Immuno-proteomic analysis of human immune responses to experimental Neisseria meningitidis outer membrane vesicle vaccines identifies potential cross-reactive antigens

Vaccine 32 (2014) 1280–1286

Contents lists available at ScienceDirect

Vaccine journal homepage: www.elsevier.com/locate/vaccine

Immuno-proteomic analysis of human immune responses to experimental Neisseria meningitidis outer membrane vesicle vaccines identifies potential cross-reactive antigens Jeannette N. Williams a , Vincent Weynants b , Jan T. Poolman b,1 , John E. Heckels a , Myron Christodoulides a,∗ a Neisseria Research, Molecular Microbiology, Division of Clinical and Experimental Sciences, Sir Henry Wellcome Laboratories, University of Southampton Medical School, Southampton SO166YD, United Kingdom b GlaxoSmithKline Vaccines, Rixensart, Belgium

a r t i c l e

i n f o

Article history: Received 26 July 2013 Received in revised form 25 November 2013 Accepted 19 December 2013 Available online 30 January 2014 Keywords: Neisseria meningitidis Outer membrane vesicle Immuno-proteomics Vaccine

a b s t r a c t Human volunteers were vaccinated with experimental Neisseria meningitidis serogroup B vaccines based on strain H44/76 detoxified L3 lipooligosaccharide (LOS)-derived outer membrane vesicles (OMV) or the licensed Cuban vaccine, VA-MENGOC-BC. Some volunteers were able to elicit cross-bactericidal antibodies against heterologous L2-LOS strain (760676). An immuno-proteomic approach was used to identify potential targets of these cross-bactericidal antibodies using an L2-LOS derived OMV preparation. A total of nine immuno-reactive spots were detected in this proteome: individuals vaccinated with the detoxified OMVs showed an increase in post-vaccination serum reactivity with Spots 2–8, but not with Spots 1 and 9. Vaccination with VA-MENGOC-BC induced sera that showed increased reactivity with all of the protein spots. Vaccinees showed increases in serum bactericidal activity (SBA) against the heterologous L2-LOS expressing strain 760676, which correlated, in general, with immunoblot reactivity. The identities of proteins within the immuno-reactive spots were determined. These included not only wellstudied antigens such as Rmp, Opa, PorB and FbpA (NMB0634), but also identified novel antigens such as exopolyphosphatase (NMB1467) and ␥-glutamyltranspeptidase (NMB1057) enzymes and a putative cell binding factor (NMB0345) protein. Investigating the biological properties of such novel antigens may provide candidates for the development of second generation meningococcal vaccines. © 2014 Elsevier Ltd. All rights reserved.

1. Introduction Neisseria meningitidis (meningococcus) infections are still of considerable concern worldwide, leading to significant mortality and morbidity [1]. Capsular polysaccharide (CPS)-conjugate vaccines have proven success in virtually eliminating disease caused by serogroup C meningococci and serogroup A, Y and W CPSconjugate vaccines are expected to deliver a similar outcome in those countries where they are introduced [2–4]. The development of meningococcal serogroup B (MenB) vaccines has proved more intractable, due to the poor immunogenicity of the B capsule and its molecular mimicry of human fetal NCAM that raises concerns over inducing auto-immune responses [5]. Studies on the sub-capsular outer membrane (OM) have led to the development and use of OM vesicle (OMV) vaccines, which are based on detergent extraction

∗ Corresponding author. Tel.: +44 2380798896. E-mail address: [email protected] (M. Christodoulides). 1 Present address: Crucell, Leiden, The Netherlands. 0264-410X/$ – see front matter © 2014 Elsevier Ltd. All rights reserved. http://dx.doi.org/10.1016/j.vaccine.2013.12.070

of OM to reduce lipopolysaccharide content and reactogenicity [6]. Such vaccines have been successful in controlling clonal outbreaks of MenB disease, e.g. in Cuba [7], Brazil [8] and more recently in New Zealand, through the use of a strain-specific vaccine MeNZB (NZ98/254, P1.7–2,4, ST41/44) [9]. MeNZB is also a component of Bexsero® , which has been recently licensed as a MenB vaccine by the European Union [10,11]. Inclusion of MenZB is believed to provide an immuno-adjuvant effect for the defined recombinant antigens – Neisseria heparin binding antigen (NHBA, genome-derived Neisseria antigen (GNA) 2132), factor H binding protein (fHbp or lipoprotein (LP)2086, GNA1870), and Neisseria adhesin A (NadA, GNA1994) [12] – and this adjuvant effect may be related to the lipid vesicular structure itself and the presence of immunomodulatory OM antigens. MeNZB is included also to provide additional protection against ST41/44 clonal complex strains, particularly since low level expression of fHbp protein on the surfaces of meningococcal strains of clonal complex 41/44 has been reported [13–15], suggesting the possibility that such strains may be more difficult to kill with a vaccine response towards one of the recombinant proteins within

J.N. Williams et al. / Vaccine 32 (2014) 1280–1286

Bexsero® . The proteome of the meningococcal OM/OMV is complex, containing >200 proteins [16–20] and the hypervariable PorA protein is the immunodominant antigen that generates serum bactericidal antibodies, the generally accepted correlate of protection [21]. It is also possible that immunomodulatory components of the OM/OMV synergistically contribute to the stimulation of bactericidal antibodies. In the current study, we used an immuno-proteomics approach to investigate pre- and post-vaccination sera from human volunteers vaccinated with experimental meningococcal serogroup B vaccines [22]. The trial assessed the immunogenicity of five formulations of detoxified L3 derived lipooligosaccharide OMV (TrL3 or L7 OMV vaccines) produced in strain H44/76 background alongside the licensed Cuban VA-MENGOC-BC vaccine [7]. The modified LOS vaccines lacked PorA and tended to be less immunogenic than VAMENGOC-BC, which contains PorA (P1.15,19). Regardless, paired sera were identified from individuals vaccinated with TrL3 or L7 OMV that showed significant increases in post-vaccination serum bactericidal activity (SBA), which could be used in the immunoproteomic study. The trial also highlighted that the response of individuals to VA-MENGOC-BC, which expresses the L3 LOS immunotype, was high towards a heterologous L2 LOS strain (760676). Since no cross-reactivity has been observed between the L2 and L3 immunotypes [23], the authors suggested that the crossprotective responses towards the L2 strain could be directed against as yet uncharacterized minor OM proteins. Thus, our immunoproteomic study relates increases in serum reactivity with the proteome of the heterologous L2-LOS derived OMV with increases in post-vaccination SBA against the corresponding L2 strain, in order to potentially identify those minor OM antigens with crossreactivity.

1281

other antigens in these OMV vaccines was identical, as previously reported [22,24]. The VA-MENGOC-BC vaccine is not genetically modified and contains wild-type OMVs of serogroup B strain CU385 (B:4:P1.19,15; L3,7,9 ST32 complex) produced by a detergent extraction method, in combination with serogroup C polysaccharide [7]. 2.2. Heterologous OMV preparation for immuno-proteomics A heterologous L2-LOS OMV preparation was made from N. meningitidis strain 760676 (B:2a:P1.5,2:L2 ST11 complex) by sodium deoxycholate (NaDOC, 0.1% w/v) extraction, as described previously [22]. The strain was genetically modified [22] to produce blebs that lacked PorA and capsule and contained a non-sialylated LOS, thanks, respectively, to porA, siaD, msbB and lst gene deletion. The strain was also grown in the presence of iron (to avoid expression of iron binding proteins like FrpB and TbpA); the rationale for this was to concentrate the immuno-proteomic search on the probably minor cross-reactive, and possibly more conserved, antigens without focusing on the responses to the more variable iron-regulated proteins. The choice of L2-LOS derived heterologous OMV was also made to reduce the possibility of having immunereactive spots linked to a LOS response and thus to focus on the anti-protein response. 2.3. Serum bactericidal activity (SBA) Bactericidal activity of pre- and post-vaccination sera was determined against the homologous H44/76 and heterologous 760676 strains using the GlaxoSmithKline Vaccines Laboratory standard SBA-MenB assay with human complement, as described previously [25].

2. Materials and methods 2.1. OMV vaccines and serum samples A total of 26 paired pre- and post-vaccination serum samples were available from young adults vaccinated in an open, randomized phase I trial with experimental N. meningitidis serogroup B (MenB) vaccines obtained from strain H44/76 (B:15:P1.7,16:L3,7,9 ST32 complex) detoxified L3 derived lipooligosaccharide (LOS) OMV (TrL3 or L7 OMV): specifically, 8 paired samples were from subjects receiving TrL3 OMVs and 18 paired samples from subjects receiving L7 OMVs [22]. An additional 9 paired sera were from individuals vaccinated with the VA-MENGO-BC vaccine (Finlay Institute, Cuba) [7]. Subjects were selected based on their preimmunization versus post-immunization dose III SBA titers and selecting for the highest vaccine responders. The experimental TrL3 and L7 OMV vaccines differ only in the size of the LOS ␣-chain and they were obtained by inactivation of the lst and lgtB genes, respectively [22,24]. They also do not contain either PorA or FrpB protein, which were both removed by genedeletion. The rationale for PorA and FrpB deletion was to improve the immunogenicity of minor OM proteins, including Hsf (also known as NhhA, neisserial autotransporter) and Tbp (transferrin binding protein)A, which were over-expressed in the OM by genetic manipulation and growth under iron depletion, respectively [22]. The relative content of Hsf and TbpA were estimated against PorB content using SDS-PAGE and were found to be present in the OM at ∼5% and ∼15% of the PorB content, respectively [22]. In addition, data from a previous study [23] have shown that in the absence of up-regulation and using classical culture, i.e. no iron chelation, the level of Hsf and TbpA were so low that they could not be detected by SDS-PAGE. Based on one-dimensional SDS-PAGE, the expression of

2.4. 2-Dimensional (2D) gel electrophoresis and western blot assay L2-LOS OMV samples were subjected to 2D gel electrophoresis in triplicate as described previously [26]. For each serum sample examined, one 2D gel (reference gel) was stained with ProteoSilver plus staining kit (Sigma) to visualize proteins. The two unstained 2-D gels were electroblotted onto polyvinylidene difluoride membranes, stained with MemCode protein stain (Perbio Science) and the blots scanned to produce a reference map of proteins. The membrane stain was then reversed, following the manufacturer’s instructions, and the membranes incubated with individual serum (1/300 dilution) samples and immunoreactivity detected as described previously [26]. Membranes were then scanned, and the profile of immuno-reactive proteins matched to 2D gel images of the same sample using PDQuest software (BioRad). The signal intensities of individual antigen reactions were compared and scored semi-quantitatively by three independent investigators on a scale of 0 to 5, as described previously [26]. The profiles of immuno-reactive proteins were matched to 2D gel images of the same sample and selected protein spots were excised from the stained 2D gels, digested in-situ with trypsin [27] and subjected to mass spectrometry fingerprinting. 2.5. Mass spectrometry (MS) and data processing LC–MS–MS was done as described previously [20] and MS–MS data were searched against a protein translation of both the MC58 genome and the NCBI non-redundant database in a FASTA format using MASCOT (Matrix Science, London, U.K.). The significance

1282

J.N. Williams et al. / Vaccine 32 (2014) 1280–1286

threshold for search results was set at a P value of 0.05, which indicates identity or extensive homology. 2.6. In silico characterization of proteins The predicted physicochemical parameters of proteins identified in this study were derived using the protein sequences extracted from the MC58 genomic database (http://cmr.jcvi.org/ cgi-bin/CMR/GenomePage.cgi?org/gnm). Protein hydrophobicity (GRAVY) values were determined using the ProtParam program (http://www.expasy.ch/tools/protparam.html) and theoretical isoelectric pH (pI) and molecular weight measurements were calculated using MASCOT software (data not shown). PSORTb version 2.0 (http://psort.org/) was used to predict the cellular locations of proteins. 3. Results and discussion In this study, we used immuno-proteomics as a method to identify potentially cross-reactive antigens in meningococcal OMV. This was done by examining the reactivity of paired sera, taken from young adult volunteers before and after immunization with experimental detoxified L3 derived LOS OMV vaccines (either TrL3 (n = 8 subjects) or L7 (n = 18 subjects) LOS) based on strain H44/76, or VAMENGOC-BC vaccine (n = 9 subjects), against a heterologous L2-LOS OMV preparation. The vaccine given to each numbered individual is shown in Supplementary Table 1. The blots for each serum sample (pre- and post-vaccination) were analyzed for spots that showed increases in immuno-reactivity as well as spots showing prevaccination reactivity. Fig. 1 shows a representative 2D gel with the positions of serum-reactive protein spots identified from the heterologous L2-LOS derived 760676 OMV preparation. Surprisingly, despite using biochemical methods to improve protein solubility and gel resolution [26], the 2D electrophoretic pattern of this NaDOC L2-LOS OMV preparation (Fig. 1) showed a lower number of protein spots when compared to the proteomes reported elsewhere for meningococcal OM prepared from either LPS-expressing and LPS-deficient strains or from NaDOC-extracted OMV, including the Cuban vaccine [19,26,28]. There are several limitations to the study that predominantly revolve around the use of 2D gel electrophoresis. It is possible that proteins could be under-represented in the NaDOC L2-LOS derived OMV because of their hydrophobic nature and alkaline isoelectric pH [29,30]. 2D gel electrophoresis also poorly resolves very basic and high molecular weight proteins and integral OM proteins with several transmembrane-spanning regions [31]. The technology can also suffer from dynamic-range effects, which are caused by the presence of a limited number of highly abundant individual proteins within an OM complex [32]. Thus, presentation of OM proteins in dynamically low copy number [33] could impact on the immuno-reactive proteome. In addition, differences in the purification method used, e.g. whether employing detergent or no detergent-extraction, ultracentifugation versus ultrafiltration methods, have been reported to influence protein content and OMV composition [34]. It is also possible that serum functional antibodies directed towards conformational epitopes do not react with denatured proteins on 2D gels. A total of nine human serum immuno-reactive spots were detected by sera from individuals vaccinated with either the TrL3 or L7-LOS derived OMV preparations or VA-MENGOC-BC. The selected immuno-reactive spots were excised from the 2D electrophoretic patterns and analyzed by mass spectrometry to determine identity (Table 1). After vaccination with the TrL3 or L7 OMVs, Spots 2–8 increased in signal intensity, whereas no immuno-reactivity was observed for Spots 1 and 9 (Fig. 2A). The highest increases, in

Fig. 1. Gel positions of proteins in heterologous L2-LOS derived OMV preparation selected for mass spectrometry analysis. The figure shows a representative (of n > 30) 2D SDS-PAGE gel of the heterologous L2-LOS derived OMV preparation (strain 760676). Proteins were separated according to their pI by using an immobilized pH gradient (3–10 nonlinear, NL) in the first dimension and by Mr in the second dimension (SDS-PAGE gels, 12–14% acrylamide). Molecular size markers are indicated on the left (in kilodaltons). Proteins were visualized by silver staining and the immuno-reactive areas selected for mass spectrometry fingerprinting are indicated by numbered circles.

terms of number of vaccinees responding and spot intensity, were observed with Spots 3, 5, 7 and 8, whereas fewer vaccinee sera showed increased reactivity with Spots 2, 4 and 6. The possibility should be noted that the observed increased responses to Spots 3, 5, 7 and 8, compared to Spots 2, 4 and 6 could be related to protein abundance. Nevertheless, 10 individuals showed geometric mean (GM) post-vaccination increases in intensity for Spot 3 of +1 unit (95% confidence limits (CL) +1, +2); 11 individuals recognized Spot 5 (GM increases in intensity +2 units (95%CL +2, +3); 9 individuals recognized Spot 7 (GM increases in intensity +1.5 (95%CL +1, +2) and 18 individuals recognized Spot 8 (GM increases in intensity +1.5 (95%CL +1, +2). By contrast, only 3 individuals showed increases in reactivity of +1/+2 with Spots 2 and 4 and only 2 individuals showed increases of +3 against Spot 6. In addition to reactivity with Spots 2–8, vaccination with VA-MENGOC-BC induced antibodies that could react with Spots 1 and 9 (Fig. 2B). Although the number of vaccinees receiving VA-MENGOC-BC was smaller, the highest increases in post-vaccination reactivity (in intensity and number of positive individuals) were observed against Spots 3 and 8, with the other protein spots showing lower responses (in number of positive individuals or spot intensity). For Spot 3, six individuals showed GM increases in spot intensity +3 (95% CL +2, +4) and for Spot 8, seven individuals showed GM increases +1.5 (95% CL, +1, +2.5). For particular individuals, there was some reactivity of pre-vaccination sera with L2-LOS derived OMV antigens, and this reactivity did not

J.N. Williams et al. / Vaccine 32 (2014) 1280–1286

1283

Fig. 2. Reactivity of pre- and post-immunization sera with heterologous L2-LOS derived OMV preparation blot. Reactivity of sera from young adults vaccinated with (A) TrL3 or L7 OMVs or (B) VA-MENGOC-BC. The number on the left-hand side of each individual graph represents the study subject number. The x axis shows the immunoreactive spots (as numbered on Fig. 1) and the y axis the signal intensities of individual antigen reactions scored semi-quantitatively on a scale of 0 to 5 [26]. The stacked bars represent the intensity of blot immunoreactivity of each spot: open bars denote pre-immunization reactivity values, the cross-hatched bars denote no change in reactivity (i.e. pre-immunization reactivity value = post-immunization reactivity value), the black bars above the open bars denote increases in post-pre immunization reactivity values and where the black bar is below the white bar this denotes a reduction from the pre-immunization reactivity value to the post-immunization reactivity value.

1284

J.N. Williams et al. / Vaccine 32 (2014) 1280–1286

Table 1 Immuno-proteome of detoxified L2-LOS derived OMV (strain 760676) identified from 2D gel spots. Protein spots (Fig. 1) on gels corresponding to immunoreactivity on blots were excised and identified by mass spectrometry fingerprinting. Spot no.

NMB

Protein name

Peptides

1

1057 0928 1594

23 15 8

2

1057 2039 0546 1429 0382 0382

Gamma glutamyltranspeptidase Hypothetical protein NMB0928 Spermidine putrescine ABC transporter periplasmic spermidine putrescine binding protein Gamma glutamyltranspeptidase Major outer membrane protein PorB Alcohol dehydrogenase Outer membrane protein PorA Rmp protein Rmp protein Opacity related protein POPM1 OS Neisseria meningitidis Serogroup C GN opr PE 3 SV 1 Opacity related protein POPM1 OS Neisseria meningitidis Serogroup C GN opr PE 3 SV 1 Hypothetical protein NMB2050 Major outer membrane protein PorB Outer membrane protein PorA Iron III ABC transporter periplasmic binding protein Exopolyphosphatase Histidinol phosphate aminotransferase Putative cell binding factor 30S ribosomal protein S2 Outer membrane protein NspA Hypothetical protein NMB1164 Hypothetical protein NMB1126 30S ribosomal protein S2 Iron III ABC transporter periplasmic binding protein Outer membrane protein NspA Thiol disulfide interchange protein DsbC

3 4 5 6 7

8

9

2050 2039 1429 0634 1467 1582 0345 2101 0663 1164 1126 2101 0634 0663 0550

6 5 4 1 24 4 5 8 14 12 9 8 8 4 24 3 2 8 8 3 3 2 2

Note: The protein name is from the MC58 genome annotation located at www.tigr.org/ and the NMB number corresponds to annotations in the MC58 genome database. Protein identifications are shown in terms of the number of peptide matches to data predicted from protein sequence information.

increase following vaccination, remaining unchanged or reduced (Fig. 2A and B). The SBA of pre- and post-vaccination sera was assessed against the heterologous strain 760676 (Table 2) and there was some correlation between the measured increases in SBA after vaccination and protein spot intensity. SBA titres ≥4, quantified using human complement bactericidal assays are generally considered to correlate with protection [21]. Serum from individuals vaccinated with TrL3 or L7 OMVs that showed the highest pre-to post-vaccination increases in SBA, i.e. individuals 61 and 176, were reactive with multiple protein spots. Moreover, sera showing high pre-vaccination SBA titres ≥4, e.g. from individuals 101, 145, 173, 187, 212 and 239, also showed some reactivity with protein spots. By contrast, sera from individuals showing no SBA reactivity (<2), e.g. 133, 169 and 210, showed little or no reactivity with any protein spots. However, the correlation between SBA and immuno-proteome spot development for some of the TrL3 or L7 OMV vaccinees was not perfect, e.g. sera from individuals 137, 182, 192, 201 and 245 were non-bactericidal, yet showed some reactivity with protein spots (Table 2; Fig. 2A). By contrast, individuals vaccinated with VA-MENGOC-BC, all showed increases in SBA post-vaccination, and all vaccinee sera showed reactivity with heterologous protein spots (Table 2; Fig. 2B). Thus, our data suggest that quantitative serum reactivity with protein spots, correlating with increased SBA titres, appears to be influenced by the nature of the vaccine used. It is possible that the superior performance of VAMENGOC-BC is due to the presence of PorA antigen, which is absent in the TrL3 or L7 OMV vaccines. Moreover, for those vaccinees with no detectable SBA, it is possible that the observed immunoreactivity with protein spots correlates with other measures of protection, such as antibody opsonophagocytosis or T-cell mediated responses. For comparison, the SBA of pre- and post-vaccination sera against the homologous H44/76 strain is shown in Supplementary Table 1 and, with few exceptions, the observed increases

in SBA against the heterologous strain (Table 2) are matched by increases in post-vaccination SBA against the homologous strain. Several of the protein spots contained well-characterized meningococcal antigens. Spot 3 and 4 was identified as Rmp protein, and these two spots are likely to represent different reduction modifiable forms of the protein. Notably,10 individuals vaccinated with TrL3 or L7 OMV vaccines showed increased reactivity (+1 to +3) to Spot 3 Rmp, whereas only two individuals showed increased reactivity (+1, +2) towards Spot 4 Rmp. In addition, seven individuals vaccinated with VA-MENGOC-BC showed increases (+2 to +4) in reactivity towards Spot 3 Rmp, compared with 4 individuals showing reactivity (+1 to +4) with Spot 4 Rmp (Fig. 2). Antibodies to Rmp (formerly the Class 4 OM) protein have been shown to inhibit the bactericidal activity of both normal human sera and a PorA monoclonal antibody towards meningococci [35]. However, there was no evidence that vaccination with a Norwegian meningococcal OMV vaccine containing Class 4 protein induced such blocking antibodies [36]. Spot 5 and 6 were identified as Opa-related protein, after searching against a translation of all six frames of the MC58 genome sequence [20]. These spots represent two phases of Opa denaturation following SDS-PAGE, with the denatured form of Spot 5 being the most abundant, which could account for the greater serum immunological reactivity against Spot 5 compared to Spot 6. Opa proteins are abundant adhesins in the OM and antibodies to Opa are observed after natural meningococcal infection or vaccination, albeit with wide variation in antibody levels [37]. Recently, a study of the ability of 14 different recombinant Opa proteins to induce bactericidal antibodies suggested that a combination of six particular Opa proteins would provide theoretical meningococcal strain coverage of 90% in the UK [38]. Possible reservations to the inclusion of Opa in meningococcal vaccines are the observed significant phase and antigenic variation of Opa and evidence that these proteins can suppress T-cell immune activation [39,40].

J.N. Williams et al. / Vaccine 32 (2014) 1280–1286 Table 2 Serum bactericidal activity (SBA) of antisera of individuals vaccinated with either TrL3 or L7 OMV vaccines or VA-MENGOC-BC against the heterologous strain. Vaccine used for immunization

Subject number

TrL3 or L7 OMVs

61 78 90 99 101 123 133 137 145 165 169 173 176 182 187 188 192 195 201 205 207 210 212 221 239 245 22 67 69 103 146 156 179 190 211

VA-MENGOC-BC

760676 SBA titre Pre

Post

3 14 <2 <2 54 IR <2 <2 85 21 <2 8 <2 <2 8 <2 <2 <2 <2 <2 <2 <2 17 <2 4 <2 <2 5 5 44 5 <2 3 <2 38

184 57 <2 4 40 3 <2 2 92 14 <2 8 80 2 14 9 2 3 <2 3 4 <2 9 13 5 <2 22 56 27 104 65 55 21 32 63

Note: Pre-vaccination and post-vaccination sera were tested for bactericidal activity against heterologous strain 760676 using human serum as an exogenous source of complement. The numbers denote the reciprocal SBA titre, defined as the dilution that showed >50% killing of meningococci. For reference, the exact TrL3 or L7 OMV used to vaccinate each individual and the SBA responses towards the homologous strain H44/76 are shown in Supplementary Table 1.

However, several protein spots contained meningococcal antigens that have not been investigated, to our knowledge, as potential vaccine antigens. The immuno-reactivity observed against Spot 8 was likely due to the presence of antibodies directed against the putative cell binding factor (NMB0345), based on the high number of peptide matches. In contrast, the protein associated with immuno-reactivity to Spot 7 was ambiguous, with hypothetical protein NMB2050, PorB, iron III ABC transporter periplasmic binding protein (NMB0634, FbpA, ferric binding protein) and an exopolyphosphatase (NMB1467) showing similar peptide match numbers (between 8 and 14). Specific PorA reactivity within Spot 7 cannot be excluded and could arise from previous exposure and natural immunity. The nature of the hypothetical protein NMB2050 has not been determined. The proteins associated with Spots 1 and 9 are also ambiguous (Table 1): Spot 1 contains the ␥-glutamyltranspeptidase (GGT) enzyme, an uncharacterized hypothetical protein (NMB0928) and another ABC transporter (involved in polyamine transport of spermidine/putrescine), whereas Spot 9 contains predominantly two uncharacterized hypothetical proteins (NMB1164 and NMB1126). Of note, a previous study of sera taken from individuals following colonization with meningococci, also showed reactivity against NMB0345 and NMB0634 proteins [26]. The ability of PorB and meningococcal iron-regulated OM proteins, such as the transferrinbinding proteins, to induce SBA has been reported, but these

1285

antigens have serious limitations as vaccine candidates capable of inducing cross-protection [41–44]. Meningococci grown in the presence of haemoglobin show increases in NMB0345 gene transcription [45] and the ability of this antigen to induce SBA in animal models is being evaluated (corresponding authors’ unpublished work). A recombinant FbpA protein has been reported to induce murine bactericidal antibodies using a variety of different adjuvants, but the levels of killing were low and without significant cross-strain protection [46]. To our knowledge, the exopolyphosphatase NMB1467 protein, which is required for cleavage of cellular polyphosphate, has not been investigated for its ability to induce SBA, although the enzyme appears to be required for resistance against complement-mediated lysis [47,48]. Expression of the GGT enzyme has a potential role in virulence, since it has been shown to provide an advantage for meningococcal growth in cerebrospinal fluid [49], but its ability to induce SBA is unknown. In summary, the present study represents an initial, exploratory investigation using immuno-proteomics to identify antigens with potential cross-reactivity, which could form the basis for more in-depth studies with other meningococcal vaccines. A limited collection of OM antigens were identified that showed increases in heterologous post-vaccination immuno-reactivity. Although the ability of several of these identified antigens to induce bactericidal antibodies has already been investigated, current studies are focusing on the biology of some of these new proteins, including the putative cell binding factor. Investigating the biological properties of such antigens may provide candidates for inclusion in new meningococcal vaccines. Acknowledgements This study was funded by GlaxoSmithKline Biologicals S.A., who covered all costs. MC and JEH are named inventors on meningococcal vaccine patents owned by the University of Southampton and GlaxoSmithKline. VW and JTP are, or were at the time of the study, employees of the GlaxoSmithKline group of companies. JTP and VW are designated inventors on patents owned by GlaxoSmithKline. VW owns shares and options to shares in GlaxoSmithKline. We are grateful to Dr Paul Skipp, Centre for Proteomic Research, Southampton, for mass spectrometry and data processing and analyses. Appendix A. Supplementary data Supplementary data associated with this article can be found, in the online version, at http://dx.doi.org/10.1016/j.vaccine. 2013.12.070. References [1] Rouphael NG, Stephens DS. Neisseria meningitidis: biology, microbiology, and epidemiology. Methods Mol Biol 2012;799:1–20. [2] Tan LK, Carlone GM, Borrow R. Current concepts. Advances in the development of vaccines against Neisseria meningitidis. N Engl J Med 2010;362(16):1511–20. [3] Snape MD, Perrett KP, Ford KJ, John TM, Pace D, Yu LM, Langley JM, McNeil S, Dull PM, Ceddia F, Anemona A, Halperin SA, Dobson S, Pollard AJ. Immunogenicity of a tetravalent meningococcal glycoconjugate vaccine in infants: a randomized controlled trial. JAMA 2008;299(2):173–84. [4] Frasch CE, Preziosi MP, Laforce FM. Development of a group A meningococcal conjugate vaccine, MenAfriVac(TM). Hum Vaccines Immunother 2012;8(6):715–24. [5] Finne J, Leinonen M, Makela PH. Antigenic similarities between brain components and bacteria causing meningitis. Implications for vaccine development and pathogenesis. Lancet 1983;(ii):355–7. [6] Holst J, Martin D, Arnold R, Huergo CC, Oster P, O’Hallahan J, Rosenqvist E. Properties and clinical performance of vaccines containing outer membrane vesicles from Neisseria meningitidis. Vaccine 2009;27(Suppl. 2):B3–12. [7] Sierra GVG, Campa HC, Varcacel NM, Garcia IL, Izquierdo PL, Sotolongo PF, Casanueva GV, Rico CO, Rodriguez CR, Terry MH. Vaccine against group B Neisseria meningitidis: protection trial and mass vaccination results in Cuba. NIPH Ann 1991;14:195–210.

1286

J.N. Williams et al. / Vaccine 32 (2014) 1280–1286

[8] Demoraes JC, Perkins BA, Camargo MCC, Hidalgo NTR, Barbosa HA, Sacchi CT, Gral IML, Gattas VL, Vasconcelos HD, Plikaytis BD, Wenger JD, Broome CV. Protective efficacy of a serogroup B meningococcal vaccine in Sao Paulo, Brazil. Lancet 1992;340:1074–8. [9] Oster P, Lennon D, O’Hallahan J, Mulholland K, Reid S, Martin D. MeNZB: a safe and highly immunogenic tailor-made vaccine against the New Zealand Neisseria meningitidis serogroup B disease epidemic strain. Vaccine 2005;23(17–18):2191–6. [10] Giuliani MM, du-Bobie J, Comanducci M, Arico B, Savino S, Santini L, Brunelli B, Bambini S, Biolchi A, Capecchi B, Cartocci E, Ciucchi L, Di Marcello F, Ferlicca F, Galli B, Luzzi E, Masignani V, Serruto D, Veggi D, Contorni M, Morandi M, Bartalesi A, Cinotti V, Mannucci D, Titta F, Ovidit E, Welsch JA, Granoff D, Rappuoli R, Pizza M. A universal vaccine for serogroup B meningococcus. PNAS 2006;103(29):10834–9. [11] Serruto D, Bottomley MJ, Ram S, Giuliani MM, Rappuoli R. The new multicomponent vaccine against meningococcal serogroup B, 4CMenB: immunological, functional and structural characterization of the antigens. Vaccine 2012;30(Suppl 2):B87–97. [12] Gorringe AR, Pajon R. Bexsero: a multicomponent vaccine for prevention of meningococcal disease. Hum Vaccines Immunother 2012;8(2):174–83. [13] Fletcher LD, Bernfield L, Barniak V, Farley JE, Howell A, Knauf M, Ooi P, Smith RP, Weise P, Wetherell M, Xie XL, Zagursky R, Zhang Y, Zlotnick GW. Vaccine potential of the Neisseria meningitidis 2086 lipoprotein. Infect Immun 2004;72(4):2088–100. [14] Jiang HQ, Hoiseth SK, Harris SL, Mcneil LK, Zhu DZ, Tan CW, Scott AA, Alexander K, Mason K, Miller L, DaSilva I, Mack M, Zhao XJ, Pride MW, Andrew L, Murphy E, Hagen M, French R, Arora A, Jones TR, Jansen KU, Zlotnick GW, Anderson AS. Broad vaccine coverage predicted for a bivalent recombinant factor H binding protein based vaccine to prevent serogroup B meningococcal disease. Vaccine 2010;28(37):6086–93. [15] Pajon R, Beernink PT, Harrison LH, Granoff DM. Frequency of factor H-binding protein modular groups and susceptibility to cross-reactive bactericidal activity in invasive meningococcal isolates. Vaccine 2010;28(9):2122–9. [16] Ferrari G, Garaguso I, Adu-Bobie J, Doro F, Taddei AR, Biolchi A, Brunelli B, Giuliani MM, Pizza M, Norais N, Grandi G. Outer membrane vesicles from group B Neisseria meningitidis delta gna33 mutant: proteomic and immunological comparison with detergent-derived outer membrane vesicles. Proteomics 2006;6(6):1856–66. [17] Mendum TA, Newcombe J, McNeilly CL, McFadden J. Towards the immunoproteome of Neisseria meningitidis. PLoS One 2009;4(6). [18] Gil J, Betancourt LZ, Sardinas G, Yero D, Niebla O, Delgado M, Garcia D, Pajon R, Sanchez A, Gonzalez LJ, Padron G, Campa C, Sotolongo F, Barbero R, Guillen G, Herrera L, Besada V. Proteomic study via a non-gel based approach of meningococcal outer membrane vesicle vaccine obtained from strain CU385: a road map for discovering new antigens. Hum Vaccines 2009;5(5):347–56. [19] Vipond C, Suker J, Jones C, Tang C, Feavers IM, Wheeler JX. Proteomic analysis of a meningococcal outer membrane vesicle vaccine prepared from the group B strain NZ98/254. Proteomics 2006;6(11):3400–13. [20] Williams JN, Skipp PJ, Humphries HE, Christodoulides M, O’Connor CD, Heckels JE. Proteomic analysis of outer membranes and vesicles from wild-type serogroup B Neisseria meningitidis and a lipopolysaccharide-deficient mutant. Infect Immun 2007;75(3):1364–72. [21] Vermont C, van den Dobbelsteen G. Neisseria meningitidis serogroup B: laboratory correlates of protection. FEMS Immunol Med Microbiol 2002;34:89– 96. [22] Bonvehi P, Boutriau D, Casellas J, Weynants V, Feron C, Poolman J. Three doses of an experimental detoxified L3-derived lipooligosaccharide meningococcal vaccine offer good safety but low immunogenicity in healthy young adults. Clin Vaccine Immunol 2010;17(9):1460–6. [23] Weynants VE, Feron CM, Goraj KK, Bos MP, Denoel PA, Verlant VG, Tommassen J, Peak IR, Judd RC, Jennings MP, Poolman JT. Additive and synergistic bactericidal activity of antibodies directed against minor outer membrane proteins of Neisseria meningitidis. Infect Immun 2007;75(11):5434–42. [24] Weynants V, Denoel P, Devos N, Janssens D, Feron C, Goraj K, Momin P, Monnom D, Tans C, Vandercammen A, Wauters F, Poolman JT. Genetically modified L3,7 and L2 lipooligosaccharides from Neisseria meningitidis serogroup B confer a broad cross-bactericidal response. Infect Immun 2009;77(5):2084–93. [25] Boutriau D, Poolman J, Borrow R, Findlow J, Domingo JD, Puig-Barbera J, Baldo JM, Planelles V, Jubert A, Colomer J, Gil A, Levie K, Kervyn AD, Weynants V, Dominguez F, Barbera R, Sotolongo F. Immunogenicity and safety of three doses of a bivalent (B:4:p1.19,15 and B:4:p1. 7-2,4) meningococcal outer membrane vesicle vaccine in healthy adolescents. Clin Vaccine Immunol 2007;14(1):65–73. [26] Williams JN, Skipp PJ, O’Connor CD, Christodoulides M, Heckels JE. Immunoproteomic analysis of the development of natural immunity in subjects colonized by Neisseria meningitidis reveals potential vaccine candidates. Infect Immun 2009;77(11):5080–9.

[27] Shevchenko A, Jensen ON, Podtelejnikov AV, Sagliocco F, Wilm M, Vorm O, Mortensen P, Shevchenko A, Boucherie H, Mann M. Linking genome and proteome by mass spectrometry: large-scale identification of yeast proteins from two dimensional gels. PNAS 1996;93(25):14440–5. [28] Uli L, Castellanos-Serra L, Betancourt L, Dominguez F, Barbera R, Sotolongo F, Guillen G, Pajon FR. Outer membrane vesicles of the VA-MENGOC-BC vaccine against serogroup B of Neisseria meningitidis: analysis of protein components by two-dimensional gel electrophoresis and mass spectrometry. Proteomics 2006;6(11):3389–99. [29] Santoni V, Molloy M, Rabilloud T. Membrane proteins and proteomics: un amour impossible? Electrophoresis 2000;21(6):1054–70. [30] Post DM, Zhang D, Eastvold JS, Teghanemt A, Gibson BW, Weiss JP. Biochemical and functional characterization of membrane blebs purified from Neisseria meningitidis serogroup B. J Biol Chem 2005;280(46):38383–94. [31] Lee EY, Choi DS, Kim KP, Gho YS. Proteomics in gram-negative bacterial outer membrane vesicles. Mass Spectrom Rev 2008;27(6):535–55. [32] Cordwell SJ. Technologies for bacterial surface proteomics. Curr Opin Microbiol 2006;9(3):320–9. [33] Gorg A, Obermaier C, Boguth G, Harder A, Scheibe B, Wildgruber R, Weiss W. The current state of two-dimensional electrophoresis with immobilized pH gradients. Electrophoresis 2000;21(6):1037–53. [34] van de Waterbeemd B, Mommen GP, Pennings JL, Eppink MH, Wijffels RH, van der Pol LA, de Jong AP. Quantitative proteomics reveals distinct differences in the protein content of outer membrane vesicle vaccines. J Proteome Res 2013 Feb 25. [35] Munkley A, Tinsley CR, Virji M, Heckels JE. Blocking of bactericidal killing of Neisseria meningitidis by antibodies directed against class 4 outer membrane protein. Microb Pathog 1991;11:447–52. [36] Rosenqvist E, Musacchio A, Aase A, Hoiby EA, Namork E, Kolberg J, Wedege E, Delvig A, Dalseg R, Michaelsen TE, Tommassen J. Functional activities and epitope specificity of human and murine antibodies against the class 4 outer membrane protein (Rmp) of Neisseria meningitidis. Infect Immun 1999;67(3):1267–76. [37] Mandrell RE, Zollinger WD. Human immune response to meningococcal outer membrane protein epitopes after natural infection or vaccination. Infect Immun 1989;57:1590–8. [38] Callaghan MJ, Lewis S, Sadarangani M, Bailey SES, Chan H, Ferguson DJP, Derrick JP, Feavers I, Maiden MC, Pollard AJ. The potential of recombinant Opa proteins as vaccine candidates against hyperinvasive meningococci. Infect Immun 2011;79(7):2810–8. [39] Lee HS, Boulton IC, Reddin K, Wong H, Halliwell D, Mandelboim O, Gorringe AR, Gray-Owen SD. Neisserial outer membrane vesicles bind the coinhibitory receptor carcinoembryonic antigen-related cellular adhesion molecule 1 and suppress CD4+ T lymphocyte function. Infect Immun 2007;75(9): 4449–55. [40] Sadarangani M, Pollard AJ, Gray-Owen SD. Opa proteins and CEACAMs: pathways of immune engagement for pathogenic Neisseria. FEMS Microbiol Rev 2011;35(3):498–514. [41] Wedege E, Bolstad K, Wetzler LM, Guttormsen H. IgG antibody levels to meningococcal porins in patient sera: comparison of immunoblotting and ELISA measurements. J Immunol Methods 2000;244(1–2):9–15. [42] Wright JC, Williams JN, Christodoulides M, Heckels JE. Immunisation with recombinant PorB outer membrane protein induces a bactericidal immune response against Neisseria meningitidis. Infect Immun 2002;70:4028–34. [43] Rokbi B, Mignon M, Maitre-Wilmotte G, Lissolo L, Danve B, Caugant DA, Quentin-Millet MJ. Evaluation of recombinant transferrin-binding protein B variants from Neisseria meningitidis for their ability to induce cross-reactive and bactericidal antibodies against a genetically diverse collection of serogroup B strains. Infect Immun 1997;65(1):55–63. [44] Ala’aldeen DAA, Stevenson P, Griffiths E, Gorringe AR, Irons LI, Robinson A, Hyde S, Borriello SP. Immune responses in humans and animals to meningococcal transferrin-binding proteins: implications for vaccine design. Infect Immun 1994;62:2984–90. [45] Jordan PW, Saunders NJ. Host iron binding proteins acting as niche indicators for Neisseria meningitidis. PLoS One 2009;4(4):e5198. [46] Gomez JA, Criado MT, Ferreiros CM. Bactericidal activity of antibodies elicited against the Neisseria meningitidis 37-kDa ferric binding protein (FbpA) with different adjuvants. FEMS Immunol Med Microbiol 1998;20(1):79–86. [47] Tinsley CR, Gotschlich EC. Cloning and characterization of the meningococcal polyphosphate kinase gene-production of polyphosphate synthesis mutants. Infect Immun 1995;63:1624–30. [48] Zhang Q, Li Y, Tang CM. The role of the exopolyphosphatase PPX in avoidance by Neisseria meningitidis of complement-mediated killing. J Biol Chem 2010;285(44):34259–68. [49] Takahashi H, Hirose K, Watanabe H. Necessity of meningococcal gammaglutamyl aminopeptidase for Neisseria meningitidis growth in rat cerebrospinal fluid (CSF) and CSF-like medium. J Bacteriol 2004;186(1):244–7.