Biology Reference
In-Depth Information
Figure 5.2
Examples of regulatory network motifs including miRNAs. (A) Examples
of simple double-negative feedback loops involving one miRNA and one transcription
factor. These small motifs are typically embedded in more complex networks. (B)
Examples of coherent feedforward loops involving a few of the miRNAs mentioned
in the text. (C) An incoherent feedforward loop involving a miRNA that has been
implicated in providing robustness to a broader gene regulatory network.
This type of configuration has two possible outcomes. First, if the miRNA
is able to fully turn off the TF, and the TF is able to shut down expression of
the miRNA, this will result in a bistable switch where only one of the two
components can be active at a given time. These relatively simple motifs are
typically embedded in more complex networks. Which one of the two
remains active will depend on additional input biasing the loop to one or the
other side, or alternatively, initial stochastic fluctuations can be amplified to
result in one or the other state. In cases where the TF is under an auto-
regulatory positive feedback, a negative feedback loop with a miRNA can
rather act as a noise filter and provide stability against fluctuations in the
level of the TF that could trigger an unwanted response, increasing the
specificity of a response.
Feedforward motifs also provide a number of advantageous properties to
gene regulatory networks. These can belong to two main classes, coherent
and incoherent FFLs. Coherent FFLs are those in which an upstream
regulator affects a target through two different paths, both of which affect
the target levels in the same direction (i.e., both activate or both repress the
target). In contrast, in incoherent FFLs, the two regulatory paths cause
opposite effects on the target level (i.e., one activates and one represses
the target). Coherent FFLs can provide robustness to a biological response
and could reinforce a switch-like effect, as is illustrated by the relationship of
Irx3
,
Olig2
, and mir-17-3p in the mammalian spinal cord (
Fig. 5.2
B). In
addition, coherent FFLs will also likely affect the dynamics of activation and
repression of the target(s).
One of the functions of incoherent FFLs is to buffer noise in gene
expression, defining and maintaining the steady-state level of a network
component. This results in more stable states, preventing random switching
to the alternative state due to stochastic fluctuations (
Fig. 5.2
C). Examples
