The anaerobic spirochete Brachyspira pilosicoli colonizes the large intestine of various species of birds and mammals, including humans. It causes “intestinal spirochetosis”, a condition characterized by mild colitis, diarrhea and reduced growth. This study aimed to sequence and analyse the bacterial genome to investigate the genetic basis of its specialized ecology and virulence.
The genome of B. pilosicoli 95/1000 was sequenced, assembled and compared with that of the pathogenic Brachyspira hyodysenteriae and a near-complete sequence of Brachyspira murdochii. The B. pilosicoli genome was circular, composed of 2,586,443 bp with a 27.9 mol% G+C content, and encoded 2,338 genes. The three Brachyspira species shared 1,087 genes and showed evidence of extensive genome rearrangements. Despite minor differences in predicted protein functional groups, the species had many similar features including core metabolic pathways. Genes distinguishing B. pilosicoli from B. hyodysenteriae included those for a previously undescribed bacteriophage that may be useful for genetic manipulation, for a glycine reductase complex allowing use of glycine whilst protecting from oxidative stress, and for aconitase and related enzymes in the incomplete TCA cycle, allowing glutamate synthesis and function of the cycle during oxidative stress. B. pilosicoli had substantially fewer methyl-accepting chemotaxis genes than B. hyodysenteriae and hence these species are likely to have different chemotactic responses that may help to explain their different host range and colonization sites. B. pilosicoli lacked the gene for a new putative hemolysin identified in B. hyodysenteriae WA1. Both B. pilosicoli and B. murdochii lacked the rfbBADC gene cluster found on the B. hyodysenteriae plasmid, and hence were predicted to have different lipooligosaccharide structures. Overall, B. pilosicoli 95/1000 had a variety of genes potentially contributing to virulence.
The availability of the complete genome sequence of B. pilosicoli 95/1000 will facilitate functional genomics studies aimed at elucidating host-pathogen interactions and virulence.
Citation: Wanchanthuek P, Bellgard MI, La T, Ryan K, Moolhuijzen P, et al. (2010) The Complete Genome Sequence of the Pathogenic Intestinal Spirochete Brachyspira pilosicoli and Comparison with Other Brachyspira Genomes. PLoS ONE 5(7): e11455. doi:10.1371/journal.pone.0011455
Editor: Niyaz Ahmed, University of Hyderabad, India
Received: May 3, 2010; Accepted: June 13, 2010; Published: July 6, 2010
Copyright: © 2010 Wanchanthuek et al. This is an open-access article distributed under the terms of the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original author and source are credited.
Funding: This study was supported by the Australian Research Council and Novartis Animal Vaccines through Linkage grant number LP0348441 for genomic sequencing and comparative genomic analysis for animal bacterial vaccine discovery, the Western Australian Government in establishment of the WA Centre of Excellence for Comparative Genomics and for support for this project, and the Mahasarakham University Research Affairs Division. The funders had no role in study design, data collection and analysis, decision to publish, or preparation of the manuscript.
Competing interests: This study included identifying bacterial proteins that could be useful for a vaccine. Part of this project was funded by Novartis Animal Vaccines. Several of the proteins that were identified are being patented by Spirogene Pty, a Murdoch University spin-out company. This does not alter the authors' adherence to all the PLoS ONE policies on sharing data and materials.
Spirochetes of the genus Brachyspira are anaerobic bacteria that colonize the large intestine of animals and birds . There are currently seven officially named species in the genus, including both pathogenic and commensal representatives. The names of the species and their phylogenetic relationships based on their 16S rRNA gene sequences are shown in Figure 1. The two main pathogenic species are Brachyspira hyodysenteriae, the agent of a major pig disease called swine dysentery, and Brachyspira pilosicoli, the cause of a condition known as intestinal (or colonic) spirochetosis. B. pilosicoli has a wider host range than B. hyodysenteriae, and colonizes a variety of species, including human beings –. Infections with B. pilosicoli are particularly common in intensively housed pigs and chickens: colonized individuals may develop focal areas of inflammation in the large intestine, with diarrhea and reduced rates of growth and production. Colonization with B. pilosicoli also occurs at a high prevalence rate in human beings living in crowded and unhygienic conditions, particularly in developing countries –, as well as amongst homosexual males . Despite the potential importance of B. pilosicoli as a pathogen, it has not been extensively studied. Progress has been hampered by the spirochete's specialised growth requirements and slow growth rate, as well as by a lack of genomic information and an absence of means for genetic manipulation.
Figure 1. Phylogenetic tree constructed using the UPGMA method showing the relationships of the seven named Brachyspira species based on the full sequence of their 16S rRNA genes.
The names of the strains that were analysed in this study are marked in bold. 0.002 on the scale bar represents two substitutions in 1000 bp.doi:10.1371/journal.pone.0011455.g001
A characteristic feature of infection with B. pilosicoli is the attachment of the spirochete by one cell end to the luminal surface of enterocytes in the large intestine; with time, dense mats of these attached bacteria may form a “false brush-border” covering the whole luminal surface of the enterocytes , . Experimentally, following attachment to cultured Caco-2 cells B. pilosicoli has been shown to induce apoptosis, actin rearrangement and increased expression of interleukin-1ß (IL-1ß) and IL-8 . Such pathological changes may contribute to the local colitis and diarrhea that is observed in vivo.
The recent publication of the genome sequence of B. hyodysenteriae strain WA1  and the availability of a near-complete genome sequence of Brachyspira murdochii type strain 56-150T (DSM 12563) in GenBank represent important opportunities to advance research on these Brachyspira species. B. murdochii is generally considered to be a harmless commensal in pigs, chickens and rats , ; however, there have been recent reports that it may have some potential to cause colitis in pigs , . In the current study the genome of B. pilosicoli strain 95/1000 was sequenced and subjected to comparative genomic analysis, particularly in relation to the sequences of WA1 and 56-150T. As the sequence of the latter was incomplete some comparisons were not appropriate. The overall aim of this work was to enhance knowledge about B. pilosicoli and its relationships to the other Brachyspira species, particularly in regard to identifying the genetic basis of their different ecologies and pathogenic potentials.
Materials and Methods
Spirochete strain and growth conditions
Brachyspira pilosicoli strain 95/1000 (ATCC BAA-1826) was originally isolated in Western Australia from the diarrheic feces of a commercial pig with porcine intestinal spirochetosis . The spirochete was purified by repeated subculture, grown to mid-log phase in pre-reduced anaerobic broth , and a cell pellet prepared.
Genomic DNA preparation, library construction and sequencing
Preparation of genomic DNA, library construction and sequencing was as previously described for B. hyodysenteriae . Sequencing was undertaken at the Australian Genome Research Facility. The first round of sequencing was performed via Sanger sequencing, with a total of 42,565 reads generated. The second round was performed using a pyrosequencing approach on a Roche-454 GS20 instrument, generating more than 25 times coverage of the genome. The quality filtered reads were assembled into contiguous sequences using the Newbler Assembler software (http://www.454.com/). To finish the genome sequence, remaining gaps were closed by PCR walking between un-linked contiguous sequences .
Sequence analysis and annotation
For both the newly sequenced B. pilosicoli 95/1000 and for the incomplete genome sequence of B. murdochii 56-150T obtained from GenBank (ABTG00000000), sequence analysis and annotation were as previously described for B. hyodysenteriae WA1 . Dot matrix plots comparing the genomes of 95/1000, 56-150T and WA1 were generated using Freckle, an in-house development of the Dotter tool . The minimum size of matched sequences was set to 20 base pairs (bp).
The complete nucleotide sequence and annotation of B. pilosicoli 95/1000 has been deposited in GenBank (accession number CP002025; Project ID: 48097). Annotations and functional assignments for B. pilosicoli 95/1000 and for B. murdochii 56-150T also can be accessed at the CCG website (http://ccg.murdoch.edu.au/spirochaetales/).
Results and Discussion
General genome features
The general features of the genome of B. pilosicoli 95/1000, together with those of B. hyodysenteriae WA1 and the near-complete genome of B. murdochii 56-150T are summarised in Table 1. The complete genome of B. pilosicoli 95/1000 consisted of a single circular chromosome of 2,586,443 bp (Figure 2), making it the smallest of the three Brachyspira genomes analysed to date (with B. murdochii being the largest). The overall G+C content of the B. pilosicoli genome was 27.9 mol%. Unlike B. hyodysenteriae WA1, B. pilosicoli 95/1000 did not contain a ~36 Kb plasmid. Based on the available sequence it also appeared unlikely that B. murdochii contained a similar plasmid. The B. pilosicoli genome encoded 2,338 genes, with an overall 85% coding region. In comparison, B. murdochii 56-150T had 3,055 predicted genes in the seven contigs that were available. The percent coding region was similar across the three genomes. The average size of the predicted genes in B. pilosicoli 95/1000 was 997 bp, and tentative function was assigned to 1,645 of these. The B. pilosicoli strain had 569 more functionally assigned genes than did B. murdochii 56-150T and 282 less than B. hyodysenteriae. B. pilosicoli 95/1000 had 655 genes with unknown function and this was somewhat fewer than for B. hyodysenteriae (704), but much fewer than the extremely high number of unmatched genes for B. murdochii (1,925). A total of 1,201 (51.4%) genes in B. pilosicoli had matches in the COG database and 1,410 (60.3%) had matches in the KEGG database (Table 1). Of these, only 426 (18.2%) were assigned to enzymes (EC number) and 635 (27.2%) were involved in KEGG pathways. Of the B. pilosicoli open reading frames (ORFs) screened, 244 were predicted to encode proteins with signal peptides and 48 were predicted to have transmembrane helices. Eleven genes encoded proteins predicted to have seven transmembrane helices.
Figure 2. Genome of B. pilosicoli 95/1000.
Circles outer to inner: Genes, forward strand, COG coded; Genes, reverse strand, COG coded; tRNA; rRNA; GC skew; AT skew. All genes are color-coded according to Cluster of Orthologous Gene Categories (COG) functions: violet for translation, ribosomal structure and biogenesis; plum for RNA processing and modification; pink for transcription; deep pink for DNA replication, recombination and repair; hot pink for chromatin structure and dynamics; wheat for cell division and chromosome partitioning; light salmon for nuclear structure; yellow for defence mechanisms; gold for signal transduction mechanisms; pale green for cell envelope biogenesis, outer membrane; spring green for cell motility and secretion; lawn green for cytoskeleton; yellow green for extracellular structures; aquamarine for intracellular trafficking, secretion, and vesicular transport; medium aquamarine for posttranslational modification, protein turnover, chaperones; cyan for energy production and conversion; deep sky blue for carbohydrate transport and metabolism; sky blue for amino acid transport and metabolism; light slate blue for nucleotide transport and metabolism; orchid for coenzyme metabolism; medium orchid for lipid metabolism; dark orchid for inorganic ion transport and metabolism; blue violet for secondary metabolites biosynthesis, transport and catabolism; slate grey for general function prediction only; grey for function unknown; grey for not in COGs; black for tRNA.doi:10.1371/journal.pone.0011455.g002
Table 1. General genomic features predicted for B. pilosicoli 95/1000 and comparison with those of B. hyodysenteriae WA1 and B. murdochii 56-150T.doi:10.1371/journal.pone.0011455.t001
The three Brachyspira species had similar numbers of tRNA genes, representing all 20 amino acids (Table 1). As with B. hyodysenteriae , only single copies of the 16S, 23S, and 5S ribosomal RNA genes were found in B. pilosicoli and B. murdochii. In B. pilosicoli 95/1000, the rrs and rrl genes were closely linked, with the rrl (16S) gene being about 645 Kb from the other two genes. This rRNA gene organisation has been noted earlier; it distinguishes the Brachyspira species from other spirochetes, and presumably pre-dates speciation in the Brachyspira genus . Although spirochetes have a monophyletic origin, the copy number and organisation of rRNA genes differ in the different genera. For instance, the Treponema pallidum rRNA genes appear to be arranged in two typical rrn operons , . A single rRNA locus is found in most Borrelia species, with rrs separated from rrl and rrf by a small segment of DNA (~4 Kb). In Borrelia burgdorferi the rrf-rrl cluster is duplicated and is found immediately adjacent to the rrs-rrl-rrf cluster . Pathogenic Leptospira species possess two copies each of rrs and rrl and one copy of rrf . The non-pathogenic Leptospira biflexa contains two copies of each rRNA gene , and these are dispersed around the genome. Differences in the sequences of the 16S rRNA genes have been used to examine phylogenetic relationships between spirochete genera and species . On this basis, in the case of the Brachyspira species, B. pilosicoli is more distantly related to B. hyodysenteriae than is B. murdochii (Figure 1).
The origin and terminus of replication in the genome of B. pilosicoli 95/1000 were predicted based on the position of dnaA (BP951000_0595), as bacterial chromosome replication origins are typically located near this gene. The putative oriC origin of replication was identified in an AT-rich intergenic region upstream of dnaA, in the vicinity of a cluster of hypothetical DnaA boxes. Similar putative oriCs with comparable DnaA boxes have been identified in other spirochete genomes (Figure 3). The original of replication in B. murdochii was not found due to the genome sequence being incomplete. The putative origins of replication are located centrally within the most highly conserved and synteneous regions of the various spirochete genomes (Figure 3). B. pilosicoli contained the genes grpE, dnaK, hyp, ark, hyp, arg and gyrA downstream of the dnaA, an arrangement similar to B. hyodysenteriae WA1 . B. pilosicoli had a unique hypothetical coding region immediately upstream of the dnaA gene, whereas there were three hypothetical protein encoding genes upstream of dnaA in B. hyodysenteriae. This finding suggests that in B. pilosicoli 95/1000 and B. hyodysenteriae WA1 oriC relocated during evolution, presumably as the result of a DNA rearrangement. The differences in the origins of replication compared to other spirochetes suggest that there could be different mechanisms for replication of the spirochete chromosomes. Experimental studies will be required to verify the origin of replication in B. pilosicoli, as has been accomplished for Borrelia burgdorferi .
Figure 3. Comparison of the B. pilosicoli and B. hyodysenteriae putative origins of replication (oriC) region with that of other spirochetes.
Homologous genes commonly found at bacterial origins are indicated in similar colors. dnaA is indicated in red and hypothetical proteins are indicated in grey. Conservation of a cluster of genes is located around the origin of replication in several spirochetes. The genes encoding the putative proteins and the origins of replication are indicated.doi:10.1371/journal.pone.0011455.g003
A dot plot comparison of the genomes of B. pilosicoli, B. hyodysenteriae and B. murdochii showed no large scale conservation of gene order, but only conservation of some genes in clusters (Figure 4). Gene cluster conservation appeared to be greatest between B. hyodysenteriae and B. murdochii, consistent with these two species being phylogenetically more closely related to each other than to B. pilosicoli. A 26 Kb region of unknown significance that is partially conserved in B. hyodysenteriae WA1, B. pilosicoli 95/1000, Enterococcus faecalis and Escherichia coli , also was identified in B. murdochii. Alignment of the two complete Brachyspira genomes using the Artemis Comparison Tool  identified extensive gene rearrangements between them (Figure S1 in supporting information), consistent with the findings of an earlier comparison of partial physical maps of the two species .
Figure 4. Dot matrix plots comparing the genomes of 95/1000, 56-150T and WA1 generated using Freckle.
The output displays a two dimensional plot with dots representing matched regions between the three genomes. The minimum size of matched sequences was set to 20 bp.doi:10.1371/journal.pone.0011455.g004
An exception to the apparent general lack of conservation of gene order between the three Brachyspira genomes was a high conservation of the operons encoding ribosomal proteins. B. pilosicoli 95/1000 contained genes for 33 r-proteins organised in a 18 Kb region, whereas the ribosome clusters in the B. hyodysenteriae and B. murdochii strains included a total of 32 and 36 genes encoding r-proteins organised in 15 Kb and 18 Kb regions, respectively (Figure 5). The sequences of the B. pilosicoli r-proteins in this cluster were highly similar to the B. hyodysenteriae and B. murdochii homologs. Ribosomal protein genes are interesting because of their ubiquity and similar conservation rate such that horizontal transfer between lineages is unlikely. Apparently there has been a positive selection for clustering of these physically interacting proteins in the sequenced Brachyspira species, whilst there has been no absolute requirement for juxtaposition of other genes in the genomes. Synteny therefore has been lost at a much faster rate than is useful for prediction of gene function. This is consistent with the situation in other bacterial species, where orthologous genes are not necessarily located at the same relative position in the genomes, and only certain gene clusters are syntenic . Genomes of closely related species typically maintain a high degree of synteny , whereas genomes of moderately distant species, such as these Brachyspira species, usually have no significant overall synteny .
Figure 5. Conservation of the gene cluster for ribosomal proteins in B. pilosicoli 95/1000, B. hyodysenteriae WA1 and B. murdochii 56-150T.doi:10.1371/journal.pone.0011455.g005
Functional predictions amongst the Brachyspira species
The COG categories of the protein-coding genes are shown in Table 2. Generally there were few significant differences amongst the three genomes, although the smaller genome of B. pilosicoli 95/1000 contained more genes than the other two species in five categories: (N) Cell motility, (U) Intracellular trafficking, secretion, and vesicular transport, (C) Energy production and conversion, (H) Co-enzyme transport and metabolism, and (I) Lipid transport and metabolism. On the other hand, B. hyodysenteriae had more genes functioning in (K) Transcription, (V) Defense mechanisms, (T) Signal transduction mechanisms, (O) Posttranslational modification, protein turnover, chaperones, (G) Carbohydrate transport and metabolism, (E) Amino acid transport and metabolism, (P) Inorganic ion transport and metabolism, and (Q) Secondary metabolites biosynthesis, transport and catabolism. B. murdochii had more genes involved with (L) Replication, recombination and repair, and (M) Cell wall/membrane/envelope biogenesis. It was noteworthy that B. pilosicoli had the largest number of genes involved in energy production and conversion, and this capacity may enhance its potential to colonize various hosts.
Table 2. Distribution of Cluster of Orthologous Genes (COGs) in B. pilosicoli 95/1000, B. hyodysenteriae WA1 and B. murdochii 56-150T, showing the number and percentage of proteins within a genome assigned to each functional group (the B. murdochii genome is incomplete).doi:10.1371/journal.pone.0011455.t002
Previously it has been shown that B. hyodysenteriae lacks many genes for the biosynthesis of small molecules, and therefore must acquire these from the environment . Interestingly, B. pilosicoli was predicted to contain substantially fewer genes involved in inorganic ion transport and metabolism (P) than both B. hyodysenteriae and B. murdochii (53 versus 74 and 73, respectively; Table 2). Furthermore, it is not known which B. pilosicoli genes may act to compensate for the shortage of such biosynthetic pathways, since the substrates of many of the genes regarded as encoding transporters due to their possession of the motif sequences were unknown. These differences were unexpected, given that B. pilosicoli has a wider host range than the other two Brachyspira species, and hence might be predicted to have a greater metabolic capacity and flexibility in order to survive in these more varied nutritional environments.
Global gene comparisons between the Brachyspira species
As expected, comparative analysis of the B. pilosicoli genome across available microbial genomes in the non-redundant (nr) database at NCBI identified greatest similarities with B. hyodysenteriae and B. murdochii. However, as with B. hyodysenteriae , the next highest levels of similarities at the protein level were with Clostridium species (~10%) and E. coli (~5%). The three sequenced Brachyspira species contained a total of 1,087 conserved or “core” genes (Figure 6). It would be instructive to determine whether these same genes are conserved in other Brachyspira species. The majority of genes in B. pilosicoli (1,769, 77%) were conserved in B. hyodysenteriae; therefore, most of the genome-inferred metabolic potential of B. hyodysenteriae described previously can be extrapolated to B. pilosicoli. Only an additional 99 genes were conserved between B. pilosicoli and B. murdochii, whereas an additional 311 were conserved between B. hyodysenteriae and B. murdochii. B. hyodysenteriae had 1,014 genes that were not shared with B. pilosicoli, whereas B. murdochii contained a remarkable 1,900 genes not found in B. pilosicoli. On this basis, B. pilosicoli seemed to be more similar to B. hyodysenteriae than to B. murdochii. A large proportion of the unique genes in both B. hyodysenteriae and B. murdochii were of unknown function (Table 1).
Figure 6. Global gene conservation in the three Brachyspira species strains.
Each circle represents the total number of gene types in each genome. Overlapping regions depict the number of genes types shared between the respective genomes. The numbers outside the circles indicate the total number of CDS identified in each genome.doi:10.1371/journal.pone.0011455.g006
Although the B. hyodysenteriae genome was ~450 Kb larger than that of B. pilosicoli, both species contained roughly the same number of gene across the identified functional gene categories, and only a relatively small number of genes in B. pilosicoli were obviously different from those in B. hyodysenteriae. Of the 525 potential genes that were specific for B. pilosicoli, 319 were of unknown function. The remaining 206 functionally annotated B. pilosicoli-specific genes included those predicted to be involved in energy production, carbohydrate metabolism, amino acid metabolism, capsule biosynthesis (although B. pilosicoli is not known to have a capsule), or encoding transcriptional regulators, transporters and predicted surface antigens, as well as forming a bacteriophage region. Selected differences found in B. pilosicoli that appear to be potentially significant are described below.
Brachyspira pilosicoli 95/1000 contained a 25 Kb region (location 1,595,515–1,615,675 bp) encoding components of an integrated bacteriophage (Figure 7). The 29 associated genes in the region were absent from B. hyodysenteriae and B. murdochii, apart from the OrfG VSH-1 protein homolog that has been associated with a gene transfer agent in B. hyodysenteriae. Sixteen genes were identified as encoding hypothetical proteins, whilst 13 were bacteriophage-associated: putative integrase XerDC family protein (BP951000_1455); OrfG VSH-1 phage protein homolog (BP951000_1459); putative Lys - endolysin; glycoside hydrolase (BP951000_1461); putative RNA polymerase domain-containing protein (BP951000_1463); phage tail tape measure protein homolog, TP901 family (BP951000_1469); bacteriophage TP901-1 like family protein homolog (BP951000_1474); putative DNA packaging, Phage QLRG family protein (BP951000_1476); phage major capsid protein homolog, HK97 family (BP951000_1477); phage prohead protease (BP951000_1478); putative phage portal protein, HK97 family (BP951000_1479); DNA methylase N-4/N-6 domain protein homolog (BP951000_1480); putative phage terminase, small subunit, P27 family protein (BP951000_1481); and putative phage terminase, large subunit (BP951000_1482). The GC content in this bacteriophage region was 29.65%, which was slightly higher than the average for the whole genome (27.90%). It is not yet clear whether this integrated bacteriophage is functional, or capable of transferring genetic material between B. pilosicoli strains. If it is, it could be developed into a tool for genetic manipulation of the spirochete. By analogy with other prophages, it potentially also could be involved in lysogenic conversion, for example modifying the spirochete phenotype to enhance its virulence . It will be important to determine how widespread and conserved this bacteriophage region is amongst B. pilosicoli strains. B. hyodysenteriae WA1 also contained two genes predicted to be bacteriophage-associated: phage terminase large subunit (xtmB, BHWA1_01969) and integrase (BHWA1_02688), but these were distinct from the bacteriophage genes found in B. pilosicoli. A predicted integrase-recombinase protein (4083292.C14.orf00918) was found in B. murdochii, but it only had 25.61% identity at the protein level.
Figure 7. Organisation of the bacteriophage genome in B. pilosicoli 95/1000.doi:10.1371/journal.pone.0011455.g007
The novel bacteriophage region in B. pilosicoli 95/1000 was distinct from the smaller prophage-like gene transfer agent (GTA) that previously was described in the genomes of B. pilosicoli 95/1000 and in B. hyodysenteriae WA-1 , and which was originally described in B. hyodysenteriae B204 as VSH-1 , . The genes of the GTAs in 95/1000 and WA1 were contiguous, but showed gene rearrangements. Elements of the GTA also have been detected in different contigs of a partial genome sequence of B. intermedia HB60 . In the current analysis OrfE was found immediately in front of the other 17 GTA genes that were previously described in B. pilosicoli 95/1000 . B. murdochii 56-150T was found to have a cluster of 31 genes on contig 4083292.C21 (orf00108–orf00140) that included many of the described components of the GTA, as well as 17 hypothetical proteins. Genes for Hvp19/Hvp22, Hvp13, Hvp38, and Hvp28 were not identified in the partial genome of B. murdochii. Again, the known genes showed rearrangements from the order found in B. hyodysenteriae and B. pilosicoli, and it is unclear whether the B. murdochii GTA is functional.
Genes of the glycine reductase complex system.
B. pilosicoli 95/1000 and B. murdochii 56-150T differed from B. hyodysenteriae WA1 in that they possessed nine genes within a glycine reductase complex (grd cluster): this arrangement is similar to that in the anaerobic bacterium Clostridium sticklandii  (Figure 8). The locus in B. pilosicoli included genes encoding the GrdX protein (grdX, BP951000_1852), thioredoxin reductase (trxB, BP951000_1853), glycine reductase (grdE; BP951000_1854), sarcosine reductase (grdA, BP951000_1855), glycine reductase complex selenoprotein B (grdB, BP951000_1856), two copies for the glycine/sarcosine/betaine reductase complex, component C, alpha subunit (grdC, BP951000_1857 and BP951000_1858), and two copies for a sodium:alanine symporter family protein (BP951000_1859 and BP951000_1860). In B.murdochii, the locus consisted of genes encoding putative glycine reductase complex component (grdX, 4083292.C42.orf00552) thioredoxin reductase (trxB, 4083292.C42.orf00553), glycine/sarcosine/betaine reductase component B alpha/beta subunit (grdE; 4083292.C42.orf00554), two copies of glycine reductase complex selenoprotein A (grdA, 4083292.C42.orf00556 and 4083292.C42.orf00557), two copies of selenoprotein B, glycine/betaine/sarcosine/D-proline reductase family (grdB, 4083292.C42.orf00558 and 4083292.C42.orf00559), 3-oxoacyl-(acyl-carrier-protein) synthase III (fabH, 4083292.C42.orf00560), and fatty acid/phospholipid biosynthesis enzyme (plsX, 4083292.C42.orf00561).
Figure 8. Physical map of the glycine reductase genes and adjacent regions in B. pilosicoli 95/1000, B. murdochii 56-150T and Clostridium sticklandii.doi:10.1371/journal.pone.0011455.g008
ORF grdX preceded the thioredoxin/glycine reductase gene cluster in both B. pilosicoli and B. murdochii, and it is known to be co-transcribed in C. sticklandii . An additional two copies of trxB were identified in the B. pilosicoli genome (BP951000_1651 and BP951000_1853). Other genes in B. pilosicoli and B. murdochii that were outside the cluster and which encoded proteins with predicted selenocysteine sites were thioredoxin (BP951000_0519 and BP951000_1639; 4083292.C42.orf00619 and 4083292.C42.orf00482) and glutathione peroxidase (BP951000_2065 and BP951000_2066; 4083292.C16.orf00268). Given the potential role of these selenoproteins in antioxidant functions, deficiency of selenium may have detrimental effects on their ability to withstand oxidative stress.
The glycine reductase complex catalyses the reductive deamination of glycine to acetylphosphate and ammonia with the generation of ATP from ADP and orthophosphate . B. pilosicoli and B. murdochii therefore were predicted to be able to utilise glycine anaerobically as a sole source of carbon and energy, and hence have a distinct energy-conserving mechanism using an internal reaction in which glycine serves as an electron donor during oxidation by a glycine cleavage system, or as an electron acceptor when being reduced by glycine reductase. In contrast, the lack of this complex in B. hyodysenteriae suggested that it is unable to ferment glycine to act as a carbon and energy source. The difference between the species suggests that high levels of glycine might favor B. pilosicoli and B. murdochii populations over B. hyodysenteriae, either in vitro or in vivo. The glycine reductase complex is predicted to play an important role in allowing B. pilosicoli to successful colonize a broader range of host species than B. hyodysenteriae, and it also may confer advantages to B. murdochii. Clearly, further research is required to determine whether glycine does enhance growth of B. pilosicoli and B. murdochii, as predicted.
The absence of the grd cluster in B. hyodysenteriae may make it unable to synthesise glutathione and glutaredoxins, and therefore it was predicted to be less able to withstand oxidative stress than the other two species. Only two copies of a gene encoding thioredoxin reductase were identified in B. hyodysenteriae (trxB, BHWA1_02087 and BHWA1_00602). The predicted ability of B. pilosicoli to rapidly respond to an oxidative stress or a redox insult was consistent with its reported ability to survive outside the host for a longer period than B. hyodysenteriae , , and also may explain why it is able survive in more oxygenated host tissues than B. hyodysenteriae, such as in the bloodstream of immunocompromised patients . B. murdochii also has been isolated from extra-intestinal sites . This predicted ability to withstand oxidative stress also may make B. pilosicoli more adaptable in terms of its abilities to colonize the large intestines of a wide variety of host species, where environmental conditions are likely to vary.
Three copies of genes encoding sulfatases (BP951000_0858, BP951000_0859 and BP951000_0861) were identified in B. pilosicoli 95/1000, whilst one putative copy was found in B. murdochii (4083292.C42.orf00884), and none in B. hyodysenteriae. Sulfatases play important roles in the cycling of sulfur in the environment, in the degradation of sulfated glycosaminoglycans and glycolipids in the lysosome, and in remodeling sulfated glycosaminoglycans in the extracellular space. The sulfatase genes potentially could encode for enzymes that modify gycosaminoglycans to generate binding sites required for attachment of B. pilosicoli .
A novel set of two copies of genes encoding alpha-1, 2-fucosyltransferase (fucT, BP951000_1232 and BP951000_1235) and beta-1, 3-galactosyltransferase (BP951000_1768) were identified in B. pilosicoli 95/1000: these are key enzymes in the biosynthesis of Lewis antigens, structures found on the surface of human erythrocytes and epithelial cells. The gastric pathogen Helicobacter pylori can express Lewis and related antigens in the O-chains of its surface lipopolysaccharide , and this activity is believed to be important for bacterial persistence, immune evasion, and possibly generation of inflammation . Further work is required to determine whether B. pilosicoli expresses Lewis antigens, and to investigate their potential involvement in the pathogenesis of infection by this spirochete species.
Three copies of a gene encoding a sialidase (neuraminidase) family protein homolog (nanA, BP951000_2021, BP951000_2022 and BP951000_2023) were identified in B. pilosicoli 95/1000, but not in B. hyodysenteriae or B. murdochii. NanA proteins are produced by a wide variety of mucosal pathogens, and are a potential virulence factor in bacteria , . As they are widespread and conserved among a very broad range of important human pathogens, this implies that they have a critical role in microbial ecology . These enzymes may enhance B. pilosicoli colonization or induce tissue damage.
B. pilosicoli and B. murdochii were both found to have a gene encoding aconitase (BP951000_0370 and 4083292.C42.orf00530, respectively), and this was not present in B. hyodysenteriae. Aconitase is a tricarboxylic acid (TCA) cycle protein that catalyses the conversion of citrate to isocitrate, and, amongst other things, has been implicated in the virulence of Staphylococcus aureus . It is possible that the aconitase gene may contribute to the control of virulence factor synthesis . As none of the three anaerobic Brachyspira species contained genes for a complete TCA cycle, this suggests that retention of the aconitase gene may have an important functional significance. Two other genes were located adjacent to the aconitase gene in both B. pilosicoli and B. murdochii, these being putative citrate synthase (BP951000_0368 and 4083292.C42.orf00523) and putative isocitrate dehydrogenase (BP951000_0369 and 4083292.C42.orf00525). They were not identified in B. hyodysenteriae WA1. The presence of aconitase implies that B. pilosicoli and B murdochii can utilise the TCA to produce glutamate, and this capacity is discussed later in the section on central metabolic pathways.
Potential virulence factor screening
Independent of the previous analysis, screening of all coding sequences (CDS) of the three species using Blast, SignalP, PSORT and TMHMM to look for similarities to genes reported to be associated with virulence in other bacteria identified a total of 235 genes with putative roles in virulence in B. pilosicoli 95/1000, compared to 303 in B. hyodysenteriae WA1 and 142 in B. murdochii 56-150T. The number of genes in the different categories for all three Brachyspira species is shown in Table 3. Interpretation of the results for B. murdochii was complicated by the fact that the genome was incomplete, and additional genes are likely to be identified once a complete genome sequence becomes available. Overall, apart from what is shown, the predicted gene products did not have significant similarities with those of other well-characterized toxins or adhesins described in the major species of enteropathogenic bacteria - such as in E. coli or Clostridium species. Nevertheless, it is important to remember that other virulence-associated genes are likely to exist amongst those that are currently recorded as encoding proteins that are hypothetical or of unknown function.
Table 3. Genes with potential roles in pathogenesis and virulence in B. pilosicoli 95/1000, and comparison with those in B. hyodysenteriae WA1 and B. murdochii 56-150T.doi:10.1371/journal.pone.0011455.t003
It was of interest that B. hyodysenteriae, the most pathogenic of the three species, contained more genes involved with lipopolysaccharide biosynthesis, motility and chemotaxis, and adhesion and/or surface proteins than did the other two species. The latter two did not have more of any category of these potential virulence-associated genes, apart from the bacteriophage genes in B. pilosicoli.
The outer envelope of the Brachyspira species is considered to contain lipoologosaccharide (LOS) rather than lipopolysaccharide, based on the lack of a typical Gram-negative ladder formation of repeated O-sugar components on silver-stained SDS-PAGE gels , . LOS is considered to be involved in the induction of colonic lesions associated with B. hyodysenteriae , , and hence should be considered as potentially being involved in virulence in other Brachyspira species. LOS is also antigenic, and protective immunity to B. hyodysenteriae is specific to the LOS serogroup . The importance of LOS in relation to protective immunity to B. pilosicoli is less clear, as pigs experimentally infected with B. pilosicoli 95/1000 did not develop a systemic antibody response against the spirochete .
B. pilosicoli contained a set of 25 core genes involved in the biosynthesis of LOS (Table 3). This was fewer than the 30 genes identified for B. hyodysenteriae, but more than the 10 so far identified in B. murdochii. The predicted pathways for biosynthesis of LOS and peptidoglycan for B. hyodysenteriae and B. pilosicoli are shown in Figure 9. The main difference between the species related to the presence of an rfbBADC cluster of genes on the ~36 kB plasmid in B. hyodysenteriae WA1 that encode proteins for nucleotide sugar biosynthesis (dTDP-rhamnose) that are likely to be involved in O-antigen assimilation. A similar rfbBADC cluster was not found in B. pilosicoli or in B. murdochii, implying that they were unable to produce these antigens. Similar rfb gene clusters have been linked to virulence in numerous Gram negative bacteria , , and, for example, are plasmid encoded in the case of some Salmonella enterica serovars . Further work is required to examine the potential role of the rfb cluster in the virulence of B. hyodysenteriae. It was interesting that despite the presence of genes encoding glycosyltransferases in this pathway, B. hyodysenteriae does not appear to produce ladder-like O-antigens . This discrepancy requires further investigation. The most precise way to describe the diversity of LPS/LOS in the Brachyspira species will be to make a detailed comparison of polysaccharide content and structures. It is unfortunate that no such structural data currently are available.
Figure 9. Different biosynthetic pathways for lipooligosaccharide in B. pilosicoli 95/1000 and B. hyodysenteriae WA1.
The gene name is shown in pink and corresponding predicted ORFs present in the B. hyodysenteriae genome are indicated in red, and those for B. pilosicoli in blue.doi:10.1371/journal.pone.0011455.g009
The capacity for chemotactic responses and motility are extremely important for Brachyspira species, allowing them to colonize specialized niches in the complex nutritional, physical and microbiological environment of the large intestine. Interestingly, a striking difference was found in the numbers of chemotaxis genes. B. hyodysenteriae contained 52 genes associated with chemotaxis, compared to 27 in B. pilosicoli and 39 in the partial genome of B. murdochii [Table 3]. The distribution of methyl-accepting chemotaxis genes also was different between the species, with B. pilosicoli having no mcpC genes. These differences are likely to be highly significant in relation to the chemotactic signals that the different species can respond to: they may explain the differences that have been observed in the attraction of B. pilosicoli and B. hyodysenteriae to mucin, and their tendency to occupy different local niches within the large intestine . B. hyodysenteriae colonizes the lumen of the colon, but also penetrates deep into the colonic crypts, where it enters goblet cells. Although B. pilosicoli also can enter the crypts, unlike B. hyodysenteriae it penetrates the dense mucus layer at the crypt shoulders, and attaches to the underlying colonic enterocytes. As B. hyodysenteriae has a much more limited host range than B. pilosicoli, it seems paradoxical that it has a greater number and variety of methyl-accepting chemotaxis genes.
B. pilosicoli contained 13 chemosensory transducer genes, and it was interesting that seven of these, and genes for five hypothetical proteins, were located within a cluster at a ~5.7 Kb locus. This locus comprised cheB (BP951000_0460), cheD (BP951000_0461), cheR (BP951000_0462), cheW (BP951000_0463), cheB (BP951000_0464), cheY (BP951000_0465, five copies of hypothetical proteins (BP951000_0466 to BP951000_0470), and cheX (BP951000_0471). Similarly, six of the chemosensory transducer genes of B. hyodysenteriae were within one cluster at a locus of similar size. These genes encoded chemotaxis response regulator (cheY, BHWA1_00488), chemotaxis histidine kinase (cheA, BHWA1_00489), chemotaxis protein (cheW, BHWA1_00490), chemotaxis protein methyltransferase (cheR, BHWA1_00491), chemotaxis protein (cheD, BHWA1_00492), and response regulator receiver modulated CheB methylesterase (cheB, BHWA1_00493). The other chemosensory transduction genes were located elsewhere on the two genomes, and a similar clustering of che genes was not found in B. murdochii.
Only minor difference were found in the number and types of flagellar-associated genes between the three Brachyspira strains (Table 3). B. hyodysenteriae WA1 and B. murdochii 56-150T contained four genes for FlaB core proteins while B. pilosicoli 95/1000 contained three. WA1 and 95/1000 contained three FlaA sheath protein genes while 50-150T contained two (Table 3). 50-150T also contained two copies of fleN, encoding flagellar synthesis regulator FleN, while the other two strains had single copies. As expected, the three strains had many minor differences in the predicted sequences of the flagellar proteins. When considering the predicted flagellar structure, it should be remembered that B. pilosicoli only has 4–6 periplasmic flagella at each cell end compared to 7–14 for B. hyodysenteriae , and this difference could involve the need for some modifications. Furthermore, differences in flagellin protein ratios can affect the stiffness of the periplasmic flagella in spirochetes, and this stiffness directly correlates with their motility . Hence, depending on expression of the flagellin proteins, the species are predicted to have different capacities for motility.
Lipoproteins and surface proteins.
As with other bacteria, predicted surface associated proteins and lipoproteins of the Brachyspira species are likely to be important in interactions with the host; for example, cell surface proteins are thought to be involved with the polar attachment of B. pilosicoli to colonic enterocytes . B. pilosicoli contained substantially fewer genes encoding lipoproteins than did B. hyodysenteriae (13 versus 34). Such lipoproteins, where they are surface associated, are likely to be important as potentially targets for immune recognition and could be used as subunit vaccine candidates . An example of the difference between the two fully sequenced Brachyspira species was the presence of a gene encoding a peptidoglycan-associated outer membrane lipoprotein (PAL, BHWA1_02111) in B. hyodysenteriae that was absent in B. pilosicoli. PAL is involved in maintenance of the integrity of the cell envelope , and contributes to bacterial virulence and inflammation in Gram-negative sepsis . Examples of lipoprotein-encoding genes that were shared between the species include those for basic membrane lipoprotein, outer membrane lipoprotein, apolipoprotein N acyltransferase, lipoprotein releasing system, transmembrane protein, LolC/E family, and lipoprotein releasing system ATP binding protein. The species had other predicted lipoproteins, but these were of unknown function.
B. hyodysenteriae has been reported to have a set of variable surface proteins of about 39 KDa , and strain WA1 contained nine genes encoding these proteins (VspA to VspF, and VspH), with two copies of VspA and VspD. Genes for VspA to VspF were located adjacent to each other (BHWA1_01456 to BHWA1_01601), whilst those for one copy of VspA and for VspH were located elsewhere on the genome (BHWA1_00889 and BHWA1_02382, respectively). B. pilosicoli contained single copies of the genes for VspD, VspE and VspH located in different regions, while B. murdochii 56-150T had three copies of vspG, two of vspF and one of vspG. Differential expression of these genes may be involved in immune evasion, allowing persistence of the spirochetes .
Amongst the other outer membrane proteins, B. pilosicoli 95/1000 contained three copies of bspA, predicted to encode a BspA-like surface protein. BspA-like proteins are expressed on the surface of many pathogenic bacteria: they may bind to fibronectin, stimulate a significant serological response , and induce secretion of IL-8 .
B. hyodysenteriae is strongly hemolytic, whereas B. pilosicoli and B. murdochii are weakly hemolytic . The strong hemolytic activity of B. hyodysenteriae is thought to be an important virulence trait , . In addition to the seven genes associated with hemolysis previously identified in the genome of B. hyodysenteriae WA1 , an additional putative hemolysin gene was identified (BHWA1_00962).
Only single copies of tlyA, tlyB, tlyC, and acpP (also known as hlyA ) were identified in the partial genome of B. murdochii. On the other hand, B. pilosicoli 95/1000 contained eight genes potentially involved with hemolysis, of which seven were similar to those in B. hyodysenteriae. These encoded hemolysin A (tlyA, BP951000_0123), ATP-dependent Clp protease proteolytic subunit (clpP, also known as tlyB, BP951000_1802), hemolysin C (tlyC, BP951000_1288), hemolysin related protein (hly, BP951000_0473), acyl carrier protein: contains beta-hemolysin (acpP or hylA, BP951000_0533), putative hemolysin III (BP951000_0424), and putative channel protein hemolysin III family protein (BP951000_0888). The eighth was a putative hemolysin (BP951000_2207) that was similar to a protein of Parabacteroides distasonis ATCC 8503, with ~42.9% protein identity. B. pilosicoli lacked a gene with similarity to that encoding the new putative hemolysin of B. hyodysenteriae (BHWA1_00962). From this, it is possible that the new putative hemolysin identified in B. hyodysenteriae is responsible for its strongly hemolytic phenotype. Alternatively, it has been suggested that differences between B. hyodysenteriae and the weakly hemolytic Brachyspira species such as B. pilosicoli in the FabF ACP synthase and FabG ACP reductase proteins that flank the acyl carrier protein (HlyA) may result in different lipid moieties being attached to the HlyA proteins . Acyl carrier protein-dependant fatty acylation is known to be important in activation of prohemolysin to active hemolysin in E. coli , and such differences therefore may be important for expression of the strongly hemolytic phenotype in B. hyodysenteriae.
B. hyodysenteriae WA1, B. pilosicoli 95/1000 and B. murdochii 56-150T contained similar numbers of genes predicted to encode proteases (15, 15 and 14, respectively). The proteolytic capacity in all three species was linked with the large number of ORFs encoding enzymes involved in uptake and metabolism of amino acids. These proteases are likely to be required for survival in the intestinal environment, but also may be involved in induction of local destruction of host tissues, and hence contribute to virulence. A serine protease previously described in B. pilosicoli (BP951000_0826, BP951000_1141 and BP951000_2083)  was found to have >82% sequence identity to proteases of B. murdochii 56-150T (ZP_04048074) and B. hyodysenteriae WA1 (YP_002721653).
As with B. hyodysenteriae , B. pilosicoli lacked genes for the specialized secretion systems found in pathogenic Gram-negative bacteria, although it had genes for the Sec pathway and ABC-type transporters. Six sec genes were identified: secA (BP951000_0230), secY (BP951000_1006), secF (BP951000_1532), secD (BP951000_1533), secE (BP951000_1559) and secG (BP951000_1896), and there were 36 genes encoding ABC transporters (compared to 35 and 29 for B. hyodysenteriae and B. murdochii, respectively). Ten flagella-associated genes that can form part of a type III secretary system were found in B. pilosicoli 95/1000, but needle associated genes that encode proteins required for the injection of toxins into the host cell were not identified.
Central metabolic pathways
A reconstruction of the central metabolic pathways of B. pilosicoli showed that it shared many metabolic capabilities with B. hyodysenteriae. The pathways for B. pilosicoli are shown in Figure 10, and can be compared with those for B. hyodysenteriae by examining Figure 2 in reference 15. In both species the glycolysis-gluconeogenesis metabolic axis constituted the backbone of energy production and the starting point of many biosynthetic pathways. The biosynthesis of peptidoglycan, phospholipids, aromatic amino acids, fatty acids and cofactors commenced from pyruvate, or from intermediates in the glycolytic pathway. As with B. hyodysenteriae, a complete set of genes for the non-oxidative pentose phosphate pathway, nucleotide metabolism and a respiratory electron transport chain were identified in B. pilosicoli and B. murdochii. It can be seen from the figures that some of the obvious ways in which B. pilosicoli differed from B. hyodysenteriae related to the presence of pathways associated with citrate and 2-oxo-glutarate, a pathway from acetyl-CoA to ethanol, and a lack of the rfbBADC genes located on the B. hyodysenteriae plasmid (predicted to be involved in O-antigen biosynthesis).
Figure 10. Central metabolic pathway construction for B. pilosicoli 95/1000.doi:10.1371/journal.pone.0011455.g010
Unlike B. hyodysenteriae, B. pilosicoli and B. murdochii both appeared to have the capacity to use enzymes of the incomplete TCA to synthesize glutamate. Relevant predicted enzymes in the two species were putative citrate synthase (gltA, BP951000_0367 and 4083292.C42.orf00523), putative isocitrate dehydrogenase (idh, BP951000_0368 and 4083292.C42.orf00525), aconitase (BP951000_0369 and 4083292.C42.orf00530), malate dehydrogenase (mdh, BP951000_1737 and 4083292.C15.orf00095), and two copies of fumarate hydratase in B. pilosicoli (fumA, BP951000_1272 and BP951000_1273): fumA was not indentified in B. murdochii, possibly due to the genome being incomplete. This capacity emphasized the dual function of the TCA portion of the cycle: provision of NADH as part of the complete cycle in aerobes, and provision of alpha-oxaglutarate in aerobes and anaerobes. Aconitase allows the TCA cycle to function during periods of oxidative stress, and may play a role in mediating oxidative stress and regulating cell death , . The presence of malate dehydrogenase as an alternative enzyme for converting oxaloacetate into malate is likely to provide a source of malate for cell biosynthesis in B. pilosicoli and B. murdochii.
A key branch in the partial TCA cycle of B. pilosicoli and B. murdochii involved 2-oxoglutarate, succinyl-CoA and fumarate, precursors that could proceed in either the oxidative or reductive direction. Unlike B. hyodysenteriae, B. pilosicoli and B. murdochii also had a gene set for the production of ethanol from acetyl-CoA via a branched fermentation pathway. The significance of this difference between the species is unknown, but it is likely to influence overall metabolic flexibility.
B. pilosicoli can be distinguished from all the other Brachyspira species in that it lacks beta-glucosidase activity . Consistent with this, no genes specifically encoding beta-glucosidase were identified in the B. pilosicoli genome; however, a novel system for metabolizing multiple sugars, including beta-glucosides was identified that involved alpha-galactosidase/6-phospho-beta-glucosidase(celF, BP951000_0253), PTS system, arbutin-like IIC component (glvC, BP951000_0254) and 6-phospho-alpha-glucosidase (bglA, BP951000_0255). These enzymes have specificity for 6-phospho-beta-D-glycoside substrates rather than the beta-D-glycoside substrates of beta-glucosidases , . Genes associated with beta-glucoside metabolism are likely to be important in Brachyspira species as they are known to regulate growth, adhesion and colonization in other bacterial species .
B. hyodysenteriae and B. murdochii had genes encoding phosphoenolpyruvate synthase, whereas B. pilosicoli did not appear to have the capacity to interconvert phosphoenolpyruvate (PEP) and pyruvate. Instead, it may utilize glycerate but not pyruvate for growth, and use pyruvate kinase (pyk, BP951000_0535) to catalyse the reverse conversion of PEP to pyruvic acid . This process increases markedly under microaerobic conditions, with a corresponding increase in the rate of constructive metabolism.
B. hyodysenteriae and B. murdochii possessed a complete set of genes for histidine metabolism, while B. pilosicoli did not. It is possible that B. pilosicoli has an alternative pathway for synthesizing histidine. It may however also obtain nutrients from the surrounding environment, particularly as the intestine is a rich source of amino acids, including histidine . Consistent with this requirement, three copies of a polar amino acid uptake ABC transporter, PAAT family, amino acid-binding protein (BP951000_0988, BP951000_0989 and BP951000_0990) were identified in B. pilosicoli, and not in the other two species.
A complete fatty acid biosynthesis gene set previously has been found in B. hyodysenteriae , and was identified here in B. murdochii. In contrast, an incomplete gene set was identified in B. pilosicoli, with only the genes for (3R)-hydroxymyristoyl-(acyl carrier protein) dehydratase (fabZ, BP951000_0773), malonyl CoA acyl carrier protein transacylase (fabD, BP951000_1348), and the previously mentioned locus comprising ACP synthase II (fabF, BP951000_0532), acyl carrier protein (acpP, BP951000_0533) and acyl reductase (fabG, BP951000_0534) being identified while FabZ provides the 3-hydroxymyristate moieties characteristic of lipid A. FabD is involved in elongation reactions in fatty acid synthesis. These findings suggested that B. pilosicoli does not incorporate carbon from simple carbon sources into fatty acids.
In this study the complete genome sequence of B. pilosicoli, a widespread but somewhat neglected enteric pathogen of humans and animals, was obtained and made available for analysis by the scientific community. The spirochete genome was shown to have basic similarities to the genomes of the related species B. hyodysenteriae and B. murdochii, but a number of distinct features were identified that may help to explain the different host ranges and colonization sites of the species. These included predicted differences in chemotactic responses, oxygen sensitivity, the use of some different substrates, the presence of different surface proteins, and predicted differences in LOS content and structure. The first description of genes for an apparently complete bacteriophage in B. pilosicoli also was made: this finding has important practical implications, since if the bacteriophage can be shown to be functional it could be used for genetic manipulation of B. pilosicoli. The current lack of such a genetic tool for B. pilosicoli is a constraint that limits functional studies of the genes that have been identified.
Whole-genome comparisons between B. pilosicoli 95/1000 and B. hyodysenteriae WA1 displayed using the Artemis Comparison Tool. Regions of DNA:DNA similarity are joined by lines.
(2.24 MB TIF)
The authors thank Drs Peter J. Wilson and Annette McGrath from the Australian Genome Research Facility for undertaking the B. pilosicoli sequencing and for initial bioinformatics analysis in support of the sequencing effort. We also wish to acknowledge the generosity of Dr S. Lucas and colleagues for making the partial genome sequence of B. murdochii DSM 12563 freely available in GenBank.
Conceived and designed the experiments: MIB TL DJH. Performed the experiments: PW TL. Analyzed the data: PW MIB TL KR PMM BC MB DS AH RB NP DJH. Contributed reagents/materials/analysis tools: MIB DS AH. Wrote the paper: PW DJH.
- 1. Stanton TB (2006) The genus Brachyspira. In: Falkow S, Rosenberg SE, Schleifer KH, Stackebrant E, editors. The Prokaryotes (Volume 7). New York: Springer. pp. 330–356.
- 2. Lee JI, McLaren AJ, Lymbery AJ, Hampson DJ (1993) Human intestinal spirochaete are distinct from Serpulina hyodysenteriae. J Clin Microbiol 31: 16–21.
- 3. Trott DJ, Stanton TB, Jensen NS, Duhamel GE, Johnson JL, et al. (1996) Serpulina pilosicoli sp. nov., the agent of porcine intestinal spirochetosis. Int J Syst Bacteriol 46: 206–215.
- 4. Trott DJ, Mikosza ASJ, Combs BG, Oxberry SL, Hampson DJ (1998) Population genetic analysis of Serpulina pilosicoli and its molecular epidemiology in villages in the Eastern Highlands of Papua New Guinea. Int J Syst Bacteriol 48: 659–668.
- 5. Ochiai S, Adachi Y, Mori K (1997) Unification of the genera Serpulina and Brachyspira, and proposals of Brachyspira hyodysenteriae comb. nov., Brachyspira innocens comb. nov., and Brachyspira pilosicoli comb. nov. Microbiol Immunol 41: 445–452.
- 6. Hampson DJ, Oxberry SL, La T (2006) Potential for zoonotic transmission of Brachyspira pilosicoli. Emerg Infect Dis 12: 869–870.
- 7. Lee JI, Hampson DJ (1992) Intestinal spirochaetes colonising Aborigines from communities in the remote north of Western Australia. Epidemiol Infect 109: 133–141.
- 8. Trott DJ, Combs BG, Oxberry SL, Mikosza ASJ, Robertson ID, et al. (1997) The prevalence of Serpulina pilosicoli in humans and domestic animals in the Eastern Highlands of Papua New Guinea. Epidemiol Infect 119: 369–379.
- 9. Margawani KR, Robertson ID, Brooke CJ, Hampson DJ (2004) Prevalence, risk factors and molecular epidemiology of Brachyspira pilosicoli in humans on the island of Bali, Indonesia. J Med Microbiol 53: 325–332.
- 10. Munshi MA, Traub RJ, Robertson ID, Mikosza ASJ, Hampson DJ (2004) Colonization and risk factors for Brachyspira aalborgi and Brachyspira pilosicoli in humans and dogs on tea-estates in Assam, India. Epidemiol Infect 132: 137–144.
- 11. Nelson EJ, Tanudra A, Chowdhury A, Kane AV, Qadri F, et al. (2009) High prevalence of spirochetosis in cholera patients, Bangladesh. Emerg Infect Dis 15: 571–573.
- 12. Trivett-Moore NL, Gilbert GL, Law CLH, Trott DJ, Hampson DJ (1998) Isolation of Serpulina pilosicoli from rectal biopsy specimens showing evidence of intestinal spirochetosis. J Clin Microbiol 36: 261–265.
- 13. Trott DJ, Huxtable CR, Hampson DJ (1996) Experimental infection of newly weaned pigs with human and porcine strains of Serpulina pilosicoli. Infect Immun 64: 4648–4654.
- 14. Naresh R, Song Y, Hampson DJ (2009) The intestinal spirochete Brachyspira pilosicoli attaches to cultured Caco-2 cells and induces pathological changes. PLoS One 4: e8352.
- 15. Bellgard MI, Wanchanthuek P, La T, Ryan K, Moolhuijzen P, et al. (2009) Genome sequence of the pathogenic intestinal spirochete Brachyspira hyodysenteriae reveals adaptations to its lifestyle in the porcine large intestine. PLoS One 4: e4641.
- 16. Stanton TB, Fournie-Amazouz E, Postic D, Trott DJ, Grimont PA, et al. (1997) Recognition of two new species of intestinal spirochetes: Serpulina intermedia sp. nov. and Serpulina murdochii sp. nov. Int J Syst Bacteriol 47: 1007–1012.
- 17. Hampson DJ, La T (2006) Reclassification of Serpulina intermedia and Serpulina murdochii in the genus Brachyspira as Brachyspira intermedia comb. nov. and Brachyspira murdochii comb. nov. Int J Syst Evol Microbiol 56: 1009–1012.
- 18. Weissenbock H, Maderner A, Herzog AM, Lussy H, Nowotny N (2005) Amplification and sequencing of Brachyspira spp. specific portions of nox using paraffin-embedded tissue samples from clinical colitis in Austrian pigs shows frequent solitary presence of Brachyspira murdochii. Vet Microbiol 111: 67–75.
- 19. Jensen TK, Christensen AS, Boye M (2010) Brachyspira murdochii colitis in pigs. Vet Pathol 47: 334–338.
- 20. Kunkle RA, Harris D, Kinyon JM (1986) Autoclaved liquid medium for propagation of Treponema hyodysenteriae. J Clin Microbiol 24: 669–671.
- 21. Wilson K (1990) Preparation of genomic DNA from bacteria. In: Ausubel FM, Brent R, Kingston RE, Moore DD, Smith JA, Seidman JG, Struhl K, editors. Current Protocols in Molecular Biology. New York: Wiley. pp. 241–242.
- 22. Sonnhammer EL, Durbin R (1995) A dot-matrix program with dynamic threshold control suited for genomic DNA and protein sequence analysis. Gene 167: GC1–10.
- 23. Zuerner RL, Stanton TB (1994) Physical and genetic map of the Serpulina hyodysenteriae B78T chromosome. J Bacteriol 176: 1087–1092.
- 24. Fukunaga M, Masuzawa T, Okuzako N, Mifuchi I, Yanagihara Y (1990) Linkage of ribosomal RNA genes in Leptospira. Microbiol Immunol 34: 565–573.
- 25. Fukunaga M, Okuzako N, Mifuchi I, Arimitsu Y, Seki M (1992) Organization of the ribosomal RNA genes in Treponema phagedenis and Treponema pallidum. Microbiol Immunol 36: 161–167.
- 26. Schwartz JJ, Gazumyan A, Schwartz I (1992) rRNA gene organization in the Lyme disease spirochete, Borrelia burgdorferi. J Bacteriol 174: 3757–3765.
- 27. Zuerner RL, Herrmann JL, Saint Girons I (1993) Comparison of genetic maps for two Leptospira interrogans serovars provides evidence for two chromosomes and intraspecies heterogeneity. J Bacteriol 175: 5445–5451.
- 28. Paster BJ, Dewhirst FE (2000) Phylogenetic foundation of spirochetes. J Mol Microbiol Biotech 2: 341–344.
- 29. Picardeau M, Lobry JR, Hinnebusch BJ (1999) Physical mapping of an origin of bidirectional replication at the centre of the Borrelia burgdorferi linear chromosome. Mol Microbiol 32: 437–445.
- 30. Motro Y, Dunn DS, La T, Phillips ND, Hampson DJ, et al. (2008) Intestinal spirochaetes of the genus Brachyspira share a partially conserved 26 kilobase genomic region with Enterococcus faecalis and Escherichia coli. Microbiol Insights 2008: 1, 1–9.
- 31. Rutherford K, Parkhill J, Crook J, Horsnell T, Rice P, et al. (2000) Artemis: sequence visualization and annotation. Bioinformatics 16: 944–945.
- 32. Zuerner RL, Stanton TB, Minion FC, Li C, Charon NW, et al. (2004) Genetic variation in Brachyspira: Chromosomal rearrangements and sequence drift distinguish B. pilosicoli from B. hyodysenteriae. Anaerobe 10: 229–237.
- 33. Tamames J (2001) Evolution of gene order conservation in prokaryotes. Genome Biol 2: RESEARCH0020.
- 34. Eckardt NA (2001) Everything in its place: Conservation of gene order among distantly related plant species. Plant Cell 13: 723–725.
- 35. Wang Y, Diehl A, Wu F, Vrebalov J, Giovannoni J, et al. (2008) Sequencing and comparative analysis of a conserved syntenic segment in the Solanaceae. Genetics 180: 391–408.
- 36. Little JW (2005) Lysogeny, prophage induction, and lysogenic conversion. In: Waldor MK, Friedman DI, Adhya SL, editors. Phages: Their role in bacterial pathogenesis and biotechnology. Washington: ASM Press. pp. 37–53. Chapter 3 In:.
- 37. Motro Y, La T, Bellgard MI, Dunn DS, Phillips ND, et al. (2009) Identification of genes associated with prophage-like gene transfer agents in the pathogenic intestinal spirochaetes Brachyspira hyodysenteriae, Brachyspira pilosicoli and Brachyspira intermedia. Vet Microbiol 134: 340–345.
- 38. Humphrey SB, Stanton TB, Jensen NS, Zuerner RL (1997) Purification and characterization of VSH-1, a generalized transducing bacteriophage of Serpulina hyodysenteriae. J Bacteriol 179: 323–329.
- 39. Matson EG, Zuerner RL, Stanton TB (2007) Induction and transcription of VSH-1, a prophage-like gene transfer agent of Brachyspira hyodysenteriae. Anaerobe 13: 89–97.
- 40. Graentzdoerffer A, Pich A, Andreesen JR (2001) Molecular analysis of the grd operon coding for genes of the glycine reductase and of the thioredoxin system from Clostridium sticklandii. Arch Microbiol 175: 8–18.
- 41. Andreesen JR (1994) Glycine metabolism in anaerobes. Anton Van Leeuw 66: 223–237.
- 42. Oxberry SL, Trott DJ, Hampson DJ (1998) Serpulina pilosicoli, water birds and water: potential sources of infection for humans and other animals. Epidemiol Infect 121: 219–225.
- 43. Boye M, Baloda SB, Leser TD, Møller K (2001) Survival of Brachyspira hyodysenteriae and B. pilosicoli in terrestrial microcosms. Vet Microbiol 81: 33–40.
- 44. Trott DJ, Jensen NS, Saint Girons I, Oxberry SL, Stanton TB, et al. (1997) Identification and characterization of Serpulina pilosicoli isolates recovered from the blood of critically ill patients. J Clin Microbiol 35: 482–485.
- 45. Hampson DJ, Robertson ID, Oxberry SL (1999) Isolation of Serpulina murdochii from the joint fluid of a lame pig. Aust Vet J 77: 48.
- 46. Piotrowski J, Slomiany A, Murty VL, Fekete Z, Slomiany BL (1991) Inhibition of Helicobacter pylori colonization by sulfated gastric mucin. Biochem Int 24: 749–756.
- 47. Wang G, Rasko DA, Sherburne R, Taylor DE (1999) Molecular genetic basis for the variable expression of Lewis Y antigen in Helicobacter pylori: analysis of the alpha (1,2) fucosyltransferase gene. Mol Microbiol 31: 1265–1274.
- 48. Wang G, Boulton PG, Chan NW, Palcic MM, Taylor DE (1999) Novel Helicobacter pylori alpha 1,2-fucosyltransferase, a key enzyme in the synthesis of Lewis antigens. Microbiology 145: 3245–3253.
- 49. Lichtensteiger CA, Vimr ER (1997) Neuraminidase (sialidase) activity of Haemophilus parasuis. FEMS Microbiol Lett 152: 269–274.
- 50. Soong G, Muir A, Gomez MI, Waks J, Reddy B, et al. (2006) Bacterial neuraminidase facilitates mucosal infection by participating in biofilm production. J Clin Invest 116: 2297–2305.
- 51. Vimr ER, Kalivoda KA, Deszo El, Steenbergen SM (2004) Diversity of microbial sialic acid metabolism. Microbiol Mol Biol Rev 68: 132–153.
- 52. Somerville GA, Chaussee MS, Morgan CI, Fitzgerald JR, Dorward DW, et al. (2002) Staphylococcus aureus aconitase inactivation unexpectedly inhibits post-exponential-phase growth and enhances stationary-phase survival. Infect Immun 70: 6373–6382.
- 53. Somerville G, Mikoryak CA, Reitzer L (1999) Physiological characterization of Pseudomonas aeruginosa during exotoxin A synthesis: glutamate, iron limitation, and aconitase activity. J Bacteriol 181: 1072–1078.
- 54. Halter M, Joens LA (1988) Lipooligosaccharides from Treponema hyodysenteriae and Treponema innocens. Infect Immun 5: 3152–3156.
- 55. Lee BJ, Hampson DJ (1999) Lipo-oligosaccharide profiles of Serpulina pilosicoli strains and their serological cross-reactivities. J Med Microbiol 48: 411–415.
- 56. Nuessen ME, Joens LA, Glock RD (1983) Involvement of lipopolysaccharide in the pathogenicity of Treponema hyodysenteriae. J Immunol 131: 997–999.
- 57. Greer JM, Wannemuehler MJ (1989) Comparison of the biological responses induced by lipopolysaccharide and endotoxin of Treponema hyodysenteriae and Treponema innocens. Infect Immun 57: 717–723.
- 58. Joens LA, DeYoung DW, Glock RD, Mapother ME, Cramer JD, et al. (1985) Passive protection of segmented swine colonic loops against swine dysentery. Am J Vet Res 46: 2369–2371.
- 59. Hampson DJ, Robertson ID, La T, Oxberry SL, Pethick DW (2000) Influences of diet and vaccination on colonisation of pigs with the intestinal spirochaete Brachyspira (Serpulina) pilosicoli. Vet Microbiol 73: 75–84.
- 60. Whitfield C (1995) Biosynthesis of lipopolysaccharide O antigens. Trends Microbiol 3: 178–85.
- 61. Wildschutte H, Wolfe DM, Tamewitz A, Lawrence JG (2004) Protozoan predation, diversifying selection, and the evolution of antigenic diversity in Salmonella. Proc Natl Acad Sci USA 101: 10644–9.
- 62. Keenleyside WJ, Perry M, Maclean L, Poppe C, Whitfield C (1994) A plasmid encoded rfb O:54 gene cluster is required for biosynthesis of the O:54 antigen in Salmonella enterica serovar Borreze. Mol Microbiol 11: 437–448.
- 63. Naresh R, Hampson DJ (2010) Attraction of Brachyspira pilosicoli to mucin. Microbiology 156: 191–197.
- 64. Li C, Sal M, Marko M, Charon NW (2010) Differential regulation of the multiple flagellins in spirochetes. J Bacteriol 192: 2596–2603.
- 65. Trott DJ, Alt DP, Zuerner RL, Wannemuehler MJ, Stanton TB (2001) The search for Brachyspira outer membrane proteins that interact with the host. Anim Health Res Rev 2: 19–30.
- 66. Movahedi A, Hampson DJ (2008) New ways to identify novel bacterial antigens for vaccine development. Vet Microbiol 131: 1–13.
- 67. Rodríguez-Herva JJ, Ramos-Gonzalez MI, Ramos JL (1996) The Pseudomonas putida peptidoglycan-associated outer membrane lipoprotein is involved in maintenance of the integrity of the cell envelope. J Bacteriol 178: 1699–706.
- 68. Hellman J, Roberts JD Jr, Tehan MM, Allaire JE, Warren HS (2002) Bacterial peptidoglycan-associated lipoprotein is released into the bloodstream in gram-negative sepsis and causes inflammation and death in mice. J Biol Chem 277: 14274–1480.
- 69. McCaman MT, Auer K, Foley W, Gabe JD (2003) Brachyspira hyodysenteriae contains eight linked gene copies related to an expressed 39-kDa surface protein. Microbes Infect 5: 1–6.
- 70. Sharma A, Sojar HT, Glurich I, Honma K, Kuramitsu HK, et al. (1998) Cloning, expression, and sequencing of a cell surface antigen containing a leucine-rich repeat motif from Bacteroides forsythus ATCC 43037. Infect Immun 66: 5703–5710.
- 71. Onishi S, Honma K, Liang S, Stathopoulou P, Kinane D, et al. (2008) Toll-like receptor 2-mediated interleukin-8 expression in gingival epithelial cells by the Tannerella forsythia leucine-rich repeat protein BspA. Infect Immun 76: 198–205.
- 72. ter Huurne AA, Gaastra W (1995) Swine dysentery: More unknown than known. Vet Microbiol 46: 347–360.
- 73. Hsu T, Hutto D, Minion FC, Zuerner RL, Wannemuehler MJ (2001) Cloning of a beta-hemolysin gene of Brachyspira (Serpulina) hyodysenteriae and its expression in Escherichia coli. Infect Immun 69: 706–711.
- 74. Issartel JP, Koronakis V, Hughes C (1991) Activation of Escherichia coli prohaemolysin to the mature toxin by acyl carrier protein-dependent fatty acylation. Nature 351: 759–761.
- 75. Muniappa N, Duhamel GE (1997) Outer membrane-associated serine protease of intestinal spirochetes. FEMS Microbiol Letts 154: 159–164.
- 76. Moeder W, Del Pozo O, Navarre DA, Martin GB, Klessig DF (2007) Aconitase plays a role in regulating resistance to oxidative stress and cell death in Arabidopsis and Nicotiana benthamiana. Plant Mol Biol 63: 273–287.
- 77. Fellström C, Gunnarsson A (1995) Phenotypical characterization of intestinal spirochaetes isolated from pigs. Res Vet Sci 59: 1–4.
- 78. Anderson RL (1972) Cellobiose metabolism in Aerobacter aerogenes. 3. Cleavage of cellobiose monophosphate by a phospho-beta-glucosidase. J Biol Chem 247: 3420–3423.
- 79. Dale M, Kopfler W, Chait I, Byers L (1986) Beta-glucosidase: substrate, solvent, and viscosity variation as probes of the rate-limiting steps. Biochemistry 25: 2522–2529.
- 80. Kilic AO, Tao L, Zhang Y, Lei Y, Khammanivong A, et al. (2004) Involvement of Streptococcus gordonii beta-glucoside metabolism systems in adhesion, biofilm formation, and in vivo gene expression. J Bacteriol 186: 4246–4253.
- 81. Podkopaeva DA, Grabovich M, Dubinina GA (2003) Oxidative stress and antioxidant cell protection systems in the microaerophilic bacterium Spirillum winogradskii. Mikrobiologiia 72: 600–608.
- 82. Juliao PC, Marrs CF, Xie J, Gilsdorf JR (2007) Histidine auxotrophy in commensal and disease-causing nontypeable Haemophilus influenzae. J Bacteriol 189: 4994–5001.