a The activity of a given target can be evaluated in the presence of a cognate miRNA. ‘Switch’ targets are essentially inactive following miRNA-mediated repression, whereas ‘tuning’ targets produce functional protein in the domain of miRNA activity. Both of these terms apply to evolutionarily conserved target relationships. Non-conserved but functional sites may mediate repression that is incidental or otherwise compensated for by other regulatory mechanisms; these are termed ‘neutral’ targets. Non-conserved sites may occasionally mediate beneficial, species-specific regulation. Not shown are ‘anti-targets’, transcripts for which miRNA-mediated repression is deterimental, such that the acquistion of target sites is avoided. b Another means of classification is to evaluate whether loss of function of a specific miRNA is associated with a phenotype, or whether the removal of target sites from a specific target causes a phenotype. For many conserved miRNAs and their conserved targets, the repression is likely to be beneficial enough to be selected during evolution but, overall, to be auxiliary to other means of gene regulation. Only a subset of miRNAs, with a small number of targets, mediate regulation with genetically defined consequences; we call these ‘miRNA genetic switches’, c One may consider the total number of targets whose repression by a miRNA in a given setting can be detected by quantitative means. Alternatively, a more stringent classification would be to consider only those targets whose repression by a miRNA is of detectable phenotypic consequence.

The activation of a miRNA can direct or aid a transition from state A to state B (a). Loss of a miRNA, or of its binding sites in a target, can lead to the cell becoming stuck in state A (b) or can result in a mixed state with characteristics of both state A and state B (c). An ectopic miRNA can prematurely induce state B or can cause cells to overcommit to this state (d).

The lin-4 miRNA promotes the transition between the first and second larval stages (L1 and L2, respectively) by repressing the transcription factor LIN-14 and the RNA-binding protein LIN-28. Similarly, three related miRNAs with homology to let-7 (miR-48, miR-84 and miR-241) act in concert to repress the transcription factor HBL-1, promoting the transition from L2 to L3. Finally, let-7 represses the TRIM protein LIN-41, and possibly HBL-1, to promote the transition from L4 to the adult.

a Potent inhibition by a miRNA can have a substantial phenotypic effect in some cases, but can lead to only minor phenotypic consequences in other cases. b Relatively weak repression by a miRNA might often have minor consequences, but could have substantial effects if the biological system is sensitive to minor fluctuations in target activity. c Although most animal miRNAs are likely to influence a battery of targets in a given cell, the phenotype caused by loss of a miRNA might demonstrably be due to one or a few targets. d In some cases a miRNA might exert relatively equivalent effects across a large number of targets, the collective deregulation of which might lead to a detectable phenotype. However, the loss of miRNA-mediated repression of an entire target battery can also be fully compensated for by other regulatory mechanisms. RISC, RNA-induced silencing complex.

a Signal-dependent alleviation of miRNA-mediated repression — for example, at synapses following neural activity, or in response to intracellular signals — might also trigger target re-activation, b Cell-type-specific RNA-binding proteins can antagonize miRNA activity For example, in the germ line, dead end 1 (DND1) can prevent the interaction between a miRNA and targets such as the nanos mRNA. RISC, RNA-induced silencing complex.