AbstractGene expression is a fundamental cellular process. Its stochastic fluctuations due to intrinsic and extrinsic sources, known generically as ‘gene expression noise’, trigger both beneficial and harmful consequences for the cell behavior.Controlling gene expression noise is of interest in many applications in biotechnology, biomedicine and others. Yet, control of the mean expression level is an equally desirable goal. Here, we analyze a gene synthetic network designed to reduce gene expression noise while achieving a desired mean expression level. The circuit combines a negative feedback loop over the gene of interest, and a cell-to-cell communication mechanism based on quorum sensing. We analyze the ability of the circuit to reduce noise as a function of parameters that can be tuned in the wet-lab, and the role quorum sensing plays. Intrinsic noise is generated by the inherent stochasticity of biochemical reactions. On the other hand, extrinsic noise is due to variability in the cell environment and the amounts of cellular components that affect gene expression. We develop a realistic model of the gene synthetic circuit over the population of cells using mass action kinetics and the stochastic Chemical Langevin Equation to include intrinsic noise, with parameters drawn from a distribution to account for extrinsic noise. Stochastic simulations allow us to quantify the mean expression level and noise strength of all species under different scenarios, showing good agreement with system-wide available experimental data of protein abundance and noise inE. coli. Ourin silicoexperiments reveal significant noise attenuation in gene expression through the interplay between quorum sensing and the negative feedback, allowing control of the mean expression and variance of the protein of interest. Thesein silicoconclusions are validated by preliminary experimental results. This gene network could have important implications as a robust protein production system in industrial biotechnology.Author SummaryControlling gene expression level is of interest in many applications in biotechnology, biomedicine and others. Yet, the stochastic nature of biochemical reactions plays an important role in biological systems, and cannot be disregarded. Gene expression noise resulting from this stochasticity has been studied over the past years bothin vivo, andin silicousing mathematical models. Nowadays, synthetic biology approaches allow to design novel biological circuits, drawing on principles elucidated from biology and engineering, for the purpose of decoupled control of mean gene expression and its variance. We propose a gene synthetic circuit with these characteristics, using negative feedback and quorum sensing based cell-to-cell communication to induce population consensus. Ourin silicoanalysis using stochastic simulations with a realistic model reveal significant noise attenuation in gene expression through the interplay between quorum sensing and the negative feedback, allowing control of the mean expression and variance of the protein of interest. Preliminaryin vivoresults fully agree with the computational ones.
Fitness effects of altering gene expression noise in Saccharomyces cerevisiae
