Basic HTML Version
Full Abstracts – REGULATION OF GENE EXPRESSION
128
25
Patched
together: sophisticated multi-modular
cis
-regulatory circuitry underlies a seemingly simple "constitutive" response to Hedgehog signaling.
Scott Barolo. Department of Cell & Developmental Biology, University of Michigan Medical School, Ann Arbor, MI.
The
Drosophila
gene
patched
and its mammalian ortholog
PTCH1
encode the receptor for the Hedgehog morphogen. Both
patched
and
PTCH1
are
constitutive responders to Hedgehog signaling: they both contain binding sites for Hedgehog-regulated Gli transcription factors (Ci in flies), and both are
activated in response to Hedgehog signaling in all tissues. In principle, this represents the simplest possible regulatory circuit: where Hedgehog signaling is
active, Glis directly activate
patched
.
However, our dissection of the
patched
locus shows that this seemingly simple response depends on unexpectedly complex
cis
-regulatory circuitry, with
many Hedgehog-responsive enhancers spread over tens of kilobases. These enhancers employ variable numbers of Gli/Ci binding sites of widely varying
affinities: our previous work has shown that Gli binding affinity is a key patterning determinant in the tissue-specific Hedgehog target genes
dpp
and
wingless
. We have identified separate
patched
enhancer modules responsible for expression in the developing wing, eye, gut, segmental stripes, embryonic
mesoderm, salivary gland, gonadal stem cells, etc., suggesting that
patched
's "constitutive" Hedgehog response is in fact composed of many tissue-specific
elements with diverse types of regulatory logic.
Interestingly, a large number of separate
cis
-elements contribute to
patched
expression in the wing, suggesting a "shadow enhancer"-like mechanism of
transcriptional control. We will present the latest results of our investigation into (1) the role of Gli/Ci sites in this multi-modular Hedgehog response; (2)
mechanisms by which tissue-specific responses are integrated into a constitutive response pattern; (3) the role of "shadow enhancers" in shaping the response
to Hedgehog signaling in the wing; and (4) similar enhancer logic of the fly and mouse
patched
genes.
26
The spatial and temporal activity of enhancers depends on combinatorial binding of transcription factors.
Zhe Xu
1
, Hongtao Chen
1
, Paolo Struffi
2
,
Constance Mei
1
, Darren Huang
1
, Steve Small
1
. 1) Biology, New York University, New York, NY; 2) European Commission, Joint Research Center.
Although extensive research has been done on a few well-known regulatory elements, it is still poorly understood how they are activated or silenced in
general. To better understand the activation/repression mechanisms used by enhancers that pattern the early embryo, we conducted a genome-wide search for
Bcd-dependent enhancers by looking for clusters of predicted Bcd-binding sites or in vivo Bcd-binding, or both. More than 75 fragments were tested by
reporter gene assays, and thirty-two new enhancers were identified in the search. However, sixteen of bio-informatically predicted enhancers showed no Bcd
binding in vivo and did not direct early embryonic expression. By comparing sequences of positive and negative fragments, we found a strong correlation
between enhancer activity and the number of predicted binding sites (TAGteam sites) for the ubiquitous maternal protein Zelda (Zld). Fragments activated
by Bcd in the early embryo showed an enrichment of TAGteam sites compared to those that are not activated. Adding TAGteam sites to some negative
fragments successfully converted them into Bcd-dependent enhancers in the early embryo. Adding more TAGteam sites increased Zld- and Bcd-binding to
these enhancers, and resulted in earlier activation and stronger expression in regions of limiting Bcd concentration. Preliminary data shows change of histone
modification with addition of Zld binding sites to negative enhancers, indicating that Zld possibly functions by modifying chromatin states. Our work
underscores the importance of combinatorial binding in determining the spatial and temporal activity of enhancers, and provides data that will refine our
ability to predict regulatory activity based on DNA sequence.
27
Assessing the basic building blocks necessary to generate
cis
-regulatory modules (CRMs) that regulate spatio-temporal expression.
Jelena Erceg,
Charles Girardot, Eileen Furlong. Genome Biology Unit, European Molecular Biology Laboratory, Heidelberg, Germany.
cis
-regulatory modules (CRMs) regulate the spatio-temporal activity of target genes through the recruitment of transcription factors (TFs) via a collection
of TF binding sites (TFBSs). Understanding the rules of how different TFBSs are organized to drive specific activity is key to elucidate the logic of gene
expression patterns. Are CRMs generally a composite of TFBSs as basic building blocks with very constrained positions or ‘grammar’ (enhanceosome) to
which highly ordered TF complexes bind? Or is it a more relaxed and flexible composite, as suggested by the billboard model? Do novel spatio-temporal
patterns arise through an increase in motif complexity? We addressed these questions using a synthetic biology approach focused on the key TFs essential
for
Drosophila
mesoderm development. Synthetic CRMs with multimers of a single TFBS (homotypic) were assessed for their ability to drive expression in
transgenic reporter assays. We found that homotypic synthetic CRMs were sufficient to recapitulate the endogenous expression patterns for some TFs, but
not others. We further tested synthetic CRMs with a pair combination of TFBSs (heterotypic) where TFBSs were chosen for factors with evidence of
protein-protein interaction (PPI). To assess the possible constraints at the sequence level due to impinging protein interactions, changes in both orientation
and spacing were systematically explored for four pairs of TFBSs. Our data revealed emerging developmental patterns, which were not simply additives for
the corresponding patterns obtained with homotypic synthetic CRMs. Furthermore, we found that minor variations in the spacing and orientation produce
specific spatio-temporal patterns, suggesting constraints imposed by direct PPI. These findings provide valuable insights into the general toolkit required to
generate CRMs, which drive restricted developmental patterns.