Population analysis of retrotransposons in giraffe genomes supports RTE decline and widespread LINE1 activity in Giraffidae

The majority of structural variation in genomes is caused by insertions of
transposable elements (TEs). In mammalian genomes, the main TE fraction is
made up of autonomous and non-autonomous non-LTR retrotransposons commonly
known as LINEs and SINEs (Long and Short Interspersed Nuclear Elements).
Here we present one of the first population-level analysis of TE
insertions in a non-model organism, the giraffe. Giraffes are ruminant
artiodactyls, one of the few mammalian groups with genomes that are
colonized by putatively active LINEs of two different clades of non-LTR
retrotransposons, namely the LINE1 and RTE/BovB LINEs as well as their
associated SINEs. We analyzed TE insertions of both types, and their
associated SINEs in three giraffe genome assemblies, as well as across a
population level sampling of 48 individuals covering all extant giraffe
species. Results The comparative genome screen identified 139,525 recent
LINE1 and RTE insertions in the sampled giraffe population. The analysis
revealed a drastically reduced RTE activity in giraffes, whereas LINE1 is
still actively propagating in the genomes of extant (sub)-species. In
concert with the extremely low activity of the giraffe RTE, we also found
that RTE-dependent SINEs, namely Bov-tA and Bov-A2, have been virtually
immobile in the last 2 million years. Despite the high current activity of
the giraffe LINE1, we did not find evidence for the presence of currently
active LINE1-dependent SINEs. TE insertion heterozygosity rates differ
among the different (sub)-species, likely due to divergent population
histories. Conclusions The horizontally transferred RTE/BovB and its
derived SINEs appear to be close to inactivation and subsequent extinction
in the genomes of extant giraffe species. This is the first time that the
decline of a TE family has been meticulously analyzed from a population
genetics perspective. Our study shows how detailed information about past
and present TE activity can be obtained by analyzing large-scale
population-level genomic data sets.