Let’s see some ID: simulated molecular bouncer helps track protein movement across the nuclear membrane

If the human cell is a nightclub, then the nucleus is a VIP lounge. Inside are the DNA, fiercely maintained by the nuclear pore complex, which is charged with the task of protecting every strand. To do this, the NPC acts as gatekeeper against unworthy proteins.

By modeling a dynamic simulation of the NPC, physicists at the University of Illinois Urbana-Champaign have developed a theory that explains why some proteins enter the nucleus more readily than others.

Their work appears in Nature Communications.

The NPC is one of the largest and most complex protein structures in the cell. It consists of 34 distinct protein types called nucleoporins, many of which are intrinsically disordered, forming a protein-based mesh that spans the pore’s interior. This mesh is responsible for selecting which biomolecules can enter the nucleus, which can exit, and which are forbidden from the VIP lounge altogether.

“If a protein wants to get in or out of the nucleus, it must either go through the lipid membrane or through the nuclear pore complex,” said David Winogradoff, the study’s lead author and a postdoctoral researcher at UIUC during the time of the research. “The mesh of disordered nucleoporins is responsible for selecting what proteins can passively transport across the NPC.”

Proteins that are prohibited from passive transport across the NPC can be actively transported.

“During active nuclear pore transport, the cell spends energy to get through the central mesh of the nuclear pore — similar to paying off a club’s bouncer,” said lead investigator Aleksei Aksimentiev, a researcher at the Beckman Institute for Advanced Science and Technology and a professor of biological physics at UIUC.

Thus, the NPC is responsible for determining which proteins are allowed inside the nucleus as well as keeping DNA safely inside. While its build is ideal for such a weighty task, the NPC’s size, complexity, and disordered, meshy interior make it a difficult structure for researchers to study experimentally.

So, this research group turned to simulation. Winogradoff and colleagues wanted to improve on previous NPC models by building a simulation that could accurately predict its biophysical properties; namely, the physical laws that dictate which proteins may be excluded from passively crossing based on their shape and size.

“There were two main prongs of our work,” Winogradoff said. “One of those was running simulations designed to specifically model the disordered nucleoporins in the NPC. The other was the development of a percolation theory to explain why certain proteins cross the NPC more readily than others.”

Toward the first aim, Winogradoff and colleagues simulated the disordered mesh of nucleoporins using coarse-grained molecular dynamic simulations. Coarse-grained simulations simplify complex biological molecules by grouping atoms together to reduce degrees of freedom. The program used to generate these simulations, called ARBD, was designed by collaborator Christopher Maffeo, a research scientist at the Beckman Institute.

This model helped Winogradoff and colleagues simulate how a group of globular, structured proteins spanning a diverse range of molecular weights and sizes moved across the NPC — like a small-scale round of Red Rover.

Prior experiments have shown that the NPC is a high-traffic area of the cell, with thousands of proteins commuting across a single pore per second at any given time. Winogradoff’s simulations showed that small proteins readily cross the NPC through passive diffusion (meaning that no helper proteins assisted with their transport) many times within the simulated timescale of several milliseconds.

Larger proteins, however, were predicted to translocate once or not at all over this same timescale. Brute-force simulations were chosen to study these translocations because they are unbiased; in these simulations, proteins of all sizes are monitored to determine the rate at which they cross the NPC, but no bias is placed on large proteins to cross the barrier at all.

“We observed that the mesh of nucleoporins is dynamic,” Winogradoff said. “The nucleoporins are constantly moving around and creating new free paths for proteins to travel across the pore.”

Han-Yi Chou, a collaborator on the study and former graduate researcher at UIUC, developed a void analysis to determine the biophysical basis for the observed size-exclusion phenomenon. This type of analysis monitors the space within the pore not being taken up by the disordered nucleoporins at any given time. Combined with information about the size and shape of a protein crossing the NPC, it can be determined how often a free path exists from one side of the membrane to the other for any given protein.

“For a small protein, there’s a higher likelihood that there will always be a free path available to cross the membrane,” Winogradoff said. “But once a protein reaches a certain size, that continuous path is not always going to exist, so the protein must wait for a larger path to open.”

Thus, for large proteins, a free path through the NPC is much less likely to exist over the time periods simulated. Because simulations spanning a timescale of several milliseconds can take weeks to run, it is not practical to model the passive transport of larger proteins.

“Part of what makes our work novel is that we modeled our system using realistic physiological timescales,” Winogradoff said. “Our model shows that it’s possible for larger proteins to cross the nuclear membrane without any assistance, but that’s not practical on a physiological timescale.”

While this study sheds light on the biophysical properties of one of the most complex protein complexes in the cell, many other inherently disordered protein complexes remain to be studied, perhaps by analogous computational methods.

“We believe that what we’ve shown in our NPC model is going to be a generalizable property to any mesh consisting of disordered proteins,” Winogradoff said.

Furthermore, this work sets the stage for analogous studies on active transport across the NPC.

“My group is interested in investigating the routes that viruses take to infect the nucleus as well as the mechanisms that enable export of protein blueprints through the NPC to protein production plants,” Aksimentiev said. “Equipped with the modeling and visualization software developed by at Beckman, we are constructing a fully atomistic model of the NPC, which promises to chart the pathways of nuclear pore transport with unprecedented resolution.”

Editor’s note:

The paper titled “Percolation transition prescribes protein size-specific barrier to passive transport through the nuclear pore complex” is available online at https://doi.org/10.1038/s41467-022-32857-1

Funding for ARBD, the algorithm used to model the NPC in this study, came from the NIH Center for Macromolecular Modeling and Bioinformatics. The director of this center, Emad Tajkhorshid, also leads the Theoretical and Computational Biophysics Working Group at the Beckman Institute for Advanced Science and Technology.

Research reported in this press release was supported by the National Institute of General Medical Science of the National Institutes of Health under award number R01-GM136015. The content is solely the responsibility of the authors and does not necessarily represent the official views of the National Institutes of Health.

Supercomputer time was sponsored by a leadership allocation at TACC Frontera, XSEDE, and by NCSA at the BlueWaters petascale supercomputer.