Host Pathways of Hemostasis that Regulate Group A Streptococcus pyogenes Pathogenicity (original) (raw)

. Author manuscript; available in PMC: 2020 Nov 17.

Abstract

A hallmark feature of severe Group A Streptococcus pyogenes (GAS) infection is dysregulated hemostasis. Hemostasis is the primary pathway for regulating blood flow through events that contribute towards clot formation and its dissolution. However, a number of studies have identified components of hemostasis in regulating survival and dissemination of GAS. Several proteins have been identified on the surface of GAS and they serve to either facilitate invasion to host distal sites or regulate inflammatory responses to the pathogen. GAS M-protein, a surface-exposed virulence factor, appears to be a major target for interactions with host hemostasis proteins. These interactions mediate biochemical events both on the surface of GAS and in the solution when M-protein is released into the surrounding environment through shedding or regulated proteolytic processes that dictate the fate of this pathogen. A thorough understanding of the mechanisms associated with these interactions could lead to novel approaches for altering the course of GAS pathogenicity.

Keywords: Group A Streptococcus, hemostasis, inflammation, pyogenes, pathogen, M-protein

1. INTRODUCTION

Group A Streptococcus pyogenes (GAS) causes >700 million cases of human pharyngeal and dermal infections worldwide/year, ranging from simple antibiotic-sensitive pharyngitis and impetigo/dermatitis to severe antibiotic-resistant forms of disseminated infectious diseases, e.g., necrotizing fasciitis and streptococcal toxic shock syndrome. In addition, post-infection non-pyogenic sequelae, e.g., rheumatic fever and glomerulonephritis, can occur. Due to the existence and continuous emergence of highly virulent strains of GAS, identification, characterization, and regulation of GAS virulence factors are topics of high interest. While there are >250 serotypes of GAS, among the most prevalent is a Pattern D skin trophic group, which conscripts components of the human innate immune system, including hemostasis and complement factors, in order to survive and disseminate in the host. Traditionally, the hemostasis system has been exclusively associated with mechanisms involved in regulating blood flow. Only within the past few decades, a role for components of this system has been identified as participants in other biological processes. Most of these advances have been made from the utilization of isolated components of this pathway in in vitro cell biology and biochemical studies as well as in vivo models utilizing gene-altered mice. Understanding the relationships between hemostasis and GAS in survival and dissemination of this bacteria can lead to novel therapeutic approaches in arresting the pathogenic processes of GAS infection.

2. GROUP A STREPTOCOCCUS PYOGENES

GAS is a common commensal bacterium of humans that typically colonizes the skin and mucosal surfaces, often leading to self-limiting skin and respiratory tract infections, e.g., impetigo and pharyngitis, that are sensitive to antibiotics. However, when treatment is delayed or absent, GAS infections can result in post-infection pyogenic or nonpyogenic sequelae, some of which evoke autoimmune responses that lead to glomerulonephritis and rheumatic heart disease [1]. In the most severe infections, when GAS breaches the epithelial barrier and invades blood and other deep tissue sites, life-threatening complications, e.g., necrotizing fasciitis and toxic septic shock, can occur [2]. The ~250 M-protein (M-Prt)-based serotypes of GAS that have been identified to date, possess highly evolved mechanisms to combat host innate immune defenses, and this human-selective microorganism continuously evolves to adjust to the more general defenses employed by the host that are genetically less focused on a battle with a single microorganism. GAS is genetically well constructed to combat host defenses. While this Gram+ microbe does not possess a protective outer membrane, it does contain a thick peptidoglycan (PGN)-rich semi-porous cell wall (CW) layer around its cytoplasmic membrane (CM). This CW is coated by a nonimmunogenic hyaluronic acid (HA) capsule. The (CW+HA) provides structural strength to GAS and presents a barrier to osmotic lysis. Importantly, the CW is also a matrix with functional groups to which proteins and carbohydrate can covalently tether.

The surface of GAS is studded by projections that extend beyond the HA capsule, which consists of four major types, viz.:(1) sortase A-dependent CW covalently-anchored LPXTG proteins, e.g., M-protein (M-Prt) [3], as well as glycopolymers, e.g., CW-bound teichoic acid, that serve as adhesins and also stabilize the Gram+ cell surface [4]; (2) non-covalently-bound glycolytic enzymes normally found in the cytosol, e.g., enolase (Eno) and GAPDH, provide ATP at the GAS surface, as well as serve as receptors for host proteins [5, 6]; (3) CM-bound moieties, e.g., leucine-rich repeat (LRR) lipoproteins (Lp) [7, 8], which are inserted in the CM through their N-terminal lipid components and bind GAS to host cell extracellular matrix (ECM) components [9]; (4) pathogen-associated molecular patterns (PAMP)s, e.g., CW PGN and lipotechoic acid (LTA), the latter also anchored in the CM by its N-terminal lipid moiety [10], that provides strength to the CW [11]. Some LTA is released from the CM and appears both in the cell media and on the cell surface where it is bound to positively charged proteins, e.g., M-Prt [12, 13]. This cell surface LTA provides hydrophobicity and functions in biofilm formation, as well as in adhesion [10, 14]. LTA can also engage CD14 and toll-like receptor (TLR)-2, thus upregulating inflammatory mediators [15, 16]. To colonize within various host niches, GAS employs ~30 one-component (OC) and ~13 two-component (TC) regulatory systems to sense, respond, and accommodate gene transcription to local environmental changes [17]. Control of human plasminogen/human plasmin (hPg/hPm) binding to GAS must be regulated since uncontrolled protease activity is harmful to the bacteria as they encounter disparate environments during different stages of infection. As a relevant example, M-Prt expression is positively controlled by the OC transcriptional regulator, Mga, which is most active during the bacterial growth phase. Intracellular Mga responds to the environment of GAS, likely by sensing metabolites in growing cells, e.g., glucose, and regulates ~10% of the GAS genome [18]. TC transcriptional regulatory systems normally consist of an extracellular environmental sensor (S) and intracellular responder (R) proteins. The most widely studied TC system is the control of virulence (Cov) RS system, which regulates ~18% of GAS transcription [19, 20]. An example of an important gene regulated by CovRS is the major secreted cysteine proteinase, streptococcal pyrogenic exotoxin B (SpeB). Overall, many GAS strains engage the hemostasis system to assist in its survival, and the human host also employs hemostasis as an innate immune defense.

3. EFFECTS OF COMPONENTS OF HEMOSTASIS ON GAS

Hemostasis, a component of human innate immunity, combines pathways of coagulation (intrinsic and extrinsic), anticoagulation, fibrinolysis, and platelet activation, to seize bleeding and maintain blood flow within the vessels. After infection, many components of hemostasis, in collaboration with inflammatory systems, can also be exploited by the pathogen for survival benefits of the microbe.

3.1. Coagulation and GAS Infection

The contact activation pathway (intrinsic coagulation pathway) consists of three serine proteases, Factor (F) XIIa, FXIa, and plasma kallikrein, as well as a non-enzymatic cofactor, high molecular weight kininogen (HMWK). Activation of this pathway is initiated by FXII binding to a negatively charged surface, which results in the autoactivation of FXII to FXIIa. FXIIa then activates prekallikrein and the clotting factor FXI, in combination with HMWK. Kallikrein cleaves HMWK which generates bradykinin (BK), a vasoactive proinflammatory peptide [21, 22]. Further proteolytic processing, i.e., neutrophil elastase, of HMWK generates antimicrobial peptides (AMPs) which assist in host defense responses against invading pathogens (Fig. 1).

Fig. (1).

Fig. (1).

The contact activation system (intrinsic coagulation pathway) consists of three serine proteases, FXII, FXI, and plasma kallirein (Ka), as well as a non-enzymatic cofactor, high molecular weight kininogen (HMWK). Activation of this pathway is initiated by FXII binding to a charged surface, which results in autoactivation of FXII. FXIIa then activates prekallikrein (PK) and FXI in complex with HMWK. Ka cleaves HMWK, generating bradykinin (BK), a vasoactive proinflammatory peptide, and kininogen can be further proteolytically degraded to generate anti-microbial peptides (AMPs).

GAS, specifically the cell wall PAMPs, via toll-like receptors (TLR)s, produce inflammatory mediators (IM) that originate from various host cells, e.g., monocytes [23–25]. This results in activation of the extrinsic pathway of hemostasis by the expression of Tissue Factor (TF) on monocytes and liberation of TF-containing microparticles [26], culminating in the upregulation of coagulation proteases, e.g., activated FXa and thrombin, which also enhance IM generation through binding to cellular protease-activated receptors (PARs) [27, 28]. Activation of endothelial cells (EC)s occurs during this process, which further upregulates TF [29] and plasminogen activated inhibitor (PAI)-1 [30], and downregulates anticoagulant activated Protein C (aPC) [31], which in combination enhances further FXa/thrombin expression. Stressed ECs, and other cell types, also release their intracellular contents (DAMPs), further stimulating inflammation [32]. This cycle of coagulation-inflammation is propagated until systemic inflammation, DIC, and eventually, lethal organ failure can occur.

Thrombin also originates from the intrinsic coagulation system, and is important in severe GAS infections [33–35]. It has been shown that coagulation FXII interacts with the GAS cell surface [36], where it is capable of a low degree of activation to FXIIa [37, 38]. Both FXI and PK (substrates of FXIIa) are bound to the GAS surface through HMWK, which in turn is bound to GAS by some M-Prts [39]. Ka [40], and/or a bacterial secreted protease, SpeB [41], converts HK to bradykinin (BK). The activation of Factor XII and Factor XI, critical components of the contact activation system, has also been shown to occur in the presence of GAS. Clinically, low levels of Factor XII and prolonged aPTTs are seen in patients with septic shock, an end-stage event of severe GAS infection [42].

3.2. Fibrinolysis and GAS Infection

Human fibrin (hFn), a consequence of thrombin/FXIIIa generation, serves other roles, e.g., hFn engulfs and confines the bacteria, thus hindering bacterial dissemination [43]. GAS responds by secreting a protein, streptokinase (SK), which is highly specific for activation of the human fibrinolytic system. This leads to the degradation of hFn, thus liberating the bacteria (Fig. 2). The human fibrinolytic system, consisting of hPg, hPm, fibrinogen (hFg), and activators and inhibitors of fibrinolysis, regulates hFn formation, which is important for a variety of normal processes [44–46], e.g., thrombotic stroke prevention, that require control of microthrombosis [47–49]. hPg is an 810-amino acid residue single-chain modular protein consisting of a 19-residue signal peptide, followed by five consecutive ~80 residue kringle (K) domains, which function to bind fibrinolytic effectors, e.g., Cl-, and lysine [50, 51], and a cryptic serine protease chain. hPg is activated by cleavage of the R580-V581 peptide bond (numbering from methionine 1 of the signal peptide) forming hPm, which contains two peptide chains, viz., the five-kringle nonprotease heavy chain (HC; E20-R580) double disulfide-linked to a trypsin-like serine protease light chain (LC; V581-N810) [52]. The essential features for GAS needed to enlist the host fibrinolytic system in its virulence are first to specifically and functionally interact with hPg and then to activate hPg on its surface to hPm. Many skin trophic strains of GAS can directly, and with very high affinity, bind hPg and hPm through the lysine binding site (LBS) of the K2 domain of hPg/hPm (K2hPg) and a small N-terminal a1a2 module of the M-Prt, plasminogen binding Group A streptococcal M-like protein (PAM) [53–56]. SK is arranged into two clusters, SK1 and SK2, while SK2 is subdivided into subtypes SK2a and SK2b. SK2b is selectively generated in skin-trophic Pattern D GAS strains that express hPg binding PAM. Studies have indicated that SK2b-mediated activation of hPg is enhanced by its interaction with the surface-expressed PAM and hPm that is generated remains associated with the cell surface which protects it from interacting with Pm inhibitors, i.e., α2-antiplasmin [57, 58] (Fig. 3A). Thus, a stable proteolytic surface is formed on GAS that provides the bacteria with a potent weapon against host innate immunity. Other strains of GAS with nasopharynx trophicity contain surface M-Prts, i.e., serotype M1, that first binds hFg through the D-domain of hFg and the B-domain of M-Prts [59]. hPg then binds to GAS via the bound hFg, through the E domain of hFg/hFn and the K1/K4/K5 domains of hPg/hPm (Fig. 3B). Other less virulent strains of GAS do not tightly bind to hPg or hFg. hPm bound to GAS functions in its defense against several host innate immune responses. For example, Pm allows GAS to penetrate the epithelial cell (EpC) and endothelial cell (EC) tight junctions (TJ), thus aiding the spread of the bacteria [60, 61]. hPm can also activate pro-metalloproteinases (pMMP) to active MMPs, which degrade the ECM, as well as EpC and EC TJs, thus allowing GAS dissemination [62]. In addition, hPm engages human cell integrin α5β1 and rearranges the cell cytoskeleton, promoting bacterial invasion of cells [63], thus allowing GAS to persist intracellularly. Lastly, mechanisms exist whereby hPm interferes with complement C3b deposition [64, 65], thereby attenuating opsonophagocytosis of GAS by phagocytic cells.

Fig. (2).

Fig. (2).

In an attempt to inhibit the dissemination of GAS in host tissue, activation of the host coagulation cascade occurs which results in engulfing GAS in a fibrin mesh (blue hash markings). GAS responds to this entrapment by binding host plasminogen (Pg) to M-Prts (orange triangles), directly or indirectly, exposed on the surface of GAS. GAS-derived SK is then released and surface-bound Pg is activated to plasmin (Pm) which facilitates in the degradation of fibrin and the release of GAS.

Fig. (3).

Fig. (3).

Plasminogen assembly on plasminogen binding Group A streptococcal M-like protein (PAM) and M proteins from GAS occurs either directly (A) through interaction with the a1a2 repeats of PAM through the lysine binding site of Kringle (K2) of human plasminogen (hPg) or indirectly (B) through the B domain of M-Prts interacting with the D domain of fibrinogen (Fg). Fg, through its E domain, then interacts with K1, 4, 5 of hPg. GAS expressing PAM is skin trophic while GAS expressing M-Prts, e.g., M1, that bind Fg are nasopharynx trophic. HVR = hypervariable region; LPXTG = Sortase A recognition site.

In an attempt to mimic the human condition during GAS infection in an animal model, a murine model was developed [66]. Since mouse Pg (mPg) binds weakly to PAM and is not activated by the GAS plasminogen activator, SK, a model of GAS infection was developed in which mice express hPg, through a hPg transgene. The transgene is directed for expression in the liver by the insertion of albumin gene regulatory sequences upstream of the hPg gene. When challenged with a GAS infection, these mice (expressing ~17% hPg) demonstrated a lower survival rate than WT mice (~80% mortality utilizing AP53 GAS). Additionally, genetically altered bacteria that do not express SK or PAM result in enhanced survival in these hPg expressing mice. However, depleting host fibrinogen Fg in these mice leads to enhanced mortality. The results from this study indicate that a major mechanism for bacterial invasion and dissemination is SK-mediated host Pg activation allowing for degradation of fibrin barriers.

3.3. Platelets and GAS Infection

Previous studies have demonstrated that platelets, major components of a thrombus, can interact with GAS leading to host tissue injury and inflammation [67–69]. Dysregulated hemostasis is correlated with severe GAS infections [2]. During infection, M1 protein is released from the bacteria, through both proteolytic processing and shedding, which binds plasma Fg [70]. The Fg receptors of platelets interact with these M1/Fg complexes on the platelet surface. Additionally, the presence of antibodies to M1 protein on these complexes allows for the activation of platelets through interaction with Fc receptors on platelets [69]. The ultimate result is the formation of platelet aggregates and microthrombi. Quantitative mass spectrometry from GAS infections identified M1 in plasma in association with Fg, IgG3, and complement component C1q [71]. Through flow cytometry, these components were also identified on the platelet surface with resultant complement activation. Monocyte-mediated M1-activated phagocytosis was also enhanced demonstrating a novel mechanism for platelet activation with resultant platelet consumption, a hallmark feature of sepsis. Clinically, platelet aggregates, IgG, and M1 have been identified in biopsy samples from patients with S. pyogenes toxic shock syndrome [69]. Additionally, tissue biopsies at the site of S. pyogenes infection show colocalization of bacteria and platelets in microthrombi, indicating that platelet/bacteria interaction and platelet activation may occur at the primary site of invasion during the early stages of infection [72]. Murine studies using a GAS model of sepsis (AP1 clinical isolate) demonstrated an increase in platelet-neutrophil complexes and plasma levels of CD62P (P-selectin, a marker of platelet activation) during bacterial dissemination in the development of sepsis which is decreased in the late stages of the disease. This correlated with organ damage and accumulation of platelets in the liver.

3.4. Regulating GAS Pathogenicity Targeting GAS Components and Host Hemostasis

It has been reported that natural GAS infections result in the development of protective antibodies. It is not surprising that children are more susceptible to the incidence of symptomatic infection which declines as a result of aging. These protective antibodies have been derived from exposure to sequences within the M-protein and other conserved antigens of GAS. Currently, there is no licensed vaccine for GAS which may be the result of these vaccines generating antibodies that induce an autoimmune effect towards host tissue, i.e., ARF. Additionally, the complexity of the disease in terms of the variety of emm types, host location of the infection, and prevalence of disease burden based on geographic location also contribute to the lack of development of an effective vaccine. Many approaches at developing a vaccine have targeted epitopes within the M-Prt that elicit a protective (opsonic) effect without human tissue cross- reactivity [73]. Some approaches have involved generating vaccines that contain a fusion protein with 4-30 N-terminal peptides of M-Prt from different serotypes in association with a carrier protein [74, 75].

In contrast to focusing on the N-terminal sequences of M-Prt, other approaches involve targeting conserved sequences within the M-Prt, i.e., the C-domain. These approaches include utilizing the entire C-domain [76] or 12 amino acid minimal B-cell or B-cell/T cell epitopes within this domain. In some cases, these peptides have been conjugated to the cholera toxin B subunit [76–78]. An epitope within the C-domain (P145) was found to be recognized by individuals in a highly endemic area and these antibodies have been shown to have opsonic activity [79–81]. Further characterization of this epitope determined that a 12 amino acid sequence within the P145 sequence (J8i) is a B cell epitope and does not stimulate T cell responses [82, 83]. However, due to structural restrains of this small peptide, it was found to be poorly immunogenic. This peptide was further modified (J8) and conjugated to diphtheria toxin to generate an immunogenic reagent (J8-DT) in mice [84]. This complex was able to induce B cell memory cells, protect mice from GAS challenge [85], and also prevent pyoderma during GAS skin infection [86]. The antibodies induced by J8-DT did not cross-react with human tissue [81]. Additionally, a related peptide, J14, was able to inhibit throat colonization after an intranasal challenge with GAS [87, 88].

Interestingly, J8-DT was shown to be ineffective against GAS containing mutations in CovR/S two-component regulatory system. Mutations in CovR/S lead to changes in the expression of virulence factors, some of which alter the immune response. These strains degrade CXC chemokines and therefore inhibit neutrophil chemotaxis. A vaccine was then developed that recognized two components of GAS, the J8-DT and an inactive streptococcal CXC protease, Streptococcus pyogenes cell envelope proteinase (SpyCEP). This vaccine protected against pyoderma and bacteremia with 100-1000-fold reduction in bacterial burden after challenge. [86]. However, given the amount of effort in understanding the biochemistry of critical surface virulence proteins of GAS, there are no licensed vaccines to combat GAS infection.

With regard to host hemostasis, it is known that a hallmark feature of severe GAS infection is dysregulated hemostasis and clearly demonstrates that hemostasis plays a major role in GAS infection and invasion in host. Mice with genetic alterations of components of hemostasis have largely assisted in determining key players during disease progression [66, 89]. One of these studies looked at the effects of both plasma and platelet Factor V, a major cofactor in prothrombin activation, on infection [89]. Relative to plasma FV, platelet FV has enhanced procoagulant properties and is more resistant to degradation by activated Protein C. The results from these studies demonstrated a unique role of platelet FV in regulating GAS infection and supports a critical role of thrombin generation in host defense. Additionally, while humans expressing a mutant FV Leiden (resistant to degradation by the anticoagulant protein, Protein C) were resistant to severe sepsis, mice carrying this mutation had no survival advantage when infected with GAS [89–91].

Fibrin has been shown to associate with GAS and inhibit its dissemination in the host. However, fibrin can also assist GAS by acting as a shield against host innate immune responses [92–97]. Although this has only been demonstrated in Staphylococcus aureus infection, which expresses a coagulase that forms a complex with host thrombin (staphylothrombin). Similar observations have not been observed in GAS infections. Other studies have shown that fibrinogen deficient (FG−/−) mice are more susceptible to GAS infection than wild type mice [89]. It has been shown that Fg can also function as a modulator of leukocyte function by interacting with the leukocyte integrin receptor, MAC-1 [98, 99]. In an infection model using mice expressing a Fg mutant (Fibgamma (390-396A)) that can no longer interact with MAC-1, but still has procoagulant activity, demonstrated a compromised host inflammatory response but a delayed and less severe mortality than observed in FG−/− mice [99]. This indicates that these two functions of Fg are important for regulating host responses to infection. It is known that components of the contact activation pathway can bind to GAS. Recent studies in mice have demonstrated that domain-5 of HMWK can bind to GAS, inhibit the contact activation pathway, and protect mice from developing lung pathology [100]. Therefore, blocking bacteria/host interactions, specifically those interactions involving bacteria and host hemostasis components, may be a reasonable approach towards developing new therapies for arresting GAS invasion.

CONCLUSION AND AUTHORS INSIGHT ON THE TOPIC

Hemostasis has traditionally been viewed simply as a method for regulating bleeding. However, through a number of in vitro studies using purified components of hemostasis and the development of mice deficient for or expressing mutant forms of these proteins, a role of hemostasis has been identified in a number of physiological and pathophysiological processes. Not surprisingly, hemostasis also plays a role in regulating GAS infection by utilizing components of this pathway to protect itself from host immune responses and to invade the surrounding barrier tissue. Recent studies have begun to unravel the mechanisms associated with these functions by identifying specific interactions between host and bacteria. Therefore, elucidation of these critical interactions may serve to identify new modalities of arresting bacterial infection and invasion.

ACKNOWLEDGEMENTS

FUNDING

This work was supported by a grant from the National Institute of Health (NHLBI) HL013423 to FJC and VAP.

Footnotes

CONSENT FOR PUBLICATION

Not applicable.

CONFLICT OF INTEREST

The authors declare no conflict of interest, financial or otherwise.

REFERENCES