Target Location by DNA-Binding Proteins: Effects of Roadblocks and DNA Looping

The model of facilitated diffusion describes how DNA-binding proteins, such as transcription factors (TFs), find their chromosomal targets by combining 3D diffusion through the cytoplasm and 1D sliding along nonspecific DNA sequences. The redundant 1D diffusion near the specific binding site extends the target size and facilitates target location. While this model successfully predicts the kinetics measured in test tubes, it has not been extended to account for the highly crowded environment in living cells. Here, we investigate the effect of other DNA-binding proteins that partially occupy the bacterial chromosome. We show how they would slow down the search process, mainly through restricted sliding near the target. This implies that increasing the overall DNA-binding protein concentration would have a marginal effect in reducing the search time, because any additional proteins would restrict the search process for each other. While the presence of other proteins prevents sliding from the flanking DNA, DNA looping provides an alternative path to transfer from neighboring sites to the target efficiently. We propose that, when looping is faster than the initial search process, the auxiliary binding sites further extend the effective target region and therefore facilitate the target location.

💡 Research Summary

The paper extends the classic facilitated diffusion model—where transcription factors (TFs) alternate between three‑dimensional (3D) diffusion in the cytoplasm and one‑dimensional (1D) sliding on nonspecific DNA—to the crowded intracellular environment of bacteria. In vitro, this mechanism enlarges the effective target size because a TF that lands on DNA near a specific site can slide into it, thereby accelerating target location. However, in vivo the chromosome is heavily occupied: estimates suggest that 20–30 % of the DNA is bound by other proteins, creating “roadblocks” that impede sliding.

The authors first treat roadblocks as static obstacles. They define the vacancy fraction (v) (the proportion of DNA that is free) and the average footprint (d) of a bound protein. Assuming random placement, the gaps between roadblocks follow an exponential distribution with mean (v/d). For a given gap size (s), the effective target size contributed by sliding is
\