Transcript Positional information: fields, boundaries, and gradients

Positional information: fields, boundaries, and gradients
Positional information is specified by (1) subdivision of larger
fields of cells into smaller fields, and (2) specifying
the "address" of each cell within the field.
This is a recursive process that requires translation of gradients of
gene expression into sharp boundaries, and initiation of
new gradients by these boundaries
What mechanisms are responsible for these transitions?
The French flag model
Single gradient
Threshold responses to the Dpp morphogen gradient
hnt
(Lost in dpp / - )
ush
Problems with the morphogen gradient model?
1. Precision: If there is a direct correspondence between
morphogen concentration and the activation of downstream genes,
then the organism has to precisely control at least three parameters:
Rate of morphogen production
Rate of morphogen diffusion
Rate of morphogen degradation
2. Scale: l = D/w regardless of the absolute size of the
morphogenetic field.
How can this mechanism deal with variations in size, temperature,
random fluctuations, etc.?
The roles of maternal gradients in Drosophila
Gradients form from maternally deposited transcripts
by diffusion or transport in a cell-free environment
Opposing maternal gradients in the Drosophila egg
Production of the Bicoid protein gradient in syncytial blastoderm
Bicoid and the French Flag model
Quantification of hunchback gradient
Cycle 13
Cycle 14
Quantification of Bicoid gradient
Average intensity of sliding window the size of 1 nucleus
Variability of Bicoid gradient
30% EL
0.23 of max
intensity
Variability of Bicoid gradient
Threshold
Slope of exponential decay
Half the embryos deviate from the "norm" by more than 5 nuclei
So what happens to the downstream target of this gradient?
s (Bcd) = 0.07 EL
s (Hb) = 0.01 EL
2/3 of embryos deviate from the "norm" by less than 1 nucleus
What about scale-independence?
Bcd gradient is not scaled
Hb gradient is scaled
Can cooperative activation explain the increased precision?
0.1 EL
~30%
decrease
Cooperativity of binding is described by
Hill coefficient (steepness of saturation
curve, plotted against concentration, at
50% saturation)
log(q /1-q) = log[L] - logKd
HC = 1 means no cooperativity (each
molecule binds independently)
100%
decrease
HC > 1 means positive cooperativity
(binding of one molecule increases
affinity for additional molecules)
Bcd / Hb interaction would require HC
of at least 10…
Temperature does not compromise precision
(as predicted by cooperativity?)
Bcd
s = 0.016
s = 0.013
Hb
s = 0.010
s = 0.011
What does it take to lose precision?
Regulation by Nanos gradient?
Mutual repression by other gap genes?
Autoregulation by Hunchback?
*
*
*
Scaling also remains unaffected (r = 0.7 - 0.76)
staufen affects precision of Bicoid / Hunchback translation
stauHL
stauD3
staur9
Staufen is an RNA and mtb
binding protein localized to
both poles of the egg, and
required for localizing other
mRNAs
Bicoid binding sites are sufficient for sharpening the
expression boundary of an artificial transgene
lacZ expression is driven by a synthetic enhancer that contains 3
artificial high-affinity binding sites for Bcd (and nothing else)
The sharpness is about the same as for native hb and otd genes
(5.4 vs 5.8 vs 4.6)
Deletions of Bcd activation domains compromise sharpening
Mutant Bcd proteins expressed under the control of wild-type
regulatory sequences of bicoid
If does not even have to be Bicoid
Synthetic transcription factors fused to bicoid 3' UTR
Precision is maintained by enhancers and transcription
factors that have no homology to Bicoid and hunchback
hb: 50.2%  1.5% EL
otd: 72.1%  1.4% EL
Bcd3-lacZ: 71.4%  1.6% EL
Gal4-3CGN4 / UAS-lacZ:  1.7% EL
Gal4-2Q / UAS-lacZ:  1.9% EL
So it must be a general feature of the morphogen gradient…
Precision is compromised in staufen maternal mutants
Boundary refinement is a general feature of TF gradient???
Possible explanations:
A gradient of general transcriptional repressor?
Cooperative interaction between TFs and transcriptional machinery?
Changes in chromatin state? Cellularization?
Or is this all an artifact of detection?
Refining the response to a morphogen gradient
Dpp signaling represses the expression of its own antagonist, Brinker
Brinker, repressed by Dpp, represses Dpp target genes
brinker silencer causes repression of heterologous enhancers
by Dpp signaling
brinker silencer represses gene expression in response to Dpp
signaling in a dosage-sensitive manner
Co-transfection of S2 cells with (1) lacZ reporter, (2) plasmid
expressing Su(H) and constitutively active Notch, (3) variable
amounts of plasmid expressing constitutively active Thickveins, and
(4) luciferase plasmid for normalization of b-Gal levels
Spatial pattern of gene expression is determined by the
balance of activator and silencer activities
Components of Dpp signal transduction required for brk silencing
TF binding mediates response to Dpp gradient
This was as far as I
got…
Role of receptors and signal transduction in measuring
morphogen gradients
Smo signals to activate Hh targets
Ptc, when not bound to Hh ligand,
blocks Smo signaling
Binding to Hh inactivates Pct,
releasing Smo to signal
Loss of Ptc leads to ectopic
activation of Hh targets
Cells determine their position by
measuring the concentration of
active (unbound) Ptc, or the ratio
of unbound to bound Ptc
Testing the receptor ratio model
Constitutively active
PtcDloop2 cannot bind
Hh but can inactivate
Smo
Question:
Does the minimum
amount of PtcDloop2
needed to shut down
the Hh pathway
depend on the
presence of ligandbound Ptc?
Different transgenes produce different levels of expression within the
normal physiological range
Unbound Ptc represses Hh targets in a concentration-dependent manner
Unliganded Ptc blocks Hh transduction in cells lacking
endogenous Patched
y w hsp70–FLP UAS–GFPnls; FRT42D ptc IIW /dpp–lacZ FRT42D Tuba1–Gal80;
rp49 > CD2, y+ > ptc – hsp70 3'UTR / Tuba1–Gal4
In L > PtcD2, Hh transduction requires both Hh and endogenous Ptc
(since Hh targets are OFF in ptc- clones, or in ptc+ cells away from compartment border)
So ligand-bound Ptc is not equivalent to the absence of Ptc - instead,
it titrates out the inhibitory activity of unbound Ptc
Unliganded Ptc blocks Hh transduction in posterior
compartment cells
nub-Gal4 / UASCiDZnf.GFP
Posterior compartment cells are exposed to
uniformly high levels of Hh
Activation of Hh targets depends on the ratio of ligand-bound and
unbound Ptc receptor
Bound and unbound Ptc
expressed in different
ratios in the absence of
endogenous Ptc
A two-fold change in
this ratio is sufficient to
change Hh targets from
ON to OFF state
This range is within the
normal range of Ptc
concentrations (700%
=> 100% over a few
cells
y w hsp70–FLP UAS–GFPnls; rp49 > CD2,y+ > ptcDloop2–hsp70 3'UTR FRT42D ptcIIW
/dpp–lacZ FRT42D Tuba1–Gal80; Tuba1 > CD2,y+ > hh–ptc–Tuba1 3' UTR