Determination of the Genome Sequence of Porphyromonas gingivalis Strain ATCC 33277 and Genomic Comparison with Strain W83 Revealed Extensive Genome Rearrangements in P. gingivalis

The gram-negative anaerobic bacterium Porphyromonas gingivalis is a major causative agent of chronic periodontitis. Porphyromonas gingivalis strains have been classified into virulent and less-virulent strains by mouse subcutaneous soft tissue abscess model analysis. Here, we present the whole genome sequence of P. gingivalis ATCC 33277, which is classified as a less-virulent strain. We identified 2090 protein-coding sequences (CDSs), 4 RNA operons, and 53 tRNA genes in the ATCC 33277 genome. By genomic comparison with the virulent strain W83, we identified 461 ATCC 33277-specific and 415 W83-specific CDSs. Extensive genomic rearrangements were observed between the two strains: 175 regions in which genomic rearrangements have occurred were identified. Thirty-five of those genomic rearrangements were inversion or translocation and 140 were simple insertion, deletion, or replacement. Both strains contained large numbers of mobile elements, such as insertion sequences, miniature inverted-repeat transposable elements (MITEs), and conjugative transposons, which are frequently associated with genomic rearrangements. These findings indicate that the mobile genetic elements have been deeply involved in the extensive genome rearrangement of P. gingivalis and the occurrence of many of the strain-specific CDSs. We also describe here a very unique feature of MITE400, which we renamed MITEPgRS (MITE of P. gingivalis with Repeating Sequences).Key words: Porphyromonas gingivalis, whole genome sequence, genome rearrangement, conjugative transposon, MITE     

Analysis of the prtP gene encoding porphypain, a cysteine proteinase of Porphyromonas gingivalis.

The cloning and sequencing of the gene encoding porphypain, a cysteine proteinase previously isolated from detergent extracts of the Porphyromonas gingivalis W12 cell surface, are described. The prtP gene encoded a unique protein of 1,732 amino acids, including a putative signal sequence for protein secretion. The predicted molecular mass for the mature protein was 186 kDa, which was close to the observed molecular mass of 180 kDa. There was one copy of prtP in the genomes of seven P. gingivalis strains examined. The gene was located 5' to a region with a high degree of homology to the insertion element IS1126 in P. gingivalis W12. The PrtP protein had regions of high homology to HagA, a hemagglutinin of P. gingivalis, and to several purported proteinases of P. gingivalis that have Arg-X specificity. A detailed comparison of genes encoding the latter and cpgR suggested that rgp-1, prpR1, prtR, agp, cpgR, and possibly prtH were derived from identical genetic loci. Although an rgp-1-like locus was detected in seven P. gingivalis strains by Southern blot analyses, agp and cpgR were not detected, not even in the strains from which they were originally isolated. In addition, at least 20 copies of a repeat region common to PrtP, the Rgp-1-like proteins, and HagA were observed in each of the seven genomes examined. The repeat region hybridization patterns for strains W83 and W50 were very similar, and they were identical for strains 381 and ATCC 33277, providing further evidence that these strains are closely related genetically.     

Complete genome sequence and comparative analysis of the fish pathogen Lactococcus garvieae

Lactococcus garvieae causes fatal haemorrhagic septicaemia in fish such as yellowtail. The comparative analysis of genomes of a virulent strain Lg2 and a non-virulent strain ATCC 49156 of L. garvieae revealed that the two strains shared a high degree of sequence identity, but Lg2 had a 16.5-kb capsule gene cluster that is absent in ATCC 49156. The capsule gene cluster was composed of 15 genes, of which eight genes are highly conserved with those in exopolysaccharide biosynthesis gene cluster often found in Lactococcus lactis strains. Sequence analysis of the capsule gene cluster in the less virulent strain L. garvieae Lg2-S, Lg2-derived strain, showed that two conserved genes were disrupted by a single base pair deletion, respectively. These results strongly suggest that the capsule is crucial for virulence of Lg2. The capsule gene cluster of Lg2 may be a genomic island from several features such as the presence of insertion sequences flanked on both ends, different GC content from the chromosomal average, integration into the locus syntenic to other lactococcal genome sequences, and distribution in human gut microbiomes. The analysis also predicted other potential virulence factors such as haemolysin. The present study provides new insights into understanding of the virulence mechanisms of L. garvieae in fish.     

Bacterial genomic islands: organization, function, and role in evolution

Data on the structural organization and evolutionary role of specific bacterial DNA regions known as genomic islands are reviewed. Emphasis is placed on the most extensively studied genomic islands, pathogenicity islands (PAIs), which are present in the chromosome of Gram-negative and Gram-positive pathogenic bacteria and absent from related nonpathogenic strains. PAIs are extended DNA regions that harbor virulence genes and often differ in GC content from the remainder of the bacterial genome. Many PAI occur in the tRNA genes, which provide a convenient target for foreign gene insertion. Some PAI are highly homologous to each other and contain sequences similar to ISs, phage att sites, and plasmid ori sites, along with functional or defective integrase and transposase genes, suggesting horizontal transfer of PAI among bacteria.     

Pathway analysis for intracellular Porphyromonas gingivalis using a strain ATCC 33277 specific database

Background  

Porphyromonas gingivalis is a Gram-negative intracellular pathogen associated with periodontal disease. We have previously reported on whole-cell quantitative proteomic analyses to investigate the differential expression of virulence factors as the organism transitions from an extracellular to intracellular lifestyle. The original results with the invasive strain P. gingivalis ATCC 33277 were obtained using the genome sequence available at the time, strain W83 [GenBank: AE015924]. We present here a re-processed dataset using the recently published genome annotation specific for strain ATCC 33277 [GenBank: AP009380] and an analysis of differential abundance based on metabolic pathways rather than individual proteins.     

Bacterial Genomic Islands: Organization,Function, and Evolutionary Role

Data on the structural organization and evolutionary role of specific bacterial DNA regions known as genomic islands are reviewed. Emphasis is placed on the most extensively studied genomic islands, pathogenicity islands (PAIs), which are present in the chromosome of Gram-negative and Gram-positive pathogenic bacteria and absent from related nonpathogenic strains. PAIs are long DNA regions that harbor virulence genes and often differ in GC content from the remainder of the bacterial genome. Many PAI occur in the tRNA gene loci, which provide a convenient target for foreign gene insertion. Some PAI are highly homologous to each other and contain sequences similar to ISs, phage att sites, and plasmid ori sites, along with functional or defective integrase and transposase genes, suggesting horizontal transfer of PAI among bacteria.     

Intraspecific variability of the terminal inverted repeats of the linear chromosome of Streptomyces ambofaciens

The sequences of the terminal inverted repeats (TIRs) ending the linear chromosomal DNA of two Streptomyces ambofaciens strains, ATCC23877 and DSM40697 (198 kb and 213 kb, respectively), were determined from two sets of recombinant cosmids. Among the 215 coding DNA sequences (CDSs) predicted in the TIRs of strain DSM40697, 65 are absent in the TIRs of strain ATCC23877. Reciprocally, 45 of the 194 predicted CDSs are specific to the ATCC23877 strain. The strain-specific CDSs are located mainly at the terminal end of the TIRs. Indeed, although TIRs appear almost identical over 150 kb (99% nucleotide identity), large regions of DNA of 60 kb (DSM40697) and 48 kb (ATCC23877), mostly spanning the ends of the chromosome, are strain specific. These regions are rich in plasmid-associated genes, including genes encoding putative conjugal transfer functions. The strain-specific regions also share a G+C content (68%) lower than that of the rest of the genome (from 71% to 73%), a percentage that is more typical of Streptomyces plasmids and mobile elements. These data suggest that exchanges of replicon extremities have occurred, thereby contributing to the terminal variability observed at the intraspecific level. In addition, the terminal regions include many mobile genetic element-related genes, pseudogenes, and genes related to adaptation. The results give insight into the mechanisms of evolution of the TIRs: integration of new information and/or loss of DNA fragments and subsequent homogenization of the two chromosomal extremities.     

Identification of genes encoding hypothetical proteins in open-reading frame expressed sequence tags from mammalian stages of Trypanosoma cruzi

Approximately 50% of the predicted protein-coding genes of the Trypanosoma cruzi CL Brener strain are annotated as hypothetical or conserved hypothetical proteins. To further characterize these genes, we generated 1161 open-reading frame expressed sequence tags (ORESTES) from the mammalian stages of the VL10 human strain. Sequence clustering resulted in 435 clusters, consisting of 339 singletons and 96 contigs. Significant matches to the T. cruzi predicted gene database were found for ~94% contigs and ~69% singletons. These included genes encoding surface proteins, known to be intensely expressed in the parasite mammalian stages and implicated in host cell invasion and/or immune evasion mechanisms. Among 151 contigs and singletons with similarity to predicted hypothetical protein-coding genes and conserved hypothetical protein-coding genes, 83% showed no match with T. cruzi EST and/or proteome databases. These ORESTES are the first experimental evidence that the corresponding genes are in fact transcribed. Sequences with no significant match were searched against several T. cruzi and National Center for Biotechnology Information non-redundant sequence databases. The ORESTES analysis indicated that 124 predicted conserved hypothetical protein-coding genes and 27 predicted hypothetical protein-coding genes annotated in the CL Brener genome are transcribed in the VL10 mammalian stages. Six ORESTES annotated as hypothetical protein-coding genes showing no match to EST and/or proteome databases were confirmed by Northern blot in VL10. The generation of this set of ORESTES complements the T. cruzi genome annotation and suggests new stage-regulated genes encoding hypothetical proteins.     

Optical mapping and sequencing of the Escherichia coli KO11 genome reveal extensive chromosomal rearrangements, and multiple tandem copies of the Zymomonas mobilis pdc and adhB genes

Escherichia coli KO11 (ATCC 55124) was engineered in 1990 to produce ethanol by chromosomal insertion of the Zymomonas mobilis pdc and adhB genes into E. coli W (ATCC 9637). KO11FL, our current laboratory version of KO11, and its parent E. coli W were sequenced, and contigs assembled into genomic sequences using optical NcoI restriction maps as templates. E. coli W contained plasmids pRK1 (102.5 kb) and pRK2 (5.4 kb), but KO11FL only contained pRK2. KO11FL optical maps made with AflII and with BamHI showed a tandem repeat region, consisting of at least 20 copies of a 10-kb unit. The repeat region was located at the insertion site for the pdc, adhB, and chloramphenicol-resistance genes. Sequence coverage of these genes was about 25-fold higher than average, consistent with amplification of the foreign genes that were inserted as circularized DNA. Selection for higher levels of chloramphenicol resistance originally produced strains with higher pdc and adhB expression, and hence improved fermentation performance, by increasing the gene copy number. Sequence data for an earlier version of KO11, ATCC 55124, indicated that multiple copies of pdc adhB were present. Comparison of the W and KO11FL genomes showed large inversions and deletions in KO11FL, mostly enabled by IS10, which is absent from W but present at 30 sites in KO11FL. The early KO11 strain ATCC 55124 had no rearrangements, contained only one IS10, and lacked most accumulated single nucleotide polymorphisms (SNPs) present in KO11FL. Despite rearrangements and SNPs in KO11FL, fermentation performance was equal to that of ATCC 55124.     

Porphyromonas gingivalis Strain Specific Interactions with Human Coronary Artery Endothelial Cells: A Comparative Study

Both epidemiologic and experimental findings suggest that infection with Porphyromonas gingivalis exacerbates progression of atherosclerosis. As P. gingivalis exhibits significant strain variation, it is reasonable that different strains possess different capabilities and/or mechanisms by which they promote atherosclerosis. Using P. gingivalis strains that have been previously evaluated in the ApoE null atherosclerosis model, we assessed the ability of W83, A7436, 381, and 33277 to adhere, invade, and persist in human coronary artery endothelial (HCAE) cells. W83 and 381 displayed an equivalent ability to adhere to HCAE cells, which was significantly greater than both A7436 and 33277 (P<0.01). W83, 381, and 33277 were more invasive than A7436 (P<0.0001). However, only W83 and A7436 were able to remain viable up to 48 hours in HCAE cell cultures, whereas 381 was cleared by 48 hours and 33277 was cleared by 24 hours. These differences in persistence were in part due to strain specific differences in intracellular trafficking. Both W83 and 381 trafficked through the autophagic pathway, but not A7436 or 33277. Internalized 381 was the only strain that was dependent upon the autophagic pathway for its survival. Finally, we assessed the efficacy of these strains to activate HCAE cells as defined by production of IL-6, IL-8, IL-12p40, MCP-1, RANTES, TNF-α, and soluble adhesion molecules (sICAM-1, sVCAM-1, and sE-selectin). Only moderate inflammation was observed in cells infected with either W83 or A7436, whereas cells infected with 381 exhibited the most profound inflammation, followed by cells infected with 33277. These results demonstrate that virulence mechanisms among different P. gingivalis strains are varied and that pathogenic mechanisms identified for one strain are not necessarily applicable to other strains.     

In silico comparative study of the genomic islands of Vibrio cholerae MJ1236 with those of Classical and El Tor N16961 strains of Vibrio cholerae

The evolution of microbial genomes is greatly influenced by horizontal gene transfer (HGT), where large blocks of horizontally acquired foreign sequences, often encoding virulence determinants, occur in chromosomes of pathogenic bacteria. A program design-island developed in our laboratory was used on three completely sequenced Vibrio cholerae genomes, V. cholerae Classical O395, El Tor N16961 and MJ1236, in order to identify the putative horizontally acquired regions. The putative genomic islands (GIs) were graphically represented and analyzed. The study identified distinct regions in the GIs of V. cholerae MJ1236 which were shared either with the Classical or the El Tor strain of V. cholerae. A cluster comprising of 38 ORFs was common to V. cholerae strains of MJ1236 and Classical O395 but absent in El Tor N16961. About 5% of the predicted GIs of V. cholerae MJ1236 were unique to itself. Among these unique ORFs, a region of mostly hypothetical genes was identified, where the ORFs were present in a large cluster. The results show that the HGT had played a significant role in the evolution and the differentiation of V. cholerae MJ1236.     

A novel type of two-component regulatory system affecting gingipains in Porphyromonas gingivalis

We surveyed the Porphyromonas gingivalis W83 genome database for homologues of FimS, the first two-component sensor histidine kinase, which could possibly control virulence factors. Including fimS, we found six putative sensor kinase genes in the genome. The gene encoding one of the homologues was cloned from a P. gingivalis plasmid library, sequenced, and analyzed using its mutants. Two gene-disruption mutants were created in strain ATCC 33277 by introducing a drug cassette into the gene. The mutants formed nonpigmented colonies, indicating that they might be defective in proteinase production, a characteristic of this organism. Proteinase activities, measured as arginine- and lysine-specific (Rgp and Kgp gingipains, respectively) activities, of the mutants were almost half those of the parent strain. Unlike the parent and wildtype strains, most of the gingipain activities were detected in the culture supernatant, not in cells, of the mutants. Abnormal production of gingipains was confirmed by sodium dodecyl sulfate polyacrylamide gel electrophoresis and Western blot analyses. These results strongly suggest that this newly-discovered two-component sensor kinase is involved in maturation and proper localization of gingipains to the outer membrane through an unknown mechanism. The gene encoding the sensor histidine kinase was designated gppX, which represents regulation (X) of gingipains and black pigmentation in P. gingivalis.     

The opcA and ΨopcB regions in Neisseria: genes, pseudogenes, deletions, insertion elements and DNA islands

Previous data have indicated that the opc gene encoding an immunogenic invasin is specific to Neisseria meningitidis (Nm) and is lacking in Neisseria gonorrhoeae (Ng). The data presented here show that Nm and Ng both contain two paralogous opc-like genes, opcA, corresponding to the former opc gene, and (psi)opcB, a pseudogene. The predicted OpcA and OpcB proteins possess transmembrane regions with conserved non-polar faces but differ extensively in four of the five surface-exposed loops. Gonococcal OpcA was expressed weakly under in vitro conditions, and it is unknown whether these bacteria can express this protein at high levels. Analysis of the sequences flanking opcA and (psi)opcB revealed a framework of conserved housekeeping genes interspersed with DNA islands. These regions also contained several pseudogenes, deletions and IS elements, attesting to considerable genome plasticity. Both opcA and (psi)opcB are located on DNA islands that have probably been imported from unrelated bacteria. A third island encodes the dcmD/dcrD R/M genes in Ng versus a small open reading frame in most strains of Nm. Rare strains of Nm were identified in which the R/M island has been imported. DNA islands in Nm and Ng seem to have been acquired by recombination via conserved flanking housekeeping genes rather than by insertion of mobile genetic elements.     

Distribution and organization of auxotrophic genes on the multichromosomal genome of Burkholderia multivorans ATCC 17616

The Burkholderia multivorans strain ATCC 17616 carries three circular chromosomes with sizes of 3.4, 2.5, and 0.9 Mb. To determine the distribution and organization of the amino acid biosynthetic genes on the genome of this beta-proteobacterium, various auxotrophic mutations were isolated using a Tn5 derivative that was convenient not only for the determination of its insertion site on the genome map but also for the structural analysis of the flanking regions. Analysis by pulsed-field gel electrophoresis revealed that 20 out of 23 insertion mutations were distributed on the 3.4-Mb chromosome. More detailed analysis of the his, trp, arg, and lys mutations and their flanking regions revealed the following properties of these auxotrophic genes: (i) all nine his genes were clustered on the 3.4-Mb chromosome; (ii) seven trp genes were organized within two distinct regions, i.e., a trpEGDC cluster on the 3.4-Mb chromosome and a trpFBA cluster on the 2.5-Mb chromosome; (iii) the leu gene cluster, leuCDB, was also located close to the trpFBA cluster; and (iv) lysA and argG genes were located on the 2.5-Mb chromosome, in contrast to the argH gene, which was located on the 3.4-Mb chromosome. Southern hybridization analysis, allelic exchange mutagenesis of ATCC 17616, and complementation tests demonstrated that all of the genes examined were functional and existed as a single copy within the genome. The present findings also indicated that the 2.5-Mb chromosome carried various auxotrophic genes with no structural or functional counterparts on the remaining two chromosomes.     

Intraspecies Variability Affects Heterotypic Biofilms of Porphyromonas gingivalis and Prevotella intermedia: Evidences of Strain-Dependence Biofilm Modulation by Physical Contact and by Released Soluble Factors

It is well known that strain and virulence diversity exist within the population structure of Porphyromonas gingivalis. In the present study we investigate intra- and inter-species variability in biofilm formation of Porphyromonas gingivalis and partners Prevotella intermedia and Prevotella nigrescens. All strains tested showed similar hydrophobicity, except for P. gingivalis W83 which has roughly half of the hydrophobicity of P. gingivalis ATCC33277. An intraspecies variability in coaggregation of P. gingivalis with P. intermedia was also found. The association P. gingivalis W83/P. intermedia 17 produced the thickest biofilm and strain 17 was prevalent. In a two-compartment system P. gingivalis W83 stimulates an increase in biomass of strain 17 and the latter did not stimulate the growth of P. gingivalis W83. In addition, P. gingivalis W83 also stimulates the growth of P. intermedia ATCC25611 although strain W83 was prevalent in the association with P. intermedia ATCC25611. P. gingivalis ATCC33277 was prevalent in both associations with P. intermedia and both strains of P. intermedia stimulate the growth of P. gingivalis ATCC33277. FISH images also showed variability in biofilm structure. Thus, the outcome of the association P. gingivalis/P. intermedia seems to be strain-dependent, and both soluble factors and physical contact are relevant. The association P. gingivalis-P. nigrescens ATCC33563 produced larger biomass than each monotypic biofilm, and P. gingivalis was favored in consortia, while no differences were found in the two-compartment system. Therefore, in consortia P. gingivalis-P. nigrescens physical contact seems to favor P. gingivalis growth. The intraspecies variability found in our study suggests strain-dependence in ability of microorganisms to recognize molecules in other bacteria which may further elucidate the dysbiosis event during periodontitis development giving additional explanation for periodontal bacteria, such as P. gingivalis and P. intermedia, among others, to persist and establish chronic infections in the host.     

Whole-genome plasticity among Mycobacterium avium subspecies: insights from comparative genomic hybridizations

Infection with Mycobacterium avium subsp. paratuberculosis causes Johne's disease in cattle and is also implicated in cases of Crohn's disease in humans. Another closely related strain, M. avium subsp. avium, is a health problem for immunocompromised patients. To understand the molecular pathogenesis of M. avium subspecies, we analyzed the genome contents of isolates collected from humans and domesticated or wildlife animals. Comparative genomic hybridizations indicated distinct lineages for each subspecies where the closest genomic relatedness existed between M. avium subsp. paratuberculosis isolates collected from human and clinical cow samples. Genomic islands (n = 24) comprising 846 kb were present in the reference M. avium subsp. avium strain but absent from 95% of M. avium subsp. paratuberculosis isolates. Additional analysis identified a group of 18 M. avium subsp. paratuberculosis-associated islands comprising 240 kb that were absent from most of the M. avium subsp. avium isolates. Sequence analysis of DNA regions flanking the genomic islands identified three large inversions in addition to several small inversions that could play a role in regulation of gene expression. Analysis of genes encoded in the genomic islands reveals factors that are probably important for various mechanisms of virulence. Overall, M. avium subsp. avium isolates displayed a higher level of genomic diversity than M. avium subsp. paratuberculosis isolates. Among M. avium subsp. paratuberculosis isolates, those from wildlife animals displayed the highest level of genomic rearrangements that were not observed in other isolates. The presented findings will affect the future design of diagnostics and vaccines for Johne's and Crohn's diseases and provide a model for genomic analysis of closely related bacteria.     

Frame-shift mutations in NAD(P)H flavin oxidoreductase encoding gene (frxA) from metronidazole resistant Helicobacter pylori ATCC43504 and its involvement in metronidazole resistance

Metronidazole is a critical ingredient for combination therapies of Helicobacter pylori infection, the major cause of peptic ulcer and gastric cancer. It has been recently reported that metronidazole resistance from H. pylori ATCC43504 is caused by the insertion of a mini-IS605 sequence and deletion of sequences in an oxygen insensitive NAD(P)H nitroreductase encoding gene (rdxA). We also found that an additional gene (frxA) encoding NAD(P)H flavin oxidoreductase in the same strain was truncated by frame-shift mutations. To assess whether the frxA truncation is also involved in metronidazole resistance, metronidazole sensitive H. pylori strains ATCC43629 and SS1 were transformed by the truncated frxA gene cloned from strain ATCC43504. All transformed cells grew on agar plates containing 16 microg ml(-1) of metronidazole. The involvement of the frxA gene in metronidazole resistance was also confirmed by insertion inactivation of frxA and/or rdxA genes from strain ATCC43629 and one metronidazole sensitive clinical isolate H. pylori 2600. In addition, the frxA gene cloned from the H. pylori 2600 showed metronidazole nitroreductase activity in Escherichia coli and rendered ordinary metronidazole resistant E. coli to metronidazole sensitive cell. These results indicate that the frxA gene may also be involved in metronidazole resistance among clinical H. pylori isolates.     

Distribution and sequence analysis of a family of type ill-dependent effectors correlate with the phylogeny of Ralstonia solanacearum strains

In Ralstonia solanacearum, we previously have reported on the characterization of popP1 and popP2 genes. These genes encode type III-dependent pathogenicity effectors related to the large family of AvrRxv/YopJ cysteine proteases that are shared among pathogens of plants and animals. In this study, we identify a third gene, named popP3, that is inactivated in the genome sequence of strain GMI1000 by insertion of a copy of the insertion sequence ISRso13. The three popP genes are localized on two large chromosomal pathogenicity islands, with popP1 and popP2 being present on the same island. Phylogenic analysis demonstrated that the PopP2 and PopP3 proteins are clearly distinct from other effectors of this family previously characterized in plant and animal pathogens. Analysis of the distribution and allelic variations of the three genes in 30 strains representative of the biodiversity of R. solanacearum established that popP genes are distributed widely among strains from two of the three phyla previously defined on the basis of the structure of the core genome. Sequencing of the popP genes from the different strains revealed limited allelic variations at the three loci but did not show evidence of recombination between the popP genes. Limited allelic variation together with occurrence of insertion sequences within or in the close vicinity of popP genes and the presence of gene duplications in these pathogenicity islands suggest that genomic rearrangements might be a major evolutionary driving force controlling evolution of the genes encoded in these regions. The implications of these observations in terms of bacterial evolution, gene acquisition, and horizontal gene transfers are discussed.