in GBS, several TCSs regulate virulence-related genes. Some researchers speculate that the regulation of GBS virulence factors may play a role in determining whether a particular strain acts as an asymptomatic colonizer or an invasive pathogen (33).
Control of virulence: CovRS.Perhaps the most well-studied TCS in GBS is the CovRS (also called CsrRS) system, which controls the expression of more than 100 genes, depending on the strain studied (34, 35). The GBS CovRS system has 83% amino acid identity and 91% similarity with the well-studied CovRS ortholog in S. pyogenes (36). Genes regulated by the CovRS TCS allow for the transition between commensal and opportunistic lifestyles (37). When activated, CovS phosphorylates CovR at its conserved aspartic acid residue at position 53 (35, 38, 39). Consequently, CovR binds to target promoters to directly alter transcription of a large number of GBS genes (40). Although the primary signals sensed by CovS have not been completely defined, some evidence suggests that, at least in strain 2603 V/R (serotype V), CovR activation is at least partially dependent on pH fluctuations sensed by CovS (41).
Two concurrent studies first reported on the GBS CovRS TCS as a homolog of the well-studied CovRS system in S. pyogenes (group A Streptococcus [GAS]), and both studies demonstrated that CovRS mutants in three GBS strains, NEM316 (serotype III), 515 (serotype Ia), and 2603 V/R (serotype V), displayed a hyperhemolytic phenotype and increased expression of the cyl operon but decreased production of cationic antimicrobial peptide (CAMP) factor, a small secreted pore-forming toxin produced by GBS that augments the hemolytic activity of Staphylococcus aureus sphingomyelinase (β toxin) (34, 38). Both studies also examined the effects of CovRS on bacterial pathogenesis. Lamy et al. demonstrated that a ΔcovRS mutant in strain NEM316 had significantly decreased virulence in a neonatal rat model of virulent infection following intraperitoneal injection, as seen by a 3-log increase in 50% lethal dose (LD50) values for the ΔcovRS mutant (34). Jiang et al. used two different GBS strains, 515 and 2603, and evaluated LD50 using an adult murine model of GBS infection with intraperitoneal injection (38). They observed a similar 3-log increase in LD50 values for the ΔcovRS mutants in strain 515 and a 1-log increase in 2603, demonstrating decreased virulence in strains lacking CovRS. These early experiments laid the groundwork for future work examining the role of CovRS in GBS virulence (38).
Global gene expression studies have been undertaken on CovR mutants in at least three GBS strains, NEM316 (34), 2603 V/R, and 515 (42). While many genes make up a core regulon of CovRS in all three strains, there are also genes that are regulated only in certain strains, suggesting that the full CovRS regulon between strains of GBS is likely much more heterogeneous and complex than we currently appreciate (42). Interestingly, unlike the S. pyogenes CovR, which acts primarily as a repressor, data indicate GBS CovR may also act as an activator depending on the gene target (42).
Further work has examined the role of the CovRS system in a variety of tissue culture and pathogenic animal models of GBS infection. In vivo experiments using a murine urinary tract infection (UTI) model demonstrated that a CovR mutant in a serotype III GBS strain 874391 was less able to colonize the bladder and had a decreased bacterial load in the urine, demonstrating the importance of this TCS in UTI pathogenesis. Additionally, the authors found that factors regulated by CovR affect bladder cytokine production in mouse and human uroepithelial cell death (37).
Although studies like those discussed above have found that deletion of the CovRS system results in decreased virulence, in some cases, ΔcovRS mutants are hyperinvasive and hypervirulent. Using human amniotic epithelial cells (hAECs), it was shown that deletion of covR in the hypervirulent strain COH1 (serotype III) resulted in a hyperhemolytic phenotype and significantly increased inflammation and bacterial invasion. These phenotypes were associated with negative regulation of the cylE gene by CovR (43). The authors propose that the cyl operon encodes an ornithine rhamnolipid that is both hemolytic and cytotoxic and that upregulation of this pigment in ΔcovRS mutants subsequently promotes invasion into the amniotic cavity (43).
An A909 covR mutant strain was used to examine adherence and invasion into human brain microvascular endothelial cells (hBMECs). No significant differences were observed in adherence but the ability of ΔcovR mutants to invade was significantly diminished, indicating that these mutants could be impaired in invasion into the brain endothelium. Interestingly, using a murine meningitis model following intravenous injection, the authors observed hypervirulence in this strain, with increased proinflammatory signaling and bacterial penetration into the brain in a ΔcovR mutant (40).
Bacterial colonization of the mucosal surfaces is often a precursor to more invasive infections, and the ability of GBS to bind to the vaginal mucosa is an important aspect of both its pathogenic and asymptomatic lifestyles. Using a human vaginal epithelial cell (HVEC) line (VK2/E6E7), Patras et al. examined the role of CovR in the regulation of inflammatory processes and attachment and invasion of the epithelium (44). A microarray examining epithelial cell gene expression showed that increased expression of genes involved in immune regulation included proinflammatory cytokines and chemokines. They also observed that the A909 ΔcovR strain showed increased adherence but decreased invasion in cultured HVECs; although it is difficult to tease out whether this is due to decreased ability to invade or decreased intracellular bacterial viability of the mutant. In addition, the A909 ΔcovR mutant had decreased persistence in the vaginal tract in a murine model of vaginal colonization, which corresponded to increased inflammatory cytokines and neutrophil invasion in vivo (44).
Adding a layer of regulation to GBS colonization and virulence are proteins that interact with, and regulate, CovRS. The Abi domain family protein Abx1, a member of a large transmembrane protein family, couples with the histidine kinase CovS to form a signaling complex. Abx1 directly interacts with the transmembrane and extracellular domains of CovS, impacting the catalytic cytoplasmic domain and helping to maintain the phosphatase/kinase state of CovS. Interaction between Abx1 and CovS likely affects conformational changes in CovS, altering the switch between phosphatase and kinase states and subsequently altering downstream activation of CovR (45). In the NEM316 strain, overexpression of Abx1 stabilized CovS in a phosphatase conformation, while an Abx1 knockout strain locked CovS in the kinase state (45). Generally, it is thought that the equilibrium between the kinase and phosphatase states of CovS is sustained by Abx1.
Like CovS, CovR is also regulated by another factor, the serine threonine kinase Stk1 (39). Unlike CovS, Stk1 does not phosphorylate CovR at the conserved Asp D53 but at the threonine Thr65 residue, which then decreases phosphorylation at D53. Utilizing phosphomimetic and silencing substitutions of CovR in strain A909, phosphorylation at T65 rather than D53 was observed to result in decreased promoter binding by CovR (39). This is an example of an efficient fine-tuning mechanism to optimize adaptation to the environment in a strain-specific manner (42).
Data looking at the role of CovRS in pathogenesis and colonization show that the model system, route of infection, and strain can greatly affect observed outcomes. There are likely complex and strain-specific gene regulatory networks controlled by CovRS that fine-tune the ability of GBS to both colonize the host and cause active infection at different host sites. CovR has been shown to be responsible for the transcriptional repression of more than 75 genes and activation of more than 60 genes in GBS, including cylE, cell wall and metabolic components, CAMP factor, and other putative transcriptional and virulence factors (46); though there is likely a core regulon, heterogeneity also exists in the CovRS regulon between strains, and strain-specific pathogenic outcomes may be observed.
bac gene regulation system: BgrRS.bgrRS (sak_0188/0189, referred to as sak188/189 in A909) encodes a TCS found on a putative pathogenicity island approximately 8,992 bp in size, adjacent to the bac (sometimes referred to as bag) gene (47). Bac encodes C protein β-antigen, an important GBS virulence factor. β-Antigen (Bac) contains a 73-amino-acid domain in the N terminus which has been identified as the binding site for interaction with the Fc region of human IgA (48). IgA effector functions are inhibited through the binding of β-antigen, potentially contributing to GBS immune evasion (49). β-Antigen also binds human factor H (FH) (50), a complement inhibitor. It is hypothesized that β-antigen binding to FH may promote degradation of C3b, preventing proper phagocytosis and again promoting host immune evasion by GBS (50). In addition to IgA and FH, Bac also binds to a third ligand, human sialic acid-binding immunoglobulin-like lectin 5 (Siglec-5), a leukocyte cell-surface receptor. β-Antigen binding to human Siglec-5 results in impaired oxidative burst, malfunction of leukocyte phagocytosis, and decreased production of neutrophil extracellular traps, generally improving the conditions for optimal bacterial survival in the host (51). Overall, β-antigen is an important GBS virulence factor, and its regulation is important for in vivo survival.
In GBS strain 168/00 (serotype Ib), deletion of BgrR greatly decreased β-antigen expression and virulence in a mouse model using intraperitoneal infection of GBS. In the absence of BgrS, BgrR is still able to upregulate bac gene transcription and β-antigen expression, indicating potential cross talk between BgrR and other sensor kinases (52). The BgrRS TCS is thought to be found in approximately 80% to 90% of GBS strains, but some strains include naturally occurring mutations in both bgrR and bgrS that impair their ability to upregulate bac (53).
The bgrRS TCS was previously characterized as a homolog of GBS TCS-6 (sag0616/0617) (22). It appears this is a mischaracterization, as the A909 homolog of TCS-6 is Sak_0701/0702, with greater than 98% protein sequence conservation. The BgrRS system is likely a separate TCS, making the current total number of described TCSs in GBS 22 rather than 21 as previously reported (Table 1).
Heme-sensing system: HssRS.Most bacteria require heme as an iron source and a metabolic cofactor; however, high levels of heme can cause toxicity due to its redox activity. While some bacteria have their own machinery for the synthesis of endogenous heme, GBS, a heme auxotroph, does not carry the enzymes required for autosynthesis and, as such, relies on exogenous heme uptake from the environment. In many organisms, heme toxicity is controlled primarily by the heme-regulated transport efflux system, HrtBA, which is under the regulation of the HssRS TCS. HssRS homologs control the expression of HrtBA in response to environmental heme (54–56). The HssRS system senses high heme concentrations and upregulates expression of HrtBA to export heme, preventing toxicity (56).
In GBS, like in other organisms, control of heme toxicity is essential for virulence, and in the NEM316 derivative NEMJ18 strain, ΔhrtBA mutants are less pathogenic than the wild type (WT) in a mouse model of systemic infection using intravenous inoculation (57). In vivo, the GBS response to heme appears to be organ dependent. Using the intravenous infection model, the authors found that ΔhrtBA mutants are more impaired in survival in blood-rich organs such as the heart and kidney but survive similarly to the WT in the brain and liver (57). Unlike other organisms where HrtBA and HssRS are encoded by separate operons, in GBS, these four genes are located on the same operon, suggesting that both expression and activation are controlled by heme concentrations (57).
Regulation of pathogenesis and host colonization: LtdRS.GBS is a major cause of neonatal meningitis leading to postinfection sequelae and death. In order for GBS to affect the central nervous system, it must first be able to resist host immune defenses in the bloodstream and cross the blood brain barrier (BBB). The LtdRS TCS was recently shown to play a role in both GBS vaginal colonization and the development of meningitis in the hypervirulent GBS strain COH1. Infection of human cerebral microvascular endothelial cells (hCMEC) with the ΔltdR mutant caused increased secretion of proinflammatory cytokines, interleukin-8 (IL-8), CXCL-1, and IL-6. This correlated with both increased invasion into hCMEC and greater penetration of the BBB in a murine model of meningitis (58). These data indicate that LtdRS plays an important role in the pathogenesis of GBS meningitis. Interestingly, LtdRS also affects GBS mucosal colonization. The ΔltdR mutant triggered increased production of cytokines from human vaginal epithelial cells versus that by COH1 WT cells, and in a murine model of vaginal colonization, the mutant was more rapidly cleared from the vaginal tract than a WT strain (58).
Transcriptome sequencing (RNA-seq) experiments in COH1 showed differential expression of many genes in a ΔltdR mutant, including many genes involved in metabolism. In addition, scanning electron microscopy demonstrated that a ΔltdR mutant formed excessive aggregates and clumps (58). These results suggest varied processes may be controlled by LtdRS, and further characterization of this system is needed. Taken together, these data indicate that LtdR, and likely the LtdRS TCS, aids in regulating the inflammatory response to GBS in the epithelial cells of both the brain and vagina. A ΔltdR mutant is more invasive and less able to colonize the mucosa, suggesting that LtdRS potentially promotes colonization and suppresses pathogenesis of GBS.
COLONIZATION AND ADHESION TO MUCOSA
During colonization, GBS binding to mucosal surfaces is mediated via attachment to host molecules such as laminin (Lm), fibronectin (Fn), and fibrinogen (Fbg), components of the basal membrane and extracellular matrix (ECM). ECM components contribute to streptococcal adhesion to the host, thereby improving host colonization (35, 59–61). They also aid in the internalization of GBS into host tissue and evasion of host immune response. GBS expresses many surface proteins, cell wall-anchored lipoproteins, and secreted proteins which influence host binding, and many of these are regulated by TCSs.
Regulator of fibrinogen binding: RgfAC.The rgf locus of GBS is made up for four genes, rgfBDAC, homologous to the S. aureus agr quorum sensing system. rgfBDAC encode a potential ABC transporter (RgfB), a peptide (RgfD), and a TCS with a sensor kinase (RgfC) and response regulator (RgfA). RgfD was identified as a putative autoinducer quorum sensing molecule that is necessary for maximal expression of rgfC (62). The four-gene operon is cotranscribed and most highly expressed during late exponential growth (63). In addition, expression of rgfAC is repressed by the CovRS TCS in at least one strain of GBS, A909 (40).
RgfAC was initially identified in strain O90R (serotype Ia) for its role in the upregulation of factors involved in GBS binding to fibrinogen. This binding was observed to be growth phase dependent, leading to the hypothesis that RgfAC acts in a cell density-dependent manner. In addition, the authors found that the C5a peptidase gene scpB was upregulated in an rgfC mutant (63). Based on these data, the authors hypothesized that the RgfAC TCS likely regulates components of the bacterial cell surface. It was later shown that, in a highly invasive clonal complex 17 (CC17) strain, RgfAC causes upregulation of one of the two important fibrinogen-binding protein genes in GBS, fbsB (64). In addition, a ΔrgfC mutant in CJB111 (serotype V, CC1) demonstrated hypervirulence in a mouse model of meningitis (22). Together, these data suggest a role for RgfAC in promoting a colonization versus the virulence phenotype.
Portions of the rgf TCS can be found in all serotypes of GBS, but the system is not present or functional in all strains. Many GBS isolates have point mutations in rgfA, resulting in a truncated protein lacking its C-terminal DNA-binding domain, rendering RgfA nonfunctional. Even in strains such as A909, containing a truncated and nonfunctional RgfA, deletion of rgfC caused differential gene expression of more than 200 genes, including capsule genes and virulence factors. From this, it was postulated that in strains lacking RgfA, RgfC uses nonspecific kinase activity to promote virulence (65). In addition, the A909 ΔrgfC strain caused increased proinflammatory signaling, increased systemic infection in a mouse model of bacterial meningitis, and increased virulence in a neonatal rat sepsis model (65).
Interestingly, when comparing differential gene expression in ΔrgfC mutants, there is high variability between strains. For instance, while RgfC appears to regulate expression of fbsA in CC17 (64), this was not observed in CJB111, which belongs to CC1 (22) or A909 (65). Further characterization of this potential quorum sensing system and the genes it regulates in different strains of GBS is necessary to fully understand its roles in colonization and pathogenesis.
Regulation of host adherence: SaeRS.A TCS homologous to the virulence regulatory TCS SaeRS in S. aureus was recently identified in GBS. This system was initially identified based on its upregulation during murine vaginal colonization (66). Activation of GBS SaeRS-regulated genes occurred in murine vaginal lavage fluid, demonstrating the presence of putative in vivo signals activating the system. RNA-seq data from the A909 ΔsaeR mutant grown under different conditions showed that the genes regulated by SaeRS differed based on different environmental conditions, demonstrating the likelihood that multiple signals lead to changes in gene expression through SaeRS (66).
Among others, SaeRS upregulates the gene encoding the plasminogen-binding surface protein (pbsP) following signals sensed in murine vaginal lavage fluid (66). PbsP is a highly conserved plasminogen-binding adhesin promoting GBS invasion of the central nervous system, at least in a hypervirulent CC17 strain (67). Moreover, PbsP facilitates bacterial adhesion to the host tissue through the intermediate binding to ECM components such as vitronectin. Vitronectin is an important adhesion substrate (68) which interacts with GBS streptococcal surface repeat (SSURE) domains on PbsP (69). The interaction facilitates GBS binding with epithelial cells, thereby promoting in vivo colonization. A ΔpbsP deletion strain also showed decreased colonization levels and persistence time in a mouse model of vaginal colonization (66). Overall, evidence suggests that upregulation of pbsP and potentially other factors through the SaeRS TCS promotes host colonization by GBS.
Regulation of carbon metabolism and virulence: FspSR.The fructose-6-phosphate (Fru-6-P) sensor histidine kinase and response regulator (FspSR) was identified as a TCS involved in the regulation of cellular responses to nutrient availability and bacterial fitness in GBS strain CJB111 (22). In a CJB111 ΔfspR mutant, there was a strong growth defect when cells were grown on Fru-6-P as a carbon source. In the WT strain, induction of both fspSR and the neighboring PTS operon (Man/Fru/Sor family) was observed in the presence of Fru-6-P, suggesting that Fru-6-P is a signal sensed by FspSR to upregulate the PTS operon (22).
A murine model of vaginal colonization showed that an fspR-deficient strain was significantly less persistent than the WT CJB111 strain in the vaginal tract (22), indicating its role in mucosal colonization. In addition, transcription of fspSR was observed to be upregulated when cells were exposed to bacterial lysates. Among other possibilities, the authors hypothesized that Fru-6-P could be released from dying organisms in the vaginal microbiota or that FspSR senses additional signals present in bacterial lysates (22). The biological relevance of a link between use of Fru-6-P as an energy source, upregulation of a PTS, and vaginal colonization is unclear and warrants further investigation.
Cell wall and resistance to antimicrobial peptides.In the host, one of the first lines of defense against bacteria is the production of cationic antimicrobial peptides (CAMPs). CAMPs are immune factors whose role is to directly kill invading microbes, and CAMP expression increases during bacterial infections. CAMPs are produced by circulating leukocytes such as neutrophils and macrophages as well as epithelial cells. In order to cause meningitis, GBS must evade host defenses in the blood compartment, an important component of which is CAMPs. In humans, there are two types of CAMPs, defensins and cathelicidins, which are small cytotoxic pore-forming peptides possessing regions of high cationic charge. This characteristic forms the basis of their antimicrobial activity, as CAMPs are attracted to the negatively charged bacterial surface. Hydrophobic CAMPs permeabilize the membrane of the bacteria, causing cell death (70). Bacteria have adapted evasion systems that allow resistance to some CAMPs (71, 72), many of which are controlled by TCSs.
d-Alanyl-lipoteichoic acid regulation: DltRS.There are two types of anionic polymers found in the Gram-positive bacterial cell wall: teichoic acids (TA), covalently linked to peptidoglycan, and lipoteichoic acids (LTA), anchored in the cell membrane by a glycolipid moiety and containing d-Ala ester or a glycosyl residue substitution for (poly)phosphoglycerol (73). Integration of d-Ala in the cell wall of Gram-positive bacteria allows for the maintenance and protection of the cell from, for example, turgor pressure and external stress (such as antibiotics), which function to break down the cell wall (74).
In many Gram-positive organisms, d-alanylation of LTA is carried out by the genes of the dlt operon, dltABCD. The net charge of LTA is reduced upon the addition of an amino group by d-Ala esterification. Absence or mutation of genes within the dlt operon results in the absence of d-Ala in LTA. Cells without d-Ala are more electronegative and consequently bind to cationic molecules more efficiently (75). Deficiency in d-alanyl-LTA showed increased vulnerability to phagocytic cells and defensins in GBS strain NEM316 (76). Furthermore, d-alanylation of LTA in GBS strain NEM316 confers enhanced resistance to CAMPs such as colistin (76), polymyxin B, and LL-37 (77).
In GBS, the dlt operon includes an additional two genes, dltRS, which are located upstream of dltABCD and which encode a TCS. DltS is a membrane-associated sensory protein (78) that likely senses environmental changes to activate DltR. DltRS are not required for expression of the dlt operon, and dltR mutants have WT levels of d-Ala on their LTAs, but DltR is activated when LTA is d-Ala deficient. Data from GBS strain NEM316 support the hypothesis that DltRS help maintain appropriate levels of d-Ala on LTA based on environmental signals by modulating expression of the dlt operon (78).
Lipid II-interacting antibiotics: LiaSR.The LiaSR TCS is found in many Gram-positive organisms and is generally involved in sensing and organizing a response to agents that disrupt cell envelope functions such as active antimicrobials and antibiotics, particularly, lipid II cycle inhibitors such as bacitracin and nisin (79, 80). In Bacillus subtilis, the lia operon is one of the most highly induced transcriptional factors when cells are grown in the presence of cell wall-active antibiotics (81), demonstrating its importance in regulating a response to these compounds. In many organisms, including GBS, liaSR is cotranscribed on an operon with liaF. LiaF is a membrane protein that has been shown to interact with LiaS to negatively regulate LiaSR (79, 80).
Evidence from studies in Streptococcus mutans and GBS indicates LiaSR is involved in sensing disruptions in cell wall integrity, activating LiaR (82–84). In GBS serotype Ia strain A909, LiaR is predicted to regulate the response to these agents by positively regulating the expression of genes involved in cell wall synthesis (pbp2b and murN), cell membrane modifications (mprF/fmtC homologs), and pilus formation (pilus island 2b) (82). Increased sensitivity to CAMPs nisin, polymyxin B, and colistin was also reported in the A909 liaR deletion mutant. This study also investigated the role of LiaR in virulence by using two mouse models of GBS infection, an intraperitoneal model of sepsis and a pneumonia model. These experiments showed significantly less aggressive pneumonia and sepsis with the A909 liaR deletion mutant than with the WT (82).
Nisin and bacitracin resistance regulation: NsrRK/BceRS.Lantibiotics are small antimicrobial lanthipeptides that are both hydrophilic and lipophilic. They are produced primarily by Gram-positive bacteria and exhibit antimicrobial activity against other Gram-positive bacteria (85). One mode of antimicrobial activity exhibited by certain lantibiotics is the binding of lipid II molecules, specifically, the pyrophosphate moiety, to disrupt cell wall function. Nisin, a lantibiotic produced by Lactococcus lactis, binds lipid II, creating pores and aggregates in the bacterial wall leading to cell death (86).
Some GBS strains contain a nisin resistance protein, SaNSR, which recognizes nisin and binds to the C terminus on the lanthionine ring. This interaction orchestrates the assembly of the catalytic site of SaNSR and the nisin cleavage site, causing nisin cleavage and inactivation (87). SaNSR is encoded by an operon containing an ABC transporter, NsrFP, and a TCS, referred to as NsrRK (or BceRS) (88, 89). NsrRK is predicted to regulate expression of the nsr operon, upregulating SaNSR expression. By regulating genes involved in lantibiotic resistance, the NsrRK TCS theoretically enables GBS to better compete against lantibiotic-producing mucosal microbiota.
The NsrRK system, also known as BceRS in the literature, is a homolog of a well-described TCS associated with cell wall-targeting antimicrobial resistance. BceRS systems have been described to regulate vancomycin (90) and polymxyin B (91) resistance in S. aureus and bacitracin resistance in S. mutans (92) and B. subtilis (93, 94). In 2019, Yang et al. provided evidence that, in the absence of the response regulator nsrR (called bceR in this study), GBS strain STE283 (serotype III) was more sensitive to human cathelicidin, LL-37, and bacitracin (95). In comparison, sensitivity to beta-lactam and erythromycin antibiotics was unaltered. They demonstrated that a likely method for mediating resistance to these compounds was through NsrR-mediated upregulation of the ABC transporter, NsrFP (BceAB), which functions to pump AMPs from the bacterial cell (95). Further experiments revealed that a STE283 ΔnsrR mutant strain had reduced biofilm formation capability and was more susceptible to hydrogen peroxide oxidative stress (95). Moreover, in a systemic murine intraperitoneal infection model, a significant decrease in bacterial survival rate was observed in the ΔnsrR mutant. These studies indicate that the NsrRK/BceRS TCS plays an important role in GBS resistance to antimicrobial peptides and antibiotics as well as playing roles in biofilm formation and survival in the host (95).
Antimicrobial peptide resistance: CiaRH.Based on the hypothesis that GBS factors that would promote CAMP resistance would also play a role in meningitis pathogenesis, Quach et al. screened GBS strain COH1 mutants to identify factors involved in resistance to CAMPs (96). Mutants identified in the screen were further examined for their ability to adhere to, invade, and survive inside hBMECs and neutrophils, survive CAMP killing, and survive and propagate in a mouse model of meningitis (96).
CiaRH, a TCS present in many streptococcal species, was initially shown in this study to confer resistance to CAMPs (both human and murine produced). ciaR deletion mutants were more susceptible to killing by CAMPs, H2O2, hypochlorite, and lysozyme and had decreased survival in hBMECs. A murine in vivo model of bacterial meningitis further confirmed the roles CiaR plays in survival and virulence, showing that WT GBS strain COH1 was recovered at significantly higher levels in the bloodstream and brain than the ciaR deletion strain (96).
While the exact role of CiaRH in GBS pathogenesis is still unknown, Mu et al. demonstrated that, in the same GBS strain COH1, CiaR may prevent endocytic trafficking of GBS-containing vacuoles to the lysosome (97). In addition, the group identified two hypothetical genes regulated by CiaR that may play a role in this phenotype, SAN_2180 (potentially related to acid tolerance) and SAN_0039 (a putative metallopeptidase) (97). Characterization of these genes will more fully elucidate the role of CiaRH in GBS pathogenesis and host immune evasion.
CONCLUSION
Successful colonization, invasion, persistence, and translocation of GBS are dependent on its ability to adhere to, and persist on, the host mucosal surfaces. Many factors are involved in colonization, infection of mucosal surfaces, response to the immune system, and environmental stressors, and these are often regulated by TCSs (98) (Fig. 2). Colonizing and infecting bacteria must be able to sense changes to their environment and respond quickly. With the ever-changing dynamics and variability of the host environment, efficient signal transduction mechanisms such as the TCSs serve to increase survival and distribution of the bacteria.
Model of regulation by selected GBS two-component systems. Selected GBS TCSs are shown along with genes they have been demonstrated to regulate. TCSs are separated into three sections based on predicted primary cellular functions: colonization in blue, pathogenesis in purple, or cell wall modifications and resistance to cationic antimicrobial peptides in green.
This review highlights some of the TCSs that have been described to date. While at least 22 TCSs have been identified in GBS, most are not yet well characterized and much remains to be done, including identifying the signals and environmental cues that activate TCS sensors and defining the genetic and phenotypic outcomes of activation of the response regulators. A more complete understanding of these systems will prove beneficial in the development of managerial strategies for the prevention and control of GBS infection.
- Copyright © 2020 American Society for Microbiology.
REFERENCES
- 1.↵
- 2.↵
- 3.↵
- 4.↵
- 5.↵
- 6.↵
- 7.↵
- 8.↵
- 9.↵
- 10.↵
- 11.↵
- 12.↵
- 13.↵
- 14.↵
- 15.↵
- 16.↵
- 17.↵
- 18.↵
- 19.↵
- 20.↵
- 21.↵
- 22.↵
- 23.↵
- 24.↵
- 25.↵
- 26.↵
- 27.↵
- 28.↵
- 29.↵
- 30.↵
- 31.↵
- 32.↵
- 33.↵
- 34.↵
- 35.↵
- 36.↵
- 37.↵
- 38.↵
- 39.↵
- 40.↵
- 41.↵
- 42.↵
- 43.↵
- 44.↵
- 45.↵
- 46.↵
- 47.↵
- 48.↵
- 49.↵
- 50.↵
- 51.↵
- 52.↵
- 53.↵
- 54.↵
- 55.↵
- 56.↵
- 57.↵
- 58.↵
- 59.↵
- 60.↵
- 61.↵
- 62.↵
- 63.↵
- 64.↵
- 65.↵
- 66.↵
- 67.↵
- 68.↵
- 69.↵
- 70.↵
- 71.↵
- 72.↵
- 73.↵
- 74.↵
- 75.↵
- 76.↵
- 77.↵
- 78.↵
- 79.↵
- 80.↵
- 81.↵
- 82.↵
- 83.↵
- 84.↵
- 85.↵
- 86.↵
- 87.↵
- 88.↵
- 89.↵
- 90.↵
- 91.↵
- 92.↵
- 93.↵
- 94.↵
- 95.↵
- 96.↵
- 97.↵
- 98.↵
- 99.↵
- 100.↵
- 101.↵
- 102.↵
- 103.↵
- 104.↵
- 105.↵
- 106.↵
- 107.↵
- 108.↵
- 109.↵
- 110.↵
- 111.↵
- 112.↵
- 113.
Author Bios

Lamar Thomas earned her bachelor’s and master’s degrees in Biology and Biotechnology, respectively, at the University of the West Indies in Kingston, Jamaica. She worked in Dr. Paula Tennant’s laboratory for both degrees, where she studied plant diseases with focus on viruses and fungi. Lamar emigrated to the United States thereafter and is currently a Ph.D. candidate in Microbiology at Binghamton University. Lamar joined Dr. Cook’s lab in 2018, where she studies host-pathogen interactions. Specifically, Lamar’s research aims to characterize a novel protein identified in Streptococcus agalactiae, believed to be pivotal for group B Streptococcus adherence to human host vaginal mucosa.

Laura Cook grew up in Rochester, MN, and earned her B.A. degree in English Literature and B.S. degree in Microbiology from the University of Minnesota in 2005. As an undergraduate, she worked in the laboratory of Dr. Patrick Schlievert for 3 years, studying staphylococcal superantigen toxins. She stayed at the University of Minnesota to obtain her Ph.D. in Microbiology studying enterococcal conjugation and biofilm formation in the lab of Dr. Gary Dunny in 2006. Laura began her postdoctoral training at the University of Illinois—Chicago in the lab of Dr. Michael Federle in 2012. While there, she began her work with streptococci, examining cell-cell communication between streptococcal species and streptococcus-host interactions. In 2018, Dr. Cook moved to Binghamton University to begin a position as an Assistant Professor. Her lab focuses currently on two species of pathogenic streptococci, S. pyogenes and S. agalactiae, and their interactions with host mucosal surfaces.