Genome Sequencing Reveals Widespread Virulence Gene Exchange among Human Neisseria Species Authors: Pradeep Reddy Marri, Mary Paniscus, Nathan J. Weyand, María A. Rendón, Christine M. Calton, Diana R. Hernández, Dustin L. Higashi, Erica Sodergren, George M. Weinstock, Steven D. Rounsley, Magdalene So Published: 2010, PLOS ONE DOI: 10.1371/journal.pone.0011835 ABSTRACT Commensal bacteria comprise a large part of the microbial world, playing important roles in human development, health and disease. However, little is known about the genomic content of commensals or how related they are to their pathogenic counterparts. The genus Neisseria , containing both commensal and pathogenic species, provides an excellent opportunity to study these issues. We undertook a comprehensive sequencing and analysis of human commensal and pathogenic Neisseria genomes. Commensals have an extensive repertoire of virulence alleles, a large fraction of which has been exchanged among Neisseria species. Commensals also have the genetic capacity to donate DNA to, and take up DNA from, other Neisseria . Our findings strongly suggest that commensal Neisseria serve as reservoirs of virulence alleles, and that they engage extensively in genetic exchange. INTRODUCTION The genus Neisseria is a large group of β-Proteobacteria that are obligate symbionts of humans and animals. At least eight species of commensal Neisseria colonize human mucosal surfaces [1] , [2] . These sites are also infected by two pathogenic Neisseria : Neisseria meningitidis , which causes epidemics of meningitis and septicemia, and Neisseria gonorrhoeae , a sexually transmitted bacterium. Because of their importance to global public health, research has focused mainly on the two pathogens, leading to the identification of many virulence factors that are important for infection in humans. Recent reports indicate that commensal Neisseria also possess virulence genes [3] , [4] , [5] . However, these studies focused either on a limited number of virulence genes or on only one commensal genome, that of Neisseria lactamica . The total virulence gene content of the Neisseria genus is unknown. The rich diversity of Neisseria species at human mucosal surfaces raises the possibility of genetic exchange among these bacteria [6] , [7] . Indeed, interspecies exchange of antibiotic resistance genes can occur in vitro [8] . These considerations, combined with the limited glimpse of the virulence gene content of commensals, led us to undertake a comprehensive analysis of commensal Neisseria genomes and to determine the extent of interspecies genetic exchange. RESULTS We generated high quality draft genome sequences of eight species of human commensal Neisseria , and compared them to 11 published N. meningitidis , N. gonorrhoeae and N. lactamica genomes (see Table S1 for all strains used and their corresponding accession numbers). Commensal genomes were sequenced using a combination of the Roche 454 and Illumina platforms. Commensal genomes range between 1.8 and 2.8 Mb and contain 2,000 to 2,842 predicted genes. In contrast, pathogen genomes are ∼2.2 Mb and encode ∼2,000 predicted genes ( Table S2 ). Genes for common metabolic processes, DNA replication, recombination, transcription and translation are conserved in all commensals ( Data Set S1 and Text S1 ). The core Neisseria genome – the set of genes present in all Neisseria species – consists of 896 genes, mostly with housekeeping function assignments. To determine the phylogenetic relationship of Neisseria species, we concatenated the DNA sequences from a subset of 636 Neisseria core genes that are shared with Chromobacterium violaceum , the outgroup used to root the tree. The N. meningitidis and N. gonorrhoeae genomes form a distinct monophyletic clade that is derived from the commensal genomes ( Figure 1A ). The topology of this tree is consistent with that generated by concatenating all 896 Neisseria core genes ( Figure 1B ), and with previously derived single gene phylogenies [2] . (A) Rooted maximum likelihood Neisseria species tree (GTR + I + γ) based on concatenating the DNA sequences of a subset of 636 core Neisseria genes that are shared with the outgroup C. violaceum . A dagger denotes a bootstrap value of 100. The three N. meningitidis carrier strains are denoted by asterisks. (B) Maximum likelihood Neisseria species tree (GTR + I + γ) based on concatenating the DNA sequences of all 896 core Neisseria genes. Repetitive elements are important features of pathogenic Neisseria genomes, playing key roles in genetic transformation, gene expression and genome rearrangements [9] , [10] . The 10 bp DNA Uptake Sequence (DUS) GCCGTCTGAA is essential for genetic transformation of N. meningitidis and N. gonorrhoeae . As in pathogenic Neisseria , the DUS is the most prominent repetitive element in commensals, occurring over 2,000 times in most species ( Figure 2A ). Intriguingly, Neisseria sicca and Neisseria mucosa , which form a distinct clade on the Neisseria species tree ( Figure 1 ), have far fewer copies of the canonical DUS. Instead, they have >3,400 copies of a variant DUS (DUS1) with a one base mismatch: G t CGTCTGAA. This DUS1 functions in DNA uptake/transformation in N. meningitidis and N. gonorrhoeae , albeit less efficiently than the canonical sequence [11] , [12] . Thus, all commensals appear to have functional DNA uptake sequences. (A) Numbers of repeat sequences (RS), insertion sequences (IS), Correia elements, and DNA Uptake Sequences (DUS) are given for each Neisseria species, with complete copies in parentheses. DUS refers to the canonical DNA uptake sequence GCCGTCTGAA. DUS1 refers to the DUS variant G t CGTCTGAA. Note the high copy number of DUS1 in N. sicca and N. mucosa , highlighted in bold. (B) Absence of repeat elements in the pgd/lpxC locus region in commensals. Nel: N. elongata ; Nsi: N. sicca ; Nmu: N. mucosa ; Nsu : N. subflava ; Nfl: N. flavescens ; Nci: N. cinerea ; Npo : N. polysaccharea ; Nla: N. lactamica 23970; Nme: N. meningitidis MC58. Black bars represent dRS3 elements; blue bars represent Correia elements. Orthologous genes are the same color; orange represents a γ-glutamyltranspeptidase present in four commensal genomes; grey represents hypothetical proteins unique to each genome. (C) Absence of large repeat arrays in the murI-tbpA/B-potD-1 locus in commensals. Ngo: N. gonorrhoeae FA1090. Forward slashes (//) represent a break in chromosome contiguity. Orthologous genes are the same color; light blue denotes a hypothetical protein. All other features are the same as in (B). Correia and dRS3 elements [13] are the second most abundant repeats in commensal Neisseria . These elements function in gene regulation and sequence variation in pathogenic Neisseria [14] , [15] . Commensals have fewer Correia and dRS3 elements than the pathogens. The only exception is N. sicca , which has more Correia elements than either of the pathogens ( Figure 2A, 2B and 2C ; Text S1 ). Moreover, commensal dRS3 elements are not arranged into large Intergenic Mosaic Element arrays, unlike their counterparts in pathogenic Neisseria ( Figure 2B and 2C ). This suggests that expansion of dRS3 elements occurred more recently in the pathogens. In pathogenic Neisseria , repetitive elements facilitate two important immune evasion mechanisms: phase and antigenic variation. Many N. meningitidis and N. gonorrhoeae genes for outermembrane proteins undergo high frequency ON/OFF (phase) switching that results from intragenic recombination of repeat sequences in or near the coding sequence. Over 70% of the genes (52/72) known or hypothesized to be phase variable in pathogenic Neisseria [14] , [16] , [17] are present in commensals ( Table S3 ). However, many of these (26/52) do not have tandem repeats, indicating that recombination-based phase switching of surface proteins occurs less frequently in commensals. Studies of a few handpicked genes and whole genome microarrays centered on N. lactamica detected virulence genes in commensals [3] , [4] , [5] . To examine virulence gene content in the newly sequenced commensal genomes, we determined the distribution of 177 genes that have been reported to play a role in Neisseria virulence ( Table S4 ) [4] , [18] , [19] . Seventy-five percent of these genes (133/177) are present in one or more commensals ( Figure 3A ), and 40% (70/177) are in all of them ( Data Set S2 ). A large subset of Neisseria virulence genes is therefore present in the commensals. (A) Number of genes present in different Neisseria species, out of the 177 genes known or hypothesized to be important for virulence (see methods for details). (B) Heatmap of AU test p -values obtained by comparing individual virulence gene tree topologies. The map is a 69×69 matrix of 69 virulence genes that are present in all Neisseria species (y-axis) and their corresponding topologies (x-axis). Blue indicates a p -value of 1 (trees are similar); red indicates a p -value of 0 (trees are significantly different). Among the virulence genes found in all commensal species are those encoding the Type IV pilus (Tfp), a surface structure that in pathogenic Neisseria promotes twitching motility, DNA uptake/genetic transformation, attachment, and host cell signaling [20] . Commensals have a complete set of Tfp biogenesis genes ( Figure S1 ), including: pilE , encoding the Tfp structural subunit; pilD , encoding a pilin peptidase; pilF , encoding a pilin assembly ATPase; and pilT, encoding an ATPase required for Tfp retraction and DNA uptake. Commensals also have one or more pilin modification genes. The four commensals closest to the pathogen clade, N. lactamica , Neisseria polysaccharea , Neisseria cinerea , and Neisseria flavescens , have a pilC ortholog. However, the function of commensal pilC cannot be deduced due to sequence differences from pathogenic Neisseria pilC1 and pilC2 . Commensals therefore have the genetic capacity to produce Tfp. The functions of commensal Tfp, if expressed, remain to be elucidated. Tfp systems of commensal and pathogenic Neisseria differ in one major respect. In pathogenic Neisseria , pilin antigenic variation results from recombination of pilE with a silent variant pilin pseudogene, or pilS [21] . Pathogenic Neisseria have as many as 19 copies of pilS , while commensals have only 2–5 copies ( Table S5 ). Moreover, in commensals the region upstream of pilE lacks the guanine-repeat element that is essential for pilE / pilS recombination [22] . Pilin antigenic variation is therefore unlikely to occur in commensals, at least via the guanine-repeat element mechanism. As pseudogenes lacking a function are rapidly lost from the population [23] , the presence of pilS in commensal genomes is curious. The phylogeny of Neisseria ( Figure 1 ) indicates that commensals are basal to the two pathogenic species, which form a single, derived monophyletic clade. The most parsimonious explanation for these observations is that pilE / pilS antigenic variation is a novel function that arose in the pathogens from a system originally evolved for another, yet unidentified purpose. The Opa family of outermembrane proteins promote N. gonorrhoeae and N. meningitidis attachment, invasion, immune cell signaling and inflammation. Pathogen genomes harbor multiple variant opa s; N. meningitidis has 3–4 opas , while N. gonorrhoeae has approximately 11 [24] . Most commensals lack opa genes and thus do not interact with host cells via this virulence factor. However, N. polysaccharea , N. flavescens and N. lactamica have 1, 2 and 3 variant opas , respectively ( Table S5 ). All of these contain variable numbers of the CTCTT pentameric repeat that, in pathogenic Neisseria , undergoes slip-strand misrepair resulting in Opa phase and antigenic variation [25] , [26] . Commensal Opas are therefore likely to undergo phase switching, but limited antigenic variation. Iron scavenging from the host is an important virulence attribute of bacterial pathogens [27] . Commensal Neisseria have a diverse arsenal of iron acquisition genes ( Figure 4 ). N. lactamica and N. cinerea have genes for acquiring iron from human transferrin and lactoferrin ( tbpA/tbpB and lbpA/lbpB , respectively) [28] . These loci are important for N. gonorrhoeae infection and fitness [28] . Commensals also have hmbR and hpuAB , which are required for acquiring hemoglobin iron [28] . In N. meningitidis , HmbR promotes replication in the blood [29] . All commensals have the tonB/exbB/exbD iron transport locus, and fur, the iron-responsive regulatory element. Notably, commensals most distal to the pathogens have iron uptake genes that are missing from pathogenic Neisseria . These include genes for transport of hemin, Fe +3 and Fe +2 , and for siderophore receptors and transport ( Figure 4 ). Because Fe +2 is often complexed with protein in foodstuff, the Fe +2 uptake system, if expressed, may allow oral commensals to absorb iron from the host diet [28] . Thus, Neisseria species possess a diverse array of iron uptake genes. This diversity may help different Neisseria species to colonize the same niche without being affected by antibodies directed against the iron acquisition components of other species [30] . Species and strain names are listed in the phylogenetic tree. N. meningitidis carrier strains are marked with asterisks. Each box represents an individual gene. Genes of the same color belong to the same iron utilization system. Genes connected by a horizontal line are contiguous in the chromosome. Loci confirmed by sequencing to be pseudogenes have a forward slash (/). N. meningitidis and N. gonorrhoeae gene names appear below each column. Genes are listed as locus tags if their iron utilization functions were deduced from other bacteria. General functions appear below the gene names. Several genes are found uniquely in commensals. Only four genes - encoding hypothetical proteins - are present in all commensals, while several are present in one or more commensals. Intriguingly, some commensal-specific genes encode proteins that function as hemolysins, adhesins or invasins in other bacterial pathogens ( Data Sets S3 and S4 ). Others encode components of the Type IV and Type VI Secretion Systems and lipooligosaccharide modification enzymes ( Text S1 ). Commensal-specific genes are generally smaller, have a lower %G+C content than Neisseria core genes, and have homologs in other bacterial genera, indicating that they were acquired recently via horizontal gene transfer. We also identified several pathogen-specific genes ( Data Set S5 ). Only 16 genes are present in all sequenced N. meningitidis genomes, and 81 in all sequenced N. gonorrhoeae genomes. Genes for biosynthesis of the polysaccharide capsule, which confers resistance to complement killing and phagocytosis, are unique to N. meningitidis ( Figure S2 ). The ctrABCD and lipAB loci, encoding polysaccharide transporters, are found in some commensals ( Text S1 ). However, the siaABC locus, which is required for capsule production, appears only in N. meningitidis . The tdfF locus which is essential for intracellular iron acquisition and thus for Neisseria intracellular replication, is found only in pathogen genomes ( Figure 4 ) [31] . Some genes were found in all sequenced N. meningitidis and/or N. gonorrhoeae genomes, but only rarely in commensal species. For example, iga , encoding an IgA protease that cleaves human IgA and lysosomal membrane protein Lamp1 [20] , is present only in pathogenic Neisseria genomes with one exception: an intact iga2 allele is found in both N. lactamica genomes. The 57 kb Gonococcal Genetic Island (GGI) [32] is largely absent from commensals. However, genes with varying degrees of nucleotide sequence identity to GGI encoded dsbC, topB, and parA are present in all commensal genomes ( Text S1 ). In N. gonorrhoeae , dsbC and parA play roles in genetic transformation by secreting DNA [32] . Nf, an M13-like filamentous phage, is the only feature that distinguishes hyperinvasive N. meningitidis isolates from noninvasive ones [33] , [34] . Nf genes are detected in N. flavescens and N. lactamica . N. flavescens has only a subset of Nf genes, while N. lactamica contains all the major M13-like life cycle genes. Finally, it should be noted that NMB1646, encoding a putative hemolysin, was proposed to be specific to pathogenic Neisseria . Microarray studies found NMB1646 in N. meningitidis and N. gonorrhoeae , but absent from N. lactamica [4] . In our study, NMB1646 is also missing from N. lactamica . However, it was found in all other commensal genomes. Thus, NMB1646 is present in both commensal and pathogenic Neisseria . Our study shows that commensals have a large number of virulence genes, as well as genes for DNA secretion and uptake, and an abundance of DUS. Additionally, previous work has demonstrated gene exchange among Neisseria species [8] , [35] , [36] , [37] . Combined, these results suggest that genetic exchange among Neisseria may be more widespread than previously thought. We therefore determined the extent of genetic exchange among Neisseria species, using a phylogenetic method to analyze 69 virulence genes shared by all 19 Neisseria genomes ( Data Set S2 ). This method relies on the phylogenetic congruence of individual gene trees: a gene tree that does not match the species tree indicates the gene in question has been transferred laterally. We generated individual gene trees for each of the 69 virulence genes using PhyML [38] and looked for the rejection of a set of these topologies by the 69 genes using an Approximate Unbiased (AU) test [39] . The p -value heatmap ( Figure 3B ) indicates that most gene trees showed divergent histories (depicted as groups of topologies with dissimilar p -value patterns) that are most likely due to horizontal gene transfer. A small group of topologies (topologies 58–69) were rejected by all the genes. Consistent with these findings, 45% (31/69) of the virulence genes rejected the concatenated tree, compared to 34% of the core genes (average of three replicates of 69 randomly selected core genes). These results suggest that there is widespread genetic exchange among Neisseria species, with virulence genes being exchanged at a slightly higher frequency than core genes. The AU test relies on the rejection of a tree by a gene, rather than on its acceptance, and less divergent or saturated alignments may result in false predictions [40] . Moreover, high rates of recombination between N. meningitidis strains can also result in conflicting phylogenies. Therefore, we also looked for evidence of horizontal gene transfer using the program RDPv3.18 [41] to detect homologous intragenic recombination in the 69 virulence genes mentioned above. Evidence of recombination was detected in 53 of the 69 genes, and over half of these events (31/53) involved at least one commensal genome ( Table S6 ). The intragenic recombination and AU tests provide strong evidence of genetic exchange among commensal and pathogenic Neisseria , suggesting that commensals act as reservoirs of new virulence alleles. DISCUSSION In this comprehensive analysis of human commensal Neisseria genomes, we have identified the gene content of commensal Neisseria species, generated an extensive phylogenetic tree of this genus, and determined the distribution of virulence genes in its members. Commensal Neisseria have an extensive repertoire of virulence alleles from pathogenic Neisseria as well as other bacterial genera. Moreover, they have the genetic machinery for exchanging DNA with each other, and nearly half of Neisseria virulence genes have undergone intra- and interspecies recombination. High frequency horizontal gene transfer can increase pathogen fitness, accelerate host adaptation, and affect bacterial virulence. The prevalence of numerous Neisseria species in the same niches in the human body provides an opportunity for DNA exchange. The repertoire of virulence genes in Neisseria populations is therefore likely to be dynamic. The large virulence gene set in commensal Neisseria raises an intriguing question: why are these bacteria generally not pathogenic? Although there are no ready answers to this question, we can offer possible explanations. First, a productive infection is determined by multiple factors, including the immediate environment of the infection site and host immune status. Commensals may not be able to express the entire constellation of virulence genes necessary for initiating infection under these circumstances. In this context, it is interesting to note that N. lactamica , which occasionally causes infections, has the largest set of virulence alleles of all the commensals. Secondly, there may be additional factors governing commensalism that have yet to be defined. N. meningitidis can cause asymptomatic infections in humans [42] . Yet, there is very little difference between the genomes of carrier and invasive strains [18] . This suggests that the ON/OFF phase switching of genes may be an important pathogenesis determinant. Indeed, many commensal orthologs of phase-variable genes lack the repeat elements that participate in phase variation. Third, a virulence allele in a commensal may differ from its pathogenic counterpart in its ability to cause disease. For example, N. gonorrhoeae expressing the N. lactamica porB allele is less able to invade cells [43] . Finally, our results add to the growing body of evidence that suggests the definition of virulence factor needs to be reassessed. Virulence factors have been defined as components whose absence attenuates bacterial virulence but not viability [44] . For example, Tfp is considered a virulence factor because it promotes attachment [45] . However, recent studies demonstrate that Tfp activates pro-survival pathways in epithelial cells [46] , [47] , a situation that benefits both host and microbe. The presence of Tfp biogenesis genes in commensals supports this idea. Thus, some virulence factors may be more appropriately termed “host adaptation” factors, as they allow bacteria to interact with and adapt to the host environment, but do not specifically promote virulence. The distinction between a true virulence factor versus a host adaptation factor will be clarified as the genomes of more commensal bacteria are sequenced and characterized. Genome sequences generated in this study were deposited in GenBank. See Table S1 for the GenBank accession numbers for all genome sequences discussed in this paper. Updated versions of the eight commensal Neisseria genome sequences generated in this study are available at http://www.u.arizona.edu/~pradeepm/data.html . MATERIALS AND METHODS The following strains were obtained from the American Type Culture Collection (ATCC): N. lactamica (ATCC 23970), N. polysaccharea (ATCC 43768), N. cinerea (ATCC 14685), N. sicca (ATCC 29256), N. mucosa (ATCC 25996), and Neisseria elongata subsp. glycolytica (ATCC 29315). Neisseria subflava (NJ9703) was provided by J.B. Kaplan [48] . N. flavescens (NRL30031 or H210) was a gift from E.L. Aho [49] , [50] . All commensal Neisseria strains were grown on chocolate agar with 3.6% GC agar base (BD), 1% Isovitalex (BD), and 1% hemoglobin (Oxoid) for 18–24 hours in 5% CO 2 at 37°C. DNA ISOLATION Chromosomal DNA was isolated as described previously [51] . Briefly, cells from half of a 10 cm diameter petri dish were harvested with a sterile Dacron swab into 0.5 ml of 50 mM Tris-Cl, 20 mM EDTA, 50 mM NaCl (pH 8). SDS (1% final concentration) and RNAse A (Qiagen, 1 mg/ml) were added, and cells were lysed for five minutes. Two extractions with phenol-chloroform-isoamyl alcohol (25∶24∶1) and one with chloroform were used to remove proteins. DNA was precipitated by adding two volumes of isopropanol followed by suspension in TE (10 mM Tris, 1 mM EDTA, pH 8). Ammonium acetate was added to 2.5 M followed by precipitation with two volumes of ethanol. Following 70% ethanol washes and drying, DNA was suspended in water or 10 mM Tris, 0.1 mM EDTA (pH 8). GENOME SEQUENCING AND ASSEMBLY 454 sequencing was performed using the Titanium platform (Roche) according to the manufacturer's instructions. Paired-end sequencing was completed using the Illumina Genome Analyzer II following protocols specified by the manufacturer. Draft genomes were assembled using the Newbler (Roche) and Velvet [52] assemblers. GENOME ANNOTATION Annotation was performed using a combination of RAST [53] and the Washington University Genome Center's gene prediction pipeline. Select loci were manually curated using previously published N. meningitidis and N. gonorrhoeae genome sequences as references. ORTHOLOGS AND CORE GENOME Orthologous proteins were identified by BLASTP using reference proteins from previously sequenced N. meningitidis and N. gonorrhoeae genomes where possible. Predicted proteins were designated orthologs of a reference protein if the ratio of reference self-hit bitscore to predicted protein bitscore was greater than 0.4. The orthology of proteins unique to draft genomes was determined in the same manner, except randomly chosen predicted proteins were used as reference proteins. The orthology was confirmed by genome context whenever required. The 896 genes present in all 19 Neisseria genome sequences used in this study were designated as the core Neisseria genome. GENERATION OF NEISSERIA SPECIES TREE The subset of 636 core Neisseria genes also present in C. violaceum was used to find phylogenetic relationships among Neisseria species. The concatenated DNA sequences of these 636 genes were aligned using MAFFT [54] and the resulting alignment edited using Gblocks [55] to remove any gaps. A maximum likelihood tree (using the GTR + I + γ model in PAUP) was generated using the concatenated orthologous sequences from C. violaceum as the outgroup. Support for each branch was obtained by performing the bootstrap analysis with 100 replicates. Using the same method, an unrooted tree was also obtained with the concatenated sequences from all 896 core Neisseria genes. IDENTIFICATION OF REPEAT ELEMENTS Repetitive elements of commensal Neisseria were identified using the ‘fuzznuc’ application of EMBOSS [56] . DNA Uptake Sequences (DUS) were identified in each genome by looking for the pattern GCCGTCTGAA . Non-canonical DUS were found by allowing one mismatch to the canonical DUS pattern. The dRS3 elements were detected in each genome using the pattern ATTCCCNNNNNNNNGGGAAT [13] , [14] . Each identified site was then manually checked to see if it had a lone dRS3 repeat or a complete RS element. Relaxing the search to allow one or two mismatches did not detect any additional RS elements. Correia repeats (CR) were identified by searching for the following patterns [15] , [57] , with three mismatches allowed per pattern: TATAG[CT]GGATTAACAAAAATCAGGAC, TATAG[CT]GGATTAAATTTAAACCGGTAC, TATAG[CT]GGATTAACAAAAACCGGTAC, TATAG[CT]GGATTAAATTTAAATCAGGAC. Each identified repeat was then manually checked to see if it had a lone CR or a complete Correia element. IDENTIFICATION OF VIRULENCE GENES A list of 177 virulence genes was created by searching published literature and available databases for Neisseria genes that are known or hypothesized to have virulence functions [4] , [18] , [19] . The protein sequences of these 177 genes were used as queries in a TBLASTN search against all commensal genome sequences with an E-value cutoff set at 1e −5 . A gene was identified as present if it had >50% identity to the query protein and an alignment length of ≥75% of the query length. IDENTIFICATION OF PHASE VARIABLE GENES Searches of the published literature identified 72 Neisseria genes known or hypothesized to be phase variable [14] , [16] , [17] . The protein sequences of these 72 genes were used as queries in a TBLASTN search against all commensal genome sequences with an E-value cutoff set at 1e −5 . A gene was identified as present if it had >50% identity to the query protein and an alignment length of ≥75% of the query length. To confirm the presence of the phase variable (PV) repeat, the corresponding gene from the commensal genome sequence was extracted and aligned to the appropriate reference gene sequence. The alignments were manually examined for the presence of the PV repeat. GENE TREE-TO-SPECIES TREE CONGRUENCE A phylogenetic method was used to detect the extent of genetic exchange among Neisseria species by examining 69 of the 70 virulence genes that are shared across all 19 Neisseria genome sequences. The pilE gene was not analyzed, as multiple copies of pilE in several of the genomes made identifying orthologs difficult. Phylogenies for each of the 69 genes were derived in PhyML [38] using a GTR + I + γ model of evolution with BIONJ starting tree. The site likelihood for each tree was computed using baseml (PAML package) [58] . The AU test [39] was then applied using Consel [59] . DETECTION OF INTRAGENIC RECOMBINATION To provide a secondary method for measuring genetic exchange, the extent of intragenic recombination was also examined. Alignments for each of the 69 virulence genes were generated and recombination breakpoints identified using the program RDPv3.18 [41] . CONFIRMATION OF PSEUDOGENES Genes of interest were amplified from purified genomic DNA by PCR. All PCR reactions were performed in a 50 µl reaction volume using 50 ng of purified DNA as template, 1X Phusion HF master mix (Finnzymes) and 0.1 mM of the gene-specific primers ( Table S7 ). Thermocycler conditions were as follows: 1 cycle at 98°C for 30 sec; 35 cycles at 98°C for 10 sec, 60°C for 30 seconds and 72°C for 15 sec; 1 cycle at 72°C for 10 min. The resulting PCR products were purified using a column-based method (Qiagen). Purified products were then sequenced using appropriate forward and reverse primers ( Table S7 ) by the Sanger method at the University of Arizona Genetics Core (Tucson, AZ). SUPPORTING INFORMATION Supplemental text and references. (0.11 MB DOC) Human Neisseria type IV pilus (Tfp) biogenesis and pilin modification genes. Species and strain names are listed in the phylogenetic tree on the left. N. meningitidis carrier strains are denoted by asterisks. Each box represents an individual gene. Genes connected by a horizontal line are contiguous on the chromosome. A forward slash (/) indicates genes that have been confirmed as pseudogenes by sequencing. A backwards slash (\) represents hypothetical pseudogenes whose status has not been confirmed by experimentation. Genes in grey are not involved in Tfp biosynthesis or pilin modification. Tfp genes that have not yet been named are listed by their locus tag designations. The designations of commensal pilC orthologs are based on genome context and sequence homology to pilC1 or pilC2 of pathogenic Neisseria . (0.89 MB TIF) Capsular polysaccharide genes of human Neisseria species. The Neisseria capsular locus is made up of five regions: A, B, C, D, and E. Some sequences also contain a duplicate of the D region, denoted as D'. A composite of the multi-species capsular locus is depicted at the bottom of the figure and includes gene names based on published nomenclature. Each box represents an individual gene, except for numbered boxes, which represent genes of the number listed. Genes of the same color are part of the same pathway; grey boxes denote genes whose functions are unknown or unrelated to the capsule. Genes connected by a horizontal line are contiguous on the chromosome. A backwards slash (\) represents hypothetical pseudogenes whose status has not been confirmed by experimentation. Note that in N. subflava , the C and E regions as well as gene NLA0370 are in an inverted orientation compared to the other sequences. Also, whether N. elongata , N. sicca , and N. mucosa contain a D or D' region cannot be ascertained from the available data. (0.74 MB TIF) Genome sequences used in this study. (0.07 MB PDF) General characteristics of Neisseria genomes. (0.07 MB PDF) Presence or absence of phase variable genes in commensal Neisseria . Targets were identified by searching the literature for genes known or hypothesized to be phase variable in pathogenic Neisseria . (0.17 MB PDF) Genes known or hypothesized to be involved in Neisseria virulence. The 177 targets were identified by searching the literature for genes previously identified as having a role in Neisseria virulence. Genes in bold type are present in all 19 Neisseria genomes used in the study. (0.13 MB PDF) Copy number of pilE , pilS and opa genes in Neisseria species. See supplemental references ( Text S1 ) for additional information on N. meningitidis MC58 [3] , [20] and N. gonorrhoeae FA1090 [20] , [21] . (0.08 MB PDF) Recombination of virulence genes among Neisseria species. The RDPv3.18 program was used to test for recombination in 69 virulence genes that are shared by all Neisseria species. (0.13 MB PDF) Primers used for PCR amplification and sequencing of potential pseudogenes. (0.07 MB PDF) Common metabolic pathways present in commensal Neisseria . Genes from the N. meningitidis MC58 genome are included as a reference. (0.18 MB XLS) Virulence genes present in all 19 Neisseria genomes. Note that the N. subflava and N. cinerea genome sequences both contain two copies of pilE ; the second copy is listed directly below the first (see rows 74–75). (0.06 MB XLS) Neisseria strain-specific genes. The name on each tab indicates the species included for the given table. (0.26 MB XLS) Commensal Neisseria -specific genes. Genes present in two or more commensal genomes but absent from all N. meningitidis and N. gonorrhoeae genomes. See supplemental reference [19] ( Text S1 ) for abbreviations of the COG categories. (0.22 MB XLS) Pathogenic Neisseria -specific genes. Tab “N. meningitides” includes the seven N. meningitidis strains analyzed; tab “N. gonorrhoeae” includes the three N. gonorrhoeae genomes studied. See supplemental reference [19] ( Text S1 ) for abbreviations of the COG categories. (0.04 MB XLS) REFERENCES [1] Janda WM Knapp JS 2003 Neisseria and Moraxella catarrhalis . Murray PR Manual of clinical microbiology Washington, D.C. ASM Press 585 608 8th ed. [2] Tonjum T 2005 Genus Neisseria . Garrity GM Bergey's manual of systematic bacteriology New York Springer-Verlag 777 798 2nd ed. [3] Bennett JS Thompson EA Kriz P Jolley KA Maiden MC 2009 A common gene pool for the Neisseria FetA antigen. Int J Med Microbiol 299 133 139 [4] Snyder LA Saunders NJ 2006 The majority of genes in the pathogenic Neisseria species are present in non-pathogenic Neisseria lactamica , including those designated as ‘virulence genes’. BMC Genomics 7 128 [5] Stabler RA Marsden GL Witney AA Li Y Bentley SD 2005 Identification of pathogen-specific genes through microarray analysis of pathogenic and commensal Neisseria species. Microbiology 151 2907 2922 [6] Knapp JS 1988 Historical perspectives and identification of Neisseria and related species. Clin Microbiol Rev 1 415 431 [7] Saez Nieto JA Marcos C Vindel A 1998 Multicolonization of human nasopharynx due to Neisseria spp. Int Microbiol 1 59 63 [8] Catlin BW 1960 Interspecific transformation of Neisseria by culture slime containing deoxyribonucleate. Science 131 608 610 [9] Liu SV Saunders NJ Jeffries A Rest RF 2002 Genome analysis and strain comparison of correia repeats and correia repeat-enclosed elements in pathogenic Neisseria . J Bacteriol 184 6163 6173 [10] Snyder LA McGowan S Rogers M Duro E O'Farrell E 2007 The repertoire of minimal mobile elements in the Neisseria species and evidence that these are involved in horizontal gene transfer in other bacteria. Mol Biol Evol 24 2802 2815 [11] Goodman SD Scocca JJ 1988 Identification and arrangement of the DNA sequence recognized in specific transformation of Neisseria gonorrhoeae . Proc Natl Acad Sci U S A 85 6982 6986 [12] Qvarnstrom Y Swedberg G 2006 Variations in gene organization and DNA uptake signal sequence in the folP region between commensal and pathogenic Neisseria species. BMC Microbiol 6 11 [13] Parkhill J Achtman M James KD Bentley SD Churcher C 2000 Complete DNA sequence of a serogroup A strain of Neisseria meningitidis Z2491. Nature 404 502 506 [14] Bentley SD Vernikos GS Snyder LA Churcher C Arrowsmith C 2007 Meningococcal genetic variation mechanisms viewed through comparative analysis of serogroup C strain FAM18. PLoS Genet 3 e23 [15] Snyder LA Cole JA Pallen MJ 2009 Comparative analysis of two Neisseria gonorrhoeae genome sequences reveals evidence of mobilization of Correia Repeat Enclosed Elements and their role in regulation. BMC Genomics 10 70 [16] Jordan PW Snyder LA Saunders NJ 2005 Strain-specific differences in Neisseria gonorrhoeae associated with the phase variable gene repertoire. BMC Microbiol 5 21 [17] Snyder LA Saunders NJ Shafer WM 2001 A putatively phase variable gene ( dca ) required for natural competence in Neisseria gonorrhoeae but not Neisseria meningitidis is located within the division cell wall ( dcw ) gene cluster. J Bacteriol 183 1233 1241 [18] Schoen C Blom J Claus H Schramm-Gluck A Brandt P 2008 Whole-genome comparison of disease and carriage strains provides insights into virulence evolution in Neisseria meningitidis . Proc Natl Acad Sci U S A 105 3473 3478 [19] Tettelin H Saunders NJ Heidelberg J Jeffries AC Nelson KE 2000 Complete genome sequence of Neisseria meningitidis serogroup B strain MC58. Science 287 1809 1815 [20] Merz AJ So M 2000 Interactions of pathogenic neisseriae with epithelial cell membranes. Annu Rev Cell Dev Biol 16 423 457 [21] Hill SA Davies JK 2009 Pilin gene variation in Neisseria gonorrhoeae : reassessing the old paradigms. FEMS Microbiol Rev 33 521 530 [22] Cahoon LA Seifert HS 2009 An alternative DNA structure is necessary for pilin antigenic variation in Neisseria gonorrhoeae . Science 325 764 767 [23] van Passel MW Marri PR Ochman H 2008 The emergence and fate of horizontally acquired genes in Escherichia coli . PLoS Comput Biol 4 e1000059 [24] Dehio C Gray-Owen SD Meyer TF 1998 The role of neisserial Opa proteins in interactions with host cells. Trends Microbiol 6 489 495 [25] Muralidharan K Stern A Meyer TF 1987 The control mechanism of opacity protein expression in the pathogenic Neisseriae. Antonie Van Leeuwenhoek 53 435 440 [26] Murphy GL Connell TD Barritt DS Koomey M Cannon JG 1989 Phase variation of gonococcal protein II: regulation of gene expression by slipped-strand mispairing of a repetitive DNA sequence. Cell 56 539 547 [27] Payne SM 1993 Iron acquisition in microbial pathogenesis. Trends Microbiol 1 66 69 [28] Cornelissen CN 2004 Iron transport system in pathogenic bacteria: Neisseria . Cronin A Iron transport in bacteria Washington, D.C. ASM Press [29] Stojiljkovic I Hwa V de Saint Martin L O'Gaora P Nassif X 1995 The Neisseria meningitidis haemoglobin receptor: its role in iron utilization and virulence. Mol Microbiol 15 531 541 [30] Snyder LA Davies JK Ryan CS Saunders NJ 2005 Comparative overview of the genomic and genetic differences between the pathogenic Neisseria strains and species. Plasmid 54 191 218 [31] Hagen TA Cornelissen CN 2006 Neisseria gonorrhoeae requires expression of TonB and the putative transporter TdfF to replicate within cervical epithelial cells. Mol Microbiol 62 1144 1157 [32] Hamilton HL Dominguez NM Schwartz KJ Hackett KT Dillard JP 2005 Neisseria gonorrhoeae secretes chromosomal DNA via a novel type IV secretion system. Mol Microbiol 55 1704 1721 [33] Bille E Zahar JR Perrin A Morelle S Kriz P 2005 A chromosomally integrated bacteriophage in invasive meningococci. J Exp Med 201 1905 1913 [34] Kawai M Uchiyama I Kobayashi I 2005 Genome comparison in silico in Neisseria suggests integration of filamentous bacteriophages by their own transposase. DNA Res 12 389 401 [35] Bowler LD Zhang QY Riou JY Spratt BG 1994 Interspecies recombination between the penA genes of Neisseria meningitidis and commensal Neisseria species during the emergence of penicillin resistance in N. meningitidis : natural events and laboratory simulation. J Bacteriol 176 333 337 [36] Linz B Schenker M Zhu P Achtman M 2000 Frequent interspecific genetic exchange between commensal Neisseriae and Neisseria meningitidis . Mol Microbiol 36 1049 1058 [37] Maiden MC 1998 Horizontal genetic exchange, evolution, and spread of antibiotic resistance in bacteria. Clin Infect Dis 27 Suppl 1 S12 20 [38] Guindon S Gascuel O 2003 A simple, fast, and accurate algorithm to estimate large phylogenies by maximum likelihood. Syst Biol 52 696 704 [39] Shimodaira H 2002 An approximately unbiased test of phylogenetic tree selection. Syst Biol 51 492 508 [40] Susko E Leigh J Doolittle WF Bapteste E 2006 Visualizing and assessing phylogenetic congruence of core gene sets: a case study of the gamma-proteobacteria. Mol Biol Evol 23 1019 1030 [41] Martin DP Williamson C Posada D 2005 RDP2: recombination detection and analysis from sequence alignments. Bioinformatics 21 260 262 [42] Glitza IC Ehrhard I Muller-Pebody B Reintjes R Breuer T 2008 Longitudinal study of meningococcal carrier rates in teenagers. Int J Hyg Environ Health 211 263 272 [43] Bauer FJ Rudel T Stein M Meyer TF 1999 Mutagenesis of the Neisseria gonorrhoeae porin reduces invasion in epithelial cells and enhances phagocyte responsiveness. Mol Microbiol 31 903 913 [44] Casadevall A Pirofski LA 1999 Host-pathogen interactions: redefining the basic concepts of virulence and pathogenicity. Infect Immun 67 3703 3713 [45] Craig L Pique ME Tainer JA 2004 Type IV pilus structure and bacterial pathogenicity. Nat Rev Microbiol 2 363 378 [46] Howie HL Glogauer M So M 2005 The N. gonorrhoeae type IV pilus stimulates mechanosensitive pathways and cytoprotection through a pilT-dependent mechanism. PLoS Biol 3 e100 [47] Lee SW Higashi DL Snyder A Merz AJ Potter L 2005 PilT is required for PI(3,4,5)P3-mediated crosstalk between Neisseria gonorrhoeae and epithelial cells. Cell Microbiol 7 1271 1284 [48] Kaplan JB Fine DH 2002 Biofilm dispersal of Neisseria subflava and other phylogenetically diverse oral bacteria. Appl Environ Microbiol 68 4943 4950 [49] Aho EL Keating AM McGillivray SM 2000 A comparative analysis of pilin genes from pathogenic and nonpathogenic Neisseria species. Microb Pathog 28 81 88 [50] Berger U Husmann D 1972 Studies on the normal occurrence of Neisseria flavescens (Branham, 1930). Med Microbiol Immunol 158 121 127 [51] Seifert HS Ajioka RS Paruchuri D Heffron F So M 1990 Shuttle mutagenesis of Neisseria gonorrhoeae : pilin null mutations lower DNA transformation competence. J Bacteriol 172 40 46 [52] Zerbino DR Birney E 2008 Velvet: algorithms for de novo short read assembly using de Bruijn graphs. Genome Res 18 821 829 [53] Aziz RK Bartels D Best AA DeJongh M Disz T 2008 The RAST Server: rapid annotations using subsystems technology. BMC Genomics 9 75 [54] Katoh K Kuma K Toh H Miyata T 2005 MAFFT version 5: improvement in accuracy of multiple sequence alignment. Nucleic Acids Res 33 511 518 [55] Castresana J 2000 Selection of conserved blocks from multiple alignments for their use in phylogenetic analysis. Mol Biol Evol 17 540 552 [56] Rice P Longden I Bleasby A 2000 EMBOSS: the European Molecular Biology Open Software Suite. Trends Genet 16 276 277 [57] Correia FF Inouye S Inouye M 1986 A 26-base-pair repetitive sequence specific for Neisseria gonorrhoeae and Neisseria meningitidis genomic DNA. J Bacteriol 167 1009 1015 [58] Yang Z 1997 PAML: a program package for phylogenetic analysis by maximum likelihood. Comput Appl Biosci 13 555 556 [59] Shimodaira H Hasegawa M 2001 CONSEL: for assessing the confidence of phylogenetic tree selection. Bioinformatics 17 1246 1247