Molecular sequestration as a way to control variability and induce crosstalk: from theory to experiment

The mechanism of molecular sequestration is involved in many biological processes, ranging from growth factors signalling to transcriptional and post-transcriptional regulation. This kind of dynamics involves two types of molecular species, targets and sequesters, that bind to form a complex. In the framework of mass-action law, key features of this kind of systems appear to be threshold-like profiles of the amounts of free molecules as a function of the parameters determining their possible maximum abundance. In biology, stochastic fluctuations, i.e. noise, play an undisputed role at the molecular level. Such noise can be usually divided into an intrinsic component, due to the probabilistic nature of biochemical processes, and an extrinsic one, related to the coupling with the variability of the environment in which reactions take place. Several studies highlighted the relevance of the variability induced by extrinsic fluctuations in shaping cell decision making and differentiation in molecular networks. Bimodal distributions of gene expression levels are a common feature of this kind of processes. Indeed, the two modes of the distribution often indicate the presence of two different physiological states of the system. In this thesis, we investigate the consequences of the introduction of a source of extrinsic noise onto a system governed by molecular sequestration, focussing on the appearance of bimodal distributions. To do that, we first study a minimal stochastic model of molecular sequestration and introduce extrinsic noise through a fluctuating parameter. In this framework, we analytically show how bimodal distributions can appear and characterize them as a result of noise filtering mediated by the threshold response. We then investigate the behavior of the correlations between targets of the same sequester and show how extrinsic fluctuations can induce a positive correlation that counterbalances the negative one due to competition. Given these results, we move to investigate the appearance of bimodal phenotypes in the context of microRNA (miRNA)-mediated gene regulation. MiRNAs are small noncoding RNA molecules that downregulate the expression of their target mRNAs. The interaction between miRNAs and targets is based on molecular sequestration and threshold-like responses are a known feature of this system. Recent studies suggested that miRNAs can be involved in the appearance of bimodal expression distributions of their target genes. To investigate this phenomenon, we characterize the system through an analytic and numerical approach and introduce extrinsic noise as a fluctuating miRNA transcription rate. We observe that bimodal distributions of target expression can appear for a wide range of parameters in presence of extrinsic noise. Furthermore, we show how the bimodal shape of the distribution can be tuned by the interplay between different target mRNAs competing for a common miRNA. In conclusion of this thesis we present some synthetic-biology experiments that are aimed at studying the role of extrinsic fluctuations in the appearance of bimodal distributions in the context of miRNA-mediated regulation.