Post-genomics of Neisseria meningitidis: an update

Abstract

The following article is a correction to a previously published version: Bernardini G, Braconi D, Lusini P, Santucci A. Postgenomics of Neisseria meningitidis: an update. Expert Rev. Proteomics 6(2), 135–143 (2009). These corrections were made owing to concerns being raised regarding similarity between sections of the text with previously published works.

For clarity, the corrected article is published in full below. The sections in bold text correspond to the corrected sections and are therefore different to the previously published version.

Neisseria meningitidis infection still remains a major life-threatening bacterial disease worldwide. The availability of bacterial genomic sequences generated a paradigm shift in microbiological and vaccines sciences, and post-genomics (comparative genomics, functional genomics, proteomics and a combination/evolution of these techniques) played important roles in elucidating bacterial biological complexity and pathogenic traits, at the same time accelerating the development of therapeutic drugs and vaccines. This article summarizes the most recent technological and scientific advances in meningococcal biology and pathogenesis aimed at the development and characterization of vaccines against the pathogenic meningococci.

Keywords::

• DNA microarray
• meningococcus
• proteomics
• systems biology
• vaccine development

The Gram-negative bacterium Neisseria meningitidis is a strictly human pathogen asymptomatically colonizing the nasopharynx of approximately 10% of healthy individuals. Occasionally, however, hyperinvasive meningococcal clones penetrate the mucosal membrane and enter the bloodstream, causing endemic cases, local outbreaks or even epidemics, which remain a major public health problem throughout the world, with associated high levels of morbidity and mortality. Although effective polysaccharide vaccines for capsular groups A, C, Y and W-135 are available, a vaccine that elicits a protective response to all group B (Men B) strains has yet to be developed. However, a number of methods to generate a universal vaccine that will induce a cross-reactive response are currently in progress[1].

The introduction of vaccination more than 200 years ago prevented illness and death for millions of individuals. Notwithstanding the enormous progress in socioeconomic and health conditions in developed countries, nowadays vaccination is still the best strategy for treating, preventing and trying to overcome infectious diseases, as well as the development of microbial resistance.

Reporting of complete genome sequences for a variety of bacteria, which is still ongoing, have fuelled rapid developments in microbial post-genomics. At the beginning of 2009, according to the Genome OnLine Database (GOLD [101]), 916 completed genomes have been published and there are 2430 ongoing (incomplete) sequencing projects. The success of genome sequence efforts made it possible for system-levels characterization to be established in molecular biology and biochemistry through the development of post-genomics technologies.

The advent of various post-genomics technologies, high-throughput ‘omics’ technologies and bioinformatics have had a major impact on all areas of biological research. These techniques, increasingly applied in laboratories, also offer tremendous opportunities in vaccines research; they can accelerate the rapid generation of a set of potential vaccine candidates that would normally be produced by a microbial pathogen in restricted quantities and potentially overlooked or under-represented using traditional antigen discovery techniques.

The development of post-genomic approaches, whose real power arose from the combination of comparative genomics, molecular biology, transcriptomics, proteomics, biophysics and metabolomics, has accelerated the discovery of protein functional information and the selection of virulence factors and potential antigen candidates, in particular for those pathogens that are still waiting for an effective vaccine to be available.

There have been impressive efforts in the biology, epidemiology and pathogenesis of Neisseria spp. by means of genomics, transcriptional profiling and proteomics (for detailed reviews, see [2,3]).

In this article, we will present an update of the most recent advances in meningococcal functional genomics studies focused on increasing the knowledge of the bacterial biological process, virulence factor discovery and vaccine development.

Neisserial comparative genome hybridization

Comparative genomics arises from the improved in silico analytical tools acquired thanks to the increasing number of available genomic sequences. This approach allows the rapid detection of intra- and inter-species differences and/or similarities, addressing important evolutionary, drug-resistant and pathogenic issues, with implications for vaccine discovery. Unlike traditional methods, changes regarding the acquisition, combination or deletion of genes can be monitored in an accurate and high-throughput fashion.

The genome of most microbial species is a dynamic entity, in which a phenomenon such as genetic horizontal transfer induces a genetic variability often connected to virulent or pathogenic behavior of many disease-related bacterial strains. In particular, the analysis of this genetic variability can detect the presence of DNA tracts in pathogens that are absent in closely related nonpathogenic strains, and that encode for a set of genes potentially responsible for bacterial colonization, invasion and infection in the human host[4]. From this point of view, detection and sequence analysis of genes acquired by genetic horizontal transfer has practical implications in vaccine discovery and therapeutic intervention, with their protein products being suitable candidates as protein vaccines or as vectors for vaccine delivery. Moreover, disease-related gene sequence analysis can disclose important mechanisms involved in the evolution of bacterial pathogenesis, such as phase variation[5], gene duplication[6]and tendency to loss of rudimental functions[7].

The comparative genomic hybridization (CGH) approach, by means of DNA microarray technology, is a powerful tool to investigate genome diversity and relatedness of bacteria, viruses and parasites, circumventing the need for multiple closely related genome sequences. CGH technology uses microarrays composed of DNA from a sequenced reference microorganism, which is compared with genomes of unsequenced isolates (test DNA) by detecting genes that are conserved between the test and reference DNA.

However, CGH has intrinsic technical limitations, including that detection is restricted to the DNA spotted on the array, and it is unable to detect acquisition events with respect to the reference strain; moreover, this technique would also fail to detect genomic rearrangements that may affect the expression levels or minor polymorphisms that would result in truncated or lack of protein. On the other hand, CGH analyses permit a more accurate evaluation of the genetic stability of a great number of bacterial pathogens [8]. Typically, such experiments produce large datasets that are difficult for biologists to handle. Only recently, Carter and colleagues [9] and Budinska and coworkers [10] developed a simple and robust CGH microarray data analysis process to overcome the bottleneck persisting in accurate processing and mathematical analysis of data.

Microarray-based CGH (mCGH) approaches were applied to pathogenic and nonpathogenic Neisseria spp. to assess the genetic diversity potentially associated with virulence or to characterize new meningococcal clinical isolates.

The availability of meningococcal[11–13]and other pathogenic neisserial species (GenBank accession number: AE004969) genome sequences enabled the design of a pan-Neisseria microarray to address all coding regions so far identified in the pathogenic Neisseria spp.[14].

The pan-Neisseria microarray-v2 has been specifically designed for use in comparative studies examining expression changes among different strains, and for the CGH assessment of the gene complements of collections of neisserial strains. The 2845 probes included make this the most comprehensive microarray currently available. In detail, it contains probes to the genes from the three meningococcal sequenced genomes (MenA strain Z2491, MenB strain MC58 and MenC FAM18), and from the gonococcal genome (strain FA1090), together with probes to the genes from the gonococcal genetic island (strain MS11), from minimal mobile elements and to other neisserial genes from the GenBank/EMBL database.

In a paper by Snyder and Saunders, the chromosomal DNA of 13 unrelated strains of Neisseria lactamica was investigated for positive genome hybridization to the pan-Neisseria microarray-v2 to assess the virulence-associated genetic content present in this nonpathogenic species[15]. The authors highlighted the high similarity of the genetic repertoire between the neisserial pathogenic and nonpathogenic species (75% of the pan-Neisseria microarray-v2 probes and 93% of the probes common to all of the four pathogenic Neisseria spp. hybridize in at least one strain of N. lactamica). Among these, 67% of the so-called ‘virulence genes’ are present. The authors suggested that the virulence of the pathogenic species does not depend on the genetic content alone but rather on a ‘genetic personality’ (i.e., different combinations of known and as yet unknown virulence genes, altered functions or expression levels of virulence genes).

With the aim to analyze the genome composition and genetic background of serogroup C meningococci in China, Peng and colleagues[16]assembled a whole-genome microarray of N. meningitidis isolate 053442 and analyzed the genome composition differences among 81 serogroup C isolates that were isolated from 14 provinces of China during 1966–2005. The whole genome of the serogroup C clinical isolate used in this study had been also sequenced by the same authors (GenBank accession number: CP000381). The CGH results showed that the genome compositions of nearly all serogroup C isolates were similar to those of 053442, thus providing a valuable resource with which to analyze the genome composition of serogroup C meningococci in China belonging to different lineages.

The same authors also compared the MenC 053442 genome with those of three meningococci, MenA Z2491, MenB MC58 and MenC FAM18, defining a common backbone of genes that may be responsible for the colonization and spread among members of the human host[17]. In addition, they performed an extensively CGH analysis on 28 clinical isolates belonging to the the ST-4821 complex, the most predominant and hypervirulent lineage in China[17]. The meningococcal disease-associated island was detected in almost all the ST-4821 complex clinical isolates (21 from healthy carriers and eight from patients), thus confirming its important role in meningococcal pathogenesis as reported by Bille et al.[18]. Nevertheless, the absence of relevant differences among the samples from patients or from healthy carriers suggests that infection might be triggered by nonbacterial factors, such as host susceptibility or environmental factors[19].

In a study by Hotopp et al., the authors developed a microarray platform containing the predicted genes of MenB strain MC58, plus the unique genetic region of MenA strain Z2491, MenC strain FAm18 and Neisseria gonorrhoeae FA1090[20]. This platform was used to perform large CGH studies on numerous neisserial isolates (one Neisseria cinerea, two N. lactamica, two N. gonorrhoeae and 48 N. meningitidis isolates). The study represents a very useful and comprehensive example of CGH data managing and interpretation, when submitted to a stringent validation and to a proper analytical procedure, with a wide range of applications: serogroup assignment, phylogenetic classification, detection and analysis of islands of horizontally transferred DNA, virulence and pathogenicity-related factors, and characterization of metabolic pathways. The authors also highlighted that mCGH results strongly depend on the platform used, on the bacterial strains analyzed and on the data analysis approach adopted, consequently pointing out a lack of correspondence among different studies. In particular, an important drawback of the technique is that substantially divergent sequences (i.e., different allelic versions of genes) will not hybridize with the probes on the microarray. Since differences greater than at least 20% divergence lead to a lack of hybridization with the probe, only a positive hybridization can be reliably interpreted. On the other hand, a negative result can arise from a number of factors that may not be related to the absence of the gene.

Notwithstanding these technical drawbacks, the high-throughput and versatile mCGH approach offers a huge range of important applications in the discovery of virulence and pathogenic factors, such as comparative analysis across species, phylogenetic studies, insertion/deletion events analysis, the correlation between islands of horizontally transferred DNA and bacterial phenotype, and characterization of the bacterial genomic repertoire.

Transcriptomics

Although comparative genomic approaches possess the ability to reveal the molecular differences that underlie potential phenotypic discrepancies among bacterial species or strains, they are insufficient in uncovering the complex host–pathogen interactions and the underlying basis of the infectious disease. To such an aim, post-genomics technologies, such as transcriptomics and proteomics, allow us to monitor global changes in gene and protein expression in both the pathogen and host during the infectious process.

A DNA microarray for N. meningitidis has been established and has been applied to transcriptome analysis (reviewed in [21]). Between 2005 and 2008, many studies employed transcriptome analysis for the characterization of diverse aspects of N. meningitidis physiology, in particular regulation by the two-component signal transduction NMB0595/NMB0594 [22,23] and iron availability [24,25], as well as the host cell response to a live bacterial infection [26–30].

Two-component regulatory systems control gene expression in a number of bacteria in response to environmental stimuli. Such systems consist of a membrane-associated sensor kinase protein and a cytoplasmic transcriptional regulator, which, in response to external stimuli, regulate and coordinate transcriptional changes affecting many cellular biological events, including pathogenesis and virulence[31–33].

Among the numerous bacterial two-component regulatory systems, a fundamental role in virulence-associated gene regulation was found for the PhoPQ system from Salmonella spp.[34]and its functional homologs, such as the meningococcal NMB0595/NMB0594 system[31–33]. As a consequence, microarray transcription profile comparison has been frequently used to explore the meningococcal NMB0595/NMB0594 two-component regulatory system activity at the physiological level[35].

Newcombe and colleagues reported that inactivation of this two-component system, encoded by NMB0595/NMB0594, designated misR/misS, in a serogroup C meningococcal strain significantly modifies the expression of 281 genes in the mutant compared with the parental strain [22]. Among them, there are several genes involved in lipo-oligosaccharide (LOS) synthesis (these genes did not include lgtG, proposed by Tzeng et al., responsible for conferring polymyxin sensitivity on their NMB0595 mutant [23], and several virulence genes, including nspA. However, the microarray observations by Newcombe and colleagues [22] were not validated by other biochemical or genetic means, nor were the direct regulatory targets of MisR identified.

To characterize a regulon, a combination of transcriptome analyses and bioinformatics with biochemical and genetic experiments is often used. Therefore, genes and operons that are directly regulated by a transcriptional regulator can be identified. This facilitates the consistent building of models, owing to the ability to distinguish between direct and indirect regulatory effects that are generated as a consequence of primary regulatory events.

Recently, to assess the scope of MisR regulation, Tzeng and colleagues carried out transcriptional profile analyses of a wild-type MenB parent strain and the respective MenB misR mutant, with a total of 78 genes and 39 genes being up- and down-regulated in the MenB misR mutant, respectively [36]. To define the minimal regulon, real-time reverse-transcription (RT)-PCR, reporter assays and an electrophoretic mobility shift assay were performed to confirm the potential regulatory effects of MisR on a panel of 25 genes identified by microarray. The MisR/S system directly or indirectly regulates genes involved in a diverse array of functional categories, such as protein folding (dnaJ, clpB and fkpA), chaperones, metabolism (hprA), iron assimilation (bfrA, tdfH and hmbR), type I protein transport (mtr, hlyB and hlyD), and sensitivity to oxidative stress and human serum, with many implicated in meningococcal pathogenesis.

Ren et al. investigated the role of a transcription factor, namely NMB0573 (annotated AsnC), a member of the Lrp-AsnC family of regulators, by transcriptomic analysis of the wild-type and NMB0573-knockout serogroup B strain MC58 [37]. The growth phenotype of the mutant strain was compared with wild-type N. meningitidis grown in both rich (gonococcal media with supplements) and nutrient-restricted defined media (RPMI with ferric nitrate and bicarbonate). The predominant change in the NMB0573-knockout compared with the wild-type strain was a reduction of transcript abundance (74 out of 91 genes; 81%), suggesting that this protein, if acting directly, is a global regulator of gene expression and primarily a transcriptional activator. Therefore, it was proposed that NMB0573 broadly represents a neisserial equivalent of the Lrp global regulator, controlling an adaptive response to low intracellular amino acid availability.

Recent studies based on DNA array technology investigated the repertoire of host cellular responses in response to an infection with pathogenic Neisseria. cDNA arrays were used to profile gene expression in meningothelial cells in response to interaction with N. meningitidis or to secreted meningococcal proteins [38,39]. In addition, they were used in epithelial cells to compare responses towards piliated and nonpiliated N. gonorrhoeae cells and piliated encapsulated meningococci [40].

Bonnah et al. used a specialized cDNA microarray, the ‘IronChip’, to discover the alteration of several host genes involved in iron homeostasis, demonstrating that this Gram-negative pathogen alters the iron regulatory network of epithelial cells [26]. The same authors also demonstrated that Neisseria LOS plays a limited role in alterations of cellular iron homeostasis in epithelial cells, since IronChip analysis shows that epithelial cells have a similar stress response when infected with either N. meningitidis wild-type strain 8013.3 or the corresponding mutant lacking LOS [27].

Alterations of gene-expression profiles, mediated by lipopolysaccharide (LPS), were also investigated in human monocytes [28]. To address this, an LPS-deficient mutant of the MenB reference strain H44/76 was developed. This was a valuable tool to investigate the specific biological effects of LPS integrated in the outer membrane (OM) versus the effects of other inflammation-inducing molecules in the N. meningitidis cell wall. The results showed the presence in human monocytes of ‘particularly LPS-sensitive genes’, some of which were involved in both the Toll-like receptor and the JAK/STAT signaling pathways, and thus influence several biological processes such as chemotaxis, cell motility and immune responses.

Schubert-Unkmeir et al. performed a transcriptional analysis of human brain microvascular endothelial cells, key cells affected in the pathogenesis of meningococcal disease, 4 h (saturation of bacterial adhesion) and 8 h (maximum of bacterial internalization) after interaction with meningococcus [29]. The authors emphasized bacterium-mediated effects on the host cell function, cytoskeleton organization, monolayer integrity and cell receptor abundance or abundance of secreted molecules other than cytokines, since endothelial damage and capillary leakage were shown to be the basis of tissue injury in septicemia [41] and the endothelial cell cytoskeleton has been shown to participate in cell invasion by N. meningitidis[42]. A preliminary investigation to shed light on the commensal relationship of meningococci with their hosts was undertaken by Linhartova and colleagues by examining the effects of pilus-mediated adhesion and of the production of two secreted proteins (FrpC and FrpC-like) on human cells[30]. This was achieved by comparing the transcriptomes of human umbilical vein endothelial cells (HUVECs), noninfected or infected with either the adherent wild-type N. meningitidis MC58, or its nonadherent pilD deletion mutant, which is unable to produce type IV pili, or its FrpC/FrpC-like double-deletion mutant. First, the authors estimated the number of genes with altered expression as a consequence of bacterial presence in general and selected a large number of adhesion-independent upregulated genes, some of which had been previously described by others[43]. After that, the authors investigated the alterations of the FrpC/FrpC-like-mediated or the pilus-mediated transcriptomic profiles in infected HUVECs. While the presence of the secreted FrpC/FrpC-like proteins did not significantly alter the cellular expression profiles, pilus-mediated adherence was found to strongly upregulate genes involved in important cellular processes such as cell proliferation, transcriptional regulation, apoptosis, cell adhesion and inflammatory mediator production, altogether acting as cytoprotective signals aimed at contrasting the apoptotic signals induced by infection. Iron plays a prominent role in a variety of microbial metabolic pathways in bacterial pathogenesis. Pathogens possess and coordinate cellular mechanisms for iron acquisition and homeostasis, many of which in Neisseria are under the control of the iron-responsive transcriptional regulator protein Fur (reviewed in [44]). To define the Fur regulon of N. meningitidis, many comparative transcriptomics studies have been carried out [24,45–47]. Gene expression of bacterial cultures supplemented with ferric nitrate and depleted of iron have been compared [45]. A recently available fur deletion mutant allowed Delany and colleagues to perform global analysis of differentially expressed genes in the presence or absence of Fur protein and in response to iron limitation [47]. This allowed the identification of target genes affected by the Fur transcriptional regulator, and to distinguish those genes that are regulated by iron in a Fur-independent manner and by Fur in an iron-independent manner. Moreover, through footprinting analysis, the authors also biochemically investigated the hierarchy of direct and indirect Fur-mediated mechanisms of control.

Proteomics

The most remarkable advantage of the proteomic approach is that it analyzes proteins that are present in the sample at a given time point, documenting that genes are not only transcribed but also translated into proteins.

In 2006, an exhaustive proteomic work on N. meningitidis serogroup C was published by Basler and collaborators [25]. As mentioned previously, iron acquisition systems represent important virulence factors of N. meningitidis, the expression of which must be tightly regulated. To obtain comprehensive information about iron regulation in meningococci, the authors used whole-genome DNA microarray, 2DE and MALDI-TOF/mass spectrometry (MS) techniques to investigate in parallel the transcriptome and the proteome of a clinical isolate of N. meningitidis serogroup C under different conditions of iron availability. DNA microarray revealed that 152 genes were regulated by iron availability, 87 of which were upregulated and 65 downregulated, and 154 protein spots were found to be statistically up- or down-regulated by proteome analysis. The two methods were found to be highly complementary, given that the overlap of the datasets was rather limited (19 proteins/genes). A low degree of overlap in parallel profiling of transcripts and proteins is a well-documented problem [48]. Moreover, the use of techniques other than 2DE, such as gel-free techniques, do not eliminate the low correlation [49], suggesting that other factors, besides the technical limitations of 2DE, come into play to generate these discrepancies.

Nonetheless, the combined use of at least two different approaches appears to be desirable to obtain more comprehensive information on the transcription and translation profiles of entire genomes.

In the last few years, the main efforts in neisserial proteomic analysis have been directed towards the search for an effective vaccine against this pathogen.

Vaccination is the only way to control meningococcal disease. Conjugate vaccines against serogroups A, C, Y and W-135 are currently in clinical development and are expected to be licensed shortly[50]. On the other hand, serogroup B capsular polysaccharide-based vaccines are poorly immunogenic and may induce autoimmune responses. An alternative approach to an antimeningococcus B vaccine focused on the use of subcapsular antigens, in particular on surface-exposed proteins contained in OM vesicles (OMVs) [51], which have been shown to confer immune protection with various levels of efficacy. Several systematic proteomic analyses of commercial OMV-based vaccines have been carried out [2,3]. Their major goal was to identify proteins that should be carefully considered for future meningococcal vaccines. At the same time, they monitored the quantity and quality of protein patterns that can affect the immune response, in order to correlate these with important relapses for OMV efficacy and safety [2,3].

Driven by the low yields of current methods for the production of OMVs, Mukhopadhyay and coworkers investigated alternative manufacturing methods to increase production yields and, as part of this aim, developed process analytical technology methods to determine whether changes under fermentation condition alter the profile of the key immunogenic proteins in OMVs [52]. The novel technology of SELDI-TOF-MS was used, which is particularly suitable to analyze low-molecular-weight proteins and has the advantages of speed, accuracy and use at low protein concentrations, allowing a more complete characterization of protein profiles when compared with SDS-PAGE. The results demonstrated that SELDI-TOF can be successfully used to select media for vaccine production and to obtain a better understanding and control of the overall fermentation process. Most importantly, it can provide information on the quality, consistency and stability of the vaccine production process and can be used to detect protein degradation.

Meningococcal OM preparations from serogroup B strain MC58 and from the LPS-deficient mutant were analyzed by SDS-PAGE and nanocapillary liquid chromatography-MS/MS (GeLC-MS/MS) [53], with the aim to investigate the potential use of an OM-based vaccine from the mutant strain. The most abundant proteins in the OM preparation from the parental strain (PorA, PorB, Opc, Rmp, Opa and PilQ) were present at similar levels in the mutant, while markedly increased levels of several other proteins were detected in the OM preparation from the LPS-mutant strain. Nonetheless, the most important finding in this paper was that the GeLC-MS/MS approach increases the number of proteins that can be identified in a relatively unbiased manner, providing an indication of relative abundance based on the number of peptide fragments per protein. As demonstrated by other authors [54], GeLC-MS/MS offers a good alternative, and/or complement, to the classical 2DE/MS. The importance of this methodological integration is based on the number and the range of subcellular localization and functional classification (e.g., 33 vs 1 out of a total of 55 ribosomal proteins supposed to be expressed in N. meningitidis were identified by GeLC-MS/MS and 2DE/MS, respectively) of the identified proteins.

An alternative approach to a vaccine for serogroup B meningococcal disease has been applied by Finney and colleagues in the development of an OMV vaccine based on the commensal N. lactamica, a close taxonomic relative of N. meningitidis[55]. Immunological and epidemiological evidence suggests that carriage of N. lactamica contributes to the age-related development of natural immunity against meningococcal disease, most probably due to the development of antibodies against many surface structures shared with N. meningitidis[56,57]. Proteomic analysis of the N. lactamica OMV vaccine showed considerable similarity to meningococcal OMV vaccines, and antisera raised against N. lactamica OMVs cross-reacted with meningococcal proteins.

A critical flaw in the proteomic analysis of OMV vaccines lies in the fact that the native antigenic structures within vaccine candidates are not examined – thus, important conformational epitopes represented by protein complexes may remain undetected[58].

Experimental evidence exists about the importance of some conformational epitopes in bactericidal responses to N. meningitidis[59], thus a better understanding of the native conformation, location and association of OM proteins may be helpful for the design of more effective vaccines. Sánchez and coworkers [60] used cross-linking and proteomics (diagonal electrophoresis and immunoblotting) to explore OM proteins association in N. meningitidis and N. lactamica[60], and demonstrated the presence of three main protein complexes (95, 165 and 450 kDa) in the meningococcal membrane. Multiprotein complexes participate in a number of cellular processes. The identification and analysis of these components can explain the organization of the protein repertoire of a cell into functional units.

Expert commentary & five-year view

As generally defined, systems biology ascertains the complete repertoire of genes (genome), mRNAs (transcriptome), proteins (proteome) and fluxes of an organism or cell, together with their interactions on all levels based on global analysis techniques, subsequently integrating these data into predictive models[61]. Current systems biology approaches are mostly focused on experimental data collection and integration, but there is still the need for proper models to be developed that can be bridged to experimentally proven cellular properties.

Many technical and conceptual efforts still have to be made to achieve a real global view of bacterial systems biology.

Microbiology in the new millennium represents a frontline subject in which biological questions trigger technological/scientific answers and vice versa. If, in one respect, the pan-genome concept originated from a highly innovative mathematical model able to revolutionize the classical concept of species, in another respect, the metagenomics approach[62], dissecting the study of microbial gene repertoires of complex communities directly in their natural habitats (soil, fresh and marine water, and specific niches of the human host), pointed out the need for more and more powerful bioinformatic tools to manage the flood of data produced. Moreover, metagenomics can address important issues in clinical microbiology, such as the interconnections among commensal or pathogenic bacterial populations and their hosts[63].

Ideally, scientists should study microorganisms in their natural state and environment. To date, published transcriptomic studies have primarily described the host response to an infection, while the pathogen responses within infected host cells are still poorly documented. Moreover, because of the very limited amount of bacterial mRNA and its inefficient extraction from tissues, most microarray studies have been initiated from organisms grown in vitro. Considering the influence of growth conditions on microbial gene expression, it should be pointed out that the expression results achieved in synthetic media are likely to reproduce the real in vivo situation. Because of these considerations, traditional biological studies remain important [64], and many technical efforts still need to be carried out to establish standardized and systematic methodologies, and to develop in vitro model systems of bacterial infection. The same problems also affect proteomic studies.

Notably, the most important advantage of proteomics over genomics and transcriptomics is its capacity to analyze post-translational protein modifications that may not be evident from nucleotide sequence data analysis. Post-translational modifications have been demonstrated to play an important role in many aspects of bacterial pathogenesis. Moreover, since post-translational modifications provide an effective means to generate diversity, an important aspect of proteomics in the near future will be to assess how this diversity can influence antigenicity and pathogenicity.

There is a notable lack of meningococcal post-genomic studies focused mainly on strains isolated from meningitis-affected patients. However, for a more comprehensive understanding of meningococcal phatophysiology, it will be imperative to systems biology scientists to also focus their attention on those strains that can penetrate compartments other than the leptomeninges, such as the pericardial sac [65] and the peritoneal compartment [66]. In fact, meningococcal penetration into the pericardial and peritoneal compartments is almost unexplored, as are the environmental, bacterial and host factors that lead to such an intricate phenomenon.

In addition, experimental evidences exist for the occurrence of cytokine activation within these compartments [65,66], thus suggesting that high-throughput profiling of cytokines, by means of Luminex™-based technologies, could help researchers and physicians to shed light on this intriguing phenomenon [67].

In conclusion, recent advances in global genome, mRNA and protein analysis provide access to certain aspects of biological complexity, but many efforts are still needed to link such composite data and the dynamic metabolic and physiological aspects of cellular systems.

Key issues

• • Neisseria meningitidis is a human pathogen causing life-threatening diseases worldwide with high levels of morbidity and mortality.

• • Vaccination remains an excellent strategy for preventing infectious diseases, but no ‘universal’ vaccines against meningococcus are currently available.

• • The comparative genomic hybridization approach can assess the genetic diversity of pathogenic and nonpathogenic Neisseria spp.

• • Systems biology is based on global analysis techniques, such as genetics, transcriptomics and proteomics.

• • Many technical and conceptual efforts are needed to build robust simulation and prediction models of cellular systems.

Financial & competing interests disclosure

The authors have no relevant affiliations or financial involvement with any organization or entity with a financial interest in or financial conflict with the subject matter or materials discussed in the manuscript. This includes employment, consultancies, honoraria, stock ownership or options, expert testimony, grants or patents received or pending, or royalties.

No writing assistance was utilized in the production of this manuscript.