Fitness-driven scaling laws between mRNA and protein levels

Abstract

The connection between protein and mRNA abundance represents a central question in gene expression. While high correlations between mRNA and protein levels are widely observed across different organisms, the evolutionary forces driving these correlations remain elusive. Here, by balancing the cost and benefit of gene expression, we identify the optimal scalings between mRNA and protein levels maximizing cell fitness, where the scaling exponent is positively related to the toxicity effects of protein overexpression. Notably, we predict a lower bound for the exponent, which is 0.5. For the model organism Saccharomyces cerevisiae, our model predictions, which incorporate biologically relevant parameters, align well with experimental data. By analyzing genome-wide data across organisms, we find that all organisms exhibit a scaling exponent above the predicted lower bound. These universal phenomena, together with our cost-benefit trade-off model, rationalize the dominance of transcriptional regulation. Our work demonstrates how evolutionary driving forces govern the principles of gene expression regulation.