Genome evolution in host-adapted bacteria
The availability of a large number of microbial genomes from a broad range of organisms has shaped our current understanding of the dynamics of genome structure from pathogenic and non-pathogenic bacteria. Such, it has been suggested that close adaptation towards a host in a symbiotic or pathogenic relationship results in small, minimalist genomes, which are required for survival in a host. Our group studies the genomes from related host-adapted and potentially free-living bacteria in order to gain insight into the molecular mechanisms that has driven the speciation process from free-living last common ancestors to the obligatory pathogenic species that we see today.
Epsilon-Proteobacteria show a highly variable genome structure
Our current understanding of bacterial pathogenicity mechanisms is largely based on the knowledge of the genomic inventory that is being shared by the most serious bacterial agents. Since many of these pathogenic organisms are strictly host-adapted, their genomes have undergone a degrading process leading to small, minimalist genomes. This process of deleting genetic information has resulted in orphaned cellular processes that can only be understood in their ancestral context. Such, to extend our knowledge on the origin and the emergence of a pathogen, it is essential to analyze the genomic inventory of living relatives with non-degraded genomes that have largely maintained the gene pool of a last common ancestor. Wolinella succinogenes, a rather unknown rumen dweller, is such an organism. Phylogenetically interspersed between Helicobacter pylori and Campylobacter jejuni, it constitutes an out-group to both pathogens, which have been demonstrated to cause serious illnesses in humans, such as gastric cancer or the Guillain-Barré -Syndrome. In our analysis, we show that the genome of W. succinogenes is 30% larger than those of H. pylori and C. jejuni. Since pseudo-genes were found to be rare in W. succinogenes, we believe its genome resides in a non-degrading state. A wealth of genetic information is found in addition to its relatives, which is highly reminiscent of free-living bacteria. This is particularly evident for genes that share a high degree of homology with cyanobacterial genes, such as a cluster of nif genes, as well as extended signaling networks. Surprisingly, the non-pathogenic W. succinogenes also contains complete sets of genes, which are homologous to known virulence factors and are clustered on genomic islands on the chromosome.
By a differential approach we now can identify genes that are unique in each of the three organisms and may confer host specificity to each of the organisms. Many of these species-specific genes co-localize with virulence genes on the chromosome and are therefore potential candidates for functional analysis.
By studying those genes that are being shared by all three organisms, we can identify essential molecular mechanisms used by symbiotic, commensal or pathogenic bacteria to maintain themselves in a vertebrate host environment.
The predatory bacterium Bdellovibrio keeps its genes to itself.
In a further approach we are testing whether results from the epsilon-proteobacterial system can be applied to other bacteria, which are not host-adapted to mammals, but rather more prey on other bacteria. The organism in question is called Bdellovibrio, which would translate into a "curved leech", a delta-Proteobacterium. In its lifecycle it depends on other Gram-negative bacteria for nutriment and building material. Bdellovibrio therefore invades its bacterial prey and devours it from inside, while initiating its own growth. Proteins, lipids and nucleotides of the host thereby serve as a substratum for the predator’s own growth and eventually the generation of several progeny.
Despite its life cycle being a parasitic one, Bdellovibrio has a twice as large genome compared to the one from pathogenic epsilon-Proteobacteria. A further striking difference is found in the almost complete absence of horizontal gene transfer (HGT) between the predator and the prey, which could mechanistically easily occur since the predator had direct access to the prey’s genetic information. Comparison of strains of Bdellovibrio, which are exclusively predatory or axenically growing, shows an unexpected stability of these genomes (e.g. lack of recombination).