The genetic circuits that regulate cellular functions are subject to stochastic fluctuations, or `noise', in the levels of their components. Noise, far from just a nuisance, has begun to be appreciated for its essential role in key cellular activities. Noise functions in both microbial and eukaryotic cells, in multicellular development, and in evolution. It enables coordination of gene expression across large regulons, as well as probabilistic differentiation strategies that function across cell populations. At the longest timescales, noise may facilitate evolutionary transitions. Here we review examples and emerging principles that connect noise, the architecture of the gene circuits in which it is present, and the biological functions it enables. We further indicate some of the important challenges and opportunities going forward.

Developmental patterning relies on morphogen gradients, which generally involve feedback loops to buffer against perturbations caused by fluctuations in gene dosage and expression. Although many gene components involved in such feedback loops have been identified, how they work together to generate a robust pattern remains unclear. Here we study the network of extracellular proteins that patterns the dorsal region of the Drosophila embryo by establishing a graded activation of the bone morphogenic protein (BMP) pathway. We find that the BMP activation gradient itself is robust to changes in gene dosage. Computational search for networks that support robustness shows that transport of the BMP class ligands (Scw and Dpp) into the dorsal midline by the BMP inhibitor Sog is the key event in this patterning process. The mechanism underlying robustness relies on the ability to store an excess of signalling molecules in a restricted spatial domain where Sog is largely absent. It requires extensive diffusion of the BMP-Sog complexes, coupled with restricted diffusion of the free ligands. We show experimentally that Dpp is widely diffusible in the presence of Sog but tightly localized in its absence, thus validating a central prediction of our theoretical study.

restricted source, and the rate of its production may Weizmann Institute of Science vary due to genetic alterations or fluctuations in temper-Rehovot 76100 ature or nutrients. Yet, a reliable pattern is established Israel despite such fluctuations. Temporal averaging could provide one mechanism for buffering fluctuations in gene expression. However, such a mechanism would not apply to cases of persistent changes, such as alter-Summary ation in gene dosage. In the Drosophila wing imaginal disc, for example, it was shown that intermediate in-Morphogen gradients provide long-range positional crease in Dpp expression has little effect on wing and information by extending across a developing field. To thorax patterning (Morimura et al., 1996). While similar ensure reproducible patterning, their profile is invariexperiments were not yet done for Wg and Hh morphoable despite genetic or environmental fluctuations. gens, it is known that wing patterning is precise in het-Common models assume a morphogen profile that erozygous mutants that have only one functional allele decays exponentially. Here, we show that exponential of Hh or Wg. profiles cannot, at the same time, buffer fluctuations An emerging theme is that feedback mechanisms play in morphogen production rate and define long-range a prominent role in shaping morphogen gradients (Freegradients. To comply with both requirements, morphoman, 2000; Perrimon and McMahon, 1999). Regulatory gens should decay rapidly close to their source but mechanisms were identified at all levels of morphogen at a significantly slower rate over most of the field. function, including movement away from the source Numerical search revealed two network designs that support robustness to fluctuations in morphogen pro-(Bellaiche et al., 1998; Burke et al., 1999; Chen and duction rate. In both cases, morphogens enhance their Struhl, 1996), stability (Cadigan et al., 1998; Gerlitz and own degradation, leading to a higher degradation rate Basler, 2002; Giraldez et al., 2002), and the sensitivity close to their source. This is achieved through reciproof the receiving cells to morphogen signaling (Campbell cal interactions between the morphogen and its recepand Tomlinson, 1999; Jazwinska et al., 1999). Recently, tor. The two robust networks are consistent with propthe roles of receptors in shaping morphogen gradients erties of the Wg and Hh morphogens in the Drosophila received much interest. Theoretical analysis demonwing disc and provide novel insights into their function. strated that high binding affinities may hinder ligand diffusion, but biologically relevant gradients can still be formed by diffusion, when receptor-mediated ligand Introduction degradation is taken into account (Kerszberg and Wolpert, 1998; Lander et al., 2002). Feedback regulation Morphogens are signaling molecules that induce disof receptor expression was identified for all three mortinct cell fates at different concentrations. During develphogens patterning the Drosophila wing disc (Cadigan, opment, gradients of morphoge...