Biology Reference
In-Depth Information
2.3. Feedforward loops and miRNAs
Genome-wide studies have found FFLs that contain miRNAs are network
motifs (
Re
et al.
, 2009
;
Shalgi
et al
., 2007
;
Tsang
et al
., 2007
). Thus,
miRNAs could provide robustness by participating in FFLs (
Fig. 9.5
C).
Nevertheless, the existence of a FFL topology is no guarantee that the
processing property predicted for a FFL containing a miRNA follows the
principles established for other kinds of FFLs. This is because the properties
of such loops not only depend on the pattern of interactions but also on the
molecular stoichiometry of the nodes and relative kinetics of the links.
Finally, it is worth noting that not every functional FFL containing a
miRNA might necessarily satisfy the computational assumptions made by
Mangan and Alon (2003)
for protein-based FFLs.
Several experimental examples of miRNA-containing FFLs have been
described. One such example involves
Drosophila
miR-7 and its action in
the eye differentiation network. Three FFLs are contained within the YAN
network (
Graham
et al
., 2010
;
Li
et al
., 2009
)(
Fig. 9.6
). The first of these has
a topology where miR-7 is in the middle of the long arm of the loop; the
TF
Pnt-P1
directly represses
YAN
and indirectly represses
YAN
through
miR-7 (
Li and Carthew, 2005
;
Li
et al
., 2009
). This FFL was predicted to
buffer variations of
Pnt-P1
, only accepting persistent
Pnt-P1
changes. Inter-
locked in the opposite direction is another FFL, where YAN directly
represses
miR-7
and indirectly represses
miR-7
by preventing
Pnt-P1
tran-
scription. This FFL is predicted to buffer
miR-7
expression from sudden
nonpersistent changes in YAN abundance. In a third FFL,
Pnt-P1
directly
activates
miR-7
and indirectly activates
miR-7
by repressing
YAN
. This last
FFL is predicted to buffer
miR-7
expression from sudden nonstable changes
in
Pnt-P1
. Collectively, these three interlocked FFLs could buffer sudden
and nonpersistent changes in the abundance of two key regulatory factors,
imparting robustness to the network.
The roles of these FFLs in providing robustness were tested in
miR-7
mutant flies (
Li
et al
., 2009
). When mutant animals were subjected to
temperature fluctuations,
YAN
showed abnormal overexpression and
there were errors in differentiation. Under uniform temperature,
YAN
expression was normal. This result suggests that miR-7 acting in FBL and
FFLs plays a specific role in buffering the network against environmental
perturbation.
FFLs with a TF at the beginning of the loop, a miRNA in the middle of
the long path, and a target gene at the end of the loop could provide
robustness through several mechanisms. If both TF and miRNA repress
the target, they would augment target repression asynchronously since the
kinetics of synthesis and action between TFs and miRNAs are different.
This could cause a delay or an acceleration in the expression of a target, thus
reducing or increasing the time necessary to trigger a response. Second,
