Arborescence and the Rate of Evolution in Plants

A recent article by Korall et al. (2010; Evolution Online Accepted) reveals a convincing deceleration in the rate of evolution for DNA sequence associated with the origin of arborescence (tree-like growth form and life history) in ferns. This is similar to the pattern found in seed plants, where an association between the rate of sequence evolution and growth form is already known (e.g., Smith & Donoghue 2008; Soria-Hernanz et al. 2008).

In both plants and animals, simple population genetics theory predicts that for a neutrally evolving locus the rate of substitution should be equal to the per generation neutral mutation rate, μ. Since germline cells are sequestered from somatic cells in animals, and germline cells undergo a fixed number of replications that is independent of generation time, theory thus predicts that, in animals, the rate of nucleotide substitution per unit time at a neutrally evolving locus will be μ/t, for generations of length t. However, for plants the prediction is less simple. This is because in plants, germline cells are not sequestered, but are instead derived from somatic tissue. As such, germ cells in older plants should in theory have more opportunity for somatic (and thus gametic) mutation.


This means that the concomitant increase in generation time that characterizes arborescent plants is insufficient in theory to explain the decreased nucleotide substitution rates estimated empirically. The authors suggest a number of possible alternative underlying causes for this pattern. For instance, they note that both arborescent seed trees and tree ferns might share a lower rate of somatic cell replications (as suggested by Soria-Hernanz et al. 2008). This represents a fully testable hypothesis which might (in part or in whole) account for the pattern found by the authors. Alternatively, Korall et al. (2010) propose that the duration of sporophyte/gametophyte life history stages in arborescent and herbaceous plants should also be considered. This is a difficult hypothesis to test comparatively, since all arborescent species have a relatively long sporophyte phase. It might be possible to study mutation accumulation in sporophyte and gametophyte life history stages in a rapidly reproducing species under laboratory conditions.

In spite of the numerous open questions that it leaves, this article extends the relationship between arborescence and slow rates of molecular evolution beyond the seed plants, and thus into a broader group of diverse organisms. This finding will surely stimulate considerable future research.