Transposon-Induced Hotspots for Genomic Instability

Genomic Rearrangements are Abundant in Drosophila

The unusually rich inversion polymorphism seen in many species of the Drosophila genus provides fertile ground for study of the mechanisms involved in generating genomic rearrangements. The rate of chromosomal reshuffling in Drosophila appears to be higher than that of any other animal or plant taxon that has been studied similarly (Ranz et al. 2001). D. buzzatii is a cactophylic species associated with host plants of the Opuntia genus. The species originated in Argentina and is distributed widely in South America. Recently this species colonized the Mediterranean region of the Old World and Australia. Chromosome 2 of D. buzzatii is polymorphic for two major structural configurations that are found throughout the species range. The standard ancestral configuration,2st, has a gene order consistent with that of other closely related species. The second form, 2j, was derived from the standard form and contains an inversion of a centrally located segment that represents approximately one-third of the euchromatic fraction of that chromosome. In earlier work with the D. buzzatii system,Caceres et al. (1999) elegantly implicated ectopic recombination between two members of a foldback (FB) transposable element family, named Galileo, as the mechanism of induction of the2j inversion. They not only showed the presence of the twoGalileo elements at the inversion breakpoints (BPs), but also identified small direct duplications of host DNA (the signatures left by a TE insertion) which provided direct proof of inversion induction by the Galileo elements. Now, in a followup study, they report a detailed molecular analysis of a 7.1-kb stretch of DNA surrounding the inversion BPs in multiple lines of flies sampled from natural populations throughout the species range. Such a detailed analysis allowed a reconstruction of the evolutionary sequence of events that occurred in these regions.

Massive Reshuffling in a Short Period of Evolutionary Time

The results of this study are remarkable in two ways. They demonstrate, first, the unprecedented frequency and complexity of molecular rearrangements in the relatively short chromosomal regions surrounding the inversion breakpoints and, second, the remarkable rapidity of these changes on an evolutionary timescale. The rearrangements identified in 30 2j chromosomes included 22 insertions of 10 TEs, 13 deletions, one duplication, and an internal inversion. The frequency of insertions (per kb per chromosome) was at least two orders of magnitude higher than those observed in earlier studies of euchromatic regions of D. melanogaster and otherDrosophila species. Furthermore, the use of multiple lines of flies from natural populations by Caceres et al. (2001) allowed detailed characterization of the complex molecular events that occurred in the hotspot regions.

The average rate of nucleotide substitution of unique DNA sequences immediately adjacent to the inversion BPs was found to be significantly lower in the inverted 2j chromosomes compared with that for the 2st chromosomes. On this basis, the observed changes in the 2j chromosome were estimated to have taken place in <84,000 years. This contrasts with 485,000 years that was estimated as the age of the 2st chromosomes. The rapidity of reshuffling in the 2j breakpoint regions is even faster than that reported for the adh1 region of maize, in which an intergenic region more than doubled in size, mainly during the last three million years, due to the activity of retrotransposons (SanMiguel et al. 1998).

What is the Mechanism of Hotspot Induction?

The absence of insertions and structural rearrangements in the homologous regions of the uninverted (2st) chromosome, in theD. buzzatii study, possibly implicates the inversion as the most likely explanation for the rearrangement hotspots. Although it is expected that TEs will accumulate around inversion BPs because of the protection afforded by the reduction of recombination in these regions (Montgomery et al. 1987), a number of observations suggest that other factors are also involved. These include: (1) the restriction of rearrangements to a very short region around the BPs, (2) the absence of retrotransposons, the most common type of Drosophila TE, in the hotspot regions, (3) the absence of insertions in homologous regions of uninverted chromosomes, and (4) the presence of deletions and other rearrangements in the hotspot regions, in addition to the insertions. The authors favor an alternative hypothesis for hot spot induction, namely the presence of the initial insertions ofGalileo. Members of the FB TE family are known to be associated with unstable mutations and chromosomal rearrangements, as well as insertions nested within one another (SanMiguel et al. 1996).

A predominant feature of the region under discussion is the frequent nesting of new TE insertions within earlier insertions. This is reminiscent of the behavior of LTR retrotransposons that are found most frequently in methylated, presumably locally heterochromatic regions of the adh region of the maize genome, often in nested clusters (SanMiguel et al. 1996). Accumulation in nested clusters is also observed in the telomeric regions of some species, for example, the silkworm Bombyx mori (Takahashi et al. 1997) and the green alga Chlorella vulgaris (Higashiyama et al. 1997), and in nested regions elsewhere in the genome, sometimes referred to as TE “landing pads” in yeast (Kim et al. 1998).

Hotspot Insertion Profiles do Not Represent DrosophilaTE Diversity

Of particular interest is the observation that the TEs identified in the hotspot regions adjacent to the inversion BPs are far from being representative of the diversity of Drosophila elements. Class II elements that transpose using a DNA intermediate are overwhelmingly represented in the identified insertions. These include the newly discovered families of FB elements (Galileo,Kepler and Newton) and representatives of thehAT (hobo, Activator, Tam) superfamily of elements. The abundance of FB elements provides some explanation for the present observations because of the known high level of instability and recombinogenic properties of these elements (Hoffman-Liebermann et al. 1985).

Interestingly, no representatives of the Class I retroelements were identified clearly in the sampled region. This is surprising given the abundance of these elements in Drosophila genomes. However,ISBu-1, one of the elements newly discovered in this study, has subsequently been identified as a member of a novel class of abundant Dipteran mobile elements named mini-me (Wilder and Hollocher 2001). These elements appear to represent a new subclass of retroposons that have the potential to provide a prolific source of microsatellite variation in many Drosophila species. In fact, the presence of the ISBu-1 insertion in the D. buzzatii 2j chromosome was cited as a prime example of the recent mobility of some members of this new subclass of TEs (Wilder and Hollocher 2001).

Remaining Questions

How are we to assess the overall evolutionary significance of these results in D. buzzatii? Are the D. buzzatii results an interesting, but rare, anomaly? This species belongs to the D. repleta group of the Drosophila subgenus. The second chromosome of repleta group species is homologous to the3R chromosome of D. melanogaster. The latter species is a member of the subgenus Sophophora which diverged from the Drosophila subgenus 40–62 million years ago. So far, no reports of any genomic regions exhibiting rapid evolution comparable to those now reported in D. buzzatii have been made following the sequencing of the genome of D. melanogaster (Adams et al. 2000) With the rapid appearance of new sequence data predicted in the near future, intra- and inter-specific genomic comparisons combined with enhanced tools for finding novel repetitive DNA, we should soon start to have some of the answers to these and related questions.