Biology Reference
In-Depth Information
these complexes at the regulatory level. Evidence for the idea that regulatory circuits of some floral
organs may be more sensitive to loss of higher-order complex formation than others comes from the
pistillata-5
(
pi-5
) mutant, in which the formation of the AP3/PI/SEP3 complex is reduced due to a
mutation in the protein interaction domain of the PISTILLATA protein. The
pi-5
mutant shows floral
homeotic conversion of petals into sepals, but has no defect in stamen identity [Yang
et al.
, 2003].
In addition, our model supports the idea that different higher-order complexes coexist, at least tran-
siently, and it also proposes that higher-order complexes and heterodimers may coexist. Different DNA
binding affinities of MADS-domain proteins would allow for the co-occurrence of different complexes
without disrupting the expression patterns of the target genes. This becomes relevant since under petal and
stamen conditions, which depend upon the formation of the heterotetramer complexes SEP/AP1/AP3/PI
and SEP/AP3/PI/AG, the formation of sepal and carpel complexes, SEP/AP1 and SEP/AG respectively,
cannot be ruled out. A goal of future research should be to obtain more molecular
in planta
support as
well as quantitative data for the physical interactions that are present in our model, in order improve the
modelling of the flowering process.
Feedback on spatial regulators
Recent evidence from whole-genome approaches to identify DNA-binding sites of transcription factors
in vivo (ChIP-seq and ChIP-CHIP) suggests a multitude of direct cross-talk between different transcrip-
tional regulators and feedback/feedforward loops. One example is the binding of SEP3 complexes to
the promoter of
AP2
and the miRNA loci which negatively regulate
AP2
. These findings suggest the
presence of feedback loops acting on the spatial regulator AP2, in addition to yet unknown processes
that achieve the early activation of
AP2
and its miRNA repressors. The result of our stochastic response
simulations on AP2, the only entity that disrupted the network's expression pattern, points to the fact that
in the current model the regulation of AP2 is rather simple, in contrast to almost every other entity. Given
the central role of AP2 in spatial patterning, it is likely that more complex elements and interactions
are involved in the transcriptional control of the
AP2
gene. Complex feedback mechanisms involving
miRNA loci were recently also described for the control of floral transition [Wu, G.,
et al.
, 2009], and
may play a general role in developmental transitions and pattern formation. Which other processes
might control the expression of spatial regulators remains an open question. In animals, concentration
gradients of morphogens are important to set developmental pre-patterns. In plants, hormones like auxin
have been suggested to act in similar fashion to morphogens in animals, however a role of this hormone
in orchestrating floral homeotic gene expression has not yet been demonstrated. Thus a major challenge
in the future will be to unravel the upstream processes driving the expression of spatial regulators of
floral homeotic genes.
ACKNOWLEDGEMENTS
This work was supported by the European Union FP6 Marie Curie training network grant TRAN-
SISTOR. We thankfully acknowledge Raymond DiDonato (BIOBASE Corp., Beverly/MA) for useful
comments on the manuscript.
REFERENCES
•
Bao, X., Franks, R. G., Levin, J. Z. and Liu, Z. (2004).
Repression of
AGAMOUS
by
BELLRINGER
in floral and
inflorescence meristems. Plant Cell
16
, 1478-1489.
