Research

Efforts to develop a capsular-based vaccine for prevention of N. meningitidis group B strains have been hampered by poor immunogenicity of the group B polysaccharide and the risk of eliciting autoantibodies. Surface-exposed noncapsular antigens also can elicit protective antibody but a major limitation of this approach has been antigenic variability and the resulting failure of protein vaccine candidates to elicit broadly protective bactericidal antibody responses. Our laboratory is investigating several new vaccine candidates for prevention of all N. meningitidis disease, including group B strains. These include Nesisserial surface protein A, several other novel conserved proteins discovered during the N. meningitidis group B MC58 genome-sequencing project (Pizza et al, Science 2000;287:1816), and improved vesicle vaccines.

Nessierial protein A (NspA)
NspA is a 18.6 kDa membrane protein of unknown function that was first described by Martin and Brodeur and colleagues (Martin et al, J. Exp. Med. 1997;185:1173). Unlike other Neisseiral surface proteins, such as PorA and Opc that also have been shown to elicit bactericidal protective antibody, NspA is highly conserved and expressed by all N. meningitidis strains tested to date (Moe et al, Infect Immun 2001;69:3762). Immunization of mice with rNspA also conferred protection against bacteremia in animals challenged with a group B strain (Martin et al, J. Exp. Med. 1997;185:1173. However, rNspA has been difficult to formulate into an effective vaccine for humans. Our studies showed that despite nspA gene conservation there are large differences in surface-accessible NspA among N. meningitidis strains (Moe et al, Infect Immun 1999; 67:5664 and 2001;69:3762. A recent crystallographic study shows that NspA adopts an 8-stranded b barrel structure when reconstituted in detergent (Vandeputte-Rutten et al, J. Biol. Chem. 2003;278:24825). In order to define the segments of NspA-containing epitopes recognized by protective murine anti-NspA antibodies, we studied the binding of two bactericidal and protective anti-NspA MAbs, AL12 and 14C7. Neither MAb binds to overlapping synthetic peptides (10-mers, 12-mers, and cyclic 12-mers) corresponding to the entire mature sequence of NspA, or to denatured rNspA, although binding to the protein can be restored by refolding in liposomes. Based on the ability of the two MAbs to bind to E. coli microvesicles prepared from a set of rNspA variants created by site-specific mutagenesis, the most important contacts between the MAbs and NspA appear to be located within the LGG segment of putative loop 3. The conformation of putative loop 2 also appears to be an important determinant, as particular combinations of residues in this segment result in loss of antibody binding. Thus, the two anti-NspA MAbs recognize discontinuous conformational epitopes that result from the close proximity of loops 2 and 3 in the three- dimensional structure of NspA. The data suggest that optimally immunogenic vaccines using rNspA will require formulations that permit proper folding of the protein.

Figure 1. Epitopes recognized by protective anti-NspA antibodies. Recently, we showed that protective anti-NspA mAbs recognize epitopes located in loops 2 and 3 of NspA (Hou et al., Infect. & Immun. 2003 71:6844). The left panel shows a ribbon cartoon of the 8-stranded b-barrel structure determined for NspA by Vandeputte-Rutten et al. (J. Biol. Chem. 2003 278:24825). The right panel shows a surface rendering of the antibody accessible portion of NspA as viewed down long axis of the molecule. In both panels loops 2 and 3 are colored magenta.

New meningococcal protein vaccine candidates identified by reverse vaccinology
"Reverse vaccinology," a term coined by Rino Rappuoli (Rappuoli et al, Vaccine 2001;19:2688), has been used to identify new vaccine candidates for prevention of group B meningococcal disease. This approach begins with the genomic sequence instead of antigens identified directly from studies of the microbial cell envelope. The group B strain MC58 meningococcal genome sequencing project identified 2158 open reading frames (Tettelin et al., Science 2000;287:1809). While this effort was underway, sequences of unassembled DNA fragments were analyzed using computer algorithms by Pizza and colleagues at Chiron Vaccines, Italy to identify open reading frames that encoded potentially novel surface-exposed proteins (Pizza et al, Science 2000;287:1816). The encoded amino acid sequences of the group B strain were compared to those in the genome data bank of a group A meningococcal strain and a N. gonorrhoeae strain. The investigators correctly deduced that if the sequences of the group B strain were similar to those of the corresponding proteins in the group A meningococcal and gonococcal strains, then the amino acid sequence also would be conserved across diverse group B strains. Meningococcal proteins predicted to be surface-exposed and conserved were cloned and expressed in E. coli. Mouse antisera were prepared to 344 recombinant proteins. Of these, 28 novel proteins, designated “genome derived antigens” (GNA), were identified that elicited antibodies that bound to the bacterial surface and/or had bactericidal activity. To date, the most promising GNA vaccine candidates include GNA33, a mimetic of a surface-exposed loop of PorA (Granoff et al, J. Immunol. 2001;167:6487), NadA, a surface-exposed molecule that may play a role in adherence (Comanducci et al, J. Exp. Med. 2002;195:1445), and two lipoproteins of unknown function, GNA2132 (Welsch et al, J. Inf. Dis. 2003;188:1730) ] and the recently identified GNA1870 (Masignani et al, J. Exp. Med. 2003;197:789)

GNA33, a novel mimetic of a surface-exposed loop of PorA.
GNA33 is a lipoprotein with amino acid sequence homology to that of membrane-bound lytic murein transglycosylase (MltA) from E. coli and Synechocystis sp. Mice immunized with rGNA33 responded with serum bactericidal titers that were of similar magnitude to those elicited in control mice by an OMV vaccine known to confer protection in humans (Pizza et al, Science 2000;287:1816). Our subsequent studies of GNA33 demonstrated that this protein elicits protective antibody as a result of mimicking an epitope on loop 4 of PorA in strains with serosubtype P1.2 (Granoff et al, J. Immunology, 2001;167:6487). Anti-GNA33 polyclonal and monoclonal antibodies passively confer protection against meningococcal bacteremia in infant rats challenged with different N. meningitidis strains with serosubtype P1.2. The protective activity against some strains was equivalent or superior to that of anticapsular or anti-PorA antibodies (Table 1). Thus, GNA33 represents one of the most effective immunogenic mimetics yet described.

To study the function of the GNA33 gene and its role in pathogenesis and virulence, a knockout mutant of a meningococcus B strain was generated (Adu-Bobie et al, Infect. Immun. 2004;72:1914). The mutant exhibited a retarded growth in vitro. Transmission electron microscopy revealed that the mutant grows in clusters, which are connected by a continuous outer membrane, suggesting a failure in the separation of daughter cells. The mutant also showed an attenuated phenotype since it was not able to cause bacteraemia in the infant rat model. Thus, GNA33 is a highly conserved lipoprotein, which plays an important role in peptidoglycan metabolism, cell separation, membrane architecture, and virulence.

Other lipoprotein vaccine candidates
GNA2132 and GNA1870 are two other putative surface lipoproteins that were identified by Pizza and coworkers during the group B MC58 genome sequencing project (Pizza et al, Science 2000;287:1816).

The GNA2132 gene was detected in 31/31 genetically diverse N. meningitidis strains, and also was present in strains of N. lactamica and N. gonorrhoeae. Serum antibody from mice immunized with rGNA2132 elicited complement-mediated bactericidal activity against group B strain 2996. Based on gene sequences from 31 genetically diverse group B strains, there are portions of GNA2132 that are variable while segments at the amino and carboxyl-terminal ends are highly conserved. Because it was unclear whether the conserved segments were sufficient to elicit cross-protective antibody against heterologous N. meningitidis strains, we immunized mice with rGNA2132 (gene from strain NZ394/98). (Welsch et al, J.Infect. Dis. 2003;188:1730). The resulting antisera were tested for their ability to bind to the surface of live bacteria, to promote deposition of complement protein C3b/iC3c on the bacterial surface, and to elicit complement mediated bactericidal activity. We also tested the ability of the antisera to confer passive protection against meningococcal bacteremia in infant rats challenged with different N. meningitidis strains. By flow cytometry, anti-GNA2132 antibody bound to the surface of live bacteria from all 7 capsular group B or C strains tested, and elicited deposition of human C3b on the bacterial surface. However, with human or infant rat complement, anti-GNA2132 had no detectable bactericidal activity (titer <1:4) against the nominal strain, NZ394/98, and was bactericidal against only 2 of the other 6 strains tested (Table 2, below). These strain differences were unrelated to GNA2132 amino acid sequence or level of protein expression.

Despite lack of bactericidal activity, anti-GNA2132 antiserum passively protected infant rats against meningococcal bacteremia after challenged with all 5 resistant strains. Thus, antibody binding and activation of C3b deposition on the bacterial surface may be sufficient for protection in the absence of complement-mediated bacteriolysis, and GNA2132 is a promising vaccine candidate for prevention of disease caused by N. meningitidis.

GNA1870.
GNA1870 is a meningococcal vaccine candidate that can be subdivided into three variants based on amino acid sequence variability (Masignani et al, J. Exp. Med. 2003;187:789). Variant group 1 accounts for ~60% of disease-producing group B isolates. The antigen went unrecognized until its discovery by genome mining because it is expressed in low copy number by most strains. To investigate the relationship between antibody binding to GNA1870 and complement-mediated protective functions, we prepared a panel of 4 murine IgG MAbs against rGNA1870 (variant 1) and evaluated their activity against 9 genetically diverse encapsulated N. meningitidis strains expressing sub-variants of variant 1 GNA1870 (Welsch et al., J. Immunol. 2004;172:5606). Based on flow cytometry with live encapsulated bacteria, surface-accessibility of the epitopes recognized by the MAbs appeared to be low in most strains (Representative data for strains MC58, MC5DGNA1870 and 4243 are shown in Figure 2, Panel A below). Yet MAb concentrations <1 µg/ml to 5 µg/ml were sufficient to elicit bactericidal activity with human complement and/or activate C3b deposition on the bacterial surface. Representative data on the ability of the MAbs to activate human C3b deposition on the bacterial surface of strains MC5D, MC5DGNA1870 and 4243 as measured by flow cytometry, are shown below (Figure 2, Panel B). There was no evidence of complement deposition when the bacterial cells were incubated with complement together with a 1:100 dilution of the negative control antiserum (row 2). The addition of complement to a 1:250 dilution of the polyclonal anti-GNA1870 antiserum (row 2) elicited strong deposition of C3b on the surface of strains MC5D and 4243, as shown by an increase in the percentages of bacteria showing strong immunofluorescence with the anti-C3c antibody, which recognizes C3b and iC3bi. With strain MC5D, as little as 0.5 µg/ml of JAR1, JAR3 or JAR4 (rows 3, 4 and 5, respectively) activated C3b deposition. In contrast, JAR5 elicited minimal activation of complement deposition, even when tested at MAb concentration of 50 µg/ml (row 6). With strain 4243, there was no significant complement deposition elicited by JAR1 or JAR5 (rows 3 and 6, respectively). However, as little as 0.5 µg/ml of JAR3, or 50 µg/ml of JAR4, elicited strong complement deposition on strain 4243 (rows 4 and 5. respectively) despite minimal binding of these MAbs to this strain (Panel A).

Figure 2. Binding of anti-GNA1870 antibody to the bacterial surface (Panel A) and activation of C3b deposition (Panel B) as measured by flow cytometry.

Several of the anti-GNA1870 MAbs conferred passive protection against bacteremia in infant rats challenged by strains resistant to bacteriolysis (Table 4), and the protective activity paralleled the ability of the MAb to activate C3b deposition. Thus, despite low GNA1870 surface-exposure, anti-GNA1870 variant 1 antibodies are bactericidal and/or elicit C3b deposition. The antibodies also confer protection against bacteremia caused by encapsulated N. meningitidis strains expressing GNA1870 sub-variant 1 proteins. The data support GNA1870 as a promising vaccine candidate for prevention of meningococcal group B disease caused by GNA1870 variant 1 strains.

Sequential immunization with vesicle vaccines prepared from heterologous Neisseria meningitidis strains (Moe et al, Infect. Immun. 2002;70:6021).
Although certain recombinant proteins are promising meningococcal vaccine candidates, an advantage of outer membrane vesicle (OMV) vaccines is that they already have been proven to confer protection in humans against developing meningococcal disease. However, immunization with vesicle vaccines prepared from a single meningococcal strain is known to elicit bactericidal antibody responses directed primarily against surface-exposed loops of PorA and, to a lesser extent, Opc. Therefore, bactericidal antibody responses, particularly in children, tend to be specific to strains with PorA variable region (VR) sequence types identical or similar to that of the strain or strains used to prepare the vaccine. In an attempt to broaden bactericidal antibody responses, we sequentially immunized mice and guinea pigs with vesicle vaccines prepared from three N. meningitidis strains that each differed from one another by capsular group, PorA VR sequence type, PorB serotype and LOS immunotypes (Moe et al, Infect. Immun. 2002;70:6021). Fig. 3, below, shows a SDS-PAGE of vesicles from each of the strains. Our hypothesis (shown schematically in Fig. 4, below) is that repeated presentation of highly conserved but normally poorly immunogenic antigens in the context of different, variable antigens such as Por A that are normally immunodominant, focuses the antibody responses to the conserved vesicle antigens. Many of these putative conserved proteins are likely to be present in low copy number, and are not evident in Fig 3.

Figure 3 (left). 15% SDS-PAGE of vesicle vaccines prepared from three heterologous N. meningtidis strainins. Figure 4 (right) Schematic Model of sequential immunication

Mice and guinea pigs were sequentially immunized with three doses of micovesicles or outer membrane vesicles prepared from three meningococcal strains that were each antigenically heterologous with respect to the two major porin proteins, PorA and PorB, and the group capsular polysaccharide. The antisera from mice or guinea pigs given sequential immunization, or control animals given three injections of a mixture of the three vesicle vaccines, showed high bactericidal antibody responses measured against the three strains used to prepare the vaccines (Moe et al, Infect. Immun. 2002;70:6021). For strains with heterologous VR sequence types, mice (Fig. 5, Panel A, below) or guinea pigs (Fig. 5 Panel B, below) given sequential immunization had higher bactericidal antibody responses than those of animals given 3 injections of a mixture of vesicles. Using knock-out strains, a portion of the bactericidal antibody was directed against the highly conserved protein, Neisserial surface protein A (NspA). Further, an anti-NspA monoclonal antibody elicited by the sequential immunization was highly bactericidal against strains that were previously shown to be resistant to bacteriolysis by anti-NspA antibodies produced by immunization with recombinant NspA. Sequential immunization with heterologous vesicle preparations offers a novel approach to eliciting broadly protective immunity against Neisseria meningitidis strains.
 

Figure 5. Reverse cumulative distribution of bactericidal titers of mice (Panel A) or guinea pigs (Panel B) measured with human C. Data are for test strains with VR sequence types heterologous to those of strains used to prepare the vaccine. P values calculated by Fisher exact test comparing proportion with titers >1:8. A) p <.07 (mice), B) p< .02 (GP) X, Negative control animals given E. coli vesicles; 0, Animals immunized with mixture of meningococcal vesicles; s, Animals given sequential immunization.

Figure 6. Relationship between antibody avidity and dose of antibody per rat required for passive protective activity (defined as a 2 log10 decrease in CFU/ml, as compared to that of control animals pre-treated with pre-immunization serum or PBS). Panel A, Challenge by group C OAc-positive strain 4243. Panel B, challenge by the group C OAc-negative strain 4335. The respective r2 values calculated by ANOVA non-linear regression were 0.60 for strain 4342 (p<0.03), and 0.69 for strain 4335 (p<0.01). Excluding the outlier for strain 4243, the r2 value was 0.92 (p<0.001) (Harris et al, Infect & Immun 2003, 71:275).

Protective activity of group C anticapsular antibodies elicited in 2 year-olds by an investigational quadrivalent Neisseria meningitidis-diphtheria toxoid conjugate vaccine (Granoff et al, Ped. Infect. Dis. J. 2004; 23:490-497. Presented at the Pediatric Academic Societies’ Annual Meeting, May 2-4, 2004, San Francisco, CA)
Quadrivalent capsular group A, C, Y and W-135 meningococcal conjugate (MC-4) vaccines are under development. To predict efficacy of an investigational MC-4 vaccine in 2 year-old children for prevention of group C disease, we measured group C antibody concentrations, avidity, bactericidal and passive protective activity in sera from 2 year-olds given one dose of MC-4 vaccine (N=30), and 3 year-olds (N=30) and adults (N=26) given one dose of meningococcal polysaccharide (MPS-4) vaccine. One month after vaccination, the geometric mean anticapsular antibody concentration of children given the MC-4 vaccine (3.1 µg/ml) was lower than that of control children (5.1 µg/ml, P<0.04) or adults immunized with MPS-4 vaccine (22.9 µg/ml, P<0.001). However, the percent of sera with protective bactericidal titers of >1:4 was higher in children given MC-4 vaccine (50%, vs. 17% in children given MPS-4 vaccine P<0.02), and was not significantly different from that of immunized adults (65%). In children, the mean antibody avidity at 1 month was higher in the MC-4 group (22 nM^-1 vs. 16 nM^-1 in the MPS-4 group, P=0.002), and at six months increased in the MC-4 group (28 nM-1, P<0.001), but not in the MPS-4 vaccine group (17 nM-1). Higher avidity antibody gave greater passive protection in the infant rat bacteremia model than did lower avidity antibody (P<0.03). In conclusion, although MPS-4 vaccine elicited higher group C serum antibody concentrations in 3 year-olds than did MC-4 vaccine in 2 year-olds, the higher antibody avidity after MC-4 vaccine resulted in higher bactericidal and passive protective activity.

Serum group A anticapsular antibodies in a Sudanese population immunized during a group A epidemic (Ismail et al, Ped. Infect. Dis. J. 2004; In press).
Vaccination during group A meningococcal epidemics is reported to decrease the number of new cases of disease. However, implementing mass vaccination is often delayed, and little information is available on whether immunity increases in a population vaccinated during an epidemic when exposure to the epidemic strain is common. We assayed sera from a convenience sample of 134 previously unimmunized Sudanese, ages 3- to 49 years, immunized with a meningococcal polysaccharide vaccine during the 1999 group A meningococcal epidemic. Their serum group A antibody responses were compared with those of 26 adults immunized in California with no known group A exposure. Before immunization, serum anticapsular antibody concentrations were 10-fold higher in Sudanese adults, and 4-fold higher in Sudanese, ages 3 to 17 years, than in North American adults (geometric means of 29 µg/ml and 13 µg/ml, respectively, vs. 3 µg/ml, P<0.001). Seventy-five percent of the Sudanese had serum bactericidal titers that correlate with protection ( >1:128). Nearly all Sudanese with low bactericidal titers before vaccination developed protective bactericidal antibody responses after vaccination, and the magnitude of the anticapsular antibody responses of the Sudanese was similar to that of the immunized North Americans. The high titers of naturally-acquired antibody in the Sudanese may reflect widespread exposure to the epidemic strain and underscore difficulties of instituting immunization before exposure occurs. Also, epidemics in Sub-Saharan African may not abate even if 75 percent of the population is immune to disease as long as the organism is transmitted widely among both immune and susceptible persons.

Developing a Meningococcal conjugate vaccine for Africa: A model for developing new vaccines for the poorest countries (Jodar et al, Lancet 2003;361:1902)
Group A meningococcal epidemics cause tens of thousands of deaths in Sub-Saharan Africa (Figure 7 below). An effective vaccine could eliminate these epidemics. Vaccine manufacturers are currently developing tria- and quadravalent meningococcal conjugate vaccines that are intended for use in premium-priced markets but these vaccines will not be affordable in Africa. We analysed the costs of construction of manufacturing capacity, process development and clinical and regulatory activities for licensure of a group A meningococcal conjugate vaccine for use in Africa. We investigated two approaches: 1) subsidizing these costs for an established vaccine manufacturer in the US or Europe in return for sale of the vaccine at a low price, or 2) purchasing intermediate components, developing the conjugation process, and transferring the technology to a developing-country vaccine manufacturer for large-scale production in return for sale of the vaccine at a low price. Established vaccine companies did not view the incentives sufficient to offset the high “opportunity costs” of the project. In contrast, developing country manufacturers viewed the second model as a positive “opportunity” and were willing to provide 25 million doses per year of a group A meningococcal conjugate vaccine at a target price of $US 0.40 per dose. The project has started with a $70 million grant from the Bill and Melinda Gates Foundation focusing on the second approach (www.meningovacc.org). In summary, large quantities of a high-quality and low price group A meningococcal vaccine will be made by technology transfer and subsidizing the costs of vaccine development for a developing country manufacturer. Widespread use of this vaccine could eliminate group A meningococcal epidemics in Africa. The project has implications as a model for the development of other vaccines for use in the developing world.