Info

Comparative Genomics

Comparative genome analyses of close relatives are a promising way to yield insights into the sources of microbial genome variability with respect to gene content, gene order, and the existence and distribution of potential virulence genes. These differences have also been identified in Neisseria species on the basis of computational and experimental techniques such as subtractive hybridization and DNA array technology.

Based on reciprocal best-hit BLASTP analyses [75] and the primary annotations as given in Refs. [12, 13, 16], 1406 ORFs are similar in all Neisseria species and may thus be an approximation to the genus-specific core genome (Fig. 11.2A). In particular, 313 of the gonococcal proteins belonging to the core proteome are more similar to the homologues N. meningitidis B MC58 proteins than to their homologues in N. meningitidis A Z2491, and 299 of these are in turn more similar to their counterparts in N. meningitidis A Z2491, according to the expect (E) values in pairwise BLASTP comparisons (Fig. 11.2B). The remaining 794 proteins have identical E values. Synteny plots (Fig. 11.2C, D) show that the gene order is well conserved over extended regions of the neisserial genomes, although there are also numerous large-scale rearrangements between the chromosomes such as transition and inversions (Fig. 11.1). For example, relative to the MC58 chromosome there is a large inversion of almost 1 Mbp around the replication origin in the Z2491 genome (Fig. 11.1, Fig. 11.2D) [13], which is in agreement with results from experimental work on physical chromosomal maps of N. meningitidis [76, 77] and N. gonorrhoeae [78, 79]. Although some large regions of the Z2491 chromosomal sequence (5-40 kbp) were found to be absent from the gonococcus FA1090 and could therefore represent meningococcus-specific islands, a significant proportion of this DNA is also absent from N. meningitidis MC58.

In addition to in silico comparisons, experimental techniques such as DNA arrays [80] or representational difference analysis [27, 81-84] have also been used

Fig. 11.2 Proteome comparison of three neisserial genomes based on the primary annotation and bidirectional best hit BLASTP analysis [75]. (A) Venn diagram of the encoded proteomes. (B) Comparison of the core proteome. Each dot represents a single N. gonorrhoeae FA1090 protein, plotted by its

BLASTP E value to the most similar protein from each of the two N. meningitidis strains depicted on the x and y axes. Symmetrical hits are positioned along a diagonal line. Panels (C) and (D) display synteny plots to compare the order of homologous genes between two annotated genomes as given.

Fig. 11.2 Proteome comparison of three neisserial genomes based on the primary annotation and bidirectional best hit BLASTP analysis [75]. (A) Venn diagram of the encoded proteomes. (B) Comparison of the core proteome. Each dot represents a single N. gonorrhoeae FA1090 protein, plotted by its

BLASTP E value to the most similar protein from each of the two N. meningitidis strains depicted on the x and y axes. Symmetrical hits are positioned along a diagonal line. Panels (C) and (D) display synteny plots to compare the order of homologous genes between two annotated genomes as given.

to analyze the genetic differences between various strains of the pathogenic Neisseria species and apathogenic N. lactamica in more detail.

Most of the genes that were shared by all virulent strains of N. meningitidis (78% of the chromosome; 1.7 Mbp) were also present in all isolates of the three species N. meningitidis, N. gonorrhoeae, and N. lactamica and thus may correspond to the neisserial core genome [80]. Of the rest, 46kbp were found to be strictly meningo-coccus-specific, that is, present in all strains of invasive meningococci and absent from all the N. gonorrhoeae and N. lactamica strains. Seventy-three kilobasepairs of the Z2491 genome are pathogen-specific, i.e., shared with the gonococcus and absent from N. lactamica. Twelve kilobasepairs are shared with N. lactamica but absent from the gonococcus.

The analysis of the genes corresponding to these sequences reveals that they correspond in general to single genes or small groups of genes scattered around the genome, since only one genetic island more than 10kbp in length could thus be detected (NMA0687-NMA0698) [80].

The majority of the experimentally identified genetic differences between N. meningitidis A Z2491 and N. gonorrhoeae FA1090 might therefore encode functions specific to bacteremia and invasion of the meninges. They are clustered in nine distinct regions on the Z2491 genome ranging in size from 1.8 kbp to 40kbp [82, 83]. These regions together comprise roughly 5% of the chromosome of N. meningitidis and contain 84 ORFs, of which 43 are homologous to ORFs already described in other species [83]. Based on the distribution of DNA uptake sequences, gene content, and sequence homology, it has been speculated that N. meningitidis-specific regions may have arisen by import from other species via recombination in the homologous flanking DNA [83], although there were no repeat structures similar to those flanking pathogenicity islands in Enterobacteria-ceae [32]. In addition, not all regions were present in a set of 13 different N. menin-gitidis strains additionally tested, and some were also present in nonpathogenic N. lactamica. In particular, region 1 contains the capsular polysaccharide biosynthesis genes; region 2, which is highly conserved among the strains compared, contains two genes with homology to genes in Bordetella pertussis encoding the filamentous hemagglutinin precursor FhaB and the accessory protein FhaC involved in the secretion of FhaA. Region 3 contains the gpxA gene encoding glutathione peroxidase and the 39.3-kbp prophage Pnm1. Region 4, which is absent from a large proportion of other N. meningitidis strains, contains two genes of unknown function, and region 5, which is also present in N. lactamica, encodes a restriction/modification system. Region 6 encodes a pseudogene and is deleted in most of the strains. With the exception of strain Z2491, it was found that region 7 contained functional copies of genes with homologies to genes of the type I secretion apparatus in other bacteria (hlyD and tolC), suggesting that this region might be involved in virulence in some of the N. meningitidis strains [83]. Region 8 contains the two genes fhuA and dsbA that were present in all strains, although fhuA is a pseudogene in some strains. fhuA is homologous to an Escherichia coli ferri-chrome-iron receptor gene that is involved in the uptake of siderophore-bound iron, and dsbA is similar to genes in E. coli that encode a disulfide oxidoreductase involved in the correct folding of secreted proteins. Finally, region 9 contains five ORFs of unknown function. However, neither resistance to complement killing nor adherence to or invasion of human umbilical vein epithelial cells (HUVEC) cells were affected when region 2, 3, 7, 8, or 9 was deleted from the test strains [83], and only deletion mutants lacking region 8 showed reduced survival in the bloodstream in an infant rat model. Thus, this region, together with region 1 encoding well-known virulence factors, may be necessary for dissemination in the bloodstream and thus for causing meningitis [83].

On a population scale, both restriction modification systems [27, 84] and plasmids [26, 85] were found to be differentially distributed among clonal groupings of meningococci possibly contributing to the genetic isolation of hypervirulent lineages.

Remarkably, none of the pathogen-specific regions experimentally identified have the characteristics typical of pathogenicity islands, bacteriophages, or compound transposons [80, 83], structures which are associated with the introduction into bacterial chromosomes of foreign DNA coding for virulence factors. There fore, dramatic differences in pathogenic potential may result from only small genetic changes.

A detailed comparative analysis of the genomic locus comprising the genes responsible for capsule synthesis, modification, and transport named cps was done with four meningococcal strains, one gonococcal strain, and two commensal Neisseria strains, i.e., N.lactamica and N. sicca (Fig. 11.3). All analyzed Neisseria species harbored the tex gene, whereas the region D containing the galE gene and the rfb genes was only found in meningococci, gonococci, and the closest relatives of meningococci among the apathogenic Neisseria species, i.e. N. lactamica. Thus, the tex gene may be the ancient core of the cps locus. The import of the lipA and lipB genes may have coincided with a duplication and rearrangement of region D, resulting in a truncated duplicate, termed D', and truncation of the gonococcal methyltransferase genes in meningococci. The evolution of the cps locus seems to r

0 0

Post a comment