Figure 1. ‘Cut-and-paste’ transposition. (a) An autonomous DNA transposon contains an open reading frame encoding an active source of transposase (TPase) enzyme (circles). TIR: terminal inverted repeats, shown as arrowheads. (b) Transposase molecules return to the nucleus and bind, often as dimers, to the ends of virtually any transposon copy present in the genome (autonomous shown at the top, or nonautonomous, at the bottom) that contains intact binding sites for the transposase (usually located within the TIRs). (c) The transposon engages in the formation of a synaptic or paired-end complex and the transposase molecules catalyse cleavage (double-strand breaks, DSB) at each end of the transposon. (d) The element is now excised out of the chromosome and transposase catalyses its reintegration elsewhere in the genome, either on the same (as shown) or on a different chromosome. (e) Integration results in the duplication of a short host DNA sequence at the target site, called target site duplication. The size of the target site duplication (TSD) varies from 2 to 10 bp and is characteristic of a given transposase superfamily (e.g. usually 8 bp for hAT superfamily). The gap left behind by the excision of the transposon is repaired by the host DNA repair machinery. Two major repair pathways are known to operate in eukaryotic cells. Under the NHEJ, the transposon will be essentially lost at the excision site, with short sequences corresponding to the termini of the transposon sometimes remaining, also known as transposon footprint. Under the homologous recombination (HR) pathway the homologous chromosome or sister chromatid may be used as a template to repair the DSB and restore the original insertion at the excision site. This process results in a net increase of one copy of the transposon. If HR is complete, a full-length copy of the excised transposon is restored. However, experiments have shown in Drosophila and plants that HR is often incomplete (abortive gap repair) and result in the restoration of an internally deleted copy of the original transposon. These shorter, noncoding elements may still be propagated if they retain the transposase-binding sites, giving rise to homogeneous families of so-called miniature inverted-repeat transposable elements (MITEs).

Figure 2. Temporal activity of human DNA transposons throughout the evolution of placental mammals. The histograms above the schematic phylogenetic tree show the amount of elements for each superfamily (total copy number for all superfamilies is in thousands) and the total number of families found in the human genome that were inserted at different evolutionary time points (from left to right): eutherian-wide (insertions shared by various placental mammals), primate-specific (insertions shared by all primates, but absent from all nonprimate eutherians examined) and anthropoid-specific (insertions shared by all anthropoid primates examined, but absent from galago). Currently, there is no evidence for the activity of any human DNA transposon families after the emergence of marmoset (New World monkeys), as indicated by ‘0’ above the branches leading to the common ancestor of human and rhesus macaque (Old World monkeys) and to human and chimpanzee (great apes). Estimates of the divergence time of the depicted lineages are shown in million years (Myr).