Complement evasion by different mechanisms is important for microbial virulence and survival in the host. One strategy used by pathogenic bacteria is to bind the soluble complement inhibitor factor H (fH) to their surfaces. In group B streptococci and pneumococci, fH binding has been shown to be mediated by the surface proteins β and Hic, respectively. We showed previously that Hic binds to the middle region of fH and protects the pneumococcus from opsonophagocytosis. As the β protein and Hic are structurally closely related, we wanted to compare the fH binding characteristics of these two proteins. By using direct binding assays with radiolabeled proteins and surface plasmon resonance analysis we show that both β and Hic bind to the short consensus repeats 8–11 and 12–14 in the middle region of fH. Peptide mapping analysis suggested that the fH-binding sites on β and Hic were composed of discontinuous and partially homologous sequences. Thus, the bacterial virulence proteins use multiple binding sites on fH to secure high avidity. Also, the functionally active sites on fH are thereby left free to inhibit C3b deposition and opsonophagocytosis. These results reveal the evolutionary conservation of an analogous immune evasion strategy in different types of pathogenic streptococci. Importantly, the respective virulence factors could be exploited in the development of protein-based vaccines against these pathogens.

Several bacteria use C regulators to evade C attack and opsonophagocytosis (1, 2). Group B β-hemolytic streptococci (GBS;3Streptococcus agalactiae) cause life-threatening infections especially during the neonatal period (3, 4). GBS are part of the normal vaginal flora in ∼25–40% of healthy women. The severe neonatal infections are mostly acquired from the mother’s normal flora. In adults, GBS can cause cutaneous and invasive infections, such as septicemia and meningitis (5, 6). The molecular mechanisms by which the bacterium causes disease and evades the immune defense of the host are only partially known. The polysaccharide capsule of GBS is an important virulence factor that also protects the microbe against the host immune system, particularly against complement-mediated phagocytosis.

GBS express several surface proteins, some of which are involved in the immune evasion. One of these proteins is the β protein (also called Bac). The β protein binds to the Fc part of human IgA (7, 8, 9, 10). Recently it was found to bind C inhibitor factor H (fH) (11).

Binding of fH has been observed also by other types of streptococci, such as group A streptococci (GAS; Streptococcus pyogenes) and pneumococci (Streptococcus pneumoniae) (12, 13). Most GAS strains express one to three different types of fibrillar M family proteins that are antiphagocytic and important for the virulence of the bacteria. Some types of M proteins bind fH and/or the classical pathway complement regulator C4b-binding protein (12, 14, 15).

Serotype 3 pneumococci express the protein Hic (fH binding inhibitor of complement) that is encoded by the pspC locus (13). It is related to other members of the PspC protein family, but has some unique features and was therefore originally not recognized as a PspC protein. Recently, it was discovered that not only Hic on type 3 pneumococci, but also serotype 2 pneumococcal PspC protein bind fH (16, 17).

Factor H is a 150-kDa fluid phase regulator of the alternative pathway (AP) of complement. It is composed of 20 short consensus repeat (SCR) domains. Each SCR domain is held together by two disulfide bridges and has ∼60 aa. Factor H regulates the AP by inhibiting the binding of factor B to C3b, acting as a cofactor for factor I-mediated cleavage of C3b (cofactor activity) and accelerating the decay of the AP convertase C3bBb (decay-accelerating activity). All these steps are essential in keeping the AP amplification loop under control. Thereby, fH efficiently prevents C3 activation and consequent opsonophagocytosis, which are key mechanisms in innate immune defense against bacteria. Prevention of the formation of the membrane attack complex is thought to be less important because the access of membrane attack complex to the cell membranes of Gram-positive bacteria is restricted by the thick peptidoglycan layer.

As fH acts at a crucial step of C activation by regulating the AP amplification loop, it has also a major role in protecting self cell surfaces and tissues. Factor H binds to sulfated glycosaminoglycans and sialic acids. It has three recognized polyanion binding sites: at SCR7, around SCR13, and at SCR20 (18, 19, 20). Microbes do not naturally produce glycosaminoglycans, but they can have sialic acid moieties. For example, serotype III group B streptococci, group B meningogocci, and Escherichia coli K1 produce capsules that are composed of polysialic acid (21, 22, 23). It is, however, still uncertain to what extent fH can bind to polysialic acid (24). Its preference for binding is to terminal sialic acid moieties. Specific resistance to the AP is mediated by surface proteins that bind fH. Once fH is bound to the surface through these molecules, C activation is restricted.

The purpose of the present study was to take a comparative approach and analyze the fH binding properties of the GBS β protein and pneumococcal Hic by locating the binding site of β on fH and the fH binding sites on both β and Hic. We found that β and Hic, although expressed by different bacterial species, show remarkable homology and similarities in their fH binding properties. These data indicate for the first time that two different species of bacteria express homologous proteins that are used for immune evasion.

Human fH and factor I were purchased from Calbiochem-Novabiochem (La Jolla, CA). Recombinant fH constructs SCR1–7, -8–11, -8–20, and -15–20 were cloned and produced in the baculovirus expression system as described previously (25, 26, 27, 28). Human CRP and porcine heparin were obtained from Sigma-Aldrich (St. Louis, MO). Purified fH, SCR1–7, -8–20, and -15–20 and the β protein were radiolabeled with iodine (125I; NEN, Boston, MA) using the Iodogen method (29).

The recombinant SCR constructs SCR11–15 and SCR12–14 were amplified by PCR from a human liver cDNA library (Stratagene, La Jolla, CA) and cloned into the yeast expression vector pPICZαB (Invitrogen, Carlsbad, CA) using standard protocols. The primers used for PCR amplification were: SCR11For, 5′-CTGCAGGACAAGTACAATCATGTGGTCC-3′; SCR12For, 5′-CTGCAGGAGAGGAGAGTACCTGTGGAG-3′; SCR14Rev, 5′-TCTAGAGATGGAATTTTTTCAACACAGAGTGG-3′; SCR15Rev, 5′-TCTAGAGAAGGAAGGCCTTCACACTGAGG-3′. The primers incorporated restriction sites for PstI and XbaI (underlined). Ten micrograms of SacI-digested pPICZαB DNA containing the fH construct was electroporated into Pichia pastoris strain X33, and transformants were selected by zeocin. The expression of the proteins was induced for 3–4 days in the presence of methanol according to the manufacturer’s protocol (Invitrogen). The recombinant constructs were heavily glycosylated and therefore treated with endoglycosidase H (Roche, Mannheim, Germany). The proteins were purified by nickel-Sepharose affinity column (Probond Resin, Invitrogen) as described previously (25).

The wild-type GBS serotype Ia strain A909, the β-negative mutant, and the transcomplemented strain ΔbacpLZbac have been described previously (11). Strains Δ435–788, ΔXPZ, and Δ879–1064 have been described previously (30) (T. Areschoug and G. Lindahl, manuscript in preparation). The structures of the β protein and the mutants are shown schematically in Fig. 1. The β protein was purified as previously described (31, 32). The GBS strains and the β protein were gifts from Prof. G. Lindahl (University of Lund, Lund, Sweden).

FIGURE 1.

The schematic structure of β and the β mutants in GBS strains used in this study. S, signal sequence; XPZ, proline-rich repetitive region; M, C-terminal membrane-anchoring region. The IgA-binding region has been mapped to aa 153–225. The fH-binding site has been located to the C-terminal half of β, encompassing aa 441-1097. The strains used in this study include the wild-type A909 strain expressing β and mutant strains carrying β deletions spanning the fH-binding area of β.

FIGURE 1.

The schematic structure of β and the β mutants in GBS strains used in this study. S, signal sequence; XPZ, proline-rich repetitive region; M, C-terminal membrane-anchoring region. The IgA-binding region has been mapped to aa 153–225. The fH-binding site has been located to the C-terminal half of β, encompassing aa 441-1097. The strains used in this study include the wild-type A909 strain expressing β and mutant strains carrying β deletions spanning the fH-binding area of β.

Close modal

A fragment of pneumococcal Hic, covering aa 1–223 was expressed as a GST fusion protein as previously described (13). To obtain GST-free Hic (Hic1–223), the fusion protein was cleaved as previously described (33). In our previous study we used amino acid numbering where the signal peptide was included. In this study we have, in accordance with the β protein numbering, used amino acid numbering after omitting the signal peptide.

Streptococcal strains were grown until mid-log phase and were washed three times with 1/3 veronal-buffered saline with 0.1% gelatin (GVB). The bacteria (1 × 1010 cells/ml, 2 × 108 cells/reaction) were incubated with the radioactive proteins (fH, SCR1–7, -8–20, and -15–20; initial sp. act., 6.9 × 106, 1.3 × 107, 2.0 × 106, and 1.9 × 107 cpm/μg, respectively) at a final concentration of 4 nM in 1/3 GVB for 10 min at 37°C with gentle mixing. After incubation, the reaction mixture (40 μl) was centrifuged (10,000 × g, 3 min) through 20% sucrose (BDH Laboratory Supplies, Poole, U.K.) in 1/3 GVB. To separate the pellets, the sucrose-containing tubes were cut, and radioactivities in the pellets and the supernatants were measured in a gamma counter. The ratios of bound to total activity were calculated.

Surface plasmon resonance measurements were performed using the Biacore 2000 instrument and were analyzed with the BIAevaluation 3.0 software (Biacore, Uppsala, Sweden). Factor H, SCR1–7, SCR8–20, β, and Hic were immobilized on carboxymethylated dextran CM5 chips (Biacore) using the amine coupling procedure according to the protocol of the manufacturer. Binding analyses were performed using 1/3 or 1/2 veronal-buffered saline (50 or 75 mM NaCl, respectively), pH 7.4, at a flow rate of 5 μl/min. Before injecting into the Biacore flowcell, fH, recombinant constructs of fH, and β were dialyzed against the flow buffer. The protein concentrations of the reagents were measured using the bicinchoninic acid protein assay (Pierce, Rockford, IL). As controls, all binding tests were also performed using a blank flow cell, which was activated and deactivated without any coupled proteins. After each binding experiment the surface was regenerated by injecting 30 μl of 3 M NaCl in acetate buffer, pH 4.6 (regeneration buffer). All binding assays were performed in duplicate using two independently coupled Biacore chips.

For peptide scanning, we chose the regions of β and Hic known to be needed for fH binding. Amino acids 337-1067 of β and 1–223 of Hic were used in this analysis. Fifteen amino acid fragments with three amino acid transitions and 12 aa overlaps were used. The peptides were synthesized as spots onto polyethylene glycol-derivatized cellulose membranes (AIMS Scientific Products, Braunschweig, Germany) using the peptide-scanning instrument AutoSpot Robot ASP222 (Abimed Analysen-Technik, Langenfeld, Germany). The membrane was incubated with either radiolabeled fH (1 × 106 cpm) or SCR8–20 (1 × 106 cpm). After washing, binding was detected by exposure on a phosphorimager plate and Fujifilm BAS 2500 instrument (Fuji Photo Film, Tokyo, Japan).

In light of the fH binding ability of the pneumococcal Hic protein, we looked for homologous proteins in the databases. We used BLAST (34) to search the nonredundant protein database at the National Center for Biotechnology Information. As a result, we found that the GBS β protein and a large number of pneumococcal PspC proteins showed homology to Hic. The amino acid sequences of three PspC proteins (GenBank accession no. AAF73789.1, AAD31043.1, and AAF73802.1) were selected and aligned with Hic (AAG16729.1) and the β protein (P27951) using the ClustalW program (Fig. 2) (35). Group B streptococcal β is more closely related to pneumococcal PspC proteins than to, e.g., group A streptococcal M proteins (data not shown).

FIGURE 2.

Part of the alignment of the amino acid sequences of β, three PspC proteins, and Hic. The enlarged alignment represents the most homologous regions of the respective proteins (accession no. AAF73789.1, AAD31043.1, AAF73802.1, AAG16729.1, and P27951). The conserved amino acids have been shaded using amino acid similarity groups DN, EQ, ST, KR, FYW, and LIVM. White font with black background indicates 100% conservation, white font with gray background indicates 80–99% conservation, and black font with gray background indicates 60–79% conservation within each similarity group. The figure was created using GeneDoc sequence editor program (45 ).

FIGURE 2.

Part of the alignment of the amino acid sequences of β, three PspC proteins, and Hic. The enlarged alignment represents the most homologous regions of the respective proteins (accession no. AAF73789.1, AAD31043.1, AAF73802.1, AAG16729.1, and P27951). The conserved amino acids have been shaded using amino acid similarity groups DN, EQ, ST, KR, FYW, and LIVM. White font with black background indicates 100% conservation, white font with gray background indicates 80–99% conservation, and black font with gray background indicates 60–79% conservation within each similarity group. The figure was created using GeneDoc sequence editor program (45 ).

Close modal

Because of homology between β and Hic and the recently observed ability of β to bind fH, we proceeded to study how different recombinant constructs of fH (Fig. 3) bind to β in comparison with Hic. First we tested the direct binding of fH and recombinant constructs thereof to three different GBS strains: A909 expressing β, the deletion mutant of A909 not expressing β, and the transcomplemented strain with reinserted β. Factor H and its recombinant constructs SCR1–7, -8–20, and -15–20 were labeled with 125I. The radiolabeled proteins (final concentration, 4 nM) were incubated (30 min, 37°C) with GBS strains (2 × 108 bacteria in 1/3 GVB). Factor H bound to A909 and the transcomplemented strain that both express β, but not to the strain in which β had been deleted (data not shown). Also SCR8–20 bound to A909 and to the transcomplemented strain, binding percentages were 19.8 ± 0.8 and 19.3 ± 2.3% (Fig. 4). SCR1–7 or SCR15–20 did not bind to any of the GBS strains (binding <2.5%). As radiolabeling abolished the binding capacity of SCR8–11 and SCR11–15, these constructs could not be used in this assay.

FIGURE 3.

The recombinant fragments of fH used in this study. Factor H consists of 20 SCR domains. The cofactor (CA) and decay-accelerating (DA) activities have been located to SCR1–4. Three C3b (SCR1–4, SCR11–15, and SCR19–20) and three heparin (SCR7, around SCR13 and SCR20) binding sites have been identified on fH.

FIGURE 3.

The recombinant fragments of fH used in this study. Factor H consists of 20 SCR domains. The cofactor (CA) and decay-accelerating (DA) activities have been located to SCR1–4. Three C3b (SCR1–4, SCR11–15, and SCR19–20) and three heparin (SCR7, around SCR13 and SCR20) binding sites have been identified on fH.

Close modal
FIGURE 4.

Factor H binds to GBS in a β-dependent fashion with its C-terminal SCR8–20 part. GBS (2 × 108 bacteria/assay) in 1/3 GVB were incubated with radiolabeled SCR1–7 and SCR8–20 for 30 min at 37°C and centrifuged through 20% sucrose. Binding was calculated as the percentage of bound protein of the whole radioactive protein input. SCR8–20 binds to A909 and the transcomplemented strain (Δbac/pLZbac), both expressing whole β, but not to the mutant strain (Δbac) where β has been deleted. No significant binding of SCR1–7 to any of the three strains was observed. The results are shown as the mean ± SD of three assays performed in duplicate.

FIGURE 4.

Factor H binds to GBS in a β-dependent fashion with its C-terminal SCR8–20 part. GBS (2 × 108 bacteria/assay) in 1/3 GVB were incubated with radiolabeled SCR1–7 and SCR8–20 for 30 min at 37°C and centrifuged through 20% sucrose. Binding was calculated as the percentage of bound protein of the whole radioactive protein input. SCR8–20 binds to A909 and the transcomplemented strain (Δbac/pLZbac), both expressing whole β, but not to the mutant strain (Δbac) where β has been deleted. No significant binding of SCR1–7 to any of the three strains was observed. The results are shown as the mean ± SD of three assays performed in duplicate.

Close modal

Binding of the β protein to complement fH was analyzed by surface plasmon resonance using the Biacore2000 equipment. The β protein was coupled to the chip, and fH (0.3 μM) was injected into the flowcell. As fH was found to bind to β (data not shown), we proceeded by injecting SCR1–7 and SCR8–20 (1–2 and 1 μM, respectively). The injection of SCR8–20 to the flowcell containing β resulted in strong binding with slow dissociation (Fig. 5,A). There was no binding of SCR1–7 to β, even at 2 μM (Fig. 5,A). In a reverse setting, we coupled SCR1–7 and SCR8–20 to the Biacore chips and injected the β protein (0.3 μM) onto the flowcells. β bound to SCR8–20, but not to SCR1–7 or to the control flowcell (Fig. 5 B).

FIGURE 5.

Surface plasmon resonance analysis of β binding to fH fragments SCR1–7 and SCR8–20. A, β was coupled to the flowcell, and SCR1–7 and SCR8–20 were injected. Injection of SCR8–20 resulted in a positive binding curve, but with SCR1–7 no binding was seen. Controls received injection of SCR1–7 and SCR8–20 into a flowcell with no coupled protein. In a reverse setting (B), SCR1–7 and SCR8–20 were coupled to the flowcell, and β was injected. β bound to SCR8–20, but not to SCR1–7.

FIGURE 5.

Surface plasmon resonance analysis of β binding to fH fragments SCR1–7 and SCR8–20. A, β was coupled to the flowcell, and SCR1–7 and SCR8–20 were injected. Injection of SCR8–20 resulted in a positive binding curve, but with SCR1–7 no binding was seen. Controls received injection of SCR1–7 and SCR8–20 into a flowcell with no coupled protein. In a reverse setting (B), SCR1–7 and SCR8–20 were coupled to the flowcell, and β was injected. β bound to SCR8–20, but not to SCR1–7.

Close modal

To analyze the β binding region in fH in more detail, we injected the recombinant fragments of fH and SCR8–11, -11–15, -12–14, and -15–20 (2 μM of each) into the flowcells containing β. As shown in Fig. 6,A, SCR8–11 bound to β, but SCR15–20 did not. In addition, SCR11–15 and SCR12–14 bound to β. However, this binding reaction was weaker than the binding of SCR8–11 and was best observed in 1/3 veronal-buffered saline (Fig. 6, B and C). None of the injected proteins bound to the control flowcell with no protein coupled. The binding of SCR8–20 to β shows both a slower association and a slower dissociation than the binding of SCR8–11 (Fig. 6 A). This would be in accordance with two binding sites in the SCR8–20 fragment resulting in an overall stronger binding reaction. The fragment SCR8–11 appears to bind to β more strongly than fragment SCR11–15 and SCR12–14. Thus, the GBS β protein has two binding sites on fH, and the one in SCR8–11 appears to have a higher affinity than that in SCR12–14.

FIGURE 6.

Localization of the β-binding sites to the middle region in fH by surface plasmon resonance analysis. A, The β protein was coupled to the solid phase, and SCR8–11, SCR8–20, and SCR15–20 were injected. Both SCR8–11 and SCR8–20 bound to β, but the binding kinetics differed considerably. The kinetics (particularly the slow association and dissociation) of SCR8–20 suggest the presence of multiple β binding sites on this fragment. In contrast, SCR15–20 did not bind to β. Controls received injections of SCR8–20, SCR8–11, and SCR15–20 into a flowcell with no coupled protein. B, SCR11–15 was injected into the flowcells coupled with β or Hic. SCR11–15 bound to both proteins, but the binding to Hic was stronger. The dashed line indicates injection into the control flowcell. C, SCR12–14 was injected into the flowcells containing β or Hic. Again SCR12–14 bound to both microbial proteins. The dashed line shows injection into the control flowcell. Note the different y-axis scales in A–C.

FIGURE 6.

Localization of the β-binding sites to the middle region in fH by surface plasmon resonance analysis. A, The β protein was coupled to the solid phase, and SCR8–11, SCR8–20, and SCR15–20 were injected. Both SCR8–11 and SCR8–20 bound to β, but the binding kinetics differed considerably. The kinetics (particularly the slow association and dissociation) of SCR8–20 suggest the presence of multiple β binding sites on this fragment. In contrast, SCR15–20 did not bind to β. Controls received injections of SCR8–20, SCR8–11, and SCR15–20 into a flowcell with no coupled protein. B, SCR11–15 was injected into the flowcells coupled with β or Hic. SCR11–15 bound to both proteins, but the binding to Hic was stronger. The dashed line indicates injection into the control flowcell. C, SCR12–14 was injected into the flowcells containing β or Hic. Again SCR12–14 bound to both microbial proteins. The dashed line shows injection into the control flowcell. Note the different y-axis scales in A–C.

Close modal

We have observed that SCR8–11 and SCR8–20 fragments of fH bind to Hic (33). The binding kinetics of the two fragments differed, suggesting the possibility of another binding site in the SCR12–20 region. With the new fragments SCR11–15 and SCR12–14 we could now also test their binding to Hic. When Hic was coupled to a flowcell, injections of SCR11–15 (2 μM) and SCR12–14 (2 μM) resulted in the binding of both these recombinant fragments to Hic (Fig. 6, B and C). This showed that Hic has a second binding site on SCR12–14 in addition to the SCR8–11 region. However, as the site in SCR12–14 seemed to bind Hic with a lower affinity than that in SCR8–11, we propose that the SCR12–14 region is a secondary binding site for Hic as well as for β.

From these results we conclude that there are two binding regions for both β and Hic on fH, one located at SCR8–11 and a second at SCR12–14. Interestingly, the two surface proteins, β and Hic, from different bacteria not only share homology (Fig. 2), but appear to have two analogous binding sites on fH.

The binding of fH on β has been located to aa 435-1064 (11). To further map the binding site of fH on β, we used three mutant strains of GBS where deletions in β spanned this C-terminal region. We tested the binding of fH SCR8–20 to these mutant strains. The deletion of aa 435–788 from β practically abolished the binding of SCR8–20 to GBS (Fig. 7). The binding was reduced by 88% compared with that of the wild-type strain. Also the deletion of the XPZ region and the aa 879-1064 reduced the binding of SCR8–20 to the mutant GBS strain, but to a lesser degree (42 and 49%, respectively). This decrease in binding can be due to the close distance of the deletion to the capsule, which can affect the accessibility of the protein. SCR1–7 binding was tested as a control. No significant binding of SCR1–7 to any of the strains was observed (Fig. 7).

FIGURE 7.

Analysis of fH SCR1–7 and SCR8–20 binding to GBS strains expressing deletion mutants of β. Each of the GBS mutant strains (1 × 108 bacteria/assay) was incubated with radiolabeled SCR1–7 or SCR8–20 for 30 min at 37°C in 1/3 GVB and centrifuged through 20% sucrose. Binding was calculated as the percentage of bound protein (radioactivity in pellet) of whole radioactivity. SCR8–20 bound strongly to the wild-type A909 strain. There is a clear (88%) reduction in binding to the mutant strains with β aa 435–789 deleted. Binding was reduced to a lesser extent to the mutant strains where the XPZ (42%) and the C-terminal (49%) regions of β had been deleted. No significant binding of SCR1–7 was observed. Results are shown as the mean ± SD.

FIGURE 7.

Analysis of fH SCR1–7 and SCR8–20 binding to GBS strains expressing deletion mutants of β. Each of the GBS mutant strains (1 × 108 bacteria/assay) was incubated with radiolabeled SCR1–7 or SCR8–20 for 30 min at 37°C in 1/3 GVB and centrifuged through 20% sucrose. Binding was calculated as the percentage of bound protein (radioactivity in pellet) of whole radioactivity. SCR8–20 bound strongly to the wild-type A909 strain. There is a clear (88%) reduction in binding to the mutant strains with β aa 435–789 deleted. Binding was reduced to a lesser extent to the mutant strains where the XPZ (42%) and the C-terminal (49%) regions of β had been deleted. No significant binding of SCR1–7 was observed. Results are shown as the mean ± SD.

Close modal

Three heparin binding sites on fH, at SCR7, around SCR13, and at SCR20, have been described. Heparin has previously been shown to inhibit the binding of fH to β (11). This correlates with the observed binding site on SCR12–14 of fH. To further characterize the heparin inhibition of the fH-β interaction, six GBS strains (1 × 108 bacteria/assay in 1/3 GVB) were incubated (30 min, 37°C) with radiolabeled SCR8–20 (4 nM) and varying amounts of heparin (0–300 μg/ml). As shown in Fig. 8 A, heparin dose-dependently inhibited the binding of SCR8–20 to GBS strains expressing β. At a dose of 300 μg/ml, ∼50% of the binding was inhibited. Also, the binding of SCR8–20 to strains expressing mutants β where either the XPZ region or aa 879-1064 had been deleted was inhibited by heparin in a dose-dependent manner. Heparin had little effect on the binding of fH fragments to mutant strains where either whole β or aa 435–788 of β had been deleted.

FIGURE 8.

The effect of heparin on the binding of fH fragments to GBS, β protein, and Hic. A, Radiolabeled SCR8–20 was incubated with the six indicated GBS strains and in the presence of heparin (0–300 μg/ml). After 30 min, the samples were centrifuged through 20% sucrose, and binding was determined as the percentage of bound SCR8–20 vs total radioactivity. The binding of SCR8–20 to GBS strains expressing whole β was dose-dependently decreased, but not totally inhibited when heparin was added. The binding to the strains in which the XPZ or the C-terminal part of β had been deleted was also inhibited by heparin. B–D, Binding of fH fragments was analyzed by Biacore. B, β was immobilized on the flowcell, and SCR8–11 (1 μM) was injected with heparin (0–100 μg/ml). Heparin had a weak effect on the binding, as indicated by the decreased association and increased dissociation. C, The effect of heparin on binding of SCR11–15 to β was studied. SCR11–15 was injected into the flowcell together with 0–100 μg/ml heparin. A concentration of 10 μg/ml heparin increased the dissociation, and 100 μg/ml heparin inhibited the binding. D, The effect of heparin on binding of SCR11–15 to pneumococcal Hic. Hic was coupled to the flowcell, and SCR11–15 with 0–100 μg/ml heparin was injected. Addition of heparin had little effect on the binding, suggesting a slight difference in the binding sites for β and Hic on SCR11–15. Note the different y-axis scales in B–D.

FIGURE 8.

The effect of heparin on the binding of fH fragments to GBS, β protein, and Hic. A, Radiolabeled SCR8–20 was incubated with the six indicated GBS strains and in the presence of heparin (0–300 μg/ml). After 30 min, the samples were centrifuged through 20% sucrose, and binding was determined as the percentage of bound SCR8–20 vs total radioactivity. The binding of SCR8–20 to GBS strains expressing whole β was dose-dependently decreased, but not totally inhibited when heparin was added. The binding to the strains in which the XPZ or the C-terminal part of β had been deleted was also inhibited by heparin. B–D, Binding of fH fragments was analyzed by Biacore. B, β was immobilized on the flowcell, and SCR8–11 (1 μM) was injected with heparin (0–100 μg/ml). Heparin had a weak effect on the binding, as indicated by the decreased association and increased dissociation. C, The effect of heparin on binding of SCR11–15 to β was studied. SCR11–15 was injected into the flowcell together with 0–100 μg/ml heparin. A concentration of 10 μg/ml heparin increased the dissociation, and 100 μg/ml heparin inhibited the binding. D, The effect of heparin on binding of SCR11–15 to pneumococcal Hic. Hic was coupled to the flowcell, and SCR11–15 with 0–100 μg/ml heparin was injected. Addition of heparin had little effect on the binding, suggesting a slight difference in the binding sites for β and Hic on SCR11–15. Note the different y-axis scales in B–D.

Close modal

The effect of heparin was also tested using the Biacore equipment. The β protein was coupled to the flowcell. Recombinant fH fragments containing SCR8–11 or SCR11–15 (1 μM) were injected onto the flowcell in the presence of varying amounts of heparin (0–100 μg/ml). As shown in Fig. 8,B, heparin slightly inhibited the binding of SCR8–11 to β, as can be seen by the decreased association and increased dissociation. In Fig. 8,C, SCR11–15 (1 μM) was injected onto β on the flowcell. A dose of 10 μg/ml heparin increased the dissociation, 30 μg/ml had no effect, but 100 μg/ml inhibited the binding of SCR8–11 to β. We have previously found that heparin has only a minor effect on the binding of SCR8–11 or SCR8–20 to the pneumococcal Hic (33). In this study we wanted to study the effect of heparin on the binding of the SCR11–15 construct to Hic. In Fig. 8 D it can be seen that heparin, at a concentration up to 100 μg/ml, does not decrease the binding of SCR11–15 to Hic, but, rather, may increase it. These data suggest that the binding sites for heparin and β around SCR13 of fH overlap and also that the binding sites for fH on β and Hic may differ somewhat in structure.

To obtain information about the putative binding sites on β and Hic at the primary structure level, we performed peptide mapping studies. Fifteen-amino acid-long peptides of β and Hic with three amino acid shifts were synthesized on a cellulose membrane. These peptides spanned the regions that in earlier experiments were found to contain the fH binding sites, i.e., aa 351-1064 of β and 1–223 of Hic. The membranes were incubated with radiolabeled fH or with the SCR8–20 construct. Binding was visualized using a phosphorimaging system. The experiment revealed five putative binding sites on β and three sites on Hic (Fig. 9). Interestingly, although it is unlikely that all the indicated peptides are involved in fH binding, some of these sites share similarities between β and Hic. Notably, the regions 693–717 in β and 137–163 in Hic that bound both fH and the SCR8–20 fragment showed clear-cut homology. As all the putative binding regions are rich in charged amino acids, most of the binding interactions with fH seem to be of ionic nature.

FIGURE 9.

Peptide spot analysis of fH binding to Hic and β. A, Peptides spanning aa 337-1067 of β and 1–223 of Hic were spotted on a cellulose membrane. Each spot contained a 15-aa peptide, and there was a 3-aa shift between spots. The membrane was incubated with 125I-labeled fH. A similar result was obtained with [125I]SCR8–20. Binding was visualized on a phosphorimager plate. B, The β protein (aa 337–756) and Hic (aa 1–202) sequences are shown as a ClustalW alignment. The boxed areas represent the putative fH binding sites. Numbers indicate the first amino acids in the binding regions.

FIGURE 9.

Peptide spot analysis of fH binding to Hic and β. A, Peptides spanning aa 337-1067 of β and 1–223 of Hic were spotted on a cellulose membrane. Each spot contained a 15-aa peptide, and there was a 3-aa shift between spots. The membrane was incubated with 125I-labeled fH. A similar result was obtained with [125I]SCR8–20. Binding was visualized on a phosphorimager plate. B, The β protein (aa 337–756) and Hic (aa 1–202) sequences are shown as a ClustalW alignment. The boxed areas represent the putative fH binding sites. Numbers indicate the first amino acids in the binding regions.

Close modal

In this study we show for the first time that two distantly related, but partially homologous, proteins from two different bacterial species bind to the same regions on complement fH. Furthermore, the microbial evasion proteins bind to multiple sites on fH making the association more stable. It seems that the respective proteins on GBS (β) and pneumococci (PspC family) have been evolutionarily conserved through species differentiation, suggesting that the binding of fH is an important and early property for the survival of pathogenic bacteria in the host.

The β protein is an ∼125-kDa surface protein expressed by many strains of serotypes Ia, Ib, II, and V GBS (7, 8, 36). β binds to serum IgA and fH (7, 8, 9, 10, 11). We have previously shown that the pneumococcal Hic protein of the PspC family binds to the SCR8–11 of fH (33). In this study we show that the β protein also binds to SCR8–11 and that both β and Hic bind to another region at SCR12–14. The sharing of these binding regions on fH further emphasizes the structural relatedness of these proteins. Multiplicity of binding along the longitudinal fH molecule also suggests an elongated structure of the fH binding proteins and explains the relatively high affinity between fH and the microbial proteins.

The presence of another separate binding site on fH for Hic could be verified after we generated the SCR11–15 and SCR12–14 fragments. Thus, like β, Hic has a second binding site on SCR12–14 in addition to SCR8–11. Because of the stronger binding of β to the SCR8–11 region, it probably represents the primary fH binding site, and SCR12–14 is a secondary site. Nevertheless, the existence of two distinct interactions between fH and β as well as between fH and Hic ensures a high affinity between these molecules. The known ligands for the SCR8–11 region are CRP, β, and Hic (33, 37). CRP does not inhibit the binding of Hic or β to fH (data not shown), so the binding site for CRP does not, at least totally, overlap the region for these bacterial proteins.

SCR12–14 of fH bind to heparin (18) (H. Jarva, T. S. Jokiranta, and S. Meri, unpublished observations) as well as to β and Hic. In the case of β, heparin also inhibits the binding of fH to β. In the Biacore experiments, it was found that the binding of SCR11–15 (and SCR12–14) to β was inhibited by heparin (100 μg/ml), and the binding of SCR8–11 to β was slightly reduced in the presence of heparin. In general, heparin tended to have a two-phase effect on the fH fragment binding to β. At lower concentrations (<10 μg/ml; not shown) heparin often increased the binding, and at high concentrations it inhibited the binding. Probably the region for heparin binding on fH at least partially overlaps the binding site for β. In accordance, the direct binding of radiolabeled SCR8–20 to strains expressing whole β was partially inhibited by heparin. Modulation of fH binding to β by heparin further suggests that ionic interactions are involved in the binding.

As β binds to serum IgA and as some PspC proteins bind secretory IgA (38), we tested the binding of serum and secretory IgA to Hic, but no detectable binding was observed (data not shown). Thus, the IgA-binding feature is not a prerequisite for fH binding, and the IgA binding property is not shared by β and Hic. IgA binding on β has been mapped to the N-terminal part, and fH binding has been mapped to the C-terminal part of the protein (11, 39). The lack of IgA binding by Hic could be explained by the fact that it lacks the regions homologous to the N-terminal part of β (Fig. 9).

By binding to two sites in the middle part of fH, the microbes may ensure that fH has functional sites free after binding to the microbial surface. As both N and C termini have important functions (binding to C3b, cofactor, and decay-accelerating activity and binding to C3d/b and heparin, respectively), it is feasible that these regions are displayed (25, 28, 40, 41, 42). After binding of fH to β or Hic, it can still maintain strong binding to C3b.

The binding to SCR8–20 was significantly stronger than the binding of SCR8–11 or SCR11–15. The binding efficiencies of the fragments SCR8–11, -11–15, and -12–14 differ, as indicated by the different scales of the y-axes in Fig. 6. This suggests that the use of two binding sites has more than a simple additive effect. SCR8–11 and SCR12–14 could independently contribute to the binding of SCR8–20 to β, but, in addition, the physical proximity and linkage of the SCRs in the SCR8–15 region of fH could affect the conformations and/or the overall arrangement of the SCRs. These can have a synergistic effect on the binding and even generate novel binding sites for β.

By using GBS strains expressing deletion mutants of β and radiolabeled SCR8–20, the major fH-binding area on β was localized to aa 435–788. The binding of SCR8–20 was also somewhat diminished to the mutants where either the XPZ or the C-terminal region had been deleted compared with that to the wild-type strain. As β is anchored to the bacterial cell wall by the C-terminal end, we assume that the deletion of these regions may have an effect on the accessibility of fH to the major binding region.

By peptide mapping, we recognized putative binding sites on both β and Hic. By direct binding experiments, the binding on β could be located to aa 435–788. On Hic, the binding region has previously been located within the 217 N-terminal aa (13). We found discontinuous stretches of putative binding sites on β and Hic (Fig. 9). In the alignment of β and Hic sequences, some of these regions overlap and even show homology to each other (especially the aa 693–717 and 137–163 regions in β and Hic, respectively). Several positively charged residues were found on each of the putative sites. Although the three-dimensional conformations of these proteins are not known, we assume that they are fibrillary or filamentous in structure. This would be in accordance with an alignment of the elongated β and Hic proteins with the chain of SCRs in fH. Previously, the fH binding region on β has been located to aa 441-1097 (11). However, one of the putative binding sites was located 50–70 aa downstream of this area. This suggests that a single site is not enough for the binding of fH, and also that the loss of one site does not markedly affect the binding.

In conclusion, we have observed that the GBS and pneumococcal Hic proteins are structurally closely related and bind to SCR8–11 and SCR12–14 of fH. No other bacterial species has yet been shown to bind to these regions of fH. However, despite the relatedness of the proteins and the same binding region on fH, β and Hic do not share the exact binding site, which could be seen in the differences in the inhibition assays. For β and Hic, there are two binding sites on fH. This is also seen, e.g., with certain GAS M proteins, as indicated by the fact that several M proteins bind to SCR7 of fH, and at least M22 also binds to the SCR8–15 region (43, 44). Apparently, the bacteria take advantage of two binding sites, thereby ensuring a stronger binding affinity. β and Hic have been conserved through evolution, and the expression of fH-binding molecules appears to be a key feature in virulence and significant for bacterial survival by preventing complement attack and opsonophagocytosis.

We are grateful to Dr. Thomas Areschoug and Prof. Gunnar Lindahl (Lund University, Lund, Sweden) for providing the β protein and GBS strains used in this study. We thank Drs. Lars Björck and Robert Janulczyk (Lund University) for the Hic protein, and Jussi Hepojoki and Hilkka Lankinen (Haartman Institute, University of Helsinki, Helsinki, Finland) for preparation of the peptide spot membrane.

1

This work was supported by the Maud Kuistila Foundation, the Sigrid Jusélius Foundation, the Academy of Finland (the MicMan and Life2000 Programs), Helsinki University Central Hospital Funds, and the Deutscher Akademischer Austauschdienst.

3

Abbreviations used in this paper: GBS, group B hemolytic streptococci; AP, alternative pathway; CRP, C-reactive protein; fH, factor H; GAS, group A hemolytic streptococci; GVB, veronal-buffered saline plus 0.1% gelatin; Hic, fH-binding inhibitor of complement; PspC, pneumococcal surface protein C; SCR, short consensus repeat.

1
Rautemaa, R., S. Meri.
1999
. Complement-resistance mechanisms of bacteria.
Microbes Infect.
1
:
785
.
2
Jarva, H., T. S. Jokiranta, R. Wurzner, S. Meri.
2003
. Complement resistance mechanisms of streptococci.
Mol. Immunol.
40
:
95
.
3
Baker, C. J., M. S. Edwards.
1995
. Group B streptococcal infections. J. S. Remington, and J. O. Klein, eds.
Infectious Diseases in the Fetus and Newborn Infant
980
. W. B. Saunders, Philadelphia.
4
Schuchat, A..
1998
. Epidemiology of group B streptococcal disease in the United State: shifting paradigms.
Clin. Microbiol. Rev.
11
:
497
.
5
Farley, M. M., R. C. Harvey, T. Stull, J. D. Smith, A. Schuchat, J. D. Wenger, D. S. Stephens.
1993
. A population-based assessment of invasive disease due to group B ztreptococcus in nonpregnant adults.
N. Engl. J. Med.
328
:
1807
.
6
Domingo, P., N. Barquet, M. Alvarez, P. Coll, J. Nava, J. Garau.
1997
. Group B streptococcal meningitis in adults: report of twelve cases and review.
Clin. Infect. Dis.
25
:
1180
.
7
Bevanger, L..
1983
. Ibc proteins as serotype markers of group B streptococci.
Acta Pathol. Microbiol. Immunol. Scand. B.
91
:
231
.
8
Johnson, D. R., P. Ferrieri.
1984
. Group B streptococcal Ibc protein antigen: distribution of two determinants in wild-type strains of common serotypes.
J. Clin. Microbiol.
19
:
506
.
9
Russell-Jones, G. J., E. C. Gotschlich, M. S. Blake.
1984
. A surface receptor specific for human IgA on group B streptococci possessing the Ibc protein antigen.
J. Exp. Med.
160
:
1467
.
10
Lindahl, G., B. Akerstrom, J.-P. Vaerman, L. Stenberg.
1990
. Characterization of an IgA receptor from group B streptococci: specificity for serum IgA.
Eur. J. Immunol.
20
:
2241
.
11
Areschoug, T., M. Stalhammar-Carlemalm, I. Karlsson, G. Lindahl.
2002
. Streptococcal β protein has separate binding sites for human factor H and IgA-Fc.
J. Biol. Chem.
277
:
12642
.
12
Horstmann, R. D., H. J. Sievertsen, J. Knobloch, V. A. Fischetti.
1988
. Antiphagocytic activity of streptococcal M protein: selective binding of complement control protein factor H.
Proc. Natl. Acad. Sci. USA
85
:
1657
.
13
Janulczyk, R., F. Iannelli, A. G. Sjoholm, G. Pozzi, L. Bjorck.
2000
. Hic, a novel surface protein of Streptococcus pneumoniae that interferes with complement function.
J. Biol. Chem.
275
:
37257
.
14
Thern, A., L. Stenberg, B. Dahlback, G. Lindahl.
1995
. Ig-binding surface proteins of Streptococcus pyogenes also bind human C4b-binding protein (C4BP), a regulatory component of the complement system.
J. Immunol.
154
:
375
.
15
Perez-Caballero, D., S. Alberti, F. Vivanco, P. Sanchez-Corral, S. Rodriguez de Cordoba.
2000
. Assessment of the interaction of human complement regulatory proteins with group A streptococcus: identification of a high-affinity group A streptococcus binding site in FHL-1.
Eur. J. Immunol.
30
:
1243
.
16
Dave, S., A. Brooks-Walter, M. K. Pangburn, L. S. McDaniel.
2001
. PspC, a pneumococcal surface protein, binds human factor H.
Infect. Immun.
69
:
3435
.
17
Duthy, T. G., R. J. Ormsby, E. Giannakis, A. D. Ogunniyi, U. H. Stroeher, J. C. Paton, D. L. Gordon.
2002
. The human complement regulator factor H binds pneumococcal surface protein PspC via short consensus repeats 13 to 15.
Infect. Immun.
70
:
5604
.
18
Pangburn, M. K., M. A. Atkinson, S. Meri.
1991
. Localization of the heparin-binding site on complement factor H.
J. Biol. Chem.
266
:
16847
.
19
Blackmore, T. K., T. A. Sadlon, H. M. Ward, D. M. Lublin, D. L. Gordon.
1996
. Identification of a heparin binding domain in the seventh short consensus repeat of complement factor H.
J. Immunol.
157
:
5422
.
20
Blackmore, T. K., J. Hellwage, T. A. Sadlon, N. Higgs, P. F. Zipfel, H. M. Ward, D. L. Gordon.
1998
. Identification of the second heparin-binding domain in human complement factor H.
J. Immunol.
160
:
3342
.
21
Kasper, D. L., J. L. Winkelhake, W. D. Zollinger, B. L. Brandt, M. S. Artenstein.
1973
. Immunochemical similarity between polysaccharide antigens of Escherichia coli O7: K1 (L) and group B Neisseria meningitidis.
J. Immunol.
110
:
262
.
22
Baker, C. J., D. L. Kasper, C. E. Davis.
1976
. Immunochemical characterization of the “native” type III polysaccharide of group B streptococcus.
J. Exp. Med.
142
:
258
.
23
Jennings, H. J., A. K. Bhattacharjee, D. R. Bundle, C. P. Kenny, A. Martin, I. C. P. Smith.
1977
. Structure of the capsular polysaccharides of Neisseria meningitidis as determined by 13C-nuclear magnetic resonance spectroscopy.
J. Infect. Dis.
136
:(Suppl.):
78
.
24
Meri, S., M. K. Pangburn.
1994
. Regulation of alternative pathway complement activation by glycosaminoglycans: specificity of the polyanion binding site on factor H.
Biochem. Biophys. Res. Commun.
198
:
52
.
25
Kühn, S., C. Skerka, P. F. Zipfel.
1995
. Mapping of the complement regulatory domains in the human factor H-like protein 1 and in factor H.
J. Immunol.
155
:
5663
.
26
Kühn, S., P. F. Zipfel.
1995
. The baculovirus expression vector pBSV-8His directs secretion of histidine-tagged proteins.
Gene
162
:
225
.
27
Prodinger, W. M., J. Hellwage, M. Spruth, M. P. Dierich, P. F. Zipfel.
1998
. The C-terminus of factor H: monoclonal antibodies inhibit heparin binding and identify epitopes common to factor H and factor H-related proteins.
Biochem. J.
331
:
41
.
28
Jokiranta, T. S., J. Hellwage, V. Koistinen, P. F. Zipfel, S. Meri.
2000
. Each of the three binding sites on complement factor H interacts with a distinct site on C3b.
J. Biol. Chem.
275
:
27657
.
29
Salacinski, P. R., C. McLean, J. E. Sykes, V. V. Clement-Jones, P. J. Lowry.
1981
. Iodination of proteins, glycoproteins, and peptides using a solid-phase oxidizing agent, 1,3,4,6-tetrachloro-3α,6α-diphenyl glycoluril (Iodogen).
Anal. Biochem.
117
:
136
.
30
Areschoug, T., S. Linse, M. Stalhammar-Carlemalm, L. O. Heden, G. Lindahl.
2002
. A proline-rich region with a highly periodic sequence in streptococcal β protein adopts the polyproline II structure and is exposed on the bacterial surface.
J. Bacteriol.
184
:
6376
.
31
Stålhammar-Carlemalm, M., L. Stenberg, G. Lindahl.
1993
. Protein Rib: a novel group B streptococcal cell surface protein that confers protective immunity and is expressed by most strains causing invasive infections.
J. Exp. Med.
177
:
1593
.
32
Larsson, C., M. Stålhammar-Carlemalm, G. Lindahl.
1996
. Experimental vaccination against group B streptococcus, an encapsulated bacterium, with highly purified preparations of cell surface proteins Rib and α.
Infect. Immun.
64
:
3518
.
33
Jarva, H., R. Janulczyk, J. Hellwage, P. F. Zipfel, L. Bjorck, S. Meri.
2002
. Streptococcus pneumoniae evades complement attack and opsonophagocytosis by expressing the pspC locus-encoded Hic protein that binds to short consensus repeats 8–11 of factor H.
J. Immunol.
168
:
1886
.
34
Altschul, S. F., W. Gish, W. Miller, E. W. Myers, D. J. Lipman.
1990
. Basic local alignment search tool.
J. Mol. Biol.
215
:
403
.
35
Higgins, D., J. Thompson, T. Gibson, J. D. Thompson, D. G. Higgins, T. J. Gibson.
1994
. CLUSTAL W: improving the sensitivity of progressive multiple sequence alignment through sequence weighting, position-specific gap penalties and weight matrix choice.
Nucleic Acids Res.
22
:
4673
.
36
Areschoug, T., M. Stalhammar-Carlemalm, C. Larsson, G. Lindahl.
1999
. Group B streptococcal surface proteins as targets for protective antibodies: identification of two novel proteins in strains of serotype V.
Infect. Immun.
67
:
6350
.
37
Jarva, H., T. S. Jokiranta, J. Hellwage, P. F. Zipfel, S. Meri.
1999
. Regulation of complement activation by C-reactive protein: targeting the complement inhibitory activity of factor H by an interaction with short consensus repeat domains 7 and 8–11.
J. Immunol.
163
:
3957
.
38
Hammerschmidt, S., S. R. Talay, P. Brandtzaeg, G. S. Chhatwal.
1997
. SpsA, a novel pneumococcal surface protein with specific binding to secretory Immunoglobulin A and secretory component.
Mol. Microbiol.
25
:
1113
.
39
Heden, L. O., E. Frithz, G. Lindahl.
1991
. Molecular characterization of an IgA receptor from group B streptococci: sequence of the gene, identification of a proline-rich region with unique structure and isolation of N-terminal fragments with IgA-binding capacity.
Eur. J. Immunol.
21
:
1481
.
40
Gordon, D. L., R. M. Kaufman, T. K. Blackmore, J. Kwong, D. M. Lublin.
1995
. Identification of complement regulatory domains in human factor H.
J. Immunol.
155
:
348
.
41
Kühn, S., P. F. Zipfel.
1996
. Mapping of the domains required for decay acceleration activity of the human factor H-like protein 1 and factor H.
Eur. J. Immunol.
26
:
2383
.
42
Sharma, A. K., M. K. Pangburn.
1996
. Identification of three physically and functionally distinct binding sites for C3b in human complement factor H by deletion mutagenesis.
Proc. Natl. Acad. Sci. USA
93
:
10996
.
43
Blackmore, T. K., V. A. Fischetti, T. A. Sadlon, H. M. Ward, D. L. Gordon.
1998
. M protein of the group A Streptococcus binds to the seventh short consensus repeat of human complement factor H.
Infect. Immun.
66
:
1427
.
44
Jokiranta, T. S., E. Giannakis, R. Ormsby, G. Lindahl, D. L. Gordon.
2002
. Structural analysis of the interaction between M5 protein and factor H domain 7 and identification of a distinct M protein binding site on domains 8–15.
Int. Immunopharmacol.
2
:
1343
.
45
Nicholas, K. B., H. B. J. Nicholas, D. W. I. Deerfield.
1997
. GeneDoc: analysis and visualization of genetic variation.
Embnew News
4
:
14
.