Loop the loop, DNA style: One- or two-way transcription depends on gene loops

In certain toy racecar tracks, sneaky players can flip a switch, trapping their opponents' vehicles in a loop of track. Cells employ a less subtle approach: they change the track's layout. In a study published online September 28 in Science, scientists at the European Molecular Biology Laboratory (EMBL) and Oxford University discovered that, by forming or undoing gene loops, cells manipulate the path of the transcription machinery -- which reads out instructions from DNA -- controlling whether it moves along the genetic material in one direction or two.

"We found that gene loops can turn bi-directional promoters into one-way systems," says Lars Steinmetz, who led the work at EMBL.

Three years ago, Steinmetz's lab discovered that when the transcription machinery lands on most genes' promoters -- the sequences that mark where transcription should start -- it moves not only along the gene, but also in the opposite direction, along the other 'lane' of DNA. But they were somewhat surprised that this didn't happen everywhere: some genes, it seemed, had promoters that worked only one-way.

Nicholas Proudfoot's group at Oxford University, in the meantime, discovered that genes can bend into a loop, so that when the transcription machinery reaches the end of the gene, it finds itself back at the beginning and starts again, like a trapped toy car. Proudfoot's group also found that inactivating a specific protein prevented these gene loops from forming.

Looking throughout the whole genome of yeast cells, Steinmetz, Proudfoot and colleagues have now found that when genes with 'one-way' promoters can't form loops, transcription from those promoters becomes bi-directional. It seems that with no transcription 'cars' trapped in a loop, more are free to move in the opposite direction. And by doing so, they can affect other genes. So by forming or undoing gene loops, cells can control not only what happens to the looping genes themselves, but also adjust the spread of regulation throughout the genome.

The researchers would now like to understand how and when loops form, for instance in response to changes in a cell's environment, and how unlooping affects the transcription of short DNA sequences within the gene. And Steinmetz has another burning question: what happens in cases where a promoter activates transcription of two genes? "Are they each in a loop? What's going on there?" he wonders.