Adaptor interactions trigger Pan1 self-assembly during clathrin-mediated endocytosis in budding yeast

Every cell is a bustling city, constantly taking in supplies from the outside world. To do this, it uses a sophisticated delivery system called endocytosis. Imagine the cell membrane as a flexible skin that can pinch inward to form a tiny bubble, or vesicle, which carries food, signals, or other materials inside. This process is not random; it requires a team of proteins to work together in perfect sync. Some proteins act as the initial scouts, spotting where the delivery should happen. Others form a rigid scaffold to shape the membrane, while a third group generates the physical force needed to pull the bubble deep into the cell. In yeast, a simple single-celled organism often used to study these processes, one protein called Pan1 acts as a central manager. It connects the scouts to the scaffolding and the force-generating machinery. For years, scientists knew Pan1 was essential for life and that it held these different teams together, but they did not understand how it managed to organize such a complex network at the right time and place without getting tangled up everywhere else.

Researchers at the University of Geneva and the University of Bern have now uncovered a surprising mechanism that explains how Pan1 works. They discovered that Pan1 has an innate ability to clump together into dynamic, liquid-like droplets, a phenomenon known as biomolecular condensation. This is not a solid lump of protein, but rather a fluid assembly where molecules can move in and out rapidly, much like a drop of water merging with another. The team found that when Pan1 is present in high concentrations, it spontaneously forms these droplets throughout the cell. These droplets are selective, gathering specific proteins needed for the later stages of the delivery process while leaving others behind. Crucially, the study shows that under normal conditions, Pan1 does not float around forming random clumps. Instead, other proteins at the cell surface act as seeds, triggering Pan1 to assemble only at the precise spots where a delivery bubble needs to form. This ensures that the complex machinery builds up exactly where it is needed, rather than clogging up the rest of the cell.

To see how Pan1 behaves, the scientists first looked at its structure. They found that the protein is mostly made of floppy, unstructured regions mixed with a few rigid, folded parts. This specific architecture is a common recipe for proteins that can form these liquid-like droplets. When they forced yeast cells to produce extra Pan1, the protein did not just spread out evenly. Instead, it gathered into distinct, glowing spheres inside the cell. These spheres were not static blobs; they were highly active. The researchers watched them fuse together like merging raindrops and split apart again. They also observed that the molecules inside the spheres were constantly exchanging with the surrounding fluid, moving in and out in about one and a half seconds. This rapid movement proved that the structures were dynamic condensates, not dead, solid aggregates. Furthermore, these droplets were sensitive to their environment. If the scientists warmed the cells up or added a chemical that weakens the gentle forces holding the droplets together, the spheres dissolved almost instantly. When the conditions returned to normal, the droplets reformed. This reversibility confirmed that Pan1's ability to clump is a regulated, fluid process.

The team then investigated what happens inside these Pan1 droplets. They found that the condensates act as a selective filter. They eagerly recruited proteins that function later in the endocytic process, such as those that help shape the membrane and generate force. However, they largely excluded the early scouts that usually arrive first. This suggests that the droplets are designed to concentrate the specific tools needed for the next phase of the job. The researchers also discovered that the droplets are not just random piles of protein; they are built through a cooperative effort. Different parts of the Pan1 protein, including its floppy regions and specific folded domains, work together to drive the assembly. If any of these key parts were removed, the protein could no longer form droplets. Conversely, removing a specific tail section of the protein actually made the droplets form more easily and grow larger, indicating that this tail normally acts as a brake to keep the assembly in check.

Perhaps the most significant finding was how the cell controls where these droplets form. If Pan1 can clump together on its own, why doesn't it do so everywhere in the cell? The answer lies in how it is recruited. Pan1 is normally pulled to the cell surface by adaptor proteins that recognize specific signals. The researchers tested this by mutating the binding sites on these adaptors, effectively blinding them to Pan1. When they removed just a few of these connections, the cell compensated, and Pan1 still assembled correctly. However, when they removed all fourteen connection points across the entire team of adaptors, the system broke down. In these cells, Pan1 stopped gathering at the cell surface and instead formed large, misplaced droplets in the middle of the cell. The number of successful delivery sites on the surface dropped dramatically. This revealed that the adaptors do not just hold Pan1 in place; they act as seeds that trigger the assembly process at the right location. Without these seeds, Pan1's natural tendency to clump takes over, but it does so in the wrong place, disrupting the cell's ability to perform endocytosis.

The study also clarified that while the assembly process is critical, the speed of the delivery itself is not directly controlled by the strength of these connections. Even when the connections between Pan1 and the adaptors were severely weakened, the timing of the delivery events remained normal. The proteins still arrived and left the cell surface at the expected pace. The defect was purely one of location and quantity; fewer sites were formed, but the ones that did form worked correctly. This suggests that the primary role of the adaptor proteins is to ensure Pan1 is present at the right spot to build the necessary structure, rather than to act as a timer for the process. Once the structure is built, the rest of the machinery proceeds on its own schedule.

These findings paint a picture of a highly organized yet flexible system. Pan1 is a protein with a built-in potential to self-assemble into a fluid network, but this potential is kept in check by the cell. The cell uses specific signals at the surface to trigger this assembly only where needed, preventing chaos. Once triggered, the protein rapidly gathers its partners to form a robust, yet dynamic, scaffold that can handle the mechanical stress of pulling a membrane inward. The research suggests that the cell does not rely on a single rigid blueprint for its machinery. Instead, it uses a combination of intrinsic self-assembly properties and external triggers to build complex structures on demand. This mechanism allows the cell to maintain a delicate balance between flexibility and stability, ensuring that the intricate dance of cellular delivery happens efficiently and without error. The work provides a clear example of how cells use the physical properties of proteins to organize themselves, turning a simple tendency to clump into a precise, life-sustaining function.