|Home | About | Journals | Submit | Contact Us | Français|
The potential protective effect of existing vaccines against serogroup B meningococci, based on outer membrane proteins, is limited by strain restriction and apparent short duration of immune responses. In contrast, meningococcal colonization is known to stimulate the production of cross-protective antibodies as defined by the development of serum bactericidal activity (SBA) against heterologous serogroup B strains. In the current study, a resource of human serum samples and meningococcal carriage strains from studies of longitudinal carriage has been subjected to immunoproteomic analysis to investigate the outer membrane protein antigens associated with the development of SBA to both homologous and heterologous meningococcal serogroup B strains. Proteins from outer membranes of homologous and heterologous strains were separated by two-dimensional electrophoresis and reacted with paired sera which showed an increase in SBA following colonization. Individuals showed differing patterns of reactivity upon colonization, with an increase in SBA being associated with increases in the number of spots detected before and after colonization and/or with increases in the intensity of individual spots. Analysis of immunoreactive spots by mass spectrometry resulted in the identification of 43 proteins potentially associated with the development of SBA against both homologous and heterologous strains. The list of protein immunogens generated included not only well-established antigens but also novel proteins that represent potentially new candidates for inclusion in defined, multicomponent serogroup B vaccines.
Neisseria meningitidis (meningococcus) is a major causative agent of potentially life-threatening meningitis and septicemia. The case fatality rate from meningococcal disease is ca. 7% (34), with infants and adolescents most at risk. Meningococci expressing capsular polysaccharides B and C have been associated with most cases of invasive disease in America, Europe, and the United Kingdom (46). Following the initial introduction of a conjugate serogroup C capsular polysaccharide vaccine in the United Kingdom, there has been a major decline in the incidence of serogroup C infections in those countries that have introduced routine vaccination (51). However, the majority of cases of meningococcal infection continue to be caused by serogroup B meningococci, for which no proven vaccine exists.
The lack of immunogenicity of the serogroup B polysaccharide capsule and its molecular mimicry of human neural cell adhesion molecules that are expressed on the developing fetal brain (10) preclude its use as a capsule-based vaccine and have placed the emphasis for serogroup B vaccine development on the potential of subcapsular antigens (20). This has led to the development of experimental vaccines based on outer membrane (OM) “blebs,” released from the surface of meningococci during growth, from which the toxic lipopolysaccharide (LPS) has been selectively removed (4). Such OM vesicle (OMV) vaccines have been used in attempts to control outbreaks of serogroup B infection caused by single strains (5, 35, 45). However, the success of such vaccines has been limited by strain restriction (49), apparent short duration of immune responses (6, 56), and poor immunogenicity in children under 2 years of age (33). Moreover, human immune responses to different components of OMV vaccines have been shown to be variable and only a proportion of the antibodies induced appeared to be protective (33, 38). Antibodies directed against the PorA OM protein, responsible for serosubtype specificity, have been demonstrated to provide protection against serogroup B meningococcal disease, but such protection is serosubtype specific (38). A fully effective vaccine against serogroup B meningococci would be expected to induce cross-protective immunity against heterologous strains of a wide range of serotypes and serosubtypes.
The generally accepted correlate of protection against meningococcal infection is the presence of serum bactericidal activity (SBA) against a colonizing strain (14), and individuals immunized with OMV vaccines produce an SBA associated with the presence of antibodies directed against PorA. In a recent study of meningococcal carriage among university students, the acquisition of carriage of serogroup B meningococci was accompanied by the development of SBA against the homologous colonizing strain and was correlated with the induction of antibodies to PorA. In addition, in some individuals, the presence of homologous SBA could also be associated with the development of an immune response to the PorB OM protein, responsible for serotype specificity (26). Importantly, in addition to the development of homologous SBA, lower levels of SBA against some of the other serogroup B strains isolated in the study were also observed. This heterologous SBA was not associated with the presence of antibodies directed against PorA, PorB, or the other major surface antigens, i.e., serogroup B capsule, LPS, Rmp, Opa, Opc, or pilin protein, suggesting that other, unidentified antigens contribute to the development of cross-protective immunity to serogroup B meningococci (26). The ability of the immune system to produce cross-protective immunity to meningococcal infection is also suggested by the observation that individuals who succumb to meningococcal infection are seldom reinfected even with strains of differing serotype and serosubtype.
A full understanding of the basis of the human immune response to meningococci has been hampered by a lack of detailed knowledge of the antigenic composition of meningococci. The availability of meningococcal genome sequences (50) and improvements in proteomic techniques have led to studies of the protein composition of OM and OMV preparations. The use of gel-based liquid chromatography-tandem mass spectrometry (LC-MS-MS), in which electrophoretic separation of proteins according to molecular weight is followed by LC-MS-MS, has revealed that both meningococcal OM and OMV preparations contained a larger number of proteins than have been previously detected by conventional one-dimensional (1-D) sodium dodecyl sulfate-polyacrylamide gel electrophoresis (SDS-PAGE) separation (59). OMV vaccine preparations have been revealed to be similarly complex by two-dimensional (2-D) gel electrophoresis (52, 54).
In the current study, we have utilized a resource of human serum samples and meningococcal carriage strains from longitudinal carriage studies (26 and unpublished data) combined with 2-D electrophoresis and immunoproteomic analysis to investigate the OM protein antigens associated with the development of SBA to both homologous colonizing meningococci and heterologous meningococcal serogroup B strains.
A resource of serum samples and strains obtained from first-year students during two longitudinal meningococcal carriage studies conducted at the same university was utilized for this study. All samples were obtained with informed consent according to this institution's ethical guidelines. The first investigation comprised 42 volunteers from whom serum samples, throat swabs, and saline mouth gargles were obtained on four occasions over a period of 31 weeks (26). A follow-up study conducted during the next academic year consisted of matched serum samples and carriage isolates obtained from a further 89 individuals at five time points, also over a 31-week time period. Meningococcal colonization was determined by culture and modified ctrA TaqMan PCR of throat swabs and concentrated gargle samples, as described previously (25). Serogroup B carriers were identified from the student cohorts, and serum samples taken prior and subsequent to colonization were screened for biological reactivity against serogroup B meningococcal strains and antigens.
Meningococcal strains MC168 (B:4:P1.5,2), MC169 (B:NT:P1.7-2,4), and MC172 (B:1:P1.22,14) were isolated from the first carriage study (26). N. meningitidis MC58 (B:15:P1.7,16b), originally isolated from an outbreak of meningococcal infection in Stroud, United Kingdom (31), has been characterized with respect to OM protein (59) (PorA+ PorB+ Opa+ Opc+ Rmp+) and LPS expression (58) and has been subjected to genome sequencing (50). Two further strains, MC179 (B:1:P1.19-2,13-1) and MC180 (B:1:P1.22,14), were obtained from the second carriage study (this study). All meningococci were cultured on proteose peptone agar plates incubated at 37°C in an atmosphere of 5% CO2 (vol/vol) for 16 to 18 h (59).
The OM from each strain was prepared by extraction of whole cells with lithium acetate as described previously (59).
The complement-mediated SBA assay was carried out essentially as described previously (24), using human serum with low bactericidal activity at a concentration of 5% (vol/vol) as the exogenous complement source (60). Serum bactericidal titers were expressed as the final serum dilution that resulted in a ≥50% reduction in numbers of CFU surviving compared with the results for an equivalent negative control containing heat-inactivated complement.
OM samples were subjected to 2-D gel electrophoresis. Isoelectric focusing (IEF) was conducted using 7-cm, pH 3-to-10, nonlinear Immobiline Drystrips (GE Healthcare, Uppsala, Sweden) and an IPGPhor isoelectric system (Pharmacia Biotech). The immobilized pH gradient Drystrips were rehydrated overnight in 137 μl of rehydration buffer (3). Subsequently, OM preparations (75 μg of protein) were suspended in rehydration buffer and applied to the immobilized pH gradient strip through a sample cup (GE Healthcare) placed at the anodic end of the strip. IEF was then carried out for a total of 36,750 volt-hours.
After IEF, the Drystrips were equilibrated for 15 min at room temperature in a 50 mM Tris HCl (pH 6.8) solution containing 6 M urea, 1% (wt/vol) SDS, 30% (vol/vol) glycerol, and 10 mg ml−1 dithiothreitol. A second equilibration step consisted of another 15 min at room temperature in the same solution, except that dithiothreitol was replaced with 10 mg ml−1 of iodoacetamide. Bromophenol blue was added as a tracer dye. The proteins were resolved in the second dimension by electrophoresis in 12-to-14% linear-gradient Excel gels (GE Healthcare), utilizing a Multiphor II system (Pharmacia Biotech). Prestained standards were applied to the gels to monitor electrophoresis and to assess the electrotransfer efficiency of proteins.
For each serum sample, the OM content was subjected to simultaneous 2-D electrophoresis in triplicate. One gel was stained with a ProteoSilver plus staining kit (Sigma) to visualize proteins and produce a reference gel. The two unstained 2-D gels were electroblotted onto polyvinylidene difluoride membranes at 4°C overnight with the current limited to 0.15 A. Both membranes were stained with MemCode protein stain (Perbio Science) and scanned to produce a reference map of proteins. The membrane stain was then reversed according to the manufacturer's instructions, and the membranes were then incubated with serum (1/300 dilution) taken from volunteers pre- and postcolonization. Immunoreactivity was determined as described previously (24), using nitroblue tetrazolium-5-bromide 4-chloride 3-indolil phosphate (BCIP) substrates.
Membranes were then scanned, and the profile of immunoreactive proteins matched to 2-D gel images of the same sample using PDQuest software (Bio-Rad). The signal intensities of individual antigen reactions were compared and scored semiquantitatively by three independent investigators on a scale of 0 to 5 as described previously (24). Selected protein spots that showed increases in immunoreactivity, as well as spots showing high levels of precolonization reactivity, were excised from the stained reference gel, digested in situ using trypsin (43), and subjected to MS.
LC-MS-MS was performed using a capillary high-performance LC system (Waters) online to a quantitative time-of-flight Global Ultima (Waters) mass spectrometer fitted with a nanoLockSpray source. Peptides were loaded to a trap column (0.3-mm inner diameter by 50-mm PepMap C18; LC Packings, Sunnyvale, CA) and washed with 5% (vol/vol) acetonitrile containing 0.1% (vol/vol) formic acid (buffer A) at a rate of 20 μl min−1 for 10 min. Peptides were eluted from the trap column onto a PepMap C18 analytical column (75-μm inner diameter by 15 mm; LC Packings), and separation of peptides achieved by using a gradient of 0-to-85% buffer B (95% [vol/vol] acetonitrile containing 0.1% [vol/vol] formic acid) over 60 min at a flow rate of 200 nl min−1. MS-MS data were acquired from 300 to 1,700 m/z, with the switching criteria for MS to MS-MS including ion intensity and charge state. ProteinLynx 2.1 (Waters) was used to convert the raw files into .pkl text files for database searching.
MS-MS data were searched against a protein translation of both the MC58 genome and the NCBI nonredundant database in a FASTA format using MASCOT (Matrix Science, London, United Kingdom). The following parameters were used: parent mass tolerance was 150 ppm, fragment mass tolerance was 0.25 Da, carbamidomethylation was set as a fixed modification, oxidation of methionine as a variable modification, and a maximum of one missed cleavage was allowed. The significance threshold for search results was set at a P value of <0.05 (indicates identity or extensive homology).
The predicted physicochemical parameters of proteins identified in this study were derived using the protein sequences extracted from the MC58 genomic database located at http://cmr.jcvi.org/cgi-bin/CMR/GenomePage.cgi?org=gnm. Protein hydrophobicity was determined using the ProtParam program (28) (http://www.expasy.ch/tools/protparam.html), which gives a grand average hydropathy (GRAVY) value calculated as the sum of hydropathy values of all the amino acids, divided by the number of residues in the sequence. Theoretical isoelectric pH (pI) and molecular weight measurements were calculated using MASCOT software. The subcellular locations of proteins were predicted using PSORTb version 2.0 (13) (http://psort.org/).
Sera and homologous carriage strains from individuals showing an increase in SBA following colonization with serogroup B meningococci were identified for further analysis. Sera from three such individuals (subjects 15, 37, and 38) were identified from the first study (26). For the second study, matched sera and carriage isolates were obtained from 89 individuals at five time points over a period of 31 weeks. Carriage of serogroup B meningococci was detected in 13 students, 5 of whom became colonized during the study. Sera from two of these five students (24 and 61) showed significant increases in bactericidal activity after detection of carriage compared to the bactericidal activity in sera from the earlier time points (Table (Table1).1). The remaining three carriers had high serum bactericidal titers on entry into the study, the levels of which did not increase following detection of colonization. Sera from the selected five individuals (15, 37, 38, 24, and 61) were also tested against heterologous strain MC58, chosen because its genome sequence was available. All showed a significant increase in SBA against this heterologous strain (Table (Table11).
The protein profiles of OM from individual strains were compared by 2-D electrophoresis of the OM preparations. Silver staining of gels revealed approximately 80 protein spots with pI values ranging from 3.5 to 10.0 and molecular weights from 25 to 100 kDa. The protein profile of strain MC58 OM is shown in Fig. Fig.1.1. Similar patterns of resolved protein spots were achieved with only minor variations between strains with regard to relative concentration, charge, and molecular weight of some proteins (data not shown).
The detection of immunoreactivity by Western blotting on membranes which had been previously stained for protein, recorded, and then destained enabled immunoreactive spots from pre- and postcolonization sera to be readily compared and then matched to the stained reference gel by comparison of electrophoretic position, size, and shape (Fig. (Fig.1).1). Figure Figure22 shows representative immunoblots of paired sera tested for immunoreactivity against the homologous and heterologous strains. Meningococcal colonization was always associated with increased immunoreactivity on the Western blots, which was reflected both in a greater number of spots and increased reactivity of existing spots. Immunoreactivity was detected over a wide range of pI values and molecular weights, with a total of 27 immunoreactive spots identified on blots across the six OM preparations.
Initially, pre- and postcolonization sera were compared against the individuals’ homologous carriage strain(s), which revealed that an increase in SBA following meningococcal colonization was always associated with increased immunoreactivity. Figure Figure2A2A shows the developing immune response to meningococcal colonization of student 38 against his homologous carriage strain. This student acquired meningococcal strain MC168 3 weeks into the study. Serum obtained prior to colonization showed four immunoreactive spots. The increased immunoreactivity observed after colonization was evident from the detection of four additional spots and the increased intensity of one spot. Similar differences between pre- and postcolonization sera were seen with each of the paired sera tested, although the position of spots varied between individuals (not shown).
The pre- and postcolonization sera were then tested against heterologous strain MC58. Figure Figure2B2B shows reactivities of sera from student 38 against the heterologous meningococcal strain, MC58. Serum reactivity was high (11 spots) prior to colonization but, as before, the number (17) and intensity of spots increased after colonization. Figure Figure2C2C shows the results for sera obtained from another student (student 15) reacted against heterologous strain MC58. Comparison with the results for the sera from student 38 demonstrates the heterogeneity of the human response to the same heterologous strain in terms of the position and intensity of the spots detected. Serum reactivities were lower for this individual's sera than for those of student 38 both before and after carriage detection. Five spots were detected prior to colonization, which then stimulated a cross-reactive immune response to a further seven antigens and increased the intensity of three of the original five spots (Fig. (Fig.2C2C).
The reactivity of all sera with homologous and heterologous strains produced spots which could be mapped to the original stained protein gels. The signal intensities of individual antigen reactions on immunoblots were compared and scored semiquantitatively on a scale of 0 to 5. A summary of these results is presented in Fig. Fig.3.3. A total of 23 immunoreactive spots were detected across the five homologous strains, of which 22 increased in signal intensity after carriage detection. The distribution of the 22 spots varied across the strains (Fig. (Fig.3),3), with seven antigen spots (designated 9, 10, 11, 12, 15, 16, and 18) present in at least three of the meningococcal isolates. Only spot 9 showed a consistent rise in immunoreactivity across all OM blots after meningococcal colonization. On reaction with heterologous strain MC58, 24 immunoreactive spots were detected, of which 21 increased significantly in intensity in at least one individual after colonization. In addition, three immunoreactive spots (24, 25, and 27) were uniquely detected in blots with MC58 OM. Figure Figure44 shows a representative 2-D gel with the positions of the selected protein spots marked. The spots were excised from the original stained gels, digested with trypsin, and submitted to MS for identification.
MS analysis of gel spots excised from 2-D gels of the six OM preparations produced a total of 1,326 peptide assignments, which resulted in the identification of 97 nonredundant proteins by one or more peptides. Of these, 43 proteins were identified by more than three peptides and showed a 95% (P < 0.05) probability of identity with the MASCOT search engine (Table (Table2).2). The proteins identified encompassed pI values ranging from 4.6 to 9.9, molecular weights from 18 to 201 kDa, and hydrophobicity values (GRAVY) from −0.858 to 0.161 (see the table in the supplemental material). Five proteins were predicted to be localized in the OM, 7 in the periplasm, and 16 in the cytoplasmic cellular compartment. PSORTb was unable to make a prediction for the cellular location of 15 proteins, and no proteins were predicted as being located in the inner membrane.
The Opa adhesion protein was absent from the 2-D Western blots. Since this protein is not annotated in the translation of the MC58 genome (41), the tryptic peptides obtained in our study were also searched against a translation of all six frames of the genome sequence (59), but the Opa protein was not detected.
In a previous study (26), we used 1-D SDS-PAGE followed by Western blotting to investigate serum antibody reactivity to meningococci and demonstrated the presence of cross-reactive SBA in response to colonization. The identification of such immunodominant and cross-protective meningococcal OM antigens should provide valuable data for the development of serogroup B meningococcal vaccines. However, in the previous study, the antigens responsible for stimulating these cross-reactive antibodies could not be identified due to the limited power of resolution of 1-D SDS-PAGE. Therefore, for the current study, we have taken advantage of the availability of matched sera from longitudinal studies of colonization, using a combination of 2-D PAGE, Western blotting, and MS to analyze serum bactericidal antibody responses. Previous studies have reported on the use of 2-D immunoblotting to identify antibody subsets present during infection of humans with Helicobacter pylori (18), Leishmania donovani (11), and Candida albicans (36), as well as cattle infected with Neospora caninum (44). In addition, Lopez and colleagues (30) have reported the use of the technique to investigate bovine responses to vaccination with OM preparations of Anaplasma marginale. Recently, Abel et al. (1) used 2-D PAGE to investigate the potential for proteins of N. meningitidis and Neisseria lactamica to induce immunologically cross-reactive antibodies in mice. Mendum et al. (32) have used 2-D immunoblotting of whole meningococcal cells to identify the antigens recognized by antibodies in serum from patients recovering from meningococcal infection but were unable to relate the immunoproteome data to SBA, the correlate of protection against meningococcal disease. To our knowledge, the current study is the first to investigate the immunoproteome associated with the development of SBA following meningococcal colonization of humans.
Meningococcal OM preparations were used in our study based on the rationale that subcapsular antigens present in the OM are most likely to be under constant immune surveillance and are also able to stimulate the production of functional antibodies. However, membrane proteins are usually hydrophobic, possess an alkaline pI and/or are expressed in low copy numbers (15, 40), and these properties can lead to their underrepresentation on 2-D gels. Nevertheless, as shown in Fig. Fig.11 and the table in the supplemental material, we largely overcame these potential technical limitations through the concentration of low-abundance proteins, efficient solubilization of hydrophobic proteins, and resolution of basic proteins. Identification of immunogens was achieved by LC-MS-MS of protein spots on 2-D gels corresponding to immunoreactive spots on replicate blots. However, the analysis of the results was complicated by the presence of multiple proteins identified with robust probability per spot, a consequence of the sensitive detection methods which has been previously reported with OM preparations of Anaplasma marginale (30). In addition, a comparison between the 2-D gel and matching Western blots showed that not all immunoreactive spots had a corresponding protein spot, demonstrating that some immunogenic proteins were not expressed in sufficient concentration to be detected on the silver-stained gels, and so the spots could not be matched for excision and identification. However, such low-abundance proteins are unlikely to be effective vaccine candidates. The observation that immunodetection on Western blots can be more sensitive than silver staining of the homologous 2-D gel is consistent with similar observations from 2-D PAGE studies with other pathogens (11, 36).
Analysis of 27 immunoreactive spots across the six meningococcal strains resulted in the identification of 43 proteins potentially associated with the development of SBA against both homologous and heterologous strains. Individuals showed differing patterns of reactivity on colonization, with an increase in SBA being associated with increases in the number of spots detected before and after colonization and/or with increases in the intensity of individual spots. As expected, the PorA OM antigen was the most ubiquitous antigen, present across the gels in multiple charge and mass variants. This observation is in accord with the results of other 2-D PAGE studies of meningococci (1, 3, 52, 54) and reflects both the abundance and immunogenicity of this porin in the OM. It is possible that posttranslational modification of PorA accounts for the presence of isoforms of the same molecular weight but differing pI values. In our study, colonization resulted in significant increases in PorA (spot 9) immunoreactivity not only on all homologous Western blots but also in some cases with the heterologous MC58 PorA. Homologous and heterologous antibodies to the other major OM porin PorB (spot 11) were also detected following colonization. The patterns of immunoreactivity to both porins observed in the current study support the results of previous studies that reported a correlation between bactericidal activity and an increase in antiporin antibodies in response to both meningococcal infection (17, 58) and colonization (23, 26). This result is also in accord with the observation that a subset of antibodies directed against PorA and PorB show heterologous as well as serosubtype specificity (58).
In addition to these major porins, increases in immunoreactivity following meningococcal colonization were also observed to other proteins predicted to be associated with the OM, namely a putative cell-binding factor, NMB0345 (50); an immunoglobulin A-specific serine endopeptidase; and PilE, the pilin subunit protein. However, reactivity toward several OM antigens that are being considered as potential vaccine candidates, such as Opa (7); NspA (19); NadA (21); and TbpA, the transferrin binding protein A (57), was absent from our analyses. Our previous analysis of MC58 has shown that OM preparations contain low or undetectable levels of these three proteins (59). The absence of Opa reactivity is consistent with the results of our previous studies that demonstrated no association between Opa-reactive antibodies and bactericidal activity induced by both meningococcal colonization (26) and infection (58). In addition, the stringent criteria of identifying proteins by at least three peptide matches, as applied in our analyses to increase the confidence of obtaining a robust protein match, resulted in the exclusion of the Opc and PilQ OM proteins. Although the Opc protein induces bactericidal antibodies as a component of experimental OM vaccines (39), interstrain variation and its absence from many disease isolates (2) probably preclude its use as a vaccine component antigen (22). In contrast, PilQ is a more promising vaccine candidate since it is an abundant, conserved OM protein that can induce bactericidal antibodies in adult humans (12).
Colonization was also associated with increases in homologous and heterologous immunoreactivity to proteins identified as being located in the meningococcal periplasm. These included three ABC transporter proteins; a thiol-disulfide interchange protein, DsbC; Lip (H.8); and a hypothetical protein, NMB0928. Some of these periplasmic proteins are promising vaccine candidates. The ABC transporter protein NMB0041 (48) has been reported to be upregulated upon meningococcal interaction with human epithelial cells, suggesting that an immune response to this protein on colonization may be important at the mucosal surface (16). The second ABC transporter, NMB0634 (FbpA), is a virulence factor with a role in iron uptake (42); however, the role for the third ABC transporter, NMB1612, is unknown despite its presence in OMV and OM preparations (59). The other hypothetical protein associated with the periplasm, NMB0928, has been reported to be present as a constituent of OMV derived from N. meningitidis (52, 53, 55, 59) and N. lactamica (53) and also to be conserved across many meningococcal serogroups and capable of inducing murine bactericidal antibodies (8). The thiol-disulfide interchange protein DsbC is potentially interesting as a vaccine candidate due to function, as it catalyzes the formation of disulfide bonds essential for the conformational stability and folding of proteins, and recently, the DsbC-encoding open reading frame has been reported to become upregulated in meningococci during infection of HeLa cells (47).
The remainder of the immunoreactive spots were associated with proteins predicted to be located in the cytoplasm or of unknown location. It is generally accepted that meningococcal OM preparations should be composed of proteins located in the OM, with perhaps some proteins from the periplasmic subcellular compartments, but that cytoplasmic proteins should be absent. However, several proteome studies have shown that the meningococcal OM and OMV are complex structures containing a large number of proteins (9, 53, 55, 59). It is possible that a proportion of the proteins not predicted to be located in the meningococcal OM may be present as experimentally induced contaminants. Alternatively, the serum antibody reactivity to these proteins observed on Western blots may be the result of the disruption of meningococcal cells by complement-mediated bacteriolysis (14) occurring following meningococcal colonization. Although the presence of cytoplasmic proteins in OM preparations may be disputed, several proteomic studies have shown them to be present in OM preparations from other gram-negative bacteria, including those from Escherichia coli (29), Legionella pneumophila (27), N. lactamica (53), and N. meningitidis (9, 37, 59). Indeed, a fluorescence-activated cell sorting analysis of whole meningococcal cells using polyclonal antibodies against corresponding recombinant proteins has shown the presence of predicted cytoplasmic proteins on the external surface of the OM (9). Thus, protein localization by PSORTb can be ambiguous (13), and the finding of immunogenic proteins of apparently cytoplasmic origin should not be dismissed. Indeed, in the current study, the most dominant cytoplasmic protein detected was the 60-kDa chaperonin, previously shown to be surface located (9), as well as metabolic enzymes, such as alcohol dehydrogenases, kinases, and dehydratases and synthetases associated with transcription and translation processes. Immunoreactivity to several other proteins with no assigned location was also detected, and these proteins included additional metabolic enzymes, some lipoproteins, and a trigger factor protein.
In summary, the current study is the first, to our knowledge, to use 2-D PAGE, Western blotting, and MS to analyze the development of natural immunity in adult volunteers following colonization with meningococci. Colonization was associated with the induction of antibodies to a number of meningococcal proteins, and there was considerable variation between individuals in their immune responses. Notably, antibodies to several meningococcal proteins showed significant cross-reactivity to heterologous strains. Such proteins represent potential targets for the development of effective vaccines against serogroup B meningococcal infection.
This work was supported by Meningitis UK and Wessex Medical Research.
Editor: J. N. Weiser
Published ahead of print on 8 September 2009.
†Supplemental material for this article may be found at http://iai.asm.org/.