Genetic characteristics of Neisseria meningitidis serogroup B strains carried by adolescents living in Milan, Italy

Before a protein vaccine is introduced into a country, it is essential to evaluate its potential impact and estimate its benefits and costs. The aim of this study was to determine the genetic characteristics of Neisseria meningitidis B (NmB) in the pharyngeal secretions of 1375 healthy adolescents aged 13–19 y living in Milan, Italy, in September 2012, and the possible protection offered by the two currently available NmB protein vaccines. Ninety-one subjects were Nm carriers (6.6%), 29 (31.9%) of whom carried the NmB capsular gene. The 29 identified strains belonged to eight clonal complexes (CCs), the majority of which were in the ST-41/44/Lin.3 CC (n = 11; 37.9%). All of the identified strains harboured ƒHbp alleles representing a total of 15 sub-variants: the gene for NHBA protein was found in all but three of the studied strains (10.3%) with 13 identified sub-variants. There were 15 porA sub-types, seven of which were identified in just one CC. The findings of this study seem to suggest that both of the protein vaccines proposed for the prevention of invasive disease due to NmB (the 4-protein and the 2-protein products) have a composition that can evoke a theoretically effective antibody response against the meningococcal strains currently carried by adolescents living in Northern Italy. The genetic characteristics of NmB strains can be easily evaluated by means of molecular methods, the results of which can provide an albeit approximate estimate of the degree of protection theoretically provided by the available vaccines, and the possible future need to change their composition.


Introduction
Neisseria meningitidis group B (NmB) is a bacterial pathogen that has two characteristics: it is one of the major causes of invasive meningococcal disease (IMD), 1 and the only one of the most virulent meningococcal serogroups for which polysaccharidebased vaccines cannot be prepared.
The NmB capsular polysaccharide is poorly immunogenic even when conjugated with a carrier protein 2 and, because its structure is identical to some components of human cells, may induce auto-immunity. 3,4 This meant that alternative vaccines had to be developed. The first were based on outer membrane vescicles and had the PorA protein as their immunodominant antigen. 5 However, when it was demonstrated that their highly subtype-specific immunity against PorA led to limited coverage against the more diverse NmB populations, 6 an approach based on reverse vaccinology was adopted. 7 The availability of whole genome sequences of NmB led to the discovery of a number of relatively well-conserved and potentially cross-reactive surface proteins capable of evoking antibodies that were effective in killing bacteria in vitro and preventing infections in experimental animals. 8 Of these, factor H-binding protein (fHBP), neisserial adhesion A (NadA) and neisserial heparin-binding antigen (NHBA) were considered the best for preparing effective NmB vaccines, although all of these components of NmB can be found in a number of variants with different immune characteristics.
However, as no data could be collected concerning the real efficacy of these protein vaccines in preventing human IMD because of its low frequency, various indirect methods were developed. Particular attention was paid to the importance of the genetic diversity of NmB strains in conditioning vaccine efficacy, the extent to which the NmB antigens cross-reacted with those present in the vaccine, the level of expression of each antigen on the cell surface, and the bactericidal activity of the antibodies evoked by the vaccines against NmB strains expressing different protein variants. All of these studies clearly showed that vaccines containing the above-mentioned proteins (with or without PorA) Before a protein vaccine is introduced into a country, it is essential to evaluate its potential impact and estimate its benefits and costs. The aim of this study was to determine the genetic characteristics of Neisseria meningitidis B (NmB) in the pharyngeal secretions of 1375 healthy adolescents aged 13-19 y living in Milan, Italy, in september 2012, and the possible protection offered by the two currently available NmB protein vaccines. Ninety-one subjects were Nm carriers (6.6%), 29 (31.9%) of whom carried the NmB capsular gene. The 29 identified strains belonged to eight clonal complexes (ccs), the majority of which were in the sT-41/44/Lin.3 cc (n = 11; 37.9%). all of the identified strains harboured ƒHbp alleles representing a total of 15 sub-variants: the gene for NHBa protein was found in all but three of the studied strains (10.3%) with 13 identified sub-variants. There were 15 porA sub-types, seven of which were identified in just one cc. The findings of this study seem to suggest that both of the protein vaccines proposed for the prevention of invasive disease due to NmB (the 4-protein and the 2-protein products) have a composition that can evoke a theoretically effective antibody response against the meningococcal strains currently carried by adolescents living in Northern Italy. The genetic characteristics of NmB strains can be easily evaluated by means of molecular methods, the results of which can provide an albeit approximate estimate of the degree of protection theoretically provided by the available vaccines, and the possible future need to change their composition.

Genetic characteristics of Neisseria meningitidis serogroup B strains carried by adolescents living in Milan, Italy
could be effective in preventing IMD due to NmB provided that they elicited antibodies capable of overcoming possible differences in the immunogenic characteristics of the protein variants expressed by NmB strains. 9 As the specificity of antibodies induced by the proteins in each vaccine depends on their amino acid sequences, and as bactericidal titers decrease as the genetic distance between the variants increases, it was presumed that a vaccine based on a single protein could fail in a number of cases because it could not cover the numerous variants with different immune properties. On the basis of these premises, two vaccines have now been prepared: one with four components (4CMenB: ƒHbp 1.1, nadA, nhba subvariant 2, porA P1.7-2.4 based on the NZ98/254 isolate), and the other with two (2MB: ƒHbp 1.55 and ƒHbp 3.45). Both preparations are immunogenic, and adequately safe and tolerable, [10][11][12] but they are at different clinical stages of development because 4CMenB was approved in Europe in December 2012 and is expected to be on the market in late 2013, whereas significantly less clinical data are available for 2MB.
However, it is not easy to evaluate the potential impact of these vaccines or follow up possible variations in the characteristics of circulating NmB strains capable of modifying vaccine efficacy. One possible surrogate method is to analyze the genetic characteristics of the NmB strains, and evaluate the degree of identity between the proteins included in each vaccine and those theoretically expressed by the genes of the circulating bacteria. Although this has the limitation of not evaluating protein expression, it can provide an albeit approximate estimate of the theoretical strain coverage of a vaccine.
The aim of this study was to determine the genetic characteristics of the NmB strains carried in the pharyngeal secretions of healthy adolescents living in Milan, Italy, in September 2012. The primary objective was to evaluate the degree of identity between the proteins included in the two NmB vaccines and those encoded by the corresponding genes identified in the carried meningococcal strains.

Strain panel composition: Multilocus sequencing typing (MLST) clonal complex (CC)
The Gene analysis ƒHbp Table 1 shows the distribution of the ƒHbp variants by CC. Among the six sub-variants in variant 1, the gene for the ƒHbp sub-variant included in the 4MB vaccine (sub-variant 1) was identified in only one case, and no strain harboured the sub-variant included in the 2MB vaccine (sub-variant 55). The gene for subvariant 1.14 was present in seven strains, whereas those for proteins 1.13, 1.15, 1.66, and 1.322 were found in only one strain each. Table 2 summarizes the data concerning the theoretical percentage of amino acid identity between the proteins encoded by the ƒHbp genes identified in the carried NmB strains and the fHBP proteins included in the 4MB and 2 MB vaccines. The amino acid identity of the six sub-variants in variant 1 was not less than 87% for 4MB,  The gene for nhba protein was found in all but three strains (10.3%), and 13 sub-variants were identified. The gene for the nhba protein included in the 4MB vaccine (sub-variant 2) was found in eight strains (100% amino acid identity), all of which had the gene for ƒHbp variant 1; the amino acid identity between the other sub-variants and the nhba protein in 4MB ranged from 64.7% to 90.9% ( Table 3). Of the three strains without nhba, one had the gene for fHbp protein variant 1 (subvariant 15), and two the gene for fHbp protein variant 2 (subvariant 24).
PorA Table 4 shows the distribution of the porA sub-types by CC. PorA proteins were identified in all of the strains: there were 15 porA sub-types, seven of which were identified in just one CC. The most common sub-types were P1.7.2,4 (n = 6, 20.  The amino acid identity between the studied PorA proteins and the protein included in the 4CMenB vaccine (P1.7.2,4) was not less than 81.56% (Table 3). NadA The nucleotide sequence of NadA protein was not detected in any of the 29 strains.

Discussion
Carriers are the only reservoir of meningococci, and adolescents are not only among the most frequent carriers of Nm, but also at high risk of developing IMD. 13 It has also been shown that the NmB strains carried by adolescents have similar genetic characteristics to those that cause IMD in infants. 14 The periodic monitoring of Nm carriage by adolescents therefore seems to be important in order to be able to identify the groups of Nm that can cause IMD in a given period and geographical area, and suggest the best preventive measures. Furthermore, in the case of NmB, monitoring the genetic characteristics of bacteria can also be a useful means of approximately evaluating the potential strain coverage of the recently developed vaccines.
Our data suggest that only a small proportion of the adolescents attending high schools in Milan during 2012 carried nasopharyngeal bacteria with the specific NmB capsular gene. The significantly lower carriage rate than that found in studies of adolescents in countries where NmB is relatively common 15,16 may be explained by the progressive reduction in the circulation of NmB strains in Italy, which is demonstrated by the decline in the incidence of NmB-induced IMD in Lombardy (the region in which the study was performed) and Italy as a whole since 2009. 17 We only evaluated the potential ability of protein vaccines to detect NmB carrier status, and did not measure the expression of proteins or the bactericidal activity of the antibodies evoked by the vaccines. However, previous studies of strains in other geographic areas have clearly indicated the CCs that are mainly associated with bacterial hyper-virulence, and the degree of crossreactivity between the variants of the proteins encoded by the genes of the different strains. [18][19][20][21] Consequently, genetic analysis of the pathogens may be sufficient to estimate their theoretical degree of virulence and the potential strain coverage offered by the NmB vaccines.
As the serogroup-specific PCR used in this study detects the capsule gene but cannot predict capsule expression, it is not possible to say whether all of the studied strains were encapsulated and therefore potentially highly virulent. However, MLST indicated that a considerable number may have been hypervirulent because more than 40% of the studied strains fell into the ST-41/44 and ST-32 CCs, two of the profiles most frequently associated with IMD worldwide. 22,23 The genetic characteristics of the studied strains seem to indicate that both of the protein vaccines coming onto the market could theoretically evoke the production of antibodies against most of the carried strains. The 4CMenB vaccine contains a single variant of the fHbp protein in the group 1 variant (sub-variant 1.1), together with an NadA-3 component, the sub-variant 2 of nhba protein, and the PorA protein (subtype P1.7-2,4). As frequently occurs, 24 the gene for NadA protein was not identified in the studied bacterial population, and so the NadA component in the 4CMenB vaccine may not play a role in eliminating carried strains. The gene encoding ƒHbp variant 1 was found in 12 of the studied strains, but it could promote the production of the variant included in the vaccine in only one case. However, most of the remaining variant 1 genes (mainly variants 1.14 and 1.15) were among those encoding for proteins capable of causing at least a 4-fold increase in serum bactericidal activity after three vaccine doses in adults, although not always in infants. 20 These results suggest that the fHbp component of In the case of strains containing the group 2 and 3 variants not covered by 4CMenB, the vaccine may offer protection as a result of the antibodies induced by the nhba and PorA proteins. All but three of the studied strains harboured the nhba gene and, given that a high degree of cross-reactivity has been found among the antibodies evoked by different variants of this protein, 20 it can be imagined that antibodies against this protein could extend the protection offered by 4CMenB to all of the studied strains with the exception of three cases. However, one of these strains had group 1 ƒHbp, and two ƒHbp of groups 2 and 3, and would theoretically be protected by the presence of the porA genes encoding proteins that are very similar to those included in 4CMenB. The vaccine based on two antigens (2MB) contains a recombinant lipidated fHBP from groups 1 and 3 (variant 1.55 and 3.45) 14 and, as the antibodies evoked by the proteins included in each family variant cross-react with most of the other members of the same family, 20 their inclusion is considered a guarantee of good protection. 14 All of the studied strains harboured ƒHbp alleles, but only three had the ƒHbp gene encoding a variant that is identical to one of the two included in the vaccine. However, the strains harbouring ƒHbp genes for different variants had marginal polymorphisms, which suggests the production of proteins that are very similar to those included in the vaccine. This seems to indicate that the 2MB vaccine is capable of evoking protective antibodies against most, if not all of the variants in the studied strains. Furthermore, Jiang et al. measured serum bactericidal activity against a different panel of clinical isolates of sera drawn from adults fully immunised with the 2MB vaccine, and found that 36 of the 45 tested strains were killed. 25 In conclusion, our findings suggest that both of the vaccines proposed for the prevention of NmB-induced IMD have a composition that is theoretically effective in evoking an antibody response to the meningococcal strains carried by adolescents currently living in Northern Italy. The genetic properties of NmB strains can be easily characterized by means of molecular methods, the results of which can provide an albeit approximate estimate of the degree of protection theoretically provided by the available vaccines, and the possible future need to change their composition. This information could be even more relevant if the proteins encoded by the individual genes were adequately measured, which could be done relatively simply and should be the subject of further investigations aimed at establishing the real strain coverage of NmB vaccines. No data are currently available concerning on the real impact of NmB vaccines on carriage, but they are urgently needed in order to make it possible to evaluate the possible role of NmB vaccines in conditioning carrier status and the risk of developing IMD. Studies regarding the induction of a mucosal immune response following NmB vaccination and those of herd immunity might significantly improve our knowledge in this regard. 26

Subjects and sample collection
The samples of pharyngeal secretions were collected from adolescents aged 13-19 y attending two high schools in Milan, Italy, during the second and third week of September 2012. Enrolment in the study was completely voluntary, but the students were given a brochure during lesson time and, in the week preceding the swabbing, science teachers reinforced the message by providing detailed explanations of the pathogenicity of NmB and the related diseases. Written informed consent was obtained from all of the participants and from one of the parents of those aged < 18 y. The swabbing was performed in the medical rooms of the two schools at the end of teaching on two consecutive days by a group of experienced pediatric nurses supervised by a pediatrician (NP) using ESwab kits (code 482CE, Copan Italia). The swabs were immediately transported to a central laboratory and processed within two hours.
Sample preparation Immediately after their arrival at the laboratory, the swabs were shaken, broken off, inverted, returned to the tube, and centrifuged for 15 min at 15 000 × g in order to draw out the trapped fluid. The swab was then discarded, and ESwab liquid (250 μL) was thawed, boiled for 10 min, cooled on ice, and then centrifuged for 10 min at 12 500 × g. The supernatant was transferred to a fresh tube and used immediately for EasyMag Automatic DNA Extraction (Biomerieux SA) (elution in 40 μL), and then stored at -80 °C until required.  Bacterial identification and real-time PCR-based serogrouping Table 5 shows the sequences of all of the primers and probes used for N. meningitidis detection and subsequent serogrouping. The primer/probe combinations detect the capsule transfer ctrA gene of N. meningitidis and the sialytransferase gene (siaD) of NmB strains (ctrA forward 1, ctrA reverse 1, and ctrA TaqMan probe 1; group B forward, group B reverse, and group B TaqMan probe). 27,28 The amplification reaction mixtures contained 1 μM of primers, 0.5 μM of probe, 3 mM of MgCl 2 , 200 μM of dATP, dCTP, dGTP, and dUTP, 0.025 U of AmpliTaq Gold/μL, 0.01 U of AmpErase UNG/μL (all supplied with the TaqMan core reagent kit, Applied Biosystems), and 10 μL of target DNA in a total volume of 25 μL. Dilutions of home-made plasmid containing the cloned ctrA or siaD gene were included in every experiment. A modified ctrA primer-probe combination (ctrA forward 2, ctrA reverse 2, and ctrA TaqMan probe 2), which detects serogroups A, B, C, X, Y, Z, W135, 29E, and some non-groupable strains, was also used. 27 These reaction mixtures contained 1 × TaqMan Universal Mix, 0.3 μM of each primer, 0.2 μM of probe (all supplied by Applied Biosystems), and 10 μL of target DNA in a total volume of 25 μL. The specificities of both assays were confirmed using extracts of Neisseria gonorrhoeae, Neisseria lactamica, Neisseria sicca, Neisseria flavescens and Neisseria cinerea (obtained from the National Collection of Type Cultures, Central Public Health Laboratory), and their presence was excluded by means of real-time PCR using specific primers (data not shown). The amplification parameters for both primer/ probe combinations consisted of 2 min at 50 °C and 10 min at 95 °C, followed by 50 cycles of 15s at 95 °C and 1 min at 60 °C, using a 7900HT sequence detector system (Applied Biosystems). Every experiment included positive controls (containing cloned ctrA or DNA extracted from group B N. meningitidis) and negative controls (water alone). The threshold cycle (the cycle at which sample fluorescence exceeds the threshold value indicating a positive result, which is proportional to the number of genome copies present) was considered for each sample.

MLST, sequence alignments and phylogenetic analyses
The strains of NmB were characterized by means of the MLST of seven housekeeping genes (abcZ, adk, aroE, fumC, gdh, pdhC, and pgm), which is considered to be the standard for strain characterization and epidemiological surveillance. MLST was performed using the method of Maiden et al., 29 and the NmB strains were assigned to clonal complexes (CCs), in accordance with the Neisseria MLST website (http://pubmlst. org/neisseria). The sequencing reactions were performed in PCR tubes using the BigDye terminator cycle sequencing kit (Applied Biosystems), and subsequently analyzed using an ABI PRISM 377 automated DNA sequencer (Applied Biosystems). BioEdit software (version 6.0.8.0) and manual adjustment were used to identify the multiple sequence alignments of the nucleotide sequences encoding the mature proteins and the deduced amino acid sequences of the mature proteins. 30,31 P-distance neighborjoining dendrograms and amino acid identity were evaluated using the MEGA software package (version 5.0), 32,33 in which the positions containing alignment gaps and missing data were eliminated by making pairwise sequence comparisons (the pairwise deletion option). The sequence bootstrap method with 500 repetitions was used to test branch stability and tree topology. After multiple nucleotide alignments of the antigen genes and phylogenetic reconstructions, the variants were grouped into main variants, each corresponding to major clades. The remaining phylogenetic analyses were made using the eBURST program (version 3, developed and hosted at Imperial College [http://eburst.mlst.net/v3/ instructions/8.asp]), 32 in which the minimum number of identical loci for group definition was set at ≥ 5.
Genotype analysis of investigational NmB protein vaccines There are various nomenclatures for vaccine protein sub-variants and their corresponding alleles. 34,35 The nomenclature for the ƒHbp and PorA sub-variants used in this study follows that of the public fhbp and PorA database (http://neisseria.org), in which new allelic sub-variants are assigned a sequentially allocated numerical identifier and a pre-existing or new (sequentially allocated) numerical protein identifier: i.e., ƒHbp 1.15 refers to Novartis variant 1, Neisseria.org protein sub-variant 15. The nhba sub-variant has not been assigned because of the absence of universal standardisation of allele sub-variants in accordance with the public database neisseria.org.
The PorA genes were sequenced following the protocols of Sacchi et al. 36 and Clarke et al. 37 The PorA variable region (VR) types were described on the basis of the nomenclature provided in the N. meningitidis PorA VR database (ht tp : //pubm lst.org /neisseria / PorA/), and reported by De Filippis et al. 38 The ƒHbp, nhba, and nadA genes were sequenced using the protocols and nomenclature provided in the N. meningitidis pubMLST database  (http://pubmlst.org/neisseria), and described by Lucidarme et al. 18,39

Disclosure of Potential Conflicts of Interest
No potential conflicts of interest were disclosed.