Sequestration-based feedbacks in gene expression systems

Expressed Transcription Factors (TFs) not only bind to sites at target promoters but also to decoy sites scattered across the genome. Binding to such “decoys" sequesters TFs critically impacting the response time and stochasticity (noise) in TF and target gene expression level. When the TF is a stable molecule, whose concentration is diluted by cellular growth, our results show that for fixed mean concentration levels, such decoy bindings can both enhance or suppress random fluctuations in TF levels depending on the source of noise (i.e., intrinsic vs. extrinsic noise) and the strength of binding (i.e., weak vs. strong decoys). We implement negative autoregulation where free (unbound) TF molecules inhibit their synthesis. Our analytical results corroborated by numerical simulations reveal that sequestration accen- tuates the effects of feedback in the sense that noise attenuation by negative feedback is higher with sequestration than in the absence of feedback. We next consider an alternative form of feedback where the TF increases the production of its own decoys, and such feedback architectures are frequently seen in endogenous gene regulation involving microRNA-TF circuits and in controlling cellular stress responses. For these feedback circuits where decoy numbers are dynamically regulated by the TF, we identify limits of noise suppression, and in many cases, these limits occur at intermediate TF-decoy binding affinities.