Developmental pattern formation is orchestrated by diffusible signaling molecules, termed morphogens, that form gradients from which cells can determine positionally appropriate fates. Stochasticity in morphogen binding, signal transduction, and gene expression create local cell-to-cell variability in the readout of morphogen gradients. However, little is known about the actual levels of noise in morphogen gradient responses, or the mechanisms that might control it. To investigate this, I quantified the transcriptional activity noise, protein expression noise, and protein half-life of optomotor blind (omb), one of the downstream targets of the morphogen Dpp in the Drosophila larval wing imaginal disc. Using combined fluorescence in situ hybridization (FISH) with intronic probes, immunofluorescence, image segmentation and image analysis, I observed a very high level of transcriptional variability characterized by coefficients of variation (CV) as high as ~110%, in the cells in which omb plays a central role in specifying the location of wing vein primordium L5. However, the half-life of the Omb protein was found to be very long, ~ 6 hours, which would be expected to provide significant temporal filtering of the transcriptional noise. I showed that the reduction in noise from transcript to protein is sufficient to account for the precision of the positional information that patterns vein L5. I also investigated why the positioning of vein L5 is remarkably robust to genetic manipulation that change the shape of the Dpp morphogen gradient. I observed that patters of Dpp signaling and Omb expression are not constant during larval development, but change continuously, and not always in concert with each other. By taking into account the long half-life of Omb it was possible to build a model that explains both these movements and the remarkable robustness of L5 patterning to changes in Dpp gradient shape.

Signalling by morphogens such as the Hedgehog family, Notch, Wingless/Wnt and various growth factors is essential during embryogenesis. The establishment of concentration gradients of these morphogens plays a key role during developmental patterning in all multicellular organisms, assuring that distinct cell/tissue types and organs appear at the right place in the right time during embryogenesis. Regulation of morphogen synthesis, trafficking and diffusion are all known to play a part in setting up these gradients, and a complex web of signaling mechanisms ensures that specific responses occur at the correct threshold concentration in the recipient cells whose fate depends on these morphogens.

During development, cell differentiation frequently occurs upon signaling from gradients of molecules, called morphogens. A simple paradigm to study morphogens is the Bicoid gradient, which determines antero-posterior patterning in fruit fly embryos. This transcription factor allows the rapid expression of its major target gene hunchback, in an anterior domain with a sharp boundary. Using the MS2 system to fluorescently tag RNA in living embryos, we were able to show that the ongoing transcription process at the hunchback promoter is bursty Surprisingly, it takes only 3 minutes, from the first hints of transcription at the anterior to reach steady state with the setting of the sharp expression border in the middle of the embryo. To better understand the role of transcription factors other than Bicoid in this process, I used a two-pronged strategy involving synthetic MS2 reporters combined with the analysis of the hunchback MS2 reporter in various mutant backgrounds. The synthetic reporter approach, indicate that Bicoid is able to activate transcription on its own when bound to the promoter but in a stochastic manner. The binding of Hunchback to the Bicoid-dependent promoter reduces this stochasticity while Caudal might act as a posterior repressor gradient. Altogether, this work provide a new light on the mechanisms insuring a precise transcriptional response downstream of Bicoid.

Morphogens are signaling molecules that play a key role in animal development. They spread from a restricted source into an adjacent target tissue forming a concentration gradient. The fate of cells in the target tissue is determined by the local concentration of such morphogens. Morphogen transport through the tissue has been studied in experiments which lead to the suggestion of several transport mechanisms. While diffusion in the extracellular space contributes to transport, recent experiments on the morphogen Decapentaplegic (Dpp) in the fruit fly Drosophila provide evidence for the importance of a cellular transport mechanism that was termed "planar transcytosis". In this mechanism, morphogens are transported through cells by repeated rounds of internalization and externalization. Starting from a microscopic theoretical description of these processes, we derive systems of nonlinear transport equations which describe the interplay of transcytosis and passive diffusion. We compare the results of numerical calculations based on this theoretical description of morphogen transport to recent experimental data on the morphogen Dpp in the Drosophila wing disk. Agreement with the experimental data is only achieved if the parameters entering the theoretical description are chosen such that transcytosis contributes strongly to transport. Analyzing the derived transport equations, we find that transcytosis leads to an increased robustness of the created gradients with respect to morphogen over-expression. Indications for this kind of robustness have been found in experiments. Furthermore, we theoretically investigate morphogen gradient formation in disordered systems. Here, an important question is how the position of concentration thresholds can be defined with high precision in the noisy environment present in typical developing tissues. Among other things, we find that the dimensionality of the system in which the gradient is formed plays an important.

Previous work has demonstrated that the Dpp-Tkv gradients from several models for the embryonic development of a fruit fly wing may not exhibit both the biological robustness and the desired multi-fate profile. This paper explores the implications of introducing non-receptors and negative feedback on receptor production to better understand the numerical findings of previous work that suggest biological robustness with respect to changes in ligand synthesis rate is indeed achievable with such a model. The model is formulated and the existence and uniqueness of a corresponding steady-state solution is established. Some criterion for a robust, multi-fate morphogen gradient is introduced and it is shown that the presence of non-receptors can lead to a biologically feasible or realistic morphogen gradient while concurrent feedback can potentially provide a marginal improvement in robustness as well. Lastly, alternative forms of feedback are considered briefly.