Biomolecular machines are huge, complex molecules composed mainly of proteins. It is astonishing that such molecules can change their structures dynamically and function precisely in thermal fluctuations. We use molecular simulation and mathematical modeling methods to elucidate the dynamics and mechanism of biomolecular machines in function. Ultimately, we aim to control the function based on the elucidated mechanism.

Seeing the moment biomolecular machines function through molecular simulation

"Seeing is believing." 

Biomolecular machines move a lot when they function. For example, motor proteins use ATP hydrolysis energy, etc. to walk on rails or rotate against a stator. Transporter proteins transport substrate molecules in and out of the membrane by changing their conformation between inward and outward open conformations. We would like to understand the mechanism by reproducing and "seeing" the dynamics at the moment of function on a computer at the atomic and molecular level. However, this is not an easy task. It is difficult to simulate the millisecond time scale motion of a huge system consisting of hundreds of thousands of atoms or more using conventional methods. We are trying to capture the movement at the moment of function by using methods such as transition path sampling, coarse-grained model, etc.

Related research:

Rotational motion of the molecular motor F1-ATPase: Okazaki and Hummer PNAS (2013)

Substrate transport dynamics of transporter Na+/H+ antiporter: Okazaki et al. Nat. Commun. (2019)

Transition Path Sampling Method: Jung, Okazaki and Hummer J. Chem. Phys. (2017) 

Lipid membrane curvature induction and sensing by Pacsin1: Mahmood, Noguchi and Okazaki Sci. Rep. (2019)

Development of a coarse-grained model Gō-MARTINI to describe protein-membrane systems: Mahmood, Poma and Okazaki Front. Mol. Biosci. (2021)

Identifying rate-limiting processes by mathematical modeling

Biomolecular machines are huge and complex molecules, so it is a challenging problem to identify which local movements are important in their functions, or in other words, what is the rate-limiting process. We are addressing this problem by applying mathematical modeling methods. We aim to identify the rate-limiting process based on the optimization of the reaction coordinates using maximum likelihood estimation or cross-entropy minimization.

Related research

Jung, Okazaki and Hummer J. Chem. Phys. (2017)

Okazaki et al. Nat. Commun. (2019)

Mori et al. J. Chem. Phys. (2020)

Aiming to control functions based on rate-limiting processes

Based on the rate-limiting process identified in the simulation, we aim to control the function. In fact, we succeeded in improving the substrate transport rate of the transporter Na+/H+ antiporter by mutating it based on the identified rate-limiting process.

Related research: 

Okazaki et al. Nat. Commun. (2019)

Extracting information from single-molecule experimental data by mathematical modeling

Single-molecule experiments provide information on the dynamics of biomolecular machines in function, although the resolution is lower than that of molecular simulations. We are working on estimating the models behind time-series data from single-molecule experiments. For example, we are developing a method to estimate the chemical-state-dependent free energy profile of a molecular motor from the time series data of its direction of motion. We are also interested in attempts to integrate single-molecule experimental data with molecular simulations.

Related research:

Okazaki, Nakamura and Iino J. Phys. Chem. B (2020)

Utilizing molecular simulation technology to create biosensors

We design novel biosensors by applying the molecular simulation technology we have developed. A biosensor is an artificial protein that consists of a fluorescent protein connected to a protein that changes its structure upon substrate binding. By using molecular simulation, we aim to create biosensors with higher performance and higher efficiency.

Related research:

Yagi-Utsumi et al. Int. J. Mol. Sci. 24(16), 12865 (2023)

Matsuda, Sakai, Okazaki, and Nagai ACS Sens. (2024)