Alu elements began to amplify early in primate evolution (8). A precursor to Alu elements may have predated the primate/rodent divergence. Rodents have B1 elements that are also derived from the 7SL RNA gene. However, the B1 elements have a monomer structure, whereas the Alu elements have a dimer structure that formed early in primate evolution. Alu elements are found even in the prosimian primates. Alu elements began to accumulate in primates about 65 million years ago. Because there is no specific mechanism for removal of Alu elements, their copy number has continued to increase throughout primate evolution.
As the copy number increased, the sequence of Alu elements has evolved as well. Figure 3 shows a schematic of Alu insertions during primate evolution. There were very high levels of Alu retroposition early in primate evolution. The current rate in human is nearly 100-fold lower than at the peak of Alu insertion. At different stages of primate evolution, the Alu elements that amplified had distinct sequence differences that allowed them to be classified into subfamilies (8). Recent studies suggest that there has been a modest increase of Alu insertions in the human lineage relative to the other great apes (9).
The most likely explanation for the formation of different Alu subfamilies at different evolutionary times is that there are extremely few "active" Alu elements at any one time. This allows the sequence of active elements to drift with time. Various hypotheses have been proposed for the features that limit Alu element activity. These include the flanking sequences of Alu elements influencing transcription rates (10,11) and subfamily-specific changes interacting with the retrotransposition machinery (12). Currently, we favor the length of the A-tail as the most important factor (6). The A-tail grows on insertion but shrinks rapidly in evolutionary terms, potentially silencing individual Alus. In addition, older Alu elements accumulate mutations that lessen their expression, as well as potentially their interaction with the retroposition proteins. Thus, only progeny from the most recent insertions are likely to maintain activity.
Because the most recent inserts are also the most likely to be active elements, many of the most active elements will be polymorphic in the human population. Thus, different individuals in a population may be more or less prone to Alu amplification and it is the entire population that serves as a reservoir for potentially active elements.
Although Alu elements are spread throughout the human genome, they show some regions of higher density. In particular, Alu elements appear to be preferentially located in GC-rich genomic "isochores," while L1 elements are located in AT-rich isochores. However, most recent data suggest that Alu elements insert relatively randomly with respect to GC content (1), and are selectively lost from the AT-rich regions (13,14), creating the bias seen. Thus, Alu elements are likely to have a fairly random insertional mutagenesis potential, but are more likely to contribute to postinsertional recombination events (see below) in the GC- and generich regions of the chromosomes.
Was this article helpful?