Stochastic steady state gain in a gene expression process with mRNA degradation control

Recent analyses with high-resolution single-molecule experimental methods have shown highly irregular and variable bursting of mRNA in a wide range of organisms. Noise in gene expression is thought to be beneficial in cell fate specifications, as it can lay a foundation for phenotypic diversification of isogenetic cells in the homogeneous environment. However, because the stability of proteins is, in many cases, higher than that of mRNAs, noise from transcriptional bursting can be considerably buffered at the protein level, limiting the effect of noisy mRNAs at a more global regulation level. This raises a question as to what constructive role noisy mRNAs can play in the system-level dynamics. In this study, we have addressed this question using the computational models that extend the conventional transcriptional bursting model with a post-transcriptional regulation step. Surprisingly, by comparing this stochastic model with the corresponding deterministic model, we find that intrinsic fluctuations can substantially increase the expected mRNA level. Because effects of a higher mRNA level can be transmitted to the protein level even with slow protein degradation rates, this finding suggests that an increase in the protein level is another potential effect of transcriptional bursting. Here, we show that this striking steady state increase is caused by the asynchronous nature of molecular reactions, which allows the transcriptional regulation model to create additional modes of qualitatively distinct dynamics. Our results illustrate non-intuitive effects of reaction asynchronicity on system dynamics that cannot be captured by the traditional deterministic framework. Because molecular reactions are intrinsically stochastic and asynchronous, these findings may have broad implications in modelling and understanding complex biological systems.