Streptococcus agalactiae GAPDH Is a Virulence-Associated Immunomodulatory Protein¹ Free

Infection with Streptococcus agalactiae, Lancefield’s group B streptococci (GBS),⁴ is a leading cause of neonatal pneumonia, sepsis, and meningitis (1, 2). Mortality due to neonatal GBS infection remains high (0.05–0.1%) despite antibiotic therapy, and 25–50% of surviving infants are left with permanent neurological sequelae (2, 3). In addition to the deleterious effects in newborns, GBS is also a frequent cause of infections in pregnant women and adults with underlying diseases (3, 4).

It has been described that extracellular proteins produced by diverse pathogenic microbes facilitate microbial colonization through modulation of the host immune system (5, 6, 7, 8, 9). These proteins were thus designated as virulence-associated immunomodulatory proteins (VIP) (10, 11). This designation reflects their ability to exert a stimulatory effect on polyclonal B cells and to elicit IL-10 and/or IL-4 production in the murine host (6, 11, 12). Lymphocyte polyclonal activation has been considered as a generalized mechanism of immune evasion among pathogens (8, 13, 14). In addition, IL-10 is a well-characterized immunoregulatory/suppressor cytokine (15, 16), and IL-4 was associated with host susceptibility to infections caused by diverse microbes (12, 17, 18, 19, 20).

These effects induced by VIP might contribute to pathogen evasion from the host immune system (5, 12, 21). Previous reports showing that immunization against a particular VIP confers protection against the infection caused by the VIP-producing microbe, further indicates the important role of VIP for the success of microbial colonization (11, 22, 23, 24, 25).

GAPDH is a glycolytic enzyme involved in bacterial energy generation that is essential for growth in the absence of neoglucogenic substrates (26, 27). In several pathogenic bacteria, GAPDH has been described as a protein associated with virulence (28, 29, 30) due to its ability to bind several host proteins (28, 31, 32, 33) or to confer resistance against reactive oxygen species produced by host phagocytic cells (34). In GBS, GADPH is localized in their surface and is highly homologous to the plasmin receptor of group A streptococci (35). Although devoid of signal peptide, streptococcal GAPDH has been described as a secreted protein (36, 37).

In this study, we isolated a B cell stimulatory protein from S. agalactiae culture supernatants that was identified as GAPDH. The gapC gene was expressed in Escherichia coli for the production and subsequent purification of a histidyl-tagged recombinant protein. We report in this study that the recombinant GAPDH (rGAPDH) is a stimulatory protein that induces the proliferation and differentiation of B cells. Moreover, we also show that IL-10, detected at higher concentrations in the serum of rGAPDH-treated mice, is important in determining the host susceptibility to S. agalactiae infection. To our knowledge, this is the first report in which an immunobiological activity was ascribed to a GAPDH.

S. agalactiae NEM316 belongs to capsular serotype III and was isolated from a neonatal blood culture (38). The complete genome sequence of this strain has been determined (39). E. coli DH5α (Invitrogen Life Technologies) was used as a host for plasmid constructions. E. coli HB101 (F⁻, hsd-20 recA13 ara-14 proA2 lacY1 galK2 rpsL20 (Str) xyl-5 mtl-1 supE44) (40) harboring plasmid pRK24 was used as a mobilizing donor to transfer plasmid pAT28 and derivatives to S. agalactiae NEM316 (41). E. coli BL21 (DE3) strain (Novagen) and the pET28a plasmid (Novagen) were used for production of a recombinant protein. S. agalactiae was grown in Todd-Hewitt broth or agar (Difco Laboratories) or RPMI 1640 (Sigma-Aldrich), and E. coli was cultured on Luria-Bertani medium. Bacteria were grown at 37°C unless otherwise stated. Antibiotics were used at the following concentrations: for E. coli, kanamycin, 50 μg/ml and spectinomycin, 60 μg/ml; for S. agalactiae, spectinomycin, 250 μg/ml; and nalidixic acid, 50 μg/ml. Gene expression from pET28 derivatives was induced by adding a final concentration of 1 mM isopropyl-β-d-thiogalactopyranoside (IPTG).

Genomic streptococcal DNA was isolated as described previously (42). DNA manipulations used in construction of plasmids were performed by standard methodology (43). Plasmid DNA was isolated with NucleoSpin Plasmid (Macherey-Nagel) and introduced into E. coli by transformation. PCRs were conducted with AmpliTaqGold DNA polymerase according to the manufacturer’s specifications (Applied Biosystems). Amplification products were purified on Sephadex S-400 columns (Pharmacia) and sequenced with an ABI 310 automated DNA sequencer, using the ABI PRISM dye terminator cycle sequencing kit (Applied Biosystems).

The promoter region of the kanamycin resistance gene aphA-3 and the gapC gene were amplified with the primer pairs as follows: Kan-EcoRI (CCGgaattcCCAGCGAACCATTTGAGG) plus Kan-KpnI (CGGggtaccGATTTTGAAACCACAAT) and GAP-KpnI (GGAAggtaccATAAGGAGGAAATCAC) plus GAP-BamHI (AGAGggatccGGATAAATCCCTCT), respectively (the restriction sites used for cloning are written in lower case). The resulting amplicons were digested with the appropriate enzymes and ligated into the multicopy shuttle vector pAT28 to give pAT28ΩgapC. This plasmid was conjugatively transferred from the mobilizing strain HB101/pRK24 to S. agalactiae NEM316 to generate a strain overexpressing the gapC gene (oeGAPDH).

Male C57BL/6, C7BL/10ScCr, and BALB/c mice (6–8 wk old) were purchased from the Gulbenkian Institute of Science (Oeiras, Portugal). Hen-egg lysozyme-Ig-transgenic (HEL-Ig) mice (44) were purchased from The Jackson Laboratory. C57BL/6 IL-10-deficient (IL-10^−/−) mice (a gift from Prof. M. J. Saraiva, Instituto de Biologia Molecular e Celular, Porto, Portugal) were obtained from B&K Universal Limited Laboratory (Grimston Alborough Hull, U.K.) and bred at the animal facilities of the Institute for Molecular and Cell Biology (Porto, Portugal). Animals were kept at the animal facilities of the Institute Abel Salazar during the time of the experiments. Animal experimentations were performed according to the European Convention for the Protection of Vertebrate Animals used for Experimental and Other Scientific Purposes (ETS 123) and 86/609/EEC Directive and Portuguese rules (DL 129/92).

S. agalactiae was disrupted by freeze-thawing twice followed by sonication (10 cycles of 30 s at 100 W) with a Branson cell disrupter; model W 185 D, on ice. The S. agalactiae sonicates were successively filtered through 0.45- and 0.2-μm pore size filters (Schleicher & Schuell Microscience) and stored in small aliquots at −80°C. Full disruption of S. agalactiae cells was checked by light microscopy.

S. agalactiae was precultured in RPMI 1640 medium (Sigma-Aldrich) overnight and subsequently cultured during 48 h in the same medium. Culture supernatants were filtered through a 0.22-μm pore size filter (Schleicher & Schuell Microscience) and concentrated by vacuum dialysis in a Visking 100/8FT dialysis membrane with a 30,000 kDa cut-off (Cole-Parmer’s Masterflex) for the collection of the extracellular proteins (EP-Sa). The absence of detectable cytosolic contaminants in EP-Sa was assessed by measuring the activity of the cytosolic isocitrate dehydrogenase using the Diagnostics Isocitrate Dehydrogenase kit (Sigma-Aldrich). The EP-Sa were then fractionated by preparative polyacrylamide native gel electrophoresis (Model 491 Prep Cell; Bio-Rad) and the fractions eluted into PBS concentrated by vacuum dialysis. All fractions were passed through a polymixin B column (Sigma-Aldrich) to remove contaminant endotoxin, and only endotoxin-free fractions, as assessed by the limulus test (E-toxate; Sigma-Aldrich), were used. Protein content of the different samples was determined by the method of Lowry (45).

The gapC gene (gbs1811; http://genolist.pasteur.fr/SagaList/) was PCR amplified in its entirety from S. agalactiae chromosomal DNA by using the primers GAP-NcoI (CCccatggTAGTTAAAGTTGG) and GAP-XhoI (CCCctcgagTTTTGCAATTTTTGC) (the restriction sites used for cloning are written in lower case). The NcoI site of the forward primer included the ATG translational start site of gapC, whereas the XhoI site of the reverse primer was used to remove the stop codon. This 1021-bp long DNA fragment was digested with NcoI and XhoI and cloned into pET28a linearized with the same enzymes to produce a rGAPDH containing a carboxylic histidyl tag. E. coli BL21(DE3) cells were transformed with the resulting recombinant plasmid (pET28aΩgapC). Following a 3-h IPTG-induced expression of the fusion protein, the cells were harvested by centrifugation and resuspended in phosphate buffer containing 10 mM imidazole. The sample was incubated on ice for 30 min in the presence of 100 μg/ml lysozyme and 10% Triton X-100. After sonication, the insoluble material was removed by centrifugation and the supernatant was filtered through a 0.45-μM pore size filter (Millipore) and applied to a His-trap column (Amersham Biosciences). The rGAPDH was eluted with imidazole under native conditions and the eluant concentrated by vacuum dialysis and equilibrated in PBS buffer before endotoxin removal on a polymixin B column as described above. E. coli BL21 (DE3) harboring the wild-type (WT) pET28 vector (i.e., an E. coli strain that does not express the GBS rGAPDH) was similarly processed and the resulting preparation, designated neControl, was used as a negative control. Heat-inactivation of rGAPDH was performed by incubating the protein preparations in a boiling water bath for 5 min.

The rGAPDH was assayed for GAPDH enzymatic activity by spectrophotometric assessment (A₃₄₀) of NADH formation measured at 30 s intervals for 10 min. The reaction mixture contained K₂HPO₄ (100 mM; pH 7.4), fructose 1,6-bisphosphate (F1,6bisP; 200 mM), EDTA (500 mM), NAD⁺ (50 mM), aldolase (30 mM), and purified rGAPDH (0.2 mg/ml) to initiate the reaction. Negative controls were performed as described above without the addition of the rGAPDH. The specific activity (unit = U) was calculated according to the following equation: U/mg of protein = (ΔAbs/Δt × V assay)/(ε_NADH × V_enzyme × |Enzyme|): where Δt = 0.5 min; volume of assay = 1039 μl; molar extinction coefficient of NADH at 340 nm ε = 6.22 × 10⁻³ M⁻¹ cm⁻¹; volume of enzyme (rGAPDH) = 166 μl; and |Enzyme (rGAPDH)| = 1.2 mg/ml.

Kinetics of rGAPDH enzymatic activity (200 μg/ml) were determined with various concentrations of F1,6bisP and with a fixed concentration of NAD⁺, and vice versa, to determine the K_M and V_máx for F1,6bisP and NAD⁺, respectively. The results were recorded as the rate analysis of NADH release at 0.5-min intervals for a period of 10 min at 340 nm. The molar extinction coefficient of NADH (6.22 × 10⁻³ M⁻¹ cm⁻¹) was used to convert absorbance (340 nm)/minute to M NADH/minute. K_M and V_máx were determined from the Lineweaver-Burk double reciprocal plots.

Spleen cells were obtained by gently teasing the organ in RPMI 1640 supplemented with penicillin (100 IU/ml), streptomycin (50 μg/ml), 2-ME (0.05 M), and 10% FBS (Sigma-Aldrich). Cells were then distributed in 96-well plates (1 × 10⁶ cells/well) and cultured for 6 h at 37°C in a humidified atmosphere containing 5% CO₂, in the medium alone, medium containing 2.5 μg/ml LPS (positive control), or medium containing 25 μg/ml of the tested proteins. For the titration experiments, different concentrations of rGAPDH were used. The cultured cells were washed and resuspended in PBS supplemented with 1% BSA and sodium azide to a 10 mM concentration. Immunofluorescence cytometric analysis of the surface Ags of these cells was performed in a FACScan with CellQuest software (BD Biosciences), using the following mAbs: goat anti-mouse FITC-conjugated IgM (Southern Biotechnology Associates), PE-conjugated hamster anti-mouse CD3 (BD Pharmingen), FITC-conjugated rat anti-mouse B220 (Ly5) (Pharmingen), and PE-conjugated hamster anti-mouse CD69 (BD Pharmingen). Dead cells were gated out using propidium iodide incorporation.

For the isolation of mononuclear cells, 5 ml of aliquots of the spleen cell suspensions were layered on 2.5 ml of Histopaque (Sigma-Aldrich) and centrifuged at 1000 × g for 20 min at room temperature. The cells were then gently removed from the medium-Histopaque interface, transferred to a sterile container, and washed in 10 ml of RPMI 1640 (Sigma-Aldrich). The isolated mononuclear cells were resuspended in 5 ml of RPMI 1640.

The MTT assay was performed as previously described (46) with some modifications. Briefly, mononuclear cells (5 × 10⁵/well) were plated in triplicate in round-bottom 96-well tissue culture plates and stimulated with 2.5 μg/ml LPS, with 2.5 μg/ml Con A (positive controls), or with 25 μg/ml rGAPDH, and incubated for 48 h. Then, a MTT stock solution (0.5 mg of MTT per milliliter of phenol red-free cell culture medium) equivalent to 1/10 of the original culture volume was added to each culture being assayed. After incubation for 4 h, the supernatants were decanted. An amount of 0.08 M HCl solution in anhydrous isopropanol equal to the original culture volume was added to each well to dissolve the resulting MTT formazan crystals. The absorbance of the purple solution at 570 nm was measured spectrophotometrically in a Multiskan plate reader (Multiskan EX; Labsystems).

Mice were injected i.p. with 50 μg of rGAPDH in 0.5 ml of PBS or with 0.5 ml of PBS alone in control animals.

Mice were injected i.p. twice, with a 3-wk intervening period, with 200 μl of a preparation containing 20 μg of rGAPDH in a 1/1 PBS/alum suspension (aluminum hydroxide gel; Brenntag; a gift from Dr. E. Lindblad, Biosector, Frederickssund, Denmark). The sham-immunized control animals received 200 μl of a 1/1 PBS/alum suspension. Sera of GAPDH-immunized mice and sham-immunized controls were collected 30 days after the second i.p. injection and pooled for Western blot analysis of GAPDH expression.

For neutralization of GAPDH, mice were immunized with rGAPDH as described above and pooled sera containing anti-rGAPDH IgG Abs were used in in vitro and in vivo assays. The Ab titer of the serum, 37,100 ELISA Titres, was determined by ELISA as described in Ref. 47 . In spleen mononuclear cell cultures, 5 μl of antiserum was added per well, whereas, for in vivo assays, 20 μl of antiserum was injected i.p. immediately before the injection of the oeGAPDH S. agalactiae strain. Control experiments were similarly performed using pooled sera from sham-immunized animals that do not contain detectable anti-rGAPDH Abs.

The numbers of splenic Ig-secreting cells were assessed by an ELISA-spot assay as described (48). Briefly, polystyrene microtiter plates (Nunc) were coated overnight at 4°C with 5 μg/ml goat anti-mouse Ig (Southern Biotechnology Associates), with 10 μg/ml S. agalactiae sonicates prepared as described above, or with 10 μg/ml rGAPDH. The wells were then saturated for 30 min with 2% BSA in PBS at 37°C. Appropriate serial suspensions of spleen cells in RPMI 1640 supplemented with 3% calf serum (Invitrogen Life Technologies/Biocult) were incubated in the plates for 6 h at 37°C in a humidified atmosphere of 5% CO₂ in air. The plates were then rinsed with 0.075% Tween 20 in water (Sigma-Aldrich) and washed four times with PBS containing 0.075% Tween 20. Ag-specific or total Ab-secreting cells were revealed by the addition of alkaline phosphatase-coupled monoclonal goat anti-mouse-Ig (Southern Biotechnology Associates) overnight at 4°C. After washing, 5-bromo 4-chloro 3-indolyl phosphate (Sigma-Aldrich) in 2-amino 2-methyl 1-propanol (Sigma-Aldrich) buffer was used as substrate for 2 h at 37°C. After washing four times with distilled water, the number of spots was quantified in triplicate wells with a dissecting microscope.

Mice sera were collected by retro-orbital bleeding, and IL-10 titers were quantified with the Quantikine M Murine IL-10 ELISA kit (R&D Systems) according to the manufacturer’s instructions.

The presence of GAPDH in the EP-Sa was visualized by Western blot analysis. For that purpose, extracellular proteins or rGAPDH contained on 10% SDS gel were blotted onto a nitrocellulose membrane (Amersham Biosciences). Transferred proteins were stained with Ponceau S. After destaining with water, the membranes were cut into strips, according to protein migration lanes and saturated in TBST buffer (0.01 M Tris, 0.15 M NaCl, 0.05% Tween 20 (pH 8.0)), containing 1% BSA, for 1 h and further incubated for 4 h with the individual mice sera diluted in TBST. GAPDH-bound Abs were detected with alkaline-phosphatase-coupled monoclonal goat anti-mouse IgG1 (Southern Biotechnology Associates), using a solution of 0.33 mg/ml NBT dye (Promega) and 0.17 mg/ml 5-bromo-4-chloroindol-2-yl phosphate (Promega), in AP buffer (0.01 M Tris, 0.01 M NaCl, 0.5 mM MgCl₂ (pH 9.5)), as substrate.

C57BL/6, BALB/c, and C57BL/6 IL-10^−/− mice were infected i.p. with 0.5 ml of PBS containing 1 × 10⁶ cells of S. agalactiae (WT or oeGAPDH strains). Five days after the infection, the liver was aseptically removed, homogenized in PBS, and serially diluted (1/10 dilutions). Bacteria were plated onto a Todd-Hewitt agar (Difco Laboratories) plate containing 0.001 mg/ml colistin sulfate and 0.5 μg/ml oxalinic acid (streptococcus selective supplement; Oxoid), and GBS CFUs were enumerated in duplicates after 48 h of incubation at 37°C.

The level of significance of the results in all groups of mice was determined by one-way ANOVA, calculated with Microsoft Excel 2000 software. Differences were considered significant at p < 0.05.

Based on the knowledge that extracellular proteins of diverse streptococci contain immunomodulatory proteins associated with virulence, we investigated the presence of VIP(s) in S. agalactiae extracellular products. For this purpose, supernatants of bacterial cultures collected during the exponential phase of growth were fractionated by preparative polyacrylamide native gel electrophoresis. Several fractions were eluted, each corresponding to the distinct protein bands shown in Fig. 1,A. The different protein fractions were then assessed for stimulatory activity on splenic T and B cells by quantification of CD69 expression on the surface of the stimulated lymphocytes. As shown in Fig. 1,B, a marked B cell stimulatory effect was observed on cultures stimulated with fraction F5. In this case, expression of the early activation marker CD69 was detected on the surface of 70.8 ± 2.3% of B cells, whereas the expression of this early activation marker was detected only on 16.7 ± 1.5% of medium-stimulated B cells. As also shown on Fig. 1,B, a slight lymphocyte stimulatory effect was also induced by some of the other fractions, albeit less intensely as compared with that observed on cultures treated with fraction F5. By SDS-PAGE analysis, fraction F5 corresponded to a single protein band with a molecular mass of 45 kDa (Fig. 2, lane 1). The protein of this fraction was unambiguously identified as a GAPDH (Gbs1811) by N-terminal amino acid sequencing (VVKVGI).

E. coli BL21 (DE3) harboring pET28ΩgapC was used to produce high levels of a rGAPDH containing a COOH-terminal histidyl tag that was purified on a nickel-chelating column. The estimated size of the purified rGAPDH protein, estimated by SDS-PAGE, was 45 KDa (Fig. 2, lane 2). GAPDH catalyzes the oxidative phosphorylation of d-glyceraldehyde-3-phosphate to form 1,3-diphosphoglycerate in the presence of NAD⁺ and inorganic phosphate. The specific activity of the rGAPDH was 39.4 U/mg of protein confirming the enzymatic functionality of the recombinant protein. Furthermore, a variation of the enzyme reaction rate with different concentrations of F1,6bisP and NAD was measured using 200 μg/ml the rGAPDH. The results were analyzed both as Michaelis-Menten and reciprocal plots according to Lineweaver and Burk (49). From these plots, the K_M for F1,6bisP and NAD⁺ was estimated to be 23.8 mM and 32.6 mM, respectively, and V_max was 12.5 M NADH/min and 10.2 M NADH/min, respectively. The maximum velocity, V_max, is the velocity obtained under conditions of substrate saturation of the enzyme under specified conditions of pH, temperature, and ionic strength (it is a constant for a given concentration of enzyme). The results described above showed a slight variation, albeit not significant, of V_max.

Given that the gene gapC is essential for bacterial growth and cannot be deleted when glucose is the only carbon source, we constructed a S. agalactiae strain overexpressing GADPH (oeGAPDH) to study the contribution of this enzyme to GBS virulence. The overexpression of GAPDH in the mutant strain was confirmed by Western blot analysis (Fig. 3). The role of GAPDH on the virulence of S. agalactiae was assessed in six groups of six C57BL/6 mice challenged with the WT strain, the oeGAPDH mutant, the oeGAPDH mutant plus control serum or anti-rGAPDH antiserum, and the WT strain 2 days after i.p. injection of either PBS or rGAPDH (Fig. 4,A). The numbers of S. agalactiae CFU were assessed in the liver of each group of infected mice 5 days after the bacterial challenge. As shown in Fig. 4,A, bacteria were detected in the liver of mice infected with the oeGAPDH strain but not in the liver of control mice challenged with the WT strain alone. Moreover, bacterial host colonization was abrogated in mice that have received anti-rGAPDH antiserum at the time of infection (Fig. 4,A). This result suggests that GAPDH contributes to host colonization by S. agalactiae. Consistently, we observed a significant increase of S. agalactiae CFU in the liver of mice that were treated with rGAPDH 2 days before the bacterial challenge with the WT strain. To rule out the possibility that the increased virulence of the oeGAPDH mutant could be due to a faster growth rate in vivo, S. agalactiae-susceptible BALB/c mice were infected i.p. with 1 × 10⁶ of either WT or oeGAPDH bacteria. As shown in Fig. 4 B, the numbers of liver S. agalactiae CFU of either strain were not significantly different up to 5 days postinfection. Moreover, the in vitro growth curves of WT and oeGAPDH S. agalactiae strains were indistinguishable (data not shown).

To investigate the lymphocyte stimulatory activity of rGAPDH, C57BL/6 mice spleen mononuclear cells were in vitro stimulated with the rGAPDH, or with control preparations obtained from E. coli cells that do not express GBS rGAPDH, and the expression of surface CD69 was assessed on B and T cells 6 h after the stimulatory treatment. To exclude the possibility that contaminating endotoxins could be the source of the observed biological effects, protein and control preparations were run through a polymixin B column and the absence of endotoxins was confirmed with the highly sensitive limulus test. As shown in Fig. 5, an up-regulation of CD69 expression was observed in lymphocyte cells stimulated with rGAPDH and this effect was more pronounced on B than on T cells. The dose-response curve shown in Fig. 5,A indicated an optimal concentration of 25–50 μg/ml rGAPDH for the activation of B cells in this assay (Fig. 5,A). A slight increase in CD69 expression was also detected on the surface of T (CD3⁺) cells in cultures stimulated with 12 μg/ml rGAPDH (Fig. 5,A). This in vitro lymphocyte stimulatory effect was markedly reduced in the presence of anti-rGAPDH antiserum (Fig. 5,B). In addition, no CD69 up-regulation was observed when heat-inactivated-rGAPDH or control preparations from rGAPDH nonexpressing E. coli cells were used in this assay (Fig. 5 B). Taken together, these results demonstrated that the observed lymphocyte stimulatory effect of rGAPDH was not due to contaminants.

To investigate whether the B cell activation induced by rGAPDH is followed by cell proliferation, C57BL/6 mice spleen mononuclear cells were stimulated in vitro with this recombinant protein for 48 h, and the proliferative effect was assessed with the “MTT proliferation assay.” As shown in Fig. 6,A, rGAPDH induced cellular proliferation as observed 48 h after stimulation. This in vitro proliferative effect of rGAPDH was also observed when spleen mononuclear cells obtained from LPS-nonresponsive C57BL/10 ScCr mice (50) were similarly treated (Fig. 6,B). To determine the effect of rGAPDH on B cell differentiation, C57BL/6 mice were treated with 50 μg of rGAPDH i.p. and 5 days later the number of splenic Ig-secreting cells was evaluated in an ELISA-spot assay. An increase in the number of Ig-secreting cells was observed in the spleen of mice treated with rGAPDH (Fig. 7,A). This B cell differentiation was not due to a rGAPDH-specific response because the number of Ig-secreting cells specific for this protein did not differ from that observed in PBS injected controls (Fig. 7,B). Therefore, to verify the BCR-independent effect of rGAPDH, spleen mononuclear cells from BCR-transgenic, HEL-Ig, mice were stimulated in vitro with 25 μg/ml rGAPDH. An up-regulation of CD69 was observed 6 h after rGAPDH stimulation on the surface of B lymphocytes and, albeit less markedly, on T cells (Fig. 8). This result further indicates that B cell activation is independent of BCR signaling.

Taking into account the observed in vitro lymphocyte stimulatory effect of rGAPDH, the surface expression of the early activation marker CD69 was assessed in vivo on splenic B and T lymphocytes of C57BL/6 mice infected i.p. with 1 × 10⁶ WT or oeGAPDH S. agalactiae cells. As shown in Fig. 9, a higher expression of CD69 was observed in the surface of B and T cells of mice infected with the oeGAPDH than with the WT strain. This stimulatory effect was more marked on the B than on the T cell population, an observation in perfect agreement with the results obtained in vitro with rGAPDH-stimulated spleen mononuclear cell cultures.

The release of IL-10 into the serum has been previously observed in mice treated with various microbial extracellular proteins (6, 9, 10). Therefore, to investigate whether rGAPDH could induce a similar effect, C57BL/6 mice were injected i.p. with 50 μg of rGAPDH and sera were collected at different times after this treatment. As shown in Table I, rGAPDH induced in C57BL/6 mice an increase in the levels of IL-10, detectable as early as 6 h, in the serum of treated mice. The serum levels of IL-10 declined rapidly and were detected slightly above control levels 12 h after rGAPDH injection but not 24 h after treatment (Table I). An early rise of IL-10 concentration was also detected in the serum of C57BL/6 mice infected with the oeGAPDH S. agalactiae strain, whereas no detectable IL-10 serum levels were found in mice infected with the WT strain (Table I). To further confirm that IL-10 could facilitate the host colonization by S. agalactiae, IL-10^−/− and WT C57BL/6 mice were infected with 1 × 10⁶ oeGAPDH S. agalactiae cells and the numbers of bacterial CFU was evaluated in their liver 5 days after the infectious challenge. As shown in Fig. 10, no bacteria was detected in the liver of IL-10^−/− mice, whereas efficient colonization of the liver of WT mice was observed.

We report in this study the purification of a 45-kDa immunomodulatory protein from the extracellular products of S. agalactiae. This protein was sequenced and identified as a GAPDH. The gene coding for this protein was cloned and the recombinant protein was produced in E. coli cells. The rGAPDH was endowed with an enzymatic activity of 34.9 U/mg protein, similar to that found in the GAPDH isolated from group A streptococci (33). The K_M values obtained were higher (23.8 mM and 32.6 mM, respectively, for F1,6bisP and NAD⁺) than the values reported for GAPDH of group A streptococci (33) (1.33 mM for G-3-P and 156.7 μM for NAD⁺), indicating a weaker affinity of the recombinant GBS enzyme for its substrates. In addition to its presence in the cytosol, GAPDH has also been described as a surface (28, 31) or a secreted (36) protein. Consistently, we demonstrated in this study by Western blot and sequencing analysis that NEM316 GAPDH was present in culture supernatants devoid of isocitrate dehydrogenase activity, a cytosolic protein used as a marker of bacterial lysis.

In this study, we report that rGAPDH induces B cell and T cell activation, as assessed by determining CD69 expression on the surface of spleen lymphocyte cells stimulated in vitro with this recombinant protein. The rGAPDH-induced lymphocyte activation was more marked on B than on T cells. An in vitro increase in the number of lymphocyte cells expressing surface CD69 was also observed upon stimulation of these cells with GAPDH isolated from S. agalactiae culture supernatants. Again, this lymphocyte activating stimulus was more marked on the B cell population. The lymphocyte activation induced by rGAPDH leads in vivo to the differentiation of murine B cells into Ig-secreting cells as revealed by an increase in the number of these cells detected in mice treated with this protein. B cell activation is a feature often associated with infection caused by diverse microbes including virus, bacteria, fungi, or protozoa (14). Different structural microbial Ags such as LPS or the staphylococcal protein A are known to induce polyclonal B cell activation. In addition, several microbial extracellular proteins including enzymes have also been described as B cell activators (8, 13, 25, 51, 52). The ability of microbial constituents to induce polyclonal B cell activation in the infected host constitutes a mechanism of host immune evasion used by pathogens to suppress the specific, potentially protective, immune responses (5, 14, 21). Therefore, GAPDH-induced B cell activation may represent a mechanism facilitating S. agalactiae survival within the host. A role of GAPDH in promoting GBS growth in the murine host is suggested by our results showing that C57BL/6 mice were more susceptible to a S. agalactiae strain that overexpresses GAPDH as compared with the WT strain. Moreover, the abrogation of S. agalactiae liver colonization observed in anti-GAPDH antiserum-treated C57BL/6 mice infected with the oeGAPDH strain confirms the direct role of this enzyme in facilitating infection. This role was also highlighted by the observation that treatment of C57BL/6 mice with rGAPDH 2 days before inoculation of the WT S. agalactiae strain increased bacterial liver colonization as compared with untreated infected controls. Although not specifically assessed in this study, the polyclonal nature of the B cell activation induced by GAPDH is suggested by the stimulatory effect of this protein on BCR-transgenic B cells. Moreover, the Abs produced by splenic Ig-secreting cells of rGAPDH-treated mice did not reacted with this protein in an ELISA-spot assay.

A rapid increase in the serum levels of IL-10 was detected in C57BL/6 mice treated i.p. with rGAPDH. A similar effect was previously described in mice treated with other microbial proteins that facilitate microbial colonization and therefore generally designated as VIP (6, 7, 9, 10). Our observation that IL-10-deficient mice are more resistant to S. agalactiae infection shows that this cytokine, a major down-regulator of inflammatory responses (16), has an important role in successful host colonization by S. agalactiae. This is further confirmed by our other observation showing that the oeGAPDH S. agalactiae strain that successfully colonize GBS-resistant C57BL/6 mice, induces in this host an increase in serum IL-10 concentration, whereas the WT strain does not. An immunosuppressive effect mediated by IL-10 production has been ascribed to microbial extracellular proteins produced by Bordetella pertussis (53) and Yersinia pestis (54). The latter was reported to occur in a TLR2/CD14-dependent manner (54). It could thus be hypothesized that a similar immunosuppressive mechanism, mediated by IL-10, could be involved in successful host colonization by S. agalactiae. In contrast, whether S. agalactiae GAPDH immune effects could result from an engagement of pattern recognition receptors could be worth investigate. Another question that remains to be elucidated concern the cell source of IL-10 found at elevated concentrations in the serum of mice early after treatment with rGAPDH or after infection with the oeGAPDH strain. The rapidity of this rise in serum IL-10 suggests that innate immune cells such as macrophages could be involved in this phenomenon. However, it could also be due to the stimulation of other cells, such as the so-called “innate-like” lymphocytes (55), because a similar rapid production and release of IL-10 by NK T or B-1 cells was previously reported (56, 57, 58, 59, 60). Moreover, considering the early expression of CD69 on B-cells after in vivo and in vitro rGAPDH stimulation, it would be interesting to determine whether this lymphocyte population could also be involved in this early rise in serum IL-10.

The reported effects of GAPDH and the possible involvement of this protein in the mechanisms leading to host susceptibility to S. agalactiae makes this protein a suitable candidate to be used as target Ag in immunoprotective approaches against this bacterium (Portuguese patent pending no. 103450).

We are indebted to Dr. Rui Appelberg for critically reviewing this manuscript.

The authors have no financial conflict of interest.