Fluorescent Probes Show Nuclear Pore Barriers in Living Cells Are Fluid, Not Gel-Like

Inside every human cell, the genetic material sits behind a double membrane, and anything that wants in or out has to pass through one of several thousand nuclear pores. The scaffold of these pores is well mapped. What has been argued over for years is what happens inside them, and a team from Johannes Gutenberg University Mainz, the Max Planck Institute for Biophysics in Frankfurt am Main and Wuhan University in China has now reported an answer in the journal Nature.

The pore works like a town gate in the Middle Ages, in the researchers' own comparison: some visitors are waved through and others are turned back. Signaling molecules and pathogens alike must go through the pores, and each pore handles tens of thousands of visitors per minute. The selecting is done by a barrier of proteins that fills the opening. That much was already known. What the barrier actually is was not.

Studies on isolated proteins had pointed in different directions. Some suggested stretched polymers, others liquid-like condensates, and others solid gels with a consistency like that of gummy bears. Edward Lemke, professor of synthetic biophysics at Mainz, who led the study with Gerhard Hummer of the Frankfurt institute, noted that such experiments could not say how the barrier behaves in a living cell. Was it a rigid comb, or something soft and flexible like yogurt or a liquid?

To find out, Lemke's team attached fluorescent dyes to the proteins in the pores of living cells. Each dye is a tiny glowing dot, and how it moves reveals how the protein carrying it moves. The dyes moved extremely quickly, which led Lemke to conclude that the interior of the pore is fluid-like. He described it as the first time intrinsic protein dynamics had been measured inside this nanocanal in a living cell.

The proteins stayed highly mobile for the entire two-hour observation. The team then ran a comparison in a test tube with isolated proteins. After about two hours, the protein mass there invariably turned into a rigid gummy-bear-like gel, or even a solid resembling the deposits found in the brains of people with neurodegenerative diseases such as Alzheimer's and Parkinson's. Lemke called this a crucial turning point, because something in the cell had to be keeping the barrier dynamic.

Two things, it turned out. A transport receptor must be present, and the barrier proteins must be glycosylated, meaning modified with sugar groups so they do not stick together too tightly. When the team added both to the model system, the mobility matched what they had seen in the nuclear pores of living cells.

The second question was where transport actually runs. In pipes and rivers, flow is faster in the center than at the edges. Earlier studies suggested nuclear pores behave the opposite way, and the community had theorized a biological reason, such as a plug in the center. Hummer's group in Frankfurt built what they describe as the most comprehensive simulation of a nuclear pore to date, using the degree of cross-linking the Mainz experiments had measured. The calculations ran on JUPITER, Europe's first exascale supercomputer, at the Jülich Research Center.

The simulations confirm that the main transport route runs along the pore's edge, Hummer said. He does not rule out a biological cause such as a plug, but said it is not necessary from a physicochemical point of view. Miao Yu of Wuhan University and Maziar Heidari of the Frankfurt institute are co-first authors.

Beyond basic cell biology, the findings bear on neurodegeneration, since those diseases also involve flexible proteins hardening. The next thing to watch is whether the receptor-and-sugar mechanism proves informative for that process.