That last post was just a warm-up; consider this my first real contribution to Nature Network.
I’m also going to take the opportunity to insert lots of links to the more scientific posts on my other blog. Purely to give you a taste of the kind of research that I’m interested in and that is likely to crop up here. I am not a blog pimp. Oh no.
Anyone who has read VWXYNot? will know that I’ve got a bit of a thing about endogenous retroviruses and other transposable (or repetitive) elements. This sometimes causes arguments with creationists, which is very entertaining and satisfying.
So I’m going to kick this blog off with a “journal club” type post about a paper by Shuang Yang and colleagues, published in PLoS Genetics (I’ll try to go with open access papers as often as possible). The paper is called “Repetitive element-mediated recombination as a mechanism for new gene origination in Drosophila“ and is likely to get those creationists all riled up again.
One gene two gene, old gene new gene
Gene duplication is often the first step in the creation of new gene function. If an essential gene exists as a single copy, any mutations to its sequence are likely to be catastrophic and will not be allowed to persist (this is known as negative, or purifying, selection). But duplicate that gene and one copy can mutate to its heart’s content, while the other continues in its original function.
Most gene duplication events that have been studied to date start off as perfect side-by-side copies. The current paper excluded these events and chose to look instead at partial duplications; events that shuffle existing sequences from different chromosomal locations together to form new, chimeric genes.
This kind of event is thought to be relatively common, but most of the examples known to date are too old to be fully informative. Yang’s group therefore screened 5 Drosophila species (I like flies) for more recent events.
*They found:
15 relevant gene duplicates*. All were expressed differently to their parent genes (changing a gene’s pattern of expression is another great way to generate new functionality).
*of these:
13 comprised chimeric gene sequences*. Interestingly, none of these events involved a combination of two existing gene sequences. Instead, the novel genes were a combination of parts of the original gene sequence with parts of the genome that were not actually incorporated into genes at their original location. So this change in location and context converted non-genic DNA into bona fide gene sequences.
*of which:
11 encoded chimeric proteins*. So the non-genic DNA from the original site is not only transcribed into RNA, but is also translated into protein. Since most genes are only functional once converted into the corresponding protein, this represents the generation of a potentially novel function in the majority of chimeric duplication events.
No neutrals in this (s)election
The authors tested these novel genes for hallmarks of evolutionary selection. This is done by comparing the relative frequencies of mutations that do and do not alter the sequence of the protein encoded by the new gene. If the frequencies are the same, it implies that the function of the protein is not important (the gene does not “care” where the mutations occur) and that the gene is under neutral selection. If there is a relative scarcity of mutations that alter the protein sequence, it is likely that the protein performs an important role in the cell that is preserved by negative selection. On the other hand, if there is an overabundance of protein-altering mutations, the gene may be under positive selective pressure to evolve a new function. This is a relatively rare occurrence.
Most of the chimeric genes appeared to be under negative selective pressure, implying that they have assumed an important new role in the cell. Two had hallmarks of positive selection. As the authors readily admit, it’s going to take a huge amount of work to validate these findings.
So where does repetitive element awesomeness come in?
The authors mapped the breakpoints of all chimeric genes back to the original sequence. There were many more repetitive elements at the breakpoints than would be expected by chance. In fact, repetitive elements (all from the same family) were implicated in 77% of all duplication events studied in this paper.
All members of a particular repetitive element family have similar sequences, and one possible mechanism involves accidental recombination between similar sequences at different chromosomal locations. The presence of so many similar sequences scattered throughout the genome is a recipe for lots of this non-allelic recombination. Another possibility is that transposable elements carry some of the adjacent genomic sequence with them when they cut themselves out of their original location, and then paste it back into another location when they move. This change of context might be enough to generate a new chimeric gene.
This is one of those papers that raises more questions than it answers. (Isn’t science great?) It’s going to take a lot more work to validate this model and tie down the underlying mechanisms. However, it does seem likely that repetitive elements play some role in the generation of chimeric gene duplicates. The completion of more primate genome sequencing projects will allow us to examine whether repetitive elements have added this mechanism to their extensive repertoire of human genome tinkering and evolutionary innovation.
