Scientists announced this week the development of a new computational method that provides a reliable way to estimate the total number of miRNA genes in different animals. The researchers used the tool to help identify 88 miRNA genes in the worm C. elegans, a model system important in the study of human genetics. They also estimate that miRNA genes comprise nearly one percent of the human genome, making miRNA genes one of the more abundant types of regulatory genes in humans. The next step, say researchers, is to investigate the roles they play in cell growth and development.
This work, from David Bartel's lab at the Whitehead Institute for Biomedical Research and Christopher Burge's lab at MIT, was published in the April 13 issue of Genes and Development.
"MicroRNAs have been controlling the regulation of other genes for a very long time," said Bartel. "Having this extra layer of gene regulation may have enabled the emergence of the multicellular body plans found in both plants and animals. The developmental processes that give rise to an adult plant or animal require a lot of turning on and off of genes."
For many years, miRNAs went undetected because they do not code for proteins--the benchmark traditionally used to define genes within a genome. Interest in RNA as a gene regulator began when researchers first discovered two small RNAs that impacted the translation of genes into proteins in worms. If these RNAs were missing, a worm's development stalled before it reached maturity.
These findings inspired researchers in Bartel's lab to take a closer look at this phenomenon. They found a new world of tiny regulatory RNAs present in a broad range of organisms, including worms, humans, fish and plants.
"The regulatory role for RNA had historically been under-appreciated, as researchers focused primarily on proteins as gene regulators. We are excited about the extent to which these small miRNAs also appear to be involved in normal gene regulation," said Bartel.
In March 2003, Bartel and Burge auditioned their new computational approach, called MiRscan. That work, published in the March 7 issue of Science, streamlined the search for miRNAs by giving researchers an estimate of how many miRNAs are nestled within the vertebrate genomes, such as those of humans and mice.
To generate this estimate, the researchers compared candidate miRNA sequences found in mice and humans to those found the puffer fish Fugu rubripes, a distant relative of mice and humans. Puffer fish share many of the same genes and regulatory sequences as humans, but have a much smaller genome, making it easier to identify key genes and regulatory sequences. The researchers found 15,000 genomic segments that were known to exist outside of the protein-coding regions in the human, mouse and puffer fish genomes but still appear to have been retained since the last common ancestor of fish and mammals.
Lee Lim, then a postdoctoral researcher in both the Bartel and Burge labs and the tool's chief architect, used MiRscan to cross-examine these 15,000 genomic segments and accurately predict which were likely to be microRNA genes. The researchers found most of the human microRNAs and estimated that there are at least 200 but no more than 250 human genes coding for microRNAs.
The paper published this week describes MiRscan and how it was developed to identify miRNAs in C. elegans. Bartel and his colleagues have now found and confirmed nearly 90 miRNA genes in C. elegans and estimate that there are fewer than 35 genes that remain either unconfirmed or undetected. The confirmed genes represent 48 gene families, of which 22 are conserved in humans. The researchers also report that each of these genes can be expressed at very high levels in the cells, with more than 1,000 miRNA molecules per cell and some miRNAs as abundant as 50,000 molecules per cell.
"The abundance of these tiny RNAs only increases the mystery as to why they hadn't been found earlier," says Bartel. The next step, he says, is to figure out the roles that microRNAs play in the machinery of cell growth and differentiation and find out what breaks down during disease. Bartel's lab has already made substantial progress towards this goal in plants, having matched up the first 16 microRNAs found in plants with target genes that they control.