|Ilut, Dan -|
|Farmer, Andrew -|
|Maki, Sonja -|
|May, Gregory -|
|Singer, Susan -|
|Doyle, Jeff -|
Submitted to: PLoS One
Publication Type: Peer Reviewed Journal
Publication Acceptance Date: June 18, 2010
Publication Date: July 16, 2010
Repository URL: http://plosone.org/article/info%3Adoi%2F10.1371%2Fjournal.pone.0011630
Citation: Cannon, S.B., Ilut, D., Farmer, A.D., Maki, S.L., May, G.D., Singer, S.R., Doyle, J.J. 2010. Nodule evolution did not depend on early polyploidy in the legumes. PLoS One. 5(7):e11630. Interpretive Summary: The legume plant family, which contains soybeans, common beans, peas, lentils, clovers and many other crop species, derives much of its value from the ability of these plants to create (or "fix") their own nitrogen fertilizer through association with nitrogen-fixing soil bacteria. Understanding how this capacity evolved is important if we wish to develop this ability in other crop species. This paper tests a hypothesis about the origin of nitrogen fixation in the legumes: that doubling of the chromosomes (and all the genes) early in the origin of the legume family provided the extra genetic material that was used by the plants to evolve special root structures ("nodules") to house and interact with the nitrogen-fixing bacteria. This paper reports that while chromosomes did double early in the legume family, there is an earlier-diverging group of nitrogen-fixing legumes that separated from the other legumes before the chromosome-doubling occurred. Therefore, the evolution of nodules and nitrogen-fixation in the legumes can't have depended on chromosome-doubling. This result is important because it means that the origin of nodules and nitrogen fixation must have depended on a smaller number of changes than doubling of every gene in the plant. If nodules and nitrogen fixation was able to evolve in this "simpler" plant (a plant with a smaller number of genes than most crop legumes), then it may be feasible to induce other crop plants outside the legumes to fix nitrogen -- and ultimately, to rely less on expensive industrial sources of nitrogen fertilizer.
Technical Abstract: Several lines of evidence indicate that polyploidy occurred by around 54 million years ago, early in the history of legume evolution, but it has not been known whether this event was confined to the papilionoid subfamily (Papilionoideae; e.g., beans, medics, lupins) or occurred earlier. Determining the timing of the polyploidy event is important for understanding whether polyploidy might have contributed to rapid diversification and radiation of the legumes near the origin of the family; and whether polyploidy might have provided genetic material that enabled the evolution of a novel organ, the nitrogen-fixing nodule. Although symbioses with nitrogen-fixing partners have evolved in several lineages in the rosid I clade, nodules are widespread only in legume taxa, being nearly universal in the papilionoids and in the mimosoid subfamily (e.g., mimosas, acacias) – which diverged from the papilionoid legumes around 58 million years ago, soon after the origin of the legumes. Using transcriptome sequence data from Chamaecrista fasciculata, a nodulating member of the mimosoid clade, we tested whether this species underwent polyploidy within the timeframe of the legumes. Analysis of gene family branching orders and synonymous-site divergence data from C. fasciculata, Glycine max (soybean), Medicago truncatula, and Vitis vinifera (grape; an outgroup to the rosid taxa) establish that the polyploidy event known from soybean and Medicago occurred after the separation of the mimosoid and papilionoid clades, and at or shortly before the Papilionoideae radiation. The ancestral legume genome was not fundamentally polyploid. Moreover, because there has not been an independent instance of polyploidy in the Chamaecrista lineage there is no necessary connection between polyploidy and nodulation in legumes. Chamaecrista may serve as a useful model in the legumes that lacks a paleopolyploid history, at least relative to the widely studied papilionoid models.