Microbiology (2006), 152, 1307–1317
DOI 10.1099/mic.0.28747-0
Streptococcus pneumoniae enolase is important for plasminogen binding despite low abundance of enolase protein on the bacterial cell surface Jan Kolberg,1 Audun Aase,1 Simone Bergmann,2 Tove K. Herstad,1 Gunnhild Rødal,1 Ronald Frank,3 Manfred Rohde4 and Sven Hammerschmidt2 1
Department of Bacteriology and Immunology, Norwegian Institute of Public Health, Box 4404 Nydalen, NO-0403 Oslo, Norway
Correspondence Jan Kolberg
2
[email protected]
Research Center for Infectious Diseases, University of Wu¨rzburg, Roentgenring 11, D-97070 Wu¨rzburg, Germany
3,4
Department of Chemical Biology3 and Department of Microbial Pathogenesis4, German Research Centre for Biotechnology, Mascheroder Weg 1, D-38124 Braunschweig, Germany
Received 6 December 2005 Accepted 20 January 2006
Enolase represents one of the anchorless surface proteins of Streptococcus pneumoniae and has previously been identified as a plasminogen-binding protein, endowing this pathogen with host proteolytic activity. In this study the mAb 245,C-6 (IgG1) was produced in a BALB/c mouse after immunizing with a protein fraction from S. pneumoniae. The mAb reacted with recombinant pneumococcal enolase both under non-denaturing and denaturing conditions. The epitope for the mAb was mapped to residues 55DKSRYGGLG63 of pneumococcal enolase using a peptide array. By applying the previously reported structure of enolase, this epitope was localized in a surface-exposed loop in each of the monomers of the octameric enolase. Previous immunoelectron microscopic studies, using polyclonal rabbit antibodies against enolase, depicted enolase on the cell surface but did not quantify the amount of surface-exposed enolase on viable pneumococci. Here, flow cytometry revealed no binding of mAb 245,C-6 to viable pneumococci, including TIGR4 and its non-encapsulated isogenic mutant, and only a minor increase of fluorescence intensity was measured when the polyclonal anti-enolase antibodies were used. In contrast, control antibodies recognizing the choline-binding proteins (CBPs) PspA and PspC showed high reactivities. The non-encapsulated TIGR4 did not show increased levels of antibody binding for mAb 245,C-6 or polyclonal anti-enolase antibodies, but revealed increased binding of polyclonal antibodies reacting with PspA or PspC. These results suggest that, compared to other surface-exposed proteins such as CBPs, the amount of enolase under the selected conditions is low. Flow cytometry, however, with FITC-labelled plasminogen demonstrated that the amount of surface-exposed enolase is important for plasminogen binding and, therefore, is also important for pneumococcal pathogenesis.
INTRODUCTION Streptococcus pneumoniae is an important human pathogen causing a variety of diseases including pneumonia, meningitis, sepsis and otitis media. A prerequisite for invasive pneumococcal diseases (infections) is the ability of the bacteria to colonize the mucosal surface and penetrate the mucosal barriers. During these processes binding of bacterial factors such as adhesins are essential for the Abbreviations: GAPDH, glyceraldehyde-3-phosphate dehydrogenase; GMFI, geometric mean fluorescence intensity; immuno-FESEM, field emission scanning immunoelectron microscopy; PspA, PspC, pneumococcal surface protein A or C.
0002-8747 G 2006 SGM
Printed in Great Britain
interactions with receptors on host cells. Among the bacterial factors that mediate direct adherence to human cells are phosphorylcholine (Cundell et al., 1995) and the choline-binding protein PspC, also known as SpsA and CbpA (Brooks-Walter et al., 1999; Hammerschmidt et al., 1997; Rosenow et al., 1997). The PspC protein binds to the secretory component of the polymeric Ig receptor or secretory IgA and this interaction mediates transmigration of pneumococci through mucosal epithelial cells (Elm et al., 2004; Zhang et al., 2000). In addition, PspC and the PspClike protein Hic have been shown to interact with complement factor H (Dave et al., 2004; Janulczyk et al., 2000). Many pathogenic bacteria also produce receptors for 1307
J. Kolberg and others
plasminogen which can be converted into the serine protease plasmin by bacterial or host plasminogen activators. This activated form of plasminogen can degrade extracellular matrix proteins and promote bacterial migration across the human extracellular matrix (Boyle & Lottenberg, 1997; Eberhard et al., 1999). Among the plasminogen-binding proteins in Gram-positive bacteria are glyceraldehyde-3phosphate dehydrogenase (GAPDH) and enolase, important glycolytic enzymes generally found in the cytosolic compartment (for review, see Pancholi & Chhatwal, 2003). Although these enzymes lack signal sequences and typical motifs for membrane anchoring, independent groups have shown the proteins to be surface-exposed on Streptococcus agalactiae (Hughes et al., 2002), Streptococcus mutans (Ge et al., 2004), Streptococcus pneumoniae (Bergmann et al., 2001, 2004b; Whiting et al., 2002), Streptococcus pyogenes (D’Costa et al., 2000; Pancholi & Fischetti, 1992, 1998) and Streptococcus suis (Jobin et al., 2004) by immunochemical and/or immunoelectron microscopic methods. The plasminogen-binding motifs in the enolase enzyme were initially traced to C-terminal lysine residues (Bergmann et al., 2001, 2004a; Derbise et al., 2004). However, a subsequent study identified a nine residue internal plasminogen-binding site in enolase (Bergmann et al., 2003). Furthermore, structural analysis indicated the surface exposure of this motif and its function as primary plasminogen binding site (Bergmann et al., 2003; Ehinger et al., 2004). The two glycolytic enzymes, GAPDH and enolase, have recently been shown to be surface-exposed and able to bind plasminogen in Listeria monocytogenes, a Gram-positive facultative intracellular pathogen (Schaumburg et al., 2004). Enolase has also been identified on the surface of fungal pathogens (Jong et al., 2003), parasites (Bernal et al., 2004; Jolodar et al., 2003) and eukaryotic cells, including some haematopoietic and cancer cells (Lopez-Alemany et al., 1994, 2003) where it binds plasminogen presumably to facilitate tissue invasion. Surface-associated streptococcal proteins like enolase have the potential to be candidates for the development of new protein-based vaccines. From the immunochemical and/or immunoelectron microscopic studies mentioned above, it is difficult to decide whether enolase is sufficiently exposed to function as a vaccine candidate. The extent of protection against systemic pneumococcal infection has been shown to be influenced by target antigen accessibility to circulating host antibodies (Gor et al., 2005). Flow cytometry is a technique that permits epitope analyses on a large number of live bacteria (Michaelsen et al., 2001; Kolberg et al., 2000, 2001, 2003). In the present study we have used both polyclonal and mAbs recognizing the pneumococcal enolase protein and other surface determinants. We demonstrate that the extent of surface exposure of the enolase protein seems to be relatively low on in vitro-grown pneumococci compared to other surface-exposed proteins. However, this amount of surface-exposed enolase is important to endow the pneumococcal surface with plasminogen and hence, with proteolytic activity. 1308
METHODS Bacterial strains. S. pneumoniae strains ATCC 11733 (Spn 51:
Cps+ serotype 2), NCTC 10319 (Spn 37: Cps+ serotype 35A), the non-encapsulated D39 derivative R6x (Tiraby & Fox, 1973), and TIGR4 (serotype 4) and its non-encapsulated mutant FP23 were used in this study. The mutant FP23 was kindly provided by Francesco Iannelli, Siena, Italy (Pearce et al., 2002). The construction of the knock-out strains for enolase and pspC has been described previously (Bergmann et al., 2003; Hammerschmidt et al., 2000). The pspA mutants were obtained by transformation of pneumococci with plasmid pMSH6.1. This plasmid carries the full-length pspA gene of strain Spn 51 disrupted by insertion of an erythromycin gene cassette at nucleotide position 668 of pspA. Pneumococci were cultured on blood agar plates (Oxoid) at 37 uC and 5 % CO2, or in Todd–Hewitt broth (Roth) supplemented with 0?5 % yeast extract (THY). When appropriate, erythromycin (5 mg ml21) was added to the growth medium. Other bacteria (besides Lactococcus lactis IL 1403) that were used in this study have been described previously (Kolberg et al., 1997b). Recombinant enolase protein. The His6-tagged S. pneumoniae
enolase was expressed in the Escherichia coli host strain M15(pREP4) and protein purification was performed as described previously (Bergmann et al., 2001). Production of mAb 245,C-6. Proteins from a heat-treated (60 uC
for 30 min) pneumococcal clinical isolate (1679/94) were extracted with detergents as described previously (Kolberg & Sletten, 1996) and run through a column with an mAb (anti-phosphoglycerate kinase) immobilized onto CNBr-Sepharose. Proteins that bound to this column were eluted with 4 M guanidine/HCl in PBS. After removal of the guanidine salt by dialysis against PBS this protein fraction was used for immunization of BALB/C mice. The procedure for mAb production and isotyping were as described previously (Kolberg & Sletten, 1996). Other mAbs used were also produced at NIPH, Oslo, Norway. mAb 145,F-2 (IgM) reacts with the phosphorylcholine determinant (Kolberg et al., 1997a), mAb 149,B-3 (IgG2a) and 159,D-7 (IgG2a) detect epitopes on PspA (Kolberg et al., 2001, 2003) and 230,B-9 (IgG1) reacts with heat-shock protein 70 from S. pneumoniae and other streptococci (Kolberg et al., 2000). Rabbit antisera. Rabbit sera were obtained after immunization with recombinant enolase protein, purified PspA or purified PspC. Their specificities were confirmed by immunoblotting using whole bacterial cell lysates and recombinant enolase, PspA or PspC protein (Bergmann et al., 2003; Hammerschmidt et al., 2000). ELISA. Measurements were performed as described previously (Kolberg & Jones, 1998). Briefly, flat-bottomed microtitre plates (Nunc Immunoplate) were coated overnight at room temperature with recombinant enolase protein (2 mg ml21) in PBS. Bound mAbs were detected with alkaline phosphatase-conjugated goat anti-mouse Ig (Sigma) and p-nitrophenylphosphate as substrate. Isolation of pneumococcal cell wall and cytosolic proteins.
This was done essentially as described by Bergmann et al. (2001) using strain Spn 51 (ATCC 11733). Different amounts of the cytosolic and membrane fractions, respectively, were spotted onto a nitrocellulose membrane using a Bio-Dot SF microfiltration apparatus (Bio-Rad) and assayed for their reactivity with the polyclonal anti-enolase antiserum (data not shown) and mAb 245,C-6, respectively. SDS-PAGE and immunoblotting. Whole bacterial cell lysates and
recombinant enolase were subjected to SDS-PAGE with 4 % stacking and 10 % separating gels, electrotransferred to a nitrocellulose or polyvinylidene fluoride membrane (Immobilon-P; Millipore) and probed with antibodies diluted in PBS containing 3 % bovine serum Microbiology 152
Pneumococcal enolase and surface expression albumin (BSA). Antibody binding was detected with peroxidaselabelled rabbit antisera against mouse immunoglobulins or peroxidase-labelled goat antisera against rabbit immunoglobulins. The polyclonal anti-enolase antibodies and mAb 245,C-6 were used at dilutions of 1 : 100 and 1 : 10, respectively, in PBS in immunoblot and dot blot experiments. Secondary antibodies, obtained from DakoCytomatation, were used at dilutions of 1 : 1000 with the substrate 3-amino-9-ethylcarbazole and hydrogen peroxide in sodium acetate buffer (pH 5?0) or binding was detected by chemiluminescence using ECL (Amersham Pharmacia). Non-immune human sera or plasma proteins were obtained from healthy individuals. Enolase peptides synthesized on a cellulose membrane and detection of mAb binding. A membrane with 141 peptides con-
sisting of 15 aa each with an offset of three amino acids, was prepared as described previously (Bergmann et al., 2003). The peptide sequences are given as supplementary material in Bergmann et al. (2003). The membrane was incubated for 7 h with blocking buffer from Genosys Biotechnologies, followed by overnight incubation with mAb 245,C-6 used as hybridoma culture supernatant diluted 1 : 5 in blocking buffer. After washing, the membrane was incubated for 2 h with peroxidase-conjugated rabbit anti-mouse immunoglobulins (dilution 1 : 200) from DakoCytomatation. After washing the membrane was incubated with Western Lightening Chemiluminescence Reagent Plus substrate from Perkin-Elmer Life Sciences and antibody binding was detected using a Kodak Image Station 2000R. Binding of antibodies to viable pneumococci using flow cytometry. The reactivity of the antibodies was tested against live
pneumococci using flow cytometry. Bacteria were grown on horse blood agar overnight at 37 uC in 5 % CO2. Colonies were collected and resuspended in Todd–Hewitt broth supplemented with 0?5 % yeast extract and cultured for 4 h at 37 uC in 5 % CO2. The bacterial suspensions were spun down by centrifugation and mixed with Hanks’ balanced salt solution supplemented with 2 mg BSA ml21 (HBSS/BSA) to an OD620 of 0?64. For the binding experiments, 10 ml bacteria were mixed with 50 ml hybridoma cell culture supernatant or a 1 : 100 dilution of rabbit antisera. The suspensions were incubated for 45 min at 37 uC and washed three times. The binding of mouse antibodies was detected by incubation with an optimal dilution of FITC-conjugated sheep anti-mouse IgG antibody preparation and rabbit antibodies using an FITC-conjugated sheep anti-rabbit IgG antibody preparation (both produced at NIPH, Oslo, Norway) for 30 min at 37 uC. Finally the bacteria were washed once and the fluorescence was analysed by flow cytometry using a CyFlow ML (Partec). The pneumococci were detected using logforward and log-side scatter dot-plots, and a gating region was set to exclude debris and larger aggregates of bacteria. Six thousand bacteria were analysed for fluorescence using log-scale amplification. The geometric mean fluorescence intensity (GMFI) of the various antibodies was recorded as a measure for binding activity. Field emission scanning immunoelectron microscopy (immunoFESEM) and post-embedding labelling. For immuno-FESEM,
bacteria (S. pneumoniae strain Spn 51) were incubated with mAb 245,C-6 (hybridoma cell culture supernatant) or polyclonal antienolase antibodies (100 mg IgG protein ml21) for 1 h at 30 uC. After washing several times with PBS, the bound antibodies were visualized by incubation (30 min at 30 uC) with protein A–gold complexes (15 nm diam.) for polyclonal antibodies and with goatanti mouse gold-complexes (15 nm diam.) for mAbs. After washing with PBS, bacteria were fixed with 5 % formaldehyde for 15 min at room temperature and adsorbed onto carbon-coated Formvar grids, washed with distilled water and air-dried. Samples were then examined in a Zeiss field emission scanning electron microscope (DSM982 Gemini) at an acceleration voltage of 5 kV using the Everhardt–Thornley SE-detector and the in-lens detector in a 75/25 http://mic.sgmjournals.org
ratio. For post-embedding detection of enolase, bacteria were fixed with 0?2 % glutaraldehyde and 0?5 % formaldehyde for 1 h on ice, washed with PBS containing 10 mM glycine and dehydrated with a graded series of ethanol following the progressive lowering of temperature protocol (PLT method). Samples were then embedded in Lowicryl K4M at 235 uC, polymerized with UV light (366 nm, 2 days at 235 uC, 2 days at room temperature) and ultrathin sections were cut with glass knives. Sections were incubated with mAb 245,C-6 (hybridoma cell culture supernatant) or polyclonal antienolase antibodies (100 mg IgG protein ml21) overnight at 4 uC, washed with PBS and incubated with the respective gold markers as above for 30 min at room temperature. After washing with PBS containing 0?01 % Triton X-100 and with distilled water, samples were air-dried. Counterstaining was done with 4 % uranyl acetate for 1 min. Samples were then examined in a Zeiss transmission electron microscope (EM910) at an acceleration voltage of 80 kV. Plasminogen binding analysed by flow cytometry. Pneumo-
cocci cultured to mid-exponential phase were incubated with different amounts of FITC-labelled plasminogen and binding was analysed by flow cytometry. Labelling of plasminogen was performed by incubation of 400 mg plasminogen with FITC in 500 ml 0?1 M sodium carbonate buffer, pH 9?2, for 1 h at 30 uC in darkness. Pneumococci were cultured to mid-exponential phase, washed with PBS and 107 pneumococci were incubated with FITC-labelled plasminogen for 30 min at 37 uC. Thereafter, pneumococci were washed and resuspended in PBS for analysis using a FACSCalibur (Becton Dickinson). Detection of pneumococci and measurement of binding activity was performed as described above.
RESULTS Reactivity of mAb 245,C-6 with enolase and mapping of the binding epitope to residues 55–63 mAb IgG1 was made after immunization of mice with a pneumococcal protein fraction eluted from a column with immobilized anti-phosphoglycerate kinase. mAb 245,C-6 reacted with recombinant pneumococcal enolase, pneumococcal cell wall and cytosolic proteins as demonstrated by immunoblotting (Fig. 1a). This enzyme must therefore have been co-purified from the immunoaffinity column with phosphoglycerate kinase which is also a glycolytic enzyme. The purification of the antigen included an elution step with guanidine/HCl which could result in mAbs reacting only with denatured enolase protein. Strong reactions of mAb 245,C-6 with proteins of the cytoplasm, membrane fraction and recombinant enolase protein under non-denaturing dot blotting (Fig. 1b and c) indicated that the epitope is also recognized in the native protein. This was further supported by strong reactions in ELISA with immobilized recombinant enolase protein (results not shown). Our immunoblot studies with mAb 245,C-6 showed that the epitope was also expressed in other streptococci, (S. agalactiae, Streptococcus mitis, S. mutans and Streptococcus sanguis), but not in S. pyogenes, Lactococcus lactis or Enterococcus faecalis (results not shown). Gram-negative bacteria such as Escherichia coli, Haemophilus influenzae and Neisseria meningitidis were non-reactive with the mAb. Epitope mapping with peptides spanning the entire pneumococcal enolase protein showed strong reactions with three peptides (Fig. 2) and their 1309
J. Kolberg and others
Evaluation of the enolase mAb-binding sites in the enolase octamer The structure of the enolase, comprising residues 1–434, has been published previously (Ehinger et al., 2004). The Protein Data Bank entry 1W6T was used to localize the identified binding epitope consisting of residues 55DKSRYGGLG63 in each monomer of the octamer structure of the pneumococcal enolase. The epitope occurs four times on each side of the octamer and is exposed on the surface of the enolase octamer (Fig. 3). Flow cytometric analyses using mAbs
Fig. 1. Immunoblotting (a) and dot blotting (b, c) with mAb 245,C-6 and polyclonal anti-enolase (a-enolase). Whole-cell lysate (WCL), and cytosolic (CP) and membrane proteins (MP) of pneumococci were analysed for their reactivity with the polyclonal anti-enolase antiserum and mAb 245,C-6 under denaturing conditions (a) and non-denaturing conditions (b). Binding of mAb 245,C-6 and polyclonal anti-enolase was also tested with different amounts of native His6-tagged enolase immobilized on nitrocellulose (c). The secondary anti-mouse or anti-rabbit peroxidase-conjugated antibodies showed no reaction with the different protein fractions and enolase, respectively (data not shown).
sequences are also listed in Fig. 2. The identified epitope consists of the 9 amino acid sequence 55DKSRYGGLG63. The epitope mapping was supported by immunoblot analyses of bacteria with enolase sequences reported in the TIGR database. The 9 amino acid residues defining the epitope are found in S. mutans and S. agalactiae which showed positive immunoblot reactions. Substitution of Gly in position 60 to Leu as reported for enolase from S. pyogenes, or Gly-60 to Asp in L. lactis enolase, resulted in failure of the mAb reaction (Fig. 2). E. faecalis with substitutions at positions 57, with Ala instead of Ser; 60, with Leu instead of Glu; and 62, with Lys instead of Leu as in S. pneumoniae, was also non-reactive with the mAb. Substitutions within the mAb-binding region were also found for the three examined Gram-negative species which did not react with the mAb (Fig. 2). 1310
The anti-enolase mAb (245,C-6) did not show a positive reaction with any of the five examined pneumococcal strains, whereas the PspA-specific mAbs showed good surface accessibility of this protein (Fig. 4). The results for the Norwegian clinical isolate 1675/94 (serotype 12F) is not shown. An mAb (230,B-9) against intracellular structures was used to see whether the pneumococci were intact or not. This mAb detects an epitope on pneumococcal heat-shock protein 70 and was non-reactive, which is in accordance with our previous data (Kolberg et al., 2000, 2001, 2003). We have compared our results with the reactivity of mAbs generated against PspA proteins. PspA is an antigenically highly variable protein and has been divided into two major families (Hollingshead et al., 2000). mAb 149,B-3 reacted only with TIGR4 belonging to PspA family 2 (Shaper et al., 2004). In contrast, mAb 159,D-7 did not react with TIGR4, but did react with all the other examined strains, including 1675/94. mAb 159,D-7 has been shown to detect an epitope within amino acid residues 67–236 of PspA from strain R6 (Kolberg et al., 2003) which belongs to family 1 (Hollingshead et al., 2000). One explanation for the non-reactivity of the anti-enolase mAb 245,C-6 could be that the epitope was hidden by other surface structures such as the capsule or surface-exposed proteins. The TIGR4 non-encapsulated isogenic mutant FP23 did not show any binding of the enolase mAb, but did show increased binding of the PspA-specific mAb 149,B-3 and the phosphorylcholine-specific mAb 145,F-2. The PspA and PspC knock-out strains were, as for their wild-type counterparts, non-reactive with the anti-enolase mAb. Deletion of PspA from strain R6x resulted in loss of binding of the PspA-specific mAb 159,D-7 as expected. Flow cytometric analyses of enolase protein surface expression using polyclonal sera The positive control rabbit anti-PspA antibodies showed strong binding to almost all bacteria in the three examined pneumococcal strains (Fig. 5). The other positive control antibodies directed against PspC showed an intermediate binding to the bacteria. The negative control serum from a non-immunized rabbit was non-reactive against all the bacteria. With the rabbit anti-enolase antibodies there was faint binding to three strains, Spn 37, Spn 51 and R6x (GMFI range 1?0–1?4), compared to the negative control serum Microbiology 152
Pneumococcal enolase and surface expression
Fig. 2. Epitope mapping of mAb 245,C-6 using peptides covering the complete enolase protein sequence. (a) Spot membrane of the 434 aa sequence of enolase divided into 141 overlapping peptides of 15 aa each, with an offset of 3 aa. mAb binding was detected on spot numbers 17, 18 and 19. (b) Sequences of spots 16–20 and binding reactivity with mAb 245,C-6. (c) Whole bacterial cells were subjected to SDS-PAGE, electrotransferred to a nitrocellulose membrane and immunostained with mAb 245,C6. Strpn, S. pneumoniae; Strmu, S. mutans; Strag, S. agalactiae; Strpy, S. pyogenes; Lacla, Lactococcus lactis; Entfa, Enterococcus faecalis; Ecoli, Escherichia coli; Nm, Neisseria meningitidis; Hi, Haemophilus influenzae.
(GMFI=0?5) and the positive PspA serum (GMFI range 18?8–68). Two strains, TIGR4 (Fig. 5) and 1675/94 (data not shown), did not bind the anti-enolase antibodies. Deletion of PspA or PspC resulted in loss of binding of the PspA- and PspC-specific antibodies, respectively, as expected (Fig. 5).
Binding of mAb 245,C-6 analysed by immunoelectron microscopy Immuno-FESEM and post-embedding labelling was employed to reveal the binding of mAb to surface-exposed enolase and enolase localized in the cytoplasm. The
Fig. 3. Epitopes recognized by mAb 245,C6 on the surface of the pneumococcal enolase protein. Each epitope DKSRYGGLG in the enolase octamer is depicted in black. Front view (a) and side view (b). The figure was prepared using PyMol (DeLano Scientific LLC).
http://mic.sgmjournals.org
1311
J. Kolberg and others
Fig. 4. Panel of histograms showing the binding of the different mAbs to different strains and mutants of live pneumococci. As a negative control, HBSS/BSA was used. The number next to the histograms reflects the geometric mean fluorescence intensity of the whole population. Strain abbreviations: FP23, non-encapsulated isogenic TIGR4 mutant; PspA” and PspC”, isogenic mutants without PspA and PspC, respectively. mAb 145,F-2 reacts with the phosphorylcholine determinant, mAbs 149,B-3 and 159,D-7 detect epitopes on PspA and 230,B-9 reacts with heat-shock protein 70.
polyclonal anti-enolase IgG has previously been shown to detect surface-exposed enolase (Bergmann et al., 2001). Immuno-FESEM indicated that enolase is present on the bacterial surface for both antibodies (arrows in Fig. 6a, polyclonal antibodies, and Fig. 6b, mAb 245,C-6). Postembedding labelling of ultrathin sections revealed the reactivity of the mAb with the cytoplasmic enolase and cell-surface-exposed enolase (arrowheads in Fig. 6e). The polyclonal anti-enolase antibodies exhibited an identical labelling pattern with higher label intensities (Fig. 6a and d) than that seen with the mAb (Fig. 6b and e). Flow cytometric analysis of plasminogen binding to pneumococci Previous immunoelectron microscopic studies demonstrated binding of plasminogen to encapsulated and nonencapsulated pneumococci, respectively, with a major role of surface-displayed enolase (Bergmann et al., 2001). The pneumococcal enolase mutant enoint/del with inactivated plasminogen-binding sites retained 50 % binding activity 1312
to immobilized plasminogen (Bergmann et al., 2003). In addition, plasmin-mediated degradation of extracellular matrix and dissolution of a fibrin matrix by this mutant was substantially reduced when compared to the isogenic wild-type (Bergmann et al., 2005). Here, flow cytometry was employed to analyse differences in plasminogen binding of pneumococcal strains and isogenic enolase mutants. Pneumococci grown to mid-exponential phase demonstrated plasminogen-binding activity independent of the serotype used. The measured increase in mean fluorescence intensity coincided with the amount of FITC-labelled plasminogen that was employed in the binding experiments with pneumococci. The mean fluorescence intensity was significantly enhanced when the bacteria were incubated with a higher dose of plasminogen (Fig. 7a). More importantly, the mean fluorescence intensity was substantially decreased for the enolase mutants. These data confirm the important role of the enolase in plasminogen binding to the bacterial cell surface, despite its low abundance on the pneumococcal surface compared to other surface-exposed proteins (Fig. 7b). Microbiology 152
Pneumococcal enolase and surface expression
Fig. 5. Panel of histograms showing the binding of the different rabbit antisera to live pneumococci. Serum from a nonimmunized rabbit was used as a negative control. The number next to the histograms reflects the geometric mean fluorescence intensity of the whole population. Abbreviations are as defined in the legend to Fig. 4.
DISCUSSION The initiation of many infections involves the adherence of bacteria to host cells followed by spreading to the local tissue. Among the mechanisms of bacterial migration through the human extracellular matrix are the binding of plasminogen and subsequent activation to the proteolytic form plasmin. The serine protease plasmin degrades extracellular matrix proteins and promotes migration (Boyle & Lottenberg, 1997; Eberhard et al., 1999; Bergmann et al., 2005). The S. pyogenes, S. mutans and pneumococcal enolases are plasminogenbinding proteins which, besides their cytosolic location, can also be surface-displayed as shown independently by several groups using immunochemical and/or immunoelectron microscopic methods (Bergmann et al., 2001; Ge et al., 2004; Pancholi & Fischetti, 1998). Our flow cytometry analyses with polyclonal rabbit antibodies showed low amounts of surface-exposed enolase protein on three out of five examined pneumococcal strains. Among the three strains that displayed some enolase protein on the surface were the non-encapsulated strain R6x and two strains which are known to have low-level encapsulation (Bergmann et al., 2005; Hammerschmidt et al., 2005). One possible explanation for this could be that the accessibility of enolase epitopes is influenced by the level of capsular material on the surface. This is most probably not the case for http://mic.sgmjournals.org
the enolase, since neither TIGR4 nor its non-encapsulated isogenic mutant FP23 showed any binding of the antienolase antibodies. In addition, previous immunoelectron microscopic studies depicted the enolase not only in proximity to the cell wall, but also at the outer edge of the capsular polysaccharide (Bergmann et al., 2001). Other surface structures like proteins might affect the availability of enolase epitopes and therefore interfere with the reactivity of the mAb 245,C-6. By using PspA and PspC knock-out strains and their isogenic wild-type counterparts, we could not see that the presence of these proteins on the bacterial surface influenced the accessibility of enolase epitopes. The PspA- and PspC-specific antibodies did not react with the mutants that lack these proteins, supporting the antibody specificities. The epitope of the anti-enolase mAb (245,C-6) was mapped by peptide analyses to residues 55–63 of the N-terminal domain of enolase. From the published crystal structure of the S. pneumoniae enolase protein (Ehinger et al., 2004), these residues seem to be surface-exposed on the enolase octamer. The recognition of this epitope by mAb 245,C-6 on the protein was insufficient to allow measurement of the enolase on viable pneumococci by flow cytometry. However, the results did not exclude a surface location of the enolase protein on the bacteria, because the epitopes might be hidden by other surface structures. It is also possible and 1313
J. Kolberg and others
Fig. 6. Immunoelectron microscopic studies. Immuno-FESEM was applied to reveal the surface location of enolase. Polyclonal antibodies (a) as well as mAb (b) show enolase (arrows) on the surface of S. pneumoniae strain Spn 51, whereas the control incubated only with goat-anti-mouse gold complexes showed no labelling (c). Post-embedding labelling also demonstrates the surface location of enolase [arrowheads in (d) (polyclonal antibodies) and (e) (mAb)] as well as the cytoplasmic localization of enolase in both (d) and (e).
most probably more likely that the mAb epitope density on the bacterial surface is too low to be detected, as indicated by the employment of polyclonal rabbit anti-enolase antibodies in our flow cytometric analysis. The results revealed only a faint staining of the enolase in some of the strains. Our immunoelectron microscopic studies indicated that mAb 245,C-6 is reactive with surface-exposed and cytoplasmic enolase as well. However, both the mAb and the polyclonal rabbit anti-enolase showed a relatively weak surface staining
of the enolase, confirming the low amount of surfaceexposed enolase. Most pneumococcal isolates are heterogeneous populations with at least two phenotypes, opaque and transparent, which express different amounts of phosphorylcholine on the surface (Weiser, 1998). This could be an explanation for the heterogeneous pneumococcal populations we found by flow cytometry with the phosphorylcholine-specific mAb 145,F-2.
Fig. 7. Binding of FITC-plasminogen to pneumococci. Wild-type S. pneumoniae (a) and the isogenic enoint/del mutants of D39 and Spn 51 (b), which contain both functionally inactivated plasminogen-binding sites (Bergmann et al., 2005), were incubated with FITC-labelled plasminogen. The GMFI values are shown when the bacteria were incubated with 1 (grey bars) or 5 mg (black bars) FITC-labelled plasminogen (a) or with 5 mg FITC-labelled plasminogen (b; grey bars, wild-type; white bars, enoint/del mutant). FITC-plasminogen binding was measured by a FACScan cytometer as the increase in fluorescent intensity. Representative data from independent experiments are shown. 1314
Microbiology 152
Pneumococcal enolase and surface expression
Secondary ‘moonlighting’ functions of streptococcal enolases have been clearly demonstrated by their ability to bind plasminogen (Bergmann et al., 2001; Ge et al., 2004; Pancholi & Fischetti, 1998). Here we have shown by flow cytometry that despite the low amount of enolase on the surface, more than 50 % of bound plasminogen seems to be related to enolase. Previous mouse infection and in vitro experiments, respectively, indicated that binding of plasminogen to pneumococci, and in particular to surface-exposed enolase, is important for in vivo virulence and plasminmediated degradation of the extracellular matrix, respectively (Bergmann et al., 2003, 2005). It is clear that bacteria respond to environmental signals, and phenotypes determined from in vitro growth cannot necessarily be extrapolated to the in vivo situation (Marra et al., 2002). Microarray analyses of pneumococcal gene expression during invasive disease has shown that the expression of most genes (about 90 %) does not change from in vitro growth medium to in vivo conditions (Orihuela et al., 2004). However, the pspA gene did show increased expression, whereas the enolase gene showed decreased transcription during bacterial growth in blood, infected cerebrospinal fluid and bacteria attached to a pharyngeal epithelial cell line. This study demonstrated that enolase protein expression is influenced by environmental conditions, but with even lower expression in vivo than in vitro. It is largely unknown how glycolytic enzymes like enolase and GAPDH that lack a signal peptide required for secretion and peptidoglycan-anchoring motifs, are transported to and associated with bacterial surface structures or which factors influence this process. It also remains open as to whether these proteins are secreted across the bacterial cell membrane by the general secretory (Sec) pathway. One study suggested that the secretion of the Listeria monocytogenes enolase, which also binds plasminogen, is SecA2-dependent (Lenz et al., 2003). Another study speculated that surface GADPH and enolase do not come from inside the bacteria but are ‘scavenged’ from other cells that have undergone programmed cell death through allolysis (Guiral et al., 2005). It has been hypothesized that the polymorphism exhibited by certain S. pneumoniae antigenic proteins is related to immunological selection and that these proteins are readily accessible for antibody binding while highly conserved antigens are generally not readily available for antibodies on the surface of the intact pneumococcus (Gor et al., 2005; Hollingshead et al., 2000; Murphy, 1993). Our flow cytometric analyses seem to support this hypothesis since the antigen-variable proteins PspA and PspC were readily detected on the surface of the examined pneumococcal strains, whereas there was low surface expression of the highly conserved enolase protein. As a candidate for pneumococcal protein vaccine formulations, the presence of antibodies against the enolase protein has been analysed in sera from young children with and without a history of pneumococcal contact (Adrian et al., 2004; Ling et al., 2004). No significant correlation between antibody titres and http://mic.sgmjournals.org
carriage or infection in the children was found (Adrian et al., 2004). Since the enolase protein is a conserved glycolytic enzyme, it was concluded that the antibodies in young children probably represented cross-reactive epitopes on commensal bacteria (Adrian et al., 2004). The accessibility of epitopes on viable bacteria is of critical importance for antibodies to protect against bacterial infections. PspA and PspC, which consistently appear to induce protection in several studies (reviewed by Poolman, 2004), were clearly surfaced-exposed in our assays. The low extent of surface exposure of the enolase protein found in our in vitro assay may question its role as a vaccine candidate. It is reasonable to assume that a certain amount of antigen must be present for antibodies to be protective. The finding of no significant correlation between antibody titres to the enolase protein and pneumococcal contact in young children as reported by Adrian et al. (2004) might be explained by low surface exposure of the protein. In relation to vaccine strategies it should be noted that there are several reports showing a marked increase in antibodies to the phylogenetically highly conserved enolase protein in a variety of autoimmune diseases (Pancholi, 2001). Fontan et al. (2000) described that antibodies against streptococcal enolase cross-reacted with human enolase expressed on inflammatory cell surfaces and thus could play an important role in the initiation of autoimmune disease related to streptococcal infections. In summary, the low amount of surface-exposed enolase protein on in vitro-grown pneumococci suggests that the enolase is not a promising vaccine candidate. However, pneumococcal colonization is a multifactorial process which also might involve changes in expression of the enolase protein and other proteins. Despite the low amount of enolase protein on in vitro-grown pneumococci, the efficiency of plasminogen binding is high and this amount is sufficient to contribute to pneumococcal pathogenesis as shown previously (Bergmann et al., 2003).
ACKNOWLEDGEMENTS We thank Anne Ramstad Alme and Gro Lermark at the Bacteriological Unit (Oslo), Christa Albert (Research Centre for Infectious diseases, Wu¨rzburg) and Ina Schleicher (German Research Center for Biotechnology) for excellent technical assistance. This work was supported in parts by the Deutsche Forschungsgemeinschaft (Sonderforschungsbereich 479-A7 to S. H. and RO 2407/1-2 to M. R.).
REFERENCES Adrian, P. V., Bogaert, D., Oprins, M., Rapola, S., Lahdenkari, M., Kilpi, T., de Groot, R., Kayhty, H. & Hermans, P. W. (2004).
Development of antibodies against pneumococcal proteins alphaenolase, immunoglobulin A1 protease, streptococcal lipoprotein rotamase A, and putative proteinase maturation protein A in relation to pneumococcal carriage and otitis media. Vaccine 22, 2737–2742. 1315
J. Kolberg and others
Bergmann, S., Rohde, M., Chhatwal, G. S. & Hammerschmidt, S. (2001). alpha-Enolase of Streptococcus pneumoniae is a plasmin-
(ogen)-binding protein displayed on the bacterial cell surface. Mol Microbiol 40, 1273–1287. Bergmann, S., Wild, D., Diekmann, O., Frank, R., Bracht, D., Chhatwal, G. S. & Hammerschmidt, S. (2003). Identification of a
novel plasmin(ogen)-binding motif in surface displayed alphaenolase of Streptococcus pneumoniae. Mol Microbiol 49, 411–423. Bergmann, S., Rohde, M., Chhatwal, G. S. & Hammerschmidt, S. (2004a). Characterization of plasmin(ogen) binding to Streptococcus
pneumoniae. Indian J Med Res 119, 29–32. Bergmann, S., Rohde, M. & Hammerschmidt, S. (2004b).
Glyceraldehyde-3-phosphate dehydrogenase of Streptococcus pneumoniae is a surface-displayed plasminogen-binding protein. Infect Immun 72, 2416–2419.
Ge, J., Catt, D. M. & Gregory, R. L. (2004). Streptococcus mutans
surface alpha-enolase binds salivary mucin MG2 and human plasminogen. Infect Immun 72, 6748–6752. Gor, D. O., Ding, X., Briles, D. E., Jacobs, M. R. & Greenspan, N. S. (2005). Relationship between surface accessibility for PpmA, PsaA,
and PspA and antibody-mediated immunity to systemic infection by Streptococcus pneumoniae. Infect Immun 73, 1304–1312. Guiral, S., Mitchell, T. J., Martin, B. & Claverys, J. P. (2005).
Competence-programmed predation of noncompetent cells in the human pathogen Streptococcus pneumoniae: genetic requirements. Proc Natl Acad Sci U S A 102, 8710–8715. Hammerschmidt, S., Talay, S. R., Brandtzaeg, P. & Chhatwal, G. S. (1997). SpsA, a novel pneumococcal surface protein with specific
binding to secretory immunoglobulin A and secretory component. Mol Microbiol 25, 1113–1124.
Bergmann, S., Rohde, M., Preissner, K. T. & Hammerschmidt, S. (2005). The nine residue plasminogen-binding motif of the pneumo-
Hammerschmidt, S., Tillig, M. P., Wolff, S., Vaerman, J. P. & Chhatwal, G. S. (2000). Species-specific binding of human secretory
coccal enolase is the major cofactor of plasmin-mediated degradation of extracellular matrix, dissolution of fibrin and transmigration. Thromb Haemost 94, 304–311.
component to SpsA protein of Streptococcus pneumoniae via a hexapeptide motif. Mol Microbiol 36, 726–736.
Bernal, D., de la Rubia, J. E., Carrasco-Abad, A. M., Toledo, R., Mas-Coma, S. & Marcilla, A. (2004). Identification of enolase as a
plasminogen-binding protein in excretory-secretory products of Fasciola hepatica. FEBS Lett 563, 203–206.
Hammerschmidt, S., Wolff, S., Hocke, A., Rosseau, S., Muller, E. & Rohde, M. (2005). Illustration of pneumococcal polysaccharide
capsule during adherence and invasion of epithelial cells. Infect Immun 73, 4653–4667. Hollingshead, S. K., Becker, R. & Briles, D. E. (2000). Diversity of
invasive human pathogens. Thromb Haemost 77, 1–10.
PspA: mosaic genes and evidence for past recombination in Streptococcus pneumoniae. Infect Immun 68, 5889–5900.
Brooks-Walter, A., Briles, D. E. & Hollingshead, S. K. (1999). The
Hughes, M. J., Moore, J. C., Lane, J. D. & 13 other authors (2002).
pspC gene of Streptococcus pneumoniae encodes a polymorphic protein, PspC, which elicits cross-reactive antibodies to PspA and provides immunity to pneumococcal bacteremia. Infect Immun 67, 6533–6542.
Identification of major outer surface proteins of Streptococcus agalactiae. Infect Immun 70, 1254–1259.
Cundell, D. R., Gerard, N. P., Gerard, C., Idanpaan-Heikkila, I. & Tuomanen, E. I. (1995). Streptococcus pneumoniae anchor to activated
that interferes with complement function. J Biol Chem 275, 37257–37263.
human cells by the receptor for platelet-activating factor. Nature 377, 435–438.
Jobin, M. C., Brassard, J., Quessy, S., Gottschalk, M. & Grenier, D. (2004). Acquisition of host plasmin activity by the Swine pathogen
Boyle, M. D. & Lottenberg, R. (1997). Plasminogen activation by
Dave, S., Carmicle, S., Hammerschmidt, S., Pangburn, M. K. & McDaniel, L. S. (2004). Dual roles of PspC, a surface protein of
Streptococcus pneumoniae, in binding human secretory IgA and factor H. J Immunol 173, 471–477.
Janulczyk, R., Iannelli, F., Sjoholm, A. G., Pozzi, G. & Bjorck, L. (2000). Hic, a novel surface protein of Streptococcus pneumoniae
Streptococcus suis serotype. Infect Immun 72, 606–610. Jolodar, A., Fischer, P., Bergmann, S., Buttner, D. W., Hammerschmidt, S. & Brattig, N. W. (2003). Molecular cloning of
D’Costa, S. S., Romer, T. G. & Boyle, M. D. (2000). Analysis of
an alpha-enolase from the human filarial parasite Onchocerca volvulus that binds human plasminogen. Biochim Biophys Acta 1627, 111–120.
expression of a cytosolic enzyme on the surface of Streptococcus pyogenes. Biochem Biophys Res Commun 278, 826–832.
Jong, A. Y., Chen, S. H., Stins, M. F., Kim, K. S., Tuan, T. L. & Huang, S. H. (2003). Binding of Candida albicans enolase to plasmin(ogen)
Derbise, A., Song, Y. P., Parikh, S., Fischetti, V. A. & Pancholi, V. (2004). Role of the C-terminal lysine residues of streptococcal surface
results in enhanced invasion of human brain microvascular endothelial cells. J Med Microbiol 52, 615–622.
enolase in Glu- and Lys-plasminogen-binding activities of group A streptococci. Infect Immun 72, 94–105.
Kolberg, J. & Jones, C. (1998). Monoclonal antibodies with specifi-
Eberhard, T., Kronvall, G. & Ullberg, M. (1999). Surface bound
plasmin promotes migration of Streptococcus pneumoniae through reconstituted basement membranes. Microb Pathog 26, 175–181. Ehinger, S., Schubert, W. D., Bergmann, S., Hammerschmidt, S. & Heinz, D. W. (2004). Plasmin(ogen)-binding alpha-enolase from
Streptococcus pneumoniae: crystal structure and evaluation of plasmin(ogen)-binding sites. J Mol Biol 343, 997–1005.
cities for Streptococcus pneumoniae group 9 capsular polysaccharides. FEMS Immunol Med Microbiol 20, 249–255. Kolberg, J. & Sletten, K. (1996). Monoclonal antibodies that
recognize a common pneumococcal protein with similarities to streptococcal group A surface glyceraldehyde-3-phosphate dehydrogenase. Infect Immun 64, 3544–3547. Kolberg, J., Hoiby, E. A. & Jantzen, E. (1997a). Detection of the
phosphorylcholine epitope in streptococci, Haemophilus and pathogenic Neisseriae by immunoblotting. Microb Pathog 22, 321–329.
Elm, C., Braathen, R., Bergmann, S., Frank, R., Vaerman, J. P., Kaetzel, C. S., Chhatwal, G. S., Johansen, F. E. & Hammerschmidt, S. (2004).
Kolberg, J., Hoiby, E. A., Lopez, R. & Sletten, K. (1997b).
Ectodomains 3 and 4 of human polymeric Immunoglobulin receptor (hpIgR) mediate invasion of Streptococcus pneumoniae into the epithelium. J Biol Chem 279, 6296–6304.
Monoclonal antibodies against Streptococcus pneumoniae detect epitopes on eubacterial ribosomal proteins L7/L12 and on streptococcal elongation factor Ts. Microbiology 143, 55–61.
Fontan, P. A., Pancholi, V., Nociari, M. M. & Fischetti, V. A. (2000).
Kolberg, J., Hoiby, E. A., Aase, A., Sletten, K., Rodal, G., Michaelsen, T. E. & Bucher, A. (2000). Streptococcus pneumoniae heat shock
Antibodies to streptococcal surface enolase react with human alphaenolase: implications in poststreptococcal sequelae. J Infect Dis 182, 1712–1721. 1316
protein 70 does not induce human antibody responses during infection. FEMS Immunol Med Microbiol 29, 289–294. Microbiology 152
Pneumococcal enolase and surface expression Kolberg, J., Aase, A., Michaelsen, T. E. & Rodal, G. (2001). Epitope
Pancholi, V. & Chhatwal, G. S. (2003). Housekeeping enzymes as
analyses of pneumococcal surface protein A: a combination of two monoclonal antibodies detects 94% of clinical isolates. FEMS Immunol Med Microbiol 31, 175–180.
virulence factors for pathogens. Int J Med Microbiol 293, 391–401.
Kolberg, J., Aase, A., Rodal, G., Littlejohn, J. E. & Jedrzejas, M. J. (2003). Epitope mapping of pneumococcal surface protein A of strain
Rx1 using monoclonal antibodies and molecular structure modelling. FEMS Immunol Med Microbiol 39, 265–273. Lenz, L. L., Mohammadi, S., Geissler, A. & Portnoy, D. A. (2003).
SecA2-dependent secretion of autolytic enzymes promotes Listeria monocytogenes pathogenesis. Proc Natl Acad Sci U S A 100, 12432–12437. Ling, E., Feldman, G., Portnoi, M., Dagan, R., Overweg, K., Mulholland, F., Chalifa-Caspi, V., Wells, J. & Mizrachi-Nebenzahl, Y. (2004). Glycolytic enzymes associated with the cell surface of
Streptococcus pneumoniae are antigenic in humans and elicit protective immune responses in the mouse. Clin Exp Immunol 138, 290–298. Lopez-Alemany, R., Correc, P., Camoin, L. & Burtin, P. (1994).
Purification of the plasmin receptor from human carcinoma cells and comparison to alpha-enolase. Thromb Res 75, 371–381. Lopez-Alemany, R., Longstaff, C., Hawley, S., Mirshahi, M., Fabregas, P., Jardi, M., Merton, E., Miles, L. A. & Felez, J. (2003).
Inhibition of cell surface mediated plasminogen activation by a monoclonal antibody against alpha-Enolase. Am J Hematol 72, 234–242. Marra, A., Asundi, J., Bartilson, M. & 7 other authors (2002).
Differential fluorescence induction analysis of Streptococcus pneumoniae identifies genes involved in pathogenesis. Infect Immun 70, 1422–1433. Michaelsen, T. E., Aase, A., Kolberg, J., Wedge, E. & Rosenqvist, E. (2001). PorB3 outer membrane protein on Neisseria meningitidis is
poorly accessible for antibody binding on live bacteria. Vaccine 19, 1526–1533. Murphy, P. M. (1993). Molecular mimicry and the generation of host
defense protein diversity. Cell 72, 823–826. Orihuela, C. J., Radin, J. N., Sublett, J. E., Gao, G., Kaushal, D. & Tuomanen, E. I. (2004). Microarray analysis of pneumococcal gene
expression during invasive disease. Infect Immun 72, 5582–5596. Pancholi, V. (2001). Multifunctional alpha-enolase: its role in
diseases. Cell Mol Life Sci 58, 902–920.
http://mic.sgmjournals.org
Pancholi, V. & Fischetti, V. A. (1992). A major surface protein on group A streptococci is a glyceraldehyde-3-phosphate-dehydrogenase with multiple binding activity. J Exp Med 176, 415–426. Pancholi, V. & Fischetti, V. A. (1998). Alpha-enolase, a novel strong
plasmin(ogen) binding protein on the surface of pathogenic streptococci. J Biol Chem 273, 14503–14515. Pearce, B. J., Iannelli, F. & Pozzi, G. (2002). Construction of new
unencapsulated (rough) strains of Streptococcus pneumoniae. Res Microbiol 153, 243–247. Poolman, J. T. (2004). Pneumococcal vaccine development. Expert Rev Vaccines 3, 597–604. Rosenow, C., Ryan, P., Weiser, J. N., Johnson, S., Fontan, P., Ortqvist, A. & Masure, H. R. (1997). Contribution of novel choline-
binding proteins to adherence, colonization and immunogenicity of Streptococcus pneumoniae. Mol Microbiol 25, 819–829. Schaumburg, J., Diekmann, O., Hagendorff, P., Bergmann, S., Rohde, M., Hammerschmidt, S., Jansch, L., Wehland, J. & Karst, U. (2004). The cell wall subproteome of Listeria monocytogenes.
Proteomics 4, 2991–3006. Shaper, M., Hollingshead, S. K., Benjamin, W. H., Jr & Briles, D. E. (2004). PspA protects Streptococcus pneumoniae from killing by
apolactoferrin, and antibody to PspA enhances killing of pneumococci by apolactoferrin. Infect Immun 72, 5031–5040. Tiraby, J. G. & Fox, M. S. (1973). Marker discrimination in transformation and mutation of pneumococcus. Proc Natl Acad Sci U S A 70, 3541–3545. Weiser, J. N. (1998). Phase variation in colony opacity by Strepto-
coccus pneumoniae. Microb Drug Resist 4, 129–135. Whiting, G. C., Evans, J. T., Patel, S. & Gillespie, S. H. (2002).
Purification of native alpha-enolase from Streptococcus pneumoniae that binds plasminogen and is immunogenic. J Med Microbiol 51, 837–843. Zhang, J. R., Mostov, K. E., Lamm, M. E., Nanno, M., Shimida, S., Ohwaki, M. & Tuomanen, E. (2000). The polymeric immunoglobulin
receptor translocates pneumococci across human nasopharyngeal epithelial cells. Cell 102, 827–837.
1317
