Prix Balzan 2016 pour les neurosciences moléculaires et cellulaires, y compris les aspects neurodégénératifs et du développement
Synthèse panoramique – Rome, 18.11.2016 – Forum (vidéo – anglais)
Our brain, or more generally our nervous system, not only controls every function of our body but it also enables us to perceive signals from the environment and to react to such signals. Whether we listen to a lecture as you do now or run for an exercise, whether we think hard to remember something or just sit there and relax or sleep – none of this would be possible without every function of our body being controlled by our nervous system. Its complexity is staggering and challenges the imagination even of experts such as myself. The nervous system of every one of you contains more than hundred billion nerve cells or neurons. However, this is just the starting point. Each neuron, on average, forms about 1000 connections to other neurons in our brain, resulting in a wiring diagram of incredible complexity. To visualize these structures we need microscopes that can magnify about 100,000fold, which can only be achieved by electron microscopes. Indeed, reconstructing the wiring of only a minuscule part of our brain – e.g. a cubic millimeter of so-called grey matter- is a daunting task that so far has not yet been achieved. With the most modern methods and computers it still requires years of work and many thousand work hours.
Moreover, the contact sites between neurons where information is transmitted are a lot more complicated than simple relay stations for signals as we know them from electronic devices. At these contacts, termed synapses, electrical signals are converted into chemical signals, termed neurotransmitters. Upon arrival of an electrical impulse, a small amount of neurotransmitters is released from the sender neuron and then “read” by the receiving neuron where the signal is again converted into electrical signals. Signal transmission between two cells requires about one thousands of a second – slow by electronic standards, but very fast by biological standards. Principally, this delay time limits the temporal performance of our nervous system. Thus, there is no hope for beating a slot machine in a casino – the number of synapses between seeing the number and executing a movement with your hand limits the temporal resolution. Some animals are a lot faster since fewer synapses are involved – this is why it is not easy to catch a fly.
In a way, the delay time is no surprise if you take a closer look at the process. At synapses, the sender neuron (or presynaptic neuron) forms an enlarged nerve terminal that contains small, membrane-enclosed organelles termed synaptic vesicles. These synaptic vesicles are the storage containers for the neurotransmitters. Upon arrival of an electrical signal, proteins in the surrounding membranes, the calcium channels, change their properties and make the cell membrane, for a very brief period, permeable to calcium ions. This works because under normal conditions the concentration of calcium ions outside the synapse is about 10,000fold higher than within the nerve terminal. Consequently, opening of the calcium channels results in a jet-like burst of calcium that rushes into the nerve terminal. The calcium ions then trigger the fusion of the synaptic vesicle with the plasma membrane, resulting in a discharge of the neurotransmitter.
As a student of biology in the seventies of last century, I was asked to give a seminar covering the mechanisms of neurotransmitter release as part of my exam requirements for becoming a high school teacher. I had absolutely no idea about the topic and thus needed to spend many hours in the library. At this time it was well established that neurotransmitter was released in small portions of defined size, termed quanta. It was also known that synaptic vesicles were capable of storing transmitter, largely based on the seminal work of Bernhard Katz. His beautiful book “Nerve, Muscle, and Synapse” had been published a few years earlier, and I studied it literally page by page. However, most scientists, including Katz himself, could not imagine that a process as fast and efficient as neurotransmitter release would involve exocytosis. Indeed, there was a major debate trying to explain how the transmitter could get out of the vesicles and out of the synapse so quickly. However, at this time exciting work was published that changed this view, with two scientists – one American (John Heuser) and one Italian (Bruno Ceccarelli from Milan) being instrumental for the change of concept. These scientists used a “classical” synapse – the synapse connecting a neuron to a muscle cell and telling it to contract. In comparison to the synapses in the brain, these synapses are gigantic and thus were easily accessible, allowing for combining sophisticated physiological experiments with electron microscopy. In hindsight, the work of these scientists was indeed instrumental and has profoundly influenced our understanding of the synaptic vesicle cycle in nerve terminals. Personally I was fascinated by these papers, and I spent many hours trying to understand every detail of the experiments.
After my exams I switched to different topics but retained an interest in the field. Six years later, when joining the laboratory of Paul Greengard at Yale, and then at Rockefeller, I was finally able to pursue my own research in this area. This was the time when the revolution in molecular biology was about to transform life sciences. Moreover, increasingly sophisticated tools had become available for studying membrane proteins such as monoclonal antibodies. Intriguingly, with the exception of one protein studied intensely in the Greengard lab (termed synapsin I) there was still no information about the protein molecules and the molecular mechanisms that could mediate a task as complex as the exocytotic release of a synaptic vesicle.
In the Greengard laboratory, we were studying the phosphorylation of synapsin I. In particular, we were interested to find out in which way phosphorylation changes the properties of the molecule – a topic that is still being investigated today. In the laboratory, I learned how to purify synaptic vesicles to a very high degree of purity, which was needed to study binding and dissociation of synapsin I from synaptic vesicles. However, it was clear that synapsin I, while certainly being a central regulator of synaptic functions, was not directly involved in the process of calcium regulated exocytosis.
In the mid-eighties, I returned to Germany where I had the opportunity to establish my own group at the Max-Planck-Institute for Psychiatry in Munich. The key question then was which proteins mediate exocytosis and how can they be identified? We argued that our best bet was to characterize the main protein constituents of synaptic vesicles, for two reasons. First, a back of the envelope calculation suggested that proteins present in at least a few copies on each synaptic vesicle were expected to be rather abundant. Our assumption was that whatever the protein machinery mediating fusion is doing, it must connect to one of the vesicle membrane proteins, and this protein must be present in every synaptic vesicle irrespective of the specific features of the neuron. Second, I had just learned how to purify synaptic vesicles in large quantities, thus providing us with enough starting material to characterize the main membrane proteins. Moreover, I had become very proficient in generating monoclonal and polyclonal antibodies, essential tools for characterizing proteins. In these years, we teamed up with two scientists: Pietro V. De Camilli, a fantastic cell biologist, then in Milan, and Thomas C. Südhof, a superb molecular biologist, then in Dallas. I had become friends with both of them during my postdoc years, and we shared a keen interest in exocytosis. We all were young and free to choose our research topics and we decided to jointly tackle the problem by taking advantage of our complementary expertise, even though the approach was criticized by some as merely being a biochemical “stamp collection” of proteins with no function. The first protein we characterized, termed synaptophysin, is still somewhat enigmatic, and we do not yet have a clear idea what it is doing. With the next ones, however, we hit “pay dirt” since these included the SNARE synaptobrevin (also referred to as VAMP), now known to be a core constituent of the protein machinery mediating exocytotic membrane fusion, and synaptotagmin, now known to be the prime calcium sensor for transmitter release in neurons.
The early nineties of the last century, in which these discoveries were made, belong to the most exciting period in my scientific career. In the time before we were virtually on our own in working on the molecular foundation of exocytosis. However, in a very short time the field literally “exploded”, with several completely different lines of research suddenly coming together. First, it became apparent that the molecules involved in vesicular traffic are all derived from common ancestors conserved among all eukaryotes. Vesicle traffic was particularly well studied in baker’s yeast by the pioneering work of Randy Schekman. For me, as a neuroscientist working on synapses this was a surprise which I did not expect. In a way, synaptic exocytosis can be considered as an adaptive modification of a mechanism that allows yeast cells to grow. No surprise that our colleagues working on yeast were joking about the brain being derived from a yeast infection of the head. A second line of research involved the development of assays, mostly in the laboratory of Jim Rothman, which allowed for studying fusion of intracellular vesicles in the test tube. Again, these assays originally had nothing to do with neuronal exocytosis but monitored transport of vesicles between intracellular membranes. Having a biochemical assay at hand, it was possible to identify soluble proteins that are required for the reaction. In a serendipitous series of experiments, the Rothman lab then used these proteins to pull out a set of proteins from brain extracts (termed SNARE proteins) that turned out to be the core of the exocytotic fusion machinery in neurons. Around the same time and independently, these proteins were identified as the targets of botulinum and tetanus toxin, with seminal contributions from an Italian colleague (Cesare Montecucco), and also from my own laboratory, together with the late Heiner Niemann. Thus, while starting out by arbitrarily picking completely unknown proteins from purified synaptic vesicles, we suddenly had our hands on a universal membrane fusion apparatus that is conserved among all higher eukaryotic cells. This was indeed a giant leap forward in a very short time! In the midst of all of this excitement, I needed to move back to the US (and lost critical research time during these competitive years) since my position at the Max-Planck Institute was non-tenure with a rather strict time-limit. I established my new laboratory at Yale University School of Medicine where I spent 6 years before again moving back to Germany to my present position.
Since these years our work has focused on achieving a detailed understanding about how the proteins involved in membrane fusion actually function. Assays were developed to study the key proteins in the test tube. Moreover, we and others determined the structures of several of the key proteins using X-ray crystallography and NMR. A concept has emerged showing that the fusogenic SNARE proteins operate like a zipper that pulls two membranes together. They do this spontaneously using energy stored in their conformations, which literally forces the membranes on top of each other and initiates fusion. Intriguingly, there is still no consensus how the calcium sensor synaptotagmin controls this core fusion machine – whether it initiates the interlocking of the SNARE zipper or whether it blocks and releases the zippering mid-way. It is one of our main interests to obtain more details about these steps in order to understand the amazing speed and efficiency of synaptic transmission at the molecular level.
In recent years, we have returned to our starting point from more than 30 years ago, namely to the study of synaptic vesicles in which the neurotransmitter molecules are stored before release. Remember that the number of neurons in our brain is already beyond imagination. If you now consider that each neuron contains on average about 1000 synapses and then consider further that each synapse contains several hundred of synaptic vesicle, you arrive at the almost astronomic amount of 1016-1017synaptic vesicles. If you would line up all synaptic vesicles of your brain on a thread (like a string of pearls) you would need a thread reaching from here to the moon and back in order to find space for all of them. Having so many of them greatly facilitates their purification, and we have taken advantage of this to take them apart in an unprecedented manner. In a major effort, we teamed up with many other colleagues to obtain a complete molecular map of such a vesicle. This map is still far from perfect but offers completely novel insights not only into the structure of the vesicle but also into the molecular composition of biological membranes. In addition, we are presently trying to understand how a vesicle, after being retrieved from the plasma membrane by endocytosis, can be filled with high concentrations of neurotransmitter within less than a minute. This is by no means a trivial question because it not only involves a high degree of specificity but also an intricate balance of ions and other osmolytes. For instance, neurotransmitters differ in their net charge – some are cations such as acetylcholine or dopamine, others are anions such as glutamate, and again others have no net charge such as GABA and glycine. However, the energy for the uptake of all of them is provided by an electrochemical proton gradient, implying that the transport mechanisms for the various neurotransmitters are different.
The project that I plan to carry out with the funds provided by the Balzan Prize will address some rather fundamental questions about neurotransmitter release. As outlined above, we know that many dozen different proteins are required to bring a synaptic vesicle to the plasma membrane and make it release its neurotransmitter content. Indeed, the more we learn the more complicated these tiny machineries seem to be, raising the question how such an extremely complex apparatus can function reliably many million times and for many years in our body. One possible answer to this question may be that the task is divided into defined steps that follow each other in sequence, with each step having a certain degree of autonomy and being able of “self-repair”. Imagine you want to construct a complex device such as a smartphone. You need teams of experts that construct the main chip, the various sensors and devices such as the camera or the microphone, the display etc.. These teams are independent, but they need to talk to each other in order to make sure that the components constructed by each team work together in the end. Finally, you then need specialist teams that put the whole device together. Assume you want to know how the teams work but you are not allowed into the construction hall, but you are able to provide the teams with some material. For instance, you construct a display that may not be quite as sophisticated as that used by the company, and then “feed” it to the assembly hall. If you did a good job, your display may be accepted by the team connecting it to the other parts. This will allow you to see whether it will work in the finally assembled phone even if it may not be as perfect as the original one. This approach is exactly what we plan to do with synaptic transmission, with all experiments being carried out by two young doctoral students. We will use standard cells of our bodies (in fact, tumor cell lines that can be easily kept in the laboratory), and then introduce specific proteins and artificial vesicles into these cells in order to see whether the cells can use these parts to carry out the task of calcium-regulated exocytosis. We hope that with these experiments we will obtain fundamental insights into the molecular organization of vesicle traffic and of exocytosis.