Intestinal Spirochaetes and Brachyspiral colitis

The “intestinal spirochaetes” are a group of anaerobic bacteria assigned to various species in the genus Brachyspira. They inhabit the large intestines of birds and animals – but also may be found in human beings. These bacteria first came to prominence in the early 1970s when a spirochaete named Treponema hyodysenteriae (now Brachyspira hyodysenteriae) was shown to be the agent of swine dysentery, a colonic infection of pigs that is endemic in many countries and is of considerable economic significance. Since the initial description, related spirochaetes have been identified and characterised and various name changes have occurred – finally resulting in the genus Brachyspira and its seven officially recognised species. Many different hosts are colonised with the various Brachyspira species, but disease is mainly reported in pigs and in adult chickens. Humans are colonised with the zoonotic Brachyspira pilosicoli and Brachyspira aalborgi. Reduced susceptibility to various antimicrobials is now starting to represent a major problem for effective control of Brachyspiral colitis in pigs and other species, and consequently attention is focusing on the development of new vaccines. The Brachyspira species have specialised growth requirements, and different species can take from three days to three weeks to form a thin film of visible growth on selective isolation plates. Genetic manipulation of individual strains remains difficult, and this has limited understanding of gene function and disease pathogenesis. Recently whole genomic sequencing projects have started to reveal much that was previously unknown about these specialised bacteria.

Currently the genus Brachyspira includes seven officially named species and a variety of unofficially proposed members. Some species have well-established pathogenic potential in certain hosts, causing forms of Brachyspira colitis, whilst others are considered to be largely commensal (see Table 1). The Brachyspira are genetically distinct from other spirochaetes, and the close similarities between some of the species in their 16S rRNA sequences suggest that speciation in the genus has occurred relatively recently and rapidly. Apart from the strength of haemolysis, there are few clear phenotypic differences between the species, and indeed the boundaries between some of the named species are indistinct both genetically and phenotypically.

Table 1. Brachyspira species, their hosts and disease associations.

For some species, such as B. hyodysenteriae, studies using multilocus sequence typing have provided clear evidence of the population structure being clonal^1,2, while for B. pilosicoli the population appears to be recombinant³. In the case of B. hyodysenteriae, the adaptation to a lifestyle in the large intestine appears to have included acquisition of various genes from Escherichia coli and Clostridium species, especially those encoding proteins associated with transport and metabolism⁴. These are likely to have been acquired in the densely populated, complex and specialised environment of the large intestine.

The existence of extensive genetic rearrangements can be observed within and between Brachyspira species, with sequence drift also generating diversity. The variation and fluidity of the genomes can be seen in the case of B. pilosicoli, where three sequenced strains had genome sizes of ~2,765, 2.890 and 2.596 Mb respectively⁵, with genome rearrangements that largely correlated with the positions of mobile genetic elements. Novel bacteriophages were detected, as they were in a previous genomic study on B. intermedia⁶. These bacteriophages, that have themselves undergone extensive gene remodelling, are involved in intra- and inter-species horizontal gene transfer, and are likely to be a major force in the evolution of the Brachyspira species. In addition, novel genetic information may be acquired through the activity of prophage-like gene transfer agents that are present in the genome of different Brachyspira species^7,8. Evidence for rapid genetic change can be seen at the farm level, where, for example, “microevolution” of B. hyodysenteriae strains resulting in changed DNA profiles has been recorded over relatively short time periods^9,10.

The basis of virulence in the various Brachyspira species is still imperfectly understood. In order for pathogenic Brachyspira species to induce disease it is essential for them to colonise the large intestine and to grow to large numbers. Their anaerobic metabolism and use of substrates has been fine tuned to allow them to thrive in the milieu of the large intestine. There are complex physical and chemical interactions that occur between components of the diet and the normal colonic microbiota: these profoundly influence the environment, and it has become clear that the resultant conditions can affect colonisation by the spirochaetes¹¹.

As part of the colonisation process Brachyspira cells must move through the mucus overlying the epithelium of the large intestine. The corkscrew-like motility of B. hyodysenteriae has long been known to be an important virulence attribute, allowing it to penetrate the mucus. In the case of B. pilosicoli, this spirochaete shows increased motility under viscous conditions¹², including mucin concentrations equivalent to those found in the colon¹³. In addition to their motility, the cells of different Brachyspira species demonstrate a chemotactic attraction to colonic mucin. Comparison of the genome sequences of B. hyodysenteriae and B. pilosicoli has shown that B. pilosicoli has fewer methyl-accepting chemotaxis genes than B. hyodysenteriae, and completely lacks mcpC genes; hence these species are predicted to have different chemotactic responses, and this in turn may help to explain their different host ranges and colonisation sites in the large intestine¹⁴. Experimentally, although cells of both B. hyodysenteriae and B. pilosicoli are attracted to and enter mucin solutions, this was reduced at mucin concentrations above 6% for B. hyodysenteriae but not for B. pilosicoli¹³, again providing a possible explanation for their different colonisation sites.

A likely virulence determinant in B. hyodysenteriae is the strong haemolytic activity of this spirochaete. This is supported by the fact that two other recently described strongly haemolytic species are also pathogenic^15,16 (see Table 1). Currently eight genes encoding proteins with predicted haemolytic activity have been described in B. hyodysenteriae^4,14, but their respective roles have not all been confirmed experimentally.

A recent in vitro study using Caco-2 cell monolayers has provided some insights into how B. pilosicoli interacts with colonic enterocytes to cause disease¹⁷, and similar detailed studies are required with B. hyodysenteriae and other species. In that study¹⁷ the Caco-2 cell junctions were shown to be the initial targets of the characteristic end-on attachment by B. pilosicoli. Colonised monolayers then demonstrated a time-dependent series of changes, including accumulation of actin at the cell junctions, loss of tight junction integrity and condensation and fragmentation of nuclear material consistent with the occurrence of apoptosis. The colonised monolayers demonstrated a significant up-regulation of interleukin-1ß (IL-1ß) and IL-8 expression. These cytokines/chemokines are likely to be responsible for attracting inflammatory cells to the colonisation site, and causing localised colitis. Potential mechanisms for inducing such cellular damage include the biological activity of lipooligosaccharides and/or the action of membrane proteases.

Sequencing of the genome of B. hyodysenteriae strain WA1 resulted in the identification of a previously unrecognised plasmid that contained 31 genes, including six rfbA-D genes that were predicted to be involved with rhamnose biosynthesis, and hence LOS structure, as well as others associated with glycosylation⁴. Subsequently avirulent strain A1 was shown to lack the plasmid, and when an Australian field isolate lacking the plasmid was selected and used experimentally to infect pigs, significantly fewer became colonised and developed dysentery compared to the pigs infected with a control strain that contained the plasmid¹⁸. The results support the likelihood that plasmid-encoded genes of B. hyodysenteriae are involved in colonisation and/or in disease expression.

Recently recombinant protein vaccines have received attention as potential vaccine components for Brachyspira species: for example, vaccination with recombinant outer-membrane lipoprotein Bhlp29.7 from B. hyodysenteriae provided a 50% reduction in the incidence of disease compared to unvaccinated controls following experimental infection¹⁹. The availability of genome sequences has provided the opportunity to broaden this approach through the application of “reverse vaccinology”, where scores of such predicted proteins can be identified from the genome sequence, screened and tested as vaccine candidates. This approach has been used successfully with B. hyodysenteriae²⁰, and it is anticipated that a new generation of commercial vaccines based on this approach will become available in the next few years.

With the recent availability of Brachyspira genome sequences and new technologies better insights into the growth requirements and pathogenic mechanisms of Brachyspira species are emerging. This information is of direct benefit for control of the infections, since, for example, information about growth and colonisation requirements derived from metabolic reconstructions of the spirochaetes can help to predict what changes in the colonic environment are likely to reduce their growth^4,14. Further detailed studies are needed to determine how the colonic microbiota is influenced by different dietary substrates, and how this impacts on colonisation by Brachyspira species. The sequence data has also allowed the use of a reverse vaccinology approach to vaccine development.