SanMiguel et al. 1998

SanMiguel P, Gaut BS, Tikhonov A, Nakajima Y, Bennetzen JL. 1998. The paleontology of intergene retrotransposons of maize. Nature Genetics 20: 43-45.

These authors dated the insertions of 16 retrotransposons inserted in the 240kb region of the maize genome near the gene adh1. The retrotransposons in this region encompass approximately 160kb; three protein-coding genes have been described from this region.

The dating method was based on divergence of the Long Terminal Repeats (LTR) of this family of retrotransposons. The two repeats per retrotransposon would have been identical in sequence when the elements inserted, but have diverged by neutral substitution since that time. If a good estimate of neutral substitution is known, then the divergence in sequence between LTR can provide an estimate of time since insertion.

Many of the retrotransposons inserted into existing retrotransposons in a hierarchical manner, leading to a prediction that LTR divergences of retros-within-retros should be lower (i.e. more recent) than retrotransposons inserted directly into the maize genome. This prediction was correct for 10 of 11 examined retros-in-retros; in the eleventh, the divergences of the LTRs in the two retrotransposons were similar, suggesting similar dates of insertion. This is at least not inconsistent with the prediction.

When the authors compared the maize adh1 region to the orthologous region in sorghum, no retrotransposons were found in sorghum. These authors estimated divergence between the two plant species at 17.4 million years ago, consistent with a previously-published estimate of 16 mya. Thus, all 16 retrotransposon insertions most likely occurred in the last 16 million years or so. The dating based on LTR divergences suggest that almost all of these insertions occurred in the last 6 million years.

The authors discuss and are able to cautiously disregard several potential sources of error, such as gene converstion altering LTR sequences and RNA recombination among distantly-related transcripts.

Their main conclusion, which is well-supported by their data, is that the 240-kb region of the adh1 gene has experienced intense retrotransposon insertion activity over the last 6 million years, greatly increasing the size of this local region. From this, and from four published LTR retrotransposon sequences in other parts of the maize genome, they extrapolate that the maize genome doubled in size from 1200 Mb to 2400 Mb over roughly the same period, a conclusion I find consistent with their results but not particularly well-supported. They end with a request for greater study of this and related phenomena in plant genomes.