Feature: Brains and brawn of endocytosis

The plasma membrane is like border security for a eukaryotic cell, and we all know how important border security is, especially in an election year. Nothing gets in or out without passing across this specialised lipid bilayer, although for Professor Sandra Schmid from the Scripps Research Institute in La Jolla, California, ‘in’ is the only direction that matters.

Schmid is a pioneering cell biologist in the field of clathrin-mediated endocytosis, which is the most studied entry point at the plasma membrane for ‘immigration’ and how all sorts of molecules communicate with the inside of a cell, often via cell-surface receptors.

Schmid started working on endocytic mechanisms 30 years ago – in fact since just after the endosome was first named. Her PhD supervisor and postdoctoral supervisors were both giants in the field of cellular trafficking and it was long before Schmid joined their ranks.

In recent years Schmid’s research focus has been a mechanochemical GTPase called dynamin, and how it participates in the final step of endocytosis. Essentially, the process of clathrin-mediated endocytosis at the plasma membrane works like a machine that detects the signal, assembles the team of intracellular proteins, exerts force such that the membrane curves and invaginates into the cell, and eventually pinches off the membrane bud to form clathrin-coated trafficking vesicles.

GTPases are characterised by a GTP switch mechanism, which enables their function to be turned on and off. This happens via one-way cycling from the active, GTP-bound form to the inactive, GDP-bound form by binding and subsequent hydrolysis of GTP nucleotide in the cell. GTPase-family members use this switching ability to mediate almost all steps of vesicular trafficking in cells and dynamin is the star of vesicle scission.

Dynamin has the unique property among GTPases of self-assembling into helical stacks of rings to form what is like a collar or spiral on the newly formed endocytic bud. Dynamin also has a high rate of intrinsic GTPase activity, which is stimulated about 200-fold by dynamin’s self-assembly on a lipid template. By virtue of its molecular structure, the subsequent GTP hydrolysis induces lengthwise extension and constriction of the collar, which progressively tightens around the vesicle neck causing it to break and release the vesicle to go on its merry way inside the cell.

Brains and brawn

Schmid presented the Keith Stanley lecture at this year’s 10th Anniversary Hunter Cellular Biology Meeting, where she reviewed the current models of dynamin’s role at the plasma membrane and then presented recent evidence from her lab that argues for a fascinating dual role for this very capable protein in endocytic fission.

The exact way in which dynamin mediates membrane fission during endocytosis has remained poorly understood and, at times, a subject of some controversy. According to Schmid, there are essentially two current models for dynamin’s role: as a mechanochemical enzyme driving membrane fission (the brawn); or as a rate-limiting regulator of endocytosis (the brains).

---PB---

“On the basis of our early discovery that dynamin self-assembles into a collar-like structure at the necks of invaginated endocytic buds, we first proposed that dynamin function as a mechanochemical switch driving the late stages of membrane fission and vesicle detachment. However, subsequent findings suggested that the basal GTPase activity of unassembled dynamin regulates some of the early stages.”

Schmid also emphasises that a key finding that must fit into all the models is that “dynamin has an assembly-stimulated GTPase activity 200 times faster than its basal activity so the idea was that, upon self-assembly, it would undergo rapid GTP hydrolysis to drive a concerted conformational change to generate force on the membrane.”

Cell-free assays have been used by Schmid and others for many years to faithfully reconstitute the endocytic steps, but until recently many controversial issues remained. Recent work in Schmid’s lab using biochemistry, cell biology, live-cell microscopy, and biophysics has convinced her and others that, in fact, both models could be correct and that dynamin plays a dual role in clathrin-mediated endocytosis.

“It acts in the early stages as a fidelity monitor to regulate clathrin-coated pit maturation and, at later stages, to directly catalyse membrane fission and vesicle release. So, we now think that dynamin is both the brawn and the brains of endocytosis, and that idea is the essence of my talk.”

Fluorescent dynamin

The problem with all of these previous experiments, according to Schmid, was that they were done by pre-assembling dynamin in a cell-free system on lipid bilayers in the absence of nucleotide, and then adding nucleotide back.

“For many years, I tried to point out that dynamin falls apart in the presence of GTP – it disassembles. But, largely, this point was ignored and everyone pretty well went on doing their experiments as usual, pre-assembling dynamin without GTP. It should be said that we still learned an awful lot about dynamin from these previous studies because many things still worked without the native system being recapitulated. The question was: did they really tell us anything meaningful about the mechanics of fission?

“So we started to look seriously at the interactions of dynamin with a membrane surface, with and without nucleotide present, to try and work this out,” says Schmid. This involved developing new assays for the detection and real-time monitoring of dynamin-membrane and dynamin-dynamin interactions. “Two real-time, fluorescence-based approaches in particular really showed us a lot.”

In the first approach, her group made fluorescent dynamin using an environmentally sensitive fluorophore that only lights up when in a nonpolar environment like a membrane lipid bilayer. In their case, the fluoro-dynamin was mixed with cell-free lipid particles or liposomes in vitro, and fluorescence was detected when the dynamin was bound.

---PB---

“Not surprisingly, when we did the binding in the absence of nucleotide, the dynamin pre-assembled and then rapidly disassembled when we added GTP. Interestingly, this disassembly was not instant; there was a little lag between release of the membrane and the disassembly of dynamin.”

These data therefore seemed to indicate that GTP hydrolysis elicited a major conformational rearrangement in self-assembled dynamin that immediately preceded dynamin disassembly, suggesting a direct role for dynamin in membrane fission.

Their next piece in the puzzle came from the second approach. Schmid’s group developed a model membrane system using what they called ‘SUPER’ templates, which stands for SUPported bilayers with Excess Reservoirs. These are loose fitting lipid bilayers attached to a silicon bead that is about five microns wide, which allows them to reconstitute dynamin-mediated membrane fission and visualise it at the light microscope level. Fluorescent lipid can also be added to the beads to monitor events in real-time.

What Schmid found was that dynamin added to these beads in the absence of nucleotide did what everyone has shown it to do: self-assemble into very long helical spirals to generate thin dynamin-coated tubules – but it never mediated fission.

“So this supported what most others had shown and thought to date,” Schmid says. “So then, we did basically the same experiment, but under physiological conditions. That is, in the constant presence of GTP. To our absolute amazement, we saw small vesicles being pinched off these beads and released. So it turns out that in the presence of GTP, dynamin alone can mediate membrane fission. So just membrane and dynamin and GTP – nothing else – and out comes the vesicle. It was very cool and established that dynamin constitutes the minimum cellular fission machinery!” This work was published in Cell in December 2008.

“So we think that the trick for dynamin-mediated fission is to create very high, local curvature of the membrane, and dynamin can generate curvature extremely efficiently. Even if you give it a big fat liposome, it will bind it and squeeze it to form a bud. Then, by having a small assembly rather than a big assembly, and local destabilisation of the membrane, dynamin self-limits its own assembly to structures able to mediate fission: brains then brawn.”

This feature appeared in the March/April 2010 issue of Australian Life Scientist. To subscribe to the magazine, go here.