Pneumococcal surface protein A is present in virtually all pneumococcal isolates, independently of the serotype. It attaches to phosphocholine residues on the cell wall through the C-terminus, while the N-terminal domain is exposed on the bacterial surface (Yother and Briles, 1992); this region presents a pattern of sequence variation that was used to classify PspA proteins into six clades, included in three families (Hollingshead et al., 2000). All families are involved with inhibition of complement activation (Tu et al., 1999; Ren et al., 2003, 2004b) in the early phases of infection (Ren et al., 2004b), although the molecular mechanism of PspA interaction with complement has not been completely elucidated. A recent study demonstrated that PspA competitively inhibits binding of CRP to the phosphocholine moieties on the bacterial wall, thus limiting complement deposition through the classical pathway (Mukerji et al., 2012).

The inhibition of C3 deposition by PspA was attested by the observation that pneumococci lacking PspA were less virulent and more susceptible to C3 deposition than the wild-type strain, while the virulence of the PspA^- strain was recovered in complement factor B-deficient knockout mice (Tu et al., 1999). This evidence suggests that the inhibitory role of PspA over complement is related to FB activation, indicating an important contribution for the alternative complement pathway. In contrast, an in vitro study from (Ren et al., 2004b) found that C3b deposition on mutant pneumococci lacking PspA was dependent on the classical pathway activation, since blockage of the classical pathway (with EGTA) – but not the alternative pathway (using sera from FB deficient mice) – completely abrogated C3 deposition in the absence of PspA. However, when both complement pathways were present, C3 deposition on the PspA^- strain was much greater than with either pathway alone (Ren et al., 2004b). The authors conclude that while the classical pathway is required for activation of complement on PspA^- pneumococci, the alternative pathway is responsible for the amplification of C3 deposition (Ren et al., 2004b).

A study investigating the effects of PspA on complement mediated defense against pneumococci in vivo demonstrated a role for complement receptors CR1/2, since bacteria lacking PspA become virulent in knockout mice for this receptor. Also, mice that did not express factor D were more susceptible to infection with the PspA^- strain, confirming the importance of the alternative pathway for PspA mediated complement inhibition (Ren et al., 2004a). Furthermore, analysis of the fragments derived from C3 degradation on the pneumococcal surface (C3b and iC3b) showed increased levels of iC3b on the mutant strain, suggesting that PspA prevents processing of C3b into iC3b (Ren et al., 2004a). Finally, it has been shown that PspA-induced inhibition on complement deposition reduces bacterial clearance by phagocytes in vitro (Ren et al., 2012).

An additional effect of PspA’s inhibition of complement deposition is the ability to reduce the immune adherence of the bacterium to erythrocytes (as described for the capsular polysaccharides), thus limiting pneumococcal clearance through phagocytosis (Li et al., 2007). This effect was further enhanced in mutant strains lacking PspA and PspC (another complement-interacting protein that will be discussed in the following section), suggesting that these proteins can act synergistically, promoting a strong complement inhibition (Li et al., 2007).

The anti-complement properties of PspA can be abrogated by host immune response against this protein (Ren et al., 2012). A strong correlation between increased survival and the levels of complement deposited onto pneumococci coated with anti-PspA antibodies, has been reported in mice immunized with PspA fragments, suggesting that the blockage of PspA contributes to pneumococcal clearance mediated by complement (Darrieux et al., 2007; Moreno et al., 2010; Goulart et al., 2011). A similar result was obtained using sera from humans immunized with PspA (Ochs et al., 2008). Furthermore, antibodies against PspA purified from mice which had been immunized with heat-killed pneumococci promoted an increase in complement deposition on the bacterial surface (Khan et al., 2015).

Once the bacteria are coated with anti-PspA antibodies and complement, they can be recognized more efficiently by phagocytes and eliminated. Goulart et al. (2011) demonstrated that antibodies induced by mouse immunization with the N-terminal region of PspA from different clades were able to increase phagocytosis of diverse pneumococcal strains by mouse peritoneal phagocytes in presence of a complement source. This effect was confirmed in later studies using different opsonophagocytosis assay (OPA) protocols (Ren et al., 2012; Genschmer et al., 2013) and reinforces the ability of anti-PspA antibodies to promote bacterial phagocytosis through complement’ activation.

Pneumococcal surface protein C is a multifunctional protein, able to interact with complement components C3 (Cheng et al., 2000), human factor H (FH) (Janulczyk et al., 2000; Dave et al., 2001) and the regulator C4BP (Dieudonne-Vatran et al., 2009), and also functions as an adhesion protein, recognizing secretory immunoglobulin A (SIgA) through the secretory component (SC) (Dave et al., 2004; Kerr et al., 2006). PspC also interacts with a laminin-integrin receptor, improving bacterial adherence and spread to other sites (Orihuela et al., 2009). Moreover, PspC displayed affinity to other components of the extracellular matrix, such as thrombospondin-1 and vitronectin, further enhancing bacterial dissemination (Voss et al., 2013; Binsker et al., 2015).

Pneumococcal surface protein C contains an N-terminal α-helical domain which is exposed at the surface of the bacterium, followed by a proline-rich region and a cell surface-anchoring region (Brooks-Walter et al., 1999). PspCs are classified into 11 groups according to its genetic variation (Iannelli et al., 2002). Different studies demonstrated that PspC from multiple clinical isolates of pneumococci exhibit strong binding to FH (Lu et al., 2006; Moreno et al., 2012).

Pneumococcal surface protein C is able to interact directly with C3 through non-covalent binding to the α and β chains of the C3 molecule, preferentially to the alpha fragment (Cheng et al., 2000). Moreover, it has been shown that PspC interacts with C3 produced by lung epithelial cells, and that this interaction potentiates bacterial adherence (Smith and Hostetter, 2000). The impact of this interaction in pneumococcal disease was confirmed by Kerr et al. (2006) working with infected C3^-/- mice; they found that pneumococci lacking PspC presented poor virulence in WT mice, while their potential to cause disease was fully restored in C3 deficient knockout mice.

Pneumococcal surface protein C bound to vitronectin inhibited MAC formation (Jarva et al., 2002; Zipfel et al., 2008; Voss et al., 2013; Kohler et al., 2015), thus acting as a complement regulator, while improving the uptake of pneumococci by host cells (Bergmann et al., 2009). Despite the reduced susceptibility of pneumococci to MAC-mediated lysis, the combined effects of PspC interaction with vitronectin over the complement system contribute to pneumococcal virulence.

Pneumococcal surface protein C is able to recruit FH to the pneumococcal surface with a consequent reduction in complement activation, as demonstrated by an in vitro opsonophagocytosis assay (Jarva et al., 2002). PspC-FH interaction has also been shown to occur in vivo (Quin et al., 2005). The interaction between PspC and FH inhibits C3bBb formation by competition with factor B (FB) (a Bb precursor) for binding to C3b; it also accelerates irreversible C3bBb decline by relocating Bb, and acts as a cofactor for factor I (FI)-catalyzed C3b cleavage, yielding iC3b that cannot bind FB (Herbert et al., 2015). More recently, the implications of FH-PspC binding were investigated in vivo, during experimental human colonization with S. pneumoniae (Glennie et al., 2016). The authors found that individuals with epithelial inflammation and increased FH levels (due to influenza virus infection) had increased bacterial loads after experimental colonization, when compared to healthy subjects. Interestingly, the presence of antibodies against PspC did not affect in vivo binding of PspC with factor H, possibly because interaction with factor H limits the recognition of PspC by specific antibodies (Glennie et al., 2016). Furthermore, FH is exploited by PspC as a bridge to interact with host cells. Simultaneous binding of FH to PspC on the bacterial surface and the CR3 integrin on polymorphonuclear leukocytes (PMN) and CR3 expressing epithelial cells promoted bacterial adhesion and uptake (Agarwal et al., 2010). While the interaction with PMN may favor bacterial clearance, the increased adherence promoted by PspC interaction with FH is a necessary step for successful colonization (Quin et al., 2007; Agarwal et al., 2010). In fact, the adherence mediated by FH was related to an increased virulence during lung infection in mice (Quin et al., 2007).

In the biofilm phenotype of pneumococci, the expression of PspC was increased as well as the recruitment of FH. Also, the alternative complement pathway activation was impaired, providing a less propitious condition for phagocytosis (Domenech et al., 2013).

The ability of PspC to avoid complement attack was confirmed by Herbert et al. (2015), who found that FH bound to PspC displays a conformational change which causes an enhanced exposure of C3 binding sites (Herbert et al., 2015). As a result, the interaction of PspC-bound FH with C3b was increased twofold, while the alternative pathway convertase disassociation was increased fivefold (Herbert et al., 2015). In summary, PspC’s ability to interact with the complement system is the most important contribution of this molecule to pneumococcal pathogenicity, and provides an interesting target for protein-based vaccines.

Pneumococcal histidine triad (Phts) is a family of proteins exposed on the surface of S. pneumoniae. This group is composed by four members, PhtA, PhtB, PhtD, and PhtE, which are characterized by the presence of histidine triads (HxxHxH) repeated 5 or 6 times in their amino acid sequences (Adamou et al., 2001). Due to the large number of histidine residues in the triads, it has been suggested that Phts may bind to nucleoside or metals. In fact, Phts demonstrate high affinity for divalent cations, with 5 to 6 zinc binding domains (Adamou et al., 2001; Rioux et al., 2011). Zinc is involved in the positive regulation of Phts. Furthermore, variations in zinc concentration on different host niches correlate with the level of Phts expression; Phts are upregulated in the nasopharynx and lungs – where zinc availability is higher – with a reduced expression in blood (Ogunniyi et al., 2009). Despite the variations in Pht levels in different host niches, immunization with Pht fragments is protective in models of bacterial colonization and sepsis (Godfroid et al., 2011).

It has been suggested that Phts interact with complement factor H. However, studies evaluating the binding of Phts to FH in vitro revealed conflicting results. Ogunniyi et al. (2009), using FH-enzyme-linked immunosorbent assay and western blot, showed that all Phts bind to factor H, although this interaction was much lower when compared to PspC (positive control). More recently, Melin et al. (2010) used FH-ELISA and demonstrated that Pht proteins did not bind to factor H in most of the serotypes tested; they concluded that the capsule type influences Phts’ ability to interact with FH. In the same work, Melin et al. (2010) used western blot analysis to show that FH was able to bind to cell lysates of Pht^- pneumococcal strains, an effect that was dependent on the presence of PspC. The contribution of PspC for FH binding was attested by the abolished binding of this complement regulator, to mutants lacking PspC; in another experiment using flow cytometry, they showed that among five Pht^- mutants of different serotypes, only the type 4 strain was able to interact with factor H. They concluded that the relative contribution of Pht proteins to complement inhibition is likely to be affected by the presence of other pneumococcal proteins and, most importantly, depends on the genetic background (Melin et al., 2010).

In summary, the putative role of Phts in the binding to FH remains unclear, although their influence in complement-mediated phagocytosis was in fact observed. Studies with mutant strains negative for all Phts show an increased susceptibility to bacterial opsonization by C3. It was also suggested that a PhtA fragment could cleave human C3 (Hostetter, 1999). Therefore, further studies are needed in order to elucidate the role of Phts in complement-mediated immunity, which should contribute to the evaluation of their potential as pneumococcal vaccines.

Murein hydrolases are enzymes that degrade and remodel the cell wall, and are classified based on the substrate they target. This family of proteins includes amidases, glucosaminidases, muramidases (also termed as lysozymes), endopeptidases and transglycosylases. S. pneumoniae exhibits three main murein hydrolases: LytA, LytB, and LytC.

LytA, B, and C are choline binding proteins attached to the pneumococcal surface, involved in pneumococcal cell division – particularly LytB (Garcia et al., 1999) and cell wall remodeling (Gosink et al., 2000; Bai et al., 2014). They have also been shown to improve the adhesion/colonization of human and rat nasopharyngeal cells (Gosink et al., 2000; Bai et al., 2014). The loss of LytB and LytC expression results in an increased susceptibility to phagocytosis by alveolar macrophages and neutrophils (Ramos-Sevillano et al., 2011).

LytA interacts with the complement system in many ways. It binds to C4BP and FH, preventing complement deposition by classical and alternative pathways (Ramos-Sevillano et al., 2015). It has also been demonstrated that LytA inhibits binding of C1q to CRP, which was related with an increased exposure of Phosphocholine molecules in the LytA-negative mutant strain (Ramos-Sevillano et al., 2015). LytA can cleave C3b and iC3b fragments deposited in the bacterial surface regardless of the presence of capsule (Ramos-Sevillano et al., 2015), reducing opsonic activity and subsequent phagocytosis. Indeed, phagocytosis by neutrophils and alveolar macrophage cells was described to be higher in the lytA mutant strain (Ramos-Sevillano et al., 2012). The mutant strain also displayed an increased complement deposition when compared with the parental strain expressing autolysin (Ramos-Sevillano et al., 2012).

LytC negative pneumococci show an increased C3b deposition on their surface, an effect that was not observed with the LytB mutant (Ramos-Sevillano et al., 2011). However, despite the absence of a direct effect of LytB alone on complement deposition, a double mutant lacking LytB and LytC revealed a higher sensitivity to C3b deposition when compared with the single LytC mutant (Ramos-Sevillano et al., 2011). This suggests a contribution of both proteins to control complement deposition on pneumococci, an effect that was related to increased virulence in models of colonization and invasive disease in presence of LytB and LytC (Ramos-Sevillano et al., 2011).

Despite the evidence of LytA, LytB and LytC involvement in complement resistance by S. pneumoniae, the exact mechanism by which they confer resistance to complement is not completely clear. It is possible that the chain-dispersing function of these hydrolases influences the amount of C3 deposition on the bacterial surface. Growing of S. pneumoniae in biofilms was shown to reduce complement deposition and confer resistance to its opsonic activity and subsequent phagocytosis, when compared to bacteria in diplococci morphology. Indeed, the absence of LytA and LytB was found to promote formation of long chains of pneumococci, while the WT strain yielded diplococci morphology (Domenech et al., 2013). In addition, treatment of long pneumococcal chains with LytB led to bacterial dispersion into diplococci or short chains (De Las Rivas et al., 2002; Dalia and Weiser, 2011; Domenech et al., 2013). This result suggests a link between colony morphology and complement deposition on pneumococci, while indicating a role for autolysins in this mechanism.

Pneumolysin (PLY) is a cytotoxic protein produced by all strains of S. pneumoniae, found in the cytosol and cell wall compartments (Price and Camilli, 2009), and also being released from the bacterium during cell growth (Balachandran et al., 2001). PLY displays many important functions during pneumococcal infections, including (i) hemolytic and genotoxic effects (Johnson et al., 1980; Rai et al., 2016); (ii) induction of apoptosis in different cells – including macrophages and brain cells (Braun et al., 2001; Srivastava et al., 2005); (iii) interaction with TLR-4 and production of inflamatory cytokynes (Malley et al., 2003); (iv) activation of NLRP3 inflamasome (McNeela et al., 2010); (v) induction of neutrophil extracelullar trap (NET) formation (Nel et al., 2016); (vi) inhibition of cilliary movements in epithelial respiratory cells (Rayner et al., 1995; Marriott et al., 2008) and (vii) complement activation (Paton et al., 1984; Mitchell et al., 1991; Benton et al., 1997; Alcantara et al., 2001; Jounblat et al., 2003; Yuste et al., 2005; Propst-Graham et al., 2007; Marriott et al., 2008; Ali et al., 2013).

Pneumolysin’s ability to activate complement – which could be interpreted as disadvantageous for pneumococci – has actually been reported as a protective mechanism for the pathogen; since PLY is only exposed to interact with complement when it is released from the bacterium (the cell wall-attached molecule is not accessible to recognition by antibodies or complement), it has been suggested that complement activation by PLY causes depletion of complement components in the serum (Paton et al., 1984; Mitchell et al., 1991), thus limiting complement binding to the pneumococcus. This effect was confirmed in vivo using mutant strains that produce truncated forms of the toxin (Rubins et al., 1995; Alcantara et al., 2001; Propst-Graham et al., 2007); in a lung infection model, a mutant strain expressing a PLY which lacks the complement activation activity displayed a marked reduction in bacterial loads in the mice lungs (Rubins et al., 1995; Propst-Graham et al., 2007). Similarly, complementation of a PLY negative mutant with a recombinant PLY presenting intact complement activation capacity but no hemolytic activity was able to restore the bacterial virulence, indicating an independent contribution of the complement activation motif of PLY for pneumococcal virulence in this infection model (Rubins et al., 1995). However, in models of pneumococcal bacteremia, a significant role for the complement activity of PLY was only observed in cirrhotic rats (which produce reduced amounts of complement), with no effect in animals exhibiting normal complement production (Alcantara et al., 1999). This result suggests that the contribution of the complement activation motif of PLY is important in host niches or systems where a limited amount of complement is available (Propst-Graham et al., 2007; Marriott et al., 2008).

Pneumolysin’s ability to activate complement occurs mainly through the classical pathway; in fact, a direct interaction of PLY with C1 has been described in vitro (Paton et al., 1984) and related to a specific region of PLY that shows homology to human CRP, located in the domain 4 (Mitchell et al., 1991). PLY also shows structural similarity with the Fc portion of immunoglobulin, being able to bind Ig (Mitchell et al., 1991). Interestingly, the region involved in interaction with Ig is the same responsible for PLY self-oligomerization, suggesting one mechanism by which PLY promotes the complement activation [reviwed in (Marriott et al., 2008)]. However, complement activation by PLY is not dependent on the presence of specific anti-PLY antibodies, since sera from unvaccinated humans and non-immune mice showed increased C3 deposition on pneumococci lacking PLY (Yuste et al., 2005). Additionally, the contribution of PLY to complement-mediated inflammatory response may be enhanced by the fragments generated from complement activation (C3a and C5a), which attract polymorphonuclear cells, as already seen in serum containing high amounts of PLY (Johnson et al., 1981).

The observation that human serum (but not mice) depleted from C1q still promotes PLY opsonization by C3 activation products suggests a role for the lectin pathway in PLY-induced complement activation (Ali et al., 2013). In fact, human L-ficolin has been shown to initiate MBL activation by PLY in human serum, contributing to the classical pathway activation (Ali et al., 2013).

In addition to its direct role in complement activation, PLY participates in the assembly of pneumococcal biofilms (Shak et al., 2013) – a phenotype that is important for successful colonization and during the early stages of invasive infections, as well as limiting the interaction of complement components with surface molecules of pneumococci (Hall-Stoodley et al., 2006; Domenech et al., 2013). Therefore, the involvement of PLY on biofilm formation indirectly contributes to evasion from complement.

In summary, the contribution of PLY to pneumococcal virulence – which is highly influenced by the ability to interact with the complement system – reinforces the importance of complement evasion for bacterial survival, as well as PLY’s potential as a vaccine candidate against this pathogen.

The pneumococcal exoglycosidases are a group of enzymes that cleave terminal sialic acid from glycoconjugates. Among five exoglycosidases expressed by pneumococci – NanA, NanB, NanC, BgaA, and StrH, three surface associated molecules have been shown to interact with complement: NanA, BgaA, and StrH.

Neuraminidase A (NanA) is able to remove the sialic acid (SA) from a range of host substrates, thereby promoting the interaction between pneumococci and host cell receptors that contribute to pneumococcal colonization (Tong et al., 2000; Brittan et al., 2012). BgaA is the only pneumococcal β-galactosidase; it is secreted and acquired by the pneumococcal surface through a sortase. BgaA exhibits affinity to terminal galactose β (1–4) linked to GlcNAc, suggesting a role in deglycosylation of glycoconjugates of human cells (?) and was shown to be an adhesin, regardless of its galactosidase activity (Limoli et al., 2011). N-acetylglucosaminidase (StrH) is an exoglycosidase found in the bacterial surface and supernatant, able to cleave N-acetylglucosamine (GlcNAc) that is linked to mannose through β-1 (Zähner and Hakenbeck, 2000). All three proteins have been reported to work sequentially in the deglycosylation of SA, galactose, N-acetylglucosamine, decrypting mannose of host glycoproteins, which favors adherence/colonization (King et al., 2006).

These proteins have also been shown to deglycosylate human glycoconjugates as lactoferrin and IgA (King et al., 2006), demonstrating a direct interaction with the immune system. Indeed, SA participates in several immune processes such as leukocyte rolling, cell activation, modulation of immune cell functions and interactions with the complement system [reviewed in (Varki and Gagneux, 2012)].

NanA, BgaA, and StrH were shown to act in conjunction to limit complement deposition on pneumococci of different genetic backgrounds (Dalia et al., 2010). Blocking of the classical pathway did not affect complement levels on the bacterial surface, suggesting that NanA acts on the alternative pathway (Dalia et al., 2010). These proteins were also shown to inhibit bacterial killing by human phagocytes in vitro. This effect was attributed to the sequential removal of sugar moieties from host glycoconjugates by NanA, BgaA, and StrH: NanA removes sialic acid that is α2–3 or α2–6 linked to galactose, BgaA excises the galactose that is β1–4 linked to N-acetylglucosamine, and StrH removes N-acetylglucosamine that is β1 linked to mannose (King et al., 2006). Since complement components contain many glycosylation sites which are necessary for their functions [reviewed in (Ritchie et al., 2002)], the sequential activities of NanA, BgaA, and StrH possibly deglycosylate one or more glycoproteins that are important for complement deposition, thus leading to inhibition of complement mediated opsonophagocytosis (Dalia et al., 2010).

Previous studies found a role for SA in subversion of complement alternative pathway. SA is required for effective binding between FH and surface-attached C3b, as well as for dissociation of C3 convertase in sheep erythrocytes (Fearon, 1978). Pathogens like Neisseria meningitidis and Haemophilus influenzae exploit this complement-evasion strategy by producing SA that binds factor H (Ram et al., 1998). Interestingly, NanA from S. pneumoniae is able to remove sialic acid from the surface of these pathogens, thus limiting their ability to escape complement attack, providing a competitive advantage for pneumococci (Shakhnovich et al., 2002).

In summary, despite the evidence of exoglycosidases role in complement evasion, further investigations are required in order to elucidate the exact mechanism by which these proteins circumvent complement activation.

Streptococcus pneumoniae expresses a pilus-like structure, encoded by the rlrA pilus islet 1, which contributes to virulence in animal models of pneumococcal infection (Barocchi et al., 2006). This structure is composed of three proteins: RrgA, RrgB, and RrgC (Barocchi et al., 2006; Hilleringmann et al., 2009). RrgA is an adhesin that interacts with lung epithelial cells in vitro and confers virulence in mice (Barocchi et al., 2006; Hilleringmann et al., 2009). It is also able to increase bacterial phagocytosis by murine and human cells (Hilleringmann et al., 2009). Orrskog et al. (2012), using flow cytometry have shown that purified complement receptor 3 (CR3) binds to pneumococci expressing RrgA, and that purified RrgA is able to interact with the CD11b integrin domain of CR3. In addition, RrgA expression was linked to a faster dissemination of pneumococci from the upper respiratory tract and peritoneal cavity to the bloodstream of challenged mice, and to an increased bacterial survival inside macrophages (Orrskog et al., 2012). Altogether, the evidence indicate that RrgA interaction with CR3 on the surface of macrophages promotes bacterial internalization, while a prolonged survival of the internalized pneumococci contributes to their dissemination during systemic infection.

Moonlighting proteins are a group of molecules able to perform distinct biological activities through a single domain. Found in both prokaryotic and eukaryotic cells, these proteins display several activities, including roles in metabolic processes, host-pathogen signaling, transcription activation, adhesion and invasion of host tissues (Henderson and Martin, 2014). Based on their potential to interact with host cells, bacterial moonlighting proteins are important virulence factors in prokaryotic organisms (Jeffery, 1999; Henderson and Martin, 2011).

One important feature shared by many moonlighting proteins is their ability to interact with plasminogen (PLG) (Henderson and Martin, 2011), a plasmatic glycoprotein that generates plasmin (PL) upon activation. PL is a serine protease which displays many functions in tissue remodeling and coagulation. The conversion of PLG into PL is accomplished by activators like urokinase type A (uPA) and tissue type plasminogen (tPA) activators (human molecules) or by bacterial proteins [reviewed in (Lahteenmaki et al., 2001)].

Plasminogen and PL interfere with many molecules of the complement system (Barthel et al., 2012; Castiblanco-Valencia et al., 2016). PL cleaves C1, C2, C3, C3b, C4, and C5 (Pillemer et al., 1953; Amara et al., 2010; Barthel et al., 2012), thereby depleting and inhibiting complement activation through all three complement pathways. Even the enzymatically inactive PLG was able to improve the cofactor activity of FH in the inactivation of C3b by Factor I (Foley et al., 2015). It also cleaves iC3b, yielding C3dg-like, which acts as an opsonin and an activator of proinflammatory cytokines in macrophages (Foley et al., 2015). PLG has been found to interact with C3, C5, C3b, C3c, and C3d (Barthel et al., 2012).

Streptococcus pneumoniae expresses moonlighting proteins (Henderson and Martin, 2013), including PepO, enolase, GAPDH, PGK, and Tuf, which are able to acquire surrounding PLG molecules of the host. This interaction has been shown to contribute to invasion of host tissues and to avoid complement attack. The actions of pneumococcal moonlighting proteins are described in more detail below.