ElfLab Molecular Dynamics
Title: ElfLab Molecular Dynamics
SNIC Project: SNIC 2021/23-132
Project Type: SNIC Small Storage
Principal Investigator: Johan Elf <johan.elf@gmail.com>
Affiliation: Uppsala universitet
Duration: 2021-02-16 – 2021-07-01
Classification: 10603
Homepage: http://elflab.icm.uu.se


The transcription factor LacI of E. coli is one of the best studied model systems for protein DNA interaction and regulation of gene expression. Its search mechanism was studied in our group for many years revealing the fact that it displays facilitated diffusion, how the crowded cellular environment influences search and by characterizing a search mechanism that includes helical sliding and instantaneous hops on the DNA that can lead to target site missing. (Elf et al Science 2007, Science 2012, Nature 2020 ) The remaining challenge is to determine microscopic on- and off-rates or binding probabilities at the target site that are possibly linked to conformational transitions in both, protein and DNA. As described above, all-atom MD simulations can be used to study conformational transitions and will help to complete the picture of LacI’s binding mechanism. Structural information of LacI-DNA complexes is available in form of crystal structures containing the full, dimeric repressor, as well as NMR-structures containing the truncated dimeric repressor headpiece. The first challenge is to dissociate the stable specific complex from the crystal structure, to obtain the starting point of the binding event, the non-specific protein-DNA complex. This has not been achieved in our all-atom simulations of the full-length dimeric repressor so far. (Liao, Q. et al. J. Phys. Chem. B 2019) Our approach is to run atomistic simulations with distance and dihedral restraints from NMR-experiments of a non-specific LacI-DNA complex (PDB-ID 1OSL11). The AAMD simulations start from the crystal structure of a specific complex (PDB-ID 1EFA13). First tests show that the number of hydrogen bonds as well has the helical structure of an important regions in the protein, that stabilize the specific complex, decreases during 50-150 ns of simulation. As soon as we obtain the path of binding/unbinding we will use umbrella sampling (US) and obtain the potential of mean force and binding energies from reweighting of the trajectories for different DNA sequences. We will compare the simulation results to experiments that will be carried out in the laboratory. Thus, we will obtain an atomistic insight into effects of DNA-sequence on binding affinity, kinetics and specificity that can guide further experimental studies. The project is a collaboration between the van der Spoel and Elf groups at Dept Cell and molecular biology, UU.