To survive and fulfill their biological functions, cells need to take in material from their environment. In this process, proteins within the cell pull inward on its membrane, forming a pit that eventually encapsulates the material in a bubble called a vesicle. Researchers from the University of Pennsylvania have now revealed a relationship that governs this process, known as endocytosis.

Their new study, published in Nature Communications, shows that the threshold at which proteins succeed at making a vesicle depends on both the quantity of membrane-bending proteins and the tension in the membrane itself. As tension on the membrane decreases, fewer proteins are needed to reach that critical mass.

Calculating where this threshold is in a given cell would be useful for understanding many biological processes. Many diseases disrupt normal endocytosis, so altering this threshold might prove to be a basis for future treatments.

This relationship between protein activity and membrane tensions may also help explain the recently discovered "ultrafast endocytosis" pathway, in which cells are sometimes able to form a vesicle in a few milliseconds, thousands of times faster than usual.

The study was conducted by Tobias Baumgart, associate professor in the Department of Chemistry in Penn's School of Arts and Sciences, and Zheng Shi, a graduate student in Baumgart's lab.

Biochemists have identified a class of proteins that facilitate endocytosis by pulling on the cellular membrane. However, exactly what role each of the members of this class of proteins play and how many are needed to form a vesicle remains unclear. Microscopy techniques that have the resolution to answer such questions eliminate the possibility of observing the endocytosis in action.

"There are powerful techniques that allow scientists to clearly see how proteins change the membrane shape at a molecular level," Shi said.

"However, it is more challenging to look at how this process changes with time with those techniques, because the samples usually must be brought into a solid state, such as by freezing them. For the same reason, it's also nearly impossible with these techniques to accurately control how tense the membranes are."

"Our approach," Baumgart said, "was based on a technique developed in our lab that can be used to control the membrane tension and look at the protein membrane-binding process in real time."

To get around the limitations in what they could see through their microscopes, the researchers developed a technique for inferring the necessary information from a model system.

They created stand-alone cell membranes, fluorescently labeled and partially sucked into a pipette. A standardized amount of suction drew a small amount of the membrane into the pipette, forming a "nose" on the otherwise spherical model cell.

The researchers exposed the model cell to a bath of membrane-bending proteins, fluorescently labeled in a different color. They attached themselves to the exterior of the model cell and began the endocytosis process in multiple places at once.

"From the perspective of the proteins," Baumgart said, "it doesn't matter that they're on the outside of the model cell. It's effectively a flat surface for them, just like the earth seems flat from our perspective."

As the proteins collectively pulled on the part of the model cell's membrane that was outside of the pipette, they were able to draw out some of the membrane that was trapped inside. This shortened the length of the "nose" left in there. By measuring this change, the researchers were able to infer the point at which endocytosis began to occur on the model cell's membrane. They could then calculate the total amount of protein involved at that point, as indicated by the intensity of their fluorescent markers.

Changing the strength of the suction pressure on the pipette also changed the overall tension on the model cells' membranes, allowing the researchers to directly observe the role tension played in regard to the number of membrane-bending proteins at work.

The interplay between these two factors means that the threshold at which endocytosis begins can be lowered not only by deploying more proteins but also by decreasing the overall tension in the membrane. Though it was not possible to study the entire mechanism in the model cells, the latter method might explain the speed with which ultrafast endocytosis begins in living cells.

"It's like getting a message to your friend by calling out rather than walking over," Baumgart said, "the tension signal literally propagates as a wave across the cell's membrane, which is much faster than making more proteins and needing them to physically get to the site where the vesicle is forming."

While their experiment only used one type of membrane-bending protein, future research using this technique will allow researchers to directly investigate the role other members of this class play in endocytosis.