Unformatted text preview: tion in vivo31, as appropriate flanking sequences are required for its activation32. So, most new Alu copies in the human genome are, by definition, non-functional fossil relics with respect to retrotransposition unless they fortuitously land in a region of the genome that confers activity to the incomplete RNA-polymerase-III promoter. Transposition of elements that are fortuitously activated might be short lived, because individual Alu elements carry 24 or more CpG dinucleotides33 that are prone to mutation as a result of the deamination of 5-methylcytosine residues34,35. Mutations in the CpG dinucleotides of a newly integrated Alu element could therefore minimize or eliminate the retrotransposition capability of a newly integrated Alu repeat. In addition, the homopolymericA-rich tails of individual Alu repeats are thought to be important in the amplification process27 and might rapidly mutate into simple sequence repeats after the integration of a new Alu element36–41. The decay of A-rich Alu tails provides a second potential mechanism for the retrotranspositional quiescence of individual Alu repeats. Therefore, individual Alu repeats seem to have very little chance of acting as long-lived amplification drivers for the expansion of Alu-element copy number30. Although the essential features that define an The rate of amplification of human Alu elements has not been uniform47. FIGURE 2 illustrates the pattern of expansion of the Alu family in primate genomes in relation to the approximate subfamily size. Most of the Alu repeats duplicated more than 40 million years ago. Early in primate evolution, there was approximately one new Alu insertion in every primate birth. By contrast, the current rate of Alu amplification is estimated to be of the order of one Alu insertion in every 200 births48. So, the rate of Alu amplification has decreased by at least two orders of magnitude throughout the expansion of the family. Although the underlying reasons behind the decrease in the amplification rate are unknown, changes in the retrotransposition potential of mobilizationcompetent Alu elements that result from altered transcription or reverse transcription might be to blame47. It might also be a consequence of a decreased availability of empty insertion sites for the integration of new Alu copies — most of these sites are already occupied by older Alu elements. Furthermore, one might speculate that the human genome has evolved towards restricting the amplification of these elements, similar to the way that genomes of model organisms, such as Drosophila melanogaster, restrict amplification of other types of mobile elements49.
Recently integrated human Alu repeats Alu elements that are unique to the human genome were initially identified on the basis that they share a higher number of diagnostic point mutations, and that 372 | M AY 2002 | VOLUME 3 www.nature.com/reviews/genetics © 2002 Nature Publishing Group REVIEWS
human genome at different times. Therefore, the time of origin of a new Alu insertion directly affects the spread of this insertion through the species or the population. Depending on when, in primate evolution, an Alu element has integrated into a primate genome, it will be shared by one or more species. But even the elements that are only found in a single species might have arisen at different times. Some members of the ‘young’Alu subfamilies have inserted into the human genome so recently that they are polymorphic with respect to the presence or absence of insertion in different human genomes51. Those relatively few elements that are present in the genomes of some individuals and absent from others are referred to as Alu-insertion polymorphisms51,53,59,60. Individual Alu elements might be found in a single population, a single family or, in the case of the de novo Alu insertions, in a single individual, depending on the genetic drift that occurs after the initial integration of that element into the human genome (FIG. 4). The ‘young’ Alu subfamilies are composed of ~5,000 Alu elements that have integrated into the human genome in the past 4–6 million years after the divergence of humans and African apes45,46,51,52,54, but most of them integrated before the African radiation of humans44–46,51,54,61. So, these Alu repeats are monomorphic for their insertion sites among diverse human genomes. However, ~25% of the young Alu repeats (~1,200 elements) have inserted into the human genome so recently that they are dimorphic for the presence or absence of the insertion, which makes them a useful source of genomic polymorphism44–46,51,54.
Alu-insertion polymorphisms Ch im pa nz ee G or illa ke y ut an m on re e n G O ra ng O w 381 copies 35 copies 79 copies 2,640 copies 1,852 copies Yc1 Ya5a2 Yb9 Ya5 Yb8 >200,000 copies Y 40,000 copies >850,000 copies Sg1 Sx & J Monomeric phase Figure 2 | The expansion of Alu elements in primates. The expansion of Alu subfamilies (Yc1, Ya5a2, Yb9, Yb8, Y, Sg1, Sx and J) is superimposed on a tree of primate evolutio...
View Full Document