The definitive Bradyrhizobium genome
This week, a new Bradyrhizobium genome sequence was published (1). It is not the first, and certainly will not be the last. It is not from the most-studied strain, and is not the most exhaustively annotated genome. Nevertheless, it is important because it is the definitive Bradyrhizobium genome. It is the genome of Bradyrhizobium japonicum strain USDA 6. This is the type strain of the species, and B. japonicum is the type species of the genus Bradyrhizobium, so the genus is defined in relation to this strain.
The authors make extensive comparisons with the genome sequence of strain USDA 110, the first strain of this species to be sequenced. The two genomes are indeed very similar, but the differences are instructive, shedding some light on the evolution of the symbiosis island, for example.
Genomes and taxonomy
Every bacterial species has a type strain, but often this is not the best-studied strain in the species, and often it is not the first to be sequenced. As sequencing costs drop, though, it becomes feasible to sequence the type strain of every described bacterial species. I’ve heard that the BGI in China has this ambition. This would, of course, allow us to define the relationships among bacterial species with far more precision than we can now. It would highlight anomalies in the exchange rate between DNA divergence and taxonomic rank in different bacterial groups: how often are bacteria placed in different genera actually more similar than bacteria within a single species elsewhere in the bacteria? It would also be a treasure chest for the study of genome evolution in bacteria, allowing us to explore the extent to which different classes of genes are prone to horizontal transfer, to selection, to loss or rearrangement, and so on.
I can imagine that, in ten years time, it might become obligatory to deposit the entire genome sequence of the type strain when describing a new bacterial species. It would actually be feasible to introduce such a rule now, and will certainly be economically acessible for most microbiologists within a couple of years. I am saying ‘ten years’ though, because taxonomists are notoriously slow and conservative beasts.
The type and the swarm
Although a bacterial species is defined in relation to a type strain, the type strain is not sufficient for a robust species definition. A species is a swarm of related individuals, and it is important to define the extent of that swarm and its boundaries with neighbouring species. The ‘space’ we are considering here is not physical, but genomic, or perhaps phenotypic. Strains which are sufficiently close to the type strain are considered to belong to the same species. Unfortunately, bacteriologists have got into the habit of using a ‘one size fits all’ strategy for defining ‘sufficiently close’ (most notoriously through the use of DNA-DNA hybridisation) rather than relating it to the size of the swarm. In fact, many publications that describe new species make little effort to explore the breadth of the species. They are often based on just a handful of isolates, sometimes only one. Of course, the ideal type strain should be somewhere near the centre of the swarm in order to be a typical representative, but the rush to publish without a thorough sampling of the diversity means that the type strain can easily turn out to be peripheral to the species.
Ideally, then, we should have not just one complete genome before publishing a new species, but a whole swarm of sequences. Fortunately, the pace of technological development is on our side, and it may not be too many years before this becomes a reality.
(1) Kaneko T., Maita H., Hirakawa H., Uchiike N., Minamisawa K., Watanabe A., Sato S. Complete Genome Sequence of the Soybean Symbiont Bradyrhizobium japonicum Strain USDA6T. Genes. 2011; 2(4):763-787.