Transcriptional Leakage vs. Noise: Binary to Graded Response Conversion in Autoregulated Genes

Abstract

The response of a gene to a signal of increasing strength is binary whenthe protein distribution changes its shape from unimodal through bimodalto unimodal. Graded response occurs when the protein distribution remainsunimodal while changing its mean. It became a common knowledge thatpositive feedback in gene regulation is an evolutionary way for obtaining bi-modality, whereas negative feedback increases the precision of gene responseand produces unimodal protein distributions in autoregulated genes.But what if the cells need to evolutionarily adapt from the environmentwhere binary response was benecial to the environment where graded re-sponse is more preferred? The change of the regulation from positive tonegative may not be the optimal way because it requires multiple muta-tions.

Using a stochastic model of autoregulated gene, we show that another,simpler mechanism is possible for the conversion between binary and gradedresponse. It is based on ne-tuning of the transcriptional leakage and ithas the opposite effect to the translational noise: An increase in the noiseconverts the response from graded to binary, whereas an increase in theleakage converts the response from binary to graded. Importantly, thechange in the leakage level can be acheved by single mutations and thereforeit seems to be a more probable evolutionary scenario than the change ofthe nature of feedback from positive to negative.