|
|
|
|
|
|
|
||||
| Home Help Feedback Subscriptions Archive Search Table of Contents | ||||
|
Fig. 7. Summary of the gene regulatory architecture in the ASE neurons.
(A) Summary of regulatory interaction in ASEL and ASER. Broken line
indicates partially penetrant feedback interaction (see
Fig. 5E). See D for
deconvolution of individual regulatory interactions. Several permissively
acting factors, i.e. factors expressed in both ASEL and ASER
(Chang et al., 2003) are not
shown here for simplicity. Such factors could, for example, activate the
ASER-expressed GCY genes in the absence of the ASEL repressors die-1
and lim-6. (B) Network motifs. A FFL motif occurs when one
gene (Gene A) controls a second gene (Gene B) and together these factors are
required to regulate a target gene (Gene T). The addition of other factors
(e.g. factor C) transforms the FFL motif to a `bi-parallel motif'
(Milo et al., 2002), which
(analagous to FFL motifs) one could also envision to work as a persistence
detector. (C) die-1 and fozi-1 may control target
genes through a FFL motif. For a target to be activated, it requires both the
presence of die-1 and the absence of fozi-1. See D for
identity of target genes. All arrows shown in this figure represent genetic
interactions and do not necessarily imply direct physical interactions.
Therefore, the identification of additional factors may alter network
architecture. For example, die-1 may not only repress fozi-1
but also an additional factor, `repressor Z', which together with
fozi-1 may repress ASEL-specific GCY genes. Such a repressor Z would
transform the network motif from a FFL motif to a `bi-parallel motif' (B).
(D) Deconvoluted regulatory motifs extracted from A. Owing to their
differential behavior upon loss of upstream regulators, ASE terminal
differentiation genes can be placed into three distinct categories, all of
which controlled by the basic FFL motif architecture shown in B. Target Gene
Category 1: ASEL-specific expression of the GCY genes gcy-6, gcy-7,
gcy-14 and gcy-20 does not require lim-6, but depends
on the loop output regulator die-1 and the downstream regulator
fozi-1. As a complete elimination of fozi-1 activity only
results in partially expressive de-repression of the ASEL-specific GCY genes,
an additional factor must be involved in repressing these GCY genes. This
factor could be an unknown repressor that cooperates with fozi-1 or,
alternatively, the failure to completely activate ASEL-specific GCY genes in
ASER may be due to the lack of an activator in ASER
(Fig. 4E). The loop output
regulator die-1 is the best available candidate for this activator as
die-1 is predominantly expressed in ASEL and die-1 mutation
leads to a completely penetrant and expressive effect on ASEL-specific gene
expression. As die-1 also regulates fozi-1, the genetic
interaction therefore may define a FFL motif. This motif is the most
parsimonious illustration of the genetic observation that ASEL-specific genes
depend on two different factors: the presence of die-1 and the
absence of fozi-1. Target Gene Categories 2 and 3: Regulation of
genes in this category is distinguishable from control of Category 1 genes by
the distinct role of the LIM homeobox gene lim-6. die-1 represses
fozi-1 expression in ASEL; in the absence of fozi-1, lim-6
is expressed. Together, lim-6 and die-1 (or a
die-1-dependent pathway) activate ASEL-specific FLP genes and repress
ASER-specific GCY genes in ASEL. This motif architecture is also a FFL motif,
but now with an additional tier of regulation. Similar to the case of Category
1 target genes, the argument for this network architecture is revealed through
the completely penetrant effect of disruption of the components of the
feedback loop (including die-1) on all downstream genes (lim-6,
fozi-1 and terminal target genes), and the incompletely penetrant and
expressive effect of lim-6 on the terminal target genes. This
incomplete penetrance and expressivity implies the need for another regulatory
factor, for which die-1 is at present the best candidate, given its
completely penetrant and expressive effect on the terminal target genes. An
additional potential feed-forward motif in the interaction of these factors is
suggested by the incomplete penetrance of fozi-1 on lim-6
expression. As lim-6 is affected by die-1 in a completely
penetrant manner, lim-6 is controlled by a potential feed-forward
loop, receiving inputs from die-1 and fozi-1. This renders
lim-6 under the same control as the above mentioned ASEL-specific GCY
genes, which are also controlled by a combination of die-1 and
fozi-1.
|
