Networks of enhancers and microRNAs drive variation in cell states

Abstract

Cell-to-cell variation in gene expression is a common feature of developmental processes. Yet, it remains unclear whether molecular mediators can generate variation and how this process is coordinated across loci to allow the emergence of new cell states. Using embryonic stem cells (ESCs) as a model of development, we found interconverting cell states that resemble developmental expression programs and vary in activity at specific enhancers, such as those regulating pluripotency genes Nanog and Sox2 but not Pou5f1 (Oct4). Variable enhancers drive expression of variable genes, including those encoding microRNAs (miRNAs). Notably, variable miRNAs increase cell-to-cell variation by acting on neighborhoods of pluripotency genes. The encoded, variable pluripotency factors bind variable enhancers, forming a feedback loop that amplifies variation and allows the emergence of new cell states. These findings suggest gene regulatory networks composed of enhancers, protein-coding genes, and miRNAs harness inherent variation into developmental outcomes.

microRNA-mediated noise processing in cells: A fight or a game?

“…These results suggest that miRNA could increase the cell-to-cell variability of their targets. If the targets are key developmental factors such as morphogens, this variability can be the trigger of cell state transitions [27]. These experimental works along with theoretical modelling showed that quantitative investigation is crucial for understanding the impact of miRNAs in managing noise.…”

“…A recent study based on a combination of theory and experiments [23] showed that selective pressure might even increase expression noise and the positively selected genes with elevated noise are also those highly regulated by transcription factors. On the same idea that cells do not necessarily buffer noise, a recent work showed that introduction of extrinsic noise in microRNA-mediated regulatory networks, i.e., increased variability in gene expression, can instead favour cell differentiation [24][25][26][27]. These recent studies arouse the possibility that cells do not only buffer noise but they rather "take advantage of stochasticity" to optimise specific needs, e.g., cell-to-cell variability, protein number precision, information flow [28,29], etc.…”

From Endogenous to Synthetic microRNA-Mediated Regulatory Circuits: An Overview

“…Given the probabilistic nature of chemical reactions and the small number of molecules involved, fluctuations in the abundance of such molecules are relevant. Thus, through stochastic modelling, several works deeply investigated the properties and biological advantages of such networks in the presence of different sources of noise [45,48,49].…”