Where Do Genes Come From?

Mutations may be the grist of evolution, but we usually imagine them as modifying extant genes. That leaves open an important question: where do genes actually come from? In many cases, new genes arise by duplication. If a stretch of DNA gets duplicated by mistake during replication, the copied genes are free to build up mutations and acquire new functions. In a paper recently published in Science, researchers from UC Davis and the University of North Carolina at Chapel Hill reported the discovery of another route to gene creation, by changes in non-coding stretches of DNA.

The team extracted and sequenced RNA from the testes of six wild strains of the fruit fly Drosophila melanogaster. Comparing the RNA transcripts identified in their data with the standard reference D. melanogaster sequence, they found 142 candidates for de novo genes -- genes which were expressed in at least one of the wild strains but not in the reference strain. These genes also weren't expressed in two other Drosophila species, D. simulans and D. yakuba; in all likelihood, these are new genes that have recently appeared in some D. melanogaster populations. "This is the first example of totally new genes still spreading through a species," said Li Zhao, a postdoctoral researcher at UC Davis and first author on the paper.

Most of the candidate genes contained an open reading frame, a region bordered by START and STOP codons which could theoretically encode a protein. The D. melanogaster reference sequence contains the same open reading frames, but they don't get transcribed into RNA. The team hypothesized that changes in regulation transformed these non-coding regions into novel genes, and in some cases they were able to identify candidate mutations in the regulatory region upstream of the new gene.

Different selective forces shape the fate of coding and non-coding stretches of DNA; once the new genes evolved, they would have been exposed to selection. "If it has a beneficial effect, then it gets selected," Zhao said. To test this, the team measured the amount of variation in the new genes across the six strains. They found reduced nucleotide diveristy and lower levels of heterozygosity, both signs of selective processes at work. Surprisingly, they only found evidence of weaker selection or soft selective sweeps, although it's often thought that new mutations are subjected to strong selection.

"It's difficult to say at this point how important this phenomenon is for generating new genetic material," said Zhao. Even though it's unclear how significant this process is and how much of a contribution it makes, it's sitll a very exciting discovery. Like most other researchers, I've generally thought of genes arising through duplication and divergence, but this raises intriguing new possibilities to consider. Future studies will have to uncover how common this kind of de novo gene evolution is and its implications, and I'm looking forward to following the fruits of that research!