Ordering up gene expression by slowing down transcription factor binding kinetics

Abstract

Efficient regulation of a complex genetic response requires that the gene products, which catalyze the response, be synthesized in a temporally ordered manner to match the sequential nature of the reaction pathway they act upon. Transcription regulation networks coordinate this aspect of cellular control by modulating transcription factor (TF) concentrations through time. The effect a TF has on the timing of gene expression is often modeled assuming that the TF-promoter binding reaction is in thermodynamic equilibrium with changes in TF concentration over time; however, non-equilibrium dynamics resulting from relatively slow TF-binding kinetics can result in different network behavior. Here, I highlight a recent study of the bacterial SOS response, where a single TF regulates multiple target promoters, to show how a disequilibrium of TF binding at promoters results in a more complex behavior, enabling a larger temporal separation of promoter activities that depends not only upon slow TF binding kinetics at promoters, but also on the magnitude of the response stimulus. I also discuss the dependence of network behavior on specific TF regulatory mechanisms and the implications non-equilibrium dynamics have for stochastic gene expression.

Keywords: DNA damage; Kinetics; LexA; Promoter; SOS response; Transcription factor.

Fig. 1

Timing of promoter activity in simple transcription regulation networks. a. Transcription factor (TF) cascade. Multiple TFs act in series to temporally order the transcription of early, middle, and late genes. In this example, each TF is an activator. The graph shows the relative concentrations of the different TFs (dotted lines) and the relative activities of the promoters they act upon (solid lines). Line colors indicate the different TFs and promoters shown in the schematic. Timing differences between promoters emerge here largely due to the time delay imposed by each TF’s own transcription, translation, folding, maturation, and degradation rates. b. Single-input module (SIM). One TF acts on multiple target promoters to temporally order the transcription of early, middle, and late genes. In this example, the TF is a repressor and each promoter has a similar maximal activity, but with a different equilibrium binding affinity (K_d) for the TF. Here, the TF-promoter interaction is in thermodynamic equilibrium with the changing TF concentration through time. The time of each promoter’s activation (t_on) and shut-off (t_off) are determined by its K_d and occur when the TF concentration crosses an arbitrarily defined activity cut-off (shown here as the x-axis). Under these conditions, promoters with smaller K_d values are the last to turn-on and the first to shut-off and all promoters display peak activity at the time of the TF nadir (t_nadir). Large timing differences between promoter activities rely on large K_d differences and slow changes in TF concentration over time

Fig. 2

Behavior of the ‘non-equilibrium SIM’ of the SOS response. In the absence of DNA damage, steady-state LexA levels are maintained by the negative auto-regulatory circuit of the lexA gene (cyan box). After DNA damage, RecA* induces the rapid auto-proteolysis of the unbound form of LexA, marking the start of the SOS response (yellow box). The reversible LexA-promoter dissociation reaction (unshaded box) displays different binding kinetics for each SOS promoter (grey box). Top right: LexA levels fall, reach a nadir, and then re-accumulate. ‘High’ amounts of DNA damage (dotted line) result in LexA concentrations that are both lower and slower to recover than ‘low’ amounts of DNA damage (solid line). Bottom right: Traces indicate activities of promoters with different k_off values (colors), but with the same k_on values. Vertical lines indicate the time-point at which peak activity is achieved. Under ‘low’ DNA damage conditions (solid lines), re-accumulation of LexA is rapid, disequilibrium is minor, and promoter activities have similar timing. Under ‘high’ DNA damage conditions (dotted lines), a disequilibrium manifests causing promoters with slow LexA dissociation rates (lower values for k_off) to display longer time intervals until reaching peak promoter activity. In contrast to equilibrium conditions (Fig. 1b), promoters with smaller k_off values are the last to turn-on and shut-off and promoters do not display peak activity at the time of the TF nadir. Large timing differences between promoter activities rely on large k_off differences and a relatively rapid rate of LexA depletion