In January, researchers from the labs of Massachusetts Institute of Technology professors David Bartel and Chris Burge published data in Cell indicating that up to a third of the human genome may be regulated by microRNAs.
Now, according to a report in the Nov. 24 online edition of Science, the labs have found that a majority of mammalian genes, while not directly regulated by miRNAs, appear to be nonetheless influenced by the small non-coding RNAs.
In the Cell paper, Ben Lewis, a graduate student in both Bartel's and Burge's labs, "showed that about a third of human genes, roughly, are targeted by microRNAs," Andrew Grimson, a postdoc in Bartel's lab who co-authored the Science paper, told RNAi News this week. "What we are seeing is that, in addition to the targets of microRNAs, some genes are under selection to avoid being targets of microRNAs, so it's almost the opposite."
According to Grimson, his work began with a random selection of evolutionarily conserved miRNA targets 7-nucleotide long sites on the 3' UTRs of messenger RNAs that Lewis had previously identified. "In a tissue culture assay, [we evaluated] how well they worked in response to a microRNA [in order to] get an aggregate feeling for how … much [gene] repression they mediate," he said. "That was the benchmark."
"What we are seeing is that, in addition to the targets of microRNAs, some genes are under selection to avoid being targets of microRNAs."
Grimson then examined a number of non-conserved miRNA target sites "to see how much repression those mediated," he said. "If we could show that the non-conserved sites worked about as well as the conserved sites, then that would open up the possibility that the microRNAs are regulating both conserved and non-conserved sites. What I showed was that they worked just as well."
Grimson noted that "there's almost an order of magnitude more non-conserved sites of the genome than conserved sites," changing the picture from one where miRNAs regulate a limited number of conserved sites to one where non-conserved sites, which are 10 times more abundant, also mediate gene repression.
The next experiments were headed by computational biologist Kyle Farh, a graduate student in Bartel's lab, Grimson said. "He wanted to look at … whether or not he could see influences on the expression of genes that contain non-conserved sites." Mining publicly available expression data, Farh found that few genes well-expressed in a particular tissue contain sites that match up to miRNAs that are also well-expressed in that tissue, he said. "So it's a selective avoidance of sites for the well-expressed genes in the tissue."
"Genes preferentially expressed at the same time and place as a miRNA have evolved to selectively avoid sites matching the miRNA," the researchers wrote in the Science paper. "This phenomenon of selective avoidance extends to thousands of genes, and enables spatial and temporal specificities of miRNAs to be revealed by finding tissues and developmental stages in which messages with corresponding sites are expressed at lower levels."
"If you are a messenger RNA and you want to encode for a protein in a cell, you don't really want to evolve a site that will result in a microRNA latching on and downregulating expression for that gene," Grimson explained. "The idea is simply that if you want to be well-expressed in a particular tissue, you want to avoid pairing to a microRNA.
"If you contrast that with the earlier work from Ben Lewis [which involved] cases where selection has acted to maintain base pairing [in order to] allow a microRNA to come in to help to downregulate a gene, [our work has revealed] the opposite case where genes are trying to avoid being negatively regulated by the microRNA," he said.
Grimson noted that the Science paper does not address how the selective avoidance occurs, but said that it probably happens in a manner similar to any other evolutionary change. "Since the UTR sequences of genes are pretty free from a lot of selective constraints … and can shuffle around … unlike coding sequences … what we think is happening is the UTR sequence is evolving over time, over the millions of years of evolution, [and] there is selection acting on it to stop occurrences of matches to the microRNAs," he said.
"If you think about [this] in the wider picture, a little over a decade ago the field started to discover microRNAs," Grimson added. "Then, with [Lewis'] paper, we were able to conclude that about a third of the genome is regulated by microRNAs. What this shows is that … when you consider conserved targets … and genes avoiding being regulated … it looks like microRNAs are influencing a majority of genes. And that is pretty remarkable, to be able to go in a little over a decade from the first hints of this system to saying that the majority of mammalian messenger RNAs are influenced by microRNAs."
Doug Macron ([email protected])