Meneely: Advanced Genetic Analysis 
Topical updates
Figures from the bookMeneely: Advanced Genetic Analysis
Commentary on review article by Bartel (2009)
Chapter 07, February 2009
Bartel (2009). microRNAs: Target recognition and regulatory functions. Cell, 136, 215-233 [DOI: 10.1016/j.cell.2009.01.002]
Our understanding of the field of small regulatory RNA molecules is rapidly expanding. This review by Bartel, one of the leaders in identifying and analyzing microRNA (miRNA) molecules, presents a discussion of some of the recent developments in this field, with a particular emphasis on identifying the sequences to which miRNAs are targeted. Bartel points out that microRNAs are formed from processing a transcript that folds back on itself as a hairpin, which distinguishes them from the many other types of small RNAs that have been found. For most genes, the effects of miRNAs arise from translational repression with mismatches occurring between the miRNA and its target sequence. When there is extensive sequence complementarity, the miRNAs recruit the silencing complex for RNA degradation; this category of miRNA and their targets was the first to be discovered, as described in Chapter 7, but RNA degradation may not be the most common mechanism by which miRNAs regulate gene expression.
One particularly helpful section of the review is its summary of current efforts to use computational methods to identify the targets of miRNAs. microRNAs are only 21-23 nucleotides in length, and not all of these bases are essential for interaction with the target message. As a result, the most significant issue in identifying targets by computational methods is the presence of false positives, that is, mRNAs that have some sequence complementarity with the microRNA by chance but are not physiological targets. Since translation repression does not require an exact sequence match, false positive sequences are very common and present a challenge to identifying the genuine targets.
To this end, there has been considerable progress in defining the sequence positions that require exact matches between miRNAs and their targets. Positions 2-7 of the miRNA nearly always have perfect base-pairing with the target mRNA. Thus, when this sequence is used as the seed in seeking for targets, the number of false positives greatly decreases. By extending this seed by looking for exact matches at positions 8 and 9, the number of false positives is reduced even further. Extending the seed sequence to look for an exact match at position 1 does not reduce the number of false positives.
The efforts to reduce the number of false positives can also be supplemented by examining the orthologous gene from a closely related organism to identify additional positions that are evolutionarily conserved. True targets are under evolutionary constraint and are thus more likely to be conserved than are false positive sequences. Although the effort is to reduce the number of false positive sequences, Bartel notes that the number of genuine targets is often very high, particularly for miRNAs that are evolutionarily conserved. For example, for miRNA sequence families that are conserved among vertebrates, the miRNA has more than 300 targets on average in a mammal. As noted in Chapter 12 of the book, there appear to be many more targets for a miRNA that there are for a typical transcription factor. There is no doubt that regulation by microRNAs is a significant feature of metazoan gene expression, one that had been undiscovered for decades after regulation by transcription factor proteins and other regulatory mechanisms had been found and studied.
In addition to thoroughly discussing the possible compensatory roles for the other positions in the miRNA and the effects of the sequences surrounding the target site in the 3’UTR of the mRNA, Bartel also reviews a few of the biological functions of microRNAs. He estimates that more than half of mammalian messages are regulated by microRNAs. He provides an interesting speculation about the role of miRNAs as a failsafe mechanism for tuning expression levels to reduce the impact of biological noise. (The concept of biological noise in gene expression is discussed in Chapter 8.) The role of miRNAs that was originally found in worms was to shut down protein expression from pre-existing messages at the time of larval molts; in other words, transcriptional regulation of the microRNA gene switched off expression of the target gene, as shown in Figure 1A. Another role for miRNAs is that they help to set threshold levels at which the target messages are translated. In other words, transcriptional regulation of the target mRNA results in expression above the threshold set by pre-existing microRNAs, as shown in Figure 1B.
This update has provided only a brief “review of the review” by Bartel; readers are directed towards his article and the original literature that he cites. Regulation of eukaryotic gene expression by microRNAs is clearly now the rule rather than the interesting exception that it appeared to be a decade ago. As Bartel himself writes, “It may prove difficult to find a biological function or process that is not influenced at least to some degree, in some cell types, by miRNAs.” Although much remains to be elucidated about miRNAs and their targets, both as a global process affecting most genes and as a means regulating one or a few genes, any picture of eukaryotic gene regulation that does not include the role of miRNAs is likely to be incomplete.
Figure 1. miRNAs and the level of gene expression.
The regulation of gene expression by microRNAs can have different effects, depending on whether the miRNA or its target is newly synthesized. Panel A depicts the original effect of miRNA, as described in the life cycle of C. elegans. At the time of regulation, the target mRNA (in blue) pre-exists in the cell and is not subject to new transcription. The microRNA (in red) is subject to transcriptional regulation, and is newly synthesized at the time of the molt. The newly synthesized miRNA forms a dsRNA hybrid with the target RNA, which is then degraded or expressed is blocked. A different effect is shown in Panel B. In this case, the miRNA is pre-existing so the expression of any target message present at a low level of is blocked. Transcriptional activation occurs for the target message, so that its level now exceeds the effects of the miRNA. In effect, the miRNA is setting a threshold for activation of the target, which could reduce translational noise for the target message.
View print version