A Rugged RNA-Scape
Two miles from the Fields lab, a similar experiment was under way at the Fred Hutchinson Cancer Research Center. There, HHMI investigator Adrian Ferré-D'Amaré had also enlisted high-throughput sequencing, this time to chart the terrain of an RNA molecule in precise detail.
He and postdoctoral fellow Jason Pitt were exploring how nucleotide sequence influences the function of a catalytic RNA—a ribozyme—that joins RNA strands.
Because a ribozyme's nucleotide sequence is functional as is, without translation into a different molecule, the researchers did not face the same challenge as Fowler and Fields did in keeping track of which genetic sequences produced functional proteins. "We can just take the naked RNA molecules and ask them to do catalysis," Ferré-D'Amaré says.
Still, they had a vast genetic "space" of possibilities to explore: systematically substituting a different nucleotide at each of the ribozyme's 54 positions produces more than 300 nonillion (1030) alternative sequences. Starting with 60 trillion of them, the researchers tested whether each variant ribozyme could reconnect RNA molecules in a test tube. Successful variants were filtered out and sequenced—all 10 million of them.
Though the scale of this selection protocol is routine in the RNA field, using a sequencing machine to read out such a large number of results—before the variants become dominated by just a few active ones—is new. As reported in Science on October 15, 2010, this method revealed that the abundance of a particular RNA sequence corresponded to how well that RNA molecule worked, rather than to other factors, like its tendency to be replicated. Highly active variants that efficiently connected RNA strands predominated over those that worked haphazardly. The sequence differences between these millions of variants offered a high-resolution view of an RNA molecule not seen before, showing that even slight sequence changes can dramatically alter function.
"If you move a little bit in genetic space, activity of this ribozyme changes a lot," says Ferré-D'Amaré. "That is telling you that the landscape is very rugged, with valleys of low activity and peaks of high activity very close to each other."
Useful information to know when designing drugs or vaccines or when thinking about how molecules evolve, he says.
-- Michele Solis
HHMI Bulletin, February 2011