Lorne special: Dynamic dynamin and synaptic transmission

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

Some astronomers have a favourite nebula. Some chemists have a favourite molecule. Professor Phil Robinson, of the Children’s Medical Research Institute in Sydney, has a favourite protein: the small GTPase, dynamin-1. And it’s this protein that will be the focus of his talk at the Lorne Conference on Protein Structure and Function in February.

Dynamin-1 is one of a whole family of dynamins that all play the same role in controlling cell internalisation by endocytosis. Dynamin-1 is a bit special in that it controls the endocytosis of synaptic vesicles in nerve cells and, therefore, the transport of neural transmitters and other goodies that make your nerve cells fire. So, if this pivotal protein does not do its thing properly or in a timely manner, the neurons tends to get a bit tired out and not work very well in regard to synaptic transmission. Robinson has been working on dynamin for a very long time now, since his postgraduate days in fact. “And one thing we showed years ago is that dynamin-1 function or activity in neurons is controlled by phosphorylation and dephosphorylation.” This process is the attaching and detaching of phosphate groups on proteins by kinase and phosphatase enzymes, respectively. This simple phosphate-state change is key to the proper functioning of many proteins and systems in biology, including synaptic vesicle endocytosis.

A strong focus of Robinson’s work has been the signalling systems that control dynamin-1 phosphorylation and dephosphorylation because the same signalling is controlling the endocytosis of synaptic vesicles.

“At the start of the project I will talk about at Lorne, we were using mass spectrometry and lots of other tools to find all of the sites of phosphorylation in dynamin. Then we ranked those sites in terms of amount of phosphate added, which responds the most when a neuron is stimulated,” Robinson explains. “There turned out to be seven, and the two that we were most interested in are at amino acid positions 774 and 778.”

The group showed some time ago that these two interesting sites are phosphorylated between nerve cell stimuli, so when a nerve is at rest it phosphorylates dynamin on sites 774 and 778. Then, when the neuron is depolarised, within one second a phosphatase called calcineurin is activated and it dephosphorylates those same two sites. However, what is new and different about Robinsons’s findings earlier this year is that the phosphatase that turns on dynamin is calcineurin.

“So, a calcium-dependent phosphatase depolarises the neuron to activate synaptic transmission, which causes calcium influx that turns on synaptic transmission and activates calcineurin, which then dephosphorylates dynamin, thus, endocytosis is turned on.”

The readout for all of this is rates of synaptic vesicle endocytosis (SVE), which measures all the processes including exocytosis that go together for synaptic transmission in the neuron, i.e. one nerve firing onto another.

---PB---

An unexpected duality

“The surprise was in two things we discovered recently,” says Robinson. “Firstly, in all the 28 years I have studied dynamin phosphorylation, we have depolarised a neuron maybe a 100 million times and we have never trapped an intermediate state – dynamin is either phosphorylated or dephosphorylated – and never in between.

“But what we found this year was that swapping from using a mechanical stimulus to an electrical stimulus for the depolarisation meant that suddenly dynamin could be partly dephosphorylated, depending on the strength of the stimulus.”

Thus, it seemed that dynamin dephosphorylation is stimulated fully only by high-frequency stimuli – nothing occurs at low frequencies – and then at medium-frequency stimulation, dynamin is partly dephosphorylated. The important part is that endocytosis of synaptic vesicles still occurs regardless of the intensity of the stimulus.

“In other words, dynamin is not dephosphorylated at low-intensity stimulation, but endocytosis works just fine. So, how can dephosphorylation be specifically triggering endocytosis if it goes fine without it?”

Although completely unexpected, the conclusion was that the endocytosis of synaptic vesicles occurs by at least two pathways and that dephosphorylation is only involved in one of them. Further work showed one to be the classical endocytic pathway mediated through clathrin, while the other turned out to involve a process known for 20 years but not thought to be fast, reversible or really major – it is called bulk endocytosis, or BE.

The job of this type of endocytosis or BE is to retrieve a big chunk of membrane following a nerve exocytosis and release as opposed to the seemingly inefficient vesicle-mediated endocytosis. The thinking up to now by everybody in the field was that BE works like a back up system to replace all the small vehicles with one big prime mover when just too much trafficking is going on – a recovery event that endosomes mount so they can still respond slowly and steadily to a big stimulus. BE has also been described as a pathological scenario whereby if you stimulate a neuron system too far and for too long, it creates a bulk endosome that is used later to regenerate the outer cell membrane.

The kicker in this context was that dynamin dephosphorylation at sites 774 and 778 correlated exclusively with BE activation as opposed to a different dephosphorylation program that occurs when the standard sort of endocytosis is turned on.

“We found that these bulk endosomes were formed within one second and then it was all over by 10 seconds after the stimulus, and this was again stimulus-strength dependent,” says Robinson. “So, the chemical stimulus we have used in the past always turns on the BE which is why we have never seen it as different – it always appeared that just one mechanism was happening.”

---PB---

Splicing it up

This first part of this new endocytosis and dynamin story has been published, but surprise number two is part of unpublished work about calcineurin and BE. Dynamin-1 has two splice variants, something that Robinson‘s team had noticed and wondered about for years.

“However, during a fishing exercise to see what these variants might each do, we found something quite unexpected. The short splice variant is only seven amino acids long, and that exact region binds calcineurin. So, here is dynamin-1 binding to its own phosphatase, and as far as we know, other proteins that are dephosphorylated by calcineurin don’t bind it.

“It seems, therefore, that nature has invented an entire splice insert just to make dynamin a binding partner for calcineurin, and so wherever short-dyn1 goes in a cell it takes calcineurin along with it, thus the enzyme activity is ready to hit dynamin before any other protein.”

In the final piece of the calcineurin chapter, Robinson’s team has identified a peptide that blocks the binding between dynamin and calcineurin when introduced into cells. “And here is the fun part,” Robinson says. “This peptide blocks the binding, but it does not block calcineurin activity in general. When we stimulate neurons, there are other proteins dephosphorylated by calcineurin and it doesn’t block those events, but one protein is no longer dephosphorylated – and that of course is dynamin-1.”

In fact, all other known dynamins in a family of about eight proteins are still dephosphorylated by calcineurin even in the presence of this blocking peptide. “So the best part of that is this: when that peptide blocks dynamin-1 dephosphorylation and not the others, it blocks BE, but not CME. So the strange thing is that dynamin is carrying around its own bound phosphatase and that interaction is necessary for this subset of endocytosis called BE.

Another intriguing thing about this year’s findings, according to Robinson, is that it fits a lot of their old data that never made sense – the sort of data that all researchers have rattling around in their files.

“Suddenly, all these bits and pieces add up and really it makes far more sense that the phosphorylation of dynamin-1 is involved only in BE, whereas if you block dynamin itself with a drug or inhibitor, you block both CME and BE. So normal dynamin, whether it is phosphorylated or not is just doing regular endocytosis – like in non-neuronal cells – but this phosphorylation mechanism is like a souped-up adaptation catering only for the bulk form of endocytosis. It is a potentially powerful tool – a specific molecular probe to unravel the endocytic pathways in neurons – and we will be looking at that.”

---PB---

To the clinic

Robinson runs several other programmes in his lab, and although not specifically part of his talk at Lorne, he was keen to discuss the relevance of one in particular program to the science he will be presenting. His group at the CMRI has had a drug discovery program going on in parallel with their basic research for the last 10 years, specifically for the discovery of small-molecule drugs that bind dynamin and block it.

“This is a massive program in collaboration with Adam McCluskey’s group at the University of Newcastle and we have already made literally thousands of small molecules that block dynamin activity. An important point relevant to the findings mentioned above is that all of these molecules block both the bulk and clathrin-mediated forms of endocytosis.”

According to Robinson, some of the program’s molecules are looking very promising as potential drugs based on their ability to reduce brain seizures in small animal models of epilepsy, which makes sense given their mode of action.

“If you block endocytosis, the neurons would run out of synaptic vesicles eventually because there is no way to replenish them, and that is exactly what seems to happen. For me, this is really exciting stuff. To go from understanding dynamin’s role in the biology of a neuronal cell at the bench to drug discovery programs that make drugs that are not only useful in a lab application, but that look like they might also have clinical applications.”

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