From Rhizobium to Sinorhizobium, from Sinorhizobium to Ensifer?
There are many publications on the bacteria that most people know as Sinorhizobium meliloti, S. fredii, and related species. The name of their genus has changed over the years, and I used Web of Science to track the changing use of the different names in the titles of published articles.
Some key dates are:
1926: the species Rhizobium meliloti proposed.
1982: the genus Ensifer proposed for some nonsymbiotic bacteria.
1984: the species Rhizobium fredii proposed.
1988: the genus Sinorhizobium proposed for R. fredii
1994: the genus Sinorhizobium relaunched to include R. meliloti
2003: the proposed amalgamation of Sinorhizobium with Ensifer (under the rules of precedence, the combined genus should be Ensifer)
Here are the figures for the number of publications each year that used each of these names. Some papers gave two alternatives, one in parentheses. I leave the interpretation of these data as an exercise for the reader. I will probably discuss them in later posts, however.
rhizobium 
Or in other words, most of the rhizobiology community is just saying “Forget about this Ensifer stuff” to the taxonomist
I would say it is somewhat similar to the nomenclature problem of rhizobia being originally used as shorthand for the Rhizobiaceae family, so we shouldn´t use it for the betabacteria, and in case of doubt should use something like Bacteria that Nodulate Legumes and Fix Nitrogen instead.
I understand that bacterial taxonomy should be respected, but if we can´t even be sure of biologically valid species definition for bacteria, does it make any sense to be so demanding of taxonomical protocols for them?
Sinorhizobium has been used for a long while, and is the form that will be found on the vast majority of the literature, so IMHO should be accepted form at least for the symbiotic species of the genus. How to do that in a taxonomically acceptable manner is way above my paygrade…
You are quite right, Mario, that the community has not adopted Ensifer – at least, not so far, and the proposal has been around for ten years already. It is interesting that you are so positive about Sinorhizobium, showing that this name really has been adopted and become mainstream. From the graph, you can see that this only happened about fifteen years ago, and the majority of the literature is actually earlier than that, when the species were called Rhizobium. The reasons why Sinorhizobium was adopted readily, but Ensifer was not, are complex and interesting, but will have to wait until future posts.
I note that you take the opposite view from me on the way to use the term “rhizobia”. I use it to mean what you call “Bacteria that Nodulate Legumes and Fix Nitrogen”, which is rather a mouthful. You use it of bacteria in the family Rhizobiaceae (which would include Agrobacterium, Shinella, etc.), though the name does not indicate to me that it should apply to a whole family, rather than just the genus Rhizobium. Back in the days when all root-nodule bacteria were Rhizobium, and vice versa, the term “rhizobia” was unambiguous, but now that we know of many genera that make root nodules, and many strains of Rhizobium that do not, we have to make a choice – do we use “rhizobia” to describe a phenotype or a taxon? I decided that it was preferable to go with phenotype, and many others have also taken that approach, but there are others, like you, who prefer to use the word as an informal term for bacteria in a particular taxon. I guess that we just have to be careful to make it clear which we are talking about.
Peter
Peter,
It seems that my English failed me. I was really trying to “say” that rhizobia, on the old style, is much better than using that awful complete description, even though now is known to be taxonomically inaccurate due to it etymology. In other words, I agree that it is better to use rhizobia to describe the “Bacteria that Nodulate Legumes and Fix Nitrogen” phenotype, than as “bacteria from the Rhizobiaceae family that nodulate legumes and fix nitrogen”, which would be much preferrable, considering the origin of the “nickname”.
As for Sinorhizobium, if you excluded those few citations on the first half of the 90s, which probably were on the bacterial taxonomy journals, it was adopted pretty fast, on about five to six years, even though it included the venerable R. meliloti (one of the last original Rhizobium species). On no other basis than my gut feeling, I would say that having a genus name that includes the rhizobium suffix as in Brady, Azo, or Mesorhizobium accelerated and widened the acceptance of the Sinorhizobium genus.
Again, on no basis other than gut feeling, I would predict that Agrobacterium will suffer quite a lot before being accepted as the correct name for Rhizobium, and most likely won´t be accepted in the general scientific literature on rhizobiology for a long while, if at all…
Best regards from Brazil,
Mario
Mario
There is no danger that taxonomists will ever suggest that Agrobacterium is the correct name for Rhizobium – their own rules forbid it. Rhizobium is one of the oldest bacterial names, dating from 1889, while Agrobacterium was only proposed in 1942. This means that, if the two genera are merged, Rhizobium ‘wins’ and Agrobacterium species have to be renamed Rhizobium. John Young (no relation!) proposed this in 2001, provoking vehement opposition from the Agrobacterium community. The proposal has caused a certain amount of confusion, as a few people (mostly other taxonomists) have adopted the new names, but in general the community has ignored it. In the past three years, more papers have been published with Agrobacterium in the title than with Rhizobium – and that is including all the real Rhizobium.
Peter, I sure am glad to know that. If Rhizobium is the earlier name, why was this name change even avented in the literature? I read somewhere (and have really no clue where) that they were proposing Agrobacterium to be the real genus exactly because it was older than Rhizobium. Mind, if we can get them to agree, I would rather keep the non-symbiotic/pathogenic species quietly in Agrobacterium with the symbiotic in Rhizobium.
The main reason for that is that in Brazil, rhizobia (lato sensu) strains can only be used for inoculant production if sanctioned by the Federal Ministry of Agriculture and Livestock Production. If the genus includes any pathogenic, or phytopathogenic, species (such as Burkholderia) the testing requirements are way, way more stringent, and would be impossible to meet for a lot of the research teams working on the subject, mine included.
As we “know” that old-style Rhizobium is NOT-pathogenic, so there is no real risk factor in inoculating with one of them, these new requirements would not add any real safety to the indications, but could limit our inoculant selection programs needlessly.
I am not sure you know, but there is a strong current in Brazil advocating ever larger dependency on biological nitrogen fixation in agriculture (from both the economic and ecological viewpoints), and anything that limits our finding new strains, with higher BNF potentials and competitiveness would be a major handicap.
Of course, our main legume crop (soybean) wouldn´t have any limitations due to that, since as far as I know, Bradyrhizobium would be quite safe. But several other crops, such as Phaseolus vulgaris, form symbiosis with Rhizobium, and the Brazilian BNF research community has been trying to break the logjam in inoculant use for these crops for several years. It would be a pity to have a new requirement to make the job even harder (as if our usually large native, competitive rhizobial community in agricultural soils wasn´t enough of a problem for inoculant recommendation by itself).
Thanks, then, for the good news.
Mario
I would be interested to see your take on the proposal for ‘Agrobacterium fabrum’ from a couple of years ago. (PMID 21795751) Perhaps in a future post?
Hi Matt
Perhaps it reflects my taste for mischief, but I actively encouraged Xavier Nesme to publish Agrobacterium fabrum, knowing full well that it would annoy the taxonomic hegemony, who would probably not allow it to be validly published.
A future post? Maybe, if I can find a sufficiently inoffensive way to express my views!
Peter
Well, I don’t know whether it reflects oversight or annoyance, but J.P. Euzéby still hasn’t included Agrobacterium fabrum in his list of post-1980 unaccepted nomenclature (http://www.bacterio.cict.fr/nonvalid.html) much less in the validly published names.
I really enjoyed this post. It’s funny what a name means to people. I started with R. meliloti. It is nice to see numbers to see how the name Sinorhizobium has been adopted. Pretty much what I would have thought. It also shows Ensifer is not catching on. I realize it is correct, but I think I will only use it when i am told to use it. Some weird attachment to word rhizobium I guess.
Hi Peter, great blog you have here, I wish I was even half as prolific at posting as you are.
Regarding Agrobacterium fabrum, there is no great conspiracy here, the name was published without explicitly mentioning that it was a new species (sp. nov.) and without a description. I do find the requirements for publishing a new bacterial species frustrating compared to publishing a new fungus, but the trade off is that we have a really great single list of current names (the LPSN).
About the wider Agrobacterium / Rhizobium issue, I finally got my thoughts together into a blog post about this: http://www.rhizobia.co.nz/taxonomy/agrobacterium-or-rhizobium spoiler: I think we should stick with Rhizobium until we have a lot more data.