Transient non-specific DNA binding dominates the target search of bacterial DNA-binding proteins

Abstract

Summary Despite their diverse biochemical characteristics and functions, all DNA-binding proteins share the ability to accurately locate their target sites among the vast excess of non-target DNA. Toward identifying universal mechanisms of the target search, we used single-molecule tracking of 11 diverse DNA-binding proteins in living Escherichia coli . The mobility of these proteins during the target search was dictated by DNA interactions rather than by their molecular weights. By generating cells devoid of all chromosomal DNA, we discovered that the nucleoid is not a physical barrier for protein diffusion but significantly slows the motion of DNA-binding proteins through frequent short-lived DNA interactions. The representative DNA-binding proteins (irrespective of their size, concentration, or function) spend the majority (58%–99%) of their search time bound to DNA and occupy as much as ∼30% of the chromosomal DNA at any time. Chromosome crowding likely has important implications for the function of all DNA-binding proteins.

Facilitated Dissociation of Nucleoid Associated Proteins from DNA in the Bacterial Confinement

“…This property can allow them to form stable higher-level protein complexes and limit the gene accessibility effectively while facilitating the disassembly of the complexes as the protein levels fall (Figure 3). This view is also consistent with long DNA-occupancy times of HU and H-NS proteins (99%), which can assemble into large protein clusters on a DNA [59].…”

Bound2Learn: a machine learning approach for classification of DNA-bound proteins from single-molecule tracking experiments

“…We had a stable, bound fraction (diffusion coefficient ( D ) of D bound = 0.005 μm 2 /s, bound time >>> bleach time), along with a mobile fraction ( D mobile = 0.5 μm 2 /s) (Figure 1D ). We chose a D mobile of 0.5 μm 2 /s, as it represents an appropriate lower limit for a diffusing molecule and is on the order of previously used values used in simulations to benchmark measurements of DBP ( 3 , 18 , 29 ). The value for D bound was selected for similar reasons.…”

“…We also characterized our approach across a range of different simulation conditions and compared it to the commonly-used MSD analysis approach for classifying states of molecules from SPT (Figure 3 ) ( 2 , 11 , 29 ). Here we tested the performance of our approach in conditions where we reduced the proportion of DNA-bound to diffusive molecules, the signal to noise ratio of the spots, and the diffusion coefficient of diffusive molecules.…”

“…Hence, we did not simulate molecules that diffused slower than 0.25 μm 2 /s. However, this value is roughly half of the slowest diffusive DBPs in E. coli ( 29 ). Overall, these results make us confident that our ML approach can robustly classify DBP and estimate bound times across a range of conditions.…”

The needle and the haystack: single molecule tracking to probe the transcription factor search in eukaryotes

“…Indeed, LacI spends >90% of its time bound non-specifically, and visits the same binding site several times before binding to it, highlighting a potential trade-off between search speed and search accuracy [ 13 ]. Strikingly, it has been recently demonstrated that many other bacterial DNA binding proteins spend the majority of their time bound to non-specific sites on DNA in vivo [ 14 ].…”