Use of the serogroup B meningococcal capsular polysaccharide (MenB CP) as a vaccine is hampered by the presence of epitopes that cross-react with human polysialic acid. As non-cross-reactive, protective capsular epitopes have also been described, we set out to develop protein mimics of one of such epitopes using as a template a highly protective mAb (mAb Seam 3) raised against a chemically modified form of the MenB CP (N-Pr MenB CP). Using phage display, anti-idiotypic single-chain Ab fragments (scFvs) were obtained from spleen cells of mice immunized with the Seam 3 mAb. Two Seam 3-specific scFvs competed with N-Pr MenB CP for binding to either mAb Seam 3 or rabbit Abs present in typing sera. Moreover, in mice and rabbits the scFvs elicited the production of Abs reacting with both N-Pr MenB CP and whole meningococci, but not with human polysialic acid. These scFv-induced Ab responses were boostable and of the Th1 type, as shown by a predominance of IgG2a. In addition, passive immunization with sera from scFv-immunized animals partially protected neonatal mice from experimental infection with group B meningococci. In conclusion, we have produced anti-idiotypic scFvs that mimic a protective MenB CP epitope and may be useful in the development of an alternative group B meningococcal vaccine.

Neisseria meningitidis is one of the most common causes of bacterial meningitis and septic shock in infants and young adults. Mortality rates from meningococcal meningitis range from 5–10%, and neurological sequelae are present in up to 25% of the survivors (1, 2). The main virulence factor of these organisms is the capsular polysaccharide (CP),2 which protects against complement-mediated bacteriolysis and phagocytosis (3). Based on chemically and antigenically distinctive capsular properties, N. meningitidis can be divided into 12 serogroups, with serogroups A, B, C, Y, and W-135 accounting for virtually all disease-producing isolates (4). CP-protein conjugate vaccines can induce effective Ab responses and therefore hold great promise for the control of infections caused by groups A, C, W-135, and Y (5).

In contrast, group B N. meningitidis (MenB), which is responsible for up to 80% of the total cases of disease in industrialized countries, has been resistant to any attempt to develop a capsule-based vaccine due to the poor immunogenicity of the CP even after protein conjugation. This is probably due to the similarity between the MenB CP and a polymer expressed on the neuronal cell adhesion molecule of mammalian tissues (6). Both polysaccharides are α(2→8) homolinear polymers of N-acetylneuraminic acid (polysialic acid), although the bacterial CP is longer than human polysialic acid (∼200 vs 50 residues, respectively). Therefore, the MenB CP is a self-Ag unable to stimulate an immune response and with the potential to induce autoimmunity if used as a vaccine.

A chemically modified MenB CP, obtained by removal of N-acetyl (N-Ac) and substitution with N-propionyl (N-Pr) groups, can elicit, after conjugation to a protein, bactericidal Abs that passively protect animals from experimental MenB infection (7). It is believed that the N-Pr CP somehow mimics antigenic determinants that are present on the bacterial surface, but are infrequently expressed on the purified N-Ac CP. The N-Pr CP, however, also elicits a small subset of autoreactive IgGs. Although the concentration and/or avidity of the latter are generally low (8), it may be difficult to prove that such autoantibodies are absolutely innocuous. By the use of mAbs raised against either the N-Pr or the N-Ac CP, at least two different classes of capsular epitopes have been defined, both of which can be found on the intact bacterial surface. One class, expressed mainly on N-Ac polysialic acid, is cross-reactive with human tissue, whereas the other, which predominates in N-Pr polysialic acid, is protective and non-cross-reacting (9, 10, 11).

It is well established that protein molecules, such as peptides and anti-idiotypic Abs, can mimic functional features of carbohydrate Ags, including their ability to induce the production of specific Abs (reviewed in Ref. 12). The ability of murine monoclonal anti-idiotypic Abs to function as surrogate vaccines against several different encapsulated pathogens has been shown in the past (13, 14, 15). More recently, it was shown that anti-idiotypic single-chain Ab fragments (scFv) are sufficient to induce protective, capsule-specific Ab responses (16). These recombinant fragments are attractive, because they can be easily manipulated for different purposes, including DNA vaccination and/or fusion with cytokines or immunogenic peptides. In the present study, using an appropriate mAb as a template, we generated scFv mimics of a protective, non-cross-reactive epitope of the MenB CP. These constructs were able to induce anti-meningococcal Ab responses in animals and therefore may be useful in the development of a safe MenB vaccine.

Three encapsulated group B N. meningitidis strains (2996, MC58, and 8047) and the unencapsulated M7 strain (17) were provided by M. Giuliani (Chiron, Siena, Italy). Purified N-Ac MenB, N-Pr MenB, and biotin-conjugated N-Pr MenB CPs, prepared as previously described (9, 18), were provided by A. Bartoloni (Chiron). Vaccines consisting of N-Ac MenB and N-Pr MenB CPs conjugated to tetanus toxoid (N-Ac MenB-TT and N-Pr MenB-TT, respectively) were prepared as previously described (9). Briefly, after partial acid hydrolysis of the polysaccharides, intermediate size (9–25 repeating units) oligosaccharides were separated by ion exchange chromatography, oxidized, and conjugated with TT (obtained from Calbiochem, La Jolla, CA). After purification by gel filtration, saccharide to protein ratios, determined as previously described (9), were 0.19 and 0.22 for N-Pr MenB-TT and N-Ac MenB-TT, respectively. Purified murine Seam 3 mAb was provided by M. Mariani (Chiron). The Seam 3 mAb is an IgG2b raised against N-Pr MenB that strongly reacts against encapsulated MenB to produce complement-mediated bacteriolysis and has no autoantibody reactivity (9). Rabbit MenB-specific typing serum was purchased from Difco (distributed by Voigt Global Distribution, Kansas City, MO).

Preparation of Ab fragments was accomplished as previously described (16, 19), with minor modifications, using libraries of filamentous phages expressing murine scFvs on their surface. Briefly, female BALB/c mice (8 wk old) were s.c. injected on days 0 and 15 with 50 μg of mAb Seam 3 in 0.2 ml of CFA (Difco) and on days 21 and 28 with the same Ag dose in 0.2 ml of IFA (Difco). A final booster injection (50 μg of Ag in saline) was given i.p. on day 35, and 3 days later spleen cells were obtained. After incubating these cells on Seam 3-coated plates, RNA was extracted from adherent cells. After reverse transcribing purified mRNA with random hexadeoxyribonucleotides primers, a commercial system (Recombinant Phage Ab System; Amersham Pharmacia Biotech, Milan, Italy) was used to clone and express Ab fragments. Recombinant phages displaying scFv Abs on their tips were selected after absorption of the library with an irrelevant IgG2b mAb (mouse IgG2b MOPC 141; Sigma-Aldrich, Milan, Italy) as previously described (16, 19) by three rounds of panning on culture flasks that had been previously coated with mAb Seam 3. Screening of isolated colonies was performed by phage ELISA as previously described (16, 19). Seam 3-specific scFvs and the irrelevant H6 scFv (19, 20) were produced and purified as previously described (16, 19). Purity was checked using Coomassie staining of overloaded SDS-polyacrylamide gels as previously described (16). Under these conditions, only a single band with the expected molecular mass of 30 kDa was observed.

To ascertain whether different clones from the same library shared a common ancestry, restriction analysis was performed on amplified DNA fragments. To amplify a region containing the scFv genes, 25 μl of a master PCR mix was prepared with the M13 sense primer 5′-GGTAACGCCAGGGTTTTCC-3′ and the antisense M13 primer 5′-GGAATTGTGAGCGGATAAC-3′. A HotStart Taq DNA polymerase (0.25 U/tube; Qiagen, Milan, Italy) was used to perform the PCR assay, using isolated colonies directly transferred into PCR tubes. After a first cycle at 95°C for 15 min and 30 cycles at the annealing temperature of 50°C, the PCR products were digested for 3 h at 60°C with the frequently cutting restriction enzyme BstNI (5 U/PCR tube). The digested DNAs were then separated on a 3% agarose gel, and the fragment pattern was recorded.

The DNA encoding for the G1 and B5 scFvs (see Results) was sequenced using the automated Alfexpress DNA Sequencer (Amersham Pharmacia Biotech). These sequences have been deposited with GenBank (accession no. AY374127 and AY374128).

To study the ability of anti-idiotypic scFvs to compete with N-Pr MenB CP for binding to MenB-specific Abs, an inhibition ELISA was used. Briefly, wells of microtiter plates (Nunc, distributed by M. Brunelli, Milan, Italy) were coated overnight at 4°C with hen’s egg avidin (ExtrAvidin; Sigma-Aldrich) at a concentration of 4 μg/ml in PBS (0.01 M phosphate and 0.15 M NaCl, pH 7.2). After washing, the wells were incubated for 2 h at 37°C with biotinylated N-Pr MenB CP (0.5 μg/ml) in PBS containing 1% BSA (PBS-BSA; Sigma-Aldrich). After washing with PBS containing 0.05% polyoxyethylene-sorbitan monolaurate, the plates were blocked with 1% gelatin in PBS for 3 h and fixed with a 10% sucrose and 4% polyvinyl-pyrrolidone solution. Test samples, consisting of mixtures of MenB-specific Abs and putative inhibitors, were added, and the plates were incubated for 2 h at 37°C. Positive and negative controls consisted of soluble N-Pr MenB and the group-specific polysaccharide of group B streptococci, obtained as previously described (21). After washing, bound Abs were detected by HRP-conjugated anti-mouse or anti-rabbit IgG Abs (Amersham Pharmacia Biotech), followed by 2% (w/v) 2,2′-azino-bis (3-ethylbenz-thiazoline-6-sulfonic acid) in 0.05 M citric acid. Reading was performed at 405 nm on a microplate reader. Percent inhibition was calculated by comparing the absorbance values of wells with and without the inhibitors.

Female BALB/c mice (6 wk old) or New Zealand rabbits were used for immunization with scFvs. In the case of mice, a first s.c. injection of 50 μg in 0.2 ml of CFA was followed, on days 21 and 42, by two additional s.c. injections of the same Ag dose in 0.2 ml of IFA. Rabbits underwent a similar immunization regimen, except that they received 500 μg of Ag for each immunization at multiple s.c. sites. Sera were obtained from the retro-orbital plexus (mice) and the saphenous vein (rabbits) at the indicated times and were tested as described below.

Control mice were immunized three times s.c. with N-Pr MenB-TT or N-Ac MenB-TT (4 μg of sialic acid/injection) in CFA (day 0) or IFA (days 21 and 42). In additional experiments mice were immunized with the G1 scFv conjugated with keyhole limpet hemocyanin (KLH; Pierce, distributed by Celbio, Milan, Italy). The scFv (5 mg) was mixed with KLH (20 mg) in phosphate buffer (0.1 M, pH 6.8) containing 0.5% glutaraldehyde (Sigma-Aldrich). The reaction was stopped after 30 min by the addition of 1 M glycine (Sigma-Aldrich), followed by extensive dialysis using 50,000 kDa cutoff membranes. The scFv-KLH conjugate was adsorbed to Al(OH)3 gel (alum; Sigma-Aldrich). A group of five mice was immunized s.c. on days 0 and 21 with 0.1 ml of the alum-adsorbed conjugate corresponding to 50 μg of the G1 scFv. Another group of five mice received 50 μg of the G1 scFv in CFA on day 0 and the same Ag dose in IFA on day 21.

Abs in sera from scFv-immunized animals were assayed for binding to N-Pr MenB CP or to whole bacteria using ELISA. Microtiter plates were coated with avidin, followed by biotinylated N-Pr MenB, as described above (see Inhibition ELISA). To sensitize plates with whole bacteria, strains 2996, MC58, 8047, and M7 were grown at 37°C in 5% CO2 in Muller-Hinton broth to the early exponential phase, washed in PBS, heat-inactivated (56°C for 30 min), and lyophilized. Five micrograms of lyophilized bacteria per well were used to sensitize the plates. After the addition of diluted mouse or rabbit sera, bound Abs were detected with HRP-conjugated polyvalent anti-mouse or anti-rabbit IgG (Amersham Pharmacia Biotech). Isotype-specific HRP-conjugated Abs (Vector Laboratories, distributed by Labtek, Milan, Italy) were used to detect mouse IgM, IgG1, IgG2a, IgG2b, and IgG3.

Bactericidal activity was assayed in serum samples from scFv-immunized animals using 96-well microtiter plates, with modifications of a previously described procedure (22). Group B N. meningitidis (2996 strain) was grown overnight on chocolate agar plates (Oxoid, Milan, Italy). Approximately five colonies were inoculated into Muller-Hinton broth, and test organisms were grown to an A620 of 0.24 (early log phase). After washing, ∼1000 CFU in 25 μl of PBS-BSA were added to 25 μl of undiluted, freshly thawed baby rabbit complement (Cedarlane, distributed by Celbio) or heat-inactivated complement and to 25 μl of diluted test sera. After a 1-h incubation with shaking, an aliquot (10 μl) of the reaction mixture from each well was transferred onto chocolate agar plates, and CFU were counted after an overnight incubation at 37°C in 5% CO2 and 95% humidity. Each assay was conducted in triplicate. Bactericidal activity was expressed as the percent variation in CFU numbers observed in the presence of fresh complement relative to heat-inactivated complement.

Opsonophagocytic activity was measured in serum samples in an identical way, except that 5 × 103 human polymorphonuclear leukocytes (PMNs) were added to the mixtures containing serum dilutions, bacteria, and complement. To isolate PMNs, the peripheral blood from adult healthy donors was centrifuged on a Polymorphoprep gradient (Sentinel CH, Milan, Italy), as described by the manufacturer.

Ab binding to mammalian polysialic acid was assayed in sera from immunized mice using the human neuroblastoma CHP 212 cell line (provided by A. Pession, Università di Bologna, Bologna, Italy), which expresses high levels of polysialic acid (23). Cells from nearly confluent cultures were collected, washed in PBS, and counted. One aliquot was incubated with exoneuraminidase (Sigma-Aldrich; 150 U/107 cells/ml) at 30°C for 2 h to cleave surface polysialic acid. The cells from each aliquot were then dispensed into the wells of microtiter plates (105 cells/well) and fixed for 1 h at 20°C with 2% paraformaldehyde (Sigma-Aldrich) in PBS. After blocking with 1% gelatin (1 h at 37°C) and washing with PBS-BSA, diluted serum samples were added to the wells, and the ELISA was developed as described above. The Seam 26 mAb (9), known to react against mammalian polysialic acid, was used as a positive control.

To study the protective effects of passively administered sera from immunized animals, a neonatal mouse model of systemic meningococcal infection was used. Neonatal (<48-h-old) BALB/c mice were randomly assigned to control or experimental groups (30/group), marked, and kept with the mother. Pups were inoculated s.c. with 30 μl of preimmune or immune serum pools and 4 h later were challenged s.c. with 4 and 34 × 105 CFU of strain 2996 corresponding, respectively, to 50 and 90% lethal doses. Mortality was assessed every 24 h for 5 days. Deaths rarely occurred after this time. In selected experiments, MenB was confirmed as the cause of death by blood cultures. Preliminary experiments showed that this model correlated well with the infant rat model in assaying the protective activities of mouse sera.

Differences between Ab concentrations and functional activities of sera were analyzed for statistical significance using one-way ANOVA and the Student-Keuls-Newman test. Differences in lethality were analyzed with Fisher’s exact test.

To produce anti-idiotypic scFvs, BALB/c mice were immunized with Seam 3, an mAb raised against N-Pr MenB CP (9). ScFv libraries displayed on filamentous phages were prepared from spleen cells of immunized mice, and individual clones were selected using Seam 3-coated plates. A total of 48 clones were obtained and two of these clones, designated G1 and B5, were selected on the basis of strong ELISA signals. As these clones were obtained from the same library, they could share clonal ancestry. To exclude this possibility, fingerprint analysis was performed on PCR-amplified DNA fragments after BstNII digestion. The restriction pattern of the G1 fragment markedly differed from that of B5 (not shown), thus excluding shared ancestry. Nucleotide sequence analysis further confirmed that G1 and B5 were different scFvs.

Next, it was determined whether the G1 and B5 anti-idiotypic scFvs could mimic some of the antigenic properties of the nominal Ag, i.e., the N-Pr MenB CP. First, we looked at the ability of the scFvs to compete with the N-Pr MenB CP for binding to Seam 3. Fig. 1 shows the results obtained with an inhibition ELISA using N-Pr MenB CP as the coating Ag. As expected, soluble N-Pr MenB CP, used as a positive control, inhibited Seam 3 binding, with a 50% inhibitory concentration of 10 ng/ml (Fig. 1, left panel). In contrast, the negative control, consisting of group-specific polysaccharide from group B streptococci, did not produce any inhibition at doses up to 50 μg/ml. Both the G1 and B5 scFvs could block Seam 3 binding in a dose-dependent fashion, with 50% inhibitory concentrations of 7 and 9 μg/ml, respectively. In contrast, the irrelevant H6 scFv failed to inhibit Seam 3 binding at concentrations as high as 50 μg/ml.

FIGURE 1.

ScFv-mediated inhibition of Ab binding to N-Pr MenB CP-coated plates. The left and right panels show inhibition of binding of, respectively, Seam 3 mAb and MenB-specific polyclonal rabbit Abs. Mixtures of Seam 3 (250 ng/ml, final concentration) or rabbit typing serum (final dilution of 1/40) and inhibitors were incubated for 20 min at 37°C and added to N-Pr MenB CP-coated plates. Soluble N-Pr MenB CP was used as a positive control. Negative controls consisted of the group-specific polysaccharide from group B streptococci or the irrelevant H6 scFv. After washing, the ELISA was developed as outlined in Materials and Methods. Percent inhibition was calculated relative to absorbance values observed in the absence of inhibitors.

FIGURE 1.

ScFv-mediated inhibition of Ab binding to N-Pr MenB CP-coated plates. The left and right panels show inhibition of binding of, respectively, Seam 3 mAb and MenB-specific polyclonal rabbit Abs. Mixtures of Seam 3 (250 ng/ml, final concentration) or rabbit typing serum (final dilution of 1/40) and inhibitors were incubated for 20 min at 37°C and added to N-Pr MenB CP-coated plates. Soluble N-Pr MenB CP was used as a positive control. Negative controls consisted of the group-specific polysaccharide from group B streptococci or the irrelevant H6 scFv. After washing, the ELISA was developed as outlined in Materials and Methods. Percent inhibition was calculated relative to absorbance values observed in the absence of inhibitors.

Close modal

True Ag mimics should be capable of reacting with Ag-specific Abs raised in different animal species, because in the latter, idiotypes unrelated to Ag binding are unlikely to be present. Therefore, in additional experiments we assessed the ability of G1 and B5 to inhibit binding of rabbit polyclonal Abs present in MenB-specific typing sera. Fig. 1 (right panel) shows that the positive control, consisting of soluble N-Pr MenB CP, produced up to 70% inhibition of Ab binding. The G1 and B5 scFvs inhibited binding of rabbit Abs to N-Pr MenB-coated wells by 45 and 37%, respectively (Fig. 1, right panel). The inability of the G1 and B5 scFvs to completely block binding of rabbit Abs was not surprising, because polyclonal Abs are expected to recognize a number of different MenB CP epitopes, whereas G1 and B5 can each mimic only a single epitope. These data indicated that the G1 and B5 scFvs could interact with the Ab-combining sites of anti-MenB CP Abs raised in different species and therefore mimicked the Ab-binding properties of a relevant MenB CP epitope.

In additional experiments we investigated whether immunization with the G1 and B5 scFvs could induce the production of anti-N-Pr MenB CP Abs. Sera were collected from BALB/c mice at different times after immunization and tested for binding to N-Pr MenB CP-sensitized plates. Low Ab levels were detected in preimmune sera (Fig. 2, left panels). Immunization with both the G1 and B5 scFvs induced significant elevations of Abs reacting against the N-Pr MenB CP. In contrast, immunization with the irrelevant H6 scFv produced no variation in Ab titers. Both the G1- and B5-induced Ab responses were boostable and predominantly of the IgG class (Fig. 2). Isotype analysis indicated that IgG2a and, to a lesser extent, IgG1 were the predominant Abs produced (Fig. 2, right panels).

FIGURE 2.

Anti-N-Pr MenB CP Ab levels in sera from scFv-immunized mice. Groups of five animals were immunized on days 0, 21, and 42 (arrows) with the anti-Seam 3 G1 or B5 scFvs or with the irrelevant H6 scFv. For comparison, two additional groups of five mice each were immunized with N-Pr MenB-TT or N-Ac MenB-TT conjugates. Sera were collected on days 0, 15, 36, and 57; diluted 1/200 (left panels) or 1/100 (right panels); and tested for binding to N-Pr MenB CP-coated plates by ELISA. This was performed as described in Materials and Methods using polyvalent, γ-chain- and μ-chain-specific anti-mouse Ig (left panels) or anti-mouse IgG1, IgG2a, IgG2b, or IgG3 (right panels) HRP-conjugated Abs. Data represent the mean ± SD of five determinations, each conducted in duplicate on a different animal. ∗, Significantly different from anti-H6 scFv sera; §, significantly different from anti-G1 or anti-B5 scFv serum.

FIGURE 2.

Anti-N-Pr MenB CP Ab levels in sera from scFv-immunized mice. Groups of five animals were immunized on days 0, 21, and 42 (arrows) with the anti-Seam 3 G1 or B5 scFvs or with the irrelevant H6 scFv. For comparison, two additional groups of five mice each were immunized with N-Pr MenB-TT or N-Ac MenB-TT conjugates. Sera were collected on days 0, 15, 36, and 57; diluted 1/200 (left panels) or 1/100 (right panels); and tested for binding to N-Pr MenB CP-coated plates by ELISA. This was performed as described in Materials and Methods using polyvalent, γ-chain- and μ-chain-specific anti-mouse Ig (left panels) or anti-mouse IgG1, IgG2a, IgG2b, or IgG3 (right panels) HRP-conjugated Abs. Data represent the mean ± SD of five determinations, each conducted in duplicate on a different animal. ∗, Significantly different from anti-H6 scFv sera; §, significantly different from anti-G1 or anti-B5 scFv serum.

Close modal

Next, the Ab levels measured after scFv immunization were compared with those obtained after immunization with N-Ac MenB or N-Pr MenB conjugate vaccines. The N-Pr MenB-TT, but not the N-Ac MenB-TT, conjugate vaccine produced significant elevations in anti-Pr Men B CP Ab responses that were predominantly of the IgG type (Fig. 2). The N-Pr MenB-TT IgG elevations were higher than those induced by the G1 or B5 anti-idiotypic scFvs (p < 0.05; Fig. 2, left panels).

In additional experiments it was of interest to ascertain whether the scFvs were also able to induce significant anti-N-Pr MenB CP responses when using alum as an adjuvant in place of CFA. To this end, alum-adsorbed G1 scFv-KLH conjugate was employed to immunize animals. Results were compared with those observed using an identical G1 scFvs dose in CFA. Under these conditions, both immunization treatments resulted in significant anti-N-Pr MenB CP IgG1 or IgG2a elevations, with no differences in Ab titers being observed between the two types of immunization (data not shown). These data indicated that CFA was not absolutely required for immunogenic activities of the scFvs.

Next, the ability of the G1 and B5 scFvs to induce anti-N-Pr MenB CP Ab responses was investigated in rabbits (Fig. 3). As previously observed in mice, a boostable anti-N-Pr MenB CP response, which was predominantly of the IgG type, was also observed in rabbits using the G1 scFv as an immunogen. The B5 scFv also induced elevations in IgG levels, although these were not statistically different from Ab levels in animals immunized with the irrelevant H6 scFv. This lack of statistical significance was probably secondary to the small number of animals examined. However, in view of the seemingly lower immunogenicity of the B5 scFv in rabbits, in further studies we focused on the G1 scFv only.

FIGURE 3.

Binding of sera from scFv-immunized rabbits to N-Pr MenB CP. Groups of three animals were immunized on days 0, 21, and 42 (arrows) with the anti-Seam 3 G1 or B5 scFvs or with the irrelevant H6 scFv. Sera were collected on days 0, 15, 36, and 57; diluted 1/200; and tested for binding to N-Pr MenB CP-coated plates by ELISA. This was performed as described in Materials and Methods, using polyvalent anti-rabbit IgG HRP-conjugated Abs. Data represent the mean ± SD of three determinations, each conducted in duplicate on a different animal. ∗, Significantly different from anti-H6 scFv sera by ANOVA and Student-Newman-Keuls test.

FIGURE 3.

Binding of sera from scFv-immunized rabbits to N-Pr MenB CP. Groups of three animals were immunized on days 0, 21, and 42 (arrows) with the anti-Seam 3 G1 or B5 scFvs or with the irrelevant H6 scFv. Sera were collected on days 0, 15, 36, and 57; diluted 1/200; and tested for binding to N-Pr MenB CP-coated plates by ELISA. This was performed as described in Materials and Methods, using polyvalent anti-rabbit IgG HRP-conjugated Abs. Data represent the mean ± SD of three determinations, each conducted in duplicate on a different animal. ∗, Significantly different from anti-H6 scFv sera by ANOVA and Student-Newman-Keuls test.

Close modal

In additional experiments, it was of interest to determine whether the anti-NPr MenB CP Abs present in sera from G1-immunized animals could bind to the surface of whole bacteria. To this end, multiwell plates were sensitized with different strains of heat-killed N. meningitidis, and selected pre- and postimmune samples from the experiments shown in Fig. 2 were tested by ELISA. The unencapsulated strain M7 was used to control for specificity of Ab binding to the bacterial capsule. As shown in Fig. 4, there was a low level of background binding to both encapsulated and nonencapsulated N. meningitidis. However, in each of the animals tested, an increased IgG binding to encapsulated bacteria was observed after immunization with the G1 anti-idiotypic scFv. No significant elevations in anti-bacterial Ab levels were detected using the unencapsulated strain. Collectively, these data indicated that the Abs induced by scFv immunization can recognize epitopes present on both the intact group B meningococcal capsule and purified N-Pr MenB CP.

FIGURE 4.

Binding of sera from scFv-immunized animals to whole bacterial cells using different MenB strains. Serum samples from mice (upper panel) or rabbits (lower panel) immunized with the G1 scFv were collected on day 0 (pre-immune) or 57 (postimmune) after the beginning of immunization. Samples from each group were diluted 1/400 and tested for binding to bacteria-coated plates as outlined in Materials and Methods, using polyvalent anti-mouse or anti-rabbit IgG HRP-conjugated sera. Data represent the mean ± SD of three determinations, each conducted in duplicate on a different serum sample. ∗, Significantly different from preimmune sera by ANOVA and Student-Newman-Keuls test.

FIGURE 4.

Binding of sera from scFv-immunized animals to whole bacterial cells using different MenB strains. Serum samples from mice (upper panel) or rabbits (lower panel) immunized with the G1 scFv were collected on day 0 (pre-immune) or 57 (postimmune) after the beginning of immunization. Samples from each group were diluted 1/400 and tested for binding to bacteria-coated plates as outlined in Materials and Methods, using polyvalent anti-mouse or anti-rabbit IgG HRP-conjugated sera. Data represent the mean ± SD of three determinations, each conducted in duplicate on a different serum sample. ∗, Significantly different from preimmune sera by ANOVA and Student-Newman-Keuls test.

Close modal

Abs directed against the MenB CP are known to be highly heterogeneous and to contain both human cross-reactive and non-cross-reactive Abs. To ascertain whether the Abs induced by the G1 scFv were capable of reacting with mammalian polysialic acid, the serum samples shown in Fig. 4 were tested for binding to the CHP 212 human neuroblastoma cell line, which is known to express high levels of polysialic acid. In addition, sera obtained from animals immunized with the N-Pr Men B-TT conjugate were tested. Both neuraminidase-treated and untreated neuroblastoma cells were used to detect Ag binding. As shown in Fig. 6, the polysialic acid-specific mAb Seam 26 produced a strong ELISA signal using untreated, but not neuraminidase-treated, cells, indicating that specific binding to polysialic acid was being detected by the assay. None of the pre- or postimmune mouse G1 sera tested showed significant binding to either the neuraminidase-treated or untreated CHP 212 cells (Fig. 5). In contrast, serum Abs from mice immunized with N-Pr MenB-TT bound at moderate, but significant, levels to neuroblastoma cells. This binding was abrogated by neuraminidase treatment (Fig. 5), suggesting that the Abs were specific for polysialic acid. These data indicated that N-Pr MenB-TT, but not G1 scFv, immunization induced Abs that were cross-reactive with human polysialic acid.

FIGURE 6.

Effects of scFv-priming on Ab responses to killed whole meningococci. Groups of five mice were inoculated with the anti-Seam 3 G1 or the irrelevant H6 scFvs on days 0 and 21 and boosted with heat-killed MenB strain 2996 (1 × 106 CFU) on day 42. A control group was immunized with the G1 scFv, but was not boosted with bacteria. Sera were collected on days 0, 15, 36, and 57; diluted 1/200; and tested for binding to N-Pr MenB CP-coated plates by ELISA. This was performed as described in Materials and Methods using HRP-conjugated anti-mouse Ig. Data represent the mean ± SD of five determinations, each conducted in duplicate on a different animal. ∗, Significantly different from anti-H6 scFv-primed sera; §, significantly different from sera of G1 scFv-primed animals that were not boosted with meningococci.

FIGURE 6.

Effects of scFv-priming on Ab responses to killed whole meningococci. Groups of five mice were inoculated with the anti-Seam 3 G1 or the irrelevant H6 scFvs on days 0 and 21 and boosted with heat-killed MenB strain 2996 (1 × 106 CFU) on day 42. A control group was immunized with the G1 scFv, but was not boosted with bacteria. Sera were collected on days 0, 15, 36, and 57; diluted 1/200; and tested for binding to N-Pr MenB CP-coated plates by ELISA. This was performed as described in Materials and Methods using HRP-conjugated anti-mouse Ig. Data represent the mean ± SD of five determinations, each conducted in duplicate on a different animal. ∗, Significantly different from anti-H6 scFv-primed sera; §, significantly different from sera of G1 scFv-primed animals that were not boosted with meningococci.

Close modal
FIGURE 5.

Cross-reactivity with polysialic acid of Abs in sera from scFv-immunized animals. Neuraminidase-treated or untreated cells from the CHP 212 human neuroblastoma cell line were fixed in the wells of microtiter plates. Serum samples from animals immunized with the G1 scFv or the N-Pr MenB-TT conjugate were collected on days 0 (preimmune) or 57 (postimmune) after the beginning of immunization. Serum samples from each group were diluted 1/20 and tested for binding to cell-coated plates by ELISA. This was performed as described in Materials and Methods using polyvalent anti-mouse or anti-rabbit IgG HRP-conjugated Abs. The positive control consisted of undiluted hybridoma culture supernatants containing the polysialic acid-specific murine mAb Seam 26. Data represent the mean ± SD of three determinations, each conducted in duplicate on a different serum animal. ∗, Significantly different from neuraminidase-treated cells by ANOVA and Student-Newman-Keuls test.

FIGURE 5.

Cross-reactivity with polysialic acid of Abs in sera from scFv-immunized animals. Neuraminidase-treated or untreated cells from the CHP 212 human neuroblastoma cell line were fixed in the wells of microtiter plates. Serum samples from animals immunized with the G1 scFv or the N-Pr MenB-TT conjugate were collected on days 0 (preimmune) or 57 (postimmune) after the beginning of immunization. Serum samples from each group were diluted 1/20 and tested for binding to cell-coated plates by ELISA. This was performed as described in Materials and Methods using polyvalent anti-mouse or anti-rabbit IgG HRP-conjugated Abs. The positive control consisted of undiluted hybridoma culture supernatants containing the polysialic acid-specific murine mAb Seam 26. Data represent the mean ± SD of three determinations, each conducted in duplicate on a different serum animal. ∗, Significantly different from neuraminidase-treated cells by ANOVA and Student-Newman-Keuls test.

Close modal

The ability of priming with anti-idiotypes to modify subsequent responses to the mimicked Ags has been clearly demonstrated (12, 13). It was therefore of interest to determine whether the G1 scFv was capable of priming animals for increased anti-N-Pr MenB Ab responses upon encounter with the pathogen. To this end, animals were immunized twice with different scFvs before inoculation with whole killed group B meningococci in IFA. In animals primed with the irrelevant H6 scFv, inoculation with meningococci induced low grade elevations in anti-MenB CP IgM (Fig. 6, lower panel). In contrast, brisk elevations of anti-MenB CP Abs were observed in G1-primed animals (Fig. 6). These data indicated that the G1 scFv mimic of the MenB CP could effectively prime mice for increased Ab responses after exposure to whole bacteria.

In additional experiments we looked at the ability of sera from scFv-immunized mice and rabbits to passively protect infant mice from lethality produced by group B meningococci. In these experiments we used a recently developed neonatal mouse model. Preliminary experiments showed that this model correlated well with the infant rat model in testing the protective activity of murine and human sera. Fig. 7 shows that sera from G1-scFv immunized animals moderately, but significantly, increased survival in pups inoculated with a 90 or 50% lethal dose of group B meningococci. These protective effects, however, were lower (p < 0.05) than those produced by passive immunization with sera from N-Pr Men B-TT-immunized mice (Fig. 7, upper panels).

FIGURE 7.

Passive protection using sera from scFv- or N-Pr MenB-TT-immunized animals in a neonatal meningococcal sepsis model. Neonatal mice (<48 h old) received 30 μl of sera diluted 1/15 s.c. at 4 h before challenge with a 50 or 90% lethal dose of MenB strain 2996, corresponding to, respectively, 4 or 34 × 105 CFU/ml. ∗, Significantly different from preimmune serum-treated neonatal mice by Fisher exact test; §, significantly different from anti-G1 scFv serum.

FIGURE 7.

Passive protection using sera from scFv- or N-Pr MenB-TT-immunized animals in a neonatal meningococcal sepsis model. Neonatal mice (<48 h old) received 30 μl of sera diluted 1/15 s.c. at 4 h before challenge with a 50 or 90% lethal dose of MenB strain 2996, corresponding to, respectively, 4 or 34 × 105 CFU/ml. ∗, Significantly different from preimmune serum-treated neonatal mice by Fisher exact test; §, significantly different from anti-G1 scFv serum.

Close modal

The ability of sera from scFv-immunized animals to kill N. meningitidis was examined in vitro using complement-dependent bactericidal or opsonophagocytosis assays. In these experiments we used selected serum samples obtained during the experiments shown in Fig. 2. Serum bactericidal activity (Fig. 8, □) was expressed as the percent decrease in CFU observed in the presence of fresh complement, relative to heat-inactivated complement. Although sera obtained from N-Pr MenB-TT immunized mice showed significant bactericidal activity, no such activity was observed after immunization of mice or rabbits with the G1 scFv. However, significant increases in opsonizing activity (Fig. 8, ▪) were detected in serum samples from G1-immunized animals compared with their respective preimmune controls.

FIGURE 8.

Opsonophagocytic (▪) and bactericidal (□) activities of sera from animals immunized with the anti-Seam 3 G1 scFv. Sera were obtained on day 0 (preimmune) or 57 (postimmune) after the beginning of immunization with the G1 scFv or the N-Pr MenB-TT conjugate, as described in Fig. 2. Sera were diluted 1/30 (mouse sera) or 1/100 (rabbit sera) before the assay. Data represent the mean ± SD of eight (left panel) or three (right panel) determinations, each conducted in duplicate on serum samples from different animals.

FIGURE 8.

Opsonophagocytic (▪) and bactericidal (□) activities of sera from animals immunized with the anti-Seam 3 G1 scFv. Sera were obtained on day 0 (preimmune) or 57 (postimmune) after the beginning of immunization with the G1 scFv or the N-Pr MenB-TT conjugate, as described in Fig. 2. Sera were diluted 1/30 (mouse sera) or 1/100 (rabbit sera) before the assay. Data represent the mean ± SD of eight (left panel) or three (right panel) determinations, each conducted in duplicate on serum samples from different animals.

Close modal

After the first demonstration that the antigenic functions of carbohydrates can be mimicked by anti-idiotypic Abs (24), murine mAbs have been successfully used as experimental anti-idiotypic vaccines to induce protective immunity against different encapsulated bacteria (13, 14, 15, 25). More recently, short peptides also proved capable of mimicking carbohydrate Ags (12, 26, 27). The introduction of effective, protein-conjugated, polysaccharide vaccines has alleviated the need for alternative strategies to prevent infections by many encapsulated pathogens. In the case of group B meningococcal infections, however, no effective capsule-based vaccine is available due to cross-reaction of the capsular polysaccharide with host polysialic acid.

In the present study we have developed scFv mimics of a protective capsular MenB epitope that are devoid of cross-reactivity with host polysialic acid. These anti-idiotypic mimics not only compete with the nominal Ag (i.e., N-Pr MenB CP) for binding to specific mouse or rabbit Abs, but also induce, after immunization, N-Pr MenB CP-specific serum Abs. These scFv-induced Abs had the following characteristics: 1) lack of binding to human polysialic acid; 2) binding to a capsular epitope expressed on both the intact bacterial surface and the purified capsular polysaccharide; 3) opsonizing activity, resulting in increased bacterial killing by human PMN leukocytes; and 4) protective activity in a neonatal model of MenB disease. Moe et al. (12) and Shin et al. (10) used a strategy similar to the one described in this study to select peptide mimics of the MenB CP using MenB CP-specific Abs. However, the immunogenic properties of these peptide mimotopes were not described. It should be noted that none of these peptides has any sequence homology with the hypervariable regions of the anti-idiotypic scFvs described in this study.

Collectively, our data indicate that the anti-idiotypic strategy we used was effective in mimicking the immunogenic properties of a protective capsular epitope while avoiding the risks inherent to vaccination with the whole capsular Ag, which is known to contain host cross-reactive epitopes. In fact, both whole MenB cells and the purified capsular polysaccharide express human cross-reactive epitopes. Although N-Pr derivatization of the Ag can markedly increase immunogenicity after protein conjugation (7) (Fig. 2), cross-reactive epitopes are still present on N-Pr MenB CP, as shown in immunized mice by low, but detectable, levels of anti-polysialic acid serum IgGs (Fig. 5). These data are in agreement with previously described work (8, 9). A high frequency of polysialic acid-specific mAbs was observed in N-Pr MenB CP-immunized mice after fusion of spleen cells (9). Moreover, a pool of sera from N-Pr MenB TT-immunized mice showed weak, but specific, IgG binding to polysialic acid-rich neural cell adhesion molecule in embryonic brain extracts and on the surface of live cells derived from a mouse pituitary tumor (8). No binding to polysialic acid was detected, however, by other tests. Collectively, published work and the present study indicate that weak, but discernible, elevations of anti-mammalian polysialic acid Abs can be induced by N-Pr MenB TT immunization. Although the neuropathological potential of these Abs has not been shown, it may be difficult to exclude that after crossing the placenta, they can somehow damage the nervous system of the developing fetus.

In the present study the scFv-induced responses were boostable and predominantly of the IgG2a type, indicating that these anti-idiotypes were capable of producing a Th1-like response. This feature is of interest, because capsular polysaccharides are incapable of inducing Th1-dominant responses even after protein conjugation. In fact, although protein conjugation does confer T-dependent help, conjugate vaccines have limited ability to induce the Th1-associated murine IgG2a isotype, while inducing predominantly IgG1 and IgG3 isotypes (28). A Th1 response with the expression of IgG2a would be desirable because IgG2a and IgG3 isotypes have been reported to be particularly effective in conferring protection against encapsulated organisms (29). In addition, the scFv immunization primed the mice for a MenB-specific response that was rapidly activated upon encounter with whole MenB. These data suggest that scFv immunization generates a memory B cell population that can be expanded by encounter with MenB bacteria.

However, some limitations of the anti-Id strategy used in this study need to be pointed out. First, the protective activities of the sera from scFv-immunized animals were only moderate in degree and considerably lower that those produced by the Seam-3 mAb, used as a template to develop the scFv mimics. Second, the scFv-induced sera had no significant bactericidal activity, whereas Seam 3 is highly bactericidal. The reasons underlying these phenomena are not clear, but are probably related to insufficient concentration and/or affinity of the Abs induced by the scFv immunization. Different approaches are being used to overcome these problems, all taking advantage of the ease with which the Ab fragments used can be manipulated by molecular techniques. Higher Ab responses could conceivably be obtained by expressing the scFvs as fusion proteins containing T cell epitopes or adjuvants, such as IL-12 or other cytokines. Indeed, IL-12 treatment was shown to dramatically enhance Ab responses and, in particular, IgG2a responses to both unconjugated and protein-conjugated capsular polysaccharide vaccines (28). New DNA immunization strategies are also likely to result in enhanced anti-scFv responses due to increased in vivo Ag persistence and the inclusion of DNA immunostimulatory sequences in the plasmid used for vaccination. It was recently shown that DNA vaccination with mimotope-encoded plasmid induced strong IgG2a anti-carbohydrate responses (30).

An additional strategy to increase the size and affinity of Ab responses is the development of further refined, second-generation mimics that would more closely resemble, from a structural standpoint, the natural Ag. Recent advances in structural chemistry involving x-ray and nuclear magnetic resonance spectroscopy may soon allow the creation of more “faithful” molecular mimics by design (31). Finally, the scFv described in this study may be used as fusion proteins in conjunction with other meningococcal Ags. Recently, promising vaccine candidates have been identified using whole genome sequencing of group B N. meningitidis (32). As all group B strains are sensitive to anti-capsular Abs, the inclusion of a capsule-based immunogen in such vaccines may help prevent bacterial escape due to the emergence of mutant strains in vaccinated populations.

2

Abbreviations used in this paper; CP, capsular polysaccharide; KLH, keyhole limpet hemocyanin; MenB, serogroup B meningococcal; N-Ac, N-acetyl; N-Pr, N-propionyl; PMN, polymorphonuclear leukocyte; scFv, single-chain Ab fragment; TT, tetanus toxoid.

1
Jones, D..
1995
. Epidemiology of meningococcal disease in Europe and the USA. K. Cartwright, ed.
Meningococcal Disease
147
.-175. John Wiley & Sons, New York.
2
Naess, A., A. Halstensen, H. Nyland, S. H. Pedersen, P. Moller, R. Borgmann, J. L. Larsen, E. Haga.
1994
. Sequelae one year after meningococcal disease.
Acta Neurol. Scand.
89
:
139
.
3
Klein, N. J., C. A. Ison, M. Peakman, M. Levin, S. Hammerschmidt, M. Frosch, R. S. Heyderman.
1996
. The influence of capsulation and lipooligosaccharide structure on neutrophil adhesion molecule expression and endothelial injury by Neisseria meningitidis.
J. Infect. Dis.
173
:
172
.
4
Frasch, C. E., W. D. Zollinger, J. T. Poolman.
1985
. Serotype antigens of Neisseria meningitidis and a proposed scheme for designation of serotypes.
Rev. Infect. Dis.
7
:
504
.
5
Granoff, D. M., B. Forrest, R. Rappuoli.
1997
. Meningococcal polysaccharide-protein conjugate vaccines.
Int. J. Infect. Dis.
1
:
152
.
6
Finne, J., M. Leinonen, P. H. Mäkelä.
1983
. Antigenic similarities between brain components and bacteria causing meningitis: implications for vaccine development and pathogenesis.
Lancet
2
:
355
.
7
Jennings, H. J..
1997
. N-propionylated group B meningococcal polysaccharide glycoconjugate vaccine against group B meningococcal meningitis.
Int. J. Infect. Dis.
1
:
158
.
8
Coquillat, D., J. Bruge, B. Danve, M. Latour, C. Hurpin, D. Schulz, P. Durbec, G. Rougon.
2001
. Activity and cross-reactivity of antibodies induced in mice by immunization with a group B meningococcal conjugate.
Infect. Immun.
69
:
7130
.
9
Granoff, D. M., A. Bartoloni, S. Ricci., E. Gallo, D. Rosa, N. Ravenscroft, V. Guarnirei, R. C. Seid, A. Shan, W. R. Usinger, et al
1998
. Bactericidal monoclonal antibodies that define unique meningococcal B polysaccharide epitopes that do not cross-react with human polysialic acid.
J. Immunol.
160
:
5028
.
10
Shin, J. S., J. S. Lin, W. Anderson, R. A. Insel, M. H. Nahm.
2001
. Monoclonal antibodies specific for Neisseria meningitidis group B polysaccharide and their peptide mimotopes.
Infect. Imm.
69
:
3335
.
11
Pon, R. A., M. Lussier, Q. L. Yang, H. J. Jennings.
1997
. N-propionylated group B meningococcal polysaccharide mimics a unique bactericidal capsular epitope in group B Neisseria meningitidis.
J. Exp. Med.
185
:
1929
.
12
Moe, G. R., S. Tan, D. M. Granoff.
1999
. Molecular mimetics of polysaccharide epitopes as vaccine candidates for prevention of Neisseria meningitidis serogroup B disease.
FEMS Immun. Med. Microbiol.
26
:
209
.
13
McNamara, M., R. E. Ward, H. Kohler.
1984
. Monoclonal idiotype vaccine against Streptococcus pneumoniae infection.
Science
226
:
1325
.
14
Stein, K. E., T. Soderstrom.
1984
. Neonatal administration of idiotype or anti-idiotype primes for protection against E. coli K13 infection in mice.
J. Exp. Med.
160
:
1001
.
15
Westerink, M. A. J., A. A. Campagnari, M. A. Wirth, M. A. Apicella.
1988
. Development and characterization of an anti-idiotype antibody to the capsular polysaccharide of Neisseria meningitidis serogroup C.
Infect. Immun.
56
:
1120
.
16
Magliani, W., L. Polonelli, S. Conti, A. Salati, P. F. Rocca, V. Cusumano, G. Mancuso, G. Teti.
1998
. Neonatal mouse immunity against group B streptococcal infection by maternal vaccination with recombinant anti-idiotypes.
Nat. Med.
4
:
705
.
17
Stephens, D. S., J. S. Swartley, S. Kathariou, S. A. Morse.
1991
. Insertion of Tn916 in Neisseria meningitidis resulting in loss of group B capsular polysaccharide.
Infect. Immun.
59
:
4097
.
18
Bartoloni, A., F. Norelli, C. Ceccarini, R. Rappuoli, P. Costantino.
1995
. Immunogenicity of meningococcal B polysaccharide conjugated to tetanus toxoid or CRM197 via adipic acid dihydrazide.
Vaccine
13
:
463
.
19
Magliani, W., S. Conti, F. De Bernardis, M. Gerloni, D. Bertolotti, P. Mozzoni, A. Cassone, L. Polonelli.
1997
. Therapeutic potential of antiidiotypic single chain antibodies with yeast killer toxin activity.
Nat. Biotechnol.
15
:
155
.
20
Beninati, C., M. R. Oggioni, M. Boccanera, M. R. Spinosa, T. Maggi, S. Conti, W. Magliani, F. De Bernardis, G. Teti, A. Cassone, et al
2000
. Therapy of mucosal candidiasis by expression of an anti-idiotype in human commensal bacteria.
Nat. Biotechnol.
18
:
1060
.
21
Mancuso, G., F. Tomasello, C. von Hunolstein, G. Orefici, G. Teti.
1994
. Induction of tumor necrosis factor alpha by the group and type-specific polysaccharide from type III group B streptococci.
Infect. Immun.
62
:
2748
.
22
Mandrell, R. E., F. H. Azmi, D. M. Granoff.
1995
. Complement-mediated bactericidal activity of human antibodies to poly α 2–8 N-acetyl neuraminic acid, the capsular polysaccharide of Neisseria meningitidis serogroup B.
J. Infect. Dis.
172
:
1279
.
23
Glick, M. C., B. D. Livingston, G. W. Shaw, J. L. Jacobs, F. A. Troy.
1991
. Expression of polysialic acid on human neuroblastoma.
Prog. Clin. Biol. Res.
366
:
267
.
24
Rubinstein, L. J., B. Goldberg, J. Hiernauz, K. E. Stein, C. A. Bona.
1983
. Idiotype-antiidiotype regulation: the requirement for immunization with antigen and monoclonal antiidiotypic antibodies for the activation of β2→6 and β2→1 polyfructan reactive clones in BALB/c mice treated at birth with minute amounts of anti-A48 idiotype antibodies.
J. Exp. Med.
158
:
1129
.
25
Schreiber, J. R., G. B. Pier, M. Grout, K. Nixon, M. Patawaran.
1991
. Induction of opsonic antibodies to Pseudomonas aeruginosa mucoid exopolysaccharide by an anti-idiotypic monoclonal antibody.
J. Infect. Dis.
164
:
507
.
26
Westerink, M. A. J., P. C. Giardina, M. A. Apicella, T. Kieber-Emmons.
1995
. Peptide mimicry of the meningococcal group C capsular polysaccharide.
Proc. Natl. Acad. Sci. USA
92
:
4021
.
27
Agadjanyan, M., P. Luo, M. A. Westerink, L. A. Carey, W. Hutchins, Z. Steplewski, D. B. Weiner, T. Kieber-Emmons.
1997
. Peptide mimicry of carbohydrate on human immunodeficiency virus.
Nat. Biotechnol.
15
:
547
.
28
Buchanan, R. M., B. P. Arulanandam, D. W. Metzger.
1998
. IL-12 enhances antibody responses to T-independent polysaccharide vaccines in the absence of T and NK cells.
J. Immunol.
161
:
5525
.
29
Peeters, C. C., A. M. Tenbergen-Meekes, J. T. Poolman, M. Beurret, B. J. Zegers, G. T. Rijkers.
1991
. Effect of carrier priming on immunogenicity of saccharide-protein conjugate vaccines.
Infect. Immun.
59
:
3504
.
30
Kieber-Emmons, T., B. Monzavi-Karbassi, B. Wang, P. Luo, D. B. Weiner.
2000
. Cutting edge: DNA immunization with minigenes of carbohydrate mimotopes induce functional anti-carbohydrate antibody response.
J. Immunol.
165
:
623
.
31
Beninati, C., G. Teti.
2000
. A meeting of mimics.
Trends Microbiol.
8
:
492
.
32
Pizza, M., V. Scarlato, V. Masignani, M. M. Giuliani, B. Arico, M. Comanducci, G. T. Jennings, L. Baldi, E. Bartolini, B. Capecchi, et al
2000
. Identification of vaccine candidates against serogroup B meningococcus by whole-genome sequencing.
Science
287
:
1816
.