Selective transport through the nuclear pore complex presents a very complicated case—a lot of ins, a lot of outs, a lot of what-have-yous. To have any hope of unraveling the case, scientists based at the University of Groningen and Delft University of Technology knew that they would have to keep a lot of strands in their heads, that is, strands of disordered proteins called nucleoporins, which fill nuclear pores and either allow or thwart the passage of molecules. Nucleoporins are known to work quickly—they oversee the transport of some 1,000 molecules per second—but exactly how they work has been a mystery.

Rather than adopt a straightforward approach to the study of transport regulation, the scientists opted for something rather more oblique. While remaining true to simple design rules, they developed an artificial model of the nuclear pore complex.

“The nuclear pore complex is one of the biggest protein structures in the cell,” explained Patrick Onck, PhD, professor of micromechanics at the University of Groningen. “We previously studied the pores in all their complexity, but for this study, we created a drastically simplified ‘designer’ pore to investigate the essential physical mechanisms of transport.”

Ultimately, the scientists gathered that a couple of longstanding ideas about nuclear transport were not quite right. One idea was that transport was influenced by the distribution of nucleoporin types. Another idea was that transport was influenced by the amino acid composition of nucleoporin substructures. Things turned out to be a little simpler than perhaps anyone suspected.

Details of the scientists’ findings appeared March 31 in Nature Communications, in an article titled, “A designer FG-Nup that reconstitutes the selective transport barrier of the nuclear pore complex.” Besides presenting novel findings, the article describes how the scientists’ modeling approach really tied the new findings together.

First, the team analyzed the composition of the nucleoporins to design a simplified, “average” version, which they termed nucleoporin X, or NupX for short. These proteins are made up of domains comprising phenylalanine (F) and glycine (G) amino acids in tandem, and these play an essential role in transport. These FG repeats are separated by “spacers” of other amino acids. In addition to the FG repeats, some nucleoporins also contain domains of glycine, leucine, phenylalanine, and glycine—or GLFG repeats. The team designed proteins that contain both domains, separated by spacers of ten amino acids.

Researchers reduced the complexity of nuclear pores down to a single, artificial nucleoporin that they call NupX. This protein is based on the average properties of a class of nucleoporins rich in the amino acid motif glycine-leucine-phenylalanine-glycine. In these simulation snapshots, each particle represents a whole amino acid. [University of Groningen]
NupX was tested in two different systems: it was studied experimentally, attached to a surface and added to artificial nanopores that were “drilled” in a “membrane” of silicon nitride, and through molecular dynamics simulations.

Snapshot of the computer simulations, showing that nanopores filled with NupX-proteins allow specific transport proteins (chaperones) to pass. At the same time, the pores efficiently block any non-specific molecules. [University of Groningen]
“[We] illustrate the design and synthesis of an artificial 311-residue long FG-Nup, which we coin NupX, and characterize its selective behavior with respect to Kap95 (a well-characterized nuclear transport receptor, or NTR, from yeast) versus bovine serum albumin (BSA),” wrote the article’s authors. “Solid-state nanopores with the artificial FG-Nups lining their inner walls support fast translocation of Kap95 while blocking BSA, thus demonstrating selectivity. Coarse-grained molecular dynamics simulations highlight the formation of a selective meshwork with densities comparable to native nuclear pore complexes.”

Essentially, the nucleoporins were tested for interactions with nonspecific proteins and with NTRs, which are chaperone proteins that act as transport tickets through the pore. In the cell, large molecules that must be transported into or out of the nucleus can only do so when they are attached to such a chaperone. The artificial nucleoporins selectively interacted with the chaperones but not with the nonspecific proteins. This demonstrated that the NupX pores are fully functional: they are able to facilitate selective transport.

“The experiments showed that transport through the artificial pores occurs, but not what happens inside the pore,” noted Henry de Vries, a PhD student in Onck’s laboratory and one of the article’s lead authors. “With our simulations, we showed what exactly happens inside the pore.”

“Our findings,” the article’s authors emphasized, “show that simple design rules can recapitulate the selective behavior of native FG-Nups and demonstrate that no specific spacer sequence nor a spatial segregation of different FG-motif types are needed to create selective nuclear pore complexes.”

Recent studies suggested that FG and the GLFG nucleoporins would be in different places in nuclear pores, and that this arrangement might help to create selectivity, de Vries pointed out. “However, we found that they were homogenously distributed and yet we still saw selectivity.”

Another suggestion was that the amino acids that make up the spacers are important for selectivity. “Our results showed that the specific sequence of amino acids in the spacer doesn’t matter since we used random sequences,” de Vries said. “The only important part is the ratio of charged amino acids to hydrophobic amino acids within the spacers, which determines the stickiness of the proteins.”

The final conclusion of the study is that a very simple system in nucleoporins that has limited variation still produces a selective pore. “What is needed is a certain density of these FG nucleoporins,” Onck emphasized. “These form a barrier, which can only be breached by the chaperones.” This raises the question of why the pores contain a very large number of different nucleoporins in nature. “We know that nature doesn’t always come up with optimized solutions,” Onck allowed. “However, their redundancy could very well have a function in natural pores.”

The fact that the very simple artificial system already reproduces selective transport mechanisms means that scientists now have an excellent tool to study the physical principles that regulate nuclear pore function. “This could lead to new fundamental insights but also to new applications,” Onck suggested, “for example, in the creation of filtration systems, or in the design of artificial cells.”

In the conclusion to the Nature Communications article, the authors offered a more general possibility: “We envision that just like the field of de novo protein design has come to fruition with improved understanding of protein folding, the design of unstructured proteins like NupX will enable a versatile platform to study the intriguing functionality of intrinsically disordered proteins.”